Poverty Impacts of the Doha Development Agenda

Thomas W. Hertel

and

L. Alan Winters

Purdue University and Development Research Group, The World Bank

2

Motivation• One of key goals of the Doha Development

Agenda is poverty reduction• Trade reform is also one of the avenues for

reaching the Millennium Development Goals • Largest trade distortions remain in

agriculture, which is also critical for the poor:– Poorest households are heavily dependent on agr– Poor spend large share of income on food

• Yet credible assessments have proven difficult

3

Methodology

• Establish new “policy benchmark”:– Post-UR, including ATC quota elimination– Post-WTO accession for China and others– Post-EU enlargement

• Quantify the DDA scenario• Assess implications for world markets• Communicate them to national models • Implications for poverty in individual countries: 13 case

studies in Latin America, Africa and Asia – 3 described by authors today

• Supplement with 2 global studies; draw conclusions

4

Elements of the DDA Scenariobased on July Framework Agreement

• Agriculture (Anderson and Martin, chp. 2):– Extensive binding overhang:

• Nearly 2 for industrial, 2.4 for developing, 5 for LDCs • Tariff cuts must be deep to have impact on trade flows • Use non-linear (tiered) formula (as with progressive) income tax:

– For developed: marginal rates (45, 70 and 75%) change at 10, 90% tariffs– For developing: marg rates (35, 40, 50, 60%) change at 20, 60, 120% tariffs– LDCs: no cuts– By varying marginal rates across tiers, avoid tariff discontinuities

– Definition of AMS leaves lots of wiggle room on domestic support, apply tiered formula: cuts of 60% and 75% developed, 40% developing, 0% LDC

– Export subsidies abolished

• NAMA: 50% cuts in tariffs (33% developing, 0% LDC)

5

Variants on the DDA Scenario

• Doha-All: fully reciprocal cuts in tariffs by developing and least developed countries

• Special and sensitive products:– 2% of tariff lines permitted (4% developing)– Chosen based on tariff/trade flow combination

(tariff revenue at HS-6 digit level)– Subjected to minimal (15%) tariff cuts– Erodes two-thirds of the cuts in developed

country agriculture protection

6

Assessing Global Impacts

• GTAP data base, supplemented by CEPII/ITC protection data base: MAcMap– Comprehensive treatment of preferences– Estimation of a.v. equivalent of specific tariffs

• Build up tariff cuts from HS-6 level:– Bound vs. applied rates

• Assess impact in modified version of GTAP model: GTAP-AGR (Keeney and Hertel)

7

Impact Of Trade Reforms On World Exports

(percentage change in volume)

0

20

40

60

80

100

120

140

rice beef sugar oil/fats apparel autos

Full-Lib DDA

8

Impact Of Trade Reforms On World Prices

(percentage change in average price)

-5

0

5

10

15

20

25

cotton oseeds cgrns dairy apparel autos

Full-Lib DDA

9

Communicating Global Results to National Models

• More complicated than appears at first glance:– Global model has its own representation of the national

economies – don’t want to do domestic reforms twice!

– Two views of focus country supply of products to world mkt

• We live with inherent inconsistency: national model has preferred representation of export supply curve– Communicate changes in international markets as changes

in cif import prices and vertical export demand shifts

– Omits national reforms when eliciting global market effects, but includes them, along with world price/demand shifts, when national model is simulated

10

Conceptual Framework for Country Case Studies (Winters)

Pass through, competition

Taxes, regulation,distributors, procurement

Distribution, taxes,regulation, co-ops

Co-operatives, technology, random shocks

World pricesand quantities

Border price

Wholesale price

Tariffs,QRs

Retail price

Exchangerate

Household welfarePrices, wages,endowments,

profits, other income

elderly

young

males

females

Enterprises

ProfitsWages

Employment

Tariff revenue

Taxes

Spending

11

Country Case StudiesPrice Transmission:• Mexico: Nicita• Mozambique: Arndt• Vietnam: Roland-Holst

Disaggregated HHld Incidence:• Brazil: Ferreira-Filho and

Horridge• Zambia: Balat and Porto• China: Kuiper and van Tongeren

Labor Markets:• Brazil: Bussolo et al.• China: Zhai and Hertel• Indonesia: Robilliard and

Robinson

Tax Replacement:• Cameroon: Emini et al.• Philippines: Cororaton et al.

