Trade, Exchange Rate, and Agricultural Pricing Policies in ...

8399WORLD BANK I

COMPARATIVE STUDIES IF !ILE CO>Y

The Political Economy of Agricultural Pricing Policy

Trade, Exchange Rate,and Agricultural Pricing Policiesin Ghana

J. Dirck Strykerwith the assistance ofEmmanuel Dumeau,Jennifer Wohl,Peter Haymond,Andrew Cook, andKatherine Coon

A - _ :_

i1:it rOF

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

The Political Economy of Agricultural Pricing Policy

Trade, Exchange Rate,and Agricultural Pricing Policies

in Ghana

J. Dirck Strykerwith the assistance ofEmmanuel Dumeau,

Jennifer Wohl,Peter Haymond,

Andrew Cook, andKatherine Coon

WORLD BANKCOMPARATIVE STUDIES

The World BankWashington, D.C.

Copyright © 1990The International Bank for Reconstructionand Development/THE WORLD UANK

1818 H Street, N.W.Washington, D.C. 20433

All rights reservedManufactured in the United States of AmericaFirst printing February 1990

World Bank Comparative Studies are undertaken to increase the Bank's capacity to offer soundand relevant policy recommendations to its member countries. Each series of studies, of which ThePolitical Economy of Agricultural Pricing Policy is one, comprises several empirical, multicountryreviews of key economic policies and their effects on the development of the countries in which theywere implemented. A synthesis report on each series will compare the findings of the studies ofindividual countries to identify common patterns in the relation between policy and outcome-thusto increase understanding of development and economic policy

The series The Political Economy of Agricultural Pricing Policy, under the direction of Anne0. Krueger, Maurice Schiff, and Alberto Valdes, was undertaken to examine the reasons underlyingpricing policy, to quantify the systematic and extensive intervention of developing countries in thepricing of agricultural commodities during 1960-85, and to understand the effects of suchintervention over time. Each of the eighteen country studies uses a common methodology tomeasure the effect of sectoral and economywide price intervention on agricultural incentives andfood prices, as well as their effects on output, consumption, trade, intersectoral transfers,government budgets, and income distribution. The political and economic forces behind priceintervention are analyzed, as are the efforts at reform of pricing policy and their consequences.

The findings, interpretations, and conclusions in this series are entirely those of the authors andshould not be attributed in any manner to the World Bank, to its affiliated organizations, or tomembers of its Board of Executive Directors or the countries they represent.

The material in this publication is copyrighted. Requests for permission to reproduce portions of itshould be sent to Director, Publications Department, at the address shown in the copyright noticeabove. The World Bank encourages dissemination of its work and will normally give permissionpromptly and, when the reproduction is for noncommercial purposes, without asking a fee.Permission to photocopy portions for classroom use is not required, though notification of such usehaving been made will be appreciated.

The complete backlist of World Bank publications is shown in the annual Index of Publications,which contains an alphabetical title list and indexes of subjects, authors, and countries and regions;it is of value principally to libraries and institutional purchasers. The latest edition is available freeof charge fiom Publications Sales Unit, Department F, The World Bank, 1818 H Street, N.W.,Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116Paris, France.

J. Dirck Stryker, an economist with Associates for International Resources & Development, is aconsultant to the World Bank.

Library of Congress Cataloging-in-Publication Data

Stryker, J. Dirck.Trade, exchange rate, and agricultural pricing policies in Ghana !

J. Dirck Stryker ; with the assistance of Emmanuel Dumeau [et al.].

P. cm.

ISBN 0-8213-1443-21. Agricultural prices--Government policy--Ghana. 2. Coco.a trade-

-Government policy--Ghana. 3. Ghana--Commercial policy. I. Title.HD2146.Z775S76 1990338.1'8--dc2O 89-77546

CIP

Abstract

At the time it achieved independence in 1957, the West Africancountry of Ghana was the world's leading producer of cocoa and the mostprosperous nation in Sub-Saharan Africa other than South Africa. By 1984,however, per capita gross domestic product (GDP) measured in constantprices had fallen 30 percent below its level in 1950.

Ghana's economic decline during the period covered by this studyof agricultural prices obviously cannot be attributed solely to governmentprice intervention. But intervention in the workings of the cocoa sector,this study shows, contributed heavily to the country's inability to achieveprosperity and stability after 1957. More than half of Ghana's populationwas still employed in agriculture in 1984, and the sector provided abouttwo-thirds of the country's export earnings, with cocoa by far thecountry's major export. Other important factors in Ghana's difficultiesincluded a succession of military coups between the late 1960s and theearly 1980s that were ordinarily followed by repression and coercion of thegeneral population, and deterioration of the country's road system.

During the decades since independence, direct intervention inGhana's all-important cocoa sector has been in the hands of a CocoaMarketing Board (CMB), which sets annual producer prices, purchases thecrop from domestic producers, and markets it to foreign buyers. (Theimportance of cocoa to Ghana's economy during the study period wasconsiderable in most years, with the product frequently accounting for 20percent or more of the government's annual revenues, and in some years morethan 50 percent.) Although the chief reason for creating the CMB was toassure Ghana's cocoa farmers a stable and decent income, the agency'sdirect intervention helped to keep producer prices lower than they mighthave been otherwise. The problems of cocoa farmers in earning income werecompounded during much of the study period by export taxes on cocoa andovervaluation of the domestic currency, both of which also tended todepress Ghana's earnings of foreign exchange.

The government's direct and indirect intervention in the cocoamarket, according to the study, far outweighed its incentives to cocoaproducers, including free entry of imported inputs like fertilizer andpesticides, subsidized tractor services, and cheap credit. Moreover, mostof the benefits of these incentives went to large producers rather than thefar more numerous smallholders.

Another important finding of this study (which also surveys theeffects of intervention on two principal imports, rice and maize, and onthe nontraded products cassava, yams, and sorghum/millet) is thatgovernment regulation of the cocoa sector had the serious negative long-term effect of deferring the replacement of old coffee trees with new ones.

iv

By 1982, the CMB and its subsidiaries had 109,000 employees in acountry with a population of approximately 12 million, and low producerprices were encouraging the smuggling of large amounts of cocoa toneighboring Cote d'Ivoire. Large segments of the population had abandonedproductive economic activities in favor of cultivating influential patrons(e.g., those who had licenses to import goods in a country where importshad been sharply restricted), and the government was hard at work seekingemergency assistance from the International Monetary Fund (IMF). With thisassistance in hand, Ghana then began to dismantle its administrativecontrols over the price of cocoa, rice, and maize.

v

Table of Comtents

Pa-ne

FRRT CNE: EIINIMIC POLICY AND AGRICLLURE

I. CWPTER I: INTRODUCTION 1

II. O-PTER II: AIQULTFE IN FELATICN TO tHAN'S EIXN2Y 6

Background 6Agricultural Resatr-ce Endrbawnent 6Regional Cropping Patterns 7Pbpulation 9Overall Econsmic Perfor-mance 13The Balance of Payments and Exchange Rate Disequilibrium 17

The Agricultural Sector 21Importance of Agriculture in the Economy 21Production of Specific Crops 23Food Cmsumptimn 30

II. CFRPTER III: THE EVOLVING PCLITICPL ECONOMY - THE PLAYERS 32

Interest Groups 32Sole of the State 35Patrmns 3B

IV. CHAPTER IV: THE EVOLVING POLITICAL ECUONOY - THE PERIOD6 39

Liberal Regime, 1950-60 39Imposition of Cbntrols, 1961-63 43Breakdown of the System, 1964-66 48Austerity, 1966-67 50Devaluatimn and Import Liberalization, 1967-70 53Collapse of Import Liberalizatim, 1971-72 56Early Years of the NRC, 1972-75 59Ecoromic Disintegratimn, 1976-78 65Struggle for Reform, 1978-81 69PNDC and the Ecrnomic Recovery Program, 1982 - Present 76

V. CHPTER V: OEKbR%NMENT PCLICIES TOWP1DS AGRICL*TURE 86

Cocoa Pblicies 87Tradable Foxds 103Transportation and Nbntradable Foods 111Policies Related to Agricultural Inputs 112

Input Subsidies 113Fertilizer 113Machinery Services 116Improved Seeds 117Other Inputs 118

Credit 120Research and Extension 121

vi

Page

State Prnidctian 124Cbnclusions 126

FPAT W i EFFt TB OF PRICE INTERON0TIME 129

VI. OHPPTER VI: MEASURES OF INTERVENTION 129

Direct Effects 129Producer, Consumler, and Border Prices 129Relative Prices 136Measurement of Direct Price Interventions 142

Indirect Effects 150

VII. CHWPTER VII: EFFECTS ON OLTRJT, COELUFPrTI(N,PM FCEIGNJ EXWY*3E 156

Effects on Agricultural Production 156Supply Functicns 157

Cocoa 156Food Crops 164Price Elasticities of Supply 166

EquilibriLmn Levels of Output 167Effect on Consumption 179Effect on Net Foreign Exchange Earnings 184

VIII. CHAPTER VIII: GiNVEFIIENT LEUGET AND OTH}ER FE9URCE TFW1SFERS 200

Effects of Price Policy cn the Eovermnent BEdget 200Transfers of Resources Between Agriculture and the Restof the Ecconwoy 204

Estimates of Resource Transfers 204Gbvernment Investment and Expenditure Bias 217

IX. CHAPTER IX: OTHER FRICE INTEF&ENTIONI EFFECTS 219

Farm Income Effects 219Variability Effects 230

PART THRE: THE FAILLUE CF THE POLITICPL SYSTEM 248

X. CHFrMR X: HYPOTHEEES PAD CONCLULIONS 24B

The HypDtheses 248Importance of Macroeconomic Disequilibrium 248Loss of Real Incone 248Failure of the Political System 249Rise of Rent-Seeking Activity 249Neglect of Price Policy 250Failure to Achieve National Objectives 250

The Phases 250Colonial Period 251Nkrumah 251

vi i

Paqe

National Liberation Ccucil (NLC) and Busia 252Natianal Redemption Cox,cil (NFC), Sepme Military Caoncil

(SM), ard Limam2Prfvisicmal National Defence cxuricil (PNDC) 252Smnmary 253

Testing of the Hypotheses 255Macraeccxnic Disequilibriun 255Loss of Real Inccwme 257Failure of the Political System 256Rent-Seeking Activity 258Neglect of Price Policy 260Failure to Achieve National Objectives 261

Canc lusions 263

ANNEXES

AMEX 1: AmIOrC T FFXCDIWI 265System for Data Collection 265

Cocoa ~~~~~~~~~~~~~265Other Crops 269

Agricultural Production Data 270

ANNEX 2: ERWILIBRIUM EXCHANGE RATE 276Purchasing Power Parity Approach 276The Real Exchange Rate 276The Equilibrium Exchange Rate 282

Elasticities Approach 287Equilibrium Exchaige Rate Mbdel 294

Import Demand Function 294Export Supply Functions 295Adjustment for Mbnopoly Power 296The Exchange Rate Model 298

PPRENDIX TO INNEX 2: DERIVATION OF A FCRULLA FUR ESTIPFTING THEEaJILIBRIUM EXCOM RATE USING THE EATICITY PPRAH301

ANNEX 3: DERIVATION CF PRICES PAD RRICE IMDICES 305

ANNEX 4: COMA SUPLFY FUNCTICN 316

ANNEX 5: :FEMTS OF INDIFECT AND NNARICULTLUR DIFECTRICE INTEVENTIO6 329

ANNEX 6: NEr AVAILABILITY OF RICE AM M%iIZE 334

ANNEX 7: TONERS AM 9USLIDIES 335

ANNEX 8: REAL ELJILIRIUM INME 346

viii

List of Tables

Panie

Table 1 Population 10Table 2 Gross Domestic Product, 1950 - 1984 14Table 3 Economic Indicators Related to Inflation 16Table 4 Current ARccount Balance, Actual and

Equilibrium Exchange Rates 18Table 5 Agricultural Sector 22Table 6(1) Prouction Indices 24Table 6(2) Productian Indices 25Table 6(3) Production Indices 26Table 7 Food Production and Consumption Indices 31Table 8 Cocoa Sales, Marketing Costs, and Public Revenue 89Table 9 Real Producer Price of Cocoa 101Table 10 Domestic Producer Prices 133Table 11 Domestic Consumer Prices 134Table 12 Border Price Equivalents 135Table 13(l) Prevailing Relative Price Indices 137Table 13(2) Prevailing Relative Price Indices 138Table 13(3) Prevailing Relative Price Indices 139Table 13(4) Prevailing Relative Price Indices 140Table 14 Producer Price Ratios 143Table 15 Consumer Price Ratios 144Table 16 Effect of Direct Price Interventions en

Relative Producer Price Differences 146Table 17 Effect of direct Price Interventicons on

Relative Consufer Price Differences 147Table 18 Effect of Direct and Indirect Price Interventions

on Relative Prices 152Table 19 Effect of Direct and Indirect Price Interventions

on Relative Price Differences 153Table 20 Direct Effect en (btput, Short-fRn 170Table 21 Direct Effect en Ohtput, Long-Fiun 171Table 22 Direct Effect en Output, Very Long-fRn 172Table 23 Total Effect en Output, Short-Run 173Table 24 Total Effect an Output, Long-Run 174Table 25 Total Effect en Output, Very Long-fhn 175Table 26 Direct Effect on Consumption 182Table 27 Total Effect en Consumption 183Table 2B Direct, Short-fun Effect of Price Interventions

on Foreign Exchange Earnings 116Table 29 Direct, Lcng-Rufn Effect of Price Interventions

an Foreign Exchange Earnings 188Table 30 Direct, Very Long-Run Effect of Price Interventions

on Foreign Exchange Earnings 190Table 31 Total, Short-Rkn Effect of Price Interventicns

on Foreign Exchange Earnings 192

ix

Paae

Table 32 Total, Long-fin Effect of Price Interventionson Foreign Exchange Earnings 194

Table 33 Total, Very Long-Run Effect of Price Interventionson Foreign Exchange Earnings 196

Table 34 Effect of the Pricing Policy an the Budget 202Table 35 Tax on Cocoa 203Table 36 Direct and Total Nominal Short-fin Transfers Due to

Ckitput Price Interventions Into (+)/ Out of(-) Agriculture 209

Table 37 Direct and Total Nominal Long-Rhn Transfer-s Due toGOitput Price Interventions Into (+)/ Out of(-) Agriculture 210-

Table 38 Direct and Total Nominal Very Long-Run Transfers Dueto Output Price Interventions Into (+)/ Ouit of(-) Agriculture 211

Table 39 Short-Run Transfers Into (+)/ Out of (-) Agriculture 212Table 40 Long-fin Transfers Into (+)/ Out of C-) Agriculture 213Table 41 Very Long-fun Transfers Into (+)/ Ouit of (-) Agriculture 214Table 42 Gbvernmnt Investment (GIB) and Total Expenditure

(GEB) Bias 218Table 43 Real Instantaneous Income Effect of Direct Interventinu 2 721Table 44 Real Short-Run Income Effect of Direct Intervention 222Table 45 Real Long-fRun Income Effect of Direct Intervention 223Table 46 Real Very Long-Run Income Effect of Direct Intervention 224Table 47 Real Instantaneous Income Effect of Total Intervention 225Table 48 Real Short-Rfn Income Effect of Total Intervention 726Table 49 Real Long-Run Income Effect of Total Interventian 227Table 50 Real Very Long-Run Income Effect of Total Intervention 22BTable 51 Domestic Maize Prices Under Alternative Price Scenarios 232Table 52 Domestic Rice Prices Under Alternative Price Scenarios 236Table 53 Domestic Cocoa Prices Ukhder Alterative Price Srenarios 240Table 54 Per Capita Production and Consumption 244Table 55 Relative Weight of Agricultural Price Policy Objectives 254

fnnex Tables

Table 1-1 Cocoa Production and Smuggling 1960-1962 1.2Table 1-2(1) Agricultural Production 1.8Table 1-2(2) Agricultural Production 1.9Table 1-2(3) Agricultural Production 1.10Table 2-1 Exchange Rates, 1956-85 2.4Table 2-2 Balance of Payments Current Account 2.8Table 2-3 Nominal Exchange Rates and Nominal Purchasing Power 2.13

Parity Equilibrium RatesTable 2-4 Estimation of the Equilibrium Exchange Rate Using 2.18

the Elasticities ApproachTable 2-5 Norninal Exchange Rate and Alternaltive Nominal 2.25

Equilibrium Exchange RatesTable 3-1 Wholesale Market Prices 3.2Table 3-2 Official Producer Prices 3.4Table 3-3(1) Structure of Domestic Prices 3.5

x

Table 3-3(2) Structure of Domestic Prices 3.6Table 3-4(1) Stucture of Rice Border Price Equivalents 3.8Table 3-4(2) Structure of Maize Border Price Equivalents 3.10Table 3-4(3) Structure of Border Price Equivalent for Cocoa 3.12Table 3-5(1) National Consuimer Price Indices 3.13Table 4-1 Estimation of Cocoa Planting, Traditional Varieties 4.6Table 4-2 Estimation of bormal Cocoa Production 4.8Table 4-3 Regression Results for the Cocoa Supply Equation 4.11Table 5-1 Effect of Indirect and Nonagricultural Direct Price 5.3

Interventions an Relative PricesTable 5-2 Effect of Indirect and Nonagricultural Direct Price 5.5

Interventions on Relative PricesTable 6-1 Net Availability of Rice and Maize 6.1Table 7-1(1) Short-Rgn Producer Price Transfer to and from 7.1

Maize ProductionTable 7-1(2) Long-fu-i Producer Price Transfer to and from 7.2

Maize ProductionTable 7-2(1) Short-Rn Producer Price Transfer to and fromn 7.3

Rice Pr-oductionTable 7-2(2) Lanx-flun Producer Price Transfer to and fron 7.4

Rice ProducticnTable 7-3(1) Short-FRn Producer Price Transfer from Cocoa Production 7.5Table 7-3(2) Long--Run Producer Price Transfer from Cocoa Production 7.6Table 7-3(3) Very Long-Run Producer Price Transfer from Ccxoa 7.7

ProductionTable 7-4 Indirect Input Subsidies to Agriculture 7.8Table 7-5 Government Expenditure Transfers to Agriculture 7.9Table 7-6 Total, Very Long-fRn Estimation of Agricultural GDP 7.10

in the Absence of Price InterventionsTable 7-7 Ciovernment Expenditures 7.11Table 8-1(1) Large Cocoa Farmers, Direct Real Equilibrium Income 8.1Table 8-1(2) Large Conoa Farmers, Total Real Equilibrium Income 8.3Table B-2(1) Large Rice Farmers, Direct Real Equilibrium Inca.r 8.5Table 8-2(2) Large Rice Farmers, Total Real Equilibrium Income 8.7Table 8-3(1) Large Maize Farmers, Direct Real Equilibrium Income 8.9Table E-3(2) Large Maize Farmers, Total Real Equilibrium Income 8.11Table 8-4(1) Expenditure Shares and Dircet Equilibrium Price Indices 8.13Table 8-4(2) Expenditure Shares and Total Equilibrium Price Indices 8.14

xi

List of Fioures

PaFe

Figure 1 Domestic Prmducer Prices of Maize 234Figure 2 Domestic ronsumer Prices of Maize 235Figure 3 Domestic Prcducer Prices of Rice 23BFigure 4 Domestic Consumer Prices of Rice 239Figure 5 Domestic Producer Prices of Cocoa 242Figure 6 Domestic Consumer Prices of Cocoa 243Figure 7 Maize Producticn and Availability 246Figure 8 Cbnsumer Price of Maize 246Figure 9 Rice Productim and Availability 247Figure 10 Consumer Price of Rice 247

Annex Figures

Figure 11 Log Normal and Log Actual Production 4.9

PART OlE: ECONOMIC POLICY AND AGRICLLTIFE

CHPTER I: INTFUJCITIEN

At the time of independence in 1957, Ghana was perhaps the mDst

developed country in black Africa. It was the world's mDst important

producer of cocoa, which permitted the accumulatian of substantial foreign

exchange reserves, and it also exported timber, gold, bauxite, and other

products. Its physical infrastructure and educaticnal establishment were

relatively well developed, and its per capita incaume was the highest outside

of South Africa.

Two and cne-half decades later, the Ghanaian eccnrmy was in ruins.

Cocoa exports were less than half their level of the mid-1960s, the black

market exchange rate was many times the official rate, graft and corruptimn

were rampant, and real per capita incamne had fallen by 30 percent. Only in

1963 did the Ghanaian government at last begin to undertake the extensive

eco-nomic reforms required to rise out of this abyss. The succeeding four

years witnessed a slcw but steady pracess of price and exchange rate

realignment and a dismantling of many of the extensive administrative

c.ntrols that had been utilized to allocate an ever decreasing supply of

available resources.

Funtdamental to the dissolution of the Ghanaian eccwmy during this

period was agricultural price policy. To some extent this policy was the

result of deliberate decisimns regarding producer and ccnsumer prices. This

was true of cocoa, for example, as well as other export and industrial

crops. Official price policy regarding foodgrains, an the other hand, was

unimportant in relation to trade policy, especially restrictions mn imports

of cereals.

The most important influence on agricultural prices, however, was

not agricultural price and trade policy per se, but the overall

macrouconaomic situation in which these policiles were fornmuilated and

implemented. Of special importance were periodic bouts of inflation and

overvaluatimn of the exchange rate.' These were especially severe during

the last half of the 1970s and the early 1960s before the recent reforms

un-dertaken by the Rawlings government.

This study analyzes the past three decades of Ghanaian economic

history in light of the impact that agricultural price distortions have had

an the allocation of resources and the welfare of producers and consumers.

The analysis is concerned not only with the magnitude of these effects but

also with how the resulting gains and losses of different graups in Ghanaian

soriety have fed back into the political process to shape agricultural price

policy. A major ccnclusion of the analysis is that these feedback effects

did not work very well because the political system increasingly prevented

the voices of special interest groups from being heard. Instead of

resources being allocated through policy decisions influenced by these

groups, access to goods and services came to depend on a vast network of

patrcn-client relations, which depended on favoritism, personal

connections, and outright bribery. So much time and energy was devoted to

enhancing these relationships that there was little left over for productive

endeavor. As a result, the total supply of resources available for

allocation continued to dwindle until the rewards no longer justified the

effort, and people simply withdrew from the formal sector of the ecanomy.

± The use of the term overvaluation in this report with reference bothto the exchange rate and to local currency (the cedi) implies N' 1$ rate islow in relation to its equilibrium level.

2

Finally Jerry Rawlings cmsted the civilian regime in power and, after an

initial period of varillatimn and hesitancy, embarked on a process of

fundamental ecorinnic reform that is yet to be concluded.

This study is divided into three parts. The first discusses the

economic policies that have affected agriculture. Chapter II describes the

agricultural sector in relation to the rest of the eccnomy. It provides

some general background on Ghana's agricultural resource endcwrient,

regional cropping patterns, population, cverall economic performance, and

balance of payments. Included in the last section is a discussion of the

exchange rate disequilibrium that existed in Ghana during most of the past

three decades and the methods that were used to estimate the free trade

equilibrium exchange rate. These methods, including an econometric model of

the demand for and supply of foreign exchange, are described in detail in

Aninex 2. The rest of Chapter II looks quantitatively over 30 years at the

relative importance of agriculture in the Ghanaian economy, the production

of specific crops, and food consumption.

Chapters III and IV discuss the evolving political economy of

Ghana. The first of these chapters describes the major players in this

evolution, including interest groups, the state, and the patrons. Chapter

IV examines each of the periods through which the political economy has

evolved from 1950 to the mid-1980s. In particular, it analyzes

qualitatively the ways in which the policies and instituticns created during

the early years of independence created allocative and distribution

mechanisms that inhibited the exertion of influence by interest groups on

the formulation of policy in ways that would have directly increased their

inCome and wealth. Later the dependence of successive governnents on

3

patronage networks and their inability to draw upon interest grmups for

political support reduced their capacity to unrdertake fundamental reform.

Chapter V describes the government policies that have affected

agriculture, especially those related directly or indirectly to relative

prices. Emphasis is placed on price distortions introdcijed by trade and

exchange rate policies, which have often run couinter to policies designed to

influence agricultural incentives directly. Attention is also paid to the

use of administrative decisions in place of price incentives as a

mechanism for allocating resources and the rent-seeking activity that has

resulted. Policies are examined as these have affected the prices of cocoa,

tradable foods, transportation and nontradable foods, and agricultural

inputs. There is also some discussion of non-price policies regarding

credit, research and extension, and state production.

Part Two asse quantitatively the effects of price

interventions for three major crops: cocoa, rice, and maize. Chapter VI

discusses the measurement of these interventions, including both the direct

effects of trade and price policy and the indirect effects resulting frc

distortions in the exchange rate. The methods used to obtain these

measurements are described, and the results are discussed.

Chapter VII looks at the direct and indirect effects of price

interventions on output, consumption, and net earnings of foreign exchange.

The effects on output and cotsumption are measurned using economictrically

estimated supply and demand functicns for cocoa, rice, and maize. The

estimation proceduwres are described summarily in this chapter and in more

detail for cocoa in Ainex 4. Changes in output and consuimption resulting

frcom price interventions are then used together w:ith border prices in U.S.

4

dollars to estimate the effects of the interventions an net foreign

exchange earnings.

Chapter VIII examines the effects of price interventicns an the

government budget and an the welfare of producers and consmiers. Net

transfers into and out of agriculture are estimated as these result from

price distortians and government expenditures. The bias of government

investment and total expenditures an agriculture in relation to the whole

ecacnwy is also assessed.

Chapter IX estimates the impact of agricultural price

interventicns an the real income of several representative types of farmers.

It also examines the effects an urban consumers and other interest groups

but concludes that these effects were dwarfed during most of the period

studied by large distortions in real wages that were anly to a very minor

extent the result of changes in the relative prices of the crops under

ccnsideratimn. The last part of this chapter looks at the effect of price

interventions an the variability of domestic prices and quantities.

Part Thr-ee of the study integrates the qualitative discussion in

Part One with the quantitative analyses of Part Two by elaborating several

major hypotheses that may be drawn from the study. These hypotheses are

first described and then analyzed over six major historical phases, each of

which is defined by the relative priorities-given by successive governments

to various naticnal objectives. In the end, it is argued, the instruments

employed to attain these objectives were inadequate in the face of much more

powerful political and economic pressures that resulted in the demise of

each of these regimes until the present ane. Whether this pattern has now

beEn changed remains to be seen.

5

CHETER II: AGRICLLLUE IN FELATION TO G~V S EEXU'W

This chapter describes the agricultural sector in relation to

Ghana s overall econouy. The first sectim provides backgrmund on the

agricultural resource endowhent, regional cropping patterns, population

characteristics, and overall economic performance. The seccand section

examines the relative importance of agriculture in the econcomy, the

productin of specific crops, and levels of food consumption.

Backqround

APricultural Resource Endowment'

Ghana is located on the coast of West Africa, bordered in the east

by Togo, in the West by Ivory Coast, and in the north by Burkina Fasso. The

ccintry s total area is roughly 92,100 square miles, with elevaticns

generally ranging between sea level and 1000 feet. The ccxntry is divided

administratively into ten regions - Greater Accra, Eastern, Western,

Central, Brung Ahafo, Ashanti, Vol1ta, Upper East, UJpper West, and Northern

(see Map 1).

Generally, the resource endaowent ccnsists of a forest zone in the

south, occupying roughly me-third of the country, and a wooded savanna in

the norrth. The forest zone receives much nmre rainfall than the savanna.

The highest rainfall area in the southwest averages 1950-2125 mm per year.

X MDre extensive discussions of the agricultural resource endowment inGhana are cantained in Area Handbook for Ghana, prepared by Foreign AreaStudies of the American LUniversity, 1971 and World Bank, Ghana: AqriculturalSector Review, April 12, 1978, Arnex I.

6

In contrast, the northern savanna averages 800-1200 mm per year and scme

areas of the southern coast average only 625-1000 mm per year.

The principal features of rainfall in Ghana are its seasDnality

and its interannual variability. The distribution during the year varies

considerably from region to region. Two principal types may be recognized,

though adjacent types grade into one another. The first is characterized in

the scuth by relatively heavy rains in May and June and a lesser rainy

season around October; the second is a single rainy season in the north,

principally from June through September. Large variations exist between

successive rainy seasons in time of onset, duration, and amounts

received. In snme seasons, individual rains are numerous and well

distributed and in other years they are scattered and infrequent. The

result is high variability of crop production.

Soils in Ghana are of generally pmor quality and easily become

exhausted. The maintenance of the topsoil organic matter is of prime

importance in cultivating these soils. This is largely achieved through

traditional arable farming techniques involving extensive periods of

fallow. In some areas, howEver, growing population density necessitates

shortening of fallow, leading to a loss of nutrients and the possibility of

erosion.

Reqional Croopinq Patterns

There are five major geographic regions: the coastal plains in the

south, the forest uplands further north, the high plains of the far north

and northwest, the basin area surrounding Lake Volta and its assoriated

rivers, and the rugged, mountainous region of the Akwapim-Togo ranges.

7

The coastal savanna area along the ocean receives little rainfall

compared to the rest of the country but contains alluvial soils of good

fertility. Mbst of the region is characterized by subsistence farming of

maize, cassava, and grcaxuinuts. Sugar cane is grawin under irrigation. The

section of the Arccra Plains near the coast is nearly free of the tsetse fly

and has therefore become a popular area for livestock breeding. Market

gardening has also been increasing near Accra. The coastal plain to the

west of Accra is unsafe for cattle, but a number of commercial centers and

fishing villages are found in the region. Cocanut palms are also

maintained. The Volta Delta area is characterized by flat la-nd cavered with

grass and fan palms. Coconuts, oil palm, cassava, corn, and a variety of

vegetables are grown. Near Keta there is intensive cultivation of shallots.

Fishing is also important. The Densu River basin in the eastern section of

the lowlands area is a rich cocoa and food producing region.

The forested uplands of the Eastern, Western, Central, Brong

Ahafo, and Ashanti Regicns receive substantial amouts of rainfall and have

beccne Ghana s most important area for cocoa prcouction. Traditicnal oil

palm production is also faund extensively in the higher rainfall areas in

the southwesterm part of this zone. Subsistence crops include maize,

cassava, plantains, cocaiyams, and yams. The western part of the region is

least densely populated, and forestry is an important activity. The

potential for rubber production here is also caosiderable.

The high plains of the north and northwest average between 500

and 100 feet in elevation. Soils are generally more fertile than in the

Volta Basin, and population density is accordingly higher. Food crops

include yams, maize, sorghum, millet, and grounchnuts. Cash crops are

H

cottan, tobacco, and kenaf, with rice being proKuced for sale in the river

valley bottomlands. Largely because of the virtual absence of the tsetse

fly, this regimn is the cauntry s primary area for livestock producticm.

The Volta Basin regimD occupies the central part of the ccuntry

and is characterized by poor soil ccnditimns and low annual rainfall.

Pbpulatian density is quite low, and the terrain is subject to drying and

erosim. Fishing is an important activity in the volta Lake regim.

Finally, the AkwapimrTago ranges in the eastern part of the

country cansist mainly of rugged complexes of folded strata and volcanic

rocks. The area is covered with deciduous forest, and temperatures tend to

be a bit cooler than in other parts of the country. tost farming activity

is subsistence-oriented, though some cocoa and coffee are growh along with

staple craps.

Ptclatim

As Table 1 illustrates, populatimn in Ghana has been growing

rapidly. A high crude birth rate of 49 per thousand and a comparatively low

crude death rate of 13 per thousand, estimated in 1982, imply a natural rate

of increase in that year of 3.6 percent per annum. Overall growth was aily

2.9 percent, however, because of substantial cutmigration.m

The ccuntry as a whole in 1970 had an average populatimn density

of about 36 persons per square kilometer.: Outside of Accra, the highest

densities were 91 persmns in the Central Regian and 62 persons in the

Wobrld Bank, Ghana: Pblicies and Program for Adjustment, Washingtm,D.C., 1984, p. 2 6 .

~ Ghana, Central Bureau of Statistics, Statistical Yearbook 1969-70,p.7.

9

Table I

Population

Arabil LandLabor Force Adult per Agri

Total (a) Urban lb) Concentra- Rural Id) Urban as I Part Rate (f) Literacy Worker (h)Year (000) (000) tion (c) (000) Total (e) II) Rate (g) (ha)

1948 4118 326 3 3793 7.9 NIA N/A NIA1950 4368 N/A N/A N/A N/A N/A N/A N/A1960 6804 1585 39 5219 23.3 42.9 27 7.91970 8559 2507 46 6107 29.1 39.7 30 7.51980 11100 4129 N/A 7372 35.9 37.1 30 6.51984 12206 3825 N/A 8380 31.3 N/A N/A N/A

Notes to Table 1:(a) 1949 population obtained from Ghana, Central Bureau of Statistics, Statistical Yearbook, 1969-1970.

1950-1969 populations obtained from World Bank, World Tables, 1983, Vol. 1.1970-1985 population figures are from Ghana: Policies i Issues of StructuralAdjustment, March 30, 1974. Missing data are interpolated using a growthrate of 2.661. Population data for 1948 and 1960 are based on censusesconducted in those years.

(bI 1948 urban population is from Statistical Yearbook, 1969-1970, and includes all people in cities andtowns with a population of at least 5000.1960-1984 figures are derived from data in World Tables, 1983, Vol. 2.

(c) Source: Statistical Yearbook, 1969-1970. The concentration of the urban population is measured bythe number of cities in which the most concentrated 75I of the urban population lives.

(d) Same sources as (a) and (b).le) Derived from same sources as (b).(If) Source: World Tables, 1983, Vol. 2.(q) Source: World Tables, 1983, Vol. 2, for 1960 and 1970; World Banit, Ghana: Policies and Proqraes for

Adjustment, 1984, p. 28, for 1980.(h) Arable land of 14.8 million ha is defined as current cultivation plus fallow from FAD, Perspective

Study of Agriculture and Development for Ghana, 1976.

10

Eastern and Ashanti Regimns. By 1962, average population density overall

was about 50 persons per square kilometer. The highest growth rate during

this period was in the NDrth-ern Regicn at 3.4 percent per year; the Vbota

Region had the lcwest growth rate at 1.7 percent per annum.

The rate of growth in urban areas has in most years exceeded that

in rural areas, with the percentage of total population residing in urban

areas rising from, 7.9 percent in 1948 to 35.9 percent in 1980.4 This is a

higher level of urbanization than is fcaind in most middle-incame ccuntries

in Africa south of the Sahara. With the growth in urban population, there

has been increasing demand for urban services, including the marketing of

food. The rise in the number of cities making up 75 percent of the urban

population also attests to the fact that the increase in urbanization is not

restricted to Accra and a few other larger cities, hut is well distributed

througqhut the country.

The labor force participation rate has declined steadily since

1960. In part this reflects the changing age/sex stricture of the

population resulting from a high birth rate and a relatively low mortality

rate amoig ycLing children. It also undoubtedly is the result of econcmic

stagnation and a falling rate of investment, one effect of which has been to

encourage emigration of persons of working age. In 1980, about 53 percent

of the total labor force was in agriculture and 10 percent was employed in

the formal wage sector, mostly in urban areas.' With 18 percent of the

The apparent decline in urban population from 1990 to 1984, shown inTable 1, may reflect a movement back to the land resulting from the collapseof the economy, as discussed elsewhere in this report, or it may simplyindicate that growth rates extrapolated from the 1960s overestimated urbangrowth during the 1l70s.

W Wbrld Bank, Ghana: Policies and Prouram for Adjustment, p. 26 .

11

labor force officially counted as "unemployed", 19 percent of all workers

were in the informal sector. In fact, the number was much larger than this

since many agricultural, formal sector, and "unemplayed" workers also were

involved with informal sector activities.

The effect of a deteriorating eccromy is also reflected in a

number of social indicators, of which the literacy rate shcown in Table 1 is

ane example. Whereas most middle-income African cauntries made significant

progress in educating their populations during the first two decades

following independence, Ghana, which had ane of the highest literacy rates

in Africa in 1960, has achieved little since then. This situatim is

likely to be aggravated in the future, moreover, because of the lack of

recent investment in social infrastructure to keep up with expanding

population.- Equally important is the loss of Itunan resources associated

with substantial outmigratimn over the past 15 years.

The last indicator shown in Table :L is the average amount of

arable land per person emplcyed in agriculture. While this average does not

indicate any strring pressure m land resourrces, the fact that it is falling

under the influence of growing population implies that more intensive

techniques must be employed if land and labor productively are to be

increased, or at least maintained. Otherwise the reductimn of fallow will

lead to a decline in yields and eventually to degradation of the land.

Given spacial variaticns in population density, moreover, this problem is

much more acute in some areas than in others. This has obvious

implicaticns concerning the need for systems to distribute improved seeds,

fertilizers, and other inputs.

IWorld Bank, Ghana: Policies and Procram for Adjustment, p. 2 7 .

12

Overall Economic Performance

The overall economic performance of Ghana from 1950 to 1984, shown

in Table 2, was very poor. Pbr capita (MP in constant prices during the

entire period declined by 30 percent. Frnm 1950 to 1964, despite some

fluctuations, per capita GDP remained relatively constant at about N* 600

per head. Thereafter there was a reduction, which reached N 559 in 1968,

fol lowed by an increase to NW 640 in 1971. By 1975, a much sharper and more

sustained decrease occurred in per capita GDP that reached a trough of N

396 in 1983.

This poor overall performance was echoed by the behavior of

investment and savings. As a proportion of LIP, gross investment rose f ran

13 percent in 1950 to 21 percent in 1960. The absolute value of investment

continued to rise in constant prices until 1964, when it attained a peak to

which it has never returned. EBth in absolute terms and as a percentage of

GDP, investment fell during the late 1960s, peaked again in 1971, and began

a decline that became especially severe after 1977. By 1982, gross

investment was only 7 percent of GlP and net investment was probably

negative.'

The performance of savings was even worse. In 1950, gross savings

are estimated to have equaled 19 percent of GDP. This figure fluctuated

from year to year, but reached a trough of 4 percent in 1967. It recovered

and attained a maximum of 20 percent in 1972, but thereafter it declined

sharply to only 5 percent in 1980. It later increased scmewhat during the

early 19EKs.

7 WDrld Bank, Ghana: Policies and Program for Adjustment, p.4.

13

Table 2

Gross Domestic Product, 1950-84 (a)(Million New Cedis, 1975 Prices)

lP per Invt.l Savings Savings/ Imports/ Exports/Year GDP capita Invt. GDP (Z) (b) 6DP (2) Imports GDP (2) Exports GDP (2)

__ -- -- -- -- -- -- -- - - --- -- -------_ - -- - - -- -- - --- -- --------_ _ _

1950 2631 602 355 13 488 19 584 22 716 211955 2975 547 462 16 755 25 687 23 979 331959 3680 568 771 21 627 17 1067 29 923 251960 3986 586 836 21 450 11 1370 34 984 251961 4123 592 693 17 308 7 1490 36 1105 271962 4292 602 663 15 655 15 1305 30 1297 301963 4481 613 803 19 590 13 1444 32 1231 271964 4580 612 954 21 777 17 1261 28 1084 241965 4643 606 932 20 593 13 1738 37 1399 301966 4445 567 667 15 638 14 1279 29 1249 281967 4582 570 471 10 201 4 1336 29 1066 231968 4599 559 505 11 474 10 1115 24 1084 241969 4975 579 582 12 402 8 1206 25 1026 211970 5349 625 757 14 804 15 1181 22 1228 231971 5628 640 793 14 976 17 966 17 1149 201972 5488 609 418 8 1084 20 668 12 1334 241973 5646 610 544 10 904 16 968 17 1328 241974 6033 635 789 13 686 11 1146 19 1042 171975 5293 541 673 13 722 14 974 18 1023 191976 5097 509 530 10 678 13 926 18 1074 211977 5212 507 772 15 595 11 994 19 806 151978 5654 536 605 11 465 8 911 16 771 141979 5512 509 537 10 525 10 770 14 758 141980 5538 499 490 9 303 5 842 15 655 121981 5344 469 447 8 317 6 727 14 597 111982 4974 425 344 7 583 12 449 9 688 141983 4747 396 343 7 360 9 469 10 486 101984 5158 423 392 8 276 5 526 10 410 81995 5420 429 N/A N/A N/A N/A N/A N/A N/A N/A

Notes to Table 2:(a) Source: Economic Analysis and Projections Department, World Bank, except where otherwise noted.

Source for 1959a D. Walters, The National Accounts of Ghana, 1955-1961, U.N. Report I TAO/6HA/8,February 4, 1965.

(b) 6ross Domestic Savings, derived as a residual by subtracting Private Consueption and General GverneentqConsumption from Gross Domestic Product.

14

Imports and exports tell a similar story. During the early 1950s,

exports of goods and nonfactor services exceeded imports try a ccmifortable

margin. This was a period during which foreign exchange reserves were being

substantially increased. By 1957, however, the current acccwnt was in

deficit, requiring a drawing dcw of these reserves. Despite this, imports

continued to rise both in absolute ternms (in constant prices) and as a

proportion of GDP. In 1965 imports attained a peak, equal to 37 percent of

GDP, despite the fact that exports were mnly 30 percent of GDP and foreign

exchange reserves were nearly exhausted.3 From this peak, Ghana experienced

an almost continuruis decline in imports, both in absolute terms and as a

share of GAP, until they accounted for mnly 9 percent of GDP in 1982.

Exports behaved similarly and reached a low of 8 percent of GOe in 19B4.

One reason for this dismal ec:onomic performance was inflation

engendered by large government budget deficits, as shown in Table 3. The

budget was in surplus in most years until the 1960s, when rising deficits

under Nkrumah led to double digit inflation from 1964 to 1966. A

considerable effort was made to bring the budget under control during the

ensuing years, which resulted in the rate of inflation dropping to a low of

3.7 percent (0PI) in 1970. With the coming to power of the NRC, however,

the size of the budget deficit in relation to total revenuies increased

steadily until it reached a maximum of 127 percent in 1978. The inflation

rate by this time had reached triple digits. Again there was an effort to

tighten the fiscal situation after the NRC was deposed, but a massive

8 At the end of 1956, Ghana's foreign exchange reserves totaled N*366.5 million; by the end of 1964, these had declined to Nt 83.8 million,or 21 percent of the total value of imports in 1965. J. Clark Leith, ForeiqnTrade Regimes and Economic Development: Ghana, New York: Columbia UhiversityPress, 1974, p.22.

15

Table 3

Econoeic Indicators Relited to Inflation(millions NC)

I Share of SurplusBudget (Deficit) in Consumer AnnualSurplus -------------------- Nominal Price Index Ic) Rate of

Year Govt Rev (a) Govt Exp (a) (Deficit) Govt Rev GDP GDP (bl (1972lOOl Inflation (1)

1950 36 27 9 25.9 2.4 3B8 I/A NIA1951 42 34 7 17.7 1.6 473 32.2 N/A1952 78 46 32 41.0 6.8 469 37.7 17.01953 96 75 11 12.4 2.1 505 38.4 1.91954 101 96 5 5.1 .7 696 36.5 -4.81955 161 91 70 43.5 10.5 668 37.3 2.01956 106 96 10 9.0 1.4 690 38.0 2.01957 99 111 -11 -11.3 -1.5 726 39.4 3.B1958 120 106 15 12.1 1.9 766 39.8 .91959 134 124 10 7.3 1.1 1170 39.8 .01960 140 156 -16 -11.5 -1.9 869 40.9 2.61961 167 216 -49 -29.7 -5.3 730 41.4 1.31962 150 229 -79 -52.5 -9.0 997 43.7 5.71963 165 266 -101 -61.0 -9.2 1100 47.9 9.51964 245 289 -43 -17.7 -3.5 1236 53.0 10.61965 284 371 -87 -30.7 -5.9 1466 66.9 26.41966 231 273 -42 -18.2 -2.8 1519 75.9 13.31967 254 321 -67 -26.4 -4.5 1504 69.5 -9.41968 298 400 -102 -34.3 -6.0 1700 75.1 9.11969 332 395 -63 -19.1 -3.2 2001 80.4 7.11970 437 46B -31 -7.0 -1.4 2259 93.4 3.71971 451 524 -73 -16.2 -2.9 2501 91.2 9.3i972 419 543 -124 -29.7 -4.4 215 100.0 9.71973 391 549 -157 -40.2 -4.5 3501 117.9 17.91974 579 754 -175 -30.3 -3.8 4660 139.5 19.41975 910 1146 -337 -41.6 -6.4 5293 180.9 29.71976 870 1483 -613 -70.5 -9.4 6526 282.9 56.31977 1141 2137 -996 -97.3 -8.9 11163 611.7 116.31978 1392 3165 -1773 -127.3 -8.4 20996 1058.9 73.11979 2600 4296 -1696 -65.2 -6.0 28231 1635.1 54.41980 2950 4668 -1718 -58.2 -4.0 421353 2454.2 50.11991 3234 7719 -4495 -139.7 -6.2 72626 5313.4 116.51982 4903 9530 -4726 -99.4 -5.5 86151 6499.0 22.31993 10195 14755 -4570 -44.9 -2.5 194038 14482.0 122.81994 21728 26694 -4966 -22.9 -1.9 270561 20212.6 39.61985 38691 45763 -7072 -18.3 -1.9 372982 22310.9 10.4

Notes to Table 3:(a) 1950-1964 data on government revenue and expenditures are from O8G Statistical Yearbooks through 19691

1965-1995 data on governeent revenue and expenditures are from INF International Financial Statistics.(b) Source: Economic Analysis and Projections Departeent, borld Bank.(c) Coobined National Consueer Price Index from Table 3-5(1) Annex 3.

16

deficit in 1981 led ance more -to triple digit inflation. Since 1982, the

buiget has been progressively decreased, and, with the exception of the

severe drought year in 1983, inflatimn has been steadily reduced until it

was only 10.4 percent in 1985.

The Balance of Payments and Exchange Rate Disequilibrium

The inflatimnary situation in Ghana has had important consequences

for the balance of payments. As seen in Table 4, the current account has

been in deficit more years than in surplus, reflecting for the most part

inflows of capital and foreign aid. What has characterized the balance of

payments most, however, has been a persistent, and in most years growing,

overvaluatimn of the exchange rate resulting from import and exchange

controls.

This can be seen by caciparing the official exchange rate in Table

4, in either nominal or real terms, with any of several indicators of the

scarcity value of foreign exchange: the black market exchange rate, the

equilibrium exchange rate estimated using the purchasing power parity (PPP)

approach, and the equilibrium exchange rate estimated using a simulation

model based on econometrically derived demand and supply functicns for

foreign exchange. The methods used to estimate the equilibrium exchange

rates and to adjust nominal to real rates are discussed in detail in Annex

2. A brief summary of that discussion is presented in what follows.

The purchasing power parity approach to estimating the equilibrium

exchange rate perceives that rate as reflecting the ratio of the prices of

nontradable to the prices of tradable goods and services. Starting with a

situation of balance of payments equilibrium in the absence of major

17

Table 4

Current Account Dalance, Actual and Equilibriue Exchange Rates (ml

Nominal Exchange Rates (NCIIUSI Real Exchange Rates (MCK/US) (Ib

Current Actual Equilibrium Actual EquilibriusAccount -------------------- --------------------- ------------------- ---------------------

Balante (cl Official Black Model Official Black ModelYear (hillions SU9I (di Market (dl PPP (el If) 1g9 Market (1g PPP 11g lhi

19S8 41.2 .71 .71 .80 .88 .71 .71 .80 .891959 -19.4 .71 .71 .81 .99 .70 .70 .80 .901960 -94.9 .71 .71 .82 1.15 .70 .70 .80 1.131961 -135.1 .71 .71 .81 1.30 .70 .70 .80 1.201962 -62.7 .71 .71 .85 1.17 .67 .67 .90 1.111963 -107.9 .71 .56 .94 1.17 .61 .48 .80 1.001964 -74.9 .71 .59 1.02 1.30 .56 .46 .80 1.031965 -212.1 .71 .65 1.29 2.05 .44 .40 .90 1.271966 -117.5 .71 2.13 1.43- 1.58 .40 1.20 .80 .891967 -69.3 .84 1.67 1.26 1.24 .53 1.06 .J0 .791968 -42.9 1.02 1.75 1.37 1.51 .60 1.03 .80 .991969 -48.6 1.02 1.75 1.39 1.46 .59 1.01 .80 .941970 -66.0 1.02 1.64 1.36 1.57 .60 .97 .80 .921971 -146.1 1.03 1.75 1.42 1.55 .58 .99 .80 .871972 94.9 1.15 1.64 1.45 1.43 .64 .91 .80 .791973 114.2 1.15 1.49 1.47 1.35 .63 .81 .90 .741974 -285.6 1.15 1.73 1.42 1.47 .65 .98 .80 .831975 -26.9 1.IS 1.99 1.66 1.66 .56 .96 .80 .901976 -100.9 1.15 2.91 2.59 2.06 .36 .90 .90 .641977 -138.2 1.15 9.20 5.34 4.17 .17 1.38 .80 .631978 -104.2 1.51 0.76 7.76 6.29 .16 .93 .80 .651979 43.2 2.75 15.56 10.74 9.31 .21 1.16 .90 .621980 -50.4 2.75 15.87 14.95 13.71 .15 .86 .90 .741981 -503.7 2.75 26.25 30.93 36.45 .07 .69 .90 .951982 -191.1 2.75 61.67 39.74 42.41 .06 1.25 .90 .961983 -246.5 3.45 76.58 94.38 81.79 .03 .65 .90 .701994 -201.1 35.34 135.00 123.49 107.47 .23 .89 .80 .701985 -276.3 54.05 160.00 129.32 NIA .34 1.00 .80 N/A

Notes to Table 4:(a) All figures are taken or derived from data in Annex 2.(b) Real exchange rates are calculated from nominal rates by deflating by the ratio of

Ghanas nontradable CPI (inclusive of agriculturel to the Manufacturing Unit Value(MUVI index of exports by industrial earket economies to 4e turping countries,given in Table 2-1.

(ci Table 2-2.(dl Table 2-1.(el Table 2-5, column 1I1.(f1 Table 2-5, column (41.(g1 The nominal equilibrium exchange rate from Table 2-5, coluen (1), deflated as described

In note (bi of this table.(hl Nominal equilibrium exchange rate from Table 2-59 column (41, deflated as described

in note (bh of this table.

18

distortions in 1957-59, the equilibrium rate was estimated by multiplying

the actual rate by the ratio of Ghana's nantradable consumer price index

(CPI) to the manufacturing unit value (MIN) index of industrial ccountries'

exports to the developing natiuiss.9

The simulation model involves an import demand function, a cDcoa

supply function, and a noncocoa export supply function. Each of these was

estimated using annual data for at least twenty years. For the import

demanid function, the domestic price of importables relative to the price of

nontradables was estimated as a function of the level of imports determined

exogenously by government, GDP deflated by the price of nontradables, and an

index of food production. Cocoa supply was estimated as a functim of the

current official producer price of cocoa, the previous year's market price

for maize, the quantity produced of cocoa in the previcnis year, and the

"normal" level of prnduction given the existing stock of cocoa trees. The

last variable was estimated using the "vintage-matrix" model described in

Prnex 4, which takes into accaunt the number of cocoa trees of different

ages, their yields over time, and the rate of new planting, which is partly

a function of price. Finally, a supply function for an index of timber,

gold, and other exports was estimated using as independent variables the

domestic prices of these exportables and GDP, each deflated by the price of

nontradables.

In Annex 2, calculaticns were also made using a weighted average ofthe whDIlesale price index of Ghana's principal trading partners. Theresults were essentially the same as with the MIV index, which is morecomprehensive given the multiplicity of these partners. The MNV indexleaves out petroleum imports, which accounted for 10 to 15 percent of thetotal value of imports in most years since 1972. The greater rise inpetroleum prices is partially compensated, however, by a decline in therelative prices of primary foods, which are also omitted from the MUV index.

19

It was assumned in the model that Ghana would have exploited its

mmonpoly power in the cocoa market by applying an optimal export tax to

equate marginal revenue and marginal cost. This tax rate was related to

the world's elasticity of demand for cocoa, the lamg run supply elasticity

of Ghana s competitors, and Ghana's share of the wDrld market.L° The first

twD parameters were based an independent estimates and were assumed to be

canstant aver time. Ghana s market share was allowed to vary with its level

of cocoa expDrts.

The demand and supply funrctimns were then incorporated into the

model, and the exchange rate was calculated that equates the quantities

demanded and supplied of foreign exchange, assuming no capital flows for

reasmns discussed in Arinex 2. The free trade equilibrium rate was

determined by remraving all distorticns, except for the optimal tax an cocoa

expDrts, a,nd by allowing domestic prices to equal world prices times the

exchange rate.

It is obvious from Table 4 that Ghana's exchange rate has been

substantially avervalued for most of the period studied. This was first

evident at the end of the Nkrumah era in 1965 and 1966. With the

devaluation of the cedi and the partial liberalizatim that followed, the

degree of overvaluatimn decreased somewhat for a few years. All indicators

suggest, however, that it resumed its upward course by 1975 or 1976. In

1983, the equilibrium exchange rate was at least 20 times the official

rate. Oily in 1964 did the degree of overvaluation decline.

10 The long run elasticity was used rather than the shDrt runelasticity because it is assumed that a raticnal geverTiment would want toavoid undercutting its longer term position in the world market to gainshort-term profits.

20

Thus the disequilibrium that occurred in the balance of payments

was expressed not primarily in terms of a current acccunt deficit but by

the degree of exchange rate overvaluatimn in the face of import and exchange

restrictics. The price distortions introcduced had profound implications

for the allocatim of resources that were highly detrimental to economic

growth. Furthermore, the replarement of a market rate of exchange with the

controls necessary to maintain the official exchange rate meant that

substantial quantities of resources were allocated by administrative

decisions rather than by market price signals. This contributed to the

corruption of the public sector and created opportuinities for rent-seeking

behavior that wasted resources and discouraged political action aimed at

seeking policy change.

The Aciricultural Sector

Importance of APriculture in the Ecoraw

Agriculture is the most important sector of the Ghanaian ecocWXmy.

Over ne half of the available wDrk force is engaged in agricultural

production, and arcud 50 percent of total GOP is contributed by the

agricultural sector in a good crop year. Agricultural prckucts make up only

about one-fifth of the total value of imports, but the sector accounts for

approximately two thirds of total export earnings.

As Table 5 illustrates, the share of agriculture in real GDP

fluctuated scmiewhat from year to year but on average remained arLund 50

percent from 1955 through 1983. The share of agricultural workers in the

total labor force declined, however, from 67 percent in 1955 to about 53

percent in 1980. Preliminary results from the 1984 centsus also indicate

21

Table 5

Agricultural Sector (a)

Share AgriShare Agri Share Agri Exports in

Share Agri Share Agri Share Agri Imports Exports Total NoeIn NoM in Real In Labor Agri in Total Agri in Total Value Agri

BP (bl GDP (b) Force (c) leports (d) laports (d) Exports Id) Exports (dl Prod le)Year (1) ) 2 (Million NC) 1) (Hillion NC) 1) 12)

1955 57 50 67 36 20 134 70 381956 55 52 66 37 21 107 62 311957 52 51 66 43 22 104 57 301958 55 51 65 38 22 129 61 331959 56 53 65 47 21 144 63 331960 41 N/A 64 50 19 140 60 391961 35 N/A 63 62 22 146 64 441962 38 N/A 63 51 22 147 64 391963 36 N/A 62 42 16 147 68 371964 35 N/A 62 46 19 149 65 341963 41 47 61 43 13 151 67 251966 43 49 60 46 18 121 63 181967 40 50 60 52 20 161 66 271968 42 50 59 65 21 220 65 311969 46 50 59 66 19 191 57 211970 47 51 58 95 23 338 72 321971 44 50 59 82 19 230 64 211972 47 54 57 85 22 336 60 261973 49 51 57 142 27 407 56 241974 51 52 56 186 20 548 65 231975 48 48 56 122 13 645 70 261976 51 49 55 166 17 608 64 191977 56 45 55 149 12 817 70 131979 61 49 54 225 13 1127 71 91979 60 52 54 229 10 2056 75 121990 58 53 53 313 10 2030 64 81981 53 54 N/A 307 9 1196 44 31982 57 55 N/A 303 11 N/A N/A N/A1983 60 53 N/A N/A N/A N/A N/A N/A1994 49 54 N/A N/A N/A N/A N/A N/A1985 41 52 N/A N/A N/A N/A N/A N/A

Notes to Table 5:1a) Includes forestry and fishing.(bl 1955-1959 data from D. Walters, The National Accounts of Ghana, U.N. Report ITA0/GHA/t,

February 4, 1965. 1960-1978 data based on World Bank, World Tables, various issues.1979-1995 data froe Norld Dank, Ghana: Policies and Issues of StructuralAdjustment, March 30, 1987.

Ic) Data for 1960, 1965, 1970, and 1975-1981 from World Dank, World Tables, 1993, Vol. 2;estimates for all other data interpolated or extrapolated from these years usinglinear trends.

Cd) Ghana, External Trade Statistics, various years. Data for agricultural imports and exportsfor 197S, 1979-e1 frog World Dank, Ghana: Towards Structural Adjustment, Volume 11.Data for total exports for 1972, 1975, and 1979-91; and data for total imports for 1975and 1979-81 froa World Dank, Ghana: Towards Structural Adjustment, Volume 11.

Ce) Agricultural exports divided by the vale of agricultural sector 6DP obtained fromsources listed in Note (b), except for the years 1975 and 1979-91, which come fromWorld Dank, Ghanai Towards Structural Adjustment, Volume II.

22

that the Greater Accra Region grew more rapidly than the rest of the

country from 1970 to 1964,11 though the figures presented earlier in Table 1

suggest that the pace of urbanization has at least slackened, and may have

even reversed itself during the early 19EKs.

Agricultural imports consist principally of cereals, meat, fish,

sugar, oils, and fats. These imports maintained a fairly constant share of

the total value of imports until 1975, when their share began to decline.

Agricultural exports, consisting overwhelmingly of cocoa (96 percent of

the total in 1960), have fluctuated as a proportion of total expDrts because

of variations in local supply ccnditions (principally rainfall) and world

market prices. 2 There has been no consistent trend. Agricultural exports

as a share of total agricultural production, on the other hand, fell sharply

from an average of 36 percent in 1955-64 to less than 10 percent by 1978.

Productimn of Specific Crops

iMre detailed time series data on production of specific crops are

shown in Tables 6(1) thrcugh 6(3). The tables are drawn from Annex 1, which

discusses in detail the sairces of the data and assesses their reliability.

Two major issues energe from this discussion. The first is the extent to

tL' Wbrld Bank, Ghana: Towards Structural Adjustment, October 7, 1965,Vol. 2, p-3 .

M Ghana is a sufficiently large exporter of cocoa to the world marketthat these two influences are not independent. Uhtil the 1970s, theshort-run elasticity of demand abroad for Ghanaian cocoa was estimated to besomewhat less (absolutely) than -1 (Leith, Foreian Trade ReFimes andEconomic Development: Ghana, p.44). Consequently, a shortfall inproduction, for example, wculd force world prices up to the point thatexport earnings would actually increase in comparison with years of averageyield. This situation has changed considerably in the last 15 years as aresult of the erosion that has taken place in Ghana s share of the world market.

23

Table 6(1)

Production Indices (a)(1972:100)

Cereals Staples

Year Maize Rice Sarghus Millet Cassava Yai Cacoyam Plantain

1950 N/A N/A NIA N/A N/A NIA NIA N/A1951 N/A N/A NIA N/A N/A N/A N/A N/A1952 N/A N/A N/A N/A N/A NIA N/A N/A1953 N/A N/A N/A N/A N/A N/A N/A NIA1954 N/A N/A 52 100 18 71 N/A NIA1955 42 33 52 100 18 71 N/A N/A1956 42 33 52 100 18 71 N/A N/A1957 42 33 59 114 29 71 NIA N/A1958 45 43 H/A N/A 38 N/A N/A N/A1959 N/A N/A N/A N/A N/A N/A N/A N/A1960 N/A 46 N/A N/A N/A N/A N/A N/A1961 N/A 43 66 81 N/A N/A N/A N/A1962 42 44 69 66 28 147 N/A N/A1963 45 47 72 69 42 162 N/A N/A1964 42 60. 74 71 43 170 N/A N/A1965 51 46 58 58 39 155 N/A N/A1966 88 41 70 67 41 172 N/A N/A1967 69 60 54 79 41 1B0 N/A N/A1968 63 60 47 58 37 204 N/A N/A1969 75 86 63 91 48 295 N/A 421970 120 70 122 143 84 134 120 981971 116 78 113 132 84 134 120 981972 100 100 100 100 100 100 100 1001973 106 88 109 110 101 89 140 1241974 121 104 116 157 127 125 160 1211975 85 101 121 124 84 104 116 751976 71 100 124 146 64 85 82 751977 69 155 119 127 64 79 76 561978 54 154 79 99 67 80 77 561979 94 133 104 151 62 89 79 491990 95 111 87 83 82 96 69 441981 94 138 86 121 73 87 67 501982 86 51 56 77 87 87 66 451983 43 57 37 41 61 128 76 201984 143 94 N/A N/A 144 N/A N/A N/A1985 102 .128 41 55 108 71 61 40

Notes to Table 6(1):(a) Calculated from data found in Annex 1, Table 1-2(1).

Table 6(2)

Production Indices (a)(1972:100)

Edible Oil Nuts & Seeds

ground- Coco- DilpalsYear nuts nuts bunches Cowpeas Sugar Cane1950 N/A N/A N/A N__ ----------

1950 N/A N/A N/A N/A N/A1951 N/A N/A N/A N/A N/A1952 N/A N/A N/A N/A N/A1953 N/A N/A N/A N/A N/A1954 NA N/A N/A N/A N/A1955 49 N/A N/A N/A N/A1956 49 N/A NR/A N/A N/A1957 46 N/A N/A N/A N/A1959 46 N/A N/A NIA NIA1959 N5 N/A N/A N/A N/A1960 N/A N/A N/A N/A N4A1961 NIA NIA NIA NIA 491962 N/A NIA N/A N/A N/A1963 53 N/A N/A N/A 2701964 56 N/A N/A NIA 2761965 30 NIA N/A N/A 691966 43 NIA N/A N/A 1011967 44 NIA N/A N/A 2031968 69 N/A N/A N/A 1651969 68 N/A N/A NIA 2061970 114 102 99 116 771971 114 102 98 116 771972 100 100 100 100 1001973 142 103 110 81 1111974 175 104 129 119 1181975 124 105 127 119 1411976 126 77 39 133 1311977 91 51 104 121 1781978 93 54 131 B8 1871979 120 54 142 99 1311980 103 N/A N/A N/A 1321981 112 N/A N/A N/A 1311982 123 N/A N/A N/A 761993 79 N/A N/A N/A 691934 N/A N/A N/A N/A N/A1965 N/A N/A N/A N/A N/A

Notes to Table 6(2):(a) Calculated froe data found in Annex 1, Table 1-2(2).

