Interdependence of smallholders’ net market positions in crop and livestock markets: Evidence from Ethiopia

Interdependence of Smallholders’ Net Market Positions in Crop and Livestock Markets:

Evidence from Ethiopia

Moti Jaleta and Berhanu Gebremedhin Improving Productivity and Markets Success (IPMS) of Ethiopian Farmers Project,

International Livestock Research Institute (ILRI), Addis Ababa, Ethiopia..

Presented at the ILRI Scientific Seminar, Addis Ababa,15 December 2010

Outline

Motivation Objective Analytical framework Data Methodology Results Conclusions and implications

Motivation

Integrating smallholders into markets is essential for the sustainable development of the agricultural sector.

Most smallholders’ market participation analyses are done either on crop or livestock separately.

However, in mixed crop-livestock systems, household’s net position in one market might be affected by the other. Thus, net positions in crop and live animals

markets might be joint than separate.

Motivation…

Points of argument for joint decisions: Cash income from one market could be used in the other

(for consumption, investment, or saving).

Demand for cash in one market may necessitate the sale of commodities in the other.

Thus, the position in one market could influence/or be influenced by the position in the other market.

Objective

To examine how the likelihood of household’s net position in one market is affected by its net position in the other. Using simultaneous equation approach (joint decision)

and testing the mutual interdependence in the net market positions.

Analytical framework

Nine possible combinations in net positions

Net Buyer

Net seller

Position in Crop market

Autarkic

Net Buyer

Autarkic

Net seller

Position in Live animals market

Live animals

NB A NS

C

r

o

p

NB

A

NS

Analytical framework …

We have considered only two two possible combinations that have meaningful implications in the interdependence

Net Buyer

Net seller

Position in Crop market

Autarkic

Net Buyer

Autarkic

Net seller

Live animals

NB A NS

C

r

o

p

NB

A

NS

Position in Live animals market

• The two combinations are where potentially the cash from one market flows to the other.

Data

Household survey in 2009, for the 2007/08 production year.

1075 sample households in ten districts (four regional states) where the IPMS project operates.

Annual sale and purchase of crops and live animals

Annual income from other sources (dairy, apiculture, off-farm/non-farm income including remittances).

Household and farm characteristics, etc.

Sample distribution

Region District No. PAsNo. of sample households

Male headed Female headed Total

Tigray Atsbi Wenberta 12 86 34 120

Alamata 8 91 37 128

Amhara Metema 7 65 10 75

Fogera 11 89 23 112

Bure 9 83 32 115

Oromia Ada’a 9 103 23 126

Mieso 8 77 14 91

Gomma 8 62 19 81

SNNP Alaba 10 78 33 111

Dale 10 88 28 116

Total 92 822 253 1075

Methodology

Descriptive analyses

Econometric estimation models Binary simultaneous equations

Testing endogeneity of a net market position in the other (Hausman’s endogeneity test)

For a structural equation with endogenous explanatory variables, a two stage estimation procedure is followed:

Get predicted probability that a household takes a given net position in a specific market (crop or live animals).

Use the predicted probability values in the structural equation to estimate the effects.

Empirical models

CiCiLiCi uXMM 1*

10*

LiLiCiLi uXMM 2*

10*

CiLiCiCi vXXM 210*

LiLiCiLi vXXM 210*

Structural equations

Reduced form equations

Econometric specification

NB_C = f ( NS_L , X ) NS_L = f ( NB_C , X )

Net Buyer (Crop)

NS_C = f ( NB_L , X ) NB_L = f ( NS_C , X )

Net Seller(Live animals)

Net Seller(Crop)

Net Buyer(Live animals)

Identifiers: Shared and/or rented in land ….… for position in crop market Loss due to death of live animals ……. for position in live animals market

X = HH characteristics, farm characteristics, farm and non-farm income

Results

Results

Descriptive analyses results: Average value of annual crop production is Birr 16,550 per HH,

cash income from crop sale is Birr8,140 per HH, and crop purchased costs Birr1,510 per HH.