Trade, Growth and Poverty:• Russia: Tarr et al.• Bangladesh: Annabi et al.• van der Mensbrugghe et al. (global)

Cross-country Comparison: • Ivanic (15 countries)

• Green = discussed• Red = presented today

12

Conceptual Framework: Price Transmission to HHlds

Pass through, competition

Taxes, regulation,distributors, procurement

Distribution, taxes,regulation, co-ops

Co-operatives, technology, random shocks

World pricesand quantities

Border price

Wholesale price

Tariffs,QRs

Retail price

Exchangerate

Household welfarePrices, wages,endowments,

profits, other income

elderly

young

males

females

Enterprises

ProfitsWages

Employment

Tariff revenue

Taxes

Spending

13

Incomplete price transmission yielded unequal gains from Mexican trade reforms in 1990’s

Source: Nicita, 2004.

14

Doha impacts on poorest rural households in Mexico also influenced by price transmission (Nicita, 2005)

0

0.5

1

1.5

2

2.5

Doha Doha+ Doha++

BorderNorthCenter South

Doha+ = Doha and Productivity enhancementDoha++ = Doha+ and enhanced price transmission

% c

hang

e re

al in

com

e

15

Focus on Household ImpactsTrade Policy and Poverty – Causal Connections

Pass through, competition

Taxes, regulation,distributors, procurement

Distribution, taxes,regulation, co-ops

Co-operatives, technology, random shocks

World pricesand quantities

Border price

Wholesale price

Tariffs,QRs

Retail price

Exchangerate

Household welfarePrices, wages,endowments,

profits, other income

elderly

young

males

females

Enterprises

ProfitsWages

Employment

Tariff revenue

Taxes

Spending

16

Disaggregated Household Impacts of Doha on Poverty in Brazil (Ferreiro-Filho and Horridge)

• The controversy: Brazil has been shown to be a big winner from OECD agricultural reform: Will all of these benefits accrue to the big commercial farms – thereby worsening Brazil’s income distribution?

• This study examines impact on 263,938 adult members of 112,055 hhlds spread over 27 regions

• Economic activity, employment and poverty vary widely by region

• Households diversified in earners and employment – account for impact of job gains/losses on poverty

17

Poverty Headcount by Region in Brazil

Amazonas Para

MtGrosso

MinasG

Bahia

MtGrSul

Goias

Maranhao

RGSul

Tocantins

SaoPaulo

Piaui

Rondonia

Roraima

Parana

Acre

Ceara

Amapa

StaCatari

Pernambuco

Paraiba

RGNorte

EspSanto

RioJaneiro

Alagoas

Sergipe

DF

0.14 (minimum)

0.24

0.35 (median)

0.51

0.58 (maximum)

Proportion below poverty line

18

Doha boosts employment in relatively poorer regions thereby reducing poverty

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

Current Headct delEmply delPvty

SaoPaoloRioBrazilTocantinsMaranhao

National headcnt falls by 236,000(Proportion

of pop.)(Percentage change)

19

Number of households leaving poverty, by region. Different scenarios.

-100000

-50000

0

50000

100000

150000

1 Rondonia

2 Acre

3 Amazonas

4 Roraim

a

5 Para

6 Amapa

7 Tocantins

8 Maranhao

9 Piaui

10 Ceara

11 RG

Norte

12 Paraiba

13 Pernambuco

14 Alagoas

15 Sergipe

16 Bahia

17 MinasG

18 EspSanto

19 RioJaneiro

20 SaoPaulo

21 Parana

22 StaCatari

23 RG

Sul

24 MtG

rSul

25 MtG

rosso

26 Goias

27 DF

Doha Fullib

Doha1 = 235,886 persons

Fullib = 481,988 persons

N

NE

SES

CW

Poverty impact across 27 regions in Brazil

20

Brazil summary

• Gainers/losers:– Households with lowest skill level workers gain most

due to the job creation effect– Followed by commercial farmers growing export

products– Small farmers also gain– Households relying on higher skill workers in Sao

Paolo & Rio are hurt due to job loss in heavy industry

• Inequality falls in wake of trade reforms

21

Focus on Labor MarketsTrade Policy and Poverty – Causal Connections

Pass through, competition

Taxes, regulation,distributors, procurement

Distribution, taxes,regulation, co-ops

Co-operatives, technology, random shocks

World pricesand quantities

Border price

Wholesale price

Tariffs,QRs

Retail price

Exchangerate

Household welfarePrices, wages,endowments,

profits, other income

elderly

young

males

females

Enterprises

ProfitsWages

Employment

Tariff revenue

Taxes

Spending

22

Educational attainment influences functioning of labor market and enhances poverty reduction