25

Table 6131

Production Indices (a)(1972100)

SeedYear Cocoa lb) Cocoa (c) Cotton Rubber Tobacco

1950 64 61 N/A N/A N/A1951 52 49 N/A N/A N/A1952 61 58 N/A N/A N/A1953 52 49 N/A N/A 61954 54 52 N/A 25 &1955 56 54 NIA 25 61956 65 62 N/A 25 61957 51 48 N/A 25 61958 63 60 N/A 33 181959 78 74 N/A 50 N/A1960 106 101 N/A 50 N/A1961 100 96 N/A 67 NIA1962 101 97 N/A 25 411963 105 100 N/A 25 471964 132 126 N/A 25 531965 99 94 N/A 11 591966 90 87 NIA N/A 591967 102 99 N/A N/A 711968 79 78 N/A N/A 651969 99 98 NIA N/A 651970 101 102 15 5t 531971 112 112 31 83 821972 100 100 100 100 1001973 84 84 146 108 1531974 92 93 192 150 1941975 97 97 231 175 1351976 79 79 692 225 1241977 67 68 877 225 471978 65 67 346 275 471979 73 76 400 275 471990 63 67 231 142 241991 55 58 54 56 291982 44 46 38 59 351983 39 41 38 42 291994 42 44 N/A N/A N/A1985 53 53 N/A NIA N/A

Notes to Table 613):(a) Calculated from data found in Annex 1, Table 1-2(3).(b) Based on official earketing data.(c) Corrected for estimates of sauggling, as

described in Annex 1.

26

which figures on cocoa production in Ghana are underestimated because they

are based an official marketings and do not include the effects of

snuggling. Analysis of alternative estimates of cocoa smuggling suggest

that it may in recent years have amounted to as much as 20 percent of

producktian. This is probably an overestimate, hcwever, since it does not

adequately take into account the deterioration of the transportation system

or the efforts of the government to suppress smuggling. Consequently, the

Cocoa Marketing Board data on official purchases were adjusted more modestly

to show same increase in smuggling during the late 1960s, and again during

the late 1970s, to obtain a maximum of 10 percent of official marketings by

1990. L

The second issue relates to the quality of the production

statistics after the early 1970s. With the general breakdown in the

transport system and the deteriorating situation regarding the governrent

budget, it became increasingly difficult to gather accurate data on area

cultivated, yields, and production through agricultural censuses and sample

surveys. Informed judgements concerning movements in these variables

therefore played a greater role in the collection of data, with all the

possibilities that this implies for error and bias. Nevertheless, the

trends in production that occurred were so significant and were corroborated

by other information to such an extent that the orders of magnitude, at

least, appear to be correct. In addition, efforts to estimate supply

functions for maize and rice, which are described in Chapter 7 of this

report, were quite successful despite the poor quality of data.

-' The alternative indices of cocoa production in Table 6(3) are eachbased on 1972--100, when it is estimated (Annex 1) that actual productionexceeded reported production by about 5 percent.

7

Perhaps the most significant trends shown in Tables 6(1)-6(3) are

a general increase in food procduction until the early 1970s, followed by a

steep decrline during the ensuing years. The only exceptions to this are

rice and sugar cane, both crops that are irrigated. As examples of this

trend, between 1972-74 and 1981-B3, the following production decreases were

recorded: maize 33 percent, sorghum 45 percent, cassava 33 percent, and

plantain bananas 67 percent. Production of yams did less poorly, declining

by only 4 percent.

Production of rice and sugar reached a peak in 1977 and 1978, with

production falling off thereafter. For sugar, this peak was well below the

levels of production that had been attained in the 1960s. Rice production,

on the other hand, grew fairly steadily from the mid-1950s until the late

1970s. Thereafter it experienced a sharp, though somewhat erratic, decline.

Production of the cash crops - seed cotton, rubber, and

tobacco - tells a similar story. Cotton and rubber production peaked

about 1977, tobacco a few years earlier. Each of these crops suffered a

precipitous decline in output during the late 1970s and early 1900s.

Ohe reasnm for the decrease in production of tobacco and food

crops that occurred after 1974 was severe drought from 1975 to 1979, and

then again in 1982 and 1983. The last two years of drought also had a

significant effect on output of rice and sugar because of lack of water for

irrigatimn. Cbtton prcduction appears to have been less affected during the

first period of drought, perhaps because a major program was being

introduced to encourage its cultivation in the north. With improved weather

in 1984 and 1985, production of most crops revived somewhat, though it is

still unclear whether this revival will be sustainRed.

2E3

By the middle to late 1970s, crop production wa5 also influenced

by the disintegration of the system of transportatimn. Shortages of fuel,

tires, and spare parts, plus the deterioration of the road network, raised

transport costs and dramatically decreased the availability of vehicles. In

addition, the decline in the real value of producer prices, discussed later

in this report, contributed to the lack of incentives for cash crop

production.

With cocoa, by far the most important cash crop, the origins of

decline go further back in view of the long gestation period and productive

life of cocoa trees. L After having reached a level of production in excess

of 400,000 tons in 1960, and despite some fairly substantial fluctuations

due chiefly to weather, Ghana maintained output near this level for the next

12 years. Starting in 1973, however, cocoa production began a steep decline

that resulted in a level of production in 1982-64 that was less than half

that achieved earlier. Given the fact that yields of cocoa trees, with

proper maintenance, do not seriously start to decline until at least 25

years after they are planted, much of the decrease after 1972 appears to

have been due to a failure after the early 1950s to replant at a rate

sufficient to maintain prcduction. In addition, declining producer

incentives and the deterioration of the transport system led to decreased

tree maintenance, infestation by insects and disease, lower yields, and

reduced harvests. These issues are explored further in Chapter VII and in

Pnnex 4.

1 Rubber has an even longer gestation period and productive life, butthe dramatic decrease in production after 1979 appears to have bEen due moreto an inability to cover variable costs than to a decrease in the capitalstock invested in trees.

29

Food Constmntian

Indices of total and per capita food production and consumption,

from FAD Producction Yearbooks and Food Balance Sheets, are presented in

Table 7, alcng with data on the average number of calories consunmed per day

per capita. Although incomplete, the data suggest that per capita food

availability, which grew steadily during the 19605 and early 1970s, later

declined to alarming levels. In part this was because of the failure of

food production after 1972 to keep up with population growth. It was also

because the Ghanaian economy during this period was unable to increase food

imports to fill this gap.

Per capita consumption of 1769 calories per day in 1980 must be

considered very low, even if food was spread uniformly across the

population. When one considers differential access to food resulting from

variations in family income and from intrahousehold distribution, the

evidence points strcngly to a problem of severe malnutrition. ND direct

data are available from household consumption and nutrition surveys,

(mweyer, to verify this. There is, in addition, the possibility that the

data on agricultural production are sufficiently imprecise that they do not

serve as a reliable guide to the extent of the nutritional problem that

existed at this time. The food availability situation appears to have

improved after 1983, with the return of good harvests, but the problem of

lack of direct household consumption and nutrition data remains.

30

Table 7

Food Production and Consumption Indices(19722100)

Agricultural Food Agricultural FoodProduction (a) Consumption (b)

Year Total Per Capita Total Per Capita Cal/Day

1961 68 89 N/A N/A N/A1962 68 87 73 93 20151963 71 89 N/A N/A N/A1964 86 105 77 94 20331965 73 87 79 94 20341966 74 86 83 96 20991967 83 94 87 99 21391968 77 86 91 101 21771969 82 89 95 103 22241970 91 97 98 103 22401971 101 104 101 104 22601972 100 100 100 100 21661973 101 99 106 103 22411974 104 99 111 105 22721975 104 96 104 96 20721976 95 85 103 92 19861977 89 77 106 92 19961978 88 74 N/A N/A N/A1979 93 76 N/A N/A N/A1980 90 72 102 82 17691981 91 71 N/A N/A N/A1982 98 69 N/A N/A N/A1983 85 64 NIA N/A N/A1984 100 75 N/A N/A N/A

Notes to Table 7i(a) Source for Food Production data is FAO Production Yearbooks.

For source for popualtion data used to calculate per capitafigures sit Table 1.

(b) Source for Food Consueption data is FAD Food Balance Sheets.For source for popualtian data used to calculate per capitafigures see Table 1.1962 Food Consumption figures represent an average of 1961-1963 totals;19M Food Consuaption figures represent an average of 1979-1981 totals.

31

CHWTER III: TIE EVOLVING PCLITICPL EQMNY-THE RPAYERS

Phy analysis of the political econcmy of price policy in Ghana

must begin by ccotsidering the varicus players involved in its determination.

At one level, these can be described as prcxucers and consumers of different

agricultural products and the government that receives revenue frnm and

contributes resources to the farm sector. The impact of price policy an

these groups is estimated in Part II. This is insufficient to suggest,

however, why certain policies have existed and what has caused them to be

altered over time.

The history of Ghana during the past three decades has been

characterized by frequent changes of govermment and, until recently, by a

generally deteriorating economy. The reasons for this are deeply imbedded

in the sociopolitical fabric of Ghanaian scniety, and any explanatim of

policy change must take into account the role of interest grcups not only as

producers and consumers but also as political actors. It must also

investigate the role of the state as well as that of the brokers, who have

acted as political intermediaries.

Interest Groups

The best articulated interest groups are based cn residence and

occupation. The major residential distincticn is urban and rural. Among

the occupational groups in the cities, sone of the most politically vocal

are teachers and professicnals organized into variaus formal groups such as

the Ghana Associatian of Llniversity Lecturers and the Ghana Bar Associatim.

In the ccuntryside, there are large farmers, who frequently are

absentee and employ sharecroppers or wage laborers to maintain and harvest

32

cocoa trees, to extend and replant cocoa fanrs, and to cultivate food

crops. There are also smaller farmers who live on their farms and under-take

these activities themselves, though they may employ snme outside laborers or

work off their own farms during part of the year. A third group is ccarosed

of sharecroppers who are engaged by larger farmers as caretakers to tend

their cocoa farms once they are planted and to undertake some extension and

replanting. Finally, there are wage laborers hired by the day or task to

plant cocoa trees, harvest cocoa beans, weed and harvest food crops, and

undertake other agricultural tasks.

During the Nkrumah period, the United Ghana Farmers Cboperative

Counxvcil (UGFCC) was established to help organize farmers. Althcigh it was

supposed to be a nationwide organization, its activities were restricted to

the cocoa growing areas. Mbre importantly, it was an arm of the Convention

People's Party (C1P), that was designed to extend the authority of the state

into rural areas and was highly centralized and bureaucratic in its

structure. Although it co-opted a few larger farmers into its

organization, local agents were not elected and it only very imperfectly

represented farmers' interests.L The LEFOC was banned after the fall of

Nkrumah in 1966. More recently the Ghana Federation of Agricultural

Cooperatives has represented Ghanaian farmers in various discussions, such

as those involving the determination of the cocoa producer price, but,

though agricultural cooperatives achieved some importance during the

colonial period, their power was severely undermined by the LUGFCC and their

± Bjoro Eeckman, Organizing the Farmers: Cocoa Politics and NationalDevelopment in Ghana, Uppsala: Scandinavian Institute of African Studies,1976.

33

fnemtership has never included more than a small fractimn of all Ghanaian

farmers.

Aside from farmers, the most important occupational groups in

rural areas consist of traders, transporters, and shopkeepers. Sone of the

more important of these are also large farmers, who are relieved of the need

to remain on their farms by the caretakers they engage. In addition, there

are also vestiges of the urban elite, including civil servants and a few

professicnals.

During the late 1960s and the 1970s, a radical transformatim

occurred in the composition of groups within rural areas. Low cocoa prices

resulted in a marked decline in the number of wage laborers migrating into

the cocoa produring regicns from other areas, especially the north. These

workers instead emigrated to the Ivory Coast with its booming cncoa

industry, stayed in the north to work m ccumercial rice farms, or simply

remained an their subsistence food farms. Second, significant numtiers of

professionals, former politicians and civil servants, workers, and others

frustrated with ecmnomic depressim and low wages in the formal urban

ecanomy renewed their ties with or physically moved back to the

countryside. In the north, they purchased large tracts of uncultivated land

and began to grow commercial food crops, especially rice. Elsewhere, this

rural bourgeoisie invested in a variety of productive activities outside the

purview of the state.

Cutting across the urban and rural groups described here is the

binding tie of ethnicity and regimal affiliatim. Of particular

importance, in this respect, are the Ashanti in the central and west central

part of the country, the other Akan groups to the south of the Ashanti, the

34

Ewe of the vlulta Regian bordering Togo, and a number of other numerically

less important groups along the coast and in the north. Regicnal

identification is also especially strong in the Northern and Upper Regions,

which have lagged ecnonmically behind other areas of the country, and in the

Volta Region, where the Ewe span both sides of the border and are cut off

from the rest of Ghana by the Vol ta Lake. The Brong-Ahafo and Ashanti

Regions, with Kumzasi the Ashanti capital, have for years dominated the cocoa

industry and prcrvided the main resistance to the erosion of its wealth.

Role of the State

The role of the state in Ghana was greatly increased during the

early Nkrumah years as the CRP, with its socialist ideology and its

political base among the new elite of young and educkated, scught to increase

its power in every area of political and economic activity. With the

bankruptcy of this regime brcught on by its continuing efforts to mobilize

resources for its investment program in the face of depleted reserves and

the collapse of the cocoa economy, however, the government lost the

confidence and support of broad segments of the population. Efforts were

made following Nkrumah s overthrow to bolster its legitimacy by the National

Liberation Council, and particularly by the Ehsia governrent, but continued

economic difficulties and a growing lack of confidence resulted in a

narrowing of its constituency and increased centralization of

decisionm-aking in the hands of the Prime Minister.

This tred continued under the military government of Acheampong

and the National Redemption CcLricil (NRC), especially after the October 1975

purge and the creation of the Supreme Military Council (9MC), which became

35

the paramonmt decisicar-making body. By 1976, the NRC/613 had bercame highly

authoritarian and was cut off to a large extent not only from the mass of

the population but also frnm the urban elite of professicnals and civil

servants. Important segments of the populatim simply withdrew from areas

of state influence. The state, in turn, became increasingly coercive as its

power base dwindled.

AcheampoM was forced by the military to abdicate the chairmanship

of the SMC in July 1976, and General Akuffo was installed in his place.

Only minimal changes were introdurced, hnwever, in the centralized apparatus

of decisian-making. After a brief interlude of rule by the Armed Forces

Fevolutionary Counicil (AFRC), Hilla Linann assumed the presidency, but

overwhelming ecornomic difficulties and the lack of a strong political base

som led to a resumption of the centralized, personalized style of

decision-making which alienated practically the entire Ghanaian popilation.

One of the main characteristics of the state in Ghana has been the

institutionalizatimn since independence of judicial and administrative

structures inherited from the colonial regime at the expense of

representative mechanisms that would have allowed for more widespread

political participation. A major reasmn for this has been the monopoly that

these structures have held over the mobilization and distribution of scarce

resources. Furthermore, to the extent that representative structures have

existed, they have beei elitist in nature and have not had clear-cut links

to major portians of the population. Uhder Nkrumah, "the state grew at an

extremely high rate, but without developing an organized base capable of

supporting, financially and politically, its maintenance and

36

reproduction".- Sucrcessive goverrments cantinud "the separation of the

state apparatus frnm representational control that their predecessors had

set in motion."~

The administrative institutions of the state, in the meantime,

became increasingly amenable to social pressures in their control of scarce

resources. Continuation in office depended an the bureaucracy's ability to

placate powerful groups, and, in the absence of political representation for

broad segments of the population, this resulted not in policies to stimulate

growth and development but in increasing corruption and favoritism in the

distribution of the resources at the disposal of the state. This process

was exacerbated as the magnitude of those resources decreased. Eventually,

the capacity of the state to maintain power through the distribution of

resources was so eroded that it had to resort increasingly to the use of

force in place of administrative control.

As the power and influence of the state decreased, that of

alternative political institutions was strengthened. This occurred

particularly in local communities, which were well defined in terms of kin

groups, resident alien migrants, traders, and civil servants. Local

institutions were not dependent upon the distribution of rescurces by the

state at the national level but rather relied on crops grcwn on land held by

local chiefs, kin groups, and landbwiiers. This production was, in turn,

traded for the output of local industries, commercial concerns, and service

2 Beckman, Orcianizinm the Farmers..., p.239

Naomi Chazan, in Anatawv of Ghanaian Politics: Managing PoliticalRecession. 1969 - 1962, Hbulder, Colorado: Westview, 1983, p.24. Mi-ch ofthe discussion that follows is based on this work, which brings together theresults of much of the political research on Ghana during this period.

37

establisihments, through both market and collective forms of exchange. Local

institutions differ-ed froun the state apparatus most markedly with respect to

their representation, consensus, and legitimacy. They were primarily

political and not administrative in nature and were rooted in a common set

of values and guiding codes. Accountability was a key feature of leadership

status.

Patrons

As the power of the 5tate to guarantee physical security, status,

and wealth weakened, local patronage networks grew in importance. With

direct access to the state and its resources increasingly blocked, personal

ties with "big-men" assumed greater importance as a means of gaining entry

into the state orbit. Patrons whox took on the role of broker or

intermediary were at varicus times paramount chiefs, ethic leaders,

religious men, professionals, businessmen, former office holders, military

officers, union organizers, and others with access to decisicn-makers. They

played a key role in linking the top echelons of the state with specific

local constituencies.

Patron leaders arose to assist in the allocation of goods and

services when other channels for their distribution became increasingly

closed. In return, they were rewarded by deference, status aggrandizement,

gifts, favors, and outright bribery. Patron - client relations remained

voluntary, however, and patrons were expected to perform and to avoid

overstepping their boundaries. As the state became increasingly unable to

deliver rescurces through the patronage network, the patrons broke off their

relations with it and joined forces with local community leaders.

38

CHAPTER IV: THE EVOLVINE POLITICAL EOIX74W - THE PERIOD6

This section analyzes the periods thrcugh which the political

ecKormy of Ghana evolved fram 1950 to 1985. It contends that the policies

and instituticns created during the early years of independence not only

were directly inimical to growth and development but also led to the

creation of allocative and distribution mechanisms that inhibited the

exertion of influence by major segments of the populatimn on policy-making

in ways that wculd have contributed directly to their income and welfare.

Furthermore, as successive governments fcunud their resource base dwindling,

their dependence m patrmage networks and their inability to draw upm the

support of important sociopolitical groups resulted in an inability to

undertake the major reforms that would have been necKessary to strengthen the

ec-nomy and to set the stage for long run development.

Liberal Regime. 1950-60

A liberal ecKnomic regime characterized Ghana during its period of

decolonization from 1950 to 1960. The econroy at this time was centered an

the cocoa industry, which was the most important scurce of govermment

revenue and foreign exchange. High prices mn the world market, following a

decline in production capacity during the Depression and World War II,

provided a windfall of resaurces available to the Ghanaian economy.

The Nkrumah regime was strongly committed to enhancing the power

of the state and to mobilizing the resources necessary for its investment

program. To a very large extent, this meant capturing revenue from cocoa.

The regime was aided by the creation of the Cocoa Marketing Ebard (CMH)

under the colonial government during Warld War II, partly in response to

39

pressures exerted by cocoa farmers concerned over the market-sharing and

price-fixing arrangements of foreign firms buying cocoa in Ghana.9- The C(H

became the nmnoipoly buyer of cocoa at a fixed price paid to prnducers, even

though the foreign firms and a few local traders actually purchased and

marketed the cocoa as Licensed Buying Agents (LBA) of the CMP in return for

a fixed allowance per ton. The Board was not supposed to make a profit at

the expense of the producers, though surpluses were to be set aside for

purposes of price stabilization and for activities that wauld benefit the

industry such as research, disease control, credit prugrams, and

cooperatives.

The foreign firms, which were interested principally in

maintaining a steady supply of cocoa, agreed to this policy even though it

restricted their freedom of operation. With the sweeping victory of the CFP

in the election of 1951, the transitional governnient led by Nkrumah brought

the EBard increasingly under its control. The competition for cocoa

revenue was no longer between the foreign firms and the farmers, but

between the latter and the government run by the CRP. Che implication of

this was an increasing share of cocoa revenue appropriated directly by the

government. Prior to 1951, the bulk of cocoa profits had been absorbed by

the reserves of the Ebard. In 1950-51, tax rates were revised and the

government collected one-fifth of cocoa export earnings as duties. This

trend ccntinued under the CPP transition govemnment as cocoa profits were

increasingly diverted from the cocoa sector to general public investment

through a steeply graduated export tax.

a Beckman, OrnanizinQ the Farmers..., p.41. Mich of this section isdrawn from Beckman's work.

40

Duiring the next few years, the CPP struggled to maintain its place

as the party leading Ghana to independence. Che of its major political

opponents was the Naticnal Liberatimn Pbvement, established in Ashanti with

strucg support from cocoa farmers. This organization was never very

successful because of its failure to extend its regional and separatist

appeal to a naticnal constituency. Nevertheless, it served as an important

focal point for farmer discontent over the large share of cocoa revenues

being diverted from producers to the governnent and the compulsory

cutting-out of cocoa trees to control swollen shoot disease.

In an effort to extend its influence in the rural sector, the

t&zumah negime in 1953 founded the Uhited hanw Farnmer s CLoperative Cbunicil

(L1F0C). Although ostensibly supposed to cover the entire country, the

Cbunmicil s activities were confined almost entirely to the cocoa growing

areas. It was financed almost entirely from the commercial operations of

the Cocoa Purchasing Company (CPC), established by the CQB in 1952 to

purchase cocoa in competition with other licensed agents. The CPC had a

decided advantage over the LBA's, however, in that it was also responsible

for the distribution of interest-free advances and loans financed by the

Board to assist planters in redeeming farms pledged to money lenders. Many

of these loans were never repaid. In 1957, the CPC was liquidated as a

result of large-scale financial irregularities, and the UEFCC tmok over its

funicticns. Unlike the Ccmoa Purchasing Company, the L[FOC was not a

subsidiary of the E1ard and operated as any other LBA, except that it was

financed with public capital.

By 1957, the CFP had established its political preeminence in

three elections, and "state powers could now be more directly applied to the

41

largely unsolved problem of establishing an organized political base among

the agricultural producers.`"2 The 1GFCI was declared the only officially

recognized farmer organization in the cauntry. By 1959-'0, it was

purchasing 17 percent of the cocoa crop. Within two years it was granted a

mmnopoly am all purchases of cocoa from farmers within Ghana. The

extensive marketing network of private agents, traders, brokers, and other

middlemen was replaced by an urban-based parastatal organizatian of clerks

and bureaucrats. Since these private middlemen had been drawn principally

from the upper stratum of the farming cammunity, and were "...those most

likely to turn farmer organizaticns against the central government and its

heavy appropriation of cocoa incme`"3, the process of extending CPP control

into the countryside was complete.

The capture of the windfall from high cocoa prices had important

fiscal implicaticris. Government expenditures grew dramatically during these

years. In real terms, total cnnsolidated public expenditures increased by

almost six times over the decade.4 As a proportimn of GDP, government

expenditures rose from 7 tD 18 percent over the same period.' At the same

time, the share of extraordinary and development expenditures in the total

increased from 27 to 36 percent.- Mrbst of these expenditures were directed

2 Beckman, Orhanizina the Farmers*..., p. 7 2 .

B Heckman, OrCcanizinc the Far.ers...., p.107.

Stephen Htymer, "The Pblitical Ecanomy of the Gold Coast and Ghana,"in Gustav Ranis, ed., Government and Economic Development, New Haven: YaleUniversity Press, 1971, p.173. Coisolidated expenditures include those ofthe central government, Cocoa Marketing BDard, railways and harbors, andlocal government.

° See Table 3, Chapter II.

F tymer, "The Political Economy...," p.131.

42

tnwards improving the transport system, expanding health and educational

services, and providing agricultural research and extension.

Imposition of Controls, 1961-63

Cocoa prices began to slump seriously after 1957 as a result of

increases in world supply stimulated by the high prices of the post World

War II period.-- Despite substantial increases in producktion resulting from

new planting because of these earlier high prices, Ghana's export earnings

remained relatively constant. Since produiction and marketing costs

increased more or less proportionately with the expansion of output, profits

were squeezed. Public revenue from cocoa declined in nominal terms from W

67 millicn in 1957 to Nu 36 million in 1960 and NW 13 million in 196 5.a

The governnent, with its large-scale investment program in full

swing, was able to sustain its expenditures for a time by draw.ing down

reserves, but the financial base for its program was seriously eroded.

Government budget deficits accelerated sharply from 11.5 percent of total

government revenue in 1960 to 61 percent in 1963.7 As a ccrseqtence, the

government drew down heavily on CMB reserves and cut prices to prodLicers

from Nt 224/tan in 1961 to N! 167/ton in 1964.10 In this way it was able to

stabilize its cash flows for a time but at the cost of seriously depleting

the liquidity of the CMB.

' The FOB price of cocoa fell from a high of Nk 637/ton in 1954 to alow of W 262/ton in 1965. Annex 3, Table 3-4(3).

Table 34, Chapter VIII.

9 Table 3, Chapter II.

0 Table 10, Chapter VI.

43

To a ccnsiderable extent, the fall in cocoa prices was due to the

expansim of produictian in Ghana during the 1950s and early 1960s since

Ghana acccunted for about 40 percent of total world exports. Despite

Ministry of Finance fears cooncerning the wisdom of continuing to expand

output in the face of falling prices, however, the Ministry of Trade and the

Ministry of Agriculture supported measures to increase production, such as

the mass capsid spraying program of 1958 and the sustained efforts to

cmtrol swollen shoot disease.

In the meantime, public sector investment comtinued to expand as

resources were diverted from cocoa to general development. During the

1950s, emphasis had been placed m infrastructure and social services,

especially educatimn. Che consequence was an increase in requirements for

recurrent expenditures to maintain the roads and to staff the schools built

at this time. The CPP government was also strongly committed to

diversifying the economy and promoting industrialization. The Seccnd

Development plan, launched in 1959, envisaged a radical increase in public

development expenditures commnitted to this goal. Whereas the First and the

"Consolidated" Plans, covering the period 1951-59, had absorbed 118 million

(N 236 millim), the Secm d Plan called for 350 million to be spent in

five years."

Although private foreign investment was to finance a major portim

of this plan, public expenditures were also to be accelerated. By 1960-61,

government expenditures had been raised to a level that was twice that of

the mid-1950s.' Furthermore, when difficulties were experienced in finding

EB eckman, Organizing the Farmers..., p.201.

Table 3, Chapter II.

44

foreign private investment, the government remained deterTined to proceed

with its industrialization program. The problems in the wDrld cocoa market

mnly accentuated this desire because of the perceived need to diversify the

economy. "By 1961, much of the econc3mic philosophy which had marked the

'liberal' 1959 plan had been replaced by a professed belief in the

overriding importance of direct state participation in productimn and

comprehensive state planning."'l

Falling cocoa prices and a rising demand for imports by the

government, for its investment program, and by the private sector, because

of increased income coupled with a liberal trade regime, resulted in a sharp

increase in the current account deficit from $19.4 million in 1959 to $94.9

million in 1960 and $135.1 million in 1961.2o With foreign exch,ange

reserves declining sharply and its budget deficits rising rapidly, the

Nkrumah gcermnent in 1961 resorted to a series of strong measures. An

austerity budget was introduced and taxes were raised. Foreign exchange

controls were also extended, and conprehensive import licensing was

instituted later in the year as it became clear that high duties m imports

were insufficient to achieve a balanced current account. In additimn, there

was a significant reorientation of public investment away from the

infrastructure that had supported small-scale, export-oriented agriculture

during the 1950s and towards large-scale, state-owned agricultural and

industrial enterprises designed to substitute domestic productiom for

imports.

E- Eeckman, Orcanizinq the Farmers..., p.206.

§4 Table 4, Chapter I1.

45

The expansion of public sector agencies that was initiated during

this period resulted by March 1966 in 53 state enterprises, 12 joint

state/private enterprises, and 23 public boards.I5 These were involved in a

wide range of activities, such as brick and tile manufacturing, vegetable

oil milling, paper conversion, gold mining, and food marketing, to give just

a few examples. Although the state enterprises were supposed to be

profitable ventures that would contribute to public revenues, they were also

saddled with a number of political and social objectives, such as providing

jobs for party loyalists, reducing unemployment, and maintaining low

cmsumer prices. As a result of this and general managerial inefficiency,

out of 23 enterprises for which data were available on profits and losses in

1964-65, mnly 9 appeared to have been making any profits, and the net losses

of this sample of firms were in excess of YS 14 million despite the fact

that many operated as virtual monopolies in well protected markets.'6

ANnalysis of individual firms reveals, in fact, that profits were often

earned only because outputs were priced artificially high and inputs

artificially low in relation to world market prices valued in cedis at a

reasomable equilibrium rate of exchange.

The introduction of the 1961 budget and accompanying taxes and

quantitative restrictions was an important turning point in the fortuies of

the CPP goverrnent. Prior to this, Nkrumah had been supported by a fairly

broad segment of the population, especially in urban areas. The austerity

'° Killick, Dtevelopment Ecoanmics..., p.217.

6 Killick, Development Economics..., pp.219-21. Although there wassDme improvement in the profitability of these firms by the end of the1960s, partly because they had passed thrcugh their "infant" stage, netlosses were still close to Ni 10 million annually.

46

policies of 1961, however, alienated many grnups that had been close to the

party, including workers, junior civil servants, and small businessmen. In

addition, farmers were incensed by the continued decline in producer prices

in the face of rising costs of production and prices of consumer goods, as

well as by the imposed cocoa purchasing monopoly of the U1GFOC, which led to

cheating, extortimn, favoritism, misappropriatimn of funds, and other

abuses.'" This resentment was intensified in 1963, when the compulsory

savings scheme forced on cocoa farmers in 1961 was converted into an

explicit tax on cocoa, which was readily supported by the LEFCC as the

"representative of the farmers." The result was heavy criticism of the 1963

budget and of the UGFCC in the National Assembly by representatives fr-m the

cocoa growing areas, joined by backbenchers from the CRP.

Although import restrictions and exchange controls helped to

alleviate the deficit in the balance of payments, the decrease in imports

implied a decline in tax revenues from import duties, which in the

immediate pre-1961 period had replaced the tax on cocoa exports as the major

source of government revenue. This contributed further to the growth in

goverTwnEnt budget deficits, which increased from N% 49 million in 1961 to

N. 101 million in 1963.2- Despite its financial difficulties, however, the

government was determined to go ahead with its Seven-Year Development Plan

published in 1964.

' These are described in more detail in the next chapter.

@ Table 3, Chapter II.

47

Breakdcw of the System, 1964-66

With its foreign exchange reserves seriously depleted, the

governaent increased its borrowings abroad. By the time of the coup in

February 1966, the external debt totaled N 805.3 million, of which only 20

percent was in the form of long-term loans. The remainder was made up in

large part of suppliers' credits (57.9 percent), arrears of current payment

(10.6 percent), and bank loans (6.5 percent).'9 Aside frem the debt problem,

leakages in the trade licensing system resulted in imports not being

allocxated whkere they would have done the most good. Shortages of essential

intermediate inputs resulted in factory closings and underutilization of

productive capacity. Inadequate supplies at official prices of essential

foods contributed to urban unrest.

The development plan never had a chance. Despite a substantial

increase in tax revenue generated by the Nkrumah government in 1964 and

1965, rising government expenditures resulted in substantial deficits,

financed by an expansion in the money supply. World cocoa prices collapsed

in the second half of 1964 as it became apparent that West Africa had a

bumper crop. Farmers were paid a previously agreed price, how.ever, for the

largest crop in Ghana's history - 538,000 tons. After the purchasing and

marketing costs of the CQB and the UGFCC were covered, there was almost

nothing left over frem current income to be paid to the government and the

CM's liquid resources were exhausted. With the government printing money

to meet its expenses and with imports restricted and substantial purchasing

power in the hands of cocoa farmers, inflation increased to 35 percent

V9 Leith, Foreimn Trade Regimes...., p.2B.

48

annually between Orktober 1964 and July 1965. ° The external payments crises

became acute with short-term trade and suppliers credits falling due.

In the face of such pressure, the UGFOC agreed in July 1965 to

reduce the cocoa producer price by 26 percent to 40 shillings per headload,

its lowest level in years. It was also agreed that governnent subsidies an

insecticides and spraying machines would be eliminated and that farmers

would voluntarily reduce further cocoa planting. This agreement was

achieved, however, in meetings involving mnly the CMB, the Cabinet, the

Ministry of Finance, the Bank of Ghana, and other governmment bodies. The

sole voice of the farmers was through the LIFOC. --

In contrast to the situatimn in 1963, and because of the general

atmosphere of natimnal emergency, no protest was raised in the Natimnal

Assembly regarding this sharp reduction in the producer price despite clear

evidence that farmers understood that the setting aside of CM reserves was

supposed to cushimn them from such changes in world market prices.

'Instead, the debate developed into a general attack mn the Farmer's Council

as a monopoly buyer of Ghana's cocoa. Its vulnerable political basis in the

cocoa areas was highlighted by the collapse of market prices. It was

hardly a coincidence that it was the Minister of Finance who launrched the

belated attempt in early 1966 to reform the Farmer's Cbuxncil in a more

democratic and representative directimn. Eeing responsible for fiscal

policy, he had particular reason to be worried about the government's

B Beckman, Organizinci the Farmers ... , p. 21 6 .

Beckman, Orcianizinq the Farmers ... , pp.216-17.

49

political relations with the cocoa farmers and the fictitic.is nature of the

agreenents struck on their behalf by the Farmer' s Council.`2O

The impact of inflation, and especially the rise in food prices,

which increased ckring the first half of the 1960s almnst twice as fast as

those of other consumer goods,23 ccupled with the impossibility of raising

public sector wages because of the disastruLs budget situation, was the last

element in undermining political support for Nkrumah amcang the urban

population. His regime was overthrown by the military in February 1966 and

was replaced by the National Liberation Cbuncil.

Austerity. 1966-67

The officers who led the ccup had a mich clearer idea of what they

were against than what they were for. In particular, they were opposed to

loss of civil liberties, economic hardship, and widespread corruption. In

this they mirrored the bulk of the civilian populatin. They also resented

discriminatory treatment in favor of the President's Cwn Guard Regiment in

comparison with the regular army, which suffered a reduction in living

standards similar to the rest of the urban population. Abowe all, they

wanted "... to restore the political landscape which had existed at the time

of independence in 1957 and which they felt had been eroded..." by the

creation of a one-party state, concentration of power in the hands of an

2 Beckmnan, Organizing the Farmers ... , p.218.

2 Tables 13(1) and 13(2), Chapter VI.

50

executive president, erosion of the influence of the chieftaincy, and the

program of socialist development. 2 4

ohe consequence of this lack of clear focus was that the NLMC

reached out to various interest groups to help it determine what its

objectives should be and how these should be attained. A network of

advisory cOMMittees, commissions, and committees of inquiry was appointed

with representation from these groups. The traditional political structure

- the chieftaincy - was consulted, grievances were aired, and efforts were

made to correct the injuries of the CPP, including the restoration of

traditional authority (stools) and customary rules of land tenure. a In

addition, local private businesses were encouraged through abolition of the

property tax and reduction of sales taxes on local manufacturers, while

rates of taxation on imports competing with domestic production were

increased. There was also a reorientation in the pattern of investment.

The CFP had concentrated development along the coast, with emphasis on

capital-intensive, publically-owned enterprises, many of which were not

financially viable. The NLC favored profitable, labor-intensive projects

that would ease the unemployment problem and bolster the rural econouy.2

Most importantly, the NLC set about trying to stabilize the

econroy and getting inflation under control. This was, in fact, a

precondition for returning to civilian rule. Major responsibility for

economic stabilization was vested in a National Economic Cmnmittee, under

'4 FkRbert Pinkey, Ghana Uhder Military Rule. 1966-1969, Lnndon:Methuen, 1972, pp.1-3.

Pinkney, Ghana Lhder Military Rile ... , pp.21-28.

Pinkney, Ghana Lhder Military Rile ... , pp.30-31.

51

the chairmanship of E.N. Anaboe, the JOverNrent Statistician. There was a

major rescheduling of short-term external debt and the imposition of strict

controls cNer public expenditures and subsidies. The IMF agreed for the

first time to a standby arrangement and provided technical assistance to the

government.

The system for allocating import licenses was altered to try to

ensure efficient mobilization of domestic resources and adequate supplies of

essential commodities for consumers.27 Preference was given to larger

established firms with proven trade and financial contacts. In order to

avoid wasteful competition and underutilized capacity. Nevertheless,

although the corruption of the last years of the Nkrumah period was avoided,

there was still a substantial amount of arbitrariness and inefficiency in

the allocation procedure. 3

The results of this effort were enccuraging. The current accout

deficit was reduced from $212.1 million in 1965 to $42.9 million in 1966.9

Inflation was negative in 1967, principally because of a decrease in food

prices associated with good harvests, and only 8.1 percent in 1968.3° There

was an increase in unemplcyment as government workers were laid off, but

higher cocoa producer prices helped to ease the transition for workers who

returned to rural areas. Nevertheless, urban unrest increase because of

2 Leith, Foreign Trade Regimes..., pp.31-33.

a The operation of the import licensing system under Nkrumah isdescribed in more detail in the next chapter.

Table 4, Chapter II.

*° Table 3, Chapter II.

52

growing unemplcyment, reduced real incomes, and declining expenditures on

educational and health services.

Devaluation and Import Liberalization. 1967-70

The balance of payments situation remained precarious, and there

was considerable dissatisfaction with rigid controls and macroeconomic

austerity. There was also concern over the effect that the risk of changes

in the exchange rate was having on potential private foreign investment and

a realization that donor assistance was unlikely until Ghana corrected its

balance of payments. 9- Rescheduling of medium-term debt prcvided some

immediate relief but was not a long term solution.

In the meantime, civilians were gaining increased influence within

the NLC government as plans for a return to civilian rule were being made,

and many of these people were concerned about the lack of growth in the

economy and the need for liberalization. Ohe of the more influential of

these was Dr. K.A. Eiusia, long an outspoken opponent of Nkrumah's sncialist

ideology.

Busia and others who organized the Progress Party (PP), once party

activity became legal in 1969, provided the intellectual and ideological

underpinning that the NLC lacked. They saw traditional institutions as the

fcundation for continuity and viewed the state as a regulator of private

activity rather than as having an important, directly productive role. As a

national leader, Bhisia conderned the colonial presence, but he was also

steeped in the British heritage and bound to the preservation of colonial

values. He therefore had a high regard for civil liberties and

:3 Leith, Foreiqn Trade REciimes..., p.111.

53

opportu-ities for individual mDbility. He was less concerned over issues of

equality. Finally, Busia and the RP favored close relations with the West,

foreign business interests, aid private investment. 2

This philosophy was especially appealing to professionals,

businessmen, and traditional rulers because it reinforced their basic

position in society. On top of this political base, Busia built an Akan

party that was the successor to the National Liberation Movement, which had

battled the CPP on behalf of Ashanti cocoa farmers during the 1950s. The

Progress Party did not duplicate the forest-coast split of the 1950s upon

which Nkrumah had capitalized, however, but sought to unite all Akan under

one umbrella movement.32

In July 1967, the currency was devalued by about 43 percent, the

producer price of cocoa was raised by 30 percent, and wages and salaries

were increased by 5 to 8 percent.34 Import duties were lowered an some

essential commodities, and the government committed itself to liberalizing

imports over the next few years. As this occurred in the absence of

restrictive macroeconomic policies, h:ovEver, the trade balance began to

deteriorate. For the first few years, this was masked by high world cocoa

prices, substantial foreign aid inflows, and sane debt relief.

Nevertheless, cocoa production remained stagnant in the face of producer

prices that were still low in real terms compared with those that

characterized the 1950s, and non-cocoa exports were frustrated by excessive

s Chazan, An Anatomy of Ghanaian Politics ... , pp.124-25.

Chazan, An Anatomw of Ghanaian Politics .... p.772.

3' Leith, Foreign Trade Reuimes..., p. 1 1 1 .

54

regulations and inadequate incentives in the face of a still overvalued

exchange rate.'5

The Progress Party won the election in August 1969, and Busia' s

government of the Second Republic came to power. Since many members of this

gcvernment had played an important role during the last years of the NLC,

its economic policies were marked more by continuity than by change from

those of the previous regime. What did alter, however, was the economic

environment and the constraints which the government increasingly faced.

Confrcnted with a deteriorating economic situation, the Busia

government undertook a number of policy measures. Shortly after coming to

power, it attempted to alleviate unemployment by expelling all aliens from

the ccuntry thrcough passage of the Aliens Expulsion Act in rbvember 1969.

This antagonized neighboring ccuntries and deprived the economy of some of

its seasonal agricultural labor. The government also tried to accelerate

Ghanaization of small retail trade and other businesses thr-wngh passage of

the Ghana Business Bill, but lack of capital and expertise inhibited this

measure's success. Furthernmore, though the govemment was ostensibly

ccmmitted to improving incentives for farmers, the Cocoa Marketing Enard

retained its monopoly of the cocoa trade, buying from cooperatives and

indigenous Licensed Buying Agents after the LGFOC was discredited in 1966.

Mbst significantly, the goverrnment took advantage of windfall

cocoa profits resulting from high world market prices in 1970 to permit a

rapid expansion of imports and public expenditures. This increased the

balance of payments deficit and flooded markets with imports purchased

: Table 4, Chapter II suggests that the official exchange rate wasovervalued from 1967 to 1969 by 40 to 50 percent in comparison with theequilibrium rate.

55

primarily by the urban middle and upper classes. Food actually became

cheaper during this period in the cities than in rural areas because of

increased imports. As cocoa prices dropped in 1971, however, the balance of

payments deficit became acute.`:

Collapse of Import Liberalization. 1971-72

With accelerating inflation eating away at the devaluation of

1967, the real effective exchange rate on imports declined at the same time

that the import regime was being liberalized. Cocoa prices on the world

market were down and the ga%'ernment was reluctant to raise producer prices

because of its revenue needs. Despite various promoticnal schemes, non-

cocoa exports remained frustrated by neglect and unnecessary regulation.

Exports of timber and minor agricultural crops were controlled by marketing

boards of "dubious prcmoticnal value."`Y With a substantial government

deficit budgeted for 1971-72, upward pressure on prices seemed likely to

erode further import taxes and export subsidies.

In 1971, the trade accoumt surplus that GCana had rum since 1967

to cover its services and transfers deficit, as well as its debt service

payments, no longer existed. Foreign exchange reserves at the end of the

third quarter of 1971 were less than half the trade deficit over the first

three quarters of that year.3 On December 27, 1971, Prime Minister Ehsia

-~ Chazan, Anatomy of Ghanaian Politics ... , pp.159-61.

7 Leith, Foreign Trade Regimes ., p.150.

3X Leith, Foreign Trade Regimes..., p.l51.

56

announced a devaluation of the cedi frnm Ni 1.02/$tI to !K 1.82/$SE.z' At

the same time, import surcharges and taxes on current account payments were

abolished, making the net devaluation about 12 percentage points less than

the gross change in the exchange rate. The 25 percent export and tourist

bmurses were also abolished, thouxgh the affected items acccunted for less

than 10 percent of total current acccunt receipts. Although timber and

minerals received the full benefit of the devaluation, the cocoa producer

price was raised by cnly 25 percent, allowing the government to skim off a

substantial portion of the increased FOB price measured in domestic

currency.

Despite some wage increases, the devaluation implied a huge loss

of real income, especially for heavy users of importables. This was in

contrast to the 1967 devaluation, when suppression of demand and limitations

an imports over the previous 12 months lessened the real impact of the

change in the exchange rate. Instead, it resembled more the situation in

early 1966, when a tightening of import licensing resulted in a sharp fall

in the availability of external resources.° As in the earlier instance,

the result was a toppling of the government, with Eusia being replaced in a

coup in January 1972 by Colonial I.K. Acheampong and the National REdemption

Council (NRC).

In retrospect, it appears that the Busia regime was undermined as

much by its own political weaknesses as by the economic crisis. From the

day it took over the government, it faced strong opposition. First, in

5" Leith, Foreicn Trade Reaimes ... , p.152. Since the dollar haditself been devalued a few days earlier, the weighted (by trade shares)average depreciation against all currencies was 92 percent.

40 Leith, Foreign Trade Regimes ... , p.154.

57

ethnic terms, its Akan base assured the animosity of noncAkan whDo were

excluded from access to state power. Second, in urban areas it was an

elitist party without strong ties to workers and other lower income groups.

This strengthened patron-client relations as chief means by which the

nonelite could gain access to resources controlled by the sate. Third, it

made policy errors that alienated important grcups such as the muslim

community, trade unions, the civil service, students, and the military.

Mbst protests were designed, however, merely to encourage a

redistribution of state controlled resources rather than to change radically

the political system. Opposition was urban based, small in scope, and of

limited impact. Nevertheless, the Busia government s reaction was entirely

out of proportion to the severity of the threat. It broke up

demonstrations, quelled strikes, outlawed opposition newspapers, disbanded

the TUC, and detained political oppcnents. In this way the state became

increasingly isolated and ineffective, more because of its own actions than

because of the stridency of its opposition. The military takeover that

followed had been planned shortly after Busia's inauguration by a small

group of middle-level officers concerned by the erosion of the military's

status and by the decrease of their own material benefits. Its success was

made possible, however, more by the ineptitude of the government in

handling what could be considered reasonable opposition to policy measures

in time of economic difficulty than by the fundamental economic forces at

work.

58

Early Years of the NRC. 1972-75

When it first came to power by means of force, the NRC was

confronted with the problem of establishing some degree of legitimacy in its

right to rule. Demonstrations by the Trade Uhion Congress and the National

Uhion of Ghanaian Students in favor of the new regime indicated some

measure of popular support, but reacticns elsewhere were far more

ambivalent. Many people were of the opinion that the structures of the

Second Republic were fundamentally scund and that more time had been needed

to find workable solutions to the economic crisis.`'

The immediate response of the NRC was to slash prices, revalue the

currency, and reimpose strict import licensing. Debts incurred during

previous years in the form of medium-term suppliers' credits were

repudiated. Blame for the country's economic ills was cast on the

politicians, and depoliticization of the political network commEnced with

the detention and arrest of Progress Party leaders and over 1300

ex-politicians. Freedom of speech was limited, and attacks on government

action were specifically prohibited.

On the more pDsitive side, Acheampong scight to create an alliance

between the military and the civil service, arguing that what was needed was

a technocratic/administrative, rather than a political, approach to

development. He tried to suppress ethnic divisiveness, creating the most

ethnically balanced cabinet since independence, and to promote regional

balance and ties with traditional political units.

Economically, the Acheampong government emphasized self-reliance,

embodied first in the launching of Operation Feed Yourself, which was a

4 Chazan, An Anatomv of Ghanaian Politics..., p. 2 30.

59

program intended to achieve self-sufficiency in food production and to

reduce heavy expenditures of foreign exchange on essential ccmnodities. To

promote greater regional balance, Regional Development Corporations were

set up and charged with overseeing local development projects.4 This

became an important mechanism for channeling input subsidies to particular

groups of farmers, such as those growing rice in the north.

Although the NFC envisaged it role as regulatory and

incentive-oriented, its emphasis was on exhortaticm and moral incentives

rather than economic ones. Operation Feed Yourself, for example, depended

largely upon direct farmer response rather than price incentives. It was

followed in 1973 by Operation Feed Your Industries and Operation Haul the

Food to the Markets. Some success was achieved through backyard gardens and

other public efforts, but fluctuations in production during this period were

probably due more to variations in weather than to anything else.

Furthermore, despite these successes, shortages of food at official retail

prices abounded, and the governoent was compelled to subsidize food imports

and to request large amounts of food aid, especially during the period from

1972 to 1975 of very high world grain prices.

In the cocoa sector, the expensive mass spraying campaigns of the

Busia government were abandoned, and subsidies were instead offer ed for the

purchase of insecticides and sprayers. High prices on world markets also

meant that the government was able to increase producer prices

substantially without cutting into public revenues. This resulted in some

2 Chazan, An Anatomv of Ghanaian Politics .. , p.163.

60

continuation of the planting that had occurred under the NLC and Busia

following the disastrous years of the Nkrumah regime.43

On the industrial side, the government promoted self-sufficiency

by the naticnalization and indigenization of foreign owned or staffed

firms. At the same time, road maintenance all but ceased and import

restrictions severely limited the availability of spare parts.

Distributimn of petroleum products was also closely controlled. One result

was a marked rise in transport costs. The governmmnt, in addition,

attempted to curb smuggling and black marketeering, to control prices of

essential goods, and to divert retail trade from private markets to the

outlets of the Ghana National Trading Corporation.44 The widening gap

between official and parallel market prices, however,only offered an

increasingly strong incentive for rent-seeking activity. Other policies

consisted of tariff reform to increase collection of import duties,

nontraditional export incentives, interest rate ceilings, and investment

incentives.

More importantly, however, was the gcverrment s overall economic

policy between 1972 and 1975. The National F*demption Cbuncil cane to power

in 1972 committed to resuscitating the economy principally by decreasing its

dependence on imports. In addition to tightening import controls, it

attempted to decrease the demand for imports by promoting import

substitution activities, especially food producktion. In addition, efforts

were made to reduce the size of the budget deficit.

A3 Annex 4.

4 Chazan, An Anatonm of Ghanaian Pblitics ... , pp.165-66.

61

Considerable scrcess was at first achieved. In canstant dollars,

the value of imports declined from 1971 to 1972 by 43 percent.4* With

extremely favorable prices on the world market for cocoa, gold, and timber,

Ghana's traditicnal exports, the balance of payments situation improved

dramatically from a current account deficit of $146 million in 1971 to

surpluses of $95 million in 1972 and $114 million in 1973.46 Net foreign

exchange reserves increased to $210 millimn at the beginning of 1974.

Nevertheless, despite the fact that Ghana had repudiated much of its

mediue-term debt, there were still trade credit arrears of about $85

millimn, a backlog of applications to remit profits and dividends of about

$60 million, and government equity capital obligations of $50-70 million at

the end of March, 1974.-'

Furthermore, despite higher cocoa prices, there continued to be

substantial gDvernment deficits that rose from 1971 to 1975 both in absolute

terns and as a share of government revenue and GUP.48 The major causes of

these deficits were an ambitious government investment program and large

wzoe increases to public sector employees in October, 1973 and March, 1974.

The budget deficit, along with N 200 million in domestic debt which fell

due for repayment in 1973-74, was largely financed through central bank

borrowing, especially after March, 1973. Net credit to the government

Ik World Bank, Current Economic Position and Prospects of Ghana,October 18, 1974, Table 39, Statistical Appendix.

k6 Table 4, Chapter II.

4' World Bank, Current Economic..., 1974, p. 1.

1 Table 3, Chapter II.

62

during the following twelve months increased by 23 percent.49 In 1974,

credit to the central govemment rose by 53 percent and that to state

enterprises by 61 percent.

In March, 1974, Ghana and its Western creditors reached an

agreement m rescheduling of the payments dcue n Ghana s medium-term

external debt. Despite this, the balance of payments situation deteriorated

sharply in 1974 because of increased petroleum prices and because issuance

of import licenses far exceeded the import program. In September, 1974, the

value of all outstanding import licenses was reduced across the board by 50

percent and the Open General License system was terminated. Cnmmercial

banks were instructed not to increase credit to finance imports, and import

deposits were required.

As the balance of payments situatin wrsened, the governaient

tried to structure the import bill increasingly away from consumer goods and

toaards industrial raw materials and producers equipment. Restrictions on

food imports, especially rice and maize, resulted in increases in food

prices that were only partially dampened by shifts in production towards

these importables.'° Furthermore, industrial output continued to be

constrained by the lack of availability of imported inputs. All this

contributed to inflatian and undermined increases in wages, producer prices,

and other incentives.

Political support for the regime remained fairly strcrig for its

first two years. The ethnic issue was suppressed by lack of ethnically

inspired favoritism. Patrons associated with the PP who were of Ashanti and

9 World Bank, Current Economic...., 1974, pp. 3,4.

See Table 9, Chapter V.

63

Brcng-Ihafo origin were detained, but, at the same time, Acheampong himself

was Ashanti. There was a feeling of greater equaility of distribution by the

state and less dependence upon the elite-client relations of the Busia

period. There was also fairly widespread support ancig workers, farmers,

civil servants, and other state-based occupational groups. Even the

students were supportive.-L

Macroeconomic mismanagement, however, led to increasing unrest

after mid-1l74. Inflation, which was 9.7 percent per anutmj in 1972, rose

to 18.4 percent in 1974 and 29.7 percent in 1975.n= Formal sector wages,

despite the increases of 1973 and 1974, became progressively eroded in real

terms. A series of coup attempts emanated from vestiges of the CRP and PP

opposition or from the mne ethnic group that was in serious conflict with

the government - The Ewe along the border with Togo.

In Ortober 1975, Acheampong iridertook a purge of the NFC and

created the Supreme Military Council (SMC), made up entirely of military

conuanders. Decision-making was further concentrated in the hands of

Acheampong himself, and the regime, feeling itself increasingly isolated,

became more and more authoritarian. In effect, it retreated from the

broader, somewhat populist political base that it had forged when it first

came to power to its narrower original base of middle generation military

officers concerned principally with power and protection of their material

interest.

M. Chazan, An Anatowv of Ghanaian Politics ... , pp. 2 3 6- 3 7 .

am Table 3, Chapter II.

64

Economic Disintegration, 1976-78

The year 1975 was a critical turning point. Uhtil then, cocoa

prices were high an the wDrld market, food production supplementd by

commercial imports and fool aid was adequate to satisfy most consumption

needs, the macroeccnomic situation was WDrsening but still not out of

control, and state rulers retained a measure of political power.

Thereafter, however, production spiraled doKnwards, cocoa prices fell, the

transportation system deteriorated, food prices rose, and the macroeconomy

cDllapsed. The gDvernmEnt budget deficit rose to 127 percent of total

government revenue in 1976, inflation accelerated to 116 percent per annum

in 1977,53 and balance of payments deficits were held in check only by rigid

trade and exchange controls. Corruption and patronage were rampant, and

civil servants capable of advising on economic policy were without access to

decision-makers. ".w. By 1975 it was clear to even the mDst casual observers

that the state, far from commanding the heights of the economy, was rapidly

losing ccntrol of even the most rudimcntary aspects of econamic

supervision."'4 At the same time, there was widespread evidence that

mosebers of the military were benefitting enormously from the profitable

opportunities created by administrative controls and regulations.40

The economic reasons for this deterioration are apparent. Large

current account deficits, together with sharply rising capital expenditures,

53 Table 3 Chapter II.

'4 Chazan, An InataiW of Ghanaian Politics ... , p. 169.

35 Mike Oquaye, Politics in Ghana (1972-1979), Accra: Tornado, 1980.

65

forced the govermnment to borrow heavily from the Bank of Ghana.O& an the

revenue side, the tax system was heavily dePendent an proceeds frnm cocoa,

and as this sector declined because of lack of incentives, so did public

revenue. In addition, the increasing overvaluatian of the cedi (see Chapter

2) implied that there was less domestic currency to divide the FEB price

of cocoa between government and the farmer. FRvenue from cocoa in 1974, for

exanple, accounted for 46 percent of total governnent revenue; by 1979, it

was only 23 percent, and in 1930 and 1981 it was actually negative because

of the low FOB price for cocoa measured in local currency at the overvalued

exchange rate.-' Furthermore, rising costs of the Cocoa Marketing Ebard

reduced the profits that were available to contribute to government

revenue.A9 In additimn, the ad valorem equivalent of a number of indirect

taxes that were specific, or based on controlled prices, tended to be eroded

by inflation, and collection performance for statutory levies deteriorated.

From 1973-74 to 1977-78, tax revenue as a percentage of GDP declined from

15.1 to 7.8 percent.09

Equally important were such factors as the growth in gonerrnent

expenditures resulting from rising public sector employment, inadequate

administrative control procedures, and the high priority given to the

provision of health, education, and other social services. In addition,

capital expenditures constituted a growing share of the claim an government

-5I6i World Bank, Ghana: Economic Position and Prospects:- Prosperts forExports of Processed Products: Financial Structure - A Flaw of FundsApproach, June 29, 1977, p. 1 1 .

57 Table 8, Chapter V.

5a See Chapter V for further details.

I9 Wbrld Bank, Ghana: Economic Memorandum, April 24, 1979, p.15.

66

resources. Despite efforts to reduce expenditure an low priority

development projects, contracts were awarded or amended without reference to

the availability of funrds and in the absence of effective expenditure

controls.w°

Although scoe external financing was available after the debt

rescheduling agreement in March 1974, most of the deficit continued to be

financed by borrowing from the central bank. The result was a quadrupling

of credit to the government and a growth rate of the money supply (MI) that

averaged 34 percent per arnnum fram June 1973 through June 1976, and 45

percent per annum in the following year. The result was an increase in the

annual rate of inflation from 9.7 percent in 1972 to 116.3 percent in 1977.

This was the highest rate in West Africa and far exceeded increases in

public sector wage rates and in cocoa producer prices.

To check the rise in prices, the government relied mainly an its

prices and inctomes policy. Price controls were ineffective, however, except

in a few formal-sector establisthmEnts, where supplies at these prices were

very scarce. Furthernmore, despite official restraints, private sector wages

adjusted reasanably well to changes in the cost of living, resulting in

substantial disparities between public and private sector wage rates.

Traders, too, benefitted frnm inflation to the extent that they could get

access to goods at officially controlled prices. In any case, trading

margins increased drastically.,1

Low reserves and lack of access to external credit implied that

substantial balance of payments deficits were not possible during the mid

° World Bank, Ghana - Economic Memorandum, April 24, 1979, p. 1 4 .

6 World Bank, Ghana: Ecconmic Position...., 1977, pp.4-9.

67

and late 1970s and early 1980s. The level of imports thus followed the

vagaries of West African weather and the international cocoa market,

together with an overall reduc:tion in the quantities of cocoa produced and

exported because of the aging stock of trees.

By 1977, inflation in Ghana had reached triple-digit levels,

fueled by food shortages and an annual increase in the money supply of 50

percent. Corruption, shortages, and unemployment were rampant as exchange

controls were tightened, taxes on foreign travel were increased, and export

subsidies were raised to 30 percent (20 percent for traditional exports

other than cocoa).2 The smuggling of cocoa exports increased as the

disparity in producer prices for cocoa measured at the black market exchange

rate heightened between Ghana and its neighNDrs. Nevertheless, marketing

and trade of any sort was severely handicapped by the breakdown of the

transportation system due to shortages of fuel and spare parts, and lack of

road maintenance.

All of this served to alienate nearly every political pressure

group in Ghanaian society. The professicnals were at the forefront of the

opposition, and their associations flourished and became highly

politicized. They were soon joined by individual trade unions, teachers,

and students. Religious leaders and women, particularly traders, joined

those opposed to the ErC. Ethnic opposition proliferated as well, as the

Ewe were joined by the Ga, Fante, Ashanti, Brong, Akim, and some groups in

the north. Kumasi, the capital of Ashanti, became the hotbed of

World Bank, Ghana: Economic Pbmorandum, April 24, 1979 p. 18.

68

anti-governmEnt organization. Regional discontent, of ten overlapping with

ethnic dissatisfaction, was widespread.'6

In contrast to opposition during the fusia regime, which was

relatively narrow in focus, the NFC/SMC was faced with opponents from every

quarter. Although this opposition was somewhat elitist and lacked the

populist dimension that later characterized Jerry Rawlings' ascent to power,

it nevertheless was too formidable to permit continuation of the existing

regime. A series of strikes immobilized the country between May and June

1978. Food shortages became acute. Patrons withdrew from the state and

reestablished themselves in self-reliant localities. "In the waning days

of June 1978 the fragmentation was so complete that it became abundantly

clear the SMC was doomed. The only questions were whether the masses would

simply withhold all support for the regime, whether the electorate would

wait for the 1979 date for civilian rule, whether the military would itself

intervene, whether a civilian coup would take place, or whether an all-out

violent uprising would erupt."-', The next month, Acheampong was deposed by

his fellow officers and Lt. General Fred W. K. Akuffo became head of state.

Struggle for Reform. 197E-81

The Akuffo government set about putting the economy back on the

right track in preparation for the return to civilian rule in the summer of

1979. Under prompting from the IMF, the cedi was devalued in Pugust 1978 to

Ni 2.75/$U6, an Austerity budget was introduced, and interest rates were

increased. A currency reform was instituted in March 1979 to rid the

Chazan, An Anatoyw of Ghanaian Politics..., p.242.

4 Chazan, An Anatomv of Ghanaian Pblitics ... , p.269.

69

ecmnomy of some of its mniey in circulation and to sabotage smuggling and

the currency black market. The cocoa prorucer price was increased and

price controls on essential ccmnodities were strengthEned. As a result of

these policies, inflation was reduced to 54.4 percent in 1979.4°

The capacity of the Akuffo governnent to follow through on these

measures, however, was severely limited. It was never thought to be more

than a caretaker govermrnt commnitted to a return to civilian rule in

1979. As a result of heavy pressure, Akuffo decided to lift the ban on

political parties and to convene a constituent assembly. This failed to

reach the radicalized workers, who had beaome alienated from the entire

system, and those local collectivities that had disengaged themselves from

the state. Eighty strike actions were recorded between August and November

1978. Many of these were spontaneous, volatile, and intense. They

continued into 1979 and were met with raids, arrests, and harassmnmts.

Fueled by the austerity program and currency reform, urban unrest finally

resulted in early May in police firing on a group of student dem nstrators

in downtown Accra. The moment was ripe for another military takeover of the

government on June 4, 1979.4e

Unlike previous regimes, the Armed Forces Revolutionary Counlcil

(PFRC), headed by Jerry Rawlings, had strong populist roots. It was

composed of junior and non-ccmnissioned officers with a strong sense of

moral outrage at the crimes and corruption that had characterized the

6 Table 3, Chapter II.

Chazan, An Anatomy of Ghanaian Politics ..., pp. 27 6 - 6 0 .

70

previous military goverTnnent.-.' 5MC leaders, including Acheampcng and

Akuffo, were tried in special courts and executed. Sme bureaucrats were

dismissed; others were tried and convicted. The PFFC sweep was so extensive

that hardly any component of the state apparatus was left unscathed.

A major campaign was also directed against those thought to be

manipulating the economic situation to further their personal interests.

Smiggling, black marketeering, and hoarding were condemned. Market women

were harassed by soldiers and vigilantes, and the Makola market in Arcra

was razed. Controlled prices for basic commodities were closely supervised,

and the Prices and Incomes Board was reactivated to adninister price

controls. Above all, the idea was instilled that people dealing with the

public must abide by fundamental notions of probity and must put the good of

the community above their awn personal well-being.69

There were some adverse consequences of this moral and populist

outburst. Che was a withdrawal of traders from the urban economy, resulting

in high prices and lack of availability of food and other consumer goods.

