Agricultural policy reform under the WTO and Doha Kym Anderson Development Research Group, World Bank PREM Week, Washington DC, 25 April 2005.

Agricultural policy reformunder the WTO and Doha

Kym AndersonDevelopment Research Group, World Bank

PREM Week, Washington DC, 25 April 2005

Why much of the focus in DDA must be on agriculture … … even though it provides less than 4% of global GDP and 9% of int’l merchandise tradeOECD manufacturing tariffs have fallen by 9/10ths over the past 60 years to <4%, while agricultural protection has risen

Agric. applied (bound) tariffs now average nearly 5 (10) times manufactures tariffs globally

Also, the vast majority of the world’s poor rely on farming for a living, and may be hurt by agric protection policies of rich countries

Why focus on agriculture (cont.)True, the harm to some DC farmers from rich-country agricultural protection is reduced via non-reciprocal preference schemes such as the ACP’s Lome Agreement, EBA and AGOA But those schemes contravene the core WTO rule of non-discriminationIn particular, they exclude numerous populous DCs (eg Brazil, China, India, Indonesia, Pakistan, Vietnam)Hence they may harm more poor farmers (through trade diversion) than they help

Questions re. past, present, & future of agriculture in the WTO

Why the Uruguay Round (but not earlier GATT rounds) addressed agriculture

extent of pre-WTO growth in agricultural protectionism

How URAA addressed agriculture, and its economic effectsChallenges for Doha round and beyond

Why the UR (but not earlier GATT rounds) addressed agriculture

The long history of government interventions that distort agricultural marketsDistinctive features of distortions across countries and over timeReasons for those features, & for agriculture being neglected by GATT prior to 1986Why agriculture was included in the UR

History of government interventions in agricultural markets

Been going on for millenniasee Old Testament, e.g.

Sometimes to raise tax revenueSometimes to boost food self-sufficiency/food securitySometimes to reduce domestic price fluctuations

consumers concerned with peaksproducers concerned with troughs

Three past features of agricultural distortion patterns

1. The domestic-to-border price ratio was greater for agriculture relative to that for manufacturing, the higher a country’s per capita income, cet. par.

i.e. poor (rich) countries tended to depress (raise) incentives for farmers relative to manufacturers vis-a-vis international market price ratios

Three past features of agricultural distortion patterns (continued)

2. Agricultural protection was greater, the higher a country’s comparative disadvantage in agric, cet. par.

i.e. countries that would be net food exporters (importers) under free trade tended to depress (raise) incentives for farmers relative to manufacturers

Three past features of agricultural distortion patterns (continued)3. All countries tended to use trade policy measures to reduce fluctuations in domestic food prices and quantities

with agric-protective countries mainly reducing troughs in farmer prices and agric-taxing countries mainly reducing peaks in consumer prices of food

Implications for agricultural protectionism

As economies grew and their agric. comparative advantage declined, they tended to gradually reduce their discouragement of farmers (and support for food consumers), and to replace it with increasing support for farmers (at the expense of consumers and/or taxpayers)

Implications for food prices in int’l markets

Over time, the decline in agric taxation and growth in agric protectionism that accompanied economic growth put downward pressure on int’l agric pricesAnd the use of trade policy to stabilize domestic food markets exacerbated fluctuations in int’l food prices

Political economy of agricultural protection

Why was this pattern is observed across countries and over time?Since each country’s policy choice exacerbates the long-run downward trend and fluctuations in int’l food prices, it encouraged other countries to follow suit

So why did countries not agree multilaterally to desist before the 1990s?

What was different about the 1980s that brought agric to the Uruguay Round?

CAP-generated surpluses led to disposal via EU export subsidiesUS (& Canada) retaliated in kindPushed real food prices in int’l markets to century’s lowest level by 1986

which more than doubled the welfare costs of agricultural protection over the 1980s (Tyers and Anderson 1992)

Who brought agriculture into the UR?

US farmers were hurt more by EU policies than EU farmers were by US policiesAustralia/NZ and food-exporting DC farmers were affected hugely

led to formation of Cairns Group in 1986, whose sole aim was to keep agriculture high on UR agenda• its ag. exports = Japan’s man. exports

How URAA addressed agriculture

Sought commitments to reduce protectionist interventions in 3 areas:

cut agricultural export subsidiescut domestic subsidies to farmerscut barriers to agric and food imports• with SPS Agreement to reduce the

likelihood of re-instrumentalization

How URAA addressed agriculture (continued)

Explicit cuts were agreed to on all three types of measures

but in each case there was lots of ‘wriggle room’, such that in practice very little reform has occurred

1. Agric export subsidies to be cut:36% by value, 21% by volume over six years to 2000 (or, for DCs, by 2/3rds those rates by 2004)