• Intersectoral labor mobility is key to poverty reduction, and education is key to farm-nonfarm mobility in China (Zhai and Hertel): – One add’l year of schooling boosts probability of obtaining

non-farm job by 14%– Education also boosts productivity – particularly off-farm– Yet rural ed expenditures lag urban areas by 16% per capita

• Consider impact of equalizing ed’l spending at same time as Doha implemented (paid for from combination of taxes and private spending):– Doha alone: 5 million poverty reduction– Doha + Rural Education: 50 million reduction in poverty

23

Focus on GrowthTrade Policy and Poverty – Causal Connections

Pass through, competition

Taxes, regulation,distributors, procurement

Distribution, taxes,regulation, co-ops

Co-operatives, technology, random shocks

World pricesand quantities

Border price

Wholesale price

Tariffs,QRs

Retail price

Exchangerate

Household welfarePrices, wages,endowments,

profits, other income

elderly

young

males

females

Enterprises

ProfitsWages

Employment

Tariff revenue

Taxes

Spending

24

Bangladesh Case Study (Annabi et al.)

• Expected to lose from Doha scenario:– Although not from preference erosion– Due to net agr importer status (cotton, grains & oils)

• Apparel a key industry:– Accounts for more than 2/3’s exports– Employs many low income workers (esp. women)– Global trade remains quite distorted

• Bracing for fallout from abolition of quotas• Examine SR vs. LR effects• Contrast with impact of own-liberalization

25

Long term impacts of trade reform in Bangladesh: aggregate welfare

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

2005 2010 2015 2020

Per

cen

tag

e va

riat

ion

fro

m B

aU

Doha_SDT Full-Lib

26

Country Studies Summary: Near Term Poverty Impacts of Trade Reform are Mixed

-4

-3

-2

-1

0

1

2

BangladeshBrazilCam

eroonChinaIndonesiaM

exicoM

ozambique

PhilippinesRussiaW

orld

DohaFull-Lib

Perc

enta

ge c

hang

e in

hea

dcou

nt

27

Long Term Poverty Impacts of Trade Reform are Uniformly Favorable:

(LR studies incorporate impact on investment)

-25

-20

-15

-10

-5

0

Bdsh Brazil China Vietnam World

DohaFull-Lib

Perc

enta

ge c

hang

e in

hea

dcou

nt

Note: LR results only available for 4 countries and world

28

Insights from Cross-Country Analysis

• Maros Ivanic’s cross-country study (chapter 15): – Brazil, Chile, Colombia, Mexico, Peru,

Venezuela, Uganda, Malawi, Mozambique, Zambia, Bangladesh, Thailand Indonesia, Philippines, Vietnam

• Single global model with households disaggregated (7 strata * 20 vingtiles = 140 hhlds/country)

29

Impacts of Doha and Full-Lib Compared

• Doha is less poverty friendly than Full-Lib• Operate on same instruments, but differing degrees

– We assume Doha will eliminate export subsidies, and developing country applied tariffs will be barely reduced

– But while export subsidy reforms lower poverty amongst agricultural hhlds, they raise poverty amongst other groups; so national poverty rises in many cases

– On the other hand, fully reciprocal cuts in developing country tariffs (Doha-All) would lower poverty in most of the sample

• Conclusion: Engagement by developing countries in liberalizing their trade regimes would make Doha more poverty friendly

30

Conclusions I

• DDA must be ambitious to affect development • Near-term poverty impacts mixed; on balance

poverty reduced under DDA; more so in LR• Admitting special and sensitive products in

agriculture (2%) would effectively eliminate any poverty reducing potential from the DDA

• Poverty impacts could be enhanced with deeper cuts in developing country bound tariffs

31

Conclusions II

• To have a significant near term poverty impact, complementary domestic reforms are required -- enabling hhlds to take advantage of new market opportunities

• Sustained long term poverty reduction depends on economic growth:– Impact of the DDA on investment is critical– Trade reforms need to be far reaching -- reducing barriers to

services trade and investment, in addition to merchandise tariffs

• For more information: www.worldbank.org/trade go to International Trade, then click on Topics >

Poverty and Trade