Another was a fueling of class antagonism, with wtrkers and students

outraged over elite exploitation of the masses. Ethnic cleavages were also

accentuated. Rawlings, an Ewe, was perceived as heading an PFRC that was

dominated by Ga and Ewe. Reactions by Akan, and particularly by the

Ashanti, to OFRC actions were believed to be ethnically inspired. 9

6'- It was reported, for example, that "...66% of all licenses neededto deal in foreign exchange was negotiated through the central bank and thatthe Colonel took a kickback of 107.". Pick's Currency Yearbook, 1977-79,p.259.

dE Chazan, An Anatomy of Ghanaian Politics ... , p. 262.

9 Chazan, An Anatomw of Ghanaian Politics ... , pp. 2 8 2-8 3 .

71

Uhdoubtedly, the most significant outcome of the AFFC interventim

was the emergence of Jerry Rawlings as "...the embodiment of the reformist

and the true patriot. The FRC intercession fostered a peculiar distaste

for military intervention at the same time as it highlighted the Popularity

of its leader.-"7C` This image was strongly enhanced by the PFRC's rigid

adherence to the timetable for civilian electims.

The winner of that electim, which took place in early July, was

the People's Naticnal Party (PNP), constructed fron the remnants of the old

CPP organizatim by Imoru Egala, a northerner who had financed clandestine

cells of the CPP since 1966. Uhable to stand for public office himself,

Egala put forward his nephew, Dr. Hilla Limann, as the party's presidential

candidate. Supporting the party were wealthy patrons, trade-union leaders,

farmers, students, clerks, and the "urban dispossessed." The party was

ethnically heterogeneous, with strcng bases in the north and west. Party

activists had supported Acheampong, Akuffo, and even Rawlings, demmstrating

their opportunism and political survivability. The PNP's major ccmpetition

came frnm the Popular Front Party (PFP), successor to &ksia s Progress

Party, and the Lhited National Cmvention (LtC), which was similar to the

PFP socially but differed from it ethnically in that it constituted a Ga -

Ewe alliance with some Akan representation.'-

The major characteristic of these parties and the candidates they

offered was the reemergence of recognized politicians of the past. All of

these were patrons with strcng ethnic and local ties. There was little to

distinguish between party platforms, despite the ideological differences

Chazan, An Anatcmv of Ghanaian Politics -, p. 2 8 4 .

71 Chazan, An Anatomy of Ghanaian Politics .,.. pp.2Bt6-7.

72

that had characterized their predecessors. Above all the campaign failed to

reach out to the radicalized elements and local communities of Ghanaian

society and had little to do with the poverty and social tensions that

plagued it in 1979. Indeed, the small voter tumnout could be construed as a

rejection by many Ghanaian voters of the hegemony of the middle-class and

their disengagement from the political scene at the state level.-'

Dr. Hilla Limamn came to power in 1979 as the first president of

the Third Republic in a highly ambiguous situation. The econcmy was in a

state of near total collapse, with a budget deficit equal to about 65

percent of total revenue, inflation running at 54 percent annually, severe

shDrtages of all imported goods, and cocoa exports that were less than 75

percent of their level a few years earlier.-- The RNP had few political

resources and was confronted with a Ghanaian public that was cynical and

hostile vis-a-vis the state. Government institutions were in disarray

following the purges of the PfhL Finally, there was Rawlings in the wings:

"My colleagues and I on the PFRC yield the stage to you. You are at the

center of it and the world watches your performance."'4

Despite significant inprovements in the import licensing system,

good harvests, and some decrease in inflation, the government quickly fcund

itself in trouble. The announcement in the spring of 1980 of a bwo-year

agricultural program was unaccompanied by any indication as to how this

would be implemented. A major campaign for reviving the gold industry and

Chazan, An Pnatomy of Ghanaian Politics ... , p.299.

7' Table 3, Chapter II and Pnnex 1.

I Jerry Rawlings, "Address Before Parliament - Inauguration of ThirdRepublic: 24 September 1979."

73

local manufacturing ran up against the problem of how to mDbilize investment

rescurces. New taxes were introduced and tax col]ection was tightened, but

the tax base had been severely eroded When the government raised the

prices of such items as beer, cigarettes, gasoline, and water, consumers

complained that they had already suffered encugh. Efforts to reduce public

expenditures and balance the budget met resistance from workers, who induced

the government to raise the minimum wage, and f rom cocoa farmers, who wanted

higher producer prices. Improvements in the administration of the licensing

system were inadequate to control discretionary allocations and malpractice

induced by the increasing overvaluation of the cedi.

As a result the government was forced to seek larger amounts of

foreign aid. When this failed to yield sufficient revenue, a $1 billion

loan was requested from the IMF, which set as preconditions various

stabilization measures, an increased producer price for cocoa, and

devaluation. Limann balked particularly at the idea of devaluatimn,

protesting that this had led to the fall of Ghanaian governmEnts in the

past .7

Cocoa production declined steeply in 1980 and 1981. Output was at

its lowest level since 1958. At the same time, world prices plummeted.

Gbvernment revenues from cocoa were actually negative in 1980 and 1981

since the world price at the official exchange rate was less than the

producer price plus marketing costs.'* Total debt reached $1.4 billion in

1981 and the goverunnent was $400 million in arrears on short-term debt

repayments. Faced with a growing fiscal and debt crisis, the government

Chazan, An Anatomv of Ghanaian Politics ... , p.312.

' Table 8, Chapter V.

74

resorted to increased deficit spending, which equalled 139 percent of total

revenue in 1961. Inflation rose in the same year to an annual rate of 116

percent.'' The real value of public sector wages eroded despite a tripling

of the minimxm wage in November 1990. With short-term trade payment arrears

piling up, import flows were drastically reduced, resulting in more

shortages and higher prices.

Confronted with these difficulties, the Limann regime suffered

from its identification with the middle-class elite. It was also alienated

from grcnps in the core Akan regions and from the Ewe. The inadequacy of

its rescuirce base implied that the a patronage system was unable to reach

major segments of society. - -

Given the severity of the economic situation and the weakness of

its political position, the government's energies were directed principally

towards averting a further collapse of the formal economy. In its quest for

centralized control, however, the regime proved to be weak and often inept.

Its well-intentioned, but superficial, actions were completely inadequate to

deal with the overwhelming magnitude of the problems.

The failure of the govermment to handle the crisis effectively

provoked expressions of discontent from the organized opposition, which

refrained, however, from seeking immediate changes in the ruling coalition.

Urban gruaps tied to the AFRC and largely excluded from PNP

decision-making, mn the other hand, were mutch more vocal in their

opposition. Students demonstrated, iorkers struck, and farmers agitated.

The Limann government was unable to mollify these highly politicized

`' Table 3, Chapter II.

Chazan, ,n Anatowy of Ghanaian Politics ... , p.310.

75

groups. Heavy-handed tactics began to be emplioyed. Student demonstrations

were forcefully quelled, strikes were repressed, ex-members of the PFRC were

harassed, and Jerry Rawlings was disavowed. Finally, the PNP itself began

to disintegrate as it was increasingly clear that the center of the Ghanaian

state had collapsed. Early in the morning of December 31, 1981, Flight

Lieutenant Jerry Rawlings ance more assumed power and formed the Provisional

National Defense Council (PNDC) to run the government.7

PNDC and the Economic Recovery Program, 1982 - Present

The takeover by the PNDC in early 1962 was in marked contrast to

the intervention of the (FRC in 1979. Whereas the latter was accompanied by

violent bloodshed, the former was relatively quiescent. While the AFR2 was

exclusively military, the PNDC incorporated civilians. Finally, the (FRC

assumed control of the state for a delimited period of time, but the PNDC

was there to stay..°

The new government was faced immediately with a disastrcus

economic situation, includinrg massive budget deficits, spiraling inflation,

the lowest export earnings in years, no foreign exchange reserves, and a

huge foreign debt. To top it off, Nigeria had ,cut off fuel shipments for

lack of payment. The Rawliings government's immediate response was to

curtail current spending, halt the printing of money, demonetize much of

the currency in circulation, close land borders to stem smuggling, and

tighten foreign exchange controls.

79 Chazan, An Anatcov of Ghenaian Politics ... , pp. 314-21.

O Chazan, An Anatomy of Ghanaian Politics ... , p.321.

76

Nagotiations with the 1F, which had been going on under the

Limann govemmnent, were suspEnded until the new regime decided what economic

strategy to pursue. At first it appeared that the PNPC government was

determined to regulate market activity rather tightly since it began in

early 1982, as the AFRC had in 1979, to harass traders in an effort to free

up hoards of goods and to depress market prices. In addition, People' s

Defence Cbmmittees (PDC' s) were set up at plac es of work in towns and

villages to oversee management, with particular emphasis on rooting out

corruption and ensuring the people's participation. To link the PNDC and

the PDC s, an interim national coordinating committee was established with

representatives composed of students and members of the PDCs.

On the other hand, the PNDC also set up the Naticnal Economic

Review Committee (NERC), a high powered body open to varicus opinions, which

began to consider seriously the kinds of policies that would find a welcome

in the IMF and western creditor governments.9- The Committee was charged

with formulating short term policies to get economic recovery underway and

with establishing a medium term recovery program. The NERC completed its

work by May 1982, paving the way for the appointment of Kwesi Botchway,

formerly a law lecturer at the Uhiversity of Ghana and member of the NERC,

to the key post of Minister of Finance and Economic Planning.

The ccumittee' s report appeared at the end of May in the form of a

revised budget for the 1981-B2 fiscal year, which ended one month after the

report was issued. The budget statement listed in detail the failings of

the previous regime but argued that the data base for planning was so eroded

3> Economic Intelligence Unit, Quarterly Economic Review: Ghana,ND. 2, 1962.

77

that fundamental restructuring of the economy could not be undertaken as

rapidly as the situation debanded. The statement went on to list six broad,

but fairly innrcruius, ecanomic objectives. No mention was made of the

plight of the external economy or of the potential roles of either foreign

capital or the multilateral institutions, in particular the IMF. Any

immediate change in this direction was ruled out, in fact, by Rawlings

assertion in April that devaluation was not an option. Instead the policy

goals were more in line with Rawlings' views on the desirability of

self-reliance and self-sufficiency.O-

During the next few mnnths, officials in Ghana maintained contact

with the IMF and the major donors, but the price of needed aid was a change

in the exchange rate, and the government was mcre concemned with institution

building and participation by workers in decision-making than with

correcting the massive price distortions that existed in the economy.

Stronger links were also forged with the eastern block. The anly indication

of an emerging economic strategy was an outline of a three year program,

presented by Hbtchway, which lacked details but seemed to imply increased

self-reliance, self-sufficiency in cotton and major foods, increased use of

local raw materials in domestic industry, and a decrease in the consumption

of petroleum products. A greater role for state control over industry and

trade was envisaged as the primary means of accomplishing these

objectives.Os

am Economic Intelligence Uinit, Quarterly Economic Review: Ghana, Nb.3,1982.

3 Economic Intelligence Unit, Quarterly Economic Review: Ghana,Nb. 4, 1982.

78

By late 1982, the government still remained undecided as to

whether the costs of restructuring the econmWy would be sufficiently eased

by IMF lending to warrant accepting the canditions that would be imposed.

Botchway at this point appeared to favor a modest devaluation. The PNPC was

also torn between its pledge to help the overburdened cocoa farmer and the

dilemma created by converting the FOB price of cocoa into local currency at

the official exchange rate. If the existing rate were to be retained, the

CMB would have to borrow an estimated N 1.8 billion just to pay farmers the

difference between the producer price annouxced by Limann and the expected

FOB price. This would risk adding further to an already high rate of

inflation. Thus the govermeent delayed the start of the buying campaign

while it tried to reach a decision on whether to force farmers to accept a

price cut or to devalue the currency. -

During the closing weeks of 1982 and the first part of 1983,

pressures on the government tightened. Approximately one million Ghanaian

workers were forced to leave Nigeria and return to Ghana. This placed an

enormous burden on the econcmy and its food supplies, especially in view of

the maize shortfall estimated in early 1963 at 378,000 tons following the

poor rains of 1982.

As food prices rose in urban areas, BDtchway anncuriced a plan to

rehabilitate the econamy over the next four years. The emphasis was plared

on reducing foreign exchange leakages, improving tax collection, and

achieving moderate increases in production. Figures for the last quarter of

1982, however, showed that the goverrnent was operating at a large deficit,

04 Economic Intelligence Uhit, Quarterly Economic Review: Ghana,No.4, 1982.

79

with three quarters of its total spending on recurrent costs. In the

absence of capital inflows from abroad, it was unclear as to how any

substantial rehabilitation of the economy was to be achieved. Such inflows

appeared unlikely not only because of Ghana's refusal to devalue the cedi

but also because of the government' s continued emphasis on the need to

strengthEn state control over economic activity. The plan called, for

example, for expansion of the functions of the Ghana National Trading

Corporation in order to put an end to the corruption of officials involved

in import licensing and to insure that ccamissions and profits accrued to

the state. 5

In February 19E3, Rawlings put down the fourth coup attempt in

little more than a year. Negotiaticns with the IMF continued as the

government toyed with the possibility of a dual exchange rate, but it

instead opted for an intricate system of surcharges on imports and subsidies

on exports, amounting to de facto devaluation on current account. The new

1985-64 budget called for substantial increases in the official prices of

most domestically produced goDds, with rice and maize being notable

exceptions, but the T11C quickly expressed its concern about the disparity

between wage and price increases.c-

In June there was another coup attempt. Students protested the

austerity budget and lack of civil liberties, deiwanded a return to civilian

rule, and clashed with workers who supported Rawlings. The TIC continued to

back the government but pressed hard for a substantial wage increase.

E° Economic Intelligence Lhion, Quarterly Economic Review: Ghana,No.1, 1983.

a6 Economic Intelligence Unit, QuarterlY Economic Review: Ghana,Nob.2, 1983.

60

Finally, in August 1963, the IMF approved standby and compensatory

financing facility arrangements totalling $382 million. Once the Fund had

granted its seal of approval, other donors fell in line quickly with offers

of further grants and loans. This began a process of structural adjustment

embodied in the Ecanomic Recovery Program (ERP) amorunced by the government.

The main objectives of the ERP were:

1. to remove the distortions in the ecnanmy;

2. to rehabilitate the industrial, agricultural, and miningsectors;

3. to repair and restore the infrastructural base of theecanomy;

4. to use sound fiscal and monetary policies to achieve andsustain reasonable economic growth over the yearsthereafter.

Phase one of the ERP, termed the Stabilizatimn Phase, concentrated on

eliminating or reducing some of the distortimns in the ecncmry. During the

succeeding Rehabilitation Phase, capacity utilization of local industries

was to be improved by rehabilitation of existing assets. Finally, the

Liberalization Phase was supposed to diminish government interventimn in the

economy by relaxing trade restrictions, thus setting the economuy on a sounid

growth path.

On October 10, 1983, the multiple exchange rates that had been

established the previous April were unified at NW 30/$U6. Subsequently,

Ghana pursued a policy of periodic adjustment, with the rate rising to Nf

90/$4S by early 1966. In September, 1966, a second tier exchange market

was established for most transactions except those involving cocoa,

petroleum, and official debt service, with the rate determined at public

81

auction. On February 20, 1987, the official and auction exchange markets

were unified at the auction rate.

An effort was also made to establish prices, tariff rates, wages

and salaries, and interest rates at realistic levels. Cocoa prices were

raised by 67 percent in April, 1983, by 50 percent in May, 1984, and by 87

percent in May, 1985. Tariff rates for major utilities were adjusted upward

to reflect their cost structures at the new exchange rate. Importers and

domestic producers were also permitted to alter their prices periodically

to reflect the higher cost of imports and increases in wages and other

costs. In December 1983, price ceilings were renmoved an a nurtmer of

imported products, including maize, rice, and sugar, and these prices were

increased to equal the prices of domestically produced equivalents. Where

controls were maintained on 23 other items, prices were jointly determined

by the Prices and Incentives Board (PIB) and by prodxcers/traders under

guidelines established by a tripartite committee of government, employers,

and trade unions. This list was progressively reduced to 8 items by July,

1985. Prices of noncontrolled items allowed full--cost plus a profit margin

to be passed on to consumers but were subject to official surveill lance

under a "reference" price system.

With the depreciation of the cedi, the government was able to ease

restriction on imports. Almost all controls were lifted on imports of

consumer goods paid for with importers own foreign exchange under the

Special Import License scheme. In addition, restrictions on most produicer

goods were removed through access to the foreign exchange auction. In

addition to the easing of quantitative controls, import tariffs were

simplified and lowered.

E2

A number of wage increases were put into effect starting in May,

1983, to offset partially the effects of inflation. The gcvernment's policy

was to reduce the impact of the wage increases on the budget tyy reducing

levels of public sector employment. On October 10, 1983, all interest rates

were raised 3-5 percentage points. Real rates of interest became positive

in 1985 but declined slightly below zero again in 1966 as inflation

accelerated. Rates were raised again in March 1987, with the maximum

lending rate of 26 percent, about double its level fcur years earlier.°'

Fiscal discipline was a major component of the stabilization

program. This was to be achieved in the short run by curtailing the

govemrment's recciurse to the banking system and through large expenditure

cuts. Over the medium-term, fiscal policy objectives included increased

domestic resource mobilization to support larger outlays an operations and

maintenance and an increase in public investment to support the recovery

program. The government was also committed to a phased reduction of its

external payments arrears.

Rehabilitation reforms were drawn up for transport, energy, cocoa,

timber, and mining as part of the goverrnent's 1984-86 recovery program.

These reforms were designed to assure adequate incentives, to improve

management, and to provide inputs and replacement capital. The key export

sectors were allowed to retain part of their foreign exchange earmings to

inport essential raw materials, spare parts, and equipment.aO

°' World Bank, Ghana: Policies and Issues of Structural Adjustment,March 30, 1987, pp.2-3.

aI Wbrld Bank, Ghana: Managing the Transition, Nbvember 7, 1984,pp.5-7; Wbrld Bank, Ghana: Towards Structural Adjustment, October 7, 1985,pp.1-3.

E13

These policy reforms were seriously handicapped in 1913 by severe

drought and the related decrease in domestic energy production. In

addition, slow response on the part of donors greatly increased the problem

of managing Ghana's foreign exchange. Nevertheless, strict fiscal

discipline helped curb the growth of the money supply in 19B4 and 1965,

decreasing the rate of inflation, which was also aided by lower food prices

following good harvests. Furthermore, gross disbursements of medium- and

long-term loans and grants rose sharply in 1964 and 1965, resulting in

greater availability of consumer goods, spare parts, and petroleum

products. The fact that the rate of inflation fell -frcn 123 percent in 1963

to 40 percent in 1964 and 10 percent in 1965, despite substantial

devaluations, is strong evidence that prices prior to 1964 reflected

scarcity values rather than border prices at the official exchange rate.

The result was a significant revival of economic activity as real

GDP increased by 8.6 percent in 1964, 5.1 percent in 1985, and 5.3 percent

in 19B6. 9 Government revenues as a proportion olf GDP also rose from 5.5

percent in 1963 to 8.0 percent in 19B4, and 10.4 percent in 1985.5° The

budget deficit was substantially reduced and re]liance on bank financing

became almost insignificant at 0.9 percent of GDP in 19B4 and 0.8 percent in

1985. Food production decreased in 19E5 in response to the bumper harvests

and lower prices of 1964, but cocoa production rose fran 169,000 tons in

19B4-65 to 226,000 tons in 19E5-61-' as farmers responded to the sharp

W Wbrld Bank, Ghana: Pblicies and Issues..., pp.3-4.

90 Table 3, Chapter II.

9 See Pnnex 1, Table 1-2(3).

64

producrer price increase anncuinced in May 1985. This contributed to a 12

percent expansion of export earnings in 1985..2

Although the Economic Recowvery Program had been largely successful

in achieving most of its initial stabilization objectives by the end of

1986, the process of structural adjustment had just begun. The

rehabilitation of traditional export sectors, particularly cocoa, was seen

as critical to the likelihood of continued success of the ERP. This

included raising proiducer prices, increasing supplies of inputs and ccnsumer

goods as incentives to cocoa farmers, and reducing the operating costs of

the Cocoa Marketing Board. Similar programs were underway for timber and

mining. Equally important were the increased availability of tires,

batteries, and vehicle spare parts, along with the rebuilding of Ghana's

roads, railroads, and ports.

=World Bank, Ghana: Policies and Issues..., p.11.

85

CPd TrER V: G(NOE*EwW PCRLICIES TOWWFRS AGRICL1TURE

The system of government incentives affecting Ghanaian agriculture

consists of the tariffs, taxes, and quantitative restrictians an exports and

imports, as well as the officially regulated prices of outputs and inputs,

credit policies, research and extension services, and general infrastructure

established by the central government.1- The Cocoa Marketing Board has also

played an important role in the extension and marketing services it had

delivered, the input subsidies it has offered farmers, and the claims it has

made an the resources generated by the cocoa subsectar. In addition,

semi-autonomous agencies have administered specific projects and pragrams

aid in so doing have affected the prices and canditions of credit faced by

individual farmers for their autputs and inputs. Finally, there have been

the state farms and other publicly oweed agricultural enterprises.

The implementation of government policies towards agriculture has

been heavily influenced by the evolving political economy in Ghana described

in the previous chapter. Of particular importance have been the price

distortions introduced by policies regarding trade and foreign exchange,

which have often run counter to policies designed to directly influence

agricultural incentives. Equally crucial has been the relative importance

of administrative decisions versus price incentives as mechanisms for

influencing the allocation of resources. Where scarcities have developed

± Mhch of this discussion of government policies towards agricultureis taken from J. Dirck Stryker, "World Bank Western Africa RegionalProject: Ghana, Part II, Economic Incentives and Costs in Agriculture,"Nbvember 1964, which in turn is based on World Bank reports and aninterviews with Ghanaian officials and other experts in Ghana during 1974and 1975. Various Wbrld Bank reports and other studies have been used toexamine the changes that have taken place in these policies before and afterthis period.

86

because of price distortions, goods have tended to be allocated

administratively rather than in response to market signals, aid this has

given rise to bribery, extortion, and other kinds of rent-seeking behavior.

This has resulted in substantial losses in economic efficiency, a transfer

of income and wealth towards the powerful and influential, and an

undermining of the political mechanisms for influencing policy choices.

cocoa Pblicies

Producer prices and the purchasing and marketing of cocoa have

historically been controlled by the Coroa Board (designated the Ccxoa

Marketing Daard, or CMB, thrcughout most of its history), which has used its

surpluses both for stabilization purposes and as a source of government

revenue. Uhtil 1963, the Ibard limited its activities to cocoa, but in that

year it was also made responsible for palm produckts, copra and coconut oil,

shea nuts and butter, coffee, grouncdnuts, and bananas.

During Wbrld War II, the colonial government established a public

miopoly for cocoa exports, replacing the foreign firms that had previously

controlled the export trade. After the war, the export .nopoly was

retained as the Cocoa Marketing Board, while the foreign firms continued to

organize the local purchase and collection of cocoa as the Ebard's

"Licensed Biying Agents" (LBAs). Among the reasons for establishing the CMB

monopoly were:

(1) concern over market-sharing and price-fixing arrangements amongthe foreign trading firms, and

(2) a desire to stabilize domestic prices to producers in the face ofsharp fluctuations in world market prices.

87

Uhder the CMB arrangement, an official prcducer price was determined by the

gpvernnent and anuniciced before each buying seasm. The LAs received a

fixed allcwuance per ton, to cover all expenses of haidling cocoa from huying

point to the ports, plus a profit margin that varied with the price.

Surpluses generated during periods of high world prices were to be used to

finance CMB deficits when wDrld prices were low. In addition, these

surpluses could "... be used for other purposes of general benefit to the

cocoa producers and the industry', including research, cantrol of crop

diseases, credits, cooperatives, and the provision of other amenities and

facilities to the producers"'.2

From the begiming, the CQB accumulated large surpluses, mDst of

which were invested in British governnent securities in order to prevent

inflation in Ghana at a time of high world market prices. The surpluses

were also used for cocoa research and extension and to pay for the control

of swollen shoot disease, including coffpensation for the destructian of

diseased trees. Local expenditures, however, were quite small. Beginning

in 1950-51, the government increased substantially its export duties and

began, as shown in Table 8, to draw off a muich larger share of cocoa

revenue. This was partly because the specific duty previously in effect was

replaced with a graduated ad valorum tax in 1948-49, and the amount of

revenue received from this tax increased dramatically as the average

selling price per ton of Ghana' s cocoa rose by over two and ane-half times

from 1948-49 to 1953-54.3 It was also because of an upward revision in the

ad valor-un rates in 1951, which enabled the gavernment to claim almost one-

2 Beckman, Organizina the Farmers..., pp.40-42.

Eeckman, Organizing the Farmers..., pp.193-94, 279.

Table 8

CKOA Sales, narketing Costs, and Public Revenue

Share of Sales Mi)

Current Total ClPProducer Narketing Payeents Cne Public Current Total

Sales (a) Income (b) Costs to 6oYt (c) Surplus Revenue (d1 Producer Narketing Paysents CHO PublicYear lmill NC) loill NCI lmill NC) msill NC) lmill NC) (sill NC) Income (bJ Costs to Govt Surplus Revenue

1947148 93.0 30.2 3.9 .9 48.2 49.0 .36 .05 .01 .58 .591949149 75.0 65.2 6.6 3.4 -.2 3.2 .97 .09 .05 .00 .041949150 90.2 41.2 6.0 7.0 36.0 43.0 .46 .07 .08 .40 .481950151 140.6 66.2 7.4 26.8 40.2 67.0 .47 .05 .19 .29 .491951/52 103.2 60.0 6.8 29.4 7.0 36.4 .59 .07 .28 .07 .351952153 114.2 60.8 0.6 32.0 12.8 44.0 .53 .08 .28 .11 .391953154 149.4 52.0 7.2 69.0 22.2 90.2 .35 .05 .46 .15 .601954/55 155.0 54.4 7.6 76.0 16.2 93.0 .35 .05 .50 .10 .601955/56 104.6 64.8 9.0 29.2 1.6 30.8 .62 .09 .29 .02 .291956/57 101.4 77.4 10.4 24.0 -10.4 13.6 .76 .10 .24 -.10 .131957/58 125.9 52.0 9.2 52.6 13.0 65.6 .41 .07 .42 .10 .521958159 141.8 61.4 11.0 52.4 17.0 69.4 .43 .08 .37 .12 .491959/60 139.9 65.4 13.4 52.6 8.4 61.0 .47 .10 .39 .06 .441960161 143.2 83.8 21.9 49.4 -11.9 37.6 .59 .15 .34 -.08 .261961/62 138.0 83.6 20.0 54.8 -20.4 34.4 .61 .14 .40 -.15 .251962/63 139.2 81.4 20.6 55.2 -19.0 36.2 .59 .15 .40 -.14 .261963/64 153.9 91.6 22.4 43.6 6.2 49.9 .53 .15 .29 .04 .321964/65 142.4 115.0 37.6 35.6 -45.9 -10.2 .81 .26 .25 -.32 -.071965/66 105.1 75.0 17.2 N/A N/A 12.8 .71 .16 N/A N/A .121966/67 147.3 83.3 17.5 N/A N/A 46.5 .57 .12 N/A NIA .321967/68 237.7 107.4 19.5 N/A N/A 110.9 .45 .08 N/A NIA .471968/69 253.4 94.6 16.3 N/A N/A 142.5 .37 .06 N/A NIA .561969/70 342.7 122.8 19.9 N/A N/A 201.1 .36 .06 N/A N/A .591970/71 279.1 127.2 34.3 N/A N/A 117.6 .46 .12 NIA N/A .421971/72 329.2 139.9 42.0 N/A N/A 146.4 .43 .13 N/A N/A .451972173 351.8 156.3 62.3 N/A N/A 133.2 .44 .18 N/A N/A .391973/74 462.0 156.7 75.3 N/A N/A 229.9 .34 .16 N/A N/A .501974/75 666.0 193.2 124.0 NIA N/A 349.6 .29 .19 N/A N/A .521975/76 634.8 243.4 165.2 N/A N/A 226.3 .38 .26 N/A N/A .361976/77 980.0 248.1 113.9 N/A NIA 519.0 .28 .13 N/A N/A .591977/78 1,143.2 386.6 310.6 N/A N/A 446.0 .34 .27 N/A N/A .391978/79 2,973.3 762.9 426.7 N/A N/A 1,793.8 .26 .14 N/A N/A .601979/80 2,945.8 1,292.0 694.8 N/A N/A 969.0 .44 .23 N/A N/A .331990/81 1,799.2 1,136.0 1,014.7 N/A N/A, -361.5 .63 .57 NIA N/A -.201981/82 1,240.0 2,976.0 1,209.6 N/A N/A -3,025.6 2.40 1.04 N/A N/A -2.441982/83 6,050.5 2,376.0 2,178.0 N/A N/A 1,496.5 .39 .36 N/A N/A .251998314 10,380.0 3,460.0 4,671.0 N/A N/A 2,249.0 .33 .45 N/A N/A .221984/85 15,749.0 5,670.0 8,320.5 N/A N/A 1,757.5 .36 .53 N/A N/A .111995/86 27,342.2 12,791.6 12,524.0 N/A N/A 2,026.5 .47 .46 N/A N/A .07

89

Notes to Table St

Sources: 1947148 through 1964/65 froe Bjorn Becktan, Organizing the Farmers,Uppsilta Scandinavian Institute of African Studies, 1976, pp 279-80.Data are converted fron IC to L using the exchange rate .5NC/L.1965/66 through 1995/86 from Tables 3-312) and 3-413) multiplied bycocoa production adjusted for smuggling in Table 1-23).

Notes:(a) Measured as the FOB price converted to local currency at the official

exchange rate.(b1 1947/48 through 1964/65, Sales minus (Marketing Costs plus

Total Public Revenue); 1965/66 through 1985/8b, Producer Price fromTable 3-3(2) and 3-413) multiplied by cocoa production (adjusted forsmuggling) from Table 1-213).

kc) Export and local duty, voluntary contributions, compulsory savings/farsers' income tax. Loans and grants against reserves not included.

(d) 1947/48 through 1964/65, Current CNM Payments to 6overneent plusC.i Net Surplus; 1965/66 through 1985/86, Sales minus (Marketing Costsplus Producer Income).

90

half of the total sales value in 1953-54, at the same time that the

reserves of the CMB were increased. The farmers share in this year, on the

other hand, was only 32 percent (Table 8).

The cocoa export tax was further increased following the June 1954

elections, while the prcducer price was pegged to its existing level for

four years. Whereas farmers had been reasonably passive regarding the share

of the total value of cocoa sales being sipthed off by the g=vernmlent as

long as producer prices were increased in real terms by two or three times

during the years follcawing Wbrld War II, by 1951 they were pressing for an

increased share. The result was political agitation, expressed partially

through the Ghana Farmers' Congress, which the CPP at the time was trying to

mold into a reliable party organization but later abandoned in favor of the

IGFOC.4 The Coroa Ordinance of 1954 therefore provided a powerful platform

for the Ashanti and other opponents, who left the UGFOC and joined the

National Liberation Mbvement. The Ashanti, in particular, were incensed by

the fact that they shouldered a large part of the fiscal burden but

received little in return since most payments for rehabilitation and

compensation were collected by farmers outside the Ashanti region, and

public investment in general was concentrated in the south, and especially

in the capital of Qccra.

The Naticnal Liberation Mbvement used the cocoa price issue to

reduce the dcudnaince of the CPP by advocating the establishment of a federal

type of government that would redirect a greater portion of cocoa revenue to

the region of origin. This emphasis on changing the structure of government

rather than just its policies, however, enabled the UGFCC to take up the

Beckman, Organizing the Farmers ... , p.193.

91

issue of raising the producer price without being identified with the

opponents of the government. Anxics to dismiss this issue from the ensuing

political struggle, the gcpermwent abandcned its policy of price stability

and increased the producer price from 72 to 80 shillings per headlload.

Despite the decline in world cocoa prices taking place at the time, this

price was maintained thrciugh the 1956-57 season in an effort to demmstrate

that CMB reserves were being used to support the producer price. 5

The cost to the governsnent of depoliticizing the cocoa price issue

was high, as the govermment's share of cocoa sales, inclusive of the CM1

surplus, dropped from 60 percent in 1954-55 to 13 percent in 1956-57 (Table

6), but political victory in March 1957 greatly strengthened the capacity of

the CRP to tap the resources of the rural sector and to penetrate it

politically with the UGFOC. The first step was to restore the share of

cocoa revenue going to the public sector by reducing the producer price to

its 1954 level of 72 shillings per load. The secord step, which follo.wed

the introduction of the Second Development Plan in March 1959, was for the

LGhOC to aninunce on behalf of the cocoa farmers that they were prepared to

accept a reduction in the cocoa price from 72 to 60 shillings per load as

their "voluntary contribution" to the development effort. Shortly

thereafter, the Ccru.ncil was rewarded as the CMB withdrew cocoa-buying

licenses from all the foreign trading companies. Farmer opposition by this

time was fragmented and incapable of generating the political support

necessary to reverse these policies.^

- Beckman, Orcoanizina the Farmers ... , pp.1'96-98.

B Eeckman, Organizinq the Farmers ... , pp. 198-204.

92

The governnent was also subsidized at this time with soft loans

from the CM1, drastically decreasing the BDard's liquid reserves and its

ability to stabilize farmer income. This canfirmed the transformatim of

the Board into an instrument of public finance and ended the illusim that

it had an important independent role in fixing the producer price.

Partly as a result of the expansiam of Ghanaian production of

cocoa from about 220,000 tans in 1951-54 to over 400,000 tans in 1960-63

(Amnex Table 1-2(3)), world prices plummeted, creating severe financial

difficulties for the c(.2'V In an effort to restore falling public

revenues, a compulsory savings scheme was introduced in 1961. Cocoa farmers

were required to pay 10 percent of their gross earnings in exchange for

Naticnal Development Ebnds, redeemable after ten years. COtce again, the

Farmers' Council supposedly spoke for the farmers in agreeing to such a

scheme, as it also did six months later when the farmers purportedly agreed

to renounce their claims to their savings ten years hence. In 1963, the

compulsory savings scheme was replaced by a farmers' inccme tax charged at a

flat rate equal to the previous savings deduction. 0

Farmer hostility towards the L[FCC was not confined to its role in

mobilizing ccntributians from the cocoa sector. The monKopoly granted to the

LGFCC in 1961 created enormous opportunities for the Secretary-Receivers,

the clerks in charge of the local biying centers, to exploit their

7 Beckman, Oruanizina the Farmers ... , pp.204-06.

9 Beckman, Organizing the Farmers ... , pp. 2 0 7 - 1 1 .

93

positions. The list of farmer grievances was very long.9 They included

manipulation of scales and weights, imposition of unofficial levies,

misappropriation of funds, profiteering on the distributicn of farm inputs,

and extortion. Secretary-Receivers were, in turn, required to pay off their

superior officers, thus creating an pyramid of corruption. In addition,

farmers were frequently confronted with delays in selling because the

Secretary-Receivers were not at their posts. Payments were also

occasicnally made with promissory notes rather than with cash, and ultimate

payment was delayed at times up to four months.

Another problem created by the CMB, and the mcnopoly position of

the UGFCC, was the application of high quality standards without offering

farmers a corresponding price incentive. Secretary-Receivers would

sometimes refuse to accept cocoa even though the farmer cansidered it to be

perfectly good. Farmers did not know what to do with grade I! cocoa, which

was no longer accepted.

With respect to inputs such as spraying machines, the problem was

that effective demand was substantially greater than available supply at

subsidized prices. The Secretary-Receivers who handled most of these

inputs were clearly in a position to profit handsomely. After the LUGCC

took aver distribution of insecticides from the Ministry of Agriculture in

1963, mismanagement resulted in severe shortages in scme areas. Because of

considerable public pressure, responsibility for distribution wes handed

back to the Ministry of Agriculture, but the LUFCC continued to play a role

9 Many of these grievances were aired to the Committee of Enquiry onthe Local Purchasing of Cocoa, chaired by John Colemen de Graft-Johnson,which was established after the NLC came to power. The problems cited hereare taken from the discussion of the report of this committee in Beckman,Oroanizino the Farmers...

94

in identifying "legitimate farmers" to be allowed to purchase other inputs,

such as machetes, at subsidized prices through state-owned Ghana National

Trading Corporation stares." 0

Dedurtimn of local UGFCC al lowances and expenses was authorized by

the National Delegates Conference in 1962.11 Acccountability for these

funds was totally absent, however, and misappropriation was common. Delays

at the buying centers also offered abundant opportunities for extortion and

bribery, even though the reasons for the delays may have been beyond the

ccntrol of the local marketing staff. The same was true of shortages of

cash, though rn-the-side lending of these deposits also contributed to the

shortages. Extortion was also cammn as the price to be paid by

transporters in order to haul cocoa away.

With its share of sales revenue seriously reduced and its reserves

frozen in loans to the government, the CM had great difficulty in meeting

its payments during the early 1960s. With the collapse of cocoa prices in

the second half of 1964, after farmers had already been paid the agreed

producer price for a very large crop, there was nothing left to pay the

governnent its cocoa export duty after the purchasing and marketing costs of

the CM and the LUFCC had been covered. In July 1965, the Cocoa Marketing

Board requested, and the Farmers' Council agreed, that the producer price

should be reduced to 40 shillings per load, government subsidies on

insecticides and sprayers should be discontinued, and farmers should

±0 In 1965, and as a result of strong import controls, the freemarket price of machetes was quoted as being six times the official price(Beckman, OrCanizino the Farmers ... , p.1 1 7).

aJ Beckman, Organizina the Farmers ... , describes the veryutndemocratic process by which these delegates were chosen.

95

voluntarily restrict further cocoa planting. In return, farmers were to be

exempt from inccae tax, and local caoucils and other bodies were prohibited

from imposing levies an cocoa farmers without their approval.

Farmers were strcngly opposed to this agreement, but "No voice was

raised against the sharp reduction in the producer price in the Natimnal

Assembly debate on the motimn. Instead, the debate developed into a general

attack on the Farmers Crunrcil as a monopoly buyer of Gbana's cocoa."'-a As

had been true with the National Liberation Mbvement nine years earlier,

cocoa farmers were more concerned to change the structure of the system than

to influence producer prices. This illustrated the degree to which price

policy and its administration were linked with the issue of political

representatic. The vulnerability of the political base of the UGFCC in

the cocoa growing areas was highlighted by the collapse of world market

prices, and it was ultimately the Minister of Finance, worried about the

fiscal implicaticns of the governfent's relaticos with the cocoa farmers,

who belatedly tried in 1966 to push the CoLzlcil in the directimn of more

democratic representation.

The 26 percent decrease in the producer price of cocoa seriously

reduced incomes in the cocoa sectors, especially in view of the fact that

prices in late 1965 were rising at about 30 percent per year. With

production in 1965-66 returning to more normal levels following the bumper

harvest of 1964-65, producer income from cocoa decreased in real terms by

about 60 percent of its level in 1964-65. Yet salaries of UGFCC staff were

Be Eeckman, Orcanizina the Farmers ... , p.218.

96

increased in 1965, and CM employees were granted a cne mnnth salary

bEnus. 1

With the fall of Nkrumah and the coming to power of the NLC, the

LFGOC was dissolved and African individuals, partnerships, cooperatives, and

companies that could handle at least 5000 tons of the main cocoa crop were

licenced to purchase cocoa from the CMB. Because the number of buying

agents was judged insufficient to handle the crop, however, the Produce

Buying Company was established in 1966 as a subsidiary of the CMB to

participate in the purchase of cocoa. To further encourage cocoa harvesting

and marketing under the new system, the producer price was increased to 50

shillings (NX 5.00) per headload. In addition, farmers were paid a bonus

for Grade I cocoa in order to upgrade the quality being sold. Finally,

public sales of insecticides and sprayers at subsidized prices were resumed,

initially through the Ministry of Agriculture and later through the Cocoa

Marketing Board. 4

During the period from 1967 to 1977, the system for purchasing and

marketing cocoa progressively broke down as economic conditions in general

deteriorated and CMB costs became an ever greater proportion of total F(B

sales converted to local currency at the increasingly overvalued exchange

rate. As shown in Table 8, these costs in 1967-68 accounted for 8 percent

of the total value of sales, whereas by 1977 the CB s share had risen to

27 percent. Producers, on the other hand, received 45 percent of total

Bateman, Oroanizina Cocoa Farmers ... , pp.218-22.

P4Republic of Ghana, "Government Statement on the Report of theComnittee Appointed to Enquire into the Lncal Purchasing of Cocoa," W.P. ND.3/67.

97

sales in 1967-68 and 34 percent in 1977-71, while the government s share was

47 percent in 1967-68 and 39 percent in 1977-78.

The initial CMB ordinance of 1947 specified that surpluses of the

CMB should be retained as reserves to be used to stabilize producer prices,

to finance cocoa purchases, and to prmsvide assistance to farmers in all

aspects of production. New legislation enacted by the Nkrumah goverrnenet in

March 1965, however, required the CMB to transfer all operating surpluses to

the central governeent, including all reserves held at the time.-" Thus

the distinction between CMB payments to the gvernment and its net surpluses

was erased.

As the macroeconomic situatimn deteriorated during the 1970s, the

block allowance provided to the Licensed Buying Agents (LEs) became

increasingly inadequate to cover their costs. Further-more, the smaller

companies had difficulty in financing cocoa purchases following the

abolition of the chit system in 1973 and its replacement with the statutory

requirement that all purchases be paid for in cash. Censequently, the LBAs

began to withdraw from the cocoa buying campaign, and the Produce Buying

Agency (formerly the Produce Buying Company) increased its share of the

market. By 1975, the PEA was purchasing about twor-thirds of the crop, and

almost all the rest was being bought by the Ghana Cooperative Marketing

Associatimn Ltd. (GCMA) from farmer cooperatives acting as sub-agents."

This did not solve the problem, however, since lack of road maintenance and

Wbrld Bank, Eastern Region Cocoa Project - Ghana, Arnex 1, p. 3 .

IWorld Bank, ADmpraisal of Ashanti ReFion Cocoa Project - Ghana,Nbvember 18, 1975, p. 1 2 .

98

a shortage of spare parts caused transportation costs to rise more rapidly

than the share of cocoa sales allocated to the CMB.

In 1977, the PBA and the GEMQ were merged into the Produce Euying

Division of the CMB, which was granted a moiopoly on cocoa purchases,

ostensibly because of the inability of the LBQs to settle outstanding debts

totalling N65 million and because of favoritism and corruption associated

with their activities.." In practice, however, the PBD (later the Produce

Buying Ccmpany) became a very costly organization, which by 196l was

operating 4300 buying centers, each with four full-time, year-rcxund

employees.1-

The problem of burgeoning employment was not limited to the PHC.

At the time of its dissolution in 1966, the LUFCC was the largest commercial

establishment in Ghana, employing over 13,000 people. By 1982, the Cocoa

Marketing Ebard and its subsidiaries were paying the salaries of 105,000

workers, 20,000 of which were later determined to have died or to be no

longer working for the Ebard. The large fixed costs of the CMB implied

that, with declining production and an increasingly overvalued cedi, the

share of the Ebard in total sales continued to rise. In 1981-82, with the

equilibrium exchange rate equal to 15 times the official rate, CMB costs

exclusive of the price paid to the producer actually exceeded the value of

FOB sales at the official exchange rate. Even in 1985-86, after the cedi

1 Kwesi Ahwoi and Pn-andan Nar-ayanan, "Restructuring the CocoaMarketing System in Ghana - Some Application of the Malaysian Experience,"World Bank, July 1 - August 15, 19b6, p.5.

1E FAD, Ghana - Third Cocoa Project, Draft Report of the FAO/WorldBank Cooperative Programme, Investment Centre, 29 August 1986, p.15.

99

had moved frm Ng 2.75/$ to NW 90/$, CMB costs still accounted for 28

percent of the total value of sales.'

Although there are no detailed studies, such as the de Graft-

Joh-ism report cn the ULFCC, to document the administrative problems

involved in cocoa purchasing in recent years, there is evidence to suggest

that the PBC's oxnopoly m cocoa purchases has had similar problems. By the

1981-82 crop year, for example, there was a cumulative backlog of unshipped

cocoa equal to me-half the harvest of that year because of weak management

and inadequate transportation.7-0 Lack of credit, moreover, resulted in

delayed payment and farmers being issued ICUs in lieu of cash. Corruption,

embezzlement, and diversim of cash was so commnm that, in 1982-83, a system

was instituted in which farmers were paid by check. For this purpose, 14

banks were established in the cocoa growing areas. This posed severe

problems, however, for farmers who had to travel, in many cases, lang

distances to cash their checks.

As the costs of the CMB increased relative to total sales, the

share of farmer inccae declined. At first, however, there was a concerted

effort following Nkrumah's overthrow to raise produicer prices.21 As can be

seen from Table 9, the real prockacer price increased, as a result, by 35

percent frcxn 1965 to 1968. It then fell by 16 percent from 1968 to 1971

9 J. Dirck Stryker, "Determination of the Cocoa Producer Price: ATechnical Nbte," prepared for the Gcvernment of Ghana and the World Bank,October 22, 1986.

I? Eccnomic Intelligence Uhit, Quarterly Economic Review, 1982.

21 Following the initial price hike by the NLC to NV 5 per headloadin 1966, the producer price for cocoa was increased to NV 6.50 in 1967, NV7.00 in 1968, and NV 8.00 in 1969 (World Bank, Easter Reoion Cccoa....Annex 1, p.31.

100

Table 9

Real Producer Price of CocoaINC/at)

Nominal RealProducer Rural Producer

Price la) CPI lb) Price (c)INC/et) (1972=100) (in 1972 NC)

Year (1) (21 (3)

1953 269 39 6981954 287 37 7921955 299 37 7991956 280 38 7331957 269 40 6791958 239 40 5991959 224 40 5601960 224 41 5461961 224 42 5391962 220 44 5011963 202 48 4201964 197 53 3511965 187 67 2791966 224 76 2941967 254 70 3641968 284 76 3761969 293 81 3631970 293 84 3491971 293 92 3201972 366 100 3661973 439 117 3741974 499 138, 3541975 585 176 3331976 732 279 2631977 1333 606 2201978 2667 1058 2521979 4000 1685 2371980 4000 2714 1471981 12000 5683 2111982 12000 6962 1721983 20000 15985 1251984 30000 22119 1361985 56600 24064 235

Notes to Table 9:(a) From Table 3-3(2).(bi 1953-1966 figures are linked to the national combined UI

from Table 3-5(1).Source for 1967-1969 and 1971-1972 is lorld lank, Shana:

Economic Position and Prospects, vol. 1, June 29, 1977,Table 6.

Source for 1970 and 1973-1982 Is Shana Policies andProgram for Adjustment, World lank, June, 1913, Table 7.3.

Source for 1993-1994 Is Ghana Towards Structural Adjusteunt,World lank, vol. 11. October 7, 1995, p. 80.

Source for 1985 is Statistical News Letter, StatisticalService, August 18, 1996, vol. 11.

lc) Column 11) divided by coluen 12) multiplied by 100.101

under Bisia' s civilian regime, increased by 17 percent during the next two

years under the NRC, and began a prolonged plunge in 1974, which did not

bottcm out until 1983 when the real produkcer price was only 34 percent of

its level in 1972. Only in 19B5, after inflation had been seriously slowed,

was the increase in the nominal producer price sufficient to cause a

significant rise in the real price.

The share of farmer income in the total value of sales follced a

different course. Because of low world market prices, farmers earned 81

percent of sales in 1964-65 but only 36 percent in 1969-70 after world

prices had recovered. Their share subsequently increased to 44 percent in

1972-73 and then declined to 29 percent in 1974-75 as a result of the

international commodity boom. Thereafter the farmers share oscillated

between 26 and 47 percent except for 1980-81 and 1981-62, when world prices

was depressed and government revenue was actually negative.

As produicer prices declined in real terms, farmers and middlemen

were encouraged to smuggle cocoa across the frontiers with neighboring

comutries, particularly the Ivory Coast, where prices were quite favorable

at the black market rate of exchange. Estimates of the extent of this

smuggling are presented in Annex 1. They suggest that as much as 20 percent

of the crop may have been smuggled out of Ghana in recent years, though

because of poor transport conditions and periodic government crackdwnis a

more reasonable estimate is about 10 percent. The effort going into

smuggling and attempts at its control resulted in a substantial waste of

resources and contributed to the undermining of governmental authority

through bribery and other forms of corruption.

102

Tradable Foods

Imported agricultural products have over the years been subject to

a variety of controls and regulations that, in gener-al, have caused the

domestic price to consumners to exceed the CIF price plus relevant margins.

During the colonial period, food was imported into Ghana by a number of

general import/export houses with wholesale and buying branches throughaut

the country. The main depots in the regional capitals served as the

principal outlets for this food. Although these networks remained in place

after independence, the government reduced their market shares by

establishing public trading monopolies and import licensing policies that

favored small Ghanaian importers, exclusive government departmental trade,

and subsidization of nonccommercial distribution systems.~

During the Nkrumah era, the Ghana National Trading Corporation

(GNTC) was given a public ffonopoly on imports of "essential" food

commodities, such as wheat, rice, maize, sugar, and vegetable oils. These

were sold, in turn, to licensed wholesalers or thraugh the GNTC's own retail

outlets in urban areas. To the CIF price was added an import duty that

varied over time and across commodities, but averaged 10-25 percent, plus

the cost of handling, transport, and working capital. Retail prices of

these ccmmodities were supposed to be officially controlled. These controls

were generally respected in retail stores, but prices in the market were apt

to be higher.23 As inflationary pressures built up, nmreover, these

&'World Bank, "Marketing and Input Supply," Ghana: AgriculturalSector Review Background Paper 5, January 15, 1985, pp.151-2.

= R.N. Ghosh, "Price Build-ups of a Few Imported Food Commodities inGhana," Technical Publication Series No. 32, Institute of Statistical,Social and Econonic RPesearch, Legon, 1972.

103

differences were accentuated and cammodities became increasingly scarce in

the state-coned retail outlets, as well as in other stores that were

required to sell at controlled prices.

The gap between the free market price and the cost of acquiring

imports at the official exchange rate led, during the Nkrumah period, to

widespread rent-seeking behavior. ' In the granting of import licenses,

for example, bribes of 5 to 10 percent of the value of the license were the

norm unless importers had personal contact with the Ministers of Trade or

Industries. Cumbersome procedures were evolved to assure the issuance of

licenses for priority purposes, but these inevitably allciied a certain

amoumt of discretion to government officials in charge of implementation.

The result was not only a transfer of income to these officials but also

substantial inefficiencies. In some instances, for example, import licenses

were revoked after imported goods had already arrived and were awaiting

clearance at the port.

After goods were imported under official license, the government

had neither the administrative controls nor the data to influence their

allocation and to prevent profit-taking. Instead, a number of mechanisms

came into existence that bypassed any effort towards public regulation of

private sector marketing of imported goods, including food. Of particular

importance was the grcuth of sales by large distributor firms, including

the GNTC, to wholesalers rather than to ccnsumers directly through the

distributors' own retail outlets, which tended to be more tightly

controlled. Since large open sales to individual wholesalers would attract

attention, these were frequently disguised as retail sales, with people

2 WDrld Bank, "Marketing and Input Supply," p.156.

104

hired to stand in line for cash purchases. Q^Eues were also used by the

distributors as a means of discouraging retail sales at official prices

since very limited quantities of restricted items such as milk, sugar, and

rice wwld be sold at a time. Once the token public sales were made, the

rest of the goods wculd be clandestinely transferred to selected customers

at much higher prices.

Goods that were in scarce supply at visible retail outlets were

often sold in much larger quantities using the passbook system, which

originated in Ghana following the First Wbrld War. Under this system,

women would receive goods on credit from the large trading hauses and would

sell retail on commission within the informal sector. This proved to be an

excellent mechanism to subvert the price control system since passbook

holders generally sold only to clients that were well known to them

personally. Of the 20,000 or so passbook holders in 1964, very few used

shops or stores for their retail, or in sawe cases sub-wholesale, sales.

Since the passbook system was legal and yet the prices at which goods were

eventually sold to consumers were much higher then their actual cost,

district and sales managers of the distributors received substantial

kickbacks. In some cases they had passbooks operated on their cw4n behalf.

In addition, the "chits" issued by sales managers for the release of goods

were often sold separately.

Even where outright graft was not involved, scarce goods were

frequently allocated on the basis of family ties, friendships formed in

student days, and other types of favoritism. This often required repeated

visits to the person able to authorize the release of gccds, long waits

outside his office, and other wasted resources. In many cases, the goods

105

were never actually acquired by the individuals involved, but the chits were

sold to waiting dealers an leaving the office.

Although there was a major effort after Nkrumah tD clean up the

graft and corruption associated with import and exchange controls, official

price structures, and administrative allocaticn of resources, the impact of

this effort was short-lived as long as the incentives for rent-s eeking

activity existed. In the traded goods sector, these incentives are perhaps

best exemplified by the ratio of the black market to the official exchange

rate. In 1966, at the end of the Nkrtunah era, this ratio reached its first

peak of 3.0.A5 By 1969, the ratio had fallen to 1.7, and it remained at

approximately this level until 1975, when under the SMC the macroeconomic

situation began to deteriorate sharply. Thereafter, the incentives to

divert goods from the official distribution network to the private market

increased dramatically, with the ratio of the black market to the official

rate of exchange rising to a peak of 8.0 in 1977, then falling to 5.9 in

1978, and finally increasing to a new high of 22.2 in 1983 before a series

of devaluations and, ultimately, adjustment through public auctions reduced

the ratio to about 1.2 by late 19B6.

The consequences of these distortions relating to govermnEnt

policies regarding tradable foods were predictable. With increasing

differences between the CIF price in local currency and the domestic market

price of rice, maize, sugar, milk, and most other importable products, the

profits to be gained from access to scarce goods at official prices became

2 Table 4, Chapter II.

106

immense. This gave rise to rent-seeking behavior an a massive scale.2

Frequently involved were military officers, who obtained the necessary chits

that they then turned over to Ghmnaian women with access to the pr-ivate

marketing network. The right chits, especially if signed in green ink by

Colonel Acheampong, provided access to foreign exchange, import licenses,

bags of grain, or whatever else the bearer desired. By 1976, sugar, rice,

flour, and other coimmodities acquired in this way sold on the private market

for at least five times the controlled price. Less and less time was spent

in formal sector employment, including the gvernmnent, in which wage

increases did not match the rate of inflation, and an increasing amount of

effort was devoted to informal rent-seeking.

What was particularly disturbing about this system, as it

continued through successive gpvernnts into the 19EKs, was that people

who potentially had access to scarce goods through the system spent a great

deal of time and effort attempting to acquire the goods at the expense of

directly productive activity. Furthermore, periodic gcPvernment crackdowns

on smuggling and profiteering only increased the cost of rent-seeking

activity without substantially reducing its level.

As the economic situaticn continued to deteriorate, imported goods

became increasingly scarce, and time spent in rent-seeking activity became

less privately productive. It was at this point during the late 1970s and

early 1980s that a large number of Ehanaians dissociated themselves from the

kalabule system and committed themselves simply to informal sector

-6 The system of profiteering took on the name of "kalabule",probably from the Hausa expression "kere kabure", which means "keep itquiet". A detailed description of this system in contained in Mike Oquaye,Politics in Ghana (l972-1979), Accra - Tema: Tornado, 1980.

107

activities in the cities or, in many cases, in the coutryside. Imports of

tradable foods, such as rice and maize, had by this time fallen to less then

10,000 tons per year, except when food aid could be obtained, so that

domestic grain prices were largely divorced from those on the world market

and were determined instead principally by local demand and supply

conditicns.

This situaticn continued until very recently, when food imports

were increased ance more under the Economic Recovery Program. In additian,

the mnxropoly of the Ghana National Procurement Agency (GNPA) on imports of

essential foods, which had been initiated in 1976 to take advantage of bulk

buying opportunities, was broken as private traders were authorized as of

the end of 1985 to use their own foreign exchange to import fod. Despite a

35 percent margin on the CIF price, which was designed to cover

administrative overhead, working capital costs, a "ca..ission", and profit,

the GNPA was at this time in severe financial difficulties because of

mismanagement and inefficient purchasing from overseas suppliers.27

In an effort to substitute local production for food imports and

to stabilize prices for local produciers, the government for many years

tried to increase the share of oatput being handled by public trading

agencies. During the late 1960s and early 1970s, the Grains and Legumfes

Developnent Board (GDB) was responsible for buying and storing maize, rice,

and palm oil in an effort to stabilize prices. In 1975, the Marketing

Division of this agency was taken over by the Ghana Food Distribution

Corporation (GFDC), which had been established in 1971 to market perishable

foodstuffs on which it experienced substantial losses. At the same time,

IWobrld Bank, "Marketing and Input Supply", p.156.

106

the Rice Mills Uhit (FRIU), which had been part of the GDB, was made an

autciomxis body under the Ministry of Agriculture.

The grain marketing activities of the GDB/GFDC were chiefly

oriented towards implementing the minimum guaranteed price for maize.2

This price was established for foodcrops by a Committee an Agricultural

Conmmodity Prices, composed of Ministry of Agriculture and university-based

officials, on the basis of estimated costs of producticn.29 In most years

the minimum guaranteed price for maize was well belaw the market price, but

when the market price fell, neither the storage facilities nor the

financial resources of the GDBJGFDC were sufficient for the buying

activities to have a significant influence. The anuual purchases of the

GE1/GFDC, in fact, never exceeded 12 percent of estimated total

marketings. In 1974/75, for example, when there was a substantial surplus

of maize, shortages of storage facilities and lack of finance prevented the

GMB from supporting the floor price.30 The Rice Mills Uhit, in additian to

being unable to support the minimum guaranteed price for paddy for similar

reasms, also suffered from high milling costs that prevented it from

competing with private rice millers, resulting in substantial excess

capacity.

3 In 1984, maize accounted for 54 percent of total GFDC purchasesand imported rice for another 27 percent. Other purchases were dividedamong starchy staples, legumes, charcoal, and imported "rural trade" itemssuch as machetes. World Bank, "Marketing and Input Supply," pp.160-62.

-- Wbrld Bank, "Agricultural Pricing and Trade Policy Framework,"Draft Working Paper ND. 9 for the Ghana Agricultural Sector RehabilitatimProject, by Lynn Salinger, April 1986, p.22.

-C0 Whrld Bank, Ghana: Agricultural Sector Review, April 12, 1976,Vol.II, Annex 5, p.1 1 .

109

The influence of the state trading agencies on domestic marketing

of food was thus of little importance, though they occasionally had a

detrimental effect by obliging large producers and Agricultural Developnent

Bank borrowers to sell at official prices when market prices were in fact

higher. In addition, high operating costs coupled with low allowable

marketing margins resulted in severe financial difficulties and the need for

frequent government subsidies. As a result, buying of maize was reduced

franm about 30,000 tons in 1973 to a mere 6,000 tons in 1933.:3-

The GFDC was also allowed to purchase imported rice and other

foods from the GNPA and to sell these on the private market at substantial

markups in order to cover part of its losses on domestic purchases. - For a

time this created opportunities for bribery and favoritism, but the windfall

was reduced when devaluation brought border prices more in line with

domestic market prices.

In the face of fluctuating harvests, Ghana has procedures

designed to ensure adequate food security. Estimates of shortfalls made by

the Ministry of Agriculture are passed on to the Ministry of Finance, which

manages the foreign exchange budget and coordinates concessional food aid.

The Ministry of Finance is then supposed to inform the Ministry of Trade how

much food is to be imported. In reality, however, neither the governent

nor the international donors has a very clear idea of expected concessional

imports.:' As a result, there are years in which there are significant

surpluses or shortages, with corresponding price fluctuaticns that the

-' L Wbrld Bank, "Marketing and Input Supply," p.159.

32World Bank, "Marketing and Input Supply," p.161-62.

W obrld Bank, "Agricultural Pricing ... ," pp.2-29.

110

goernment is unable to control thrcugh either its domestic or its foreign

buying and selling operations.

Transnortation and Ncntradable Foods

Motre important than the activities of the grain marketing agencies

has been the effect of gavernment policy on the transportation system. From

1961 to 1966, during the Nkrumah gDvernment, the average age of trucks

increased markedly because of restrictions on imports, and the number in

operation declined because of shortages of tires and spare parts. RcNad

ccnditimis also deteriorated with inadequate maintenance. 4 Ccnditions

improved somewhat thereafter for a few years, but a survey of farmers and

traders in 1970 ftirid that road canditicns and the scarcity of vehicles were

major constraints on marketing and resulted in high transportation charges,

especially on feeder roads.~ It is estimated that the total capacity of

trucks declined from 8,271 tons in 1956 to 4,442 tons in 1968. In 1968

almost 38 percent of the trucks were more than six years old.2

By the mid-1970s the situatim had deteriorated even further.

Imports of vehicle spare parts declined by 40 percent in real terms from

1970 to 1975. Althtgh statistics on the total truck fleet were not

;54 World Bank, Current Ecnomic Position and Prospects of Dhana, Vol.V: Agriculture, C3ctober 26, 1970, p.8.

;so V.K. Nyanteng and G.J. van Apeldomrn, "Sane DevelopmentImplications of Farmers' Problems in Marketing Their Foodcrops," inI.M.Ufori (ed.), Factors of Paricultural Growth in West Africa, Institute ofStatistical, Social, and Economic Research, Iniversity of Ghana, Legon,1973, pp.26E,69.

Kadwo Ewusi, "The Rate of Inf lation, Variation in Local FoodPrices, and the Effect of Transport Facilities on Local Food Prices in Ghanain the Sixties," in Ofori (ed), Factors of Agricultural Growth ... , p.2B4.

111

maintained after 1970, figures on registration of new trucks suggest that

the total number of trucks fell 29 percent from 1960 to 1970 and 16 percent

fram 1971 to 1975.57 As a result, it was estimated in 1977 that 70 percent

of farmers had to headload crops to market.10 By 19Y83, approximately 70

percent of the road vehicle fleet was out of service because of lack of

tires and spare parts.39

The impact of the deterioration of the transportatimi system was

especially severe for the marketing of starchy staples, such as cassava and

yams, and of millet and sorghum, since these crops have relatively low value

in relation to weight. Since no government policies affected the prices of

these products directly, the most important influence that the public sector

has had has been through its policies concerning transportation.

Policies Related to Agricultural Inputs

In contrast to output price policy, which historically has tended

to discourage agriculture, the government has intervened, especially since

the early 1970s, to encourage production through the provision of subsidized

inputs and credit, through its research and extension services, and through

interventions by specialized agencies in specific areas and crops. In

addition, in contrast to other sectors of the economy, there are no direct

taxes in agriculture. Most of these interventions have favored larger, more

modern farmers.

-'7 Wbrld Bank, Ghana: APricultural Sector Review, April 12, 1978,Annex 5, p.15.

as Inception Report, Building and Road Research Institute, " HighwayResearch Programme", 1977.