On average, cash income from live animals sale is Birr1,940, and Birr 580 is spent on live animals purchase per annum.

Net balance, average inflow of Birr 6,630 from crop and Birr1,360 from live animals markets.

Average (N=1075) (per Household)

Production Sale

(Birr)

Purchase

(Birr)

Net balance

(Birr)

Crop 16,550 8,140 1,510 6,630

Live animals 6.5 TLU(Average inventory)

1,940 580 1,360

Sample distribution in market position

Net position in

live animals market

Net position in

crop market

Total Net buyer Autarkic Net seller

Net buyer 22 (2.0) 0 (0.0) 96 (8.9) 118 (11.0)

Autarkic 68 (6.3) 0 (0.0) 172 (16.0) 240 (22.3)

Net seller 220 (20.5) 0 (0.0) 497 (46.2) 717 (66.7)

Total 310 (28.8) 0 (0.0) 765 (71.2) 1075 (100.0)• No autarkic HH in crops market• Larger proportion of the sample HHs are net sellers in both.

Simultaneity test results

Household net seller position in crop market and net buyer position in live animal market are simultaneously determined

But,

Net Buyer (Live animals)

Net Seller(Crop)

Net Buyer(Crop)

Net Seller(Live animals)

Household net buyer position in crop market and net seller position in live animal market are not determined simultaneously.

Explanatory variables Position in crop market Position in live animal marketNet buyer Net seller Net buyer Net seller

Coef.Std. Err. Coef.

Std. Err. Coef.

Std. Err. Coef. Std. Err.

Age of household head (years) 0.008* 0.005 -0.018*** 0.007 -0.016*** 0.006 0.001 0.004Sex of household head (1=male; 0=female) -0.177 0.118 0.219* 0.119 -0.143 0.146 0.105 0.113Education of household head (1=literate; 0=illiterate) 0.037 0.108 -0.146 0.121 -0.156 0.124 0.012 0.097Family size (persons) 0.054** 0.026 -0.034 0.028 0.078** 0.030 -0.042* 0.024Family labor available for agriculture (persons)

-0.107*** 0.035 0.123*** 0.035 -0.037 0.041 0.051 0.032Land owned (ha) -0.045 0.042 0.041 0.041 -0.010 0.048 -0.045 0.036Animals owned (TLU) -0.009 0.016 0.006 0.013 -0.043*** 0.015 0.070*** 0.012Value of crop production (1000Birr) -0.054*** 0.004 0.056*** 0.004 -0.003 0.004 -0.006** 0.002Income from honey and its products (1000Birr) -0.033 0.061 0.060 0.060 0.029 0.046 -0.002 0.042Off and non-farm income (1000Birr) 0.021** 0.009 -0.029*** 0.010 -0.017 0.015 0.010 0.006Income from dairy products sale (1000Birr) 0.036** 0.017 -0.032* 0.016 0.016 0.015 -0.007 0.013Dummy_ land rented or shared in (1=Yes; 0=No) -0.306*** 0.102 0.281*** 0.102Animals lost due to death (TLU) 0.147*** 0.054 -0.031 0.049Net seller in live animal (1=Yes; 0=No) 0.216** 0.102Net seller in live animals (predicted value) -0.695 0.604Net buyer in live animals (1=Yes; 0=No) 0.296* 0.161Net buyer in live animal (predicted value) -3.816** 1.720Net seller in crop (1=Yes; 0=No) 0.091 0.158Net seller in crop (predicted value) 1.476*** 0.537Net buyer in crop (1=Yes; 0=No) 0.133 0.114Net buyer in crop (predicted value) 0.528 0.375Constant 0.265 0.429 0.719* 0.399 -1.626*** 0.386 -0.034 0.242

Hausman’s endogeneity test results of the simultaneous equations (N=1075)

Interdependence in the net market positions

Marginal effects of the explanatory variables on the household net positions in crop markets

Explanatory variables Net position in crop markets

Net buyer Net seller

dy/dxStd. Err. dy/dx

Std. Err.