How URAA addressed agriculture (continued)

2. Amber box domestic subsidies to farmers to be cut by 20% in aggregate by 2000 (or 13.3% for DCs by 2004)

but blue box and green box and de minimis exceptions ensure almost no cuts have taken place

How URAA addressed agriculture (continued)

3. Import market access:tariffication of NTBstariffs bound and reduced by 36% (unweighted average) and by 15+% on each itemminimum access of 3-5% of domestic market to be guaranteed by tariff rate quotas (TRQs)

How URAA addressed agriculture (continued)

‘Dirty tariffication’ meant very little increased market access in practiceIt also left most countries with the opportunity to vary their applied tariffs upward if desired (e.g. to keep domestic price from falling)

so the hoped-for reduction in international price fluctuations did not materialized

How URAA addressed agriculture (continued)

Tariff rate quotas (TRQs) have several undesirable features:

they legitimize a role for STEsthey generate quota rents

• recipients of which now oppose TRQ expansion and cuts to applied out-of-quota tariffs

they can discriminate between import-supplying countriesthey reduce welfare more than similarly protective tariffs (especially as int’l prices fall)

Challenges for Doha and beyond

The UR brought agric into the GATT mainstream, but:

export subsidies are still alloweda form of QR still restricts importsfew OECD countries have reduced their assistance to farmers since 1995

Hence agriculture remains by far the most protected goods sector post-UR

Challenges ahead (continued)

If tariff rate quotas in agric prove as difficult to remove as QRs on textile trade, they may be still with us in 2050!43 WTO members have TRQs, and more than half use themThe gap between the in-quota and out-of-quota tariffs provides huge gains to license holders

which means some previous supporters of agric trade reform are now less so

The Doha round’s progress

Rocky start (Seattle 1999, Cancun 2003), but by July 2004 WTO members had put together a Framework agreement that focused mostly on resolving agric issuesIf implemented, how much economic impact would it have, including relative to a move to complete free trade?

This was the subject of a DECRG research project over the past 12 months

What differentiates our new study?Its point of departure is the WTO’s July 2004 Framework agreement It examines in detail each of the 3 agricultural pillars plus preferences, cotton subsidies, non-agricultural tariffs, and S&D for DCs’ reformIt ‘adds up’ the consequences of current policies and prospective Doha reforms using data from CEPII/ITC & Bank’s Linkage model, incorporating:

• bound as well as applied tariffs at the HS6 level• non-reciprocal as well as reciprocal preferential tariffs• key trade policy changes to the start of 2005

Questions addressedWhat are the potential welfare gains from full goods trade reform, by country/region, due to:

developed relative to developing countries’ policies?agriculture relative to manufacturing policies?within agric., tariffs relative to export subsidies and domestic support?

How close might Doha get to completely freeing merchandise trade, in welfare and trade terms, based on July 2004 Framework agreement?

Modeling Doha reform packages using World Bank’s Linkage Model

Recursive dynamic CGE modelWe start with GTAP 2001 protection data and project on-going reforms from 2001 to end-2004

Uruguay Round including ATCEU25 enlargementWTO accession for China, etc.

Then we assume no further reform as global economy grows to 2015 (according to World Bank population, income, etc. projections), to get our global baseline scenario for 2015, against which to compare reform scenarios

Comparison with earlier studiesWelfare effects are smaller than when GTAP Version 5 database for 1997 is used (as in GEP2004, e.g.) because

Much liberalization since 1997, including implementation of unilateral reforms and regional and UR agreementsNon-reciprocal preferences are now in databaseNew provider (CEPII/ITC) of protection data

Current applied tariffs (%)

Agriculture and food

Textiles and

clothing

Other merchandi

se

Low-income countries

22 18 15

Middle-income countries

17 17 7

High-income countries

16 8 1

Linkage model’s gain by 2015 from removing current protection policies

Global benefit from removing current tariffs on all goods plus agricultural subsidies would be $287 billion per year by 2015

(Would have been about $350 billion if we included key reforms during 2001-04)2/3rds accrues to high-income countriesBut as % of GDP, the benefit to DCs is twice that for developed countries

Full liberalization: global gain ($bn)

$ billion due to reform by:

Agric & food

Textiles clothing

Other manuf

TOTAL

High-income countries

133 16 9 159(55%)

Developing countries

42 24 58 126 (45%)

All countries’ policies

182(62%)

38(14%)

67(24%)

287(100%)

Full lib’n: gains to developing countries

$billion due to reform by:

Agric & food

Textiles &

clothing

Other manuf.