W obrld Bank. Ghana: Policies and Program for Adjustment, 1984, p.63.

112

Input Subsidies.

There are no import tariffs mn fertilizers, insecticides, and

agricultural machinery and equipment, though some inputs not used solely in

agriculture, such as trucks, are subject to duty. Domestic prices of most

inputs, however, have been influenced less by trade taxes than by government

policy and practice in relation to input distribution. The overvaluation of

the cedi has had an especially important effect on the pricing of inputs -

favoring capital-intensive, mechanized techniques aver those that absorb

more labor.

Fertilizer. Numerous agricultural inputs have wver the years been

distributed at subsidized prices by the Ministry of Agriculture. The

subsidy for fertilizers, for example, was initiated in 1968 and has evolved

as shown in the following table,

Fertilizer Cost, Price, and Subsidy(Nf/mt) (a)

Ccwnxouid Ammoniumn Sulphate

Sales Sub- % Sub- Sales Sub- % SubYear Cost(b) Price(c) sidvyd) sidy(e) Cost(b) Price(c) sidy(d) sidy (e)

1970 110.6 56.0 54.6 49% 81.4 40.0 41.4 517.1971 122.3 56.0 66.3 547. 85.6 40.0 46.5 54%1972 163.9 56.0 107.9 667. 110.7 40.0 70.7 647.1973 183.2 56.0 127.2 697. 155.2 40.0 115.2 74%.1974 353.7 56.0 297.7 847. 293.1 40.0 253.1 867.1975 408.6 56.0 352.6 86. 275.6 40.0 235.6 85%1976 297.5 56.0 241.5 81% 227.5 40.0 187.5 E8Z.1977 306.0 130.0 176.0 567. 296.0 100.0 196.0 66%

N/A N/A N/A N/A N/A N/A N/A N/A1984 2E000.0 9000.0 19000.0 6EMX. 20,100.0 7000.0 13100.0 65%

113

NOTES:(a) Source is WDrld Bank, Ghana: Aciricultural Sector Feview, April 12,

1978, Vol. III, Annex 7, Appendix Table I, and calculaticns forWorld Ebank, Ghana: Agricultural Sector Review, Background PaperNb.l: Incentives and Comparative Advantage, January 15, 1985.

(b) Cost ex-warehouse Tamale, Upper Region.(c) W2.80 per bag of 50 kg for compound fertilizer and N&2.O0 per bag

of 50 kg for ammonium sulphate from 1970 to 1976.In 1977, the prices per bag were raised to NW6.50 and NW5.00respectively, By 1984 these prices had attained !S450 and N350.

(d) Cost minus Sales Price.(e) Subsidy as a percent of cost.

It is clear from the table that there was a tendency for the official price

of fertilizer to remain fixed for a number of years, during which time the

subsidy became increasingly important as a percentage of delivered cost.

Over the years for which we have data, the subsidy rate ranged from 49 to 86

percent. This does not include, of course, the additional indirect subsidy

resulting from overvaluation of the cedi.

A major problem was inadequate supplies of fertilizer, imported by

the Ghana Natimnal Procurement Agency (GNRA) and distributed by the Ministry

of Agriculture.4" During the 1960s and early 1970s the problem was largely

mne of the GNPA s inability to gain access to adequate foreign exchange. As

foreign aid was tied increasingly to fertilizer imports, however, those

imports rose from 9980 tons in 1971 to 69,630 tons in 1976.4- Utilization

increased less rapidly, however, and 17,900 tons were carried over from 1976

into 19/7. Small farmers, particularly, obtained timely access to

*° Prior to the end of 1976, the Ministry of Agriculture was alsoresponsible for importing fertilizer.

4L Fertilizer imports were only 2900 tons in 1962 and 6340 tons in1969, indicating the absence of any concerted effort to develop agricultureother than cocoa bef-re the 1970s. Wirld Bank, Ghana: Agricultural SectorReview, August 12, 1978, Vol.III, Amnex 7, Appendix Table 5.

114

fertilizer with difficulty because of late arrivals, complex administrative

procedures, and heavy competition from large farmers at the subsidized

prices. There was also some smuggling of fertilizer across the borders. As

a result, the village market price paid by small farmers for fertilizer in

March 1977 was NP9.00 per bag, while the official price was Ni2.00., 4

By 1984, the situation was no better, with shipments arriving too

late for planting because of the Ministry of Agriculture's inability to

finance the imports, slow mobilization of transport to inland distribution

points at least partly because of low official transport rates, and

indecision an pricing under the government's new desubsidization policy.43

As a result, fertilizer distribution was transferred to government-uwned

Farmers' Services CAmpanies (FPSCOM) in the Lpper East, Upper West, and

Volta Regions."4

Aside from inefficiency and inequity in allocatimn and delays in

distribution, the fertilizer subsidy grew to be an important public sector

expense. Government outlays increased from NW0.4 millimn in 1970 to W9.2

million in 1976. In 1976/77, the subsidy amounted to 25 percent of the

current budget for all agricultural development, excluding cocoa. '

Although the government was conunitted to phasing the subsidy out over the

period from 1976 to 1990, this still had not been accomplished by 1984.

42 World Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vol.Il, Annex 5, p.21.

Wobrld Bank, Ghana: Auricultural Sector Review, BackgrouxJd Paper5: Marketing and Input Supply, January 15, 1985, p. 1 5 3 .

4 I World Bank, Ghana: Agricultural Sector Review, August 6, 1985,p. 4 2 .

0 Wobrld Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vol.III, Amnex 7, pp.8-9.

115

Machirery Services. In order to encourage farm mechanization, all farm

machinery in Ghana is imported duty free. In addition, tractor services

were for imany years provided at subsidized rates by the Mechanization and

Transport Division of the Ministry of Agriculture (MEY). These services

included clearing, tillage, and harvesting, and the aouint of subsidy during

the early 1970s was about 50 percent of private sector costs. Again, it was

the large farmers who benefitted most from these services. The candition of

the equipment was so poor and the inefficiencies of the services were so

great, however, that the government rapidly declined in importance relative

to private tractor operators, who charged rates considerably in excess of

those of the governmeit. In 1969, for example, it was estimated that only

about 40 percent of the govermEnt's wheel tractors were serviceable and

each of these was annually able to prepare an area of land that averaged

only 12 to 18 hectares. Private tractor naiers, in contrast, were able to

prepare up to 240 hectares.46 As a result, by 1976 the goverrwnmt owied

only 3 percent of all wheel tractors and harvesters in Ghana.4`

Subsidies an mechanized land preparation and harvesting services

provided by the MtA varied in 1977 from 3 to 74 percent of estimated costs,

depending on the service rendered. Except for a -few favored farmers, these

subsidies were irrelevant because of the very limited extent to which the

services were available fram the M1O, though the govermEnt did operate

about 90 percent of the crawler tractors used for clearing at subsidy rates

6 k WDrld Bank, The Current Econcmic Position and Prospects of Ghana,October 26, 1970, Vol 5: Agriculture, pp.14-15.

4' Ghana Ministry of Agriculture, Economic Research and PlanningService, "Agricultural Price Policy in Ghana, " by Frederick D. Kerfker,April 1976, p. 9 .

116

of 0 to 17 percent.49 Far more important in determining the prices of

mechanized services, hwever-, was the influence of Ghana's overvalued

exchange rate an private operator costs.

Improved Seeds. The Ministry of Agriculture operated during the early

1970s a seed multiplication and import program, which primarily benefitted

producers of maize and rice. Farmers in 1974 paid ane-third of the cost of

maize seed and three-fourths of the cost of paddy, though only 10 to 20

percent of all farmers were reached.49 By 1976, the subsidy rate had been

reduced to nil for paddy and 50 percent for maize."Q

Seed price policy lacked adequate flexibility to cope with

varying grain prices. Paddy seed was sold by the Rice Mills Uhit in 1976,

for example, for mnly one-half the market price, resulting in many cases in

the seed being purchased for direct consumption rather than for planting.

Similarly, the MOh offered registered seed grawers N!55/bag of maize seed

while the market price for seed of lower quality was about N80, and the M0A

consequently was unable to buy any of the certified seed. 8 9

Because of the administrative difficulties of running a seed

program from the Ministry, the state-owned Ghana Seed Company (GSC) was

established in the mid-1970s to produce fou,dation seed from the breeder' s

4 World Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vol.III, Amex 7, p.11.

World Bank, The Current Economic Position and Prospects of Ghana,October 18, 1974, p. 1 6 .

eC) Ghana, Ministry of Agriculture, "Agricultural Price Policy...,"p.9.

51 World Bank, Ghana: Agricultural Sector Review, April 12, 1976,Vol.III, Annex 7, p.7.

117

seed made available by the Crps Fsearch Institute. The fotundaticn seed

was then issued to registered private farmers to produce certified seed on

ccntract. This seed was in turn cleaned, processed, and sold by the GEE.

The G5C proved to be no more adept at seed pricing than the MEA, however,

and by 1984 it was in severe finamcial trouble, with large unsold seed

stocks. At the same time, there was a sharp decline in seed producticn

because of financial problems, unavailability of inputs and spare parts, and

a decline in the number and acreage of participating growers.3-

Other Inputs. The MOA also maintained subsidies cn insecticides and

hand tools, but quantities supplied were insufficient in relation to demand

at the subsidized price, as evidenced by market prices two or three times

official levels. Despite the establishment of two machete factories in

Ghana, farmers by the end of the 1970s were continuing to experience

difficulty purchasing this basic tool at the official price. Smuggling to

neighboring countries also appeared to be a major problem.53

Cocoa farmers benefitted fr-om a number of additional subsidies

provided by the Cocoa Marketing Board. These were temporarily suspended in

1965, as noted earlier, but were later resumed under the NLC. In 1974, the

CMB sold Gammalin, the main insecticide used on cocoa farms, at a price

equal to about one-quarter of its costs. Sprayers that cost the Board

V12B.00 were sold, in turn, for e3O.0O and were repaired free of charge.

Cocoa pods and seedlings were also sold at subsidized prices, estimated at

Om World Bank, Ghana: Aaricultural Sector Review, Background PaperNo.5: Marketing and Input Supply, January 15, 1965, p.167.

53 World Bank, Ghana: Restorina Economic Growth, October 22, 1981,pp.10E-09.

118

about 50 percent of cost. In addition, the CMB financed the constructian

and repair of feeder roads in cocoa growing areas. By 1977, in the face of

constant official input prices and rising inflation, subsidy rates at the

official exchange rate had increased to 81-95 percent.s4

Input supplies and local support services furnished by the CMB

were inadequate. Between 1970 and 1975, sufficient insecticides were

available to spray only 150,000 hectares, compared with the 1.2 to 1.8

million hectares in cocoa. Ghanaian seed multiplicatian farms had a total

annual production capacity of 1.65 million cocoa pods in 1976, which was

enough to replant only 20,000 ha per year.55

The major reason for these insufficient supplies of inputs was the

gcvernment subsidy, which posed financial problems for the CMB and led to

smuggl1ing. It was estimated that at 1976 import levels the subsidy annually

cost the government NV- 2.4 million for sprayers and W 15.9 million for

insecticides.-~ Despite the announ,ced intention to eliminate all input

subsidies as part of the Economic Recovery Program, the CMB in 1985 imported

sprayers at a landed cost of W14,000 apiece and resold them to farwers for

W3,400.-7 The reason for this ostensibly was to help offset low producer

prices, but the incentives for smuggling under this policy were enormous.

54 World Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vol.III, Annex 7, p.9.

54 World Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vol.Il, Amex 2, pp.9-

TM World Bank, Ghana: Agricultural Sector Review, April 12, 1976,Vol.Ill, Ann-ex 7, p.10.

M' World Bank, Ghana: Towards Structural Adjustment, October 7, 1985,Vol.1, p.48.

119

Credit.

In addition to the informal credit network, agricultural credit in

Ghana is formally provided by a number of institutions, including conmner-ial

banks, the Agricultural Developmnnt Bank (ADB), and the National Investment

Bank (NIB). In addition, the Bank of Ghana initiated a scheme in 1969

whereby it guaranteed twr-thirds of the value of loans granted by

commercial or development banks to small borrowers. Loans by commercial

banks and the NIB have gone primarily to larger farmers and agroindustrial

enterprises. Small farmers have been limited to the ADB, but most of its

loans have gone instead to middle-size farnms.0 Overall, less than 10

percent of farmers in Ghana receive institutional credit.>9

Conditions of credit have varied markedly. Most ccunercial bank

loans have been short- or medium- term, whereas the loans of the NIB have

been almDst exclusively for periods in excess of five years. The ADB in

1972 granted 70 percent of its credit as medium-term loans, 18 percent as

s-hort-term, and 12 percent as long-term. Commercial bank credit was

generally available to agriculture at this time at an interest rate of about

14 percent. Interest at 11 percent was charged on medium-term ccmmercial

bank loans guaranteed by the Bank of Ghana. The NIB and ADB, on the other

hand, provided subsidized credit to agriculture at 6 percent. Since

inflation at the time was ruming at about 10 percent per annum, real rates

of interest to agriculture were negative.

WE In 1968, the average loan of the ADB was for Ni 11,000.H. Mettrick, Policies and Institutions in Ghanaian Pariculture, LUiiversityof Reading, Department of Agricultural Economics and Management, June 1971,p.60.

- World Bank, Ghana: Aaricultural Sector Review, Val.III, Anmex 6,p.l.

120

Repayment rates have varied over time, but the ADB in particular

had problems recovering loans. As a result, the Bank initiated in 1969 a

Commcdity Credit Scheme in which a group of farmers would apply for a loan

in co,m,n in order to invest in the production of a particular commodity.

The loan was guaranteed by all members of the group. The scheme was

designed to reach the small farmer and to assist in the expansion of staple

food cultivation. By 1975, however, only 6.6 percent of all small farmers

in Ghana had been reached by the scheme. Loan recovery rates were about

K0-90 percent.4°

The major problem with rural credit in Ghana has been that high

rates of inflation lowered real rates of interest to negative levels,

especially after 1977 when inflation reached triple digits. This encouraged

misuse of credit and made it extremely difficult for credit institutions to

remain financially viable. The result was a virtual cessation of the flow

of credit resources out of national financial institutions and greater

reliance on rural banks established by the Bank of Ghana beginning in 1976

as private institutions dependent on local savings.62

Research and Extension.

Agricultural research has traditionally been the responsibility of

the Faculty of Agriculture and the Economics Department at the universities

and of several research institutions under the Council for S&ientific and

Industrial Research (ORIS). Research activities related to cocoa and cotton

Ib World Bank, Ghana: APricultural Sector Review, April 12, 1976,vol.III, Annex 6, p. 1 2 .

tb World Bank, Ghana: Agricultural Sector Review, April 12, 1978,vol.III, Annex 6, p.16.

121

come under the Cocoa Marketing Board and Cotton Development Bcard

respectively. In addition, the Crop Production Division of the Ministry of

Agriculture has certain responsibilities with respect to field trials.

Eecause of this fragmentation and lack of coordination of research activity,

thowever, the Ministry of Agriculture has tended to ignore the work of the

universities and research institutes as not being very relevant and has gone

to the other extreme "of relying very heavily on ad hoc introductions of

crop varieties without adequate attention to the need for local, adaptive

research.

The situation is exacerbated by lack of an effective extension

service. During the colonial period the British were primarily interested

in agricultural research and extension as these related to cocoa. By the

1950s, however, a national extension service concerned with a broader range

of crops had been developed within the Ministry of Agriculture. The Nkrimnah

gavernment uprooted this service in 1962 and transferred it to the Uhited

Ghana Farmers'Cooperative Cnoncil (1UGFC), the farmers' wing of the

Convention People' s Party that had previously only been concerned with the

marketing of cocoa. Other extension staff were transferred to state

farms. As Tony Killick has described it,

The results were disastrnLxs . The UIGFC lacked the expertise,administrative capabilities and motivation to operate an extensionservice. Its officials used their positions to cheat the farmers,who became increasingly hostile to the organization which wassupposed to help them. Extension work was further hamstrung bysericus shortages of imported supplies such as machetes,fertilizers and seed. The result was that such extension service

World Bank, The Current Economic Position and Prospnets of Ghana,(ktober 26, 1970, Vol.V: Agriculture, p.11.

122

as had existed at the beginning of the sixties deteriorateddrastically..3

After the Nkrumah era, respansibility for extension was

transferred back to the Crop Producrtian Divisimn of the Ministry of

Agriculture, which was also responsible for the distributicn of seeds and

fertilizers..4 This comnplicated considerably the work of extension agents

and took their attention away from advising farmers on agricultural

tech-iology. In addition, agents were handicapped by inadequate means of

transport and lack of a clear focus for their activities. To bridge the gap

between research and extension, the Grains and Legumes Development Board

embarked on a number of activities, such as demn:stration and foundaticn

seed prcduction, that partly cverlapped with those of the Ministry, creating

additional confusion.

During the late 1960s and early 1970s, an increasing amount of

research and extension took place within the context of particular project

organizations. In 1969, the FAO Fertilizer Program began fertilizer and

rotation trials and extension an maize, cotton, cassava, groundnuts, and

ccwpeas in the Ashanti, Central volta, and Erung Ahafo Regions, which led to

substantial increases in fertilizer utilization. The Ghanaian/German

Agricultural Development Project in the Nbrthern Region undertook research

on rice,-5 and the World Bank financed an important project in the Uipper

Region involving applied research and extension. While these and other

Killick, Development Economics ... , p. 191.

64 World Bank, The Current Economic Position and Prospects of Ghana,October 26, 1970, Vol, V: Agriculture, p. 1 2 .

is World Bank, Ghana: Agricultural Sector Review, April 12, 1978,Vb1.III, Annex1X, p.3.23

123

projects were more successful in instituting adaptive research and in

ccncentrating extension on particular goals, they were also plagued by

shortages and late delivery of inputs, high local currency costs,

insufficient farmer incentives, and rapid turnover of project personnel.

Since the mid-1970s, declining real income and research

facilities have induced a large exodus of trained research staff to other

countries. At the same time, there has been overstaffing at lower levels,

which has consumed budgetary funds at the expense of the money required to

operate research programs. This, coupled with the usual problems of

coordination and extension, has limited the impact of agricultural research

on Ghanaian farming techniques.4-

State Production

As part of its general orientation toaards public ownership and

industrialization, the Nkrumah government under-took a major effort to shift

Ghanaian agriculture away from its predcminant form - small-scale,

cwnaer-operated farms. The State Farms Corporation was created, which

rapidly increased the number of its farms and its total land holdings. By

1965 it was managing 105 farms, about half of which were new, the rest being

former demonstration and experiment stations marnaged by the Ministry of

Agriculture. In addition, the U[FCC was responsible for mechanized

co-operative farms, and the Workers' Brigade operated 10 mec-hanized

farms. ^' "These enterprises affected a very small proportion of the

,1 World Bank, Ghana: Agricultural Sector Review, August 6, 1985,pp.3 4-35.

e' Killick, Development Eccnomics...., p.19 2.

124

population engaged in agriculture and accounted for only about 1 percent of

total agricultural production during the period, but they absorbed large

anmunts of capital and diverted resources, manpower and facilities away from

the smalIholder sector of the economy".19

The state farms, for example, cost Ng 19.8 million in subsidies in

1963-65. Despite the fact that they had the services of many of the MIA s

agricultural officers and had much better access to capital and intermediate

inputs than did small farmers, both their yields and their labor

productivity were only about one-fifth as great. Partly as a result of

this, between 1966 and 1971 the number of state farms was cut to 33.

Nevertheless, the financial deficits of the Corporation still averaged NY

1.4 million in 1969 and 1970. 9 Since then, the State Farms Corporation

has continued in existence, along with other government parastatals involved

with agriculture, such as the Ghana Tobacco Cbmpany, Ghana Rubber Estates

Limited, Ghana Sugar Estates Limited, the State Fishing Corporation, and the

Volta River Authority. Although today they are not the major focus of

gavernment development efforts in agriculture, these enterprises

nevertheless constitute a continuing drain on the public treasury and absorb

a disproportionate share of management time.

tb World Bank, The Current Economic Position and Prospects of Ghana,October 26, 1970, Vol.V: Agriculture, p.13.

9 Killick, Development Economics..., p.1 93-9 4.

125

Conclusions

This discussimn of the history of policy interventions in Ghana

leads to a numrber of conclusions. First, and of greatest importance, the

general deterioration in the economy, accompanied until recently by rising

inflation and an increasingly overvalued currency, had a profound impact on

agricultural incentives. This became especially evident during the last

part of the 1970s and the early 19E0s, when the rate of inflation was in

triple digits and the exchange rate was overvalued by a factor of at least

twenty. Cne result was an increasingly hopeless struggle between

government, farmers, and the Cocoa BDard to extract revenue from the

plunging local currency value of cocoa exports.

On the import substitution side, agriculture benefitted from the

restrictions un grain and other food imports, which caused their domestic

prices to rise steeply in relation to CIF prices converted at the official

rate of exchange. Oh the other hand, this created abundant opportunities

for corruption and other rent-seeking behavior on the part of officials

responsible for food imports. In addition, the governent's frequent

efforts to control domestic prices and to intervene in the food

distribution system led to confusion and marketing disincentives.

Furthermore, the scarcity of foreign exchange that accompanied the

overvalued exchange rate led to a collapse of the transportation system,

which severely disrupted the transmission of higher prices to farmers for

both tradable and nontradable foods.

Second, the disenfranchisement of Ghana s cocoa farmers and the

penetration of the LEFOC into the cocoa growing areas urnder the Nkrumah

regime led to a system for allocating resources that depended on favoritism

126

and bribery rather than on the exertion of political pressure and influence.

Suaceeding regimes attempted at first to correct this, but their narrow

political basis and their macroeconomic managerial incompetence soon led to

a reversion to the old system under the auspices of the Produce Buying

Ccmpany's at first de facto, and later de jure, monopoly on cocoa purchases.

The main difference between the LUFOC mrnopoly and that of the PBC was that

the former had strong political overtones related to the desire of the CFP

to neutralize the cocoa farmers, whereas the latter was ideologically

neutral and was designed simply to keep the ever dwindling supplies of cocoa

moving as the government wrestled to maintain its share of sales revenue.

The end result, however, was the same - resource allocation based on

influence and corruption rather than an political process. As in the import

competing sector, the result was wasteful rent-seeking rather than

productive activity.

Third, the bias against tradable agriculture resulting from trade,

price, and exchange rate policies was only to a very minor extent offset by

input subsidies and cheap credit. These were often channeled through rural

development projects in which officials developed strong vested interests,

abetted in many cases by foreign donors. The projects were a weak base on

which to develop the agricultural sector, however, because they required

high levels of managerial skill and were vulnerable to the price distortions

that increased in intensity with the collapse of the macroeccnoimy. In the

end, input subsidies and cheap credit never effectively reached the smaller

farmers who were responsible for most of the nation s agricultural output.

The other alternative to small farmers - the state farms and

agricultural parastatals established under Nkrumah - proved to be a

127

disaster. Productivity was lower than on private farms, and yet these

public sector activities continued to exert a disproportionate claim on

managerial and financial resources even after Nkrumah.

With the advent of the Economic RecoverY Program, steps were taken

to improve the incentive structure. The cocoa pr-oducer price was increased

significantly in real terms for the first time in 1986. Efforts were made

to improve the transportation system and to alleviate some of the problems

of input supply. Nevertheless, years of neglect had exacted their toll, and

it was clear that considerable time would be necessary before the

agricultural sector could be fully revived.

128

PART 11T: EFFEC`TS OF PRIC INTERVENTIONS

OHPTER VI: fEfFES OF INTEFENTION

In order to quantitatively measure the effects on incentives of

governwent policy, a number of indicators have been calculated. These

pertain both to the direct effects an relative prices of trade and price

policy related to agricultural outputs and inputs and to the indirect

effects of distortions in the economy that result in a deviation of the

equilibrium from the official exchange rate. This chapter discusses the

methods used in calculating the indicators of price distortions and presents

the results. Subsequent chapters examine the effects of these distortions

an production, consumption, foreign exchange f low3s, gcverrnent revenue,

producer and consumer welfare, and producer incoom.

Direct Effects

The direct effect of policy occurs as a result of taxes and

subsidies on trade, quantitative import restrictions, and domestic price

policies. Import taxes and restrictions act to increase domestic prices in

relation to those an the wDrld market; export taxes and restrictions depress

domestic prices relative to thDse at the border. In addition, price policy

has also been important, especially for cocoa, in determining farmer

incentives.

Producer, Consumer, and Border Prices

To measure these incentives, data were gathered on the domestic

and border prices of six major agricultural ccomsodities. These data are

presented and the adjustments made to them are detailed in Annex 3. Of

129

these six crops, three are tradable and three are ncontradable. Of the

tradable crops, cocoa was chosen because it is by far the major export crap

and its fortunes are closely linked with the rest of the economy. Rice has

been the major imported food, with imports in many years exceeding local

production. Maize is also an imported food, but imports are smaller and

production and consumptim are much more important than for rice. In

additim to these tradable crops, three major nontradable foods are

included in the analysis of this chapter to see to what extent mEvements in

the domestic price of the tradable foods, rice and maize, have been

correlated with price movements for the nontradables. Since these

nontradables are not directly influenced by governoent policy, however, they

are not included in the analysis of subsequent chapters. Of the naitradable

crops, millet and sorghum, are the major foods produced and consumed in the

north. These two cereals are closely related in production and consumption,

and sorghum alcne is included here. In the south, the rot crops, cassava

and yams, are important nontradable food crops for which relatively long

price series exist. Since experience with these crops has differed

somewhat, both are retained in the analysis.

Domestic prices for most of the crops shown - maize, rice,

sorghum, cassava, and yams - are based m monthly surveys of wholesale

markets in Accra, Kumasi, and Tamale. The prices are those actually

observed rather than official minima or maxima. As such they fluctuate

relatively freely. Adjustments to arrive at producer prices were made on

the basis of detailed data obtained for 1975 on transport, storage,

handling, and other costs involved in moving aommodities from farm to

wholesale market. The 1975 figures were then adjusted to other years using

130

the overall CcLnumer Price Index- This matched fairly well with

fragmentary data on these costs in other years. L Retail prices were

estimated by adjusting wholesale prices for data cn average percentage

markups available for a few years.

For cocoa, the official producer price was used at the farm level.

To this were added puiblished Cocoa Marketing Board costs of collection and

delivery to the port of Tema in order to obtain the domestic price of cocoa

at the border. As discussed in the previous chapter, these costs increased

markedly over the years as a result of swelling employment within the CM.

In addition, the distinction between the cMB and the central government has

often been fairly arbitrary as input delivery, extension, and research

activities have been shifted back and forth between the ministries and the

Board. It is impossible, however, to dissociate over time purely marketing

costs from the other elements in the CMB's budget. Cbnsequently, the

domestic price of cocoa at the border in most years is overestimated in

comparison with what it wculd be if it included only actual marketing costs.

Border prices were estimated on the hasis of unit values

calculated from trade data, with some corrections, interpolations, and

extrapolations based on changes in world market prices where unit value data

are missing or clearly in error. Pbrt handling charges available for 1975

were adjusted to other years using the CPI in order to obtain the whDlesale

price equivalent of the border price. Retail margins similar to those used

± Although transpDrt costs tended to rise somewhat relative to theCPI, for reasons explained earlier in Chapter V, the cost of services,including those involved in marketing, had a tendency to fall. In addition,food purchased in urban areas is heavily weighted in the CPI, implying thatany increases in the cost of transport and marketing were largely reflertedin the CPI.

131

for domestic prices were applied to estimate retail price equivalents to

consumers. The same collection, procressing, and distribution costs used

for donestic prices were subtracted frcmN the wholesale price equivalents to

obtain producer price equivalents.

Tables 10 and 11, derived from PInnex 3, show the sharp increases

in food prices that have occurred in Ghana, especially since 1975.

Mbvements in relative prices are discussed later, but for now it is useful

to point to the growth in the margin between producer and ccrnsumer prices

that has occurred. The best example is cocoa, which was discussed in the

previcous chapter. Although increases in this margin were less important

for other crops because, unlike cocoa, they were subject to competitive

marketing, these increases were significant nonetheless. This especially

true of the starchy staples for which transpDrtaticn costs are important in

relation to the value of output. Yam producers, for example, rereived 70

percent of the price to consumers in 1953 but crly 56 percent in 1984.

Maize producers received 69 percent of the retail price in 1953 and 60

percent in 1984. One result of these rising costs is that producer price

equivalents of border prices for rice and maize, shown in Table 12, are

negative in the later years of the period studied as a result of

subtracting high marketing costs frnm low border prices measured at the

overvalued official exchange rate. This is also true of cocoa in 1981, when

CMB costs were in excess of the FEO price converted to cedis at the official

exchange rate.

132

Table 10

Domestic Producer PricesINC per Metric Ton)

Year Maize (a) Rice (a) Sorghum (a) Cassava (b) Yam (bl Cocoa Ib)

1953 46 130 61 N/A 54 2691954 37 118 72 N/A 49 2871955 34 99 48 N/4 52 2991956 44 100 49 N/A 56 2801957 25 101 59 N/A 52 2691958 39 100 70 N/A 58 2391959 29 86 70 N/A 54 2241960 28 69 38 NIA 54 2241961 47 63 55 N/A 53 2241962 44 54 79 N/A 57 2201963 50 137 85 15 50 2021964 66 97 99 14 52 1871965 86 135 114 29 63 1871966 85 128 113 23 70 2241967 37 125 83 15 60 2541968 73 116 83 14 63 2841969 119 160 134 22 71 2931970 87 185 133 20 72 2931971 87 186 138 31 89 2931972 137 275 159 38 95 3661973 142 332 235 34 117 4391974 151 355 243' 41 154 4891975 185 497 230 72 213 5851976 469 1109 659 181 302 7321977 970 1355 1477 486 622 13331978 834 1611 1430 319 1007 26671979 1134 1399 1558 261 1160 40001980 3261 5777 5654 848 1735 40001981 5845 6545 7035 2274 2791 120001982 5659 15530 10038 2423 4148 120001983 31537 45165 31239 10935 11436 200001984 16118 61571 37872 3721 11716 300001985 11819 28453 N/A 1742 62104 56600

…----Notes to Table 10:

(a) Data from Annex 3, Table 3-3(1).(b) Data froa Annex 3, Table 3-3(2).

133

Table 11

Domestic Consumer Prices(NC per Metric Ton)

Year Maize (a) Rice (a) Sorghum (a) Cassava (b) Ya (Ib) CNoa (bI

1953 67 190 09 N/A 77 3011954 57 176 102 N/A 70 3201955 54 155 74 N/A 75 3351956 66 158 77 N/A s0 3171957 44 159 89 N/A 75 3061958 61 158 101 N/A 82 2771959 49 143 101 N/A 77 2601960 49 124 65 N/A 78 2591961 71 121 86 N/A 78 2591962 70 115 115 N/A 83 2591963 77 211 124 34 76 2451964 98 174 143 35 80 2291965 126 233 169 58 97 2301966 129 236 172 53 109 2711967 71 224 134 41 95 3001968 115 221 136 42 101 3331969 170 278 198 53 112 3381970 135 310 198 52 114 3721971 138 320 209 67 137 2811972 199 433 238 79 147 5121973 212 519 334 90 190 6501974 231 571 356 95 231 6031975 287 782 t3 14 315 9821976 656 1604 912 311 456 10681977 1368 2269 2032 796 949 24041978 1396 3083 2223 733 1559 41591979 1979 3509 2685 854 1934 61201980 4763 9495 7842 1931 2895 75731981 8914 13701 10992 4489 5109 172001982 9190 25409 15092 5060 7131 230001983 42245 68767 43833 17920 18520 470001984 26885 94313 54600 11167 20761 740241985 22790 58617 N/A 9474 91908 112016

Notes to Table 11:(a) Data from Annex 3, Table 3-3(1).Ib) Data from Annex 3, Table 3-3(2).

134

Table 12

Border Price Equivalents(NC Per Metric Ton)

Rice (a) Maize (b) Cocoa (c)

Producer Retail Producer Retail Producer Retail

Price Price Price Price Price PriceYear Equivalent Equivalent Equivalent Equivalent Equivalent Equivalent

1954 131 191 67 92 604 6371955 106 164 103 133 400 4361956 94 151 97 127 354 3911957 101 160 114 146 595 6321959 100 159 90 119 512 5501959 85 143 66 92 403 4391960 76 133 93 123 307 3421961 70 128 103 135 293 3181962 63 125 80 111 298 3371963 67 132 24 47 314 3571964 47 116 69 102 236 2781965 59 146 29 61 219 2621966 62 159 24 59 349 3961967 90 184 29 62 516 5621968 153 264 113 161 712 7611969 140 255 115 166 773 8191970 112 225 18 55 598 6431971 88 206 50 95 600 6881972 111 244 367 463 678 8241973 218 388 42 97 1083 12941974 411 634 131 208 1374 168B1975 209 450 119 212 1129 15261976 237 600 104 236 2260 25961977 -118 575 234 521 2871 39421978 -305 890 384 879 8904 103961979 -475 1354 1014 1841 7000 91201980 -931 1781 312 1372 2727 63001981 -3488 2163 -321 1823 -200 50001982 -4762 2074 -881 1669 19558 305581983 -12081 2933 -2331 3296 33000 600001984 -3639 19322 4415 13427 39299 833231985 -749 25034 6266 16404 65567 120983

Notes to Table 12:(a) Data from Annex 3, Table 3-4(1).(b) Data from Annex 3, Table 3-4(2).(c) Data from Annex 3, Table 3-4(3).

Retail Price Equivalent is theFOB price for cocoa.

135

Relative Prices

Domestic price ratios of these crops were calculateJ at the

producer and conruiier levels fron the prices presented in Tables 10, 11, and

12. In addition to the ratios of agricultural prodcit prices, the ratios of

the price of each product to the nonagricultural consumer price index was

also calculated. The derivation of this index is explained in the footnotes

to Table 3-5(1) in Anex 3. It is the mDst relevant index to compare with

agricultural prices fron both the producer and consumer perspectives, in

comparison with the GDP deflator, which measures changes in value added

prices.

The price ratios are presented in Tables 13(1) - 13(4) in the form

of indices calculated with 1972 as a base yar. This facilitates the

comparison of trends between the various series. The indices are presented

at both producer and consumer levels.

It is evident from the tables that, relative to the price of

nonagricultural goods and services, foodcrop prices have fluctuated markedly

and have shown a tendency to increase somewhat over time, especial ly since

about 1974. Cocoa prices, an the other hand, fell markedly to farmers

during the late 1950s and early 1960s, when producer prices dropped by 56

percent of their average level in 1953-55. Following this there was a

modest improvement, followed by fluctuaticns arcound a level that was

approximately me-half that experienced during the early 1950s.

Looking at the price of rice, the principal tradable crop,

relative to the prices of ncntradable foods, there are marked changes from

year to year but no clear overall trend. This suggests that restricticns on

rice imports have been sufficiently important that the overvalued exchange

136

Table 13(l)

Prevailing Relative Price Indices (a)(1972 100)

Maize/Non-Ag Rice/Nan-Ag Sorghum/Non-Ag

Producer Retail Producer Retail Producer RetailDate Price Price Price Price Price Price

1953 81 81 114 106 93 901954 69 73 109 103 115 1091955 62 67 90 89 75 781956 79 81 89 89 75 791957 42 52 86 87 88 881958 67 71 84 85 102 991959 49 58 73 77 102 991960 47 56 57 65 54 621961 78 81 52 63 78 811962 69 74 42 56 105 1031963 68 72 92 91 99 961964 83 84 60 68 105 1021965 94 96 73 81 107 1061966 84 88 63 74 96 981967 36 48 60 69 70 751968 66 71 52 63 64 711969 101 100 68 75 98 971970 73 78 77 82 95 951971 69 75 74 80 94 951972 100 100 100 100 100 1001973 92 94 106 106 130 1241974 79 84 93 94 109 1071975 76 82 102 102 81 861976 143 137 168 154 172 1591977 203 197 141 150 264 2441978 97 112 93 114 143 1491979 87 105 54 85 103 1191990 172 173 152 158 256 2381981 140 147 78 104 145 1511982 119 133 162 169 181 1821983 338 312 240 233 287 2691984 96 111 183 178 194 1871985 55 73 65 86 N/A N/A

Notes to Table 13(1):(a) Based on crop prices from Tables 10 and 11 and the Non-Agricultural

Consuser Price Index from Annex 3, Table 3-S11).

137

Table 13(2)

Prevailing Relative Price Indices la)(1972 100)

Cassava/Non-Ag Yam/Non-Ag Cocoa/Non-Ag

Producer Retail iroducer Retail Producer Retail

Date Price Price Price Price Price Price

1953 N/A N/A 136 126 179 1421954 N/A N/A 131 121 199 159

1955 N/A N/A 137 126 203 163

1956 N/A N/A 144 132 187 151

1957 N/A N/A 129 119 173 141

1958 N/A N/A 142 130 152 126

1959 N/A N/A 132 122 143 119

1960 N/A N/A 129 120 139 115

1961 N/A N/A 1Z6 119 137 113

1962 N/A N/A 128 120 129 107

1963 73 79 98 95 102 89

1964 63 75 93 92 97 76

1965 115 109 99 99 77 67

1966 e1 90 99 100 83 72

1967 51 69 54 86 92 78

1968 46 66 53 85 96 s0

1969 66 78 88 89 93 77

1970 60 74 87 89 91 83

1971 89 92 102 101 87 60

1972 100 100 100 0oo 100 1oo1973 79 89 109 109 106 112

1974 76 86 116 112 96 112

1975 106 104 126 121 90 108

1976 196 163 132 129 83 87

1977 362 283 187 184 104 134

1978 133 148 169 169 116 130

1979 72 114 129 139 115 126

1990 160 167 132 142 79 107

1991 195 186 96 114 108 110

1982 182 184 126 139 94 1291913 419 331 176 184 80 134

1914 79 IS 101 115 67 1181 91 29 76 414 352 98 138

Notes to Table 13(2):(a) Based on crop prices from Tables 10 and 11 and the Non-Agricultural

Consumer Price Index from Annex 3, Table 3-5(l).

138

Table 13(3)

Prevailing Relative Price Indices (al(1972:100

Rice/Cassava Rice/Sorghum Rice/Yams

Producer Retail Producer Retail Producer RetailDate Price Price Price Price Price Price

1953 N/A N/A 123 128 83 911954 N/A N/A 95 100 83 901955 N/A N/A 120 115 66 711956 N/A N/A 119 112 62 671957 N/A N/A 99 98 67 721958 N/A N/A 93 56 59 651959 N/A N/A 71 73 55 591960 N/A N/A 105 91 44 471961 N/A N/A 66 64 41 431962 N/A N/A 40 41 33 351963 126 115 93 95 94 971964 95 91 57 59 65 651965 64 74 68 69 74 741966 79 82 66 65 64 631967 119 101 87 81 72 701968 114 96 81 74 63 611969 103 96 69 70 78 761970 129 110 91 81 89 871971 84 97 78 77 72 731972 100 100 100 100 100 1001973 134 119 82 86 98 991974 122 110 85 87 90 821975 9b 99 125 119 81 841976 85 94 98 105 127 1301977 39 53 53 59 75 761979 70 77 65 63 55 551979 75 75 52 45 42 391980 95 95 59 64 115 1071991 40 56 54 52 81 691992 89 92 90 89 129 1161983 58 70 84 89 136 1301984 230 155 94 98 181 1591985 227 113 N/A N/A 16 19

Notes to Table 13(3).(a) Based on crop prices tram Tables 10 and 11.

139

Table 13(4)

Prevailing Relative Price Indices (a)(1972=100)

Rice/Maize Rice/Cocoa (a) Yams/Cocoa (a)

Producer Retail Producer Retail Producer RetailDate Price Price Price Price Price Price

1953 140 131 64 75 77 891954 159 142 55 65 66 761955 145 133 44 55 67 771956 113 110 48 59 77 871957 204 167 50 62 74 851958 126 120 55 68 93 1031959 147 134 51 65 92 1031960 120 117 41 57 93 1041961 66 77 38 55 92 1051962 61 76 33 52 100 1121963 135 125 90 102 96 1071964 73 a1 69 90 107 1211965 78 84 96 120 129 1471966 75 84 76 103 120 1401967 167 144 65 88 91 1101968 79 88 54 79 86 1061969 67 75 73 97 94 1161970 105 105 84 98 95 1071971 106 106 84 135 117 1691972 100 100 100 100 100 1001973 116 112 101 94 103 961974 117 113 97 84 121 1001975 134 125 113 94 140 1121976 117 112 202 178 159 1491977 69 76 135 112 190 1371978 96 101 80 88 146 1311979 61 81 47 68 112 1101980 s8 91 192 148 167 1331981 56 70 73 94 90 1031982 136 127 172 131 133 1081983 71 75 301 173 221 1371914 190 161 273 151 151 981995 120 118 67 62 423 255

Notes to Table 13(4):(a) Based on crop prices fros Tables 10 and 11.

140

rate has not resulted in cheap food imports flooding the Ghanaian market

for any sustained periods of time. On the other hand, there have been

shorter periods during which the price of rice has been low in relation to

the prices of ncntradable foods. The first of these was fromu 1964 until

about 1967. The data an rice availability (Annex 6) do not indicate any

increase in dcmestic production or imports over previouis years, but the

price series in Tables 10 and 11 suggest that there was a fairly sharp rise

in sorghum and cassava prices, especially in 1965 and 1966, which may have

been at least partly due to a moderate decline in prockicticn of these crops

in 1965. This, of ccurse, was a critical period for the Nkrumah regime.

What is important is not that government policy regarding food imports

changed drastically at this time but that it did not change in response to

the domestic pressures created by inflation and poor harvests.

During the next few years, the price of rice rose quite sharply in

relation to the prices of nentradable foods. Again, this was not due to

production shortfalls or import ccnstraints since net availability of rice

increased fram an average of 56 tans per years in 1964-67 to 70 tons per

year in 1968-74. Rather these were years of relatively good harvests,

which caused the prices of nontradable foods to decline in real, if not

nominal, terms. They were also years during which the difference between

the black market or equilibrium exchange rate and the official rate

decreased, which tended to increase the price of tradables relative to that

of nontradables.

The succeeding decade was characterized by a reversal in these

price movements. With generally poor harvests and an increasingly

overvalued exchange rate, the domestic price of rice fell not mnly in

141

relatim to the prices of sorghum and cassava but also to that of maize.

In fact, niany of the movements in the price of rice relative to those of

millet and sarghum have been similar to muvements of the prices of rice and

maize. This is because maize was until recently traded internationally by

Ghana only in small quantities and had a relatively high cross-price

elasticity of demand with the other coarse grains. After 1981 maize

imports increased substantially and over the next four years equaled 20

percent or more of total availability (Annex 6).

The relationship between the price of rice relative to the price

of yams has been scnewhat different. Although there have been year to

fluctuations, the broader movements that have characterized the price of

rice relative to that of sorghtum, cassava, and maize have not been so

apparent with yams. If anything the price of rice might have risen in

recent years relative to that of yams. The reascns for this are not clear

since the productim data show no strcng increase in the output of yams.

What is clear is that the prices of rice and yams have risen stroigly

relative to the price of cocoa. This is especially evident after

1974. Since the prices of maize and the nmtradable foods have tended to

rise relative to those of rice and yams, cocoa prices have fallen to an even

greater extent in comparism with these crops.

Ptasurement of Direct Price Interventims.

Tables 14 and 15 present a series of ratios, measured at both the

producer and ccnsumer levels, for domestic and border prices. The dcnestic

price ratios, measured in cedis per metric ton, are essentially the same as

those used to construct the indexes of Table 13, and there is no need to

142

Table 14

Producer Price Ratios la)

RicelNon-Aq cKoa/Non-Ag Raise/Non-Ag Rice/Sorghum Rice/Yam Ricelmtaize Cocoa/Rite Cocoalflaize

Domestic (a) Border (lii Domestic (a) Border (bi DOmestic (a) Border (bi Domestic (c) Border (di Domestic (ci Border (di Domestic (ci Border (di Domestic (ci Border (di Domestic (ci Border (d)

1954 2.99 3.32 7.29 15.34 .93 1.71 1.63 1.91 2.42 2.69 3.20 1.94 2.44 4.62. 7.90 9.9

1955 2.46 2.64 7.45 9.96 .94 2.57 2.07 2.22 1.99 2.03 2.93 1.03 3.03 3.77 9.99 3.99

1956 2.44 2.29 6.94 9.64 1.07 2.36 2.02 1.90 1.79 1.67 2.29 .97 2.90 3.79 6.40 3.66

195 2.36 2.39 6.33 14.00 .59 2.67 1.69 1.71 1.94 1.95 4.10 .99 2.69 5.97 10.97 5.23

195 2.32 2.34 5.57 11.94 .91 2.09 1.43 1.44 1.72 1.73 2.54 1.12 2.40 5.10 6.10 5.71

1959 2.00 1.9 5.22 9.39 .67 1.55 1.23 1.23 1.60 1.59 2.97 1.29 2.61 4.72 7.75 6.06

MO6 1.56 1.73 5.09 6.99 .65 2.12 1.91 2.00 1.27 1.41 2.41 .91 3.27 4.04 7.99 3.29

196 1.42 1.57 5.02 6.35 1.06 2.31 1.14 1.26 1.19 1.31 1.34 .69 3.54 4.05 4.73 2.75

2962 1.15 1.34 4.67 6.32 .94 1.70 .69 .90 .95 1.11 1.22 .79 4.07 4.72 4.97 3.72

196 2.53 1.25 3.75 5.93 .93 .44 1.60 .79 2.73 1.35 2.71 2.95 1.49 4.65 4.01 13.29

196 1.66 .92 3.19 4.01 1.13 1.19 99 .40 1.99 .92 1.47 .69 1.92 4.IM 2.92 3.41

1965 2.02 .99 2.90 3.29 1.29 .44 1.19 .52 2.15 .94 1.57 2.03 1.39 3.70 2.19 7.51

1966 1.74 .93 3.03 4.72 1.25 .33 1.13 .55 1.94 .89 1.52 2.55 1.74 5.65 2.64 14.42

2967 1.66 1.20 3.39 6.97 .49 .39 1.50 1.09 2.09 1.50 3.36 3.14 2.04 5.75 6.95 19.04

1969 1.44 1.90 3.51 9.90 .90 1.40 1.40 1.95 1.93 2.42 1.60 1.35 7.45 4.64 3.99 6.29

1969 1.97 1.64 3.42 9.02 1.39 1.34 1.29 1.05 2.25 1.97 1.35 1.22 1.93 5.51 2.47 6.71

1970 2.12 1.29 3.35 6.93 1.00 .20 1.39 .84 2.57 1.55 2.12 6.35 1.59 5.34 3.36 33.93

1972 2.02 .95 3.19 6.53 .95 .54 1.34 .63 2.10 .99 2.14 1.75 2.59 6.95 3.37 12.03

1972 2.75 1.11 3.66 6.79 1.37 3.67 1.73 .69 2.90 1.17 2.01 .30 1.33 6.13 2.69 1.95

1973 2.92 1.92 3.87 9.54 1.25 .37 1.41 .93 2.93 1.96 2.33 5.20 1.32 4.97 3.09 25.93

1974 2.55 2.94 3.50 9.94 1.09 .94 1.46 1.69 2.31 2.67 2.35 3.13 1.39 3.35 3.23 10.47

1975 2.90 1.17 3.29 6.36 1.04 .67 2.16 .91 2.34 .99 2.69 1.75 1.19 5.43 3.16 .9.41

p. 1976 4.61 .9B 3.04 9.39 1.95 .43 1.69 .36 3.67 .79 2.37 2.29 .66 9.56 1.56 21.92

1977 3.97 -.34 3.90 9.19 2.77 .67 .92 -.09 2.19 -.19 1.40 -.50 .9 -24.33 1.37 12.29

1979 2.57 -.49 4.26 14.21 1.3.3 .61 1.13 -.21 1.60 -.30 1.93 -.79 1.66 -29.23 3.20 23.19

1979 1.47 -.50 4.21 7.37 1.19 1.07 .90 -.30 1.21 -.41 1.23 -.47 2.96 -24.74 3.53 6.90

19910 4.17 -.67 2.99 1.97 2.35 .23 1.02 -.16 3.33 -54 1.77 -2.99 .69 -2.93 1.23 9.74

1991 2.15 -1.14 3.94 -.07 1.9 -.11 .93 -. 50 2.35 -1.25 1.12 10.95 1.93 .06 2.05 .62

1992 4.46 -1.37 3.45 5.62 1.63 -.25 1.55 -. 47 3.74 -1.15 2.74 5.40 .77 -4.11 2.12 -22.20

1993 6.61 -1.77 2.93 4.93 4.62 -.34 1.45 -.39 3.95 -1.06 1.43 5.19 .44 -2.73 .63 -14.16

1994 5.03 -.30 2.45 3.21 1.32 .36 1.63 -. 10 5.26 -.31 3.92 -.92 .49 -10.90 1.96 9.90

1995 1.90 -.05 3.59 4.15 .75 .40 N/A N/A .46 -.01 2.41 -.12 1.99 -97.51 4.79 10.46

Notes to Table 14:(a) Domestic price lrom Annex 3, Table 3-3(11 for rice and maize and from Table 3-3i2i

for cocoa, deflated by the Nonagricultural Indexr from Table 3-5(1).(hi Producer Price Equivalent (DER) from Table 3-4(1) for rice, Table 3-4(2) for maize,

and Table 3-4(3) for cocoa, deflated by the Nonagricultural Index from Table 3-5M1.

(c) Ratio of producer prices from Table 3-3(1) and Table 3-3(21.(d) Ratio of Producer Price Equivalents from Table 3-4(2), Table 3-4(21, and Table 3-4(3),

except for sorghum and yams for which the producer prices from Table 3-3(1) andTable 3-3(2) are used.

Table 15

Consuwr Price Ratios

Rice/Non-Ag Cocoa/Non-Ag Naize/Non-Ag Rice/Sorqhus Rice/Yam Rice/Kaize Cocoa/Rice Cocoa/Naize

Dmestic (a) Border (bI Domestic (a) Border bI) Domestic la) Border (b) Domestic (c) Border id) Domestic (c) Border (d) Domestic (c) Border (d) Domestic (c) Border (dl Domestic (c) Border (dl

1954 4.47 4.05 9.13 16.19 1.44 2.33 1.73 1.87 2.52 2.73 3.10 2.09 1.82 3.33 5,63 6.941I55 3.87 4.07 8.34 10.86 1.33 3.32 2.09 2.20 2.09 2.19 2.90 1.23 2.16 2.67 6.26 3.271956 3.86 3.69 7.74 9.55 1.60 3.09 2.05 1.96 I.99 1.90 2.41 1.19 2.00 2.59 4.92 3.091957 3.75 3.77 7.20 14.87 1.03 3.44 1.79 1.90 2.13 2.15 3.64 1.10 1.92 3.95 7.00 4.32195 3.69 3.71 6.46 12.82 1.41 2.77 1.57 1.59 1.93 1.94 2.61 1.34 1.75 3.46 4.57 4.641959 3.34 3.33 6.06 10.23 1.15 2.15 1.42 1.41 1.96 1.85 2.92 1.55 1.91 3.07 5.29 4.751960 2.82 3.01 5.88 7.77 1.11 2.91 1.91 2.04 1.60 1.71 2.54 1.07 2.09 2.58 5.31 2.77Itl 2.71 2.89 5.81 7.13 1.60 3.04 1.40 1.49 1.55 1.65 1.69 .95 2.15 2.48 3.63 2.351962 2.43 2.65 5.50 7.15 1.49 2.35 .99 1.09 1.37 1.50 1.65 1.13 2.26 2.69 3.72 3.05196.3 3.92 2.45 4.55 6.62 1.43 .96 1.71 1.07 2.90 1.75 2.74 2.84 1.16 2.70 3.17 7.671964 2.95 1.99 3.B9 4.72 1.67 1.73 1.22 .82 2.19 1.46 1.77 1.15 1.32 2.39 2.33 2.741965 3.49 2.19 3.45 3.93 1.90 .92 1.39 .87 2.39 1.50 1.94 2.39 .n 1.79 1.82 4.28196 3.19 2.16 3.66 5.35 1.74 .80 1.38 .93 2.17 1.46 1.93 2.69 1.15 2.48 2.10 6.691967 2.99 2.45 3.99 7.49 .95 .82 1.68 1.39 2.37 1.94 3.14 2.9 1.34 3.05 4.20 9.111968 2.73 3.26 4.12 9.40 1.42 2.00 1.62 1.94 2.19 2.61 1.93 1.64 1.51 2.88 2.90 4.71

H 196 3.25 2.99 3.95 9.55 1.99 1.94 1.40 1.29 2.48 2.27 1.64 1.54 1.22 3.21 1.99 4.93Xs 1970 3.54 2.5B 4.25 7.35 1.54 .62 1.56 1.14 2.71 1.98 2.30 4.13 1.20 2.85 2.76 11.79

1971 3.48 2.25 4.14 7.48 1.50 1.03 1.53 .99 2.34 1.51 2.32 2.17 1.19 3.33 2.77 7.251972 4.33 2.44 5.12 8.24 1.98 4.63 1.82 1.02 2.94 1.66 2.18 .53 1.18 3.39 2.59 1.781973 4.57 3.42 5.73 11.40 1.87 .85 1.55 1.16 2.99 2.16 2.44 4.00 1.25 3.34 3.06 13.351974 4.09 4.54 5.75 12.09 1.66 1.49 1.60 1.78 2.47 2.75 2.47 3.04 1.41 2.66 3.47 8.101975 4.41 2.53 5.53 8.59 1.62 1.19 2.15 1.24 2.48 1.43 2.72 2.13 1.26 3.39 3.42 7.211976 6.67 2.50 4.44 10.79 2.73 .98 1.76 .66 3.52 1.32 2.45 2.55 .67 4.32 1.63 11.011977 6.47 1.64 6.86 11.25 3.90 1.49 1.12 .28 2.39 .61 1.66 1.10 1.06 6.95 1.76 7.561979 4.92 1.40 6.64 16.59 2.23 1.40 1.39 .40 1.98 .56 2.21 1.00 1.35 11.81 2.99 11.831979 3.70 1.43 6.45 9.61 2.08 1.94 1.31 .50 1t81 J7 1L77 .74 1.74 6.74 3.09 4.95IS90 6.86 1.29 5.47 4.55 3.44 .99 1.21 .23 3.28 .62 1.99 1.30 .80 3.54 1.5S 4.591981 4.49 .71 5.64 1.64 2.92 .60 1.25 .20 2.68 .42 1.54 1.19 1.26 2.31 1.93 2.741982 7.30 .60 6.61 8.78 2.64 .4B 1.68 .14 3.56 .29 2.76 1.24 .91 14.73 2.50 18.311983 10.07 .43 6.88 8.78 6.1B .48 1.57 .07 3.71 .16 1.63 .89 .68 20.46 1.11 18.201984 7.71 1.58 6.05 6.81 2.20 1.10 1.73 .35 4.54 .93 3.51 1.44 .78 4.31 2.75 6.211985 3.71 1.59 7.08 7.65 1.44 1.04 N/A N/A .72 .31 2.57 1.53 1.91 4.83 4.92 7.38

Notes to Table 15:(al Domstic price from Annex 3, Table 3-3(11 for rice and maize and the Rendered Port

Price from Table 3-3(21 for cocoa, deflated by the Nonagricultural Index fromfrom Table 3-5M11.

(bI Retail Price Equivalent (DER) from Table 3-4(11 for rice, Table 3-4(2) for maize,mad FOB price from Table 3-413) for cocoa deflated by the Nonagricultural Index fromTable 3-5(1M.

Cc) Ratio of retail prices from Table 3-3(1) and Table 3-3(2).(d) Ratio of Rttail Price Equivalents froo Table 3-4(11, Table 3-4(2). and Table 3-4(31,

except for sorghum and yams for which the retail prices from Table 3-3(1) andTable 3-3(2) are used.

discuss them further here. It is useful, however, to offer a few comments

concerning the bDrder price ratios.

First of all, it is very evident that the border prices of

tradable products have declined relative to the nonagricultural CPI and to

the domestic prices of nontradable foods. This is principally because of

the overvaluation of the exchange rate. Where producer bDrder price

equivalents from Table 12 are negative because high transport and marketing

costs are greater than prices at the border converted to local currency at

the official rate of exchange, the ratios are of course negative as well.

Looking at relative border prices of the tradable crops, it

appears that, despite year to year fluctuations, there is no clear trend in

the price of rice compared with that of maize. There is also no obviaus

trend in the terms of trade between cocoa and the traded cereals, thowgh any

such trend might be overwhelmed by wide price fluctuaticns, particularly in

recent years.

In Tables 16 and 17, the difference between the dcmestic price

ratio and the border price ratio, shown in Tables 14 and 15, is divided by

the border price ratio. This is equivalent to the ratio of the nominal

rates of protection (NRP) of the tWiD products or of the product and the

index as presented. Where the nonagricultural CPI is the denominator of the

price ratio, however, the figure shown is simply the NRP of the crop

indicated.

Table 16 suggests that direct price interventions, in this case

principally trade and exchange controls for rice and maize and the producer

price established for cocoa, have in most instances caused the domestic

price to the procnier to be less than its border price equivalent. While

145

Table 16

Effect of Direct Price Interventions onRelative Producer Price Differences (a)

…----- ----- ------ ----- ------ -----

Rice/Non-Ag Cocoa/Non-Ag Maize/Non-Ag Rice/Maize Cocoa/Rice Cocoa/Maize

1954 -.10 -.52 -.45 .65 -.47 -.131955 -.07 -.25 -.67 1.85 -.20 1.291956 .07 -.21 -.55 1.36 -.26 .751957 -.01 -.55 -.78 3.60 -.54 1.101958 -.01 -.53 -.56 1.27 -.53 .071959 .00 -.44 -.57 1.31 -.45 .281960 -.10 -.27 -.70 1.96 -.19 1.401961 -.09 -.21 -.54 .97 -.13 .721962 -.14 -.26 -.45 .55 -.14 .331963 1.02 -.36 1.13 -.05 -.68 -.701964 1.05 -.21 -.04 1.14 -.61 -.171965 1.28 -.15 1.94 -.23 -.62 -.711966 1.08 -.36 2.50 -.41 -.69 -.821967 .39 -.51 .30 .07 -.65 -.621968 -.24 -.60 -.36 .18 -.47 -.381969 .14 -.62 .03 .11 -.67 -.631970 .65 -. 51 3.95 -.67 -.70 -.901971 1.12 -.51 .74 .22 -.77 -.721972 1.48 -.46 -.63 5.68 -.78 .451973 .52 -.59 2.39 -.55 -.73 -.881974 -.13 -.64 .15 -.25 -.59 -.691975 1.39 -.48 .55 .54 -.78 -.671976 3.69 -.68 3.53 .04 -.93 -.931977 -12.48 -.54 3.15 -3.77 -1.04 -.891978 -6.29 -.70 1.17 -3.44 -1.06 -.861979 -3.95 -.43 .12 -3.64 -1.19 -.491980 -7.21 .47 9.45 -1.59 -1.24 -.861981 -2.88 -61.00 -19.19 -.90 30.97 2.301982 -4.26 -.39 -7.42 -.49 -1.19 -1.101983 -4.74 -.39 -14.53 -.72 -1.16 -1.041984 -17.92 -.24 2.65 -5.64 -1.05 -.791985 -38.97 -.14 .89 -21.13 -1.02 -.54

Notes to Table 16:(a) (Domestic Price Ratio - Border Price Ratio)/Border Price Ratio, froe Table 14.

146

Table 17

Effect of Direct Price Interventions onRelative Consumer Price Differences (a)

Rice/Non-Ag Cocoa/Non-Ag Naize/Non-Ag Rice/Maize Cocoa/Rice Cocoa/Maize

1954 -.08 -.50 -.38 .49 -.45 -.191955 -.05 -.23 -.60 1.37 -.19 .911956 .05 -.19 -.48 1.02 -.23 .561957 -.01 -.52 -.70 2.33 -.51 .621958 -.01 -. 50 -.4? .95 -.49 -.011959 .00 -.41 -.47 .88 -.41 .111960 -. 06 -. 24 -.60 1.37 -.19 .921961 -.06 -. 19 -.47 .78 -.13 .541962 -.08 -.23 -.37 .46 -.16 .221963 .60 -.31 .66 -.04 -.57 -.591964 .49 -.18 -.03 .54 -.45 -.151965 .59 -.12 1.06 -.23 -.45 -.571966 .48 -.32 1.17 -.32 -.54 -.691967 .22 -.47 .16 .05 -.56 -.541968 -.16 -.56 -.29 .18 -.48 -.381969 .09 -.59 .02 .06 -.62 -.601970 .37 -.42 1.47 -.44 -.58 -.771971 .55 -.45 .45 .07 -.64 -.621972 .77 -.38 -.57 3.14 -.65 .451973 .34 -.50 1.19 -.39 -.62 -.771974 -.10 -.52 .11 -.19 -.47 -.571975 .74 -.36 .36 .28 -.63 -.531976 1.67 -.59 1.78 -.04 -.85 -.851977 2.95 -.39 1.62 .50 -.85 -.771978 2.50 -.60 .59 1.21 -.89 -.751979 1.59 -.33 .07 1.41 -.74 -.381980 4.33 .20 2.47 .54 -.77 -.651981 5.33 2.44 3.89 .30 -.46 -.301982 11.25 -.25 4.51 1.22 -.94 -.861983 22.44 -.22 11.82 .83 -.97 -.941984 3.88 -.11 1.00 1.44 -.82 -.561985 1.34 -.07 .39 .69 -.60 -.33

Notes to Table 17:(a) (Doeestic Price Ratio - Border Price Ratio)/Border Price Ratio, from Table 15.

147

this has been true for cocoa, it has not generally been true for rice and

maize, which have been the subject of severe import restrictions. The

reason for the apparent paradox is that the estimation of a border price

equivalent for these crops at the producer leve-l yields negative signs

during the late 1970s and early 19EKs because of high transport costs and

especially because of the highly overvalued exchange rate. The results are

therefore misleading.

Turming to Table 17, a clearer picture emerges. During the 1950s

and early 1960s, the domestic consumer price of rice was just slightly lower

than its border price equivalent, reflecting largely free trade and perhaps

minor quality differences. By 1963, however, the impact of import

restrictions was being felt and the NRP had risen to 60 percent. These

restrictions were relaxed during the late 1960s and then tightened again

during the 1970s..2 In 1975, imports of rice were c{lose to zero. Protection

increased markedly fram 1974 to 1983 except when the cedi was devalued in

1979. Rice imports (see Anm-nex 6) were sporadic but generally very low

between 1975 and 1979, and then increased to an average of about 43,000 tons

from 1980 through 1985. This was low, however, in relation to pent-up

demand, given the rise in domestic food prices that occurred during this

period. Protection decreased substantially in 1984 and again in 1985 as the

exchange rate was successively devalued.