Age of household head (years) 0.002* 0.001 -0.002* 0.001

Sex of household head (1=male; 0=female) -0.048 0.034 0.050 0.034

Education of household head (1=literate; 0=illiterate) 0.010 0.028 -0.008 0.028

Family size (persons) 0.014** 0.007 -0.016** 0.007

Family labor available for agriculture (persons) -0.028*** 0.009 0.029*** 0.009

Land owned (ha) -0.011 0.011 0.009 0.011

Animals owned (TLU) -0.002 0.004 0.005 0.003

Value of crop production (1000Birr) -0.015*** 0.001 0.014*** 0.001

Income from honey and its products (1000Birr) -0.009 0.016 0.007 0.016

Off and non-farm income (1000Birr) 0.005** 0.002 -0.005** 0.002

Income from dairy products sale (1000Birr) 0.009** 0.005 -0.009** 0.005

Dummy_ land rented or shared in (1=Yes; 0=No) -0.079*** 0.027 0.077*** 0.027

Animals lost due to death (TLU)

Net seller in live animals (predicted value) -0.124 0.157

Net buyer in live animals (1=Yes; 0=No) 0.064* 0.034

Marginal effects of the explanatory variables on the household net positions in live animals markets

Explanatory variables Net position in live animals market

Net buyer Net seller

dy/dxStd. Err. dy/dx

Std. Err.

Age of household head (years) -0.003*** 0.001 0.001 0.001

Sex of household head (1=male; 0=female) -0.027 0.028 0.018 0.038

Education of household head (1=literate; 0=illiterate) -0.028 0.023 0.003 0.035

Family size (persons) 0.013*** 0.005 -0.012 0.008

Family labor available for agriculture (persons) -0.007 0.007 0.012 0.011

Land owned (ha) -0.002 0.008 -0.019 0.013

Animals owned (TLU) -0.007*** 0.003 0.024*** 0.004

Value of crop production (1000Birr) 0.000 0.001 -0.003*** 0.001

Income from honey and its products (1000Birr) 0.005 0.008 -0.003 0.015

Off and non-farm income (1000Birr) -0.003 0.002 0.004* 0.002

Income from dairy products sale (1000Birr) 0.003 0.003 -0.001 0.005

Dummy_ land rented or shared in (1=Yes; 0=No)

Animals lost due to death (TLU) 0.025*** 0.009 -0.011 0.017

Net seller in crop (predicted value) 0.279*** 0.078

Net buyer in crop (1=Yes; 0=No) 0.077** 0.033

Results …. (interdependence)

Net Buyer (Crop)

Net Seller(Live animals)

Net Seller (Crop)

Net Buyer(Live animals)

The probability of being a net buyer in live animals increases by 2.79% for a 10% increase in the probability of being a net sellers in crop market.

6.4%

2.79% / 10%

7.7%

The probability of being a net seller in live animals increases by 7.7% for the net buyers in crops market.

The probability of being a net seller in crop market increases by 6.4% for the net seller households in live animals market.

Conclusions and implications

There exists interdependence in households’ net market positions, which imply: Cash from crop sales is used to financed livestock purchase, Crop deficit for domestic consumption is financed by livestock

sale.

Relatively, effect of net position in crop market on net market position in livestock is stronger than effect of net position in livestock market on net position in crop market, which implies: Prior to their involvement in live animal market, farm households analyze

their position in crop market and their ability to meet cash requirements through crop sale.

In the case of cash surplus from crop sales, households stock live animal asset through selling surplus crops produced.

Only in the case of crop shortage for domestic consumption, households finance crop purchase through live animals sale.

Conclusions and implications

Policies/strategies enhancing smallholder market participation in crop and live animals markets in crop-livestock mixed system should pay attention to the production and marketing of both commodities simultaneously.

Thank You!