TOTAL

High-income countries

26 15 4 44(50%)

Developing countries

27 9 6 45(50%)

All countries’ policies

54(62%)

22(27%)

10(11%)

86(100%

)

Relative importance of 3 agric pillars Welfare

gains from:

% of gain to:

Agric market access

Agric domestic support

Agric export

subsidies

All agric policies

Developing countries

106 2 -8 100%

World 93 5 2 100%

Welfare gain from full Liberalization(percentage change from baseline income in 2015)

-1.5 -0.5 0.5 1.5 2.5 3.5 4.5

USABangladesh

ChinaMexicoCanada

IndiaRest of ECA

EU-EFTARussia

Rest of S AsiaIndonesia

South AfricaAustralia

JapanOther Sub-Saharan Africa

ME&NAArgentina

Other Latin AmericaTurkey

Selected SSA countriesBrazil

Rest of E AsiaHK & SingaporeKorea & Taiwan

Thailand

Ag & food output rise from full lib’n(percentage change from baseline income in 2015)

-1.5 3.5 8.5 13.5 18.5 23.5 28.5 33.5 38.5

Brazil

Australia/ NZ

Argentina

Rest of LAC

Thailand

South Africa

Rest of SSA

Real farm income rise from full lib’n(percentage change from baseline income in 2015)

-1.5 8.5 18.5 28.5 38.5 48.5 58.5

Brazil

Argentina

Rest of LAC

Australia/ NZ

Thailand

South Africa

Rest of SSA

Effects of full lib’n on DC agric & food

% change in:

Real value

of agric and food

output

Real value of

agric and food

exports

Real net

farm income

Brazil 34 121 52

Sub-Saharan Africa

2 48 9

All developing countries

2 67 5

Effects of full lib’n on DC factor rewards

% change in:

Farm land

Unskilled wages

Skilled wages

Brazil 1.8 2.7 1.4

Sub-Saharan Africa

4.9 6.0 4.3

All developing countries

0.9 3.5 3.0

Take-away messages from full lib’n

Potential gains from further trade reform are largeEven after UR and recent accessions to WTO and EU

Must find the political will for Doha success

DCs would gain disproportionately from reformNotwithstanding non-reciprocal tariff preferencesBut as much would come from South-South as South-North trade growth, hence importance of DC lib’n too

Agricultural reforms are the highest priority for goods, from global and DC viewpoints, and if Doha is to be pro-development and pro-poor

Take-away messages (continued)Removal of agric export subsidies: great achievementRemoving cotton subsidies in US and EU would raise DC share of global cotton exports from 56% to 85%

and price of Brazil’s cotton exports by >8%

Reducing/disciplining other trade-distorting agric domestic support is crucial too, not least to prevent re-instrumentation of agric protection when tariffs are cut

But, gains to DCs from agric subsidy cuts could be multiplied many-fold by also cutting agric tariffs

with half those potential market access gains coming from South-South trade growth

Key elements of the Doha Agenda as shown in the July 2004 Framework agreement

3 agricultural pillars (including cotton)Non-agricultural market accessServicesTrade facilitationLesser tariff and subsidy cuts for developing countries (DCs) and zero cuts for least-developed countries (LDCs)

Our prospective Doha scenarios

We assume no services reform, no new trade facilitation, but:

phase out of agricultural export subsidiestiered cut to agricultural domestic supportand tiered cut to agric and non-agric bound tariffs under various alternative market access packages

Binding overhang in agric tariffs, %

Bound Applied

Mercosur 34 13

LDCs 78 13

All DCs 48 21

Agricultural market access

Tiered formula for cutting bound tariffs (with smaller cuts for DCs)

Formula sought by Harbinson yielded almost no gains to DCs• especially if lesser (15%) cuts for 2% of

products that are ‘sensitive’ and another 2% of DC products that are ‘special’

So we increased each cut by 10 percentage points more than Harbinson

Tiered ag tariff formula: line-by-line

Tiers in developed countries at 15 & 90% bound tariffs

Harbinson: cuts of 40, 50 and 60% Deeper cuts: marginal cuts 45, 70 & 75%

Tiers in developing countries at 20, 60, 120% bound tariffs

Harbinson: cuts of 25, 30, 35 and 40% Deeper formula: marginal cuts 35, 40, 50 & 60%

Agricultural domestic supportCut in bound AMS need not reduce applied support, because of binding overhang here as well (with 1986-88 ref. prices)

and overhang can be increased by abolishing admin prices used to calculate market price support

We apply a tiered reduction in bound AMS75% if AMS>20%, otherwise 60% for developed countries (40% for developing, zero for LDCs)• Leads to only 4 members reducing support:

US 28%, Norway 18%, EU 16%, Australia 10%

Non-agric market access, and extent of DC willingness to reform

50% cut in bound rates for high-income countries, 33% for DCs, 0% for LDCsWe also examine the effects of DCs (including LDCs) becoming full participants in Doha agric and NAMA cuts (Doha-All scenario)

recalling from earlier Rounds that DCs only got what they gave, in terms of increased market access (see Finger 1974, 1976; Finger and Schuknecht 2001)

Results from Doha agric reformTiered formula cut as per Harbinson gives the world $54 billion, but little goes to DCsSo we increased all cuts by 10 percentage points, which gave a $75 billion global gain

Even then, only $9 billion go to DCs& if HICs exempt just 2% ‘sensitive’ products (DCs 4%), global gain shrinks to $18 billion, and DCs’ gain disappears

• although a 200% tariff cap reduces much of that shrinkage

Small DC gains because of their (a) lesser cuts and (b) large tariff ‘binding overhang’

Adding non-agric market accessAdding 50%/33%/0% cuts to non-agric bound tariffs boosts global gain from agric tiered formula cut from $75 to $96 billion paThat $96 billion gets the world 1/3rd of the way to the potential gains from complete free trade in merchandise (but that share is smaller as % of gains from

removing also all services trade barriers, unless services markets also are opened up)

If DCs and LDCs fully participate in market access, global gain goes up to $119 billion

Effects of Doha lib’n on DC applied tariffs

% applied tariff in:

Baseline 2015

Doha (with lesser

cuts by DCs)

Doha-All

Brazil 9.5 9.2 8.5

Middle-income countries

7.2 6.3 5.6

Low-income countries

15.6 14.6 13.4

Effects of full & Doha lib’n on DC welfare

% change in real income in:

Full global lib’n

Doha (with lesser

cuts by DCs)

Doha-All

Brazil 1.5 0.55 0.59

Middle-income countries

0.8 0.15 0.21

Low-income countries

0.8 0.18 0.30

Doha welfare gains(Percent change from baseline income in 2015)

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5

Mexico

Bangladesh

United States

China

Canada

Russia

South Africa

India

Turkey

EU 25 plus EFTA

Hong Kong and Singapore

Indonesia

Argentina

Australia & New Zealand

Japan

Brazil

Thailand

Korea and Taiwan

Importance of 3 agric pillars to LICs’ welfare gain from Doha

Under Doha with S&D, gain to low-income countries is $3.6 billion per year If agric export subsidies and domestic support were not cut, gain would still be $3.6 billion

=> which confirms that most of gains from agric reform come from increased market access, not from subsidy cuts

Effects of full & Doha lib’n on DCs’ exports

$ billion p.a. change in exports of:

Full global lib’n

Doha (exports

to all countrie

s)

Doha (exports to just high-

income countries)

Doha (exports to other

DCs)

Agriculture and food

210 41 31 10

Other merchandise

399 37 31 6

Effects of full & Doha lib’n on DC share of agric and food production that is exported

% in:

Baseline 2015

Full global lib’n

Doha (with lesser

cuts by DCs)

Brazil 17 29 22

Middle-income countries

7 11 8

Low-income countries

8 12 8

Key cotton findings for DCs

Removal of cotton subsidies in US and EU would:

raise DC share of global cotton exports from 56% to 85%, and

raise Brazil’s export price by >8%, but SSA’s by <2%

Lessons and implicationsCuts in agric tariffs and domestic support bindings need to be large to get beyond binding overhangEven large cuts in agric tariffs do little if ‘sensitive’ and ‘special’ products are subjected to lesser cuts

Unless a tariff cap of, say, 100% is enforced or there’s a large expansion in TRQs of ‘sensitive’ products

DCs would have to make few cuts because of their huge binding overhang

So can afford to tone down their demands for lesser cuts (and ‘special’ products) and exchange it for greater access to OECD agric markets including ‘sensitive’ products

Lessons and implications (cont)

Adding non-agric market access to Doha package could double the welfare gains to DCs even with their lesser cuts, and it helps balance the North-South exchange of ‘concessions’

Some LDCs could lose slightly, as could some households within DCs that gain, if they reform little – the focus of the following presentations

New working papers and forthcoming book

Anderson, K. and W. Martin, ‘Agricultural Trade Reform and the Doha Development Agenda’, The World Economy September 2005 (forthcoming, also as a WB Policy Research Working Paper, May 2005)Anderson, K., W. Martin and D. van der Mensbrugghe, ‘Would Multilateral Trade Reform Benefit Sub-Saharan Africa?’ (forthcoming as a WB Policy Research Working Paper, May 2005)Anderson, K. and W. Martin (eds.), Agricultural Trade Reform and the Doha Development Agenda, Washington DC: World Bank, forthcoming summer 2005 but draft chapters now available on World Bank website