The overall pattern of maize protection has been similar to that

of rice, but the degree of protection has fluctuated to a muich greater

extent because the domestic market is less integrated with import trade and

I The apparent relaxation suggested by the figures in 1973 and 1974was due to very high prices on international markets rather than to anysignificant increase in imports during those years.

148

there are large supply, and therefore domestic price, fluctuatims due to

variations in rainfall. There are also longer term movements revealed by

the data that are important. During the late 1950s and early 1960s, nominal

protection for maize was negative, as it was for rice, but its absolute

magnitude was much greater for maize than for rice because imports of maize

equalled less then 1 percent of domestic production whereas for rice they

were in most years greater than local production (Aninex Table 6-1). Maize

was essentially a nantradable, and its price was influenced by domestic

demand and supply, which established a local price well below the CIF price

of imports. Over the next twenty years, imports of maize remained very low

in relatimn to local productiin, but domestic prices increased relative to

border prices so that protection in most years was positive and in many

years it was greater than that for rice. After 1979, maize imports

increased rapidly to almost 24 percent of domestic production in 1982 as

protecticn for maize decreased substantially relative to that for rice.

Rates of protection on cocoa, have bee negative in almost every

year but have varied substantially with fluctuaticns in world market prices

and with changes in the domestic prodcLcer rice. SevLral phases can be

identified. The first was one of heavy taxation of cocoa exports during the

1950s when world prices were relatively high. Following this, there was a

period of relatively low taxation during the early 1960s as world prices

plummeted and the Nkrumah regime adjusted the producer price duwnward to a

lesser extent. Thereafter, rates of taxation increased steadily upwards as

world prices increased but producer prices failed to keep up with

accelerating dcmestic inflation. In 1979, world prices once again fell

sharply, and in 1980 and 1981 the exchange rate was so overvalued and the

149

world price was so low that the government ended up by subsidizing cocoa

exports, though most of this subsidy was necessary only because of inflated

CM1 costs. The following year world prices recovered, and rates of

protection once again were negative. Successive devaluations thereafter

left room for increases in both producer prices and government revenue from

cocoa as the tax rate was once more increased.

Indirect Effects

The indirect effects of policy on the agricultural sector occur as

a result of its impact both in agriculture and elsewhere in the ecormWy on

the exchange rate. In most countries this takes the form of overvaluation

resulting from tariffs and quantitative restrictions on imports. In Ghana

the exchange rate has also been influenced by the taxation of cocoa exports.

This has tended to cause the cedi to be undervalued, but this effect has

been dwarfed by import restrictions, which have worked in the opposite

direction.

Indirect price interventions are measured by adjusting domestic

prices for exchange rate disequilibria. This adjustment is made for rice,

maize, and cocoa in An,nex 5, which also describes the methodology employed.

He-re the combined effect of direct and indirect price interventions an

relative prices is measured by multiplying border prices, rather than

domestic prices, by the ratio of the equilibrium to the official exchange

rate. The equilibrium exchange rate used in these calculations is that

obtained from the simulation model described in Chapter II, with the results

presented in Table 4.

150

These adjusted border prices for each crop are divided by the

nmagricultural CPI adjusted for direct distortions and for exchange rate

disequilibria. The details ccerning these adjustments are cotained in

Pninex 5 and in the footnotes to Table 18, which presents the resulting

ratio, P* / P*i NA

Large fluctuatims from year to year make it difficult to draw

many inferences from Table 18 regarding trends, but there is me tendency

that seems fairly pronounced. That is the decline in cocoa producer prices

that occurred because of the widening margin taken up by the Cocoa Marketing

Bard. From 195B to 1964, there was a severe decline in the relative price

of cocoa at both the border and the producer level because of the fall in

world market prices. When cocoa prices revived after 1972, hoever, the

producer did not share in this improvement. The same tendency is true to a

lesser extent for rice and maize because of the general deterioratim of the

transportation system, though costs of private agricultural marketing did

not rise to nearly the same extent as did those of the CM.

Table 19 shcw.s the relative price differences betwen the

distorted domestic price ratios in Tables 14 and 15 and the adjusted ratios

in Table 18:

P, P.,* P,*

Phw P* p w

For rice and maize the net effect of dirert price distorticns and indirect

exchange rate disequilibria varies cmsiderably from year to year, but there

are scane patterns that emerge. In most years, for exanple, the overall

effect for maize is negative. Durwing the early years of the period under

151

Table 18

Effect of Direct and Indirect PriceInterventions on Relative Prices

haize Rice Cocoa

Year Producer la) Consueer (b) Producer (a) Consuser (b) Producer la) Consumer (b)_-- -- - -- - - -- - -- - - -- - -- - - -- - -- - - ------------ _--_-_-_-_-_ -

1958 2.56 3.29 3.00 4.44 9.99 10.851959 2.10 2.76 2.92 4.34 8.54 9.321960 3.22 4.03 3.00 4.38 6.81 7.531961 3.83 4.74 3.14 4.54 6.47 7.151962 2.67 3.43 2.54 3.94 6.06 6.821963 .79 1.24 2.41 3.71 5.69 6.431164 2.17 2.84 2.05 3.34 3.51 4.171965 1.46 2.04 3.80 5.39 4.82 5.391966 .94 1.48 2.87 4.42 4.69 5.291967 .67 1.16 2.22 3.64 4.93 5.551968 2.18 2.90 3.25 4.82 5.55 6.161969 2.06 2.76 2.79 4.33 7.91 8.441970 .41 .85 2.39 3.82 5.00 5.501971 .92 1.47 1.90 3.33 3.58 4.531972 4.60 5.70 1.60 3.01 2.86 4.321973 .47 .97 2.41 3.97 6.15 8.001974 1.23 1.82 3.93 5.65 5.82 8.011975 1.06 1.64 2.11 3.61 3.24 5.471976 1.02 1.69 2.77 4.62 6.32 7.781977 3.27 4.44 2.98 5.36 4.35 7.261978 3.93 5.12 3.13 5.62 13.69 16.131979 4.02 5.34 1.81 4.09 9.39 11.641980 2.50 3.55 3.12 5.51 7.18 9.581981 3.05 4.17 4.70 7.26 5.74 7.311982 1.82 2.79 3.99 6.53 6.52 9.341983 2.33 3.48 3.47 6.24 7.70 11.321984 1.91 2.91 2.39 4.74 5.97 9.701985 N/A N/A N/A N/A N/A N/A

Notes to Table 18:(a) Calculated as the Producer Price Equivalent at the equilibrium exchange

rate (Table 3-4(1) for rice, Table 3-4(2) for maize, andTable 3-4(3) for cocoa), divided by the Nonagricultural CPI fromTable 3-5(1), adjusted as described in Annex 5.

(b) Calculated the same way as (a) using the Retail Price Equivalent fromTable 3-4(1 for rice, Table 3-4(2) for maize, and the Optimum DomesticPrice at the border, valued at the equilibrium exchange rate, fromTable 3-4(3) for cocoa.

1 52

Table 19

Effect of Direct and Indirect Price Interventionson Relative Price Differences

maize Rice Cocoa

Year Producer (a) Consuser (b) Producer (a) Consumer (b) Producer (a) Consumer lb)

1958 -.643 -.571 -.228 -.170 -.442 -.4051959 -.679 -.586 -.316 -.230 -.388 -.3501960 -.799 -.725 -.481 -.356 -.253 -.2191961 -.723 -.662 -.548 -.405 -.223 -.1881962 -.648 -.569 -.549 -.382 -.230 -.1941963 .189 .160 .051 .058 -.342 -.2931964 -.479 -.412 -.191 -.114 -.094 -.0681965 -.121 -.073 -.469 -.352 -.419 -.3601966 .212 .173 -.394 -.277 -.354 -.3071967 -.263 -.180 -.252 -.181 -.314 -.2811968 -.588 -.511 -.558 -.433 -.368 -.3321969 -.326 -.283 -.330 -.250 -.567 -.5331970 1.422 .800 -.114 -.074 -.331 -.2281971 .025 .020 .066 .045 -.110 -.0861972 -.703 -.652 .715 .440 .278 .1851973 1.664 .934 .213 .151 -.372 -.2841974 -.121 -.088 -.351 -.276 -.398 -.2821975 -.020 -.014 .326 .221 .016 .0111976 .901 .615 .662 .442 -.518 -.4291977 -.153 -.121 .297 .208 -.126 -.0551979 -.652 -.565 -.179 -.124 -.689 -.5891979 -.703 -.610 -.184 -.097 -.551 -.4461980 -.057 -.032 .335 .245 -.598 -.4291981 -.371 -.298 -.543 -.381 -.315 -.2281982 -.109 -.052 .118 .118 -.471 -.2931983 .964 .779 .907 .613 -.620 -.3921994 -.311 -.244 1.108 .627 -.589 -.3771985 N/A NIA NIA N/A N/A N/A

Notes to Table 19:(a) Producer Price from Table 3-3M1) for rice and maize and from Table 3-3(2)

for cocoa divided by the Nonagricultural CPI froe Table 3-5(1) sinus theeffect on producer prices from Table 19, all divided by the effecton producer prices from Table 18.

Ib) Retail Price from Table 3-311) for rice and maize and RenderedPart Price from Table 3-3(2) for cocoa divided by the NonagriculturalCPI minus the effect on consumer prices from Table 18, all divided by theeffect on consumer prices from Table 18.

1 53

consideration, this was because both the direcft and indirect distortions

operated in the same direction. After 1964, when trade restrictions in most

years caused the domestic price to exceed the border price, the twD

influences on relative prices worked in opposite directions, with the

exchange rate distortions generally being dominant. This is consistent with

the fact that, until recently, maize was imported by Giana in only small

quantities and trade restrictions had less of an influence an its dcmestic

price than did local demand and supply. During the early 19E)s, moreover,

substantial increases in maize imports kept domestic prices from rising very

much, except during the drought year of 1963.

Rice imports, on the other hand, were subject to much more

restrictive trade controls, especially after 1970. In only 4 out of the

following 14 years, for example, were import restrictions sufficiently lax

that the overall effect of trade and exchange rate policies on the domestic

price of rice was negative. This is confirmed by the data in Prnex 6, which

show that imports of rice were not allowed to rise nearly as much as those

of maize.

The combined influence of direct and indirect interventions is

even more striking in the case of cocoa. The prcokdcer price was depressed

below its optimal level at the equilibrium exchange rate in every year but

two over the period studied. During those two years - 1972 and 1975 - the

producer price was increased significantly, world cocoa prices were

relatively low, and the exchange rate was not highly overvalued. Even

though the producer price was below its border price equivalent in those

years, the difference was less than at the optimal rate of export taxation.

154

In every other year, the total impact of policy was to penalize, and in most

years very heavily, the cocoa sector.

It is interesting to note, as well, the changes in the degre of

price distortions over time. By and large, the adverse impact of policy an

cocoa prices decreased during the Nkrumah years because of falling world

market prices. In 1965, however, the year before Nkrumah fell, cocoa was

heavily penalized as a result of a sharp increase in the overvaluation of

the exchange rate without any corresponding increase in producer prices.

Despite the change in government in 1966, cocoa continued to be penalized

until 1971, when low world market prices lessened the impact of the failure

of the Busia regime in that year to increase the producer price. There thEn

followed a few years of fairly high producer prices under Acheampong, though

these are not evident in Table 19 because 1973 and 1974 were years of very

high prices on the world cocoa market. Finally, there ensued, from 1976 to

1984, a period during which the cocoa price was highly distorted in relation

to its optimal level.

155

CHlTER VII: EFFECT ON Oa RJT, IWNfPTION, i1n FOFEIGN EXCH*M

Trade, price, and exchange rate policies in Ghana have affected

not only relative prices, as shlwn in the previous chapter, but also the

al location of resources. In this chapter the impact of these relative price

changes is assessed as these have influenced output, consumption, and

foreign exchange flwas.

The effects of price interventions are examined in the short run,

the long run, and the very long run. This is particularly important for the

cocoa sector, in which capital investmLent has a long gestation period. In

Ghana the capital stock of cocoa trees has been declining for a lcng period

of time. INt only has production been falling, but the average age of trees

has been increasing, indicating that replanting has not been sufficient to

offset the depreciation of the capital stock. A major issue to be explored

in this study is the extent to which this has been due to price policy.

Effects on Aqricultural Production

This section examines the effects of policy influencing relative

prices on agricultural production. Only changes in output prices are

considered because of the lack of reliable census or survey data an use of

inputs. Except for mechanized rice cultivation, hoever, labor and capital

are by far the most important production inputs, and most of the capital

consists of labor embodied in cocoa farms. For traditional, manual

techniques of production, intermediate inputs used on the farm acccunt for

less than 10 percent of the value of cocoa output and less than 5 percent

of protection on these inputs was not very far from unity, the error

156

introduced by not al lowing for direct subsidies does not appear to be very

great.

Mbre important may have been the indirect subsidies resulting from

the overvalued cedi, which made the CIF price of imported fertilizer,

tractors, and other inputs artificially low when converted to local currency

at the official exchange rate. Imports of these inputs were severely

constrained because of foreign exchange shortages, however, so that only a

few producers had access to them at these artificially low prices. Instead,

middlemen captured the rents created by these shortages as seccndary markets

developed in which inputs were sold at prices nuch higher than those

officially authorized. Finally, delivery delays were cammnn, decreasing

considerably the effectiveness of the inputs.

In the absence of detailed time-series information on the relative

importance of these various factors, it is impossible to estimate their

overall impact on produjction. It is clear, hawever, that only a few larger

farmers benefitted from either direct or indirect government subsidies and

that the vast majority of cultivators used few, if any, nontraditional

intermediate inputs. Consequently, limitation of the analysis to

consideration of the effects of distortions in output prices does not

seriously bias the results.

Supply Funictions. The impact of policies affecting output prices on

agricultural production may be estimated using the relative producer price

distortions show.n in Tables 16 and 19 and the elasticity of supply for each

of the tradable crops included in that table. Although the method for

estimating supply elasticities is relatively straight-forward, for annual

157

crops, such as rice and maize, it is considerably more complicated in the

case of cocoa. These methods and their results are described below.

Cocoa A number of estimates exist of supply elasticities for cocoa.

Perhaps the best of these for many years were those of Bateman.1- These

suggested that short-run elasticities from 1949 to 1962 were in the range of

0.39 to 0.87, depending on region, and long-run elasticities were from 0.77

to 1.2B.- This was a period, however, during which cocoa production and

acreage increased in respanse to strong positive price incentives, in

contrast to later years of declining production. Subsequent analysis by

Bateman covered a longer period from 1932/33 to 1969/70.: Price elasticity

estimates for this period were generally much laher. Short-term

elasticities for different regions, for example, ranged from 0.14 to 0.21.

Long-term elasticities were not given.

Mbre recently, the Ccmmodity Studies and Projections Division of

the World Bank estimated supply elasticities for a number of producier

countries. 4 The production decision, in these analyses, was viewed as a

x Merrill J. Bateman, Cocoa in the Ghanaian Econcmwv: An EconometricModel, Amsterdam: NorthHbl land, 1968.

2 Hossein Askari and John Thomas Cummings, Aaricultural SuWplvRrssonse: A Survey of the Eccrometric Evidence, New York: Praeger, 1976,p.404.

M errill J. Bateman, "An Ecorxmetric Analysis of Ghanaian CocoaSupply," in R.A. Kotey, C. Okali, and B.E. Rburke, (eds.), Eccnomics ofCocoa Production and Marketincq, Institute of Statistical, Social andEconomic Research, Liniversity of Ghana, Legon, 1974, pp.28E&-326.

Takamara Akiyama and Ronald C. Duncan, Analysis of the World CocoaMarket, Wbrld Bank Staff Commodity Working Paper ND. 8, 1982; T. Akiyama andA. Bowers, SuPplY Response of Cocoa in Major Produking Coaxtries, DivisionWbrking Paper Nb. 1984-3, Commodity Studies and ProjectionsDivision,Economic Analysis and Projections Department, Wbrld Bank, April 19B4.

158

two-stage process involving (1) the stock of cocoa trees and (2) the level

of production given a particular stock. Time series data on initial area

planted, new area planted, and yields were used to estimate production

capacity, which was an input along with prices into the supply function.

Short-run elasticities in Brazil, Ivory Coast, and Malaysia ranged fron 0.21

to 0.30. Long-run, steady-state elasticities, including feedback fron

acreage response over a period of 10 years, were 0.8 for Brazil and 1.8 for

the Ivory Coast.

For Ghana, the most recent estimates of cocoa supply elasticities

are contained in a draft annex prepared by Akiyama for the World Bank

Commodity Studies and Projections Division.25 These estimates are consistent

with the findings of the Cocoa Land Intensive Survey carried out during the

1970s by the Ghana Cocoa Services Division and analyzed in a Wbrld Bank

background paper., The estimates are based on the vintage matrix approach

used by the International Coroa Organization (IC02). This approach

estimates "normal" cocoa production based on a matrix of estimated acreage

of trees classified by age and average yield. A regression equation is then

estimated in the Akiyama study to explain actual production fram 1968/69 to

1963/84 by variations in normal production and other variables affecting

short-term supply, such as the producer price.

The Akiyama analysis results in short-run elasticities of cocoa

production with respect to price of 0.24 and with respect to insecticide

0 Takamara Akiyama, "Cocoa Supply Projections" Prriex C draft, July 30,1985.

' World Bank, Ghana: The Cocoa Sector, Backgrcoind Paper Nb. 1 of 4prepared for the Ghana: Policies and Program for Adjustment Report, October14, 1983.

159

sales to farmers of 0.29. The price variable is a weighted average of

prices lagged two and three years. This does not allow for any immediate

impact of the producer price an harvesting but does show saoe effect on

production thrcugh improved maintenance. The insecticide sales variable is

an unweighted average of sales over the previous three years. Elasticities

of new plantings with respect to price are also estimated at 0.5 for the

short run and 1.5 for the lang run. New plantings are calculated, however,

an the basis of hybrid seed distribution by the government, which has been

constrained tyy the capacity of its seed gardens and does not take into

account farmers use of their own traditional plant: materials. Furthermore,

much of the planting at this time was undertaken within the context of two

large cocoa projects for which the producer price was not a major

determinant of the planting schedule. The elasticity of new plantings

estimates are therefore not very reliable.

The procedure used here, and explained in detail in Annex 4,

involves emplcyying the methodology and some of the parameters developed by

Bateman to estimate new planting of traditional varieties over the period

1945/46 to 1985/86.7 New planting of hybrid varieties, which began in

1968/69, is estimated mn the basis of seed garden deliveries and is not

directly related to price. Each year's new planting hectarage of bDth

varieties is carried through the vintage matrix model, where it is

multiplied by the profile of yields over the age distribution of the trees.

Production from traditional and hybrid trees planted since 1945/46 is added

to production of trees planted earlier than 1945/46, with an adjustment to

' Bateman s estimates are for the period 1932/33 to 1969/70, butplanting of traditional varieties decreased rapidly to zero after the 1960s,and his model, as modified here, predicts this quite well.

160

take into acccunt the rate at which these trees must have been going out of

productimn to reach the area planted in them revealed by the surveys of the

1970s.

The resulting time-series of "normal" production is then compared

with actual production as estimated by official marketings.3 Same

adjustments are made to Bateman's parameters, as described in Annex 4, until

normal production corresponds reasonably well with actual producticn, due

allowance being made for the short-term influence of weather and prices.

As a final step, actual production is estimated as a function of

normal production and of current and past prices, using the following

specification:

-~ 2 co co cof

lnQt = Bo + B 1 lnNt + BElnQ*-,L + B31r7Pt + B4 lnPe + et ... (1)

where QF° is actual production, N'° is normal production, P-- is the

producer price of cocoa, Fr is the producer price of food crops, and t is

the year in which production takes place. The lagged dependent variable in

this equation represents the effect on current production of previous years'

prices. In the usual Nerlavian formulation, these prices are determinants

of the price the farmer expects to receive in the current year. In the

cocoa sector, however, the producer price is annxouced well in advance of

the main harvest season so there is no unmcertainty for the farmer. Previous

E Actual production differs from official marketings by the quantityof cocoa smuggled to neighboring countries. This quantity is judged (seeAnnex 1) to have been about 10 percent of production in recent years, but itis impossible to estimate, with any degree of accuracy, the year to yearvariatimns in smuggling because of marked shifts in the degree to whichanti-smuggling laws have been enforced. Instead, the approach used here isto treat official marketings as production and to assume that smuggling isane aspect of producers' supply response to price incentives.

161

prices do influence current production, however, because of their effects cn

weeding, pruning, spraying, canopy repair, and other types of maintenance,

the full impact of which may not be felt for several years. Cbnsequently,

this term is included to distinguish the short run price elasticity BR from

the long run elasticity EB/(1-8a=), where B6 is the coefficient of

adjustment.

The price of food crops is represented here by that of maize,

which tends to be closely correlated, as noted earlier, with the prices of

the other major food staples. The current price is used rather than the

lagged price since farmers are able to predict focd crop prices sufficiently

in advance during the growing seasm to influence their decisions as to how

intensively to harvest cocoa. To the extent that past food crop prices also

influence cocoa production, this is captured in the autoregressive term,

with the long-run elasticity being given by BE/( 1-E6).

The parameters estimated for equation (1), using data for 1944/45-

1985/86, are given by:

1nQt = - 1.480 + .498 lnN4 + .643 lnQ]_ + .223 ln P* -. 143 lnP* ... (2)

(-1.344)(2.644) (5.123) (2.953) (1.787)

R = 0.824 H - 0.547

where the figures in parentheses are the t-statistics for the regression

coefficients. All slope coefficients have the expected signs and the first

three are significantly different from zero, using a two-tailed test, at the

.01 level of significance. The coefficient of the price of maize is

significantly different from zero at the .10 level. The short-run

elasticity of output with respect to the price of' cocoa is .22, which is

162

consistent with previous estimates for Ghana and other major producing

countries. The short-run elasticity with respect to the price of maize is

-. 14. The lngr-rum elasticity of cocoa output, given the stock of cOCoa

trees, is .62 with respect to the price of cocoa and -. 40 with respect to

the price of maize. This is different from the long-rrn, steady-state

elasticity assuming that the stock of trees changes in response to price.

The latter parameter is, in fact, quite difficult to calculate,

especially for large changes in price, for several reasons. Annex 4

describes in detail the factors influencing the rate at which new planting

takes place and the impact that this has on subsequent production. Briefly

summarized, new planting depends, in the Bateman fornmulation, on (1) the

difference between the real producer price and a threshold level below

which planting would be unprofitable, (2) a nonlinear planting effort

function, and (3) the absorptive capacity of the agricultural sector to

undertake new investment linked to the size of the populaticn, the supply of

capital, and other constraining variables. Even if the relationship between

the rate of planting and produrer price could be established, the relative

production response also depends on the time period over which that response

is measured, the size and age distribution of the existing tree stock, and

the yield profile of the trees. The vintage matrix model is able to

incorporate the complexity of this price response mechanism without having

to focus on a single parameter - the long-run elasticity of supply. This

is particularly important in Ghana, where the response in recent years to

changing producer prices has not been new planting but rather the failure to

replant as praductive capacity has decreased with aging of the existing

trees.

163

The pracedure for estimating the impact that price policy

distortions have had on cocoa prnduction in Ghana therefore involves several

steps. The first is to calculate the replanting of traditicnal trees that

would have taken place with unrdistorted equilibrium prices using the

procedure described in Anex 4. This planting series is then carried

thrcugh the vintage matrix model to derive normal production under

equilibrium conditions. The final step is to calculate Qt from equation (2)

using this series on normal production and data on undistorted prices.

Food Crops As discussed in Annex 1, the quality of the production

data for food crops is very poor. Nevertheless, the magnitude of

distortions is such that large variations in prices ever time could

compensate, to a considerable extent, for the errors in the data.

Consequently, supply functions were rxun using the existing data, and the

results were assessed for reasonableness in comparison with other estimates

that have been made outside of Ghana.

The following regression equation was estimated for maize:

m m m

lnt*, = 4.304 + .142 lnQD_± + .794 lnPt.m_ -.267 lnPt - .440 lnP-±L ... (3)(3.434)(2.535) (2.519) (1.771) (1.802)

-2R = .32 H = 0.981

where the superscripts designate maize (m), cocoa (co), and cassava (ca).

All coefficients have the predicted signs and those for the lagged values of

On and Pm are statistically significant at the .05 level. The estimators

for the cross-price elasticities for cocoa and cassava are significant at

the .10 level. The short-rmu price elasticity for maize of .79 seems

164

somewhat high in relatimn to most estimates for cereals outside of Africa,v

but it is within the range of elasticities estimated in Nigeria for cottan

and groauxnduts, annual crops that may be substitutes for maize. -

Furthermore, the coefficient of the autoregressive term is fairly low,

implying that the long-run elasticity of maize with respect to its own price

is mnly .93, which does not seem unreasonable given high substitutability

in production with cocoa and other food crops such as cassava, yams, millet,

and sor-ghum.

In cmntrast to maize, rice, which is grown principally in the

river valleys of the north, does not use land that is suitable for other

major crops. Nevertheless, there is some substitutability in production

with cocoa, which competes with rice for migrant labor during the harvest

period. With both rice and cocoa prices as explanatory variables, the

following regression equation was run,

a ,- r- co

lnQG = 3.006 + .242 1nQG.- + .427 lnP*-1 - .391 inP* ... (4)(2.424) (2.773) (2.273) (2.340)

-2R = .467 H = 2.644

Al1 coefficients have the expected signs and are significant at the .05

level.

This is misleading, however, since the H statistic suggests the

presence of positive serial correlation, implying that the standard errors

9 Askari and Cummings, Agricultural SuPPly Response ... , pp.390-96.

20 Marian E. Bond, "Agricultural Responses to Prices in Sub-SaharanAfrican Comtries," IMF Staff Papers, 30 (4), December 1963. pp. 710-11.There are no reliable existing estimates of food crop elasticities forneighboring countries in West Africa.

165

of the regression coefficients are uiderestimated. L An attempt was made to

redress this situation using the Cochrane-Orcutt procedure. This led to a

decrease in the short-run elasticity of rice with respect to its oaw price

from .43 to .17 and a decrease in the long-run elasticity from .56 to .33.

The coefficients of the price of rice and cocoa, however, were in this case

not statistically significant. This implies that the null hypothesis that

the prices of rice and cocoa have no influence on rice output cannot be

reliably rejected. Nevertheless, it does not mean that equation (4) cannot

be used to predict the output of rice under a regime of different prices as

long as the same omitted variables that resulted in serial correlation would

have continued to operate during the same period with the different price

regime. The implied owu-price elasticities in the short-run of .43 and in

the long-run of .56 do not appear to be unreasonable.

Price Elasticities of Supply As a way of verifying the reasonableness

of the estimates, it is useful to construct the following matrix of

short-run elasticities:

Short-Run Supply Elasticities

Output Price

Cocoa Maize Cassava Rice

Cocoa +.22 -. 14Maize -. 27 +.79 -. 44Rice -. 39 +.43

" The H statistic, which is more appropriate than the Durbin-Watsonstatistic in the presence of an auto-regressive term, is not significant for-either the cocoa or the maize supply equations.

166

Only those coefficients that are significant at the .10 level or greater are

show.

With the possible exceptim of the elasticity of maize output with

respect to its own price, discussed earlier, the results appear reasmable

and consistent with mne another. The cross elasticity of maize with

respect to the price of cocoa, for example, is about double the elasticity

of cocoa with respect to the price of maize, which is consistent with the

synmnetry conditim (dg(/dP- = dg=/dPm) and the relative importance of the

tWo crops in production. The fact that the elasticity of rice with respect

to the price of cocoa is statistically significant but that the elasticity

of cocoa with respect to the pr-ice of rice is not significant is not

surprising in view of the much greater relative importance of cocoa.

Finally, the greater substitutability in production of maize and cassava

than of maize and cocoa is cmsistent with the relative magnitude of these

two cross-elasticities.

Equilibrium Levels of Output

Equilibrium levels of output were calculated using equatic-s (2)

through (4) for the short run, lamg run, and very lang run. In the short

run, real equilibrium prices were substituted for real actual prices, where

the equilibrium prices were obtained from Tables 14 and 18 adjusted so that

the nmagricultural price deflator equals unity in 1963, the base year used

in estimating these equations. Actual lagged values of the dependent

variable were used in this calculation, and the residuals were added to the

167

predicted values assuming stochastic variation, largely due to fluctuations

in rainfall, to be the same in equilibrium as in the distorted situation.12

The equilibrium level of output in the long run is predicted in

the same way as in the short run except that the lagged value of the

dependent variable, after the first year, is its predicted rather than its

actual value. The deviation of equilibrium prices frcm distorted prices

influences output in twD ways. First, it affects output directly in the

current year. Second, it affects autput indirectly by influencing previous

levels of output, which in turn have an impact on current output.

Attempts to predict the long-rum equilibrium level of cocoa

production in this way result in output rising to very high levels because

of the influence of the autoregressive term. This implies that the ability

of farmers to increase production through improved tree maintenance is much

greater than it is kncwn in fact to be. The problem is at least partially

that the coefficient of the log of the autoregressive term is estimated

largely on the basis of a historical decline in output resulting from lack

of maintenance in the face of low producer prices. While some reversal of

this process can be anticipated, it would not be unlimited, and diminishing

returns must set in as the trees approach their maxinum yield.

In the very long run, cocoa output is influenced by the same

variables that affect output in the long run, but prices also have an impact

on production through their effect on planting. Whereas normal production

2t This deals with the problem raised in A.0. Krueger, M. & hiff, andA. Valdes, "Note 7. On Measure.ent of Quantitative Effects and RegressicnResiduals," Memo No. 27, July 11, 1986. Use of the original equations,rather than just estimated elasticities, to calculate equilibrium levels ofautput is very important for Ghana because of the large distorticnsinvolved. These can lead to very biased results if changes conforming tothe elasticities are calculated using a single base point.

168

in equation (2) is based on estimated actual planting for the short and Iong

runs, for the very icng run is derived, using the vintage matrix model, from

the plantings that would have occurred had prices been in equilibrium.

These plantings are estimated in the same way as actual plantings, with

equilibrium prices used in place of distorted prices. The equilibrium

output level for maize and rice is the same, of course, as in the long run.

The results of this exercise are shown in Tables 20-22 for the

direct effect of price distortions on output, and in Tables 23-25 for the

total effects. From Table 20 it appears that the effect of direct price

distortions on production of maize and rice in the short run has been almost

uniformly positive, since during the early years in which the effect on

maize appears to have been negative, maize imports were very low in relation

to production and domestic prices were scarcely influenced by the cost of

these imports. Even later, when the equilibrium level of maize production

was usually higher than its actual level, there were sharp variations from

year to year, which were less a result of changes in trade policy than of

fluctuations in domestic supply. Towards the end of the 1970s, the exchange

rate became so overvalued that the price of maize and rice to producers in

most areas of the country would have reached zero if these cereals had

continued to be marketed in the major urban centers, As noted earlier, this

was because of the low border prices at the official exchange rate and the

high cost of transport and marketing.5 This gives rise to very low levels

of output for these years in Tables 20-22. In fact, of course, farmers

would have ceased to sell to urban markets as producer prices fell, and

3 In estimating the output of maize and rice for Tables 20-22, thenegative prices for these later years shown in Table 14 were set equal tounity because of the logarithmic nature of the supply function.

169

Table 20

Direct Effect on Output, Short-Run(000 at)

maize Rice Cocoa

Year 0 la) 01 (b) A D/0 Ic) O (a) 01 (b) Q/0O tc) 0 (a) 0t lb) A 0/0 kc)

1955 169 N/A NIA 23 NIA N/A 229 221 .0351956 169 363 -.534 23 22 .033 264 259 .0181957 169 238 -.289 23 16 .396 205 211 -.0271958 183 475 -.615 30 23 .287 256 280 -.0951959 N/A 392 N/A N/A 32 N/A 317 340 -.0671960 N/A N/A N/A 32 N/A N/A 430 410 .0501961 N/A N/A N/A 30 29 .038 409 416 -.0161962 169 NIA N/A 31 28 .110 413 436 -.0521963 183 238 -.232 33 29 .141 428 563 -.2401964 170 82 1.076 42 28 .476 538 589 -.0871965 206 15B .306 32 20 .574 401 534 -.2501966 353 102 2.472 29 15 .883 372 545 -.3171967 278 72 2.861 42 23 .824 423 537 -.2131968 253 131 .934 42 25 .696 333 401 -.1691969 300 260 .154 60 47 .282 419 529 -.2081970 482 307 .570 49 35 .380 434 639 -.3211971 465 84 4.568 55 34 .633 477 608 -.2151972 402 166 1.416 70 40 .747 427 428 -.0011973 427 530 -.194 62 30 1.087 357 524 -.3181974 486 94 4.176 73 41 .783 395 511 -.2271975 343 162 1.116 71 59 .208 416 514 -. 1911976 286 83 2.432 70 31 1.242 339 539 -.3711977 274 31 7.863 109 42 1.598 290 413 -.2981978 218 17 12.059 108 9 10.693 286 408 -.3001979 380 45 7.423 93 12 6.836 323 357 -.0961980 382 83 3.595 78 19 3.128 284 341 -.1661981 378 36 9.390 97 B1 .199 248 137 .8051982 346 2 209.524 36 5 5.875 198 355 -.4421983 173 1 190.515 40 4 8.174 173 349 -.5051984 574 1 529.637 66 7 8.389 189 224 -.1571985 411 10 40.514 90 10 7.856 226 248 -.090

Notes to Table 20:(a) Actual output froe Anne% 1, Table 1-2(1) for maize and rice, and Table 1-2(31) or cocoa.(b) Short-Run Equilibrium outpt trom equations (2)-(41 as described in text.Ic) Relative change in outputX. 0/A obtained by dividing CO-OR) by 01.

170

Table 21

Direct Effect on Output, Long-Run(000 at)

Maize Rice Cocoa

Year a (a) 01 (b) '4 0/9 (c) 0 (a) Ot 1(b) A /0 (c) Q (a) 1 (b) A. 0/a (c)

1955 169 N/A N/A 23 N/A N/A 229 221 .0351956 169 363 -.534 23 22 .033 264 254 .0411957 169 265 -.362 23 16 .407 205 205 -.0021958 183 506 -.639 30 21 .397 256 279 -.0841959 N/A 453 N/A N/A 30 N/A 317 359 -.1181960 N/A 373 N/A 32 64 -.503 430 444 -.0321961 N/A 462 N/A 30 34 -.124 409 424 -.0361962 169 631 -.732 31 29 .075 413 446 -.0741963 183 287 -.363 33 28 .161 428 592 -.2771964 170 87 .947 42 27 .531 538 726 -.2581965 206 144 .435 32 18 .745 401 648 -.3811966 353 97 2.655 29 13 1.154 372 742 -.4981967 278 60 3.641 42 19 1.197 423 843 -.4901968 253 105 1.405 42 20 1.052 333 632 -.4731969 300 229 .307 60 39 .525 419 815 -.4061970 482 296 .631 49 32 .529 434 1006 -.5691971 465 78 4.968 55 30 .808 477 1077 -.5571972 402 129 2.113 70 35 1.015 427 745 -.4271973 427 451 -.053 62 25 1.474 357 772 -.5381974 486 95 4.136 73 33 1.220 395 866 -.5441975 343 129 1.669 71 48 .466 416 880 -.5271976 286 72 2.946 70 28 1.460 339 900 -.6231977 274 25 9.771 109 34 2.231 290 804 -.6391978 218 12 17.302 108 7 14.530 286 822 -.6521979 380 30 11.727 93 6 14.219 323 739 -.5631980 382 58 5.594 78 10 6.977 284 614 -.5371981 378 28 12.581 97 49 .982 248 240 .0341982 346 1 303.915 36 4 7.112 198 370 -.4651983 173 .4 430.476 40 3 14.225 173 553 -.6871984 574 .5 1,254.9178 66 4 17.147 189 504 -.6251985 411 4 113.352 90 5 16.859 226 496 -.544

Notes to Table 21:(a) Actual output froo Annex 1, Table 1-2(1) for maize and rice, and Table 1-2(3) for cocoa.(b) Long-Run Equilibrius output from equations (2)-(4) as described in text.(c) Relative change in output Ai Q/ obtained by dividing (Q-O1) by 01.

17 1

Table 22

Direct Effect on Output, Very Long-Run(000 It)

maize Rice Cocoa

Year Q (a) 0Q (b) AO /0 (c) 0 (a) f l(b) Q/Q Cc) 0 (a) 0Q (b) A Q/Q (c)

1955 169 N/A N/A 23 N/A N/A 229 221 .0361956 169 363 -.534 23 22 .033 264 253 .0421957 169 265 -.362 23 16 .407 205 205 -.0021958 183 506 -.639 30 21 .397 256 279 -.0841959 N/A 453 NIA N/A 30 N/A 317 360 -.1191?60 N/A 373 N/A 32 64 -.503 430 444 -.0321961 N/A 462 N/A 30 34 -.124 409 424 -.0361962 169 631 -.732 31 29 .075 413 446 -.0741?63 183 287 -.363 33 28 .161 428 592 -.2771964 170 67 .947 42 27 .531 538 726 -.2591965 206 144 .435 32 18 .745 401 648 -.3811966 353 97 2.655 29 13 1.154 372 742 -.4991967 278 60 3.641 42 19 1.197 423 845 -.5001968 253 105 1.405 42 20 1.052 333 637 -.4771969 300 229 .307 60 39 .525 419 828 -.4941970 482 296 .631 49 32 .529 434 1034 -.5801971 465 78 4.968 55 30 .808 477 1130 -.5781972 402 129 2.113 70 35 1.015 427 806 -.4701973 427 451 -.053 62 25 1.474 357 865 -.5871?74 486 95 4.136 73 33 1.220 395 1013 -.6101975 343 129 1.669 71 48 .466 416 1084 -.6161976 296 72 2.946 70 28 1.460 339 1179 -.7131977 274 25 9.771 109 34 2.231 290 1124 -.7421978 218 12 17.302 1o8 7 14.530 286 1227 -.7671979 380 30 11.727 93 6 14.219 323 1177 -.7261980 382 58 5.594 78 10 6.977 284 1037 -.7261981 378 28 12.581 97 49 .982 248 429 -.4211982 346 1 303.915 36 4 7.112 198 694 -.7151983 173 .40 430.476 40 3 14.225 173 1086 -.8411984 574 .46 1,254.918 66 4 17.147 189 1031 -.8171985 411 4 113.352 90 5 16.859 226 1055 -.786

Notes to Table 22:(a) Actual output from Annex 1, Table 1-2(1) for saize and rice, and Table 1-2(3) for cocoa.(b) Very Long-Run Equilibrium output from equations (2)-(4) as described in text.(c) Relative change in output A Q/Q obtained by dividing (0-02) by Ql.

172

Table 23

Total Effect on Output, Short-Run(000 It)

Maize Rice Cocoa

Year 0 (a) as (b) a 0/0 (c) 0 (a) 0 (b) A 0/0 (c) 0 (a) 01 (b) A Q/g Icl

1955 169 N/A N/A 23 N/A N/A 229 203 .1281956 169 406 -.584 23 26 -.124 264 235 .1241957 169 261 -.352 23 19 .219 205 194 .0541958 183 494 -.629 30 25 .215 256 262 -.0221959 N/A 476 N/A N/A 37 N/A 317 320 -.0091960 N/A N/A N/A 32 N/A N/A 430 387 .1101961 N/A N/A N/A 30 36 -.165 409 393 .0401962 169 N/A N/A 31 39 -.206 413 408 .0131963 183 363 -.496 33 39 -.149 428 519 -.1751964 170 141 .209 42 40 .053 538 526 .0221965 206 235 -.124 32 25 .261 401 498 -.1951966 353 292 .209 29 30 -.025 372 470 -.2091967 278 193 .442 42 46 -.084 423 450 -.0771968 253 229 .104 42 38 .093 333 338 -.0141969 300 384 -.218 60 62 -.038 419 482 -.1311970 482 457 .056 49 49 -.007 434 538 -.1931971 465 177 1.634 55 56 -.021 477 490 -.0271972 402 320 .255 70 75 -.072 427 339 .2611973 427 714 -.402 62 41 .503 357 406 -.1211974 486 131 2.718 73 55 .327 395 436 -.0941975 343 245 .400 71 87 -.187 416 412 .0101976 286 136 1.106 70 48 .462 339 432 -.2151977 274 69 2.958 109 80 .356 290 285 .0161978 218 62 2.499 108 63 .702 286 311 -.0801979 380 178 1.136 93 73 .278 323 312 .0341980 382 165 1.321 78 59 .324 284 329 -.1351981 378 66 4.747 97 74 .309 248 239 .0371982 346 55 5.260 36 39 -.079 198 214 -.0761983 173 18 8.565 40 27 .503 173 223 -.2231984 574 23 23.830 66 37 .763 189 203 -.0701985 411 N/A N/A 90 N/A N/A 226 N/A N/A

Notes to Table 23:(a) Actual output from Annex 1, Table 1-2(1) for maize and rice, and Table 1-2(3) for cecoa.(b) Short-Run Equilibrium output from equations (2)-(4) as described in text.(cl Relative chanqe in output.J9/0 obtained by dividing (0-0*) by 0*.

173

Table 24

Total Effect on Output, Long-Run(000 at)

Maize Rice Cocoa

Year 0 (a) Os (b) A i0/ (c) 9 (a) as (b) A 0/0 (c) 0 (a) 01 (b) A/0/3 (ci

1955 169 N/A N/A 23 N/A N/A 233 203 .1471956 169 406 -.584 23 26 -.124 264 217 .2141957 169 295 -.428 23 19 .181 205 172 .1941958 183 535 -.658 30 24 .265 256 233 .1001959 N/A 554 N/A N/A 35 N/A 317 301 .0541960 N/A 504 N/A 32 79 -.593 430 375 .1481961 N/A 697 N/A 30 45 -.328 409 360 .1371962 169 1100 -.846 31 43 -.279 413 376 .1001963 183 473 -.613 33 42 -.214 428 488 -.1231964 170 161 .057 42 42 -.006 538 573 -.0611965 206 233 -.117 32 25 .259 401 518 -.2261966 353 297 .188 29 28 .031 372 555 -.3291967 278 180 .478 42 46 -.078 423 596 -.2911968 253 217 .167 42 39 .072 333 427 -.2191969 300 375 -.201 60 61 -.022 419 576 -.2731970 482 471 .023 49 50 -.012 434 677 -.3591971 465 176 1.642 55 56 -.025 477 674 -.2921972 402 279 .441 70 76 -.078 427 453 -.0581973 427 678 -.370 62 42 .474 357 513 -.3041974 486 140 2.482 73 50 .457 395 606 -.3481975 343 205 .671 71 80 -.109 416 609 -.3171976 286 126 1.266 70 49 .422 339 631 -.4631977 274 62 3.445 109 74 .476 290 442 -.3441978 218 50 3.325 108 58 .871 286 426 -.3281979 380 144 1.630 93 63 .487 323 424 -.2381980 382 143 1.662 78 54 .458 284 414 -.3141981 378 57 5.604 97 68 .433 248 324 -.2341982 346 42 7.185 36 36 .004 198 271 -.2691983 173 13 11.893 40 27 .505 173 289 -.4011984 574 16 34.699 66 34 .946 189 301 N/A1985 411 N/A N/A 90 N/A N/A 226 N/A N/A

Notes to Table 24:(a) Actual output from Annex 1, Table 1-2(1) for maize and rice, and Table 1-213) for cocoa.(b) Lonq-Run Equilibrius output from equations (2)-(4) as described in text.(c) Relative change in output, il/9 obtained by dividing (0-01) by 01.

174

Table 25

Total Effect on Output, Very Long-Run(000 It)

maize Rice Cocoa

Year Q (a) t I(b) A 0/Q (c) Q (a) Ot (b) A 0/0 lc) 0 (a,) Of (b) 0/0 Cc)

1955 169 N/A N/A 23 N/A N/A 229 203 N/A1956 169 406 -.584 23 26 -.124 264 217 N/A1957 169 295 -.428 23 19 .181 205 172 N/A1958 183 535 -.658 30 24 .265 256 233 N/A1959 N/A 554 N/A N/A 35 N/A 317 301 -.0401960 N/A 504 N/A 32 79 -.593 430 375 .0701961 N/A 697 N/A 30 45 -.328 409 360 .1031962 169 1100 -.846 31 43 -.279 413 376 .0701963 £83 473 -.613 33 42 -.214 428 488 -.1431964 170 161 .057 42 42 -.006 538 573 -.0961965 206 233 -.117 32 25 .259 401 519 -.2401966 353 297 .188 29 28 .031 372 555 -.3321967 278 188 .478 42 46 -.078 423 598 -.3051968 253 217 .167 42 39 .072 333 430 -.2381969 300 375 -.201 60 61 -.022 419 587 -.2971970 482 471 .023 49 50 -.012 434 704 -.3791971 465 176 1.642 55 56 -.025 477 723 -.3401972 402 279 .441 70 76 -.078 427 491 -.1351973 427 678 -.370 62 42 .474 357 570 -.3811974 486 140 2.482 73 50 .457 395 693 -.4351975 343 205 .671 71 80 -.109 416 715 -.4261976 286 126 1.266 70 49 .422 339 761 -.5691977 274 62 3.445 109 74 .476 290 616 -.5561978 218 50 3.325 108 58 .871 286 666 -.5971979 380 144 1.630 93 63 .487 323 729 -.5841980 382 143 1.662 78 54 .458 284 769 -.6521981 378 57 5.604 97 68 .433 248 641 -.6351982 346 42 7.185 36 36 .004 198 563 -.6671983 173 13 11.893 40 27 .505 173 627 -.7441984 574 16 34.699 66 34 .946 189 680 -.7391985 411 N/A N/A 90 N/A N/A 226 N/A NIA

Notes to Table 25:(a) Actual output from Annex 1, Table 1-2(1) for maize and rice, and Table 1-2(3) for cocoa.(b) Very Long-Run Equilibrium output from equations (2)-(4) as described in text.(c) Relative change in outputAO 9/ obtained by dividing (0-02) by a1.

175

marketing costs wculd have been lowered, maintaining producer prices at

positive levels for sales within the local rural area.9*

The positive effect of price distortions on producition of maize

and rice occurred for two reasons. First, from 1963 cnwards, inflation and

restrictions on imports resulted in domestic producer prices exceeding

their border price equivalents in most years. Second, distortions in the

producer price of cocoa encouraged production of alternative crops such as

maize and rice.

The figures on cocoa production confirm this hypothesis regarding

the allocation of resources. During the period up to 1962, the domestic

price of cocoa was in some years substantially lower than its border price,

but this was offset in some cases by a domestic price of maize that was much

lower than its border price. As a result, the direction of the effect of

price distortions on cocoa output varied from year to year. After 1963, a

combination of inflation and import restrictions resulted in a domestic

price for maize that was higher in most years than its border price. This,

coupled with continued taxatimn of cocoa exports, led to a decline in cocoa

production below the equilibrium level in every year but one.-s'

Tables 21 and 22, which show the direct effect of distortions on

output in the long and very long run, confirm these tendencies in more

±4 Subtracting all transport and marketing costs from the CIF price ofcereals imports implicitly assumes that consumption takes place in Accra.In fact, most maize production and a substantial share of the output of riceare consumed in the interior so that transport costs would have to be addedto, rather than subtracted from, the border price in order to obtain itsproducer price equivalent. Existing data do not permit, however, estimationof the quantities involved.

Lf The only year in which there was an exception to this was 1981,when the wDrld price for cocoa was so low, at the official exchange rate,that the export tax was actually negative.

176

exaggerated form. They suggest that if free trade had bee permitted at the

official rate of exchange, prcducticn of rice and maize wold have virtually

ceased by the early 1980s, and that of cocoa wsuld have expanded, with new

plantings, to almost five times its actual level. These results are based,

hoever, on extrapolation well beyond the range over which equations (2)-

(4) were originally estimated. They must therefore be treated with a great

deal of caution. Expansion of cocoa production to over 1 million tons per

year, for example, would have encountered severe constraints in terms of

both the effect this would have had on wDrld market prices and the

availability of suitable land. Despite these caveats, however, it is clear

that the direct effect of distortions on output has been very considerable.

Table 23, shows the total effect of price distortions an output in

the short-run. Since the cedi in Ghana was always overvalued during the

period under consideratio, the equi l ibrium level of producer prices is

higher relative to the actual level than when this distortion is not taken

into accoLmt. The impact this has on production depends, however, on the

relative importance of the own-price and cross-price effects. For maize and

rice, a comparison of Table 23 with Table 20 suggests that these price

effects result in a consistently higher level of equilibrium output relative

to its actual level when all distortions are taken into accat. This is

because of the own-price effects, which tend to increase the equilibrium

level of cereals output compared with the situation in which the change in

the exchange rate is ignored. The cross-price effects resulting from the

difference between the actual and equilibrium price of cocoa, on the other

hand, are more complicated. While the impact of overvaluation of the

currency tends to raise the equilibrium producer price of cocoa in relation

177

to its actual level, the introduction of an optimun export tax for the

purpose of calculating the total effect, tends to lower the equilibrium

price. Mbving fron the direct to the total effect does not, therefore,

always result in an increase in the cocoa price distortion, and its effect

on output of cereals varies from year to year.

The same is of course also true of its influence on cocoa output.

Except for 1981, when cocoa exports were subsidized rather than taxed, the

equilibrium level of cocoa production, taking into acconmt both the

overvaluation of the exchange rate and the optimal export tax, was less than

its level would have been if producer prices had equalled their border

price equivalents. Not only would the equilibrium price of cocoa have been

lower because of the optimal export tax but also the exchange rate effects

would have increased the equilibrium producer price of maize, drawing

resources away from cocoa. As a result, there are a number of years in

which actual cocoa output was greater than equilibrium output in the short

run.

This pattern changes as allowance is made for the responsiveness

of tree maintenance (long run) and new planting (very long run) to prices,

as shown in Tables 24 and 25. After an initial period up until 1963, when

Ghana might have benefitted from restricting output so as to hold up wDrld

cocoa prices, the total effect of cocoa price policy was uniformly negative,

especially in the very long run because of the disastrous effect of policy

on planting. Even if Ghana had optimally restricted cocoa exports, the

level of production in the mid-19E8s would have been 3.5 times its actual

level. With rice and maize, the decreased level of equilibrium, in

178

cmnparison with actual, output in the long run is less a functian of cwn

price response than of farmers being influenced by higher cocoa prices.

Effect on Consumptimn

Goverrvint policy influencing relative prices also affects

consumption of tradable foods. This influence can be estimated by

incorporating the equilibrium consumer prices used in Tables 15 and 18 into

estimated demand functions to predict the levels of consumption that would

have existed in the absence of direct and total distortion. ND distinction

is made, however, between short- and long-rum elasticities since it is

assumed that the entire consumption response would occur within one year.

Estimates of own-and cross-prices elasticities of Ghanaian demand

are available for maize, sorghum/mi llet, cassava, cocyyams/yams, and rice

from Haessel. - Prior information on income elasticities was combined by

Haessel with annual net import, price, and production data from 1953 to

1970, using the Theil-Goldberger mixed estimation tecrhnique, and two-stage

least-squares regression analysis was performed. X The price elasticity for

rice of about -1.25, obtained by Haessel, was highly significant and robust

under alternative specifications. The elasticity for maize, on the other

hand, was very high, but statistically insignificant. This is partly

1 Walter Haessel, "The Demand for Agricultural Commodities inGhana: An Application of Nonlinear Two-Stage Least Squares with PriorInformation," American Journal of Agricultural Economics, 58(2), May 1976,pp. 341-45.

t' It is reasonable to base the estimated demand function on theperiod from 1953 to 1970, rather then the whole period under consideration,because by the mid-1970s the quantity consumed of rice and maize was highlyconstrained by quantative restrictions on imports, with price the endogenc/usdependent variable and quantity consumed the exogenous independent variable.

179

because maize and the other coarse grain cereals, millet and sorghum, are

substitutable for nme another so that their prices tend to move closely

together. This makes it difficult to distinguish the separate influence of

the maize price variable. When the coarse grain cereals were lumped

together, the resulting price elasticity of -2.323 was almost significant,

with a standard error of 1.58.

When the root crops, cassava and yams, were combined with the

coarse grain cereals, the best results were obtained as follows:

lnC* = 2.960 - 1.689 lnPf - 0.823 lnPt + 0.427 lnPr + 0.917 lnY*. ... (5)(.14) (0.99) (0.92) (0.59) (0.54)

. ~ ~~c t r

lnct = 1.90 - 0.362 ln Pf - 0.746 lnPt - 1.256 lnPr + 0.874 lnYt ... (6)(.933) (0.62) (0.58) (0.41) (0.49)

where C is consumption, P is price, Y is per capita income, and the

superscripts designate coarse grain cereals (c), roots and tubers (t), and

rice (r). Figures in parentheses are the standard errors of the

coefficients. None of the coefficients in equation (5) is statistically

significant at the 0.10 level, though the coefficients of the cereals price

and per capita income terms are almost so. The coefficient of the logarithm

of the price of rice is significant at the 0.01 level in equatim (6) and

that of the income term is significant at the 0.10 level. None of the

cross-price elasticities is significant in this equaticns.

In order to simulate levels of consumption under equilibrium

prices, the following elasticities were used from the original equation:

180

Demand Elasticities

Consumption Own-Price Per Capita Income

Maize -1.689 40.917

Rice -1.256 +0.874

The con-price and income elasticities of total coarse grain cereals

consumption were used in place of the elasticities estimated directly for

maize because of the high correlation between the pr-ices of different coarse

grains and the more reasonable results that were obtained with all these

cereals aggregated together. These elasticities were incorporated into the

following regression equation:

InC*. - ,.lnP,, - mlnY+. = 0o + u. ...-(7)

where is the estimated intercept coefficient and ut is the residual.

Equation (7) was estimated for each food over the entire period (1955-85)

for which there are data on consumption, estimated as net availability in

Annex Table 5-1. Equilibrium prices at the retain level from Amiex Tables

3-4(1) and 3-4(2) were then inserted into this equation to determine the

direct and total effects of distortions on consumption.

The results are shown in Tables 26 and 27. Equilibrium levels of

consumption fluctuate widely from year to year, partly because of the

relatively high price elasticities that have been estimated and partly

because the magnitude of distortions varies with local production conditions

and the world market. As explained earlier, this is particularly true of

maize, for which imports are a small share of total consumption.

In the absence of cross-price effects, which make the influence of

price distortions on production relatively complicated, their direct and

indirect effects on consumption are fairly clear. In most years, especially

181

Table 26

Direct Effect on Consumption(000 *t)

Maize Rice

Year C (a) Ct (b) I C/C (c) C (a) CS (b A C/C (d)

1955 144 31 3.69 17 16 .071956 145 48 2.04 21 22 -.041957 145 19 6.67 31 31 .011958 156 50 2.12 31 31 -.011959 N/A N/A N/A N/A N/A N/A1960 N/A N/A N/A 46 42 .101961 N/A N/A N/A 62 58 .081962 144 66 1.18 88 79 .111963 156 366 -. 57 45 81 -. 451964 148 140 .06 61 101 -. 391965 176 602 -. 71 47 84 -.441966 305 1133 -. 73 62 101 -.381967 244 313 -. 22 63 81 -.231968 215 121 .78 53 42 .261969 257 267 -. 04 61 68 -.111970 415 1909 -. 78 80 119 -.331971 399 749 -. 47 65 113 -.421972 342 81 3.20 63 130 -.521973 364 1370 -.73 87 126 -. 311974 413 494 -.16 78 68 .151975 292 490 -.41 40 80 -.511976 254 1432 -.82 43 134 -.681977 233 1186 -.80 69 386 -. 821978 185 405 -.54 105 507 -. 791979 323 364 -.11 52 172 -. 701980 337 2762 -.88 109 893 -. 881981 348 5061 -.93 85 862 -.901982 376 6701 -.94 36 837 -.961983 259 19251 -. 99 55 2897 -. 981984 617 1997 -.69 87 637 -.861985 449 781 -.43 110 321 -. 66

Notes to Table 26:(a) Actual consueption C estiaated as net availability from Annex 6, Table 6-1.(b) Equilibrium consueption CS obtained from equation (7) as described in text.(c) Relative change in consumption A C/C obtained by subtracting equilibrius

consumption from actual consumption and dividing by equilibrium consumption.

182

Table 27

Total Effect on Consumption(000 It)

Maize Rice

Year C (a) CS (b) A C/C (c) C (a) Ct (b) 6 C/C (c)1-14-- N/A 17 N/A _ /-

1955 145 N/A N/A 17 N/A N/A1956 145 N/A N/A 21 N/A N/A1957 145 N/4 N/A 31 N/A N/A1958 156 37 3.19 31 25 .241959 N/A N/A N/A N/A N/A N/A1960 N/A N/A N/A 46 26 .761961 N/A N/A N/A 62 32 .921962 144 35 3.14 88 48 .831963 156 200 -.22 45 48 -.091964 148 60 1.45 61 52 .171965 176 156 .13 47 27 .731966 305 399 -.24 62 41 .521967 244 175 .39 63 49 .271968 215 64 2.34 53 26 1.061969 257 146 .76 61 43 .431970 415 1122 -.63 B0 73 .101971 399 414 -.04 65 69 -.061972 342 57 4.97 63 100 -.371973 364 1109 -.67 87 104 -.161974 413 354 .17 78 52 .511975 292 285 .02 40 51 -.231976 254 572 -.56 43 62 -.311977 233 187 .25 69 87 -.211978 185 45 3.08 105 89 .181979 323 66 3.92 52 46 .131980 337 319 .06 109 144 -.241981 348 191 .83 95 46 .841982 376 343 .10 36 41 -.121983 259 685 -.62 55 100 -.451984 617 385 .60 87 160 -.461985 449 N/A N/A 110 N/A N/A

--- -- -- -- --- -- -- -- -- --- -- -- -- -- --- -- -- -- -- --- -- _ - --_ - -- -- -- --

Notes to Table 27:(a) Actual consueption C estimated as net availability from Annex 6, Table 6-1.(b) Equilibrius consumption Ct obtained from equation (7) as described in text.lc) Relative change in consumption A C/C obtained by subtracting equilibrium

consueption from actual consumption and dividing by equilibrium consumption.

183

after 1962, import restrictims resulted in consumption of maize and rice

that was substantially below the levels that would have occurred at the

overvalued exchange rate if imports of these foods had been freely admitted

(see Table 26). On the other hand, the considering impact of all

distortions, including the overvalued rate of exchange, the picture is

mixed (Table 27). In 7 out of the 24 years for which comparisons can be

made, the equilibrium level of maize consumption would have been greater

than its actual level. In other years, Ghana would have cansumed so little

maize that it would have bEen self-sufficient or would have had a surplus

for export. The situation regarding rice is somewhat different, especially

after 1970 when Ghana in equilibrium would have consumed more rice that it

actually did in 10 out of 14 years. Furthermore, the fact that in

equilibrium there would have been a substantial growth in consumption from

the early 1960s to the early 1980s is consistent with the general perception

that rice consumption has been growing in West Africa for structural reasonis

that are independent of economic policy. *

Effect on Net Foreign Exchanae Earninis

The effect of policy on net foreign exchange earnings is estimated

by multiplying the changes in output and consumption that result from price

distortions, as shown in Table 20-27, times the relevant FOB or CIF price.

An adjustment is also made for the impact of the change in Ghan' s output

(exports) of cocoa on its world market price by multiplying that change

times the existing world price and dividing by the price elasticity of world

93 Scott R. Pearson, J. Dirck Stryker, Charles P. Ftunphreys, et al,Rice in West Africa: Policy and Economics, Stanford: Stanford UniversityPress, 1961.

164

demand for cocoa times total world consumption (see notes to Tables 2B-33).

In addition, there may be an effect on net foreign exchange earnings that

occurs because of changes in imports of intermediate inputs induced by the

changes in output. As discussed earlier, however, intermediate inputs are

unimportant for the crops considered here except for mechanized rice

cultivation. Even in this instance, the value of these inputs is only about

10 percent of the value of rice output measured in domestic prices. The

tradable component of these inputs, nmreaver, accounts for only perhaps

one-half of their total value, so that the adjustment necessary to take into

account changes in imported inputs would have a minimal influence on flows

of foreign exchange.

The results are presented in Tables 26-33 for direct and total

effects in the short, long, and very long runs. In general the net effects

of price interventions on foreign exchange earnings are positive in the

short run when only direct effects are considered and become increasingly

negative as the length of run increases and as indirect, as well as direct,

effects are taken into acccunt. For cocoa the effects of distortions on

foreign excnange earnings due to changes in output are usually negative, but

these effects tend to be at least partially offset by the positive effects

that lower Ghanaian cocoa exports would have had on world prices. On

balance, however, and especially after the mid-1970s, the net effect on

earnings from cocoa is negative..

As expected, this effect on cocoa earmings increases the longer

the period of supply response. Somewhat surprisingly, however, the direct

effect is greater than the total effect despite the fact that both direct

and indirect effects, in the case of cocoa, operate in the same direction.

185

Table 28

Direct, Short-Run Effect of Price Interventions on Foreign Exchange Earnings

Effect an Export Crop Revenue Due to: Prop of Total------------------------------------- Incremental Increaental Incremental Incremental Prop of Effect Effect on ForeignChange in Change in Prop of Effect Change in Change in Change in Change in Effect on on Food Crop in Exchange EarningsQuantity World Total on Export Crop maize Maize Rice Rice Imported Value of Total in Value of Total

Exported (a) Prices (b) Change (c) in Value of Tot Prod (e) Cons (f) Prod (g) Cons (h) Food Crops (i) Exports (I) Exports (ZI(bill USS1) (ill US$) (sill USS) Exports (%I (d) (000 at) (000 at) (000 at) (000 at) (sill US) li) (kI)

Year (11) (2) (3) (4) (5) (6) (7) (B) (9) . (10) (11)

1955 N/A N/A N/A N/A N/A 113 N/A 1 N/A N/A N/A1956 N/A N/A N/A N/A -194 97 0 -1 N/A N/A N/A1957 N/A N/A N/A N/A -69 126 4 0 N/A N/A N/A1958 -18 20 2 1 -292 106 4 0 -54 -18 -171959 -14 16 2 1 N/A N/A N/A N/A N/A N/A N/A1960 10 -13 -4 -1 N/A N/A N/A 4 N/A N/A N/A1961 -3 3 0 0 N/A N/A I 5 N/A N/A N/A1962 -11 13 2 1 N/A 78 2 9 N/A N/A N/A1963 -68 83 15 5 -55 -210 3 -37 14 4 9

00 1964 -20 29 9 3 8B 8 9 -40 16 5 81965 -49 44 -5 -2 48 -425 8 -37 37 12 101966 -96 99 2 1 251 -828 9 -39 71 25 261967 -76 81 4 1 206 -69 12 -18 20 7 91968 -51 42 -9 -3 122 94 11 11 4 1 -21969 -89 101 12 3 40 -10 9 -7 10 3 61970 -129 132 2 1 175 -1495 9 -39 68 16 161971 -87 94 6 2 381 -350 14 -48 60 19 201972 0 0 0 0 236 260 19 -67 7 2 21973 -187 161 -26 -4 -103 -1005 21 -39 70 12 81974 -171 157 -14 -2 392 -81 21 10 72 11 B1975 -131 117 -14 -2 18 -199 8 -41 69 9 71976 -451 340 -111 -14 203 -1179 25 -91 250 32 181977 -423 303 -120 -13 243 -953 44 -317 529 59 461978 -843 534 -309 -35 201 -219 64 -401 398 43 91979 -113 82 -32 -3 335 -41 53 -121 256 24 211980 -130 75 -54 -5 299 -2424 38 -784 1264 115 1101981 201 -105 96 14 342 -4713 10 -777 2012 283 2971982 -1745 705 -1040 -171 344 -6325 20 -800 1764 291 1191983 -3069 1228 -1841 -418 172 -18992 23 -2842 6204 1410 9921904 -83 35 -47 -8 573 -1380 38 -550 750 133 1241985 -50 20 -30 -5 431 -332 52 -211 257 41 36

Notes to Table 28:(a) Difference between actual output and equilibrius output of cocoa

(from Table 20) eultiplied by the FOB price of cocoa (from Table 3-4(3)),divided by the official exchange rate (from Table 4).

(b) Actual output of cocoa (from Table 20) *ultiplied by the change in its

in its FOB price, expressed in US$ at the official exchange rate. This priceprice change equals coluen (1) divided by 0.3 tiees total world exports of cocoa(froa Sill & Duffus, Cocoa Statistics, April 1981).

(c) Column (1) plus column (21.(d) Coluen (31 divided by Exports FOB (from Table 2-2).(el Difference between actual output and equilibrium output of maize (from Table 20).

(f) Difference between actual consumption and equilibrium consumption of maize (from Table 26).

lg) Difference between actual output and equilibrium output of rice (from Table 20) multiplied

by .65 to convert paddy into rice.(hI Difference between actual consumption and equilibrium consueption of rice (from Table 26).Ii) Column (5) - column (6) times CIF saize lconverted to USH using the OER)

plus coluen (7) - column (1) times CIF rice (converted to US, using the OER).(jI Column (9) divided by Exports FOB (from Table 2-2).Ik) The sue of coluens (4) and (10).

187

Table 29

Direct, Long-Run Effect of Price Interventions on Foreign Exchange Earnings

Effect on Export Crop Revenue Due to: Prop of Total--------------------------------- Incremental Incremental Incremental Incresental Prop of Effect Eftect on Foreign

Change in Change in Prop of Effect Change in Change in Change in Change in Effect on on Food Crop in Exchange EarningsQuantity World Total on Export Crop Haile Naize Rice Rice Imported Value of Total in Value of Total

Exported (a) Prices (b) Change (c) in Value of Tot Prod (e) Cons (f) Prod (g) Cons (h) Food Crops (i) Exports (Z) Exports (2)(till UBS) (sill US$) (mill US$) Exports (%) (d) (000 Mt) (000 at) (000 at) (000 Mt) (mill USS) (i (k)

Year (1) (2) (3) (4) (5) (6) (7) (9) (9 (10) (11)

1955 N/A N/A N/A N/A N/A 113 NIA I N/A N/A N/A1956 N/A N/A N/A N/A -194 97 .5 -1 NIA N/A N/A1957 N/A N/A N/A N/A -96 126 4 .2 N/A N/A N/A1959 -19 20 2 1 -323 106 6 -.3 -59 -19 -191959 -26 30 4 1 N/A N/A N/A N/A N/A NIA N/A1960 -7 9 3 1 N/A N/A -21 4 NIA N/A N/A1961 -7 9 I 0 N/A N/A -3 5 N/A N/A NIA1962 -J 6 19 3 1 -462 79 1 9 -70 -22 -211963 -92 101 19 6 -104 -210 3 -37 11 4 10

OD 1964 -73 109 34 11 93 9 9 -40 15 5 1500 1965 -91 91 -10 -3 62 -425 9 -37 38 12 9

1966 -206 211 4 2 256 -928 10 -39 72 26 271967 -201 296 15 5 219 -69 15 -19 21 7 131969 -223 195 -39 -13 149 94 14 It 9 2 -101969 -319 361 43 12 71 -10 13 -7 15 4 17-1970 -360 367 7 2 186 -1495 11 -39 69 16 191971 -401 430 29 9 387 -350 16 -49 61 19 271972 -228 207 -21 -5 2173 260 23 -67 20 5 01973 -469 402 -66 -11. -24 -1005 24 -39 76 13 21974 -692 635 -57 -9 391 -81 26 10 74 11 31975 -616 550 -65 -9 214 -199 Is -41, 75 9 11976 -1267 954 -312) -40 214 -1179 27 -91 253 32 -e1977 -1761 1262 -499 -56 249 -953 49 -317 533 60 41979 -3699 2335 -1353 -152 206 -219 66 -401 391 44 -1091979 -1381 994 -397 -36 350 -41 56 -121 265 25 -111990 -755 438 -317 -29 324 -2424 44 -794 1275 116 971991 15 -9 7 1 350 -4713 31 -777 2026 295 2961992 -1910 772 -1139 -195 345 -6325 21 -900 1764 291 1031 993V -6615- 2647 -3969 -902 173 -19992 24 -2842 6204 1410 5081994 -742 317 -425 -75 574 -1390 41 -550 751 133 591985 -604 243 -361 -57 437 -332 55 -211 260 41 -16

Notes to Table 29:(a) Difference between actual output and equilibrium output of cocoa

(froe Table 21) eultiplied by the FOR price of cocoa (from Table 3-4(3)),divided by the official exchange rate (froe Table 4).

(b) Actual output of cocoa (froe Table 21) multiplied by the change in itsin its FOB price, expressed in US$ at the official exchange rate. This priceprice change equals column (1) divided by 0.3 times total world exports of cocoaffrom Bill & Duffus, Cocoa Statistics, April 1981).

(c) Column (1) plus column (2).(d) Column (3) divided by Exports FOB (from Table 2-2).(e) Difference between actual output and equilibrium output of maize (from Table 21).(f) Difference between actual consumption and equilibrium consumption of maize (from Table 26).(g) Difference between actual output and equilibrium output of rice (from Table 21) multiplied

by .65 to convert paddy into rice.(h) Difference between actual consumption and equilibrium consumption of rice (from Table 26).li) Coluen (5) - coluen (6) times CIF maize (converted to US$ using the OER)

plus column (7) - coluen (8) times CIF rice (converted to USS using the OER).(j) Column (9) divided by Exports FOB (from Table 2-2).(k) The sum of columns (4) and (10).

189

Table 30

Direct, Very Long-Run Effect of Price Interventions on Foreign Exchange Earnings

Effect on Export Crop Revenue Due to: Prop of Total--------------------------------- Incremental Incremental Incremental Incremental Prop of Effect Effect on Foreign

Change in Change in Prop of Effect Change in Change in Change in Change in Effect on on Food Crop in Exchange EarningsQuantity World Total on Export Crop flaize Naize Rice Rice Imported Value of Total in Value of Total

Exported (a) Prices (b) Change (c) in Value of Tot Prod (e) Cons (f) Prod (g) Cons (h) Food Crops (i) Exports (Z) Exports (ii(mill USS) (sill US$) (sill US$) Exports CX) (d) (000 Mt) (000 at) (000 Mt) (000 it) (mill US$1 (i (k)

Year (1) (2) (3) (4) (5) (6) (7) (B) (9) (10) (11)

1955 N/A N/A N/A N/A N/A 113 N/A I */A N/A N/A1956 N/A N/A N/A NIA -194 97 .5 -1 NIA N/A N/A1957 N/A N/A N/A N/A -96 126 4 .2 N/A N/A N/A1959 -19 20 2 1 -323 106 6 -.3 -59 -19 -191959 -26 31 4 1 N/A NIA N/A N/A N/A N/A N/A1960 -7 9 3 1 N/A N/A -21 4 N/A NIA N/A

1961 ~~-7 9 I 0 N/A N/A -3 5 N/A N/A N/A1962 -16 19 3 1 -462 78 1 9 -70 -22 -21

J 1963 -82 101 19 6 -104 -210 3 -37 it 4 10CD1964 -73 109 34 11 93 9 9 -40 1s 5 15

1965 -91 82 -10 -3 62 -425 9 -37 39 12 91966 -206 211 4 2 256 -028 10 -39 72 26 271967 -292 298 15 5 219 -69 15 -19 21 7 131968 -227 189 -39 -13 149 94 14 11 9 2 -101969 -329 373 44 13 71 -10 13 -7 15 4 171970 -378 395 7 2 196 -1495 11 -39 69 16 le1971 -436 469 31 9 387 -350 lb -49 61 19 27197 -271 24 6 -25 -6 273 260 23 -67 20 5 -11973 -572 492 -90 -14 -24 -1005 24 -39 76 13 -11974 -909 833 -74 -11 391 -91 26 10 74 11 01975 -997 793 -94 -12 214 -199 15 -41 75 9 -21976 -1997 1429 -468 -60 214 -1179 27 -91 253 32 -291977 -2960 2049 -910 -91 249 -953 49 -317 533 60 -31197 -6490 4102 -2378 -266 206 -219 66 -401 391 44 -2231979 -2932 2039 -794 -74 350 -41 56 -121 265 25 -501990 -1726 1001 -725 -66 324 -2424 44 -794 1275 116 501991 -328 171 -157 -22 350 -4713 31 -777 2026 295 2631982 -5513 2227 -3296 -541 345 -6325 21 -900 1764 291 -2511993 -15979 6355 -1525 -2165 173 -18992 24 -2942 6204 1410 -7551984 -1996 949 -1139 -201 574 -1390 41 -550 751 133 -691995 -1956 747 -1109 -175 437 -332 55 -211 260 41 -134

Notes to Table 30:(a) Difference between actual output and equilibrium output of cocoa

(from Table 22) eultiplied by the FOB price of cocoa (froe Table 3-4(3)),

divided by the official exchange rate (from Table 4).

(b) ctual output of cocoa (from Table 22) multiplied by the change in itsin its FOB price, expressed in USf at the official exchange rate. This price

price change equals column (1) divided by 0.3 times total world exports of cocoa

(from Sill I Duffus, Cocoa Statistics, April 1991).

kc1 Column (1) plus column (2).

(d) Coluen (3) divided by Exports FOB (from Table 2-2).

(e) Difference between actual output and equilibrium output of maize (from Table 22).(f) Difference between actual consumption and equilibrium consumption of maize (from Table 26).(I) Difference between actual output and equilibrius output of rice (fros Table 22) multiplied

by .65 to convert paddy to rice.(h) Difference between actual consueption and equilibriua consumption of rice (from Table 26).(i) Column (5) - coluen (6) times CIF maize (converted to USH using the OER)

plus column (7) - coluen (8) times CIF rice (converted to USS usinq the OER).(j) Coluen (1) divided by Exports FOB (from Table 2-2).(k) The sum of columns (4) and (10).

191

Table 31

Total, Short-Run Effect of Price Interventions on Foreign Exchange Earnings

Effect on Export Crop Revenue Due to: Prop of Total------------------------------------- . Incremntal Incremental Increwntal Incremental Prop of Effect Effect on ForeignChange in Change in Prop of Effect Change in Change in Change in Change in Effect on on Food Crop in Exchange EarningsQuantity World Total on Export Crop Naize Ibize Rice Rice lported Value of Total in Yalue of Total

Exported (a) Prices (bi Change (c) in Value of Tot Prod (e) Cons (fI Prod (gi Cons (hi Food Crops lil Exports (1i Exports (X)(mill USS) Imill USS) (mill USSI Exports (XI (di (000 Mt) 1000 at) (000 at) (000 *tl (mill US$) (ij (kI

Year (1) (2) (3) (4) (5) (6) (7) (9) (9) (10) (11)

1955 NIA N/A N/A N/A N/A N/A N/A N/A NIA N/A NIA1956 NIA N/A N/A N/A -237 N/A -2 N/A NIA N/A N/A1957 N/A N/A NIA N/A -92 N/A 3 N/A N/A N/A N/A1958 -4 5 0 0 -311 11S 3 6 -60 -20 -201959 -2 2 0 0 NIA N/A N/A N/A N/A N/A N/A1960 21 -28 -8 -2 N/A N/A N/A 20 N/A N/A N/A1961 7 -8 -1 0 N/A N/A -4 30 N/A N/A N/A1962 2 -3 -1 0 N/A 109 -5 40 N/A N/A N/A

vo 1963 -46 56 10 3 -180 -44 -4 -4 -7 -2 11964 5 -7 -2 -1 29 87 1 9 -8 -2 -31965 -36 32 -4 -1 -29 21 4 20 -6 -2 -31966 -55 56 1 0 61 -94 0 21 5 2 21967 -23 25 1 0 85 69 -3 13 -2 -1 01968 -4 3 -1 0 24 151 2 27 -22 -7 -71969 -51 57 7 2 -84 111 -2 19 -30 -9 -71970 -66 67 1 0 25 -707 0 7 24 6 61971 -9 9 1 0 288 -15 -1 -4 21 6 71972 63 -57 6 2 82 294 -4 -37 -63 -i6 -i51973 -55 47 -8 -1 -297 -745 13 -17 36 6 51974 -60 55 -5 -1 355 59 12 26 35 5 41975 6 -5 1 0 98 6 -11 -12 13 2 21976 -210 159 -52 -7 150 -319 14 -19 82 11 41977 15 -11 4 0 205 46 19 -19 66 7 9.1978 -171 108 -63 -7 156 140 29 16 12 1 -61979 35 -25 10 1 202 257 13 6 -25 -2 -1

-1980 -102 59 -43 -4 217 19 12 -35 86 8 41981 16 -e 9 1 312 157 15 39 34 5 61982 -190 73 -107 -19 291 33 -2 -5 54 9 -91983 -866 347 -519 -118 155 -426 9 -46 172 39 -79I984 -34 14 -19 -3 551 231 19 -73 121 21 1l1985 N/A N/A N/A NIA N/A N/A N/A N/A N/A N/A N/A

Notes to Tible 31:(a) Difference between actual output and equilibrium output of cocoa

(from Table 23) sultiplied by the FOB price of cocoa (from Table 3-4(3)),

divided by the official exchange rate (from Table 4).(bl ctual output of cocoa (from Table 23) multiplied by the change in its

in its FOI price, expressed in US5 at the official exchange rate. This priceprice change equals column (1) divided by 0.3 tiees total world exports of(from Sill I Duffus, Cocoa Statistics, April 1981).

Ic) Coluen (1) plus column (2).(d) Coluen (3) divided by Exports FOB (from Table 2-2).(el Difference between actual output and equilibrium output of maize (from Table 23).(ft Difference between actual consumption and equilibrium consumption of maize (from Table 27).(g) Difference between actual output and equilibrium output of rice (from Table 231 *ultiplied

by b5 to convert padddy to rice.(h) Difference between actual consumption and equilibrium consueption of rice (from Table 27).(it Coluen (5) - coluen (6) times CIF maize (converted to US$ using the GER)

plus column (7) - column (8) times CIF rice (converted to USS using the OER).(jt Column (9) divided by Exports FOB (from Table 2-2).(k) The sum of columns (4) and (10).

193

Table 32

Total, Log-Run Effect af Price Interventions an Foreign Exchange Earnings

Effect on Export Crop Revenue Due to: Prop of Total--------------------------------- Incrementil Incre,ental Incremental Incremental Prop of Effect Effect on Foreign

Change in Change in Prop of Effect Change in Change in Change in Change in Effect on on Food Crop in Exchange Earningsguantity World Total on Export Crop Maize maize Rice Rice Imported Value of Total in Value of Total

Exported (a) Prices (b) Change (c) in Yalue of Tot Prod (e) Cons (fi Prod (gI Cons (h) Food Crops (i) Exports (Z) Exports (ZI(mill US$1 (mill US$) (mill US1) Exports (ZI (d) (000 St) (000 at) (000 at) (000 at) (gill US$) (il (k)

Year (1) (2) (3) (4) (5) (6) (7) (B) (9) (10) (11)…-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

1955 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A1956 N/A N/A N/A N/A -237 N/A -2 N/A N/A N/A N/A1957 N/A N/A N/A N/A -126 N/A 2 N/A N/A N/A N/A1958 19 -20 -2 -1 -352 119 4 6 -65 -22 -221959 10 -12 -2 0 N/A N/A N/A N/A N/A N/A NIA1960 27 -37 -10 -3 N/A N/A -30 20 N/A N/A N/A1961 22 -26 -3 -1 N/A N/A -10 30 N/A N/A NIA1962 le -22 -4 -1 -931 109 -B 40 -139 -43 -451963 -30 37 7 2 -290 -44 -6 -4 -12 -4 -2

~* 1964 -14 20 6 2 9 97 0 9 -10 -3 -11965 -43 39 -5 -1 -27 21 4 20 -6 -2 -31966 -102 104 2 1 56 -94 1 21 5 2 21967 -116 122 6 2 90 69 -2 13 -2 -1 219b9 -70 59 -12 -4 36 151 2 27 -20 -7 -111969 -126 143 17 5 -75 III -1 1e -21 -9 -31970 -153 156 3 1 11 -707 0 7 24 6 61971 -131 141 9 3 259 -15 -1 -4 21 6 91972 -19 17 -2 0 123 254 -4 -37 -49 -13 -131973 -175 151 -25 -4 -251 -745 13 -17 39 6 21974 -310 284 -25 -4 346 59 15 26 35 5 11975 -256 229 -27 -3 139 6 -6 -12 20 3 -11976 -659 497 -163 -21 -160 -319 14 -19 93 11 -101977 -520 373 -147 -17 212 46 23 -19 70 9 -91979 -961 609 -353 -40 169 140 33 16 19 2 -371979 -334 240 -94 -9 236 257 20 6 -5 -1 -91990 -297 172 -125 -11 239 19 16 -35 94 9 -31991 -139 72 -66 -9 321 157 19 39 39 6 -41982 -909 327 -482 -79 304 33 0 -5 57 9 -701913 -2011 905 -1206 -274 160 -426 9 -46 174 39 -2351994 -264 113 -151 -27 556 231 21 -73 124 22 -51995 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

…-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Notes to Table 32:(a) Difference between actual output and equilibrium output of cocoa

(froo Table 24) eultiplied by the FOB price of cocoa (from Table 3-4(3)),divided by the official exchange rate (from Table 4).

fb) Actual output of cocoa (from Table 24) multiplied by the change in itsin its FOB price, expressed in USS at the official exchange rate. This priceprice change equals column (1) divided by 0.3 times total world exports of cocoa(from Bill & Dulfus, Cocoa Statistics, April 19B1).

(c) Column (1 plus column (2).Cd) Coluen (3) divided by Exports FOB (from Table 2-2).le) Difference between actual output and equilibrium output of maize (frog Table 24).(f) Difference between actual consueption and equilibrium consumption of maize (from Table 27).(g) Difference between actual output and equilibrium output of rice (from Table 24) multiplied

by .65 to convert paddy to rice.(h) Difference between actual consumption and equilibrium consumption of rice (from Table 27).ti) Coluen (5) - column (6) times CIF maize (converted to US$ using the OER)

plus coluen (7) - column (8) times CIF rice (converted to USt using the OER).

1;) Column (9) divided by Exports FOB (from Table 2-2).

(k) The sum of columns (4) and (10).

195

Table 33

Total, Very Long-Run Effect of Price Interventionts on Foreign Exchange Earnings

Effect on Export Crop Revenue Due to: Prop of Total--------------------------------- Incremental Incremental Incremental Incremental Prop of Effect Effect on Foreign

Change in Change in Prop of Effect Change ini Change in Change in Change in Effect on on Food Crap in Exchange Earningsguantity World Total on Export Crop Maize Maize Rice Rice Imported Value of Total in Value of Total

Exported (a) Prices (b) Change (c) in Value of Tot Prod (e) Cons (f) Prod (g) Cons (h) Food Crops (i) Exports (Z) Exports (2)(mill US$1 (mill US$1 (mill USf) Exports (21 (d) (000 at) (000 at) (000 St) (000 at) (mill US$1 (i)(

Year (1) (2) (3) (4) (5) (6) (7) (B) (9) (10) (1

1955 -------- N/A---- ------- --------- N/A---- N---- A- N---- A- N/A---- N/A----- ------- A-- ---- ------1955 N/A N/A N/A N/A -237 N/A -2A N/A N/A N/A N/A1957 N/A N/A N/A N/A -1267 N/A -2 N/A NIA N/A N/A

1959 19 ~ 20 -2 -I -352 119 4 6 -65 -22 -221959 10 -11 -2 0 N/A N/A N/A N/A N/A N/A N/A1960 27 -37 -10 -3 N/A N/A -30 20 N/A NIA N/A1961 22 -26 -3 -1 N/A N/A -10 30 N/A N/A N/A1962 19 -22 -4 -1 -931 109 -9 40 -139 -43 -451963 -30 37 7 2 -290 -44 -6 -4 -12 -4 -2

a' 164 -14 20 6 2 9 87 -.2 9 -10 -3 -11965 -43 39 -5 -1 -27 21 4 20 -6 -2 -31966 -102 104 2 1 56 -94 1 21 5 2 21967 -117 123 6 2 90 *69 -2 13 -2 -1 21969 -72 60 -12 -4 36 151 2 27 -20 -7 -111969 -135 153 19 5 -75 III -1 18 -29 -9 -31970 -170 173 3 1 11 -707 -.4 7 24 6 61971 -164 176 12 4 289 -15 -1 -4 21 6 101972 -46 42 -4 -i 123 294 -4 -37 -49 -13 -141973 -240 206 -34 -6 -251 -745 13 -17 39 6 11974 -437 401 -36 -5 346 59 15 26 35 5 01975 -397 355 -42 -5 138 6 -6 -12 20 3 -31976 -953 718 -235 -30 160 -319 14 -19 93 11 -191977 -1117 800 -317 -36 212 46 23 -19 70 9 -291979 -2617 1657 -960 -108 168 140 33 16 19 2 -1051979 -1345 968 -377 -35 236 257 20 6 -s -1 -361980 -1111 644 -467 -42 239 18 16 -35 94 9 -341991 -715 372 -342 -49 321 157 19 39 39 6 -431992 -4057 1639 -2418 -399 304 33 .1 -5 57 9 -38919B3 -7999 3157 -4732 -1075 160 -426 9 -46 174 39 -10361994 -1157 494 -663 -117 559 231 21 -73 124 22 -951995 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

Motes to Table 33:(a) Difference between actual output and equilibrium output of cocoa

(from Table 25) multiplied by the FOB price of cocoa (from Table 3-4(3)),

divided by the official exchange rate (from Table 4).(b) Actual output of cocoa (from Table 25) *ultiplied by the change in its

in its FOB price, expressed in USS at the official exchange rate. This priceprice change equals column (1) divided by 0.3 times total world exports of cocoa(from Bill i Duffus, Cocoa Statistics, April 1981).

(cl Column (1) plus column (2).ld) Column (3) divided by Exports FOB (from Table 2-2).(e) Difference between actual output and equilibrium output of maize (from Table 25).(t) Difference between actual consumption and equilibrium consumption of maize (from Table 27).19) Difference between actual output and equilibrium output of rice (from Table 25) multiplied

by .65 to convert paddy to rice.(h) Difference between actual consumption and equilibrium consumption of rice (from Table 27).lil Column (5) - column (6) times CIF maize (converted to USH using the OER)

plus column (7) - column (8) times CIF rice (converted to US$ using the OER).Ii) Column (9) divided by Exports FOB (from Table 2-2).lk) The sum of columns (4) and (10).

197

That is, both export taxation and overvaluation of the exchange rate tended

to rechce foreign exchange earnings frtm cot:oa.. The reascn for this

apparent ananly is that the equilibrium level olf cocoa output that takes

into account distortions in the exchange rate also assumes that cocoa

exports are being taxed at the optimal rate. When the direct effect of

price interventicons alone is calculated, on the other hand, distortions are

measured with respect to the border price at the official exchange rate in

the absence of any export tax.19

The impact of policy intervention on net foreign exchange earnings

from food crops is also somewhat complicated. The direct effect of import

restrictions, coupled with the overvalued exchange rate, has generally been

to raise domestic prices, increase production ard reduce consumption. This

had tended to save foreign exchange, especially in later years. When the

total effect is considered, hc3wever, the direct ard indirect effects work in

opposite directions, resulting in an impact on foreign exchange earnings

that varies in direction from year to year, thouqh it becomes increasingly

positive tcwards the end of the period, and is on balance relatively smal 1

in magnitude.

This implies that the overall impact of interventions on net

foreign exchange earnings is dominated by food crops when only direct

effects are ccnsidered. It thus tends to be positive in the short rmun but

becomes increasingly negative as the length of rixi increases and the effects

on cocoa productions assume greater importance. The total effects, on the

'-' This is because the notion of optimality implies equating socialmarginal revenue and social marginal cost to detLermine the optimum exporttax rate. Eht this equaticn is only meaningful, in the first-best worldconsidered here, at the equilibrium exchange rate.

198

other hand, are more quickly dominated by cocoa as the supply respmse

lengthens, and the detrimental impact of policy on foreign exchange earnings

becomes more immediately obvious.

199

CHAPTER VIII: GIER%JE T ELDGET PM OTHER FESORE TRASERS

In addition to affecting production, cansumption, and net eamnings

of foreign exchange, price policy interventions have also resulted in

transfers of rescurces and welfare between the grvernmnmt, producers, and

ccnsumers. The next sectimn deals with the budgetary consequences of

agricultural price policies in Ghana. This is followed by an analysis of

the impact of these and other policies on transfers of rescurces between

agriculture and the rest of the ecaomwny.

Effects of Price Policy cn the GEbverrnoent azdcqet

The major effect of price policy in Ghana an the goverTnent

budget is the revenue that is earned an cocoa exports through the operations

of the Coroa Marketing Board. Although data on actual fiscal receipts are

unavailable, there is informatimn on CMB costs, and net revenu.e can be

estimated by subtracting these costs plus income to producers from the FoB

value of marketed cocoa. As discussed earlier-, CMB costs have varied aver

time as the responsibilities of the Board have changed. In addition to

marketing and quality control, these responsibilities have at various times

included research, extension, plantatimn management, and processing. At

other times these functions have been performed by the Ministry of

Agriculture or other public agencies, in which case CMB costs were lower and

revenue to the government was higher than when the CMB was directly

respcnsible for these activities.

Data on taxes and marketing profits for rice and maize imports are

unavailable. The schedule of custcms and excise tariffs published for 1973

lists a 50 percent ad valorum duty an imports of maize and a Nf 0.025/lb

200

specific tax on rice imports. - These taxes do not apply to food aid, nor

is there any sales tax on cereals imports. Actual tax receipts from maize

and rice imports in 1972 are estimated at Ni 1.18 million.2 This is less

than 1 percent of the revenue from cocoa exports. Given the low level of

maize imports and the fact that inflation rapidly decreased the real value

of the import tax on rice, these sources of government revenue appears to

have been of minor importance.

Agricultural inputs are exempt from tariffs. Subsidies exist, as

described earlier, but it is impossible to put together consistent time

series on their total amount. Nb effort has therefore been made to

separate them from other government transfers and expenditures.

As seen in Table 34, government revenue from cocoa has been an

important part of total public resource inflows. Its share of total revenue

has fluctuated, without any clear trend, between -63 percent and +69

percent, but in most years it has been between +20 and +50 percent. This is

a high export tax share by the standards of most African ccuintries. Cocoa

revenue as a proportion of the budget deficit has in many years exceeded 100

percent.

The per unit tax on cocoa, shown in Table 35, has increased in

nominal terms more rapidly than the total tax as a result of the decline in

exports. In 1983, the per unit tax was 129 times its nominal value in 1955,

whereas total tax revenue from cocoa was 98 times its value in 1955.

As a percentage of the FOB price, the per unit tax an cocoa has

fluctuated widely, but there are several movements of importance that

± Republic of Ghana, Customs and Excise Tariff. 1973

2 Data obtained frnm the Ministry of Finance.

201

Table 34

Effect of the Pricing Policy an the Budget(million NC)

Net Revenue as aProportion of (I)

Total (Net) Total BudgetRevenue (a) Budget (b) Deficit (c)

Year (1) (2) (3)

1955 23 22 -2311956 20 20 ;L781957 67 56 -4|461958 70 52 -6991959 57 40 3551960 36 21 731961 24 16 311962 32 20 321963 48 20 1111964 26 9 301965 13 6 311966 47 18 691967 111 37 1091968 143 43 2261969 201 46 6491970 118 26 1611971 146 35 1181972 133 34 851973 230 40 1311974 350 43 1041975 226 26 371976 518 45 521977 446 32 251978 1,784 69 1051979 969 33 561980 -362 -11 -81981 -3,026 -63 -641982 1,496 15 331983 2,249 10 4151984 1,758 5 251985 2,027 N/A N/A

Notes to Table 34:(a) FOB Price of Cocoa (from Table 3-4(3)) minus Rendered

Port Price of Cocoa (from Table 3-3(2)) multipliedby Official Cocoa Marketing Board production estieatesbased on sarketings (from Table 1-23)).

(b) Coluen (1) divided by 6overnment Revenue (from Table 3)in the succeeding year.

lc) Column i1) divided by I-) Budget Surplus (from Table 3)in the succeeding year.

202

Table 35

Tax on Cocoa

Per Unit Tax on Cocoa Total Tax- --------------------- on Cocoa

(NC/at) (a) I FOB (b) (mill NC) (c)Year (1) (2) (3)

1955 101 23 231956 74 19 201957 326 52 671958 273 50 701959 1.79 41 571960 83 24 361961 59 19 241962 78 23 321964 112 31 481963 49 18 261965 32 12 131966 125 32 471967 262 47 1111968 428 56 1431969 480 59 2011970 271 42 1181971 307 45 146I172 312 38 1331973 644 50 2701974 885 52 3501975 544 36 2261976 1528 59 5161977 1538 39 4461978 6237 60 17841979 3000 33 9691980 -1273 -20 -3621981 -12200 -244 -30261982 7558 25 14961983 13000 22 22491984 9299 11 17581985 8967 7 2027

Notes to Table 35:(a) FOB Price of Cocoa from Table 3-4(3) minus

Rendered Port Price of Cocoa fromTable 3-3(2).

(b) Coluen (1) divided by FOB Price of Cocoafrom Table 3-4(3).

(c1 FOB Price of Cocoa from Table 3-4(3) minusRendered Port Price of Cocoa fromTable 3-3(2) multiplied by official CocoaMarketing Board production estimates basedon marketings from Table 1-2(3).

203

should be noted in Table 35. First is the relatively low tax rate from

1960 to 1865, years in which Nkrumah is generally thought to have been

exploiting cocoa farmers. In fact this was a period when world market

prices were low, and the goverrwnent suffered alung with farmers. Following

this, the tax rate rose to a very high level, averaging 49 percent from 1967

to 1978.5 Thereafter, as the exchange rate became increasingly overvalued

lowering the real value of the FUH price for cocoa, the tax rate on cocoa

declined.

Transfers of Rescurces Between (Ariculture and the Rest of the Economy

There are a number of ways in which government policy has

transferred resources to and from Ghanaian agriculture. These include the

direct effects of export taxes and trade and exchange controls on output

prices and quantities, the indirect effects of the overvalued cedi on the

prices and quantities of outputs and inputs, and government recurrent and

investment expenditures.

Estimates of Rescuirce Transfers

The direct and indirect effects of trade and exchange taxes and

ccntrols are calculated in Annex 7. Separate estimates are made of changes

in producer surplus for maize (Tables 7-1(1) and 7-1(2)), rice (Tables

7-2(1) and 7-2(2)), and cocoa (Tables 7-3(1) thrnugh 7-3(3)) for the short,

long, and very long run. Using a simple, partial-equilibrium dema-id and

supply model, transfers to agriculture resulting frtm the direct effects of

distortions are given by

ULhweighted average of yearly rates.

204

(P - P ) Q + 1/2 (P - P ) (Q - Q) ...(a)

where P is the domestic price, P is the border price at the official

exchange rate, 0 is actual productim, and Q is production in the absence

of price and trade distortions at the official exchange rate. Transfers

from agriculture due to the indirect effects of the overvalued exchange

rate, are given by

(P* - P )Q* + 1/2 (P* - p )(Q* - Q*) (9)

where P* is the producer price equivalent of the border price adjusted by

multiplying the border price tby the ratio of the equilibrium exchange rate,

calculated from the simulation model described in Chapter II, to the

official exchange rate, and Q* is equilbrium production from Chapter VII.

For maize and rice, each of these two expressicns is generally

positive, especially when distorticns have been severe, and the direct and

indirect effects of these distortions tend to offset mne another. With

cocoa, an the other hand, the direct effect is in mo>st instances negative,

since the domestic price is generally less than the border price because of

the export tax. As a result, both effects tend to transfer rescuirces away

from the agricultural sector. The only exception to this is years in which

the introductian of an optimal export tax on cocoa offsets the overvaluation

of the exchange rate.

Direct and indirect producer price transfers also exist for

tradable products other than maize, rice, and cocoa. These, however, are

relatively unimportant. Production of the industrial crops - cottan,

rubber, and tobacco - has never equaled more than about 5 percent of that

of cocoa. Most production and processing of oilseeds is artisanal and

results in products that are not traded internaticnally. Domestic prices of

205

these products are influenced by gDvernnent pDlicies with respect to imports

of edible oils and tallow for the manufacture of soap, but the cross-price

elasticities of supply are unknown, and the intricacies of the input-output

relationships and uhbstitution effects involved are such that a detailed

study of this sector would be necessary in order to estimate the effects of

government policy on resource transfers.

Sugar is another industrial product for which price policy could

be important. It is an integrated industry, howdever, and price policy

applies to output of the sugar mills rather than to sugar cane. Without a

detailed study of the mills, therefore, it is impossible to assess the

transfer of resources to or frnom agriculture. This is not a critical

problem, however, since the total value of sugar cane production during

1965-1968 was never more than 4 percent of the value of cocoa production,

and output has subsequently been much lower.<

The rest of agricultural production is made up largely of

nontradable food crops. These are not directly subject to price policy.

They are influenced indirectly via substitution in production and

conssumption, though the supply and demand analysis preseted earlier

suggests that these effects are not very great.

Taxes and subsidies on agricultural inputs are also direct and

indirect. There are no import tariffs or other taxes on fertilizers,

insecticides, and agricultural machinery and equipment. There are explicit

subsidies on these inputs, but import restrictions prevent most producers

from having access to them at subsidized prices. As a result, the prices

4 M.S. Singal and J.D.N. Nartey, Sources and Metheds of Estimation ofNational Income at Current Prices in Ghana, Ghana Central. Bureau ofStatistics, 1971, pp.3-13.

206

for these inputs an the private market are quite high. Data an these

prices are extremely limited, and costructian of a time series is not

possible. In any event, the quantities involved are minimal since most

farmers do not use any of these inputs, and when they do, their value is

mnly a small part of the total cost of production.

Direct subsidies an agricultural inputs are included with other

government expenditures. An indirect subsidy also exists because of the

overvalued exchange rate. This is estimated in Table 7-4, Annex 7, by

multiplying the actual CIF value of imports of agricultural inputs times the

ratio of the equilibrium exchange rate to the official rate. The actual

value of imports is thEn subtracted fram the adjusted value to arrive at the

total value of the implicit subsidy on fertilizers, agricultural chemicals,

and agricultural machinery and equipment.

Gorvernment expenditure transfers are calculated in Table 7-5,

-nnex 7, for both recurrent and investment expenditures. These are derived

from separate budgets that are used to distinguish ongoing activities from

those that are developmental in nature. Included in the investment budget

are some current, in addition to capital, expenditures. Since these

normally are start-up costs, however, they may be cansidered as a form of

investment.

Transfers to agriculture include expenditures en roads and

waterways, in additicn to those en agriculture and non-mineral rescurces.

While not all of these transportatimn expenditures benefit agriculture, many

do, and it is impossible to isolate expenditures more closely associated

with agriculture, such as those en feeder roads. On the other hand, there

may be other government expenditures that are not included but that do

207

benefit agriculture. Examples are rail transport and storage. These are

less directly related to agriculture than roacds, however, and they have

therefore been excluded.

The results of this analysis are shown in Tables 36 through 41.

The first three tables, based an Annex 7, show the direct and total nominal

transfers out of and into agriculture because of distortions affecting

maize, rice, and cocoa in the short, long, and very lng runs. The last

three tables add to these transfers those that are not related to output

price.

Tables 36 thrcugh 38 show the magnitude of the net transfers out

of agriculture due principally to the export tax on, cocoa and the overvalued

exchange rate. Although both direct and total transfers have frequently

been positive for maize and rice because of the influence of trade controls,

these have been dominated in most years by the direct and total transfers

out of the cocoa sector. The major exception to this is the period during

the early 1980s when the extent of overvaluation was so great that it was

difficult for the governnent to extract a positive export tax from the cocoa

sector and the price of maize and rice farmers marketing their crops in

Accra would have been zero, as explained in the previous chapter, if they

had not received any trade proterticn. With the excepticn of these years,

there was always an overall price-related direct transfer out of

agriculture.

Even during these years, the direction of total transfer was out

of agriculture cnce the exchange rate effect is combined with the direct

effect. COn the other hand, there are a few years when total price-related

transfers into agriculture were positive (e.g., 1975 and 1977) . These are

20B

Table 36

Direct and Total Nosinal Short-Run Transfers Due to Output Price InterventionsInto (+) and Out of (-) Agriculture

(lillion NC)

Sum of Transfers asImportables Exportables All Products a Share of GDP (Z)

Maize Rice CocoaDirect Total Direct Total Direct Total Direct Total Direct Total

Year (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1959 -16.61 -28.00 -.01 -.52 -73.1 -54.3 -89.8 -82.9 -11.7 -10.81959 N/A N/A N/A N/A -56.8 -54.6 NIA N/A N/A N/A1960 N/A NIA N/A N/A -34.8 -44.6 N/A N/A N/A N/A1961 N/A NIA -.13 -2.03 -24.3 -43.6 N/A N/A N/A N/A1962 NIA NIA -.17 -1.67 -33.1 -40.0 N/A N/A NIA N/A1963 5.63 -.99 1.39 -.21 -55.5 -64.6 -48.5 -65.9 -4.4 -6.01964 -.36 -7.84 1.14 -.66 -27.6 -19.2 -26.9 -27.7 -2.2 -2.21965 10.31 -6.01 1.28 -2.17 -15.0 -92.9 -3.4 -101.0 -.2 -6.91966 13.75 3.81 .96 -1.45 -57.3 -67.7 -42.6 -65.3 -2.9 -4.31967 1.49 -1.29 .74 -.93 -125.8 -50.6 -123.6 -52.8 -8.2 -3.51969 -7.80 -19.04 -.81 -3.03 -157.1 -59.1 -165.7 -81.2 -9.7 -4.81969 .98 -17.48 .69 -2.63 -227.6 -172.1 -225.9 -192.2 -11.3 -9.61970 27.48 20.07 2.01 -.83 -163.7 -76.6 -134.2 -57.4 -5.9 -2.51971 10.17 5.57 2.84 .27 -166.5 -18.7 -153.5 -12.9 -6.1 -.51972 -65.46 -88.50 5.88 4.00 -133.3 16.4 -192.9 -68.1 -6.9 -2.41973 48.03 40.68 3.40 2.07 -283.5 -107.0 -232.1 -64.3 -6.6 -1.81974 5.79 .72 -2.05 -6.76 -401.0 -144.7 -397.2 -150.9 -8.5 -3.21975 16.64 2.29 12.20 4.20 -253.1 2.0 -224.2 8.5 -4.2 .21976 67.42 53.03 28.70 18.43 -670.5 -276.6 -574.4 -205.1 -8.8 -3.11977 112.24 63.72 72.28 23.92 -540.9 -97.4 -356.4 -9.9 -3.2 -.11978 52.77 -24.51 72.97 20.56 -2165.6 -1971.8 -2039.9 -1975.7 -9.7 -9.41979 25.45 -284.52 63.88 3.92 -1020.2 -1643.0 -930.9 -1923.6 -3.3 -6.81980 685.75 264.13 211.23 70.15 397.5 -2264.6 1294.5 -1930.3 3.0 -4.51981 1277.54 744.26 580.11 -381.92 2350.9 -1276.5 4208.5 -914.2 5.8 -1.31982 1136.83 909.48 271.94 -20.91 -2090.0 -3757.5 -681.2 -2869.9 -.1 -.31993 2944.91 2757.94 825.33 443.34 -3396.2 -10397.8 374.1 -7196.5 .2 -3.91984 3364.99 3145.63 1547.71 1089.30 -1920.7 -9567.2 2992.0 -4333.3 1.1 -1.61985 1168.51 N/A 950.61 N/A -2127.1 N/A -7.9 N/A -.002 N/A

Notes to Table 36:Coluens are derived as follows:(1) Transfers to maize from Annex Table 7-1(1).(2) Transfers to maize minus transfers from maize from Annex Table 7-1(1).(3) Transfers to rice from Annex Table 7-2(l).(4) Transfers to rice minus transfers from rice from Annex Table 7-2(l).(5) Minus Direct Transfers from cocoa from Table 7-3t1).(6) Minus (Direct Transfers from cocoa plus Indirect Transfers from cocoa) from Annex Table 7-3(l).(7) Sue of coluens (1), (3), and (5).(8) Sue of coluens (2), (4), and (6).(9) Column (7) divided by Nominal 6DP from Table 3.(10) Column (8) divided by Nominal GDP from Table 3.

209

Table 37

Direct and Total Noeinal Long-Run Transfers Due to Output Price InterventionsInto (+1 and Out of (-) Agriculture

(million NC)

Sue of Transfers asIeportables Exportables All Products a Share of GDP (Z)

maize Rice CocoaDirect Total Direct Total Direct Total Direct Total Direct Total

Year (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1958 -17.40 -29.64 -.01 -.49 -73.1 -55.2 -90.5 -85.4 -11.8 -11.11959 N/A N/A N/A NIA -60.5 -56.4 NIA N/A N/A N/A1960 N/A N/A -.23 -3.49 -36.3 -46.3 N/A N/A N/A N/A1961 N/A N/A -.14 - -2.46 -21.6 -43.3 N/A N/A N/A N/A1962 -14.29 -64.75 -.18 -1.78 -33.5 -40.2 -48.0 -106.7 -4.9 -10.81963 6.28 -2.09 1.38 -. 28 -57.1 -66.2 -49.4 -68.6 -4.5 -6.21964 -.36 -8.72 1.13 -. 72 -31.0 -21.1 -30.2 -30.6 -2.4 -2.51965 9.91 -5.74 1.23 -2.07 -16.8 -104.8 -5.6 -112.6 -.4 -7.71966 13.60 3.71 .92 -1.30 -69.6 -82.9 -55.1 -80.5 -3.6 -5.31967 1.43 -1.16 .69 -. 87 -165.9 -57.1 -163.7 -59.1 -10.9 -3.91968 -7.27 -17.33 -.76 -2.85 -206.6 -66.2 -214.6 -86.3 -12.6 -5.11969 .92 -16.42 .64 -2.42 -296.3 -219.9 -294.7 -238.8 -14.7 -11.91970 27.08 19.63 1.93 -.80 -219.6 -95.1 -190.6 -76.3 -8.4 -3.41971 10.06 5.58 2.73 .25 -238.6 -2.8 -225.8 3.0 -9.0 .11972 -61.16 -80.48 5.59 3.79 -182.9 51.2 -238.4 -25.5 -8.5 -.91973 44.07 37.39 3.22 1.97 -363.7 -119.7 -316.4 -80.3 -9.0 -2.31974 5.79 .52 -1.90 -5.98 -558.2 -159.9 -554.3 -165.3 -11.9 -3.51975 15.54 3.77 11.23 4.22 -352.5 57.5 -325.7 65.5 -6.2 1.21976 65.44 52.39 27.92 17.83 -946.7 -325.2 -853.3 -255.0 -13.1 -3.91977 110.23 68.04 68.34 25.83 -841.1 -50.7 -662.5 43.2 -5.9 .41978 51.69 -9.27 71.55 24.90 -3454.4 -3118.3 -3331.1 -3102.6 -15.9 -14.81979 24.54 -217.81 60.38 11.74 -1593.6 -2675.3 -1508.7 -2881.4 -5.3 -10.21980 648.60 305.81 191.35 76.58 571.3 -3516.0 1411.2 -3133.6 3.3 -7.31981 1251.19 807.10 475.90 -247.73 2975.6 -2455.9 4702.7 -1896.5 6.5 -2.61982 . 1135.16 961.76 266.67 .65 -2146.1 -4022.8 -744.3 -3060.3 -.1 -.41983 2936.40 2800.34 793.08 432.82 -4721.5 -15023.3 -992.0 -11790.1 -.5 -6.41984 3361.34 3211.44 1475.81 1087.86 -3221.1 -15731.9 1616.1 -11432.6 .6 -4.21985 1151.00 N/A 901.99 N/A -3236.1 N/A -1183.1 N/A -.3 N/A

--- --- --- --- -- --- --- --- --- -- --- --- --- --- -- --- --- --- _--- -_--- --- -- --- --- --- --- -- --- --- --- --

Notes to Table 37:Columns are derived as follows:(1) Transfers to maize from Annex Table 7-1(2).(2) Transfers to maize minus transfers from maize from Annex Table 7-1(2).(3) Transfers to rice from Annex Table 7-2(2).(4) Transfers to rice minus transfers from rice from Annex Table 7-2(2).(5) Minus Direct Transfers from cocoa from Table 7-3(2).(6) Minus (Direct Transfers from cocoa plus Indirect Transfers from cocoa) from Annex Table 7-3(2).(7) Sum of columns (1), (3), and (5).(8) Sum of columns (2), (4), and (6).(9) Column (7) divided by Nominal GDP from Table 3.(10) Column (8) divided by Nominal GDP from Table 3.

210

Table 38

Direct and Total Nominal Very Long-Run Transfers Due to Output Price InterventionsInto (t) and Out of 1-) Agriculture

(sillion NC)

Sum of Transfers aslmportables Exportables All Products a Share of GDP MU

"aize Rice CocoaDirect Total Direct Total Direct Total Direct Total Direct Total

Year (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1958 -17.40 -29.64 -.01 -.49 -73.1 -55.2 -90.5 -85.4 -11.9 -11.11959 N/A N/A N/A N/A -60.6 -56.4 N/A N/A N/A N/A1960 N/A N/A -.23 -3.49 -36.3 -46.3 N/A N/A N/A N/A1961 N/A N/A -.14 -2.46 -24.6 -43.3 N/A N/A N/A N/A1962 -14.29 -64.75 -.18 -1.78 -33.5 -40.2 -49.0 -106.7 -4.9 -10.91963 6.28 -2.09 1.38 -.28 -57.1 -66.2 -49.4 -68.6 -4.5 -6.21964 -.36 -8.72 1.13 -.72 -31.0 -21.1 -30.2 -30.6 -2.4 -2.51965 9.91 -5.74 1.23 -2.07 -16.8 -104.9 -5.6 -112.7 -.4 -7.71966 13.60 3.71 .92 -1.30 -69.6 -93.0 -55.1 -90.5 -3.6 -5.31967 1.43 -1.16 .69 -.87 -166.1 -57.1 -164.0 -59.1 -10.9 -3.91968 -7.27 -17.33 -.76 -2.85 -207.6 -66.1 -215.7 -86.3 -12.7 -5.11969 .92 -16.42 .64 -2.42 -299.4 -221.7 -297.9 -240.5 -14.9 -12.01970 27.09 19.63 1.93 -.80 -223.9 -95.3 -194.9 -76.4 -9.6 -3.41971 10.06 5.58 2.73 .25 -246.7 2.8 -234.0 8.7 -9.4 .31972 -61.16 -80.48 5.59 3.79 -192.3 60.9 -247.9 -15.7 -8.8 -.61973 44.07 37.39 3.22 1.97 -393.6 -121.1 -346.3 -81.7 -9.9 -2.31974 5.79 .52 -1.90 -5.98 -623.2 -161.7 -619.3 -167.1 -13.3 -3.61975 15.54 3.77 11.23 4.22 -409.0 87.4 -381.3 95.4 -7.2 1.01976 65.44 52.39 27.92 17.93 -1159.9 -372.3 -1066.6 -302.1 -16.3 -4.61977 110.23 69.04 68.34 25.83 -1087.5 16.7 -909.0 110.6 -8.1 1.01978 51.69 -9.27 71.55 24.90 -4719.1 -4208.9 -4595.9 -4193.3 -21.9 -20.01979 24.54 -217.81 60.38 11.74 -2249.9 -4022.2 -2165.0 -4228.3 -7.7 -15.01980 648.60 305.81 191.35 76.58 841.0 -6345.2 1680.9 -5962.8 3.9 -13.91981 1251.19 807.10 475.90 -247.73 4127.3 -6178.3 5854.4 -5618.9 8.1 -7.71982 1135.16 961.76 266.67 .65 -3371.4 -7054.1 -1969.5 -6091.7 -.2 -.71983 2936.40 2800.34 793.09 432.92 -8184.0 -29137.7 -4454.5 -25904.6 -2.4 -14.11984 3361.34 3211.44 1475.81 1097.86 -5673.7 -32279.9 -836.5 -27980.5 -.31 -10.31985 1151.00 N/A 901.99 N/A -5743.3 N/A -3690.3 N/A -1.0 NIA

Notes to Table 38:Columns are derived as followut(1) Transfers to maize from Annex Table 7-1(2).(2) Transfers to saize minus transfers from maize from Annex Table 7-1(2).(3) Transfers to rice from Annex Table 7-2(2).(4) Transfers to rice minus transfers from rice from Annex Table 7-2(2).(5) Minus Direct Transfers from cocoa from Table 7-313).(6) Minus (Direct Transfers from cocoa plus Indirect Transfers from cocoa) from Annex Table 7-3(3).(7) Sum of columns (1), (3), and (5).(8) Sum of columns (2), (4), and (6).(9) Column (7) divided by Nominal GDP from Table 3.(10) Column (9) divided by Nosinal 6DP from Table 3.

2 11

table 39

Short-Run Transfers Into (+1/ Out of (-1 Agriculturt(Sillion NC)

Total of Price-RelatedTotal of Price-Related and Non-Price Transfers

Non-price Transfers Price-Related Transfers and Non-Price Transfers as a Share of GOP (11._- - - -- - - -__ -- - - -- - - -- - - - - -- -__ -- - - -- - - - -- - - -- - - -------------------_--- -__-_ __-__ _ ____ -_- _

Public Roads and Indirect InputInvestwnt Waterways Agriculture Subsidies Subtotal Direct Total Direct Total Direct Total

tear (1I (21 (31 (49 (51 (60 (7) (a) (9t (10) (fil

1955 N/A N/A N/A NIA N/A N/A N/A N/A N/A N/A NIA195 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A1957 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A1958 6.4 9.4 14.0 .59 29.4 -89.9 -92.8 -60.4 -53.4 -7.9 -7.0195 9.9 8.4 13.5 1.94 33.6 N/A N/A N/A N/A N/A N/A1960 10.7 9.2 14.0 3.73 36.6 N/1A N/A NIA N/A N/A N/A1961 8.9 9.1 23.3 3.32 44.6 N/A NVA N/A N/A N/A N/A1962 7.5 9.0 17.4 1.94 34.9 N/A N/A N/A N/A N/A N/A1963 13.2 12.0 14.8 5.79 45.8 -49.5 -65.8 -2.7 -20.0 -.2 -1.31964 20.1 9.2 15.3 3.26 46.9 -26.9 -27.7 20.0 19.2 1.6 1.61965 24.9 6.8 20.4 13.14 65.2 -3.4 -101.0 61.9 -35.8 4.2 -2.41966 10.4 4.3 19.8 4.19 39.2 -42.6 -65.3 -3.4 -26.1 -.2 -1.71967 11.8 2.4 19.3 1.45 34.9 -123.6 -52.9 -99.6 -17.9 -5.9 -1.21969 17.9 3.8 19.7 1.05 42.5 -165.7 -91.2 -123.2 -38.7 -7.2 -2.31969 22.2 5.5 23.4 3.20 54.3 -225.9 -192.2 -171.6 -137.9 -9.6 -6.91970 30.6 5.3 25.1 3.47 64.5 -134.2 -57.4 -69.7 7.1 -3.1 .3

H 1971 32.4 4.9 25.3 3.79 66.3 -153.5 -12.9 -97.2 53.4 -3.5 2.11972 27.7 6.2 31.5 2.02 67.4 -192.9 -69.1 -125.5 -. 7 -4.5 .01973 25.7 15.9 29.1 2.15 72.9 -232.1 -64.3 -159.3 8.6 -4.5 .21974 59.2 28.7 53.9 4.62 146.3 -397.2 -150.8 -250.9 - 4.4 -5.4 -.11975 111.4 36.8 70.1 10.07 228.4 -224.2 8S.5 4.1 236.8 .1 4.51976 196.6 19.5 125.7 16.91 359.6 -574.4 -205.1 -215.9 153.5 -3.3 2.41977 233.3 42.0 343.6 84.67 703.6 -356.4 -9.9 347.2 693.9 3.1 6.21978 256.0 78.5 342.9 101.30 779.7 -2039.9 -1975.7 -1261.2 -1197.0 -6.0 -5.71979 190.5 91.4 1497.0 1130.69 807.6 -930.9 -1923.6. -43.3 -1036.0 -. 2 -3.71990 464.3 135.8 721.4 254.92 1576.4 1294.5 -1930.3 2870.9 -353.9 6.7 -.81931 375.5 131.6 903.1 926.11 2236.3 4209.5 -914.2 6444.9 1322.1 9.9 1.81992 396.2 123.4 927.6 628.48 1975.7 -691.2 -2969.9 1294.5 -993.2 1.5 -1.01993 86.9 177.8 1399.6 N/A N/A 374.1 -7196.5 N/A N/A N/A N/A1994 1851.7 423.3 936.3 N/A N/A 2 .0 -4333.3 N/A N/A N/A N/A1995 N/A N/A N/A N/A N/A -7.9 N/A N/A N/A N/A N/A

IYh3 --. 12. ----- 5.7 -4-------------------- 8 -- 4B.5- -- 65.8--------------- -2. -2. -. --- 1.8---

Notes to Table 39:Colums are derive as follows:(1) AneNx table 7-5, column 6.(29 From Annex Table 7-5.(3) From Annex Table 7-5.(49 Indirect Input Subsidie-s from Annex Table 7-4.(5) Sum of column (19 through (41.(69 Fro Table 36.979 From Table 36.18) Suo of column (51 and (6).(9) Sum of, columns a5 nd (19.(101 Column (9) divided by 4ominal SOP from Table 3.(119 Column (99 divided by Nominal 6DP from Table- 3.

Table 40

Long-Ran Transfers Into (41/ Out o1 - Aqricltureloillion ICI

Total of Price-RelatedTotal of Price-Related and Non-Price Transfers

Non-price Transfers Price-Related Transfers and Non-Price Transfers as a Share of DP (121.

Public Roads and Indirect InputInvestnent Waterways Agriculture Subsidies Subtotal Direct Total Direct Total Direct Total

Year (1 (21 131 (44 (5) (61 (7) (I8 (9) (101 (11)

1955 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/AIV56 N/A N/A N/A N/A N/A N/A N/A N/A N/A NIA NIA1957 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A1959 6.4 8.4 14.0 .59 29.4 -90.5 -85.4 -61.1 -56.0 -9.0 -7.31959 9.9 8.4 13.5 1.94 33.6 N/A N/A N/A N/A N/A N/A1960 10.7 B.2 14.0 3.73 36.6 N/A N/A N/A N/A N/A N/A1961 8.9 9.1 23.3 3.32 44.6 NIA N/l N/A N/A NiA N/A1962 7.5 8.0 17.4 1.94 34.8 -49.0 -106.7 -13.1 -71.9 -1.3 -7.31963 13.2 12.0 14.8 5.79 45.8 -49.4 -6S.6 -3.6 -22.8 -.3 -2.11964 20.1 0.2 15.3 3.26 46.9 -30.2 -30.6 16.7 16.3 1.3 1.31965 24.9 6.9 20.4 13.14 65.2 -5.6 -112.6 59.6 -47.4 4.1 -3.216 10.4 4.8 19.8 4.19 39.2 -55.1 -90.5 -15.9 -41.3 -1.0 -2.71967 11.8 2.4 19.3 1.45 34.9 -163.7 -59.1 -128.9 -24.2 -9.6 -1.69I68 17.9 3.9 19.7 1.05 42.5 -214.6 -86.3 -172.2 -43.9 -10.1 -2.6S1969 22.2 5.5 23.4 3.20 54.3 -294.7 -238.8 -240.4 -184.5 -12.0 -9.2

1970 30.6 5.3 25.1 3.47 64.5 -190.6 -76.3 -126.1 -11.9 -5.6 -.51971 32.4 4.9 25.3 3.79 66.3 -225.8 3.0 -159.5 69.3 -6.4 2.91972 27.7 6.2 31.5 2.02 67.4 -238.4 -25.5 -171.0 42.0 -6.1 1.51973 25.7 15.9 29.1 2.15 72.9 -316.4 -80.3 -243.5 -7.5 -7.0 -.21974 59.2 28.7 53.0 4.62 146.3 -554.3 -165.3 -401.0 -19.0 -8.8 -.41975 111.4 36.8 70.1 10.07 229.4 -325.7 65.5 -97.4 293.9 -1.8 5.61976 196.6 19.5 125.7 16.81 358.6 -853.3 -255.0 -494.7 103.7 -7.6 1.61977 233.3 42.0 343.6 84.67 703.6 -662.5 43.2 41.0 746.7 .4 6.71979 256.0 79.5 342.9 101.30 778.7 -3331.1 -3102.6 -2552.4 -2323.9 -12.2 -11.11979 188.5 91.4 497.0 130.69 887.6 -1508.7 -2981.4 -621.1 -1993.8 -2.2 -7.11980 44.3 135.8 721.4 254.92 1576.4 1411.2 -3133.6 2987.7 -1557.2 7.0 -3.61981 375.5 131.6 803.1 926.11 2236.3 4702.7 -1896.5 6939.0 339.8 9.6 .51992 396.2 123.4 827.6 628.48 1975.7 -744.3 -3060.3 1231.4 -1084.7 1.4 -1.31993 986.9 177.8 1389.6 N/A N/A -992.0 -11790.1 N/A N/A N/A N/A1984 1851.7 423.3 936.3 N/A N/A 1616.1 -11432.6 N/A N/A N/A N/A1985 N/A N/A N/A N/A N/A -1193.1 N/A N/A N/A N/A N/A

Notes to Table 40:Columns are derived as follows:(1) Annet Table 7-5, coluan 6.(2) Fro Annex Table 7-5.(31 From Annex Table 7-5.(41 Indirect Input Subsidies fra Annex Table 7-4.(51 Sum of columns 11( through (41.(6) Fro Table 37.(7) Froe Table 37.(8) Sue of columns (5) and (61.(9) Son of colvmns (5) and (7).(101 Coluan (9) divided by Noinal WP froe Table 3.(111 Column (9t divided by Nominal SP from Table 3.

Table 41- -- -- - -- - - - - - -- -- -- -- -- -- -- --- - -- -- - -- - -- -- -- -- -

Very Lng-Run Transfers Into 14)/ ht of (-3 Agricultureluillion ICI

Total of Price-RelatedTotal of Price-Related and Nln-Price transfers

Non-price Transfers Price-Related Transfers and Non-Price Transfers as a Share of SP (ZI.

Public Roads and Indirect InputInvestwnt WItervays Agriculture Subsidies Subtotal Direct Total Direct Total Direct Total

Year (13 (23 (3) (43 (51 (61 (7) (89 (93 (103 (11I

155 N/A N/A N/A N/A N/A N/A N/A N/A N/A NIA N/A1956 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A1957 N/A N/A N/A W/A N/A N/A N/A N/A N/A N/A N/A1951 6.4 9.4 14.0 .59 29.4 -90.5 -85.4 -61.1 -56.0 -8.0 -7.31959 9.9 0.4 13.5 1.94 33.6 NI/A N/A N/A N/A N/A N/A1960 10.7 8.2 14.0 3.73 36.6 N/A N/A N/A N/A N/A N/A191 9.9 9.1 23.3 3.32 44.6 N/A N/A N/A N/A N/A N/A1962 7.5 9.0 17.4 1.94 34.9 -48.0 -106.7 -13.1 -71.9 -1.3 -7.31963 13.2 12.0 14.9 5.79 45.8 -49.4 -69.6 -3.6 -22.8 -. 3 -2.1I164 20.1 8.2 15.3 3.26 46.9 -30.2 -30.6 16.7 16.3 1.3 1.31965 24.9 6.8 20.4 13.14 65.2 -5.6 -112.7 59.6 -47.4 4.1 -3.21966 10.4 4.8 19.9 4.19 39.2 -55.1 -80.5 -15.9 -41.4 -1.0 -2.7197 11.9 2.4 19.3 1.45 34.9 -164.0 -59.1 -129.1 -24.2 -8.6 -1.6

196I 17.9 3.9 19.7 1.05 42.5 -215.7 -96.3 -173.2 -43.9 -10.2 -2.61969 22.2 5.5 23.4 3.20 54.3 -297.8 -240.5 -243.5 -106.2 -12.2 -9.31970 30.6 5.3 25.1 3.47 64.5 -194.9 -76.4 -130.3 -12.0 -5.8 -.51971 32.4 4.9 25.3 3.79 66.3 -234.0 9.7 -167.7 75.0 -6.7 3.01972 27.7 6.2 31.5 2.02 67.4 -247.9 -15.7 -100.4 51.7 -6.4 1.81973 25.7 15.9 29.1 2.15 72.9 -346.3 -91.7 -273.5 -8.9 -7.8 -. 31974 59.2 29.7 53.9 4.62 146.3 -619.3 -167.1 -473.0 -20.8 -10.2 -.41975 111.4 36.9 70.1 10.07 229.4 -381.3 95.4 -152.9 323.7 -2.9 6.11976 196.6 19.5 125.7 16.81 358.6 -1066.6 -302.1 -709.0 56.5 -10.3 .91977 233.3 42.0 343.6 84.67 703.6 -909.0 110.6 -205.4 914.1 -1.9 7.31979 256.0 79.5 342.9 101.30 770.7 -4595.9 -4193.3 -3817.2 -3414.6 -10.2 -16.31979 199.5 81.4 487.0 130.69 897.6 -2165.0 -4228.3 -1277.4 -3340.7 -4.5 -11.i

IS90 464.3 135.8 721.4 254.92 1576.4 I19O.9 -5962.0 3257.3 -4336.4 7.6 -10.2199 375.5 131.6 903.1 926.11 2236.3 5854.4 -5619.9 09S0.7 -3362.6 11.1 -4.71992 396.2 123.4 827.6 628.48 1975.7 -1969.5 -6091.7 6.1 -4116.0 .0 -4.8193 896.0 177.0 139.6 NI/A N/A -4454.5 -25904.6 N/A N/A N/A N/A

1994 1951.7 423.3 936.3 N/A N/A -936.5 -27990.5 N/A N/A N/A N/A(995 N/A N/A NIA N/A N/A -3690.3 N/A N/A N/A N/A N/A

Notes to Table 41:Coluns are derived as follons:(11 Annen Table 7-5, coluen 6.(2) From Annex Table 7-5.(31 Fre Annex Table 7-5.141 lireKt Input Subsidie from Annex Table 7-4.(51 So of coluns (13 through (511(6 Free Table 38.(71 From Table 36.(83 So of coluns (53 and (63.(9) So of colans 151 and (71.(103 ColI (8n divided by Nominal 6DP from Table 3.(111 Colmn (9I divided by Nominal GDP fro Table 3.

years in which the optimumn domestic price for coCOa at the borTder measured

at the equilibrium exchange rate was substantially less than the FEB price

measured at the official rate.. On the whole, however, net price-related

transfers have been away from agriculture, and the magnitude of these

transfers became very substantial after 1977, especially when a very-long-

run autput response is assumed.

As a percentage of GDP, the magnitude of net transfers out of

agriculture due to output price interventions has varied markedly from year

to year. In 1958 the transfer was quite large since domestic prices were

well below world market prices for both maize and cocoa. Thereafter the

share of total price-related short-run transfers decreased from 11 percent

in 1956 to an average of 5 percent during the final four years of the

Nkrumah regime. From 1967 to 1977, total price-related short-rumn transfers

out of agriculture averaged less than 3 percent of GDP. They increased to

9.4 percent in 197E and then dropped steadily to 0.3 percent in 1962 as

government taxation of the cocoa sector eroded in the face of a highly

overvalued exchange rate and accelerating inflation.

Moving to the Icng and the very Icng run, the same trends are

evident but in much more exaggerated form because the higher response of

output to price distortions increases the magnitude of the transfers

involved. In the very long run, for example, agricultural losses associated

with price interventions frequently exceeded 10 percent of GDP, especially

after 1977. This is especially important for cocoa because of the long

gestation periods involved. It also implies that, even under the reforms

5 See Phnex Table 7-3(1).

215

of the Rawlings regime, it will be years before the effects of past policies

are fully overcome.

Price-related transfers are combined with nm-price transfers in

Tables 39-41. The latter include public investment in agriculture and

recurrent government expenditures an roads, waterways, and agriculture,

inclusive of direct input subsidies, as well as the indirect subsidy on

inputs that results from the overvalued exchange rate. In most years net

transfers were aut of agriculture and tended to increase as the supply

response period lengthened. Exceptions to this occurred in 1964 and 1965,

when public expenditures were relatively large and incawe from cocoa was

reduced because of low world prices. Again during the period 1970-79,

total price related transfers were sometimes positive for reasons explained

earlier, and even when they were negative, their magnitude was fairly low in

relation to government expenditures. In 1990 total transfers were positive

in the short and long run, largely because the cocoa price to producers was

subsidized in that year. Thereafter, though data on indirect subsidies are

incomplete, total transfers were negative and very substantial.

The fact that government expenditures and the indirect input

subsidy have at times offset price related transfers does not necessarily

imply that agriculture was better off. In many cases, investments in

agriculture were relatively unproductive, and recurrent expenditures were

related more to public sector employment than to benefitting the farmer.

This was especially true during the twD periods mentioned above, which were

each characterized by the government s strong orientation towards direct

intervention in agriculture rather than the use of price policy to induce

increased production.

216

Government Investment and Expenditure Bias

In order to estimate the degree of bias that has existed towards

agriculture with respect to government investment and total expenditures, it

is useful to campare the share of agriculture in each of these with its

share in GDP. This GOP share may be estimated for either the distorted or

the undistorted situatimns. Here we are interested in the share that would

have prevailed if prices had been undistorted by trade policy and exchange

controls.

The value of agricultural GEP in the undistorted situation is

estimated for the very long run in Table 7-6. Aijustments are made in this

table for changes in prices and quantities produced of maize, rice, and

cocoa. The price changes include both the direct effects of trade policy

and the indirect effects of exchange rate adjustments. (iiantity changes are

those estimated earlier using equaticns (2)-(4).

Government Investment Bias (GIB), or the share of agriculture in

government investment expenditures divided by its share in GDP, is shown in

Table 42. It is clear from this table that public investment has bee

strogly biased away from agriculture. In many years the ratio was less

than 0.4, especially during the Nkrumah regime. During the period from 1968

to 1972, it averaged in excess of 0.55, but then slipped scoewhat until

1981-84, when the average was 0.75. The Government Expenditure Bias (GEB)

has been even more unfavorable for agriculture, with the ratio in most years

less than 0.3. Since 1977 it has been only slightly in excess of 0.2.

217

Table 42

Government Investment (6IB) and Total Expenditure (GEB) Bias

Ni6DP /GDP (a) 61 /61 lb) BE /BE Id)A A 61B (c) A 6EB (e)

Year (1) (21 (3) (4) (5)

1958 .62.3 .194 .311 .232 .3731959 N/A .178 N/A .203 N/A1960 N/A .130 N/A .146 N/A1961 N/A .098 N/A .177 N/A1962 .557 .074 .133 .133 .2381963 .440 .126 .286 .150 .3411964 .384 .157 .409 .132 .3441965 .495 .175 .354 .144 .2901966 .510 .160 .314 .130 .2551967 .477 .171 .359 .107 .2241968 .490 .267 .546 .113 .2301969 .610 .264 .433 .116 .1901970 .539 .283 .526 .130 .2411971 .471 .312 .661 .124 .2631972 .485 .289 .595 .122 .2521973 .550 .152 .276 .092 .1681974 .584 .207 .355 .125 .2141975 .503 .253 .502 .146 .2901976 .613 .309 .503 .171 .2791977 .593 .336 .566 .202 .3411978 .830 .337 .406 .181 .2181979 .786 .317 .403 .159 .2021980 .723 .334 .462 .166 .2301981 .642 .405 .631 .137 .2141982 .692 .485 .701 .152 .2201983 .740 .655 .885 .166 .2251984 .604 .464 .768 .120 .1991985 N/A N/A N/A N/A N/A

Notes to Table 42:(a) Non-intervention Agricultural GDP from Annex 7, Table 7-6 divided by

nominal GDP from Table 3.(bl Total Investment Boverneent Expenditure Transfers to Agriculture from

Annex 7, Table 7-5 divided by overall Investment Expenditures fromAnnex 7, Table 7-7.

(ci Column (2) divided by column (1).(d) Total Recurrent plus Total Investment Government Expenditure Transfer

to Agriculture from Annex 7, Table 7-5, divided by Total GovernmentExpenditures from Annex 7, Table 7-7.

(e1 Column (4) divided by column (1).

218

CFPPTER IX: OTH0ER PRICE INTERMENTION EFFECTS

There are a number of other effects of agricultural price

interventica s. Two of these are examined in this chapter. The first is the

impact on the inclme of different types of farmers. It w.uld also be

desirable to measure the effects of price policy mn other interest groups,

biut the data do not permit this, and these effects would be swamped by the

muich larger changes in real income that all groups have experienced and that

are only partially related to agricultural prices.1 The second effect of

price interventions analyzed here is related to the variability of prices

due to changes in domestic demand and supply and to price fluctuations on

the world market.

Farm Income Effects

The effects of price policy mn farm income were estimated for

three different types of farms. The first consists of large-scale cocoa

farmers, who in 1970 harvested approximately 200 headloads of cocoa

weighting 60 posunds each.2 This quantity was adjusted for other years by

multiplying it times the ratio of total natimnal production in that year to

total producrtion in 1970. The land-owner generally retains twoi-thirds of

the revenue received. He is usually absntee and received income from other

A According to a 1974-75 hausehDld budget survey, for example, riceand wheat accounted for mnly 3.2 percent of total household expenditures inurban areas. Ghana Central Bureau of Statistics, SLumnary Reiport inHbusehold Economics Survey. 1974-1975, Accra: July 1979.

2 C. Okali, M. Owusuansha, and B.E. Rourke, "The Development Patternof Large Cocoa HDidings in Ghana: Scme Case Studies," in R.A. Kotey, C.Okali, and B.E. Rourke, Eds., Economics of Cocoa Production and Marketina,Institute of Statistical, Social and Economic Research, Uhiversity of Ghana,1974.

219

sources, but there are no data on the amnuLnts involved. Nbminal income is

therefore estimated by multiplying twe-thirds of total cocoa production per

farm by the relevant producer price. This is converted to real income by

dividing by a weighted average of the overall c1nsumer price index and the

prices of rice and maize. The weights pertain to urban consumption patterns

obtained from the results of a survey canducted during the mid 1970s.3

A similar analysis was unidertaken for large-scale rice farmers

cultivating an average of 119.3 acres, each acre yielding 7.1 bags of 160

pounds per.' The base year in this case is 1973. Production was adjusted

for the costs of labor and other inputs used in cultivation. The same

expenditure coefficients were used as for cocoa farmers, and no adjustment

was possible for off-farm income because of lack of data.

The third type of farmer grwas maize and is assumed to cultivate

10 acres, with yields of 1300 poLinds per acre.5 Data are not available by

farm type on the mix of crops groAwn so no adjustment is possible for other

sources of income. Expenditure weights pertaining to rural csansumption

weights were obtained from the same scurce as for the other farmers.

The results of this analysis are contained in Tables 43-50 for

direct and total interventions in the instantaneous, short, long, and very

long runs. The instantanecous run assumes no ou-tput response, whereas the

other runs are calculated under the set of assuffptions used in Chapter VII.

Ghana, Central Bireau of Statistics, Summarv Report..., 1979.

Fred Everett Winch III, Costs and Returns of Alternative RiceProduction Systems in Nbrthern Ghana: Implications for Oitput. Employmnentand Income Distribution, Ph.D., Dissertation, Michigan State University,1976.

Ghana, Ministry of Agriculture, Economics and Marketing Division,Report on Ghana Sample Census of Agriculture 1970, Accra: March 1972.

220

Table 43

Real Instantaneous Income Effect of Direct Intervention (a)(NC)

Large Scale Large Scale Small ScaleYear Cocoa Farmers Rice Farners Maize Farmers

1953 N/A N/A N/A1954 -.519 N/A N/A1955 -.242 -.055 -.6691956 -.203 .076 -.5441957 -.541 .008 -.7811958 -.529 .001 -. 5591959 -. 440 N/A N/A1960 -. 262 -. 087 N/A1961 -. 203 -.088 N/A1962 -. 258 -. 139 -.4441963 -. 362 1.007 1.1131964 -.212 1.041 -.0471965 -. 155 1.252 1.9131966 -. 364 1.062 2.4691967 -. 508 .388 .2971968 -.598 -.238 -.3541969 -.622 .139 .0281970 -. 513 .643 3.9181971 -.515 1.110 .7311972 -.460 1.485 -.6281973 -.597 .513 2.3721974 -.642 -.129 .1601975 -.486 1.371 .5411976 -.683 3.583 3.4231977 -.546 -12.223 3.0541978 -. 704 -6.217 1.1421979 -.433 -3.923 .1091980 .433 -7.063 9.2071981 -59.837 -2.840 -18.8331982 -.401 -4.186 -7.2761983 -.417 -4.594 -14.0071994 -.251 -17.592 2.5791985 -. 144 -38.663 .871

Notes to Table 43:(a) Real Actual Income minus Real Equilibrium Instantaneous Income

divided by Real Instantaneous Equilibrium Income (from AnnexTable 8-1(1) for Cocoa, Table 8-2(1) for Rice, and Table 8-3(1)for Maize).

221

Table 44

Real Short-Run Income Effect of Direct Intervention (a)(NC)

Large Scale Large Scale Seall ScaleYear Cocoa Farners Rice Farmers Maize Farsers

1953 N/- N/A NIA1954 N/A N/A N/A1955 -. 242 N/A N/A1956 -.215 .076 -.5441957 -.568 .361 -.6661958 -.583 .247 -.6351959 -.495 N/A N/A1960 -.252 N/A N/A1961 -.242 -.084 N/A1962 -.320 -.076 N/A1963 -.531 1.216 2.4861964 -.305 1.916 3.2501965 -.387 2.430 7.1671966 -.580 2.757 24.8681967 -.625 1.451 9.7551968 -.678 .252 1.6841969 -.711 .413 1.5491970 -.681 1.186 15.5781971 -.632 2.329 19.7021972 -.479 3.207 .9311973 -.735 2.057 4.8341974 -.733 .507 11.8901975 -.599 1.775 6.0011976 -.807 8.942 31.5981977 -.692 -29.220 76.1611978 -.800 -60.029 59.069197? -.505 -23.165 19.058L980 .154 -25.219 99.7221981 -103.635 -3.135 -398.9061982 -.677 -22.200 -2838.2081983 -.721 -32.908 -5350.3301984 -.390 -151.762 4077.9511995 -.247 -323.781 165.772

Notes to Table 44:(a) Real Actual Income minus Real Equilibrium Short-Run Income

divided by Real Short-Run Equilibrium Income (from AnnexTable 8-1(1) for Cocoa, Table 8-2(1) for Rice, and Table 8-3(1)for Maize).

222

Table 45

Real Long-Run Income Effect of Direct Intervention (a)(NC)

Large Scale Large Scale Small ScaleYear Cocoa Farmers Rice Farmers Maize Farsers

1953 N/A N/A N/A1954 N/A N/A N/A1955 -.242 N/A N/A1956 -.198 .076 -.5441957 -.557 .372 -.7001958 -.583 .354 -.6571959 -.522 N/A N/A1960 -.310 -.561 N/A1961 -.258 -.227 N/A1962 -.336 -.105 -.6801963 -.554 1.255 1.8921964 -.436 2.023 2.9861965 -.495 2.802 7.9781966 -.692 3.299 26.2291967 -.761 1.951 11.9281968 -.7?6 .514 2.3381969 -.812 .682 1.8871970 -.797 1.421 16.2211971 -.792 2.686 21.1881972 -.701 3.853 1.4881973 -.820 2.623 5.8561974 -.842 .877 11.7891975 -.765 2.368 7.8311976 -.885 9.910 36.4811977 -.842 -36.088 92.7681978 -.901 -79.400 83.1831979 -.761 -44.046 29.3091980 -.359 -47.805 143.5431981 -59.800 -4.529 -521.1161982 -.690 -26.015 -4110.3031983 -.824 -53.956 -12052.8241984 -.729 -292.386 9653.1111995 -.623 -651.935 458.383

Notes to Table 45:(a) Real Actual Income minus Real Equilibrium Long-Run Income

divided by Real Long-Run Equilibrium Income (from AnnexTable 8-111) for Cocoa, Table 8-2(1) for Rice, and Table 8-3(1)for Maize).

223

Table 46

Real Very Long-Run Income Effect of Direct Intervention (a)(NC)

Large Scale Large Scale Small ScaleYear Cocoa Farsers Rice Farsers Maize Fareers

1953 NIA NIA N/A1954 N/A N/A NlA1955 -.242 N/A N/A1956 -. ?8 ,076 -.5441957 -.557 .372 -.7001958 -. 583 .354 -. 6571959 -.523 N/A N/A1960 -.310 -.561 NIA1961 -.258 -.227 NIA1962 -.337 -.105 -.6801963 -.554 1.255 1.8921964 -.436 2.023 2.9861965 -.495 2.802 7.9781966 -.692 3.299 26.2291967 -.762 1.951 11.9281968 -.797 .514 2.3381969 -.815 .682 1.8871970 -.803 1.421 16.2211971 -.802 2.686 21.1881972 -.724 3.853 1.4881973 -.840 2.623 5.8561974 -.865 .877 11.7891975 -.810 2.368 7.8311976 -.912 9.910 36.4811977 -.887 -36.088 92.7681978 -.933 -79.400 83.1831979 -.850 -44.046 29.3091980 -.621 -47.805 143.5431981 -33.875 -4.529 -521.1161982 -.835 -26.015 -4110.3051983 -.910 -53.956 -12052.8241994 -.868 -292.386 9653.1111995 -.823 -651.935 458.383

Notes to Table 46:(a) For Cocoa: Real Actual Income minus Real Equilibrium Very

Long-Run Income divided by Real Very Long-Run EquilibriumIncome tfrom Annex Table 9-111)).For Rice and Maize: Very Long-Run in considered to be theas Long-Run so values are from Table 45.

224

Table 47

Real Instantaneous Income Effect of Total Intervention (a)(NC)

Large Scale Large Scale Snall ScaleYear Cocoa Farmers Rice Farmers Maize Farmers

1953 NIA N/A N/A1954 N/A N/A N/A1955 N/A -.062 -.6711956 N/A .068 -.5471957 N/A .000 -.7831958 -.454 -.244 -.6501959 -.420 N/A N/A1960 -.312 -.521 N/A1961 -.312 -.597 N/A1962 -.292 -.584 -.6751963 -.391 -.027 .1021964 -.153 -.242 -.5121965 -.491 -.533 -.2261966 -.393 -.429 .1421967 -.302 -.237 -.2491968 -.357 -.550 -.5811969 -.558 -.314 -.3111970 -.351 -.140 1.3511971 -.117 .057 .0171972 .274 .710 -.7041973 -.380 .196 1.6271974 -.409 -.363 -.1361975 .006 .314 -.0301976 -.503 .716 .9641977 -.178 .226 -.1991978 -.682 -.155 -.6421979 -.549 -.176 -.7001990 -.629 .235 -.1281981 -.369 -.577 -.4181982 -.530 -.002 -.2041983 -.660 .713 .7821984 -.579 1.169 -.2911985 N/A N/A NIA

Notes to Table 47:(a) Real Actual Incose minus Real Equilibrium Instantaneous

Incose divided by Real Instantaneous Equilibrium Income(from Annex Table 8-1(2) for Cocoa, Table 8-2(2) for Rice,and Table 8-3(2) for Maize).

225

Table 48

Real Short-Run Income Effect of Total Intervention (a)(NC)

Large Scale Large Scale Small ScaleYear Cocoa Farsers Rice Farmers Naize Farmers

1953 N/A N/A N/A1954 N/A N/A N/A1955 N/A N/A N/A1956 N/A .068 -.5471957 N/A .392 -.6611958 -.527 .049 -.6881959 -.490 N/A N/A1960 -.323 N/A N/A1961 -.365 -.616 N/A1962 -.365 -.623 N/A1963 -.554 -.054 .3361964 -.232 -.08B .4191965 -.636 -.327 .6291966 -.574 -.365 2.3191967 -.428 -.203 1.6061968 -.438 -.438 .1121969 -.659 -.247 .2961970 -.536 -.029 4.9671971 -.238 .180 5.4411972 .424 .815 -.1071973 -.517 1.053 2.7791974 -.525 -.032 6.7231975 -.099 .222 2.2661976 -.654 1.865 8.9431977 -.259 .897 6.6201978 -.740 .642 2.0101979 -.587 .202 .5401980 -.716 .867 3.8661961 -.420 -.368 7.0481982 -.615 .049 10.9791983 -.766 1.941 39.9961984 -.653 3.365 41.3161985 N/A N/A N/A

Notes to Table 48:(a) Real Actual Income minus Real Equilibrium Short-Run

Income divided by Real Short-Run Equilibrium Incose(from Annex Table 8-1(2) for Cocoa, Table 8-2(2) for Rice,and Table 8-3(2) for Maize).

226

Table 49

Real Long-Run Incose Effect of Total Interyention (a)

(NC)

Large Scale Large Scale S uall ScaleYear Cocoa Farmers Rice Farmers Maize Farmers

1953 N/A N/A N/A1954 N/A N/A N/A1955 N/A N/A N/A1956 N/A .068 -.5471957 N/A .348 -.7011958 -.468 .092 -.7121959 -.458 N/A N/A1960 -.300 -.777 N/A1961 -.306 -.691 N/A1962 -.310 -.657 -.8801963 -.526 -.126 .0241964 -.294 -.140 .2401965 -.651 -.328 .6421966 -.639 -.329 2.2611967 -.561 -.197 1.6701968 -.555 -.449 .1761969 -.715 -.234 .3251970 -.631 -.034 4.7801971 -.446 .175 5.4611972 .064 .803 .0251973 -.618 1.014 2.9801974 -.658 .064 6.2331975 -.391 .339 2.8991976 -.763 1.787 9.6971977 -.521 1.066 7.5591978 -.810 .B05 2.7201979 -.695 .399 .8961980 -.774 1.056 4.5821981 -.572 -.308 8.2491982 -.695 .144 14.6611983 -.819 1.944 54.2571984 -.765 3.819 59.8381985 N/A N/A N/A

Notes to Table 49:(a) Real Actual Income minus Real Equilibrius Long-Run

Income divided by Real Long-Run Equilibrium Incone(from Annex Table 8-1(2) for Cocoa, Table 8-2(2) for Rice,and Table 8-3(2) for Maize).

227

Table 50

Real Very Long-Run Incoae Effect of Total Intervention (a)(NC)

Large Scale Large Scale Seall ScaleYear Cocoa Farmers Rice Farmers Naize Farmers1953 N/_ N/A N/A

1953 N/A N/A N/A1954 N/A N/A N/A1955 N/A MIA N/R1956 N/A .068 -. 5471957 N/A .348 -.7011958 -.468 .092 -.7121959 -.459 N/A N/A1960 -.301 -.777 N/A1961 -.307 -. 691 N/A1962 -.310 -.657 -.8801963 -.526 -.126 .0241964 -.295 -.140 .2401965 -.651 -.328 .6421966 -.639 -.329 2.2611967 -.562 -.197 1.6701968 -.558 -.449 .1761969 -.720 -.234 .3251970 -.645 -.034 4.7801971 -.484 .175 5.4611972 -.018 .803 .0251973 -.656 1.014 2.9801974 -.701 .064 6.2331975 -.481 .339 2.8991976 -.804 1.787 9.6971977 -.657 1.066 7.5591978 -.879 .805 2.7201979 -.823 .399 .8961980 -.879 1.056 4.5821981 -.784 -.308 8.2491982 -.854 .144 14.6611983 -.917 1.944 54.2571984 -.896 3.819 59.8381965 N/A N/A N/A

Notes to Table 50:(a) For Cocoa: Real Actual Income minus Real Equilibrium Very

Long-Run Incose divided by Real Very Long-Run EquilibriumIncome (from Annex Table 8-1(2)).For Rice and Maize: Very Lang-Run in considered to be theas Long-Run so values are from Table 49.

228

The tables present deviations of real actual from real equilibrium income

as a perc:entage of real equilibrium income. They are based an the detailed

tables of Arnex 8. All farms are assumed to have produced the same output

in the absence of direct or total price interventions as they actually

produuced in the first year for which such comparisons can be made.

The income effect for cocoa is negative for the entire period

between 1953 and 1985 due to increasingly distorted autput prices. The high

negative values for 1981 in Tables 43-46 are due to the fact that the world

price of cocoa was so low in that year that the equilibrium producer price

in the absence of direct intervention was actually negative. As is to be

expected, the magnitude of the inccme effect for cocoa increases

dramatically as the length of the run increases. This is due to the

accelerating magnitude of the price and output effects.-

The directicn of the income effect for both rice and maize varies

over the period between 1953 and 1985. The effect of direct intervention is

very large in absolute magnitude in later years because of the problems

noted earlier of negative producer prices calculated by subtracting

transport and marketing margins from the border price at the official rate

of exchange. The income effects of total interventions, which are more

meaningful, are negative in most years in the instantaneous run. They

became positive in the short and long run in most years, especially after

1l970, because production in equilibrium is less than actual production

despite the awl-price effect, which would tend to increase equilibrium

producticn. Instead, the strong cross-price effects of higher equilibrium

d Described in Chapters VI and VII.

229

cocoa prices offset the own-price effects and result in decreased

equilibrium production.

Overall, then, the effects of price interventions generally tended

to benefit rice and especially maize farmers, but resulted in severe income

losses to cocoa producers. These different tendencies became especially

prcxounced towards the end of the three decades considered.

Variability Effects

A frequent argument for governnent price interventions in

agriculture is that it reduces price variability resulting from fluc-tuations

in prices on the world market or changing demand and supply conditions at

home. To examine this issue calculations were made of the variances of

and P./P-, P-/P P./P .

where P1, is the actual producer price, Par is the producer price in the

absence of direct interventions, P*ro is the producer price in the absence of

direct and indirect interventions, P, is the actual consu-mer price, P'. is

the consumer price in the absence of direct interventions, P*, is the

consumer price in the absence of direct and indirect interventions, and P_

is an index of nanagricultural prices. Separate calculations were made by

crop and for both producers and consumers. The results are presented in

Tables 51-53 and in the graphs that accompany these tables.

The conclusions drawn from the tables are mixed. They indicate

that both direct and indirect price interventions have decreased the

230

variability of maize prices.7 On the other hawd, these interventicns appear

to have increased the variability of rice prices. For cocoa, which is

subject to wide price swings mn the world market, these interventicns have

clearly reducced the variability of prices.

In additim to reducing the variability of prices, policy

interventicns may also have an effect on the variability of per capita food

consumptim. Table 54 presents data an per capita productim and

consumptimn of rice and maize. It indicates that the availability of these

cereals actually varied more than did their domestic produckticm. The

correlatim between the quantities produced and those cansumaed are, of

course, much higher for maize than for rice given hana s higher degree of

self-sufficiency in maize.

' The results for maize and rice are probably more reliable forcmsumer than for producer prices because of the anomaly discussed earlierccnceming their negative values in later years.

231

Table 51

Domestic haize PricesUnder Alternative Price Scenarios

Producer Producer Consuser ConsumerPrice in the Price in the Price in the Price in the

Actual Absence of Absence of Actual Absence of Absence ofProducer Direct Indirect Consumer Direct IndirectPrice Intervention Intervention Price Intervention Intervention

(Pp/Pna) (P'p/Pna) (Ptp/Pna) (Pc/Pnal IP'c/Pna) IPlc/PnalYear (a) (b) (c) (d) (e) (f)

1953 110 N/A N/A 161 N/A N/A1954 93 171 171 144 233 2331955 84 257 257 133 332 3321956 107 236 236 160 309 3091957 58 267 267 103 344 3441958 91 209 264 141 277 3401959 67 155 224 115 215 2951960 65 212 356 111 281 4461961 106 231 440 160 304 5441962 94 170 294 148 235 3771963 93 44 85 143 86 1331964 113 118 232 167 173 3041965 129 44 168 190 92 2351966 115 33 101 174 80 1591967 49 38 66 95 82 1141968 90 140 217 142 200 2881969 139 134 201 198 194 2711970 100 20 42 154 62 981971 95 54 93 150 103 1471972 137 367 462 198 463 5721973 125 37 47 187 85 971974 109 94 126 166 149 1861975 104 67 107 162 119 1651976 195 43 98 273 98 1611977 277 67 343 390 149 4671978 133 61 374 223 140 4991979 119 107 400 208 194 5311980 235 23 269 344 99 3821981 192 -11 332 292 60 4541982 163 -25 204 264 49 3121983 462 -34 254 618 49 3801984 132 36 184 220 110 2801985 75 40 N/A 144 104 N/A

Variance 5740 9140 12921 9803 10381 19533

232

Notes to Table 51:(a) Actual Producer Price froo Table 3-3(1), deflated by the

Nonagricultural Price Index froe Table 3-511(bl Producer Price Equivalent (OER) froe Table 3-4(2), deflated

Nonagricultural Price Index from Table 3-511.(c) Producer Price Equivalent (EER) from Table 3-4(2), deflited

Nonagricultural Price Index froe Table 3-5(1).ld) Actual Consuser Price from Table 3-3(1), deflated by the

Nonagricultural Price Index from Table 3-51).(e) Retail Price Equivalent lOER) from Table 3-4(2), deflated

Nonagricultural Price Index from Table 3-5(1).lf) Retail Price Equivalent JEER) from Table 3-4(2), deflated

Nonagricultural Price Index from Table 3-51).

233

L *~ -s.L _ _ * _ _ _ _ L L _ _ X.- _ * .L L L - f

LC- LL LO f . '.: ;:r L-J :Z- . Ld 0 4L Lo LCJ Uzi ;:J- L. 1L -.JD -L Q if) ' ' l LIZ, LJ LpC DL -:.0 .Zr . W r i M -,j *-.j :j -.j -.j -j -. j --1 --j --1j gi -i la. i-_.. ._1! J.7 1 CrJ C Li L,_l D1 rn 01 r1 1 0n

-> S -S- ;- --- -- -- -- -- ----.* n fi W - rilz - -1-;:rn W -- r--------- --- 0g.-

! I I I I I I Ii I I I I I I I I I I I I I II IJ I I i i I l 0 0

r-i' UU E

t, U'.- Ud,, - ,di °

,J~~~~~~~~~~Jo f

cj, sLJd -I Ii - - -n

'--31 t | t'|1 1;-11 -~-'s----'1 !-1d j @ :t r-a ;si lj t3z@-*@ Il J f'fiFr <,, 1 @~-; II r\

-~ -~ -~ L * -~ -. -~ -~ _ '- * ._ ' L -J *4 .- -- L A

,; LL ro . 0 i J 0 LJ ;0 LZ Z .LO 0 tz J LO i to '-f J Zi 0S LO0 ; .- J i; Lo . i Lg ."- LgjW w , c (i, i.j r..j- ..J --Lj - - -.j -- .- J - c: 4.. q **a F. ..-.-, c, i r n r n r i rn r n rn

vn A hD-e.i t txl-.1_tJ C.,,n 4- ',> ec- CD --_J 1-. , - 1. ;v[- rL C --.J .;, L9 4- CA

ci

IN~~~ **f~ -A - Ut

(CllJ ~ ~ 1 d 1/'-$r lii '.I . oc

It~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

f CS U d '' |l '

s.',Dljd,y"Z:'9'3;i-~ 111 .i I rI|:!g

@ ! r:-i z 1 ; ;-J ,r=-,,, .,} !--t , -4 a l Ji rl s iz I: rD '.1 J A '-> a :> 5I!I ri.1~ ~ ': ci G L n*L -i'. ~I n i

Table 52

Domestic Rice PricesUnder Alternative Price Scenarios

Producer Producer Consumer ConsumerPrice in the Price in the Price in the Price in the

Actual Absence of Absence of Actual Absence of Absence ofProducer Direct Total Consumer Direct TotalPrice Intervention Intervention Price Intervention Intervention

(Pp/Pnal lP'p/Pnal (Ptp/Pna) (PclPna) (P'c/Pna) (PIc/Pna)Year (a) (b) (c) (d) (e) (f)

1953 315 N/A N/A 459 N/A NIA1954 299 332 332 447 485 4851955 246 264 264 387 407 4071956 244 229 229 386 369 3691957 236 238 238 375 377 3771958 232 234 310 369 371 4581959 200 199 311 334 333 4621960 156 173 332 282 301 4841961 142 157 360 271 288 5211962 115 134 280 243 265 4331963 253 125 260 392 245 4001964 166 81 219 295 198 3571965 202 89 437 349 219 6191966 174 83 305 319 216 4711967 166 120 218 299 245 3591968 144 190 323 273 326 4801969 187 164 273 325 298 4241970 212 128 246 354 258 393

1971 202 95 191 348 225 3341972 275 111 161 433 244 3021973 292 192 242 457 342 4001974 255 294 402 409 454 5791975 280 117 212 441 253 3621976 461 98 264 667 250 4411977 387 -34 313 647 164 5631978 257 -49 306 492 140 5481979 147 -50 180 370 143 4071980 417 -67 336 686 129 5921981 215 -114 512 449 71 7911982 446 -137 447 730 60 7311983 661 -177 379 1007 43 6821994 503 -30 230 771 158 4571985 180 -5 N/A 371 158 N/A

Variance 14167 15996 6473 28028 11676 13336

236

Notes to Table 52:(a) Actual Producer Price from Table 3-3(1l), deflated by the

Nonagricultural Price Index from Table 3-5(1).(b) Producer Price Equivalent (OER) froe Table 3-4(1), deflated

Nonagricultural Price Index from Table 3-5(1).(c) Producer Price Equivalent JEER) from Table 3-4(1), deflated

Nonagricultural Price Index froa Table 3-51).(d) Actual Consuser Price from Table 3-3(1), deflated by the

Nonagricultural Price Index froe Table 3-5(l).(e) Retail Price Equivalent IOER) from Table 3-4(1), deflated

Nonagricultural Price Index from Table 3-5(1).(f) Retail Price Equivalent (EER) froe Table 3-4(1), deflated

Nonagricultural Price Index from Table 3-5(1)

237

Fi qu re 35: 3 mE:-.elti s R cProdu-c-er Pr'ices of lF ;ie9(,;Ae _ § ~~~~~~~~~~~~(Pp/,Pna)

-9- (P'p/Pno)

(Pp,a/Pno)

%~~~~~~~~

0063000

- - - - - - - - - - - - - - - - - - - - - - - - - - - - ------ - --- - - - -- -

43'

-300LC et 0 1 C r---c cr mD or-4 I< rt -t. ; r- CO r,- ,~ c-4 r-ry - Lf.l -¢ r--Ir x, N cx -. - ,nIDDeLfi l 0 4 C iL: fLE. 0 z§t& ~0 S s r-- r-. r- .r-r-- r-.r .r- N r. n , m

~~~~~~~~~--- - - T- - T-

In, Ch .1-Ti -;n m m 4 m roll ae CIN ci rzh m m T". m m :T., C-) Ch Ch m m C.I. m m QllrU C) .72h 11-1h

Yea r~~~~~~

6EZ

1 9 53 }19541955-1956-1957 - I1958

1962196) -O - '

1962 - '

19713

1 97.3 -. 8+_

19674 -L +A-. '

1 975 -~(:

-( 1976 -~~

1 978 - 1*E ' 1-.. * .A Li -)

'I1972-

1972 -197J4 - :c

1 9 7 e.f \ ~-~- _

1 9 /3 - jo-1984

1985 +~~~~ .,6 t

1982~~~~~~~~~~~~~~~~. . .s . .

-6

Table 53

Doaestic Cocoa PricesUnder Alternative Price Scenarios

Producer Producer Consuser ConsumerPrice in the Price in the Price in the Price in the

Actual Absence of Absence of Actual Absence of Absence ofProducer Direct Total Consumer Direct Total

Price Intervention Intervention Price Intervention Intervention(Pp/Pna) (P'p/Pna) (Ptp/Pna) (Pc/Pna1 1P c/Pna) (PIciPna)

Year (a) (b) (cI (d) (e) (f)

1953 651 N/A N/A 728 N/A N/A1954 729 1534 N/A 813 1618 N/A1955 745 996 N/A 834 1086 N/A1956 684 864 N/A 774 955 N/A1957 633 1400 N/A 720 1487 N/A1958 557 1194 1031 646 1282 11201959 522 939 910 606 1023 9941960 509 698 753 588 777 8321961 502 635 742 581 713 8201962 467 632 667 550 715 7501963 375 583 614 455 662 6941964 318 401 375 389 472 4471965 280 328 555 345 393 6191966 303 472 500 366 535 5631967 338 687 486 399 748 5471968 351 890 552 412 940 6131969 342 902 774 395 955 8271970 335 683 514 425 735 5661971 319 653 360 414 748 4551972 366 679 287 512 824 4331973 387 954 619 573 1140 8051974 350 984 597 575 1209 8221975 329 636 326 553 859 5491976 304 939 602 444 1079 7421977 380 819 457 686 1125 7631978 426 1421 1335 664 1659 15731919 421 737 933 645 961 11571980 289 197 772 547 455 10301981 394 -7 626 564 164 7961982 345 562 730 661 878 104619n3 293 483 841 688 878 12361994 245 321 575 605 681 93519"5 358 415 N/A 708 765 N/A

Variance 18243 116222 51440 17713 108695 70045

240

Notes to Table 53i(a) Actual Producer Price from Table 3-3(2), deflated by the

Nonagricultural Price Index from Table 3-5(1).(b) Producer Price Equivalent IOER) from Table 3-4(3), deflated

Nanagricultural Price Index from Table 3-511).(c) Producer Price Equivalent (EER) from Table 3-4(3), deflated

Nonagricultural Price Index from Table 3-51).(dl Rendered Port Price from Table 3-3(2), deflated by the

Nonagricultural Price Index froe Table 3-5(1).(eJ FOI price, converted at the official exchange rate, froo

Table 3-4(3) deflated by the Nonagricultural Price Index fromTable 3-5(1).

(f) Optieum Domestic Price at Border from Table 3-4(3), deflatedby the Nonagricultural Price Index from Table 3-5(1).

241

.~ a L -~ .... - * L L L -- A - J6 -A -L - L -A . -- I-. -_

co co O z z. {4 ;S LLL L O; ; 4D --Z (O L. O sL;NL 0 (o o ;D LI C.-tD iC. LO ss> Di 4;j 0 zW r m 1 m .al sJ -- ,- -, -j-, -j M m} 1. > lD lm. M -- -m .S 15) Ln rJ rn ;n ;n .

I 4- O- N) -, ' -'C -j , ,- ,-J .-Ai ,-J -, ,T co , , C, 4, ,, NaI i CT' ' ' '0'' I ul Q I Vi

P---- --------------------------------- <--D

~~~~~~~~~~~-k4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C

I,~~~~~~~~~~~~~~~~I

D Ud/-.'Jd) o C

I *d:/'S.j) I _('.'.if.71 I

D~~~~~~~~~~~ C.i: G~ ! -

' 2dJ@; - ! 1; r n@1r 1!

0 4; LO CO {D L : LO -L L -D -L L * * * -- LO LCD 'S -D * L CJ ** W L 1. t- i L .* _CO C ry m CO CO --i * _ -*i -_ J 4-- '- - j *.as CQ. u' crz C>m a/ ,c Cn -i n7 M rCTy rJi C. u n ry

ui 4. l * (a { 4- -- J { 1 t4 "J - .) - ) tP '-.J U(i 4- (A,J -D 0 D' X -.J a,) 4- t j, I , , I I I I I I I I I I I II O c

II4 + I

f: l-4d/;3,d.,' ~ ~ ~ "' O , *

f~~~~~~~~~1 ~ ~ 1

I,DLd ,"Q, d: El--

E ;) o'4C qs . ii j i _U n 3 L C<LI cU -9 nd]

Table 54

Per Capita Production and Consumption

Per/Capita Per/Capita-------------------------- Maize -------------------------- Rice

Maize Naize Consuser Rice Rice ConsuserProduction Availability Price Production iAvailability Price

tons) (ONd) lPc/Pna) (1Rs) IORd) (Pc/PnailYear (at) (a) (it) (b) (c) (at) (a) (at) (b) (d)

1955 31.1 26.5 133 2.7 3.1 3871956 29.7 25.5 160 2.6 3.8 3961957 28.5 24.4 103 2.5 5.2 3751958 29.5 25.1 141 3.1 4.9 3691959 N/A N/A 115 N/A N/A 3341960 N/A N/A 111 3.1 6.8 2821961 N/A N/A 160 2.8 9.0 2711962 23.7 20.2 148 2.8 12.3 2431963 25.1 21.3 143 2.9 6.1 3921964 22.7 19.8 167 3.6 8.2 2951965 26.9 23.0 190 2.7 6.2 3491966 45.0 38.9 174 2.4 8.0 3191967 34.6 30.3 95 3.4 7.8 2991968 30.8 26.2 142 3.3 6.5 2731969 35.6 30.5 198 4.6 7.2 3251970 56.3 48.5 154 3.7 9.3 3541971 52.9 45.4 150 4.1 7.4 3481972 44.6 37.9 199 5.0 6.9 4331973 46.1 39.4 187 4.4 9.4 4571974 51.1 43.5 166 5.0 8.2 4091975 35.1 29.9 162 4.7 4.1 4411976 20.5 25.3 273 4.5 4.3 6671977 26.7 22.7 390 6.9 6.7 6471978 20.7 17.6 223 6.7 10.0 4921979 35.1 29.9 209 5.6 4.8 3701980 34.4 30.4 344 4.6 9.8 6861981 33.2 30.6 292 5.5 7.5 4491982 29.6 32.1 264 2.0 3.1 7301983 14.4 21.6 618 2.2 4.6 10071984 47.0 50.5 220 3.5 7.1 7711985 32.6 35.6 144 4.6 8.7 371

Mean 34.0 30.4 199 3.9 6.9 436Variance 100.8 77.4 1.6 4.8

CORR between INd and ONs CORR between GRd and ORs

.94 .26

244

Notes to Table 54:(a) Production figures from Annex

population (for source of pActual production figures fto accout for conversion f

(b Availability figurn from Annpopulation.

(cl From Table 51.(d) Froa Table 52.

245

tA%,,'.m''rrt Per IJcipitcj Metric: Ton Per Cc3p.itr:* ' = ~~~~~~~~~~~~c- Lfl - M+r

1 ' Fa 7 A1 5N, ,

19- i - - 160I

1 959 t 1 9i 1 i1 -;

19 WI -- 1 960-

1 9 62 ;FO.:- 8,

1 M 4. 26+s.= - _ ,

1 gkl r, -- I .R-=- _ +F

1e O1967 E>-1--- 7

_:1969 \ z 15t'9'> --

1969 1 qt.l_

19707~P 5;6W+~_

17 -t 1971Z S&

1 97't 1 972 d= 1974 _ ^.1s9

1976-=_ 197. -a

1 9W§-a 197 -n__

9f 1 9N -1913,12,~~~~f D131 >~

M3~~ ~ 4, 1-'

1~~~~~~~~~~~~~. 934 g'A&,B

Flll.le e-i: F.lt:e ;l-;:Z;uc4rel-Xti At-.: A <jilclibillityl

-5- (CRd!

0 II~~~~~~~~~I0r i

*o -a. ixz) m, c- r4 i r = t-, . Lf, I4-> cc r> * tr t-. t-- a.% rb c% LOIn 'l UIn rJ 10 PtCAc V V W¢ !-,3 LP ;W~ t-- P-t--t- F- - 1-- , t- t,- tD:X| M M MMM l*n :n M ai ::$ LI C,t *n% M Cii *: th cJ 01c EIM at M. rn *::. Mo M lz M LT at M (M ai o

F=lg Lire -;1 0: n- '.-m r e r F;r-l l-e of R ',::: 1 I I I I I I I I I I III I' I I I I I IW; to t-- 0 - 0 - N tl> + 0 W t 11 I> M 1-- 1l / J I ,

In ,, M Mr>0wMMe V0I' - 1-t-t-., t..... , .- t-raMMI, M.. M

s~ ~~~ in MS a s X > Ma 0J Mr 13B1 J' rw n @ l:. OM IM M Z ) M an IT 'M M -M :D La M

_ _ _ _ ~~~~~~~~te _i r

247

PART THFEE: TK-E FAILLFE OF THE POLITICAL SYSTEM

CHWPTER X: HYFPEOTHESS ANM CONJLJISION

This chapter brings together Parts One and Two of the study to

suggest a number of hypotheses regarding the political economy of

agricultural price policy in Ghana. These hypotheses are examined over five

major phases of Ghana s historical development and conclusicns are dramn

with respect to their validity.

The Hypotheses

A number of significant hypotheses emerge from the study, which

are discussed briefly in this section.

Imoortance of Macroeconomic Diseouilibrium

The history of agricultural price policy in Ghana is a prime

example of the importance of macroeconomic forces in shaping macroeconcnic

incentives. The combination of inflation and a rigid exchange rate led to

massive price distortions, which in turn exacerbated imbalances in the

balance of payments and the goverment s budget. Efforts to increase

producer prices and to subsidize agricultural inputs were ineffective in the

face of such hugh distortions.

Loss of Real Income

There were no gainers in Ghana. Both agricultural produccers,

especially cocoa farmers, and urban consumers experienced large losses of

real income. Efforts to protect those incomes consumed increasing time and

248

energy, and in the end these effects were abandcned as people from virtually

every interest grcuip in Ghana withdrew from contact with the formal sector

and returned to food farming and other informal sector activities. Many

others emigrated outside the country.

Failure of the Pblitical System

The theory of political economy stresses the relationship that

exists between econanic policies, their impact on various groups in society,

and the influence that these groups have, in turn, on policy decisions. In

Ghana this feedback mechanism failed to operate as successive regimes

increasingly (1) made decisions within a political vacuum because of their

isolation from the Ghanaian populace and (2) fcund themselves confronted

with such binding macroeconomic constraints that there was virtually no

scope for independent decisions.

Rise of Rent-Seekinq Activity

Given the government's increasing inability to influence the

allocation of resources through price policy, its efforts went towards

economic management through extensive administrative controls. These took

the form of import licensing, exchange controls, marketing board operations,

input distribution, credit allocation, and other mechanisms. EnormoLus

possibilities were created for arbitrage between the informal market and

formal, government-controlled distribution channels. This led to rent-

seeking activity an a massive scale. Cbovernment regulations were subverted,

graft and corruption were rampant, and patron-client relations became

249

entrenched as the chief means by which most people could gain access to

scarce rescurces.

Neulect of Price Pblicy

In addition to the objective constraints imposed on output price

policy by macroeconomic disequilibrium, successive governments maintained a

strong bias towards intervening in agriculture through projects, public

investments, input subsidies, and cheap credit rather than by influencing

producer prices. Since rescurces were very limited and many investments

were unproductive, the result was very little increase in production and a

strong bias in favor of large farmers with contacts and influence.

Failure to Achieve National Objectives

During each of the phases described lbelow, the regime newly in

pow-er was fairly clear as to its national objectives and mDre or less how

these were weighted. Yet underlying macroeconomic constraints and the

entrenchment of rent-seeking activity within an elaborate network of patron-

client relationships soon dissipated any notions that much progress in

achieving these objectives was to be achieved. Instead, the regime found

itself increasingly embattled and unable to act decisively. As a result it

became politically isolated and was forced to rely on military weight as its

means of staying in power.

The Phases

The periods during which the political econcomy of Ghana evolved as

described in Chapter IV can be grouped together into five phases. During

250

each of these, the government in power could be characterized as having had

a distinc:t set of objectives and objective weights. Each regime, with the

exceptimn of the Rawlings gcvernment currently in power, found itself

frustrated in its attempts to achieve its objectives and was eventually

displaced.

Colonial Period

The colmnial government was essentially liberal and strongly

oriented towards export trade, and yet it created during this period the

cocoa export manopoly that was to dominate the industry. Oice the CRP came

to power, heavy emphasis was placed an extracting the resources from cocoa

that would finance its investment program and an penetrating politically the

rural sector.

Nkrumah

After inderendence in 1957, the Nkrumah government was intent an

industrialization and increased investment in urban services. It was also

highly sensitive to the welfare of its urban political base, and yet it

allowed inflation to eat away real incomes. Government revenue and foreign

exchange became increasingly important as objectives once exchange reserves

were exhausted and low world cocoa prices squeezed public revenues.

Finally, towards the end of this phase, there was heavy emphasis on large-

scale farms in the rural sector at the expense of small farmer development.

251

National Liberation CGuncil (NLC) and Busia

The National Liberation Council attached strong importance to

macroeconomic stability. Emphasis was also placed on stimulating cocoa

exports by increasing producer prices. There iwas a desire to liberalize

imports and to reduce the prevalence generally of quantitative controls.

The state was seen as a regulator of private activity rather than as having

an impor-tant directly productive role. Equality of income distribution was

given less weight, and the regime of Busia had a decided regional bias.

National Redemption Council (NRC). Supreme Military Council (SMC). andLimann

The NFR come to power committed to reversing the trend towards a

more liberal economy. Emphasis was placed on agriculture, but through

projects, public investments, and input subsidies rather than by raising

producer prices. Easing Ghana s dependence an foreign exchange was

perceived as an important objective but the mechanism to achieve it was to

tighten input restrictions and to increase self-sufficiency rather than to

increase exports. Mbre emphasis was placed on improving the distribution of

income and promoting regional equality than on maintaining macroeconomic

equilibrium.

Provisional National Defense Councril (PNDC)

After the brief interlude of the Limann government, which did

little more than attempt to cope with the disastrous economic situation, the

PNDC under Jerry Rawlings came to power with a strcng populist mandate.

Althtugh its initial views were strongly interventionist in nature, the

government fosund itself virtually obliged to adopt a more liberal framework

252

of objectives. Principal among these was macroecmnomic stability, but not

far behind were liberalization of imports and growth of exports as ways of

dealing with the foreign exchange crisis. Welfare of urban consumers was

not given high priority, nor was greater equality of regional and inccme

distributimn. Greater weight was placed mn agriculture thrnugh the

mechanism of output price policy, especially for cocoa and the industrial

crops. Less immediate attentimn was given to industry, though in the long

run its growth was perceived as essential.

Summary

Table 55 presents rough estimates of the weights attached to

various objectives during each of the phases described above. The weights

are assigned, to the extent possible, independently of the particular

cnmstraints with which each regime was faced. Thus they generally tend to

reflect relative priorities at the beginning of each phase, before the

econmic realities were fully felt.

Of particular importance in the table are the sharp shifts in

weights that occurred frum mne phase to another. Consumer welfare, for

example, was given high priority by Nkrumah, who depended on his urban base

of political power, but the military government after Busia gave this

objective much less weight. In contrast, farmer welfare was least

important to Nkrumah but gained in importance with succeeding governments as

they realized the importance of the rural sector to the econwom. Cbvernment

revenue was very important to Nkrumah as a means of enhancing the power of

the state and of financing his ambitious investment program; to other

253

gcverriments, it was required merely to keep up with the expansion of

recurr,ent expenditures in the face of rising public sector

Table 55

Relative Weight of Agricultural Price Policy Objectives

Phase(years)

Colonial Nkrumah NLC.Busia NRCISMC PNDC(1950-57) (1957-66) (1966-71) (1972-79) (1982-

Obiectives _ Present)

ConsumerWelfare .20 .25 .--) .10 .15

FarmerWelfare .05 .00 .15 .15 .20

GovernmentRevenue .25 .35 .15 .15 .15

PriceStability .20 .10 .25 .05 .270

ComparativeAdvantage .20 .0o .15 .00 .15

Food Self-Sufficiency .0o .05 .00 .20 .05

Industrial-ization .05 .25 .10 .10 .05

IncomeDistribution .00 .00 .00 .10 .05

RegionalBalance .05 .00 .00 .15 .05

Total 1.00 1.00 1.00 1.00 1.00

employment. The relative importance attached to price stability has varied

as perioids of rapid inflation have been succeeded by strong efforts to

restore macroecocrmic equilibrium.

254

Regimes have also differed markedly in the emphasis they have

placed on exploitation of Ghana' s comparative advantage versus food self-

sufficiency and import-competing industrialization. The colonial

government's emphasis on the export trade, for example, was followed by

Nkrumah's neglect of cocoa and concentration on industry. The NLC, on the

other hand, shifted back towards a more balanced view, whereas the NRC

strongly favored food self-sufficiency as the means to reduce depEndence on

foreign exchange. The present regime, in contrast, favors once more

exploitation of Ghana's comparative advantage.

Finally, equity of income distribution and regional balance have

never received very high priorities as objectives, except possibly under the

NRC. The Nkrumah and Busia governments were conspicunus in their regional

bias. The PNDC appears to be aware of distributional considerations but is

so intent on getting the economy moving again that it has not yet been able

to focus on these issues.

Testing of the Htvpotheses

The hypotheses presented earlier are tested here within the

context of the five phases just discussed.

Macroeconomic Disecnuilibrium

There is no question that until recently the effects of

macroeconomic disequilibrium overshadowed all other distortions in the

agricultural economy. Nkrumah s ambiticus investment program was severely

hampered by the decline in resources available from international trade when

would cocoa prices fell. As was shown in Chapter V, total public revenue

255

fron cocoa decreased dramatically during the period 1961-66 both in absolute

terms and as a share of the total value of sales. In real terms the loss

was even greater. Measured in 1972 constant prices, total revenue from

cocoa equalled NY 146 million in 1960 and NV 17 million in 1966. In 1965 it

was actually negative.

This decline in government revenue was not because farmers were

being let off easily. As shown in Table 9 of Chapter V, real producer

prices for cocoa decreased in real terms from S 546/mt in 1960 to N! 27B/mt

in 1965 (1972 constant prices). Yet this was a period, as demonstrated

thrcaughout Part Two, during which cocoa was relatively lightly taxed. The

unweighted average of nominal rates of protection for cocoa, with anly the

direct effects of price policy being taken into consideration, was -0.24 at

the producer level from 1960 to 1965. In five of the 32 years between 1954

and 1985 negative protection was at least 60 percent, and in 17 of the years

it was at least 40 percent.-

Despite its emphasis on price stability, the NLC/Busia regime was

never able to gain full control over the economy. Althouagh inflation was

slcwi.d, it still was close to double-digit figures by the time that the NFC

took over in 1971.2 Furthermore, the massive debt problem faced by Ghana

after Nkrumah was never resolved during this phase.

Uhder the NRC/SMC/Limann regime macroeconomic mismanagemnt proved

disastrous. Inflation reached triple-digit figures in 1977, and than again

in 1981. The real value of the exchange rate appreciated sharply with the

government s refusal to undertake any more than one nominal devaluation, and

± Table 16, Chapter VI.

2 Table 3, Chapter II.

256

as denEstrated in Part Two, this was the major cause of price distortions

during the late 1970s and early 19E0s.

The PDNC, in contrast, has mffved towards macroeconomic equilibrium

and, at the same time, institutionalized periodic adjustments in the

exchange rate. Thus even if the rate of inflation increases in the future,

the magnitude of distortions that characterized the recent past should not

be repeated.

Loss of Real Income

The losses of real income by cocoa farmers have been documented

in Chapters IV and IX. Rice and maize farmers, on the other hand,

experienced mixed effects, and in some years especially after 1972, they

gained appreciably from high rates of protection in the face of an

increasingly overvalued exchange rate and a lack of foreign exchange with

which to purchase imports.

Urban workers, on the other hand, were greatly injured by

inflation. An index of the average wage in manufacturing deflated by the

urban CPI fell from 100 in 1972 to 49 in 1976, 24 in 1979, and 11 and 1983.

Public sector real wages fell to approximately the same extent. Thus formal

sector employees were completely unable to protect themselves from real

inccmne losses. As a result, many were forced to devote an increasing amoit

of time to informal sector activities, including rent-seeking where

opportunities were available. Others simply abandoned the urban economy and

returned to the countryside, where food production offered a better hedge

against inflation.

257

Failure of the Political System

These losses in real income suggest that cocoa farmers and urban

workers had little influence on policy decisions affecting their welfare.

Yet they were much better able to organize than were most fcod farmers and

those in the informal sector. The conclusion that must be drawn, therefore,

is that the political system had broken duwn to such an extent that interest

graoups were no longer able to apply pressure effectively on those in power.

Looking at the cocoa farmers, it is clear that Nkrumah's

penetration of the Ashanti and other cocoa growing areas by the UGFOC had a

lasting political impact. His major opponent in this region was the

National Liberation Mavement (NLM), which was never very successful because

of its concentration en a regional constituency and separatist approach.

The LUFCC was therefore able to subvert the NLM s influence by focusing en

the cocoa producer price as the main issue.

Under aisia s Progress Party, cocoa farmers gained in power, but

this itself contributed to the demise of the regime. With the deterioration

of the economy that followed, there was no room for raising producer prices

further, given the increasingly overvalued exchange rate. Political

pressure, furthermore, was useless. Confronted by its own ineptitude and

corruption, the regime turned inward, relying more and more an muilitary

force.

Rent-Seeking Activity

The Nkrumah years saw the development of two major institutional

arrangements that provided abundant opportunities for rent-seeking. The

first was the Cocoa Marketing Board and the UGFCC. The first has survived

258

more or less intact; the second was dismantled in 1966 but was later

resurrected as the Produce Buying Company, albeit without the political

trappings of the UGFOC. The second arrangement was the system established

for allocating import licenses and access to foreign exchange. Each of

these is described in Chapter V, where the problems they presented are

spelled out in detail.

The introduction of these opportunities for rent-seeking behavior

into key areas of the econcomy provided an alternative to political influence

and power as a way to acquire scarce resources. Elaborate patrcn-client

relationships were established that enabled those without direct contact

with the state to nevertheless participate in the distribution of state-

controlled goods and services. Intermediaries, whether or not they were

part of the state, of course benefitted.

The growth of rent-seeking had two profound implications. First,

it drew resources away from productive activities that could have increased

the total quantity of goods and services available to society. This was

particularly true of urban areas, where the state's influence still reached.

Seccnd, it undermined the development of the political institutions and

processes at the state level that would have allowed interest groups to

influence policy-mnakers directly. Instead, political development occurred

on a mutch mDre decentralized and largely rural basis that ignored the state

and concentrated instead an local political systems. Only with the rise of

the populist leader, Jerry Rawlings, was there once more an increase in

state-level political activity. It is no accident that Rawlings and the

PNDC have loudly condemnried rent-seeking behavior.

259

Neglect of Price Policy

The combination of macroeccnomic constraints and the failure of

the political economy to operate effectively discouraged the use of output

price policy as the primary instrument for allocating resources and

transferring welfare. Instead governments relied on the mechanisms that

they could control - projects, public investments, input subsidies, and

cheap credit. Under Nkrumah this took the form primarily of the

establishment of state farms. Projects were later established to aid the

small private farmer, but these generally involved large public sector

bureaucracies, such as the central ministries or parastatals. The

NRC/SMC/Limann phase, particularly, was dominated by these plus input

subsidies and cheap credit.

The major problems with this approach were two-fold. First, the

state farms, projects, and parastatals required strang management skills,

which were and still are the scarcest resource in Ghana. Second, the

recurrent cost requirements of these institutions were enormous. Given the

severe budgetary constraints that existed in a rapidly deteriorating

economy, most financial resources were used to pay salaries and little was

left over for the operating expenses that would have enabled these

institutions to be effective.

As far as input subsidies and cheap credit are concerned,

shortages of management and operating funds implied that anly the largest,

most accessible farmers could be reached. Furthermore, these subsidies

undermined the development of private sector input marketing, which might

have gotten to the smaller farmer. This it was the NRC/SMC-gcvernment,

which originally placed greatest emphasis on more equal distributimn of

260

income, that probably contributed most to discrimination in favor of

wealthier farmers.

Failure to Achieve National Objectives

The intentions of governnents during the first four phases

described in this chapter were frustrated by the economic and political

realities of the time. The colonial regime was perhaps most successful

because of the cocoa boom of the 1950s. Yet Nkrumah's desire during this

period to mobilize resources for the state was soon confronted with the

ccnsequences of the collapse in world cocoa prices. Although some

industrialization was also achieved, it proved to be remarkably unproductive

and a severe drain on public resources. Finally, Nkrumah's emphasis on

consumer welfare, in the form of cheap food in urban areas and rising urban

wages, was undermined by the need to restrict access to foreign exchange and

the problems of balancing the budget. The nominal rate of protection on

rice and maize, for example, rose from -0.06 and -0.60 respectively in 1960

to 0.59 and 1.06 respectively in 1965.3 In the end, it was the loss of

confidence of the urban population that brought about Nkrumah's downfall.

The NLC and Bisia's government placed more emphasis on

agriculture, but it was during this period (1967-69) that negative

protection on cocoa, rice, and maize was greater than at almost any time

during the entire three decades, leading to large transfers out of

Z Table 17, Chapter VI. These are based on the effects of directprice interventions. Based on the effect of direct and indirect priceinterventions, from Table 19, the EWR was -0.36 for rice and -0.73 for maizein 1960 and -0.35 for rice and -0.07 for maize in 1965. Thus once theovervalued exchange rate is taken into account, consumers continued to besubsidized vis-a-vis world market prices in 1965, but to a lesser extentthan in 1960.

261

agriculture.4 Furthermore, confidence in the regime was eroded by declining

real urban incomes in 1969 and 1970, and rising inflation in 1971, despite

the governnent s ccnwmitment to increasing consumer welfare and maintaining

stable prices. Ohce again, it was a combination of macroeconomic crisis and

a weak political base, plus overreaction by the government to the situation,

that resulted in the take over by the NRC.

The NRC/SMC was committed to food self-sufficiency, improving

farmer welfare, and a better balance in the distribution of family and

regional income. The instruments used to achieve these objectives,

however, required intensive use of scarce management and financial

resources. As the economy deteriorated and these resourrce constraints

became more severe, the projects and programs of the government collapsed.

As output prices became increasingly distorted, moreover, cocoa farmer's

income and welfare declined.E To the extend that: greater regional balance

was achieved, this was more because of these losses than the gains achieved

by foodcrop farmers. Furthermore, food self-sufficiency was not attained

since Ghana became increasingly dependent until 1984 on food aid imports.

4 Tables 39-41, Chapter VIII.

s Tables 43-50, Chapter IX.

262

Conclusions

The hypotheses proposed here appear to be well supported by the

evidence presented in this study. It was the failure of politics to provide

an effective mechanism for influencing policy in the allocation of resources

that led ultimately to the collapse of the system. Rent-seeking took over

instead, but the diversion of resources to these activities and away from

those involving pr-oduction severely eroded the resource base. Marginal

retums to rent-seeking decreased dramatically and ultimately induced people

to minimize their contacts with the state. Decentralized political activity

in rural areas correspondingly flourished.

Rawlings came to power as a populist leader strongly committed to

the elimination of rent-seeking activities. The first attempts at doing

this by suppressing private trade were counter-productive. Cbce the

decision was made to liberalize the econaoy and to reduce price distortions,

hioever, the incentives for rent-seeking decreased. Once again resources

were allocated in accordance with relative prices rather than administrative

regulations and discretion. The process, however, is not yet complete.

NDr is it clear that institutions are being developed that will

enable Ghanaians to influence policy decisions in their favor through

political action. This will take time and the confidence of Ghanaians in

the current regime. Without this development, however, past history

suggests that the government can be expected to enccounter increasing

frustration and hostility to its actions.

263

ANNEX 1: AGRICLLTlUPL FiODUCTION

The purpose of this annex is to describe and evaluate the system

for collecting agricultural production data and to present the time series

of available data by crop. Comparisons are also made of alternative

estimates of cocoa sfftggling, and the cocoa production figures are adjusted

accordingly.

System for Data Collection

Cocoa

Llntil 1962, the Cocoa Division of the Ministry of AgricuiltLure

regularly prepared estimates of the area under and prodLuction of cocoa based

on sample measurements. ± The Cocoa Marketing Board (CMB) provided

information on the quantity and valute of cocoa purchases from producers,

marketings to foreign buyers and local grinding mills, and changes in

stocks. External Trade Statistics of Ghana, published from customs

information by the Central Eutreau of Statistics, provided data on the

quantity and valute of cocoa exports. Comparison of the data from different

sources permitted checks on the quality of the data.

After 1962, the quality of data on production deteriorated, and

better estimates of produtction were obtained by adjusting Cocoa Marketing

Board sales for changes in stocks. This ignored, however, the problem of

cocoa being smuggled across Ghana's borders. Table 1-1 presents data on

prodLction estimated by the Cccoa Marketing Board and two sets of estimates

X- This discussion is taken from M.S. Singal and J.D.N. Nartey,Sources and Methods of Estimation of National Income at Current Prices inGhana, Ghana Central ELUreau of Statistics, 1971, pp.3.-13.

265

Table 1-1

Cocoa Production and Smuggling 1960-19B2(000 At)

Smuggled Smuggled CocoaYear Production (a) Cocoa (b) Brong-Ahafo lc) Total (d)

1960 430 10 6.6 10.31961 409 8 5.9 9.41962 413 14 -1.9l -2.91963 428 11 -7.6 -12.01964 538 14 -6.1 -9.21965 401 17 .9 1.31966 368 17 37.13 57.11967 415 21 31.5 46.81968 323 17 19.2 26.51969 403 25 23.3 33.91970 413 31 17.9 34.31971 454 37 21.4 32.41972 407 42 13.8 21.91973 340 34 9.2 14.11974 376 30 15.9 26.61975 3-96 38 20.13 34.41976 320 40 21.4 35.01977 371 45 37.NA 64.31978 365 50 29.8 60.81979 296 N/A 31.8 56.01980 258 N/A 31.4 61.51981 225 N/A 24.1 46.31982 180 N/A 21.2 45.4

Notes to Table 1-1:(a) Cocoa production data are based on Cocoa Marketing Board estiaates,

the sources for which are given in Table 1-2.(b) Estioates of Ghanaian cocoa officials and international traders from

R.6. Franco, 'The Optiaal Producer Price oi Cocoa in Shana,mJournal of Developeent Economics, Vol. 8, 1981, pp.77-92.

(c) Econometric estimates from Ernesto May, Exchange Controls andParallel Market Econoeies in Sub-Saharan Africa--Focus on Shana,World Bank Staff Working Paper No.711, 1985, p.69.

(d) Derived by dividing Brong-Ahafo estisates by the share of the Brong-Ahaforegion in the production of all regions bordering neighboring countries.

266

of cocoa being siLtggled out of the country. The first consists of

guess-estimates of Ghanaian cocoa officials and international cocoa

traders. The second is derived from an econometric model developed by

Ernesto May that estimates smtuggled cocoa as a function of the price of

cocoa in the neighboring country converted to cedis at the black market

exchange rate, the official producer price of cocoa in Ghana, and cocoa

prodiuction capacity in Ghana. 2 Reasonable estimates were obtained only for

the Brong-Ahafo Regime, which borders the Ivory Coast and Luntil recently

produced almost twice as much cocoa as the other two regions bordering

neighboring coLuntries -- the Western and Volta Regions.3 Estimates for

total cocoa smuggled from Ghana were obtained by dividing May's Brong-fAhafo

figures by the share of that region in the production of all regions

bordering neighboring countries.

The two series on smuggled cocoa, while differing considerably

from year to year, do suggest that up to 5>-6O.-),CX tons of cocoa have been

smuggled ouLt of Ghana in some years. They also indicate that important

changes have taken place over time. During the period from 196:) to 1965,

for example, the quantity of cocoa being smuggled out of Ghana was always

less than 5 percent of total production. May's estimnates suggest, in fact,

that the direction of flow may have been the other way, that is that cocoa

may have been smutggled from the Ivory Coast into Ghana in some years during

this period.

2 Ernesto May, Exchanoe Controls and Parallel Market Economies inSub-Saharan Africa - Focus on Ghana, World Bank Staff Working FaperNo. 711, 1985.

3 Smuggling from regions in the interior is considered to be negligible.

267

In any event, the level of smuggling increased markedly after

1965. May' s work indicates that this took place very dramatically beginning

in 1966 because of a fall in the official producer price in Ghana and a

sharp depreciation of the cedi on the black mark:et. 4 Expert opinion seems

to be that the process was more gradual. The two sets of estimates also

differ somewhat with repect to the timing of other changes in the rate of

smiuggling between the mid-1%96s and the mid-197C)s, but they each suggest

that the average level was between 5 and 10 percen t of production. 5 They

also agree that the rate of smuggling increased again during the latter half

of the 197C)s. This was becatse increases in producer prices failed to keep

pace with inflation and the depreciation of the cedi on the black market.

By the early 198)s, smuggling may have accoLnted for close to 20 percent of

production.

It is possible that these figures are overestimated. As the

economic situation deteriorated dur-ing the 1970s, Ghana's transportation

system increasingly degenerated. In addition, the Brong-Ahafo Regicn

declined in importance as a source of supply relative to the Western Region,

which had poorer road linkages with the Ivory Coast. Finally, periodic

government crackdcons on cocoa smuggling undcoubtedly acted as a deterrent.

Experts close to the situation, in fact, have estimated smuggling during

" Until 1966, the NV/$LS rate on the black market was relatively closeto the official rate of .714 N&/$US. As a resLult of a strong tightening oftrade and exchange controls in that year, the black market rate moved to2.13 Nhl/U5. See May, Exchanae Controls..., p.129.

> One reason for the difference in timing may be learning effects andother forms of capital investment in smuggling that cause the long runresponse to variations in incentives to differ from the response in theshort run, which was estimated by May.

268

the early 199:s to be probably between 10,(X:} and 20,OCKx) tons per year.*

This woAuld amnount to approx>imately 1-0 percent of current production.

Cocoa Marketing Eoard estimates of production are adjusted for

smtggling here on the basis of this more conservative hypothesis regarding

its level during the early 1990s. As indicated in detail in the footnotes

to Table 1-2(3)! cocoa smutggling is assumed to have been nil until 1966,

when it beqan to increase as a result of substantial price incentives.

With increa,ses in producer prices in the late 1960:s and early 1970s,

smuggling is assL.uTed to have remained constant at 5 percent of official

purch-Ases from 1971 to 1975. Thereafter it rose again until it reached 10

percent of production in 193:.

Other Crop'

The Economics and Marketing Division of the Ministry of

Agriculture is responsible for estimating production of crops other than

cocoa. Beginning in the mid-1960s, the estimates were obtained by

multiplying the area Lunder cultivation by average yields for each crop under

consideration. Data on cultivated area were compiled from censuses for

large farms and from sample surveys for small farms. A broader sample

survey was Lndertaken in 1970 to measure area Lnder production of main crops

and livestock: numbers. FolloitAp surveys were then conduLcted in 1971 and

1972 to update the 1970 estimates and to obtain information on yields,

labor inputs, and farm machinery and equipment. Althcxtgh the usutal

6 World Bank, Ghana: The Cocoa Sector, Background Faper No. 1 of 4prepared for Ghana: Folicies and Proqram for Adiustment, October 14, 19E3,p.5.

269

difficulties were encountered with these surveys, the results are considered

to be reasonably reliable.

As foreign exchange constraints tightened after 1972. it became

increasingly difficult to maintain this system for data collection.

Shortages of fuel and spare parts limited the mobility required for valid

sampling. Instead, agricultural officers were obliged to report area and

production figures based on their informed opinions. There clearly, at this

point, was a deterioration in the quality of the data. Whether a bias was

introduced, however, is not so clear.

Aqricultural Froduction Data

Froduction data are contained in Tables 1-2(1) throrugh 1-2(3) for

17 crops from 193:) to 1985. Sources of data are given in the notes to the

table. The principal original sotrce is the Ministry of Agriculture, though

most of the data were obtained from World Bank Reports and the Central

Bureau of Statistics Economic Sulrvey 1969. The original source for the

cocoa statistics is the Cocoa Marketing EBard.

Except for cocoa, data on production prior to the micd-1'?96s are

taken from FPO Production Yearbooks. These data TLtst be treated with

caution since they are not based on the system of sample surveys described

above, but on cruder estimates of field personnel as corrected and inter-

polated by FAO. Nevertheless, they probably reflect reasonably well the

orders of magnitude involved.

The same may not be true of the figures on production after the

early 1970s. These show a sharp decline in food production that seems

inconsistent with the evidence presented elsewhere in this study that food

270

prices rose significantly in relation to prices of cash crops. There are at

least two possible explanations for this decline. The first is that Ghana

experienced very severe drought from 1975 to 1977. especially in the

northern area. The second reason, which has been more persistent, is the

almost total breakdown in the transport system due to lack of fuel and spare

parts and the inability of the government to maintain the road system. This

led to a strong decline in food marketings. which had been quite substantial

in earlier years. This decline was abetted by the decrease in per capita

in,comes that was occLurring across broad segments of the ulrban population.

271

Table 1-2(l)

Agricultural Production (a)(000 Its)

Cereals staples

Year maize Rice Sorghus Millet Cassava Yale Cocoyae Plantain1930 N/A -/A N/A N/A _ /A -/--N/A ---

1950 N/A N/A N/A N/A N/A N/A N/A N/A1951 N/A N/A N/A N/A N/A N/A N/A N/A1952 N/A N/A NIA N/A N/A N/A N/A N/A1953 NIA N/A NIA N/A N/A N/Ak NIA NIA1954 N/A N/A 79 99 512 481 NIA N/A1955 169 23 79 99 512 481 NIA N/A1956 169 23 79 99 512 481 N/A N/A1957 169 23 90 112 823 481 N/A N/A1958 183 30 N/A NIA 1092 N/A N/A N/A1959 NIA N/A N/A N/A N/A N/A N/A N/A1960 N/A 32 N/A N/A NIA N/A N/A N/A1961 N/A 30 100 80 N/A N/A N/A N/A1962 169 31 105 65 789 1001 N/A N/A1963 183 33 109 68 1194 1098 N/A N/A1964 170 42 113 70 1229 1153 N/A NIA1965 206 32 89 57 1100 1055 NIA N/A1966 353 29 107 66 1152 1170 NIA N/A1967 278 42 83 78 1156 1219? N/A NIA1968 253 42 72 57 1058 1382 NIA N/A1969 300 60 96 90 1350 2000 N/A 7001970 482 49 186 141 2388 909 1136 16411971 465 55 173 130 2388 909 1136 16411972 402 70 152 99 2840 679 945 16701973 427 62 167 109 2865 606 1325 20711974 486 73 177 154 3606 850 1510 20241975 343 71 185 122 2398 709 1099 12461976 286 70 189 144 1819 575 773 12561977 274 109 181 125 1811 535 722 9271978 218 108 121 98 1895 544 726 9401979 380 93 158 149 1759 602 749 8171980 382 78 132 82 2322 650 643 7341981 378 97 131 119 2063 591 631 8291982 346 36 85 76 2470 588 628 7451983 173 40 56 40 1729 866 720 3421984 574 66 NIA N/A 4083 N/A NIA N/A1985 411 90 62 54 3076 485 581 676

Notes to Table 1-2(1):(a) Data source for years prior to CBS Survey is FAO Yearbooks.

Data source for 1964-1969 (maize, rice, sorghum), 1966-1969 (millet, cassava, yam),and 1969 (plantain) is Central Bureau of Statistics, Econosic Survey 1964.

Data source for 1970-1983 is World Bank, Ghana Agricultural Sector Review,Annex 5, August 6, 1985.

Data source for 1984 and 1985 is the Ministry of Agriculture.

272

Table 1-2(2)

Agricultural Production (a)(000 its)

Edible Nuts & Seeds

Ground- Coco- OilpaleYear nuts nuts bunches Cowpeas Sugar Cane

1950 N/A N/A N/A N/A N/A1951 N/A N/A N/A N/A N/A1952 N/A N/A N/A N/A N/A1953 N/A N/A N/A N/A I/A1954 N/A N/A N/A N/A N/A1955 44 N/A N/A N/A N/A1956 44 N/A N/A N/A N/A1957 41 N/A N/A N/A N/A1958 41 N/A N/A N/A NIA1959 49 N/A N/A N/A N/A1960 N/A N/A N/A NIA N/A1961 N/A N/A N/A N/A 711962 N/A N/A N/A N/A N/A1963 47 N/A N/A N/A 3931964 50 N/A N/A N/A 4011965 27 N/A N/A N/A 1001966 38 N/A N/A N/A 1471967 39 N/A N/A N/A 2951968 62 N/A N/A N/A 2401969 61 N/A N/A N/A 3001970 102 302 696 11 1121971 102 302 696 11 1121972 89 296 711 9 1451973 127 306 782 7 1611974 157 307 917 11 1711975 111 311 901 11 2051976 113 229 281 12 1901977 8l 152 739 11 2581978 83 159 935 8 2721979 107 159 1012 9 1901980 92 H/A N/A N/A 1921981 100 N/A N/A N/A 1901982 110 N/A N/A N/A 1101983 70 N/A N/A N/A 1001984 N/A N/A N/A N/A N/A1985 N/A N/A N/A N/A N/A

Notes to Table 1-2(2):(a) Data source for 1950-1969 is FAG Production Yearbooks.

Data source for 1970-1979 (1970-1978 for sugar cane)is Norld Bank, Ghana Agricultural Sector Review,August 6, 1925, Annex 5, p. 63. Data source for

Data source for 1980-1983 (1979-1983 for sugar cane)is FAO,

273

Table 1-2(3)

Agricultural Production(000 its)

SeedYear Cocoa 'aj Cocoa (b) Cotton (c) Rubber (c) Tobacco (c)

1950 262 262 N/A N/A N/A1951 211 2-1 N/A N/A N/A1952 247 247 N/A N/A N/A1953 211 211 N/A N/A .101954 220 220 N/A .30 .101955 229 229 N/A .30 .101?56 264 264 N/A .30 .101957 206 206 N/A .30 .101958 256 256 N/A .40 .301959 317 317 N/A .60 N/A19?60 430 430 N/A .60 N/A1961 409 409 N/A .80 N/A1962 413 413 N/A .30 .701963 428 428 N/A .30 .801964 538 538 N/A .30 .901965 401 401 N/A .20 1.001966 368 372 NIA N/A 1.001967 415 423 N/A NiA 1.201968 323 333 N/A N:A 1.101969 403 419 N/A N/A 1.101970 413 434 .20 .70 .901971 454 477 .40 1.00 1.401972 407 427 1.30 1.20 1.701973 340 357 1.90 1.30 2.601974 376 395 2.50 1.80 3.301975 396 416 3.00 2.10 2.301976 320 339 9.00 2.70 2.101977 271 290 11.40 2.70 .801978 265 286 4.50 3.30 .801979 296 323 5.20 3.30 .801980 258 284 3.00 1.70 .401981 225 248 .70 .70 .501982 181 198 .50 .70 .601983 157 173 .50 .50 .501984 172 189 N/A N/A N/A1985 215 226 N/A N/A N/A

274

Notes to Table 1-2(3):(a) Source of cocoa production data is Cocoa Marketing Board, as

reported in the following: 1950-59, Merrill J. Batemn, 'An6ccnometric Analysis of Ghanaian Cocoa Supply,' in R. A. Kotey,C. Okali, and B. E. Rourke, Economics of Cocoa Production andMarketing, Institute of Statistical, Social, and EconomicResearch, University of 6hana, Legon, 1974, p.315; 1960-78,World Bank, 6hana: The Cocoa Sector, Background Paper No. I of 4prepared for the 6hana: Policies and Progras for AdjustsentReport, October 14, 1983, p.45; 1979-1984, World Bank, Ghana -Towards Structural Adjustment, Vol. 2: Statistical Appendix,October 7, 1985, p.57.

Ib1 Cocoa Marketing Board (ChB) estimates adjusted for smugglingunder the following assumptions:1950-65 negligible smuggling1966-70 smuggling increases from nil to 5 percent of CNB

estimates in equal annual increments1971-75 smuggling remains constant at 5 percent of CMB

estimates1976-80 smuggling increases from 5 to 10 percent of CMB

estimates is equal annual increments1981-85 smuggling remains constant at 10 percent of CHI

estimates(c) Source of data for 1950-69 is FAG Production Yearbooks; source

for 1970-83 is World Bank, 6hana Agricultural Sector Review,Annex 5, August 6, 1985, p.63.

275

INNEX 2: EQUILIERIUM EXCHANGE RkTE

This annex describes in detail the steps that were taken to

calculate the equilibriLun exchange rate (tt) for Ghana over the period

1953-84. The method consisted of estimating demand and sLtpply fLmncticns for

foreign exchange on current accoLnt and finding the exchange rate that would

have equilibrated the dollar value of imports with the value of exports in

the absence of taxes or quantitative restrictions on trade. ELefore

describing this methodology in detail, however, it is useful to consider two

alternative approaches to estimating the equilibrium exchange rate. These

are based on (1) the notion of purchasing power parity (FPF) and (2) the

elasticities of demand for imports and supply of exports, with adjLstment in

each case for chhanges in the terms of trade and the strcLture of capital

flows, debt service, and transfers.

Furchasina FPower Parity Aprroach

The Real Exchange Rate

This approach starts with the notion that it may be possible to

find a year in which the current account! eXclusive of unrequited transfers,

is more or less in balance, consistent with the structure of transfers and

capital flows, and to adiust the nominal exchange rate (NOR) for changes

from this base year in the absolute level of prices in the comntry being

studied compared with levels in its major trading partners. This "real"

exchange rate (RER) is the ratio of two sets of prices. The first is the

NOE, which is the price of one currency in terms of another; the second is

an indicator of the relative change between the two countries in the

purchasing power of each currency over goods and services priced in terms

276

of that currency. Miltiplying these two ratios together provides a measure

of the relative change in purchasing power over goods and services priced

in terms of a single currency.

The pu.rchasing power parity theory is derived essentially from a

monetary approach to balance of payments adiustment. To the extent, for

example, that monetary expansion generated a more rapid rate of inflation in

Ghana than existed in the rest of the world! and yet the exchange rate

remained fixed, the prices of nontradables in Ghana rose more rapidly than

the prices of tradables, causing consumers to shift towards the consumption

of tradables and producers to move in the direction of producing more

nontradables.1± The current account moved into deficit under these

circumstances, and since this could not be sustained over the long run

through capital flows or transfers, the existing exchange rate was not cne

of equilibrium.

In the absence of government price or trade policy, and assuming

no structLtrJl changes, equilibrium in the long rmn would have been achieved

throuLgh some crombination of a movement in the exchange rate and in the

absolute level of prices in Ghana, compared with its major trading partners,

that would have offset the original price changes caLsed by monetary

expansion and inflation. Regardless of whether it is the NER or prices

that woLld have adiusted, however, the RER would have remained the same.

I Nontradables are defined theoretically as goods and services forwhich, in the absence of government price or trade policy, prices indomestic currency are not the same as for their closest tradablesubstitutes. In practice, of course, most locally produced goods andservices are not perfect substitutes for tradable products so that thedistinction between tradables and nontradables becomes one of degree.

277

In Ghana's case, the major source of distortions was expansicnary

monetary and fiscal policy, accompanied by exchange and trade controls to

avoid balance of payments deficits. As discussed in the text, these

policies were unimportant until 1961, when declining reserves forced the

government to clamp dcwn on imports. Thereafter Ghana experienced

variations in the pace of expansionary pressutres and in the intensity of

controls, btLt the overall thru1st of these policies continued in the same

direction.

The base period chosen for application of the FFP approach is

1957-59, before these trade policies were imiplemented on a substantial

scale. The nominal exchange rate (NER) for the period from 19598 to 1985 was

adiLtsted for inflation in Ghana Lusing the nontradable consumer price index

(CPI) ,-4 and in its trading partners using two separate indexes. The first

was an index of NEs and wholesale prices in the United States and the

United f .ingdom, weighted by each country's share of the total value of

Ghana' s imports and exports in 1972. Since these two trading partners

account for only about one third of Ghana's total trade, however, a second

index employed was the Manufacturing Unit Value (MUV) index of exports by

induistrial market economies to developing countries.. 3

The resulting nominal and real exchange rates are presented in

Table 2-1 for both the official and black market rates. Comparing the two

nominal exchange rates suggests that overvaluation of the cedi became a

2 From Table 3-5(1). Annex 3. The nontradable CFPI in this tableincludes some tradable agricultural products. An alternative index was alsoused, therefore, which is restricted to nonagricultural nontradables. Sincethe prices of all locally produced food products increased quite rapidlyover the period studied, the two indices vary substantially.

: World Bank:, Economic Analysis and Projections Department.

278

Table 2-1

Exchange Rates, 1958-85(NC/$US) (a)

NomainalExchange Rate Real Exchange Rate (d) Real Exchange Rate (e) Price Indices (1958=100)

Black 6hana Non AgOfficial Market Official Black Market Official Black Market Non Trad Non Trad US/UK

Year (bl (c) (1) (2) (1) (2) (1) (2) (1) (2) CPI (f) CPI (9) WPI MUV (h)

1958 .71 .71 .71 .71 .71 .71 .71 .71 .71 .71 100 100 100 1001959 .71 .71 .72 .70 .72 .70 .72 .70 .72 .70 100 100 100 991960 .71 .71 .70 .70 .70 .70 .70 .70 .70 .70 103 103 101 1011961 .71 .71 .87 .70 .87 .70 .87 .70 .87 .70 104 104 127 1021962 .71 .71 .68 .68 .68 .68 .68 .68 .68 .68 110 110 104 1041963 .71 .56 .62 .61 .49 .48 .58 .57 .45 .44 120 129 105 1021964 .71 .58 .58 .56 .47 .46 .55 .54 .45 .44 132 138 107 1051965 .71 .65 .46 .44 .43 .41 .52 .50 .48 .46 170 150 110 1051966 .71 2.13 .42 .40 1.24 1.20 .47 .45 1.41 1.35 194 171 113 1091967 .84 1.67 .55 .53 1.09 1.06 .56 .55 1.11 1.08 173 170 113 1101068 1.02 1.75 .58 .60 1.01 1.03 .61 .62 1.05 1.07 186 179 107 1091969 1.02 1.75 .57 .59 .97 1.01 .60 .63 1.04 1.07 200 188 111 1151970 1.02 1.64 .58 .60 .92 .97 .61 .63 .97 1.02 208 197 117 1221971 1.03 1.75 .57 .58 .97 .99 .65 .66 1.11 1.12 229 202 127 1291972 1.15 1.64 .62 .64 .88 .91 .69 .72 .99 1.02 253 225 136 1401973 1.15 1.49 .56 .63 .73 .81 .65 .72 .84 .94 299 259 146 1631974 1.15 1.73 .56 .65 .85 .97 .62 .72 .93 1.07 352 319 173 1981975 1.15 1.99 .50 .56 .86 .96 .57 .64 .99 1.11 455 396 197 2201976 1.15 2.91 .31 .36 .78 .90 .43 .50 1.09 1.27 721 513 193 2241977 1.15 9.20 .14 .17 1.15 1.38 .32 .39 2.60 3.11 1634 726 205 2461978 1.51 8.96 .14 .16 .82 .93 .32 .36 1.90 2.15 2728 1179 250 2B21979 2.75 15.56 .19 .21 1.06 1.16 .43 .48 2.46 2.70 4279 1848 292 3201980 2.75 15.97 .15 .15 .89 .86 .37 .36 2.14 2.06 6491 2704 364 35i1981 2.75 26.25 .07 .07 .70 .68 .19 .18 1.79 1.74 13584 5307 361 3531982 2.75 61.67 .06 .06 1.24 1.25 .15 .15 3.36 3.38 17207 6349 346 3481983 3.45 76.58 .03 .03 .63 .65 .09 .09 2.03 2.10 39795 12339 327 3391984 35.34 135.00 .22 .23 .83 .88 .50 .53 1.92 2.02 51153 22246 316 3331985 54.05 160.00 .35 .34 1.03 1.00 .61 .60 1.82 1.77 53547 30224 343 335

27 9

Notes to Table 2-1:(a) Source is IMF, International Financial Statistics, except where otherwise noted.(b) The average official exchange rates in 1972 and 1973 were actually NC 1.32/US$ and NC

respectively, rather than the NC 1.151US5 indicated here, because of the dollar's devis-a-vis gold on February 13, 1973. During this period, however, the gold contentCedi was held constant.

(c) Source is Pick's Currency Yearbook, various issues, for 1958-83 and World Bank,6hana: Irrigation Sector Review, for 1984; Economist, 1985 Summary, for 1985.

(d) Nominal exchange rate divided by the 6hana nontradable CPI (1958=100) and multiplied b(1) a weighted average of the U.S. WPI index and the U.K. #PI index multiplied times

rate, with weights equal to the shares of the U.S. and the U.K. in total trade i(2) the Manufacturing Unit Value (MUYV index of exports by industrial market economi

developing countries.le) Nominal exchange rate divided by the 6hana nonagricultural, nontradable CPI (1958=100)

(1) a weighted average of the U.S. WPI index and the U.K. WP index multiplied tisesrate, with weights equal to the shares of the U.S. and the U.K. in total trade i

(2) the Manufacturing Unit Value (MUV) index of exports by industrial market economideveloping countries.

(f) CPI for nontradables (1958=100) from Annex 3, Table 3-5(1).(g) CPI for nonagricultural nontradables (1958=100) from Annex 3, Table 3-5(1)3.(h) Source is World Bank, Economic Analysis and Projection Departeent.

280

problem in Ghana for the first time at the end of the Nkrtuitah regime, when

exchange controls were for the first time strongly enforced. The black

market rate dropped somewhat after this, as the import program was expanded

and the official rate was increased. Nevertheless, even in 1973, when the

difference between the two rates was at a minimum, the black market rate

was almost ta) percent higher than the official rate. Afterwards the two

rates began to diverge sharply, despite repeated adjustments in the official

rate, as the black marhet rate rose very rapidly in the face of inflationary

pressures and quantitative trade and exchange controls. The distortion

finally reached a peak in 1982, when the black market rate was 22 times the

official rate. Thereafter, as the Economic Recovery Program was placed in

effect, several large devalutatlicris were announced and the pace of increase

in the black market rate slowed. Ey 1985 it was less than three times the

official exchange rate.

The importance of inflationary pressures in Ghana causing this

divergence in exchange rates appears to have been very strong. No matter

which index is used, deflating the exchange rate by relative price changes

in Ghana and its trading partners results in a drastic decline in the real

official exchange rate, at least up until 19B4. Cln the other hand, despite

some short-term fluctuations,the long-term trend in the real black market

rate is upwards, especially when deflated by the nonagricultural nontradable

CPI. This is consistent with the fact that the black market rate reflects

the rising scarcity value of foreign exchange, and perhaps also the risk

premium, in the face of increasing restrictions on imports, whereas the

increase in the price of nontradables relative to the price of tradables is

more of an indication of changes in the free trade eqLlilibriLun exchange

281

rate. It also suggests that the theory of purchasing power parity, coupled

with the maintenance of a relatively inflexible official exchange rate

through the use of direct trade and exchange controls, explains a great deal

of the distortions introduced into the Ghanaian economy.

The EquLilibrium ExchanQe Rate

The free trade equilibrium exchange rate is the exchange rate that

woxuld cause the demand for and sutpply of foreign exchange to be equal in the

absence of policy distortions such as trade taxes or subsidies, price

controls, and quantitative restrictions on imports and foreign exchange.

This does not necessarily imply that the current account would be balanced

since the foreign exchange market can include net flows of capital, debt

service, and unrequited transfers. For the exchange rate to be one of

equilibrium, however, these flows must be autonomous rather than

accorimodating and mLtst be sustainable and consistent with the country's

overall capital structLire.

To the extent that there are fundamental changes in capital flows,

debt service, and transfers, there must be corresponding adjustments in the

MR. The same is trtLe if there are structural changes in the domestic

economy, such as those associated with rising per capita income, or in the

terms of trade, since these can be expected to alter the exchange rate

necessary to achieve equLilibriLm in the absence of government price or trade

policy.

Over the period 1957-1959, as seen in Table 2-2, the average

current acconlt deficit was nil. In addition, there were virtually no

changes in foreign exchange reserves. This suiggests that the balance of

282

Table 2-2

Balance of Payments Current Account (a)

(millions 5US)

Current RealAccount Price of

Imports Other Total Exports Other Total Credit Cocoa Ce)FOB Debit (b) Debit (c) FOB Credit Credit Balance (d) ($/kg)

Year (1) (2) (3) (4) (5) (6) (7) (8)

1957 250 41 290 261 N/A N/A -29 2.391958 219 40 259 300 N/A N/A 41 3.081959 299 37 336 316 N/A N/A -19 2.701960 348 81 429 334 N/A N/A -95 2.131961 388 80 468 333 N/A N/A -135 1.741962 310 73 383 320 N/A N/A -63 1.691963 337 78 415 307 N/A N/A -108 2.011964 322 75 396 321 N/A N/A -75 1.791965 439 94 533 321 N/A N/A -212 1.301966 321 77 398 280 N/A N/A -118 1.741967 265 119 384 284 31 315 -69 1.961968 265 120 385 304 37 342 -43 2.541969 295 146 441 345 47 392 -49 3.151970 375 167 543 427 50 477 -66 2.141971 368 167 535 335 55 389 -146 1.561972 223 125 348 384 59 443 95 1.721973 372 168 540 585 69 654 114 2.511974 798 234 1,032 679 68 747 -286 2.781975 651 272 922 801 94 895 -27 1.941976 690 305 995 779 115 894 -101 3.131977 860 298 1,159 890) 131 1,020 -138 5.381978 780 321 1,102 893 105 997 -104 4.101979 803 318 1,121 1,066 99 1,165 43 3.581980 908 355 1,264 1,104 110 1,213 -50 2.601981 954 381 1,336 711 121 832 -504 2.071982 589 316 905 607 107 714 -191 1.761983 500 224 724 439 39 478 -247 2.201984 533 280 813 566 46 612 -201 2.531985 669 284 952 632 44 676 -276 2.36

Notes to Table 2-2:(a) Source is IMF, International Financial Statistics, except where otherwise noted.(b) 1957-66 Other Debit figures are actually net of Other Debit and Other Credit for that period.(c) Column (1) plus column (2).(d) For 1957 to 1966 credit balance is column (4) minus column (3).

For 1967 to 1985 credit balance is column (6) sinus column (3).(e) International Cocoa Organization average daily price in New York and London deflated by

Manufacturing Unit Value Index, (1980 constant $), froa World Bank, Commodity Trade andPrice Trends, 1985, Washington: 1985, p.44, and World Bank, Price Prospects for MajorPrimary Commodities, Vol. I: Summary and Implications, pp.31, 34.

283

payments was in eqLlilibriLVn at that time and that the NER was a reasonable

approximation for the EER except for the possible influence of policy

distortions.

Ghana's exports in 1957-1959 were subject to taxes equal to 25

percent of the average FOB value of exports. 4 Almost all these taxes were

on cocoa, the average tax rate on other exports being only 2 percent. The

tax on cocoa was not very bLurdensome, however, since world market prices at

the time were relatively highs especially in relation to those that

characterized the 1960s.6 In any case, prodLtcer prices at the time clearly

covered variable costs, and investment in cocoa trees that were producing in

1957-1959 had been made six or more years earlier when producer prices were

quite high. It is reasonable to argues therefore! that the export tax was

operating to stabilize producer prices aro.rnd a lang-run average world

market price for cocoa and was not adversely affecting production.

On the import side, the average tariff rate in 1957-1999 was 17

percent. This rate remained reasonably constant over the last half of the

i93:s and resulted in a lower level of imports than wculd have existed

withoLut the tariffs. The EER required to achieve balance of payments

' J. Clark Leith, Foreign Trade Recimes and Economic Development:Ghana, New York: Columbia University Press, 1974, p. 13.

As shown in Table 2-2w the world market price of cocoa in 1957-58averaged $2.72/kg in real terms (19W) prices); in 1960-69 it averaged$2.kY)/kg.

284

equilibrium in the long run, if these import tariffs were to have been

removed * can be estimated from'

E* ~~~~~~Eao + 4oQ

where E* is the equilibrium exchange rate

E is the nominal exchange r-ate;

t.. is the average rate of tariff, or tariff equivalents on

imports;

2', is Ghana's price elasticity of denand for foreign exchange;

C, is Ghana's price elasticity of sutpply of foreign exchange.

The demand elasticity ?, is assumed to equal 2.0:). The sLupply elasticity

£, is assumed to equal C.6, the long-run elasticity estimated for cocoa

exclusive of the effect of prices on new plantings, as described in Annex 4.

The resulting EER is .80 N2!/$US, compared with the official rate in 1957-59

of .71 Ne/$US. 7

, Derived from equation (A.1) in the Appendix assuming that there isno existing current account deficit and that the export tax does not haveany long run effect on the allocation of resources. Equation (1) doesassume, however, that exports are influ rnced in the long run by a change inthe exchange rate as long as this change is passed on to producers.

' These relatively crude estimates of the elasticities of demand forand supply of foreign exchange are used here for comparison sake only. Themore sophisticated'estimate of the equilibrium exchange rate, describedlater in this annex, takes into account new plantings and noncocoa exports,and is based on econometrically estimated demand and supply functions. TheEER for 1958 estimated with the more sophisticated model is .8B, comparedwith this cruder approximation of .90.

285

During the years since 1957-1999, the cLtrrent account balance has

fluctLuated greatly from year to year. In the early 1960s deficits were

financed by a decline in reserves and by increasing Ghana s foreign debt.

Neither of these practices was sustainable. Some foreign aid was received

to offset deficits during the late 196C0s and early 1970s, but this was

largely short-term program support aimed at helping Ghana to achieve

economic stabilization. In 1972 and 1973, Ghana enjoyed large surplUses as

a result of booming cocoa prices, but this period was short-lived as well.

Deficits were again the norm during the late 1970s and early 1980s, with

frequent sutrpluses cn trade account being more than offset by deficits

elsewhere. Again the situation was believed by both Ghanaians and its

creditors to be unsustainable, resulting in the launching of the Economic

Recovery Program in late 1962. This program called for significant foreign

aid inflows in return for policy reform, but disbursements were delayed and

would have had only a minor impact on the equilibrium exchange rate at the

end of the period under consideration. It is r-easonable to assume,

therefore, that the period from 1957-1959 to 1985 was one in which balance

of payments equilibrium was defined with the current account in balance.

The equilibrium exchange rate is also influenced by struLctural

changes in the terms of trade or in the domestic economy that are not the

result of policy induced distortions. In Ghana s case, the most obvious

example would be a long-term movement in the price of cocoa.. lthoLigh cocoa

is notoriously subject to long cyclical price changes, it is very difficult,

from the historical data shown in Table 2-2w to discern any significant

trend in its world price deflated by the Manufacturing Unit Value index of

the industrial countries.

286

Assiu1ing, for these reasons, that the EER was constant from

1957-1959 to 1985, its value of .9@3 NV/$US may be compared with the

estimates of the RER, both official and black market, shown in Table 2-1.

Alternatively, the EER can be adjusted by the same indexes used to adjust

the NER in Table 2-i to obtain a series of nominal equilibrium exchange

rates (NEER), estimated using the Fff approach, which can be compared

directly with the nominal official and black market rates, as shown in Table

2-3. It is clear from this table that the equilibrium rate over the longer

rn- increased less rapidly than the black market rate but at a mLuch greater

pace than the official rate.

Elasticities Approach

The purchasing power parity approach to estimating the equilibrium

exchange rate concentrates on the relationship between the prices of

tradable and nontradable goods and services as these change over time in

response to inflationary pressures. The elasticities approach, on the other

hand, focuses on the tradables sector. It estimates the change in the

exchange rate that wouLtd be necessary to equate the demand for and the

stpply of foreign exchange if unsLsstainable inbalances in the current

acccLunt and all distortions in the prices of tradable goods were to be

removed.

This definition of the equilibrium exchange rate implies that

equilibrium is maintained in the face of changing demand and supply

conditions by movements in the exchange rate alone and not by the government

allowing foreign exchange reserves to fluctuate or using commercial policy

to stabilize the foreign exchange market. This is especially important for

287

Table 2-3

Nominal Exchange Rates and Nosinal Purchasing Power ParityEquilibrium Rates

(NC/MUS)

Price Indices (1958100)

Nominal Exchange Rate Equilibrium lb) Equilibrium (c) Ghana--------------------- --------------- -------------- Ghana Non Ag

Black Non Trad Non Trad US/UKYear Official Market (1) (2) (1) (2) CPI (d) CPI (e) IPI HUV (f)1_56 .71 _7- -8- .80 -80 .8- 100 100 100 100

1958 .71 .71 .80 .80 .80 .80 100 100 100 1001959 .71 .71 .BO .81 .80 .81 100 100 100 991960 .71 .71 .81 .82 .81 .82 103 103 101 1011961 .71 .71 .66 .81 .66 .81 104 104 127 1021962 .71 .71 .84 .85 .84 .85 110 110 104 1041963 .71 .56 .92 .94 .99 1.01 120 129 105 1021964 .71 .58 .99 1.02 1.03 1.06 132 138 107 1051965 .71 .65 1.24 1.29 1.10 1.15 170 150 110 1051966 .71 2.13 1.37 1.43 1.21 1.26 194 171 113 1091967 .84 1.67 1.23 1.26 1.20 1.24 173 170 113 1101068 1.02 1.75 1.40 1.37 1.34 1.31 186 179 107 1091969 1.02 1.75 1.45 1.39 1.36 1.31 200 188 111 1151970 1.02 1.64 1.42 1.36 1.35 1.29 208 197 117 1221971 1.03 1.75 1.45 1.42 1.27 1.25 229 202 127 1291972 1.15 1.64 1.50 1.45 1.33 1.29 253 225 136 1401973 1.15 1.49 1.64 1.47 1.42 1.29 299 259 146 1631974 1.15 1.73 1.64 1.42 1.48 1.29 3.52 319 173 1981975 1.15 1.99 1.85 1.66 1.61 1.44 455 396 197 2201976 1.15 2.91 3.00 2.59 2.14 1.84 721 513 193 2241977 1.15 9.20 6.40 5.34 2.84 2.37 1634 726 205 2461978 1.51 8.96 8.75 7.76 3.78 3.35 2728 1179 250 2821979 2.75 15.56 11.78 10.74 5.09 4.64 4279 1848 292 3201980 2.75 15.87 14.30 14.85 5.96 6.19 6491 2704 364 3511981 2.75 26.25 30.17 30.93 11.79 12.08 13584 5307 361 3531992 2.75 61.67 39.94 39.74 14.74 14.66 17207 6349 346 3481983 3.45 76.58 97.76 94.38 30.31 29.26 39795 12339 327 3391984 35.34 135.00 130.16 123.49 56.61 53.70 51153 22246 316 3331985 54.05 160.00 125.19 128.32 70.66 72.43 53547 30224 343 335

Notes to Table 2-3i(a) From Table 2-1.(b) Equilibria exchange rate in 1958 of .803 multiplied by

(1) the ratio of the nontradable CPI for 6hana to the index ofwholesale prices for the US and UK derived in Table 2-1.

(2) the ratio of the nontradable CPI for 6hana to the MUV indexof industrial countries shown in Table 2-1.

(c) Equilibrium exchange rate in 1958 of .803 multiplied by(1) the ratio of the nonagricultural,nontradable CPI for Ghana to the index of

wholesale prices for the US and UK derived in Table 2-1.(2) the ratio of the nonagricultural, nontradable CPI for Ghana to the MUY index

of industrial countries shown in Table 2-1.(d) 6hana Non-Tradable CPI, from Table 3-5(1), Annex 3.(e) Ghana Non-Agricultural, Non-Tradable CPI, from Table 3-S(1), Annex 3.(f) Manufacturing Unit Value Index, (1958 constant S), from World Dank, Commodity Trade and

Price Trends, 1985, Washington: 1985, p.44, and World Dank, Price Prospects for MajorPrimary Commodities, Vol. I: Summary and lplications, pp.31, 34.

288

a country such as Ghana that is highly dependent upon an exported conmodity

subject to large variations in domestic supply and world market prices. The

Cocoa Marketing Eoard was established$ in fact, principally to stabilize

prices and incomes to producers so as to avoid resources being continually

reallocated on the basis of short-term market considerations. Similarly,

the Ghanaian government has over the years attempted to use its foreign

exchange reserves as a buffer against wildly fluctuating export revenues.

The problem, of course, is that it is very difficult to

distinguish between current accoLnt deficits/surpluses and conmercial

policies, such as the effective export tax on cocoa, that are maintained for

stabilization purposes and those that are not. The Cocoa Marketing Board's

stabilization function, for example, has for many years been overshadowed

by its role in raising government revenue. It is always possible, of

cutrse, to establish a range within which the current account balance or the

prices of tradable gcods might fluctuate withcLtt influencing the equilibrium

exchange rate, but this woltld be fairly arbitrary and the basis for

establishing the range would be disputable. It seems preferable, therefore,

to estimate the equilibriLm exchange rate by correcting the actual exchange

rate for all unisustainable current account imbalances and distortions in the

prices of tradable goods, but to avoid drawing policy concluLsions from year

to year changes in the equilibrium rate.

The foraLila for estimating the equilibrium exchange rate using the

elasticity approach, where these elasticities are defined with respect to

the distorted economy, is given by'

S See the Appendix to this annex for the derivation of this equation.

289

(M-X) + XE. +gE* )E* )

Xi, Mfl

1-t. l+tl.

where M is the demand for foreign exchange on current account;

X is the supply of foreign exchange on current accoLnt;

t. is the tax rate on exports.

The demand for and supply of foreign exchange M and X, are "total debits,"

and "total credits", respectively, of the current account from Table 2-2.

The elasticities E, and 7. retain the same values assLumed earlier.

The tax rate on imports, t,- is available from Leith for the

period from 1957 to 1971.9 With the imposition of import controls in 1961,

however, domestic prices of importables rose above the levels of the tax

adjuisted border prices becautse of the qutota premium created by import

scarcity. Direct com,parisons of domestic and border prices for manufactured

importables are available for 1972 and 1985 only.3 0 They sLugest that the

average nominal protection coefficient (domestic price/border price) for

agriculture and industry in 1972 was 1.60; in 1985, the comparable figure

9 Leith, Foreign Trade Reaimes ... , p. 1 1 .

° Scott R. Pearson, Gerald C. Nelson.! and J.Dirck Stryker,"Incentives and Comparative Advantage in Ghanaian Industry and Agriculture,"1979, and P-E International Operations Ltd in Association with ThomasW. Allen & Associates Ltd, Trade Liberalization and Incentives for IndLustrin Ghana, January 1966, Volume II, p. 109.

290

for a weighted average of industrial prodLucts was 1.53.1L In each case,

h-owever, these coefficients are based on a sample of price comparisons that

may not represent very well the entire range of traded goods.

In order to estimate the tax equivalent rate when import and

exchange controls were binding, therefore, the tariff rate of 17 percent for

1960 was extrapolated over time by mu(ltiplying by the ratio of two indexes,

each set to base year 1972: (1) a weighted average of Ghana' s Consumer Price

Index for tradable categories of goods (see Table 3-5(1)) in Annex 3) and

(2) the Manufacturing Unit Value (MUV) index of exports by industrial market

economies to developing coiuntries multiplied times the official exchange

rate. The tradable CPI for Ghana inclLtdes nontradable fcods, such as

cassava and yamns, but the prices of these have moved very closely with those

of tradable foods. The CPI is only available by category back to 1963, so

the overall index was used for 1956-62. The Manufacturing Uhit Value (MIJ)

index was used rather than the weighted average of wholesale price indexes

because, as noted above, the two give very similar results and the MUV index

covers a br-oader spectrum of Ghana's trading partners.

Leith' s import tax rates were used for the years 19586-6). After

this the tariff equivalent rates calculated as described above were Lised

since they r-ose to more than double the actual tax rates in 1965 and 1966,

after which the difference between the two narrowed considerably under the

'1 The figure for 1972 of 1.60 is the mean of the unweighted averageof 1.37 for agriculture and the weighted average of 1.51 for industry, withthe weights being the output of firms in the industrial sample, multipliedtimes the ratio of the exchange rate used to calculate this coefficient(NV 1 .2E/*US) to the 1972 exchange rate used here (N 1. 15/SUS). Theuinweighted average reflects quite well the relative proportions of thesesectors in total consumer goods imports in the early 1970s. Daring 1970-72,food products constituted 48 percent of total consumer goods imports. WorldBank, Ghana: Towards Structural Adjustment, October 7, 1985, Vol II, p.22.

291

new govemaent's liberalization policies. Beginning about 19764 however,

increasing overvaluation of the cedi resulted in a strong upward movement in

the tariff equivalent rate that was checked only by devaluaticn in 1984 and

1985. The. estimated value of the tariff equivalent rate in 1985 is 216

percent. Given the enormous distortions that have occurred over the past 25

years, it is encouraging that this is not too far from the observed tariff

equivalent rate in 1985 of 53 percent, which is not the result of a random

sample and may therefore be considerably in error.

The average tax rate on cocoa exports was calculated from Amnex 3

on the basis of producer prices, marketing board costs, and FOB prices.

Other exports were subject to a variety of taxes and subsidies, none of

which was ever greater than 2) percent. An assumed average tax rate of

zero on these exports is probably not too ffLch in error. The average tax

rate on total exports was estimated as a weighted average of the rates on

cocoa and other exports. This average rate was positive in all years but

two and normally fluctuation between 10 and 34 percent despite the

decreasing real value of the cedi.

Table 2-4 shows the estimation of the equilibrium exchange rate

using the elasticity approach. The results suggest a substantial

overvaluation of the exchange rate, especially after the mid-1970s, but the

degree of overvaluation is much less than that indicated using the FFP

approach. One reason for this is that equation (2) is valid only for small

movements away from the distorted sitLuation. DisequilibriLm in Ghana,

however, has been enormous and clearly violates this assumption.

One way of handling this problem is to estimate the equilibrium

exchange rate at the mid-point between the distorted and Lndistorted

292

Table 2-4

Estimation of the Equilibriue Exchange RateUsing the Elasticities Approach

Current Account Import Tax Export Tax Nominal Equilibrium Exchange Rate----------------------- Tradable "UV Equivalent Rate on Total Exports of Official ----------------------------Debit (a) Credit (a) CPI (b) Index(c) Rate (d) Cocoa (e) Exports (f) Cocoa (f) Export Tax Exchange Rate Distorted(i) Nid-pointli)

ueill tUS) (mill lSS) 11972=100) (1972 1001 3() (5) (mill NC) (sill NC) Rate (g) (NC/SUS) (hi (NC/$US) (NC/3USIYear (1) (2) (3) (4) (5I (6) (7) (8) (91 (10) (11) (12)

1958 259 300 39 44 le 50 209 125 30 .71 .67 .701959 336 316 39 44 15 41 226 138 25 .71 .75 .761960 429 334 40 45 17 24 232 133 14 .71 .85 .851961 408 333 40 45 19 19 230 139 11 .71 .89 .921962 383 320 43 46 23 230 134 13 .71 .86 .881903 415 307 46 45 35 31 218 136 19 .71 .94 .991964 396 321 51 46 47 18 229 136 11 .71 1.00 1.051965 533 321 67 47 8a 12 227 137 7 .71 1.35 1.451966 398 280 76 48 109 32 191 103 17 .71 1.31 1.441967 384 315 68 57 58 47 245 131 25 .84 1.17 1.261968 385 342 74 69 42 56 339 186 31 1.02 1.24 1.341969 441 392 80 73 45 59 333 221 39 1.02 1.19 1.321970 543 477 63 77 43 42 467 300 27 1.02 1.27 1.361971 535 389 91 82 48 45 496 204 19 1.03 1.48 1.561972 348 443 100 100 32 38 565 291 20 1.15 1.16 1.281973 540 654 120 116 36 SO 730 397 27 1.15 1.17 1.311974 1032 747 140 141 32 52 840 466 29 1.15 1.46 1.541975 922 895 184 157 55 36 928 551 21 1.15 1.48 1.601976 995 894 299 159 149 59 951 516 32 1.15 1.90 2.331977 1159 1020 686 175 419 39 1166 680 23 1.15 2.85 4.421978 1102 997 1206 265 502 60 1581 988 37 1.51 3.32 6.331979 1121 1165 1856 545 350 33 2737 1846 22 2.75 5.70 8.821980 1264 1213 2802 598 520 -20 3458 1942 -11 2.75 8.56 12.581981 1336 832 6199 601 1264 -244 2924 1091 -91 2.75 22.11 31.591982 905 714 7616 593 1599 25 2402 1053 11 2.75 11.37 31.901983 818 507 17406 724 3080 22 6999 3989 13 3.45 19.45 79.131984 N/A N/A 23085 7282 319 11 N/A N/A N/A 35.34 N/A N/A1985 N/A N/A 24117 NIA N/A 7 N/A N/A N/A 54.05 N/A N/A

Notes to Table 2-4:(a) Froe Table 2-2; credit figures for 1958-1966 are export earnings, FOB.(b) From Table 3-5(11.(c) Nanufacturing Unit Value Index multiplied by the Nominal Official Exchange Rate, from Table 2-1,

and set at base year 1972 100.(dl Source for 1958-1961 is J. Clark Leith, Foreign Trade Regises and Economic Development: Ghana, New York:

Columbia University Press, 1974, p1i. For other years, the rate for 1972 was adjusted by the ratio of coluen(3) to column (4). The rate for 1972 was taken froe Scott R. Pearson, Gerald C. Nelson, and J. Dirck Stryker,'Incentives and Comparative Advantage in Ghananian Industry and Agriculture', 1979.

(e) Calculated as I - (Rendered Port Price of Cocoa/FOB Price of Cocoa) times 100. The Rendered PortPrice of Cocoa is from Table 3-3(2), the FOB Price of Cocoa is from Table 3-4(3).

(f) International Monetary Fund, International Financial Statistics Yearbook 1985, pp.310-311.(g) A weighted average of the export tax rate on cocoa in column (6) and of zero for other exports, where

the weights are the value of cocoa and of other exports calculated from columns (7) and (8).lh) Fros Table 2-1.(xl Calculated from equation (9) using the following elasticities: demand 2.0, supply 0.6.(j) Calculated as described in the Annex 2.

293

situation assuming a constant arc elasticity. Solving for E*, in this

case, involves a very complex qLtadratic equation that requires the use of an

iterative process. The details of this calculation are explained in the

Appendix. The results, presented in Table 2-4, suggest that the equilibrium

exchange rate was sutbstantially higher towards the end of the period than

indicated by the EER calculated using the distorted situation as the base

period. The mid-point estimate is generally stbstantially less than the

black market r-ate.

EqUilibrium Exchange Rate MDdel

The elasticity approach suffers from the weakness of its empirical

base and from the fact that it does not capture the complexity of the demand

for and suLpply of foreign exchange in Ghana. Consequently, an alternative

approach used was to estimate these functions econometrically from

time-series data and to solve the resulting system of equations for the

exchange rate that wcould have prevailed if no distortions had existed and

the current account had been in balance each year.

Import Demand FLnction

The demand for imports in Ghana is related to the domestic price

of tradables relative to that of nontradablees and to real GDF. Since

imports have been highly controlled in Ghana, lowever, it has been the

domestic price of tradables relative to that of nontradables, rather than

the level of imports, that has varied in response to changing import

demand. In addition, the domestic production of foodstuffs, as influenced

294

by weather and other factors, has had an important influence an the demand

for imports.

The following import demand equation was estimated using annual

data from 1963 to 19873.

Pt M Yln 4.=23 - .6C4 In + 2.321 In - 2. 372 ln ... (3)

PF,, (1.845) (.211) P. (.26:)) PF^ ( 503)

R = .84 D.W. = 1.495

where P, is the domestic price of tradables (Table 3-5(1))

Pot is the domestic price of nontradables (Table 3-5(1))

M is the dollar value of cUrrent accoitnt debits (Table 2-2)

P. is the Manufacturing LUnit Value index (Table 2-1)

Y is GDP (Economic Pialysis and F'rojections Department, World Bank)

Q0, is -an index of food producticon (Table 7)

The coefficienits of this equation are all highly significant. The implicit

elasticity of demand for imports is 1.85, which is quite close to the

elasticity cf 2 assLuted earlier.

Exoort SLtpply Functions

Two export supply functions were estimated. That for cocoa is

presented and discussed in Annex 4. Independent variables in the equation

include the current official producer price of cocoa, the previouLs year's

market price for maize, the quantity produced of cocoa the previoLus year,

and the "normal" level of production given the existing stock of cocoa

trees. The last variable is estimated using a "vintage-matrix" model

described in the annex, which takes into acccLnt the number of cocoa trees

295

of different ages, their yields over time, and the rate of new planting,

which is partly a fLnction of price.

The second sutpply fLinction is for timber, gold and other

exports. Its estimation required the construction of an export quantity

index and a domestic price index for exports. The former uses world prices

in 1972 as weights, and the latter weights domestic prices by the 1972

export quantities. The domestic pr-ice index was derived by adjLsting timber

and other export prices for the tax. or sutbsidy on exports and by

incorporating the published domestic price of gold. It is deflated by the

price index for nontradables, as is the other independent variable! GDP.

The estimated equation is

PQ GDPIn XQ. = 7.716 + ..3-5) ln . .- .94C) ln

(.742) (.137) P^_ (.229) P., ... (4)

R = .466 D.W. = 1.73.9

where X. is the quantity index for noncocoa exports, and P. is the index of

domestic prices of these exports.

Adjustment for Monopoly Power

At the beginning of the period Lnder consideration, Ghana

accounted for abouLt 40 percent of the world's cocoa trade and thus was able

to influence world prices through its export tax policy. To maximize

national inccae, therefore! Ghana shotuld have applied an optimal export tax

on cocoa., and the model of the equilibriLum exchange rate should incorporate

this tax.

296

If it can reasonably be ass-mied that Ghana exports all of its

cocoak* so that domestic consLunption is Lunaffected by the export tax, Ghana

would optimize its position by setting marginal revenue on the world market

equal to marginal cost at home. Marginal revenue (OvfR) in turn! is given by

dTR X<, dF.MR = ~ = Pw + ... (5)

dX, dX,

where TR is total revenue, X. is Ghana's exports of cocoa, and P. is the

world price of cocoa. Assuming perfect competition among cocoa farmers,

marginal cost is equal to the domestic price (Pd) adjusted for the export

tax rate (t).,

Pd = P. (1-t) ... (6)

Dividing by P., yields

d F'/P,(1-t) = 1 + . .. (7)

(dXi - dX=)

X,s

where X. is total world exports and X. is exports of competing coLuntries.

Manipulation of equation (7) yields

(1-t) = 1 +

I ( 1-5) . .. (B)

s s

where e.. is the world's price elasticity of demand for cocoa, and E( is the

conpetitorss elasticity of supply.

It is assLuned that Ghana optimally chooses a tax rate that takes

into account competitors' long run supply responses over a period of about

10 years. These have been estimated for Brazil, the Ivory Coast, and

297

Malaysia at 0.5V6 1.15, and 3.0 respectively."' The Malaysian estimate is

high, however, due to low levels of production. Using the roughly equal

shares of the Ivory Coast and Brazil in the world cocoa mar.-et during the

early 1970s, the average long run sutpply elasticity of competitors may be

estimated at about 0.9. With the world's demand elasticity estimated at

approximately -:).3' and Ghana's share of the market at the beginning of the

period being about *:) percent, the optimal export tax derived u-sing equation

(8) would have been 47.6 percent. The optimal tax varies, however, as

Ghana's share in the world mark.:et changes.

The Exchanae Rate Model

The import demand function, cocoa sutpply fLunction, and noncocoa

export supply function were each incorporated into a simulation model that

calculates the exchange rate that equates the demand for and supply of

foreign exchange assLuming that the only trade distortion is the optimal

ex.port tax, which is calculated from equation (8) with Ghana's share of the

world market allowed to vary with the quantity of cocoa it exports.

Otherwise domestic prices equal world prices times the equilibrium exchange

rate.

Before looking at the results of this analysis it is useful to

recall scme of its limitations. First, the consequences of Ghana's actions

on the world market are modelled in a very simple way. A more comprehensive

X T. Akiyama and A. Etwers, SUDVplv Response of Cocoa in MajorFroducing Countries, Division Working Paper No. 1964-3. Ccmmcdity Studiesand Projections Division, World Bank, April 1984, p.25.

t Takamasa Akiyama and Ronald C. Duncan, kialvsis of the World CocoaMarket, World Bank Staff Commodity Working Faper No.8, 1982, p.13.

298

analysis might link Ghanaian policy with the World Bank's cocoa model.

Second! only the effects of changes in relative prices are taken into

accoLtnt. Maintenance of equilibriLn would also have had enormots

conseqLIences for GDP, and possibly also the production of food.

Despite the limitations, the results are interesting, especially

in comparison with the other exchange rates shown in Table 2-5. These

inclutde the official exchange rate, the black market rate, the equilibrium

rate calcLilated utsing the FPFf approach (with two different definitions of

the nontradable CPI), and the equilibrium exchange calculated using arc

elasticities. First, there is a fairly close correspondence in most years

between the equilibrium rates calculated from the model and uLsing the

elasticity approach. Second, both of these tended to be greater than the

black: market rate prior to 1966, but less than that rate thereafter. This

reflects the increasing importance of exchange controls during the early

196)s. Finally, these two eqLuilibrium rates were generally greater than the

PFP rates until 1972. at which point the diverging relative price trends

became increasingly important, and the FFF rates tended to bracket the other

two rates.

299

Table 2-5

Nominal Exchange Rate and AlternativeNominal Equilibrium Exchange Rates

(NC;USS)

Black Equilibrium (b)Official Market ------------------------------

Year (a) la) (1) (2) (3) (4)

1958 .71 .71 .80 .80 .70 .881959 .71 .71 81 .81 76 .991960 .71 .71 .82 .82 .85 1.151961 .71 .71 .81 .81 .92 1.301962 .71 .71 .85 85 .88 1.171963 .71 .56 .94 1.01 ,,99 1.171964 .71 .58 1.02 1.06 1.05 1.301965 .71 .65 1.29 1.15 1,45 2.051966 .71 2.13 1.43 1.26 1.44 1.581967 .84 1.67 1.26 1.24 1.26 1.241968 1.02 1.75 1.37 1.31 1.34 1.511969 1.02 1.75 1.39 1.31 1.32 1.46197¢0 1.02 1.64 1.36 1.29 1.36 1.571971 1.03 1.75 1.42 1.25 1.56 1.551972 1.15 1.64 1.45 1.29 1.28 1.431973 1.15 1.49 1.47 1.28 1.31 1.351974 1.15 1.73 1.42 1.29 1.54 1.471975 1.15 1.99 1.66 1.44 1.60 1.661976 1.15 2.91 2.59 1.84 2.33 2.061977 1.15 9.20 5.34 2.37 4.42 4.171978 1.51 8.96 7.76 3.35 6.33 6.291979 2.75 15.56 10.74 4.64 8.82 8.311980 2.75 15.87 14.85 6.19 12.58 13.711981 2.75 26.25 30.93 12.08 31.59 36.451982 2.75 61.67 39.74 14.66 31.90 42,411983 3.45 76.58 94.38 29.26 79.13 81.791984 35.34 135.00 123.49 53.70 N/A 107.471985 54.05 160.00 128.32 72.43 N/A N/A

Notes to table 2-5;(a) Official and black market nosinal rates from table 2-3.(bl Purchasing power parity equilibrium rates calculated in the

the following manner:(1) Equilibrium rate in 1958 of .803 tultiplied by the ratio

of the nontradable CPI for Ghana to the hUV index ofindustrial coutnries, from table 2-3.

(21 Equilibrium rate in 1958 of .803 multiplied by the ratioof the Non-Agricultural, Non-Tradable CPI for Ghana to theMUV index of industrial countries, from table 2-3.

(3) Mid-point elasticity equilibrium rate, from table 2-4.(4) Equilibrium exchange rate estimated by the simulation eodel.

300

AFENIX TO gNEX 2: DERIVATION OF A FCF'LJ FOR ESTIMTING THEECUILIERILM EXCF4GE RATE USING WhE ELAsTICIY AFfAcH

The general fornmtla for estimating the equilibrium exchange rate

using the elasticity approach is the followingA 4

(M-X) + X + ME* =( ) E ... (A.1)

Xe Mn

1-t. l+t,

where E* is the equilibrium exchange rate;

E is the actual exchange rate;

M is the actLtal level of imports;

X is the actual level of exports;

t., is the tax equivalent rate on imports;

t. is the tax equivalent rate on exports;

l is the elasticity of demand for imports;

E is the elasticity of stpply of exports.

This foramula is derived by considering the relative rates of

change in import and export prices that would occur if the economy moved

from a distorted situation to free trade equilibriLm.

'4 Maurice Schiff, "Note 1," January 23. 1986, p.16.

301

dPm d (1+tm) E= .... (A.2)

Pm ( l+tm) E

dP, d (1-t.) E

P. (1-tt) E

These relative prices changes can be substituted into an equation showing

the relationships between changes in exports, changes in imports, and the

initial trade balance as the economy moves towards eqLtilibrium.

dX -dM = M - X

d (l-t.) E d (1+tm ) EXF- M171=1M- X ...(A.4)

(1-t,) E (1+t.) E

For small changes, this formrulation can be represented by

E* - (1-t,) E E* - (l+tm) EXs - M=M -X ... (A.5)

(1-tw) E (1+t.) E

which can be solved for E* to obtain equation (A.1).

The only problem with equation (A.1) is that it expresses rates of

change at the quantities and prices that prevail in the distorted situation,

and it may not be very relevant over large changes. One alternative would

be to express all price and quantity changes with respect to the equilibriuM

situation in the absence of trade taxes and quantitative controls. In this

case equation (A.5) becomes

E- (1-t,)E E* - (t+tm)EX*S - -M *? = (M1- X) . (A.6)

ES E3

302

where X* is the equilibriLun level of exports and M* is the equilibrium level

of imports. Since X* and M* are not knawn, expressions for them mLtst be

introduced into eqLlation (A.6). These expressions are derived from

E* - (1- t) EX* - X = X*£.(A.7)

E*

E* - (i+t,) EM* - M = *77 ... (A.8)

E*

Solving for E* we obtain

?1(i+tm) X + E(1-t)ME= E ... (A.9)

(1+1) X - (1-.)M

where q=|q I

Table 2-5 in Annex 2 suqgests that the ratio E*/E is consistently

greater when the equilibrium situation is used as a base rather than the

distorted prices and quantities. A more appropriate procedure, however,

would be to choose values that lie between these two extremes. One

possibility would be to average the equilibrium exchange rates obtained from

equations (A.1) and (A.9). Another woLld be to use mid-point prices and

quantities calculated as linear averages. Equation (A.5) in this instance

becomes

E* - (1-t,) E E* - (1+t.)E(X+X*)f - (M + M*)I= (M - X) ... (A.10)

(1-t.)E + E* (1+tm)E + E*

Expressions for X* and M* must be derived in a manner similar to that used

when the equilibrium period was used as a base. Solving for E*, however,

303

involves a quadratic equation that is very complex. An alternative

procedure is to solve the following equation, derived from equation (A.10),

for E* using an iterative process:

(E*(l+E) + (1-t.)E(1-))X (E*(i+n)+(l+tm)E(1?1)) M= 0 ...(A.11)

E*tl-I)+tl-t.E(l+L) E*(l-I)+(l+t.,,)Et1+1)

By introducing different values of E* into the left hand side of this

equaticn! it is possible to arrive at that value which comes closest to

satisfying the eqLtation.