WPS3757




         Poverty Impacts of a WTO Agreement: Synthesis and

                                               Overview


                                                      by

                              Thomas W. Hertel and L. Alan Winters*




A version of this paper will appear as Chapter 1 in Poverty and the WTO: Impacts of the

Doha Development Agenda, Thomas W. Hertel and L. Alan Winters (eds.), forthcoming

from The World Bank, Washington, D.C.




World Bank Policy Research Working Paper 3757, October 2005

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the
exchange of ideas about development issues. An objective of the series is to get the findings out quickly,
even if the presentations are less than fully polished. The papers carry the names of the authors and should
be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely
those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors,
or the countries they represent. Policy Research Working Papers are available online at
http://econ.worldbank.org.



*Hertel is Distinguished Professor at Purdue University and consultant to the World Bank and Winters is
Director of the Development Economics Research Group, World Bank. This research was conducted while
Hertel was on leave with the World Bank. Financial support from the Bank-Netherlands Partnership
Program is gratefully acknowledged. The authors thank William Cline for extensive comments on this
paper as well as the research project.

Summary

         This paper reports on the findings from a major international research project
investigating the poverty impacts of a potential Doha Development Agenda. It combines
in a novel way the results from several strands of research. Firstly, it draws on an
intensive analysis of the DDA Framework Agreement, with particularly close attention
paid to potential reforms in agriculture. The scenarios are built up using newly available
tariff line data and their implications for world markets are established using a global
modeling framework. These world trade impacts, in turn, form the basis for twelve
country case studies of the national poverty impacts of these DDA scenarios. The focus
countries include: Bangladesh, Brazil (2 studies), Cameroon, China (2 studies),
Indonesia, Mexico, Mozambique, Philippines, Russia, and Zambia. The diversity of
approaches taken in these studies allows them to stress local conditions and priorities and
illustrates a wide range of facets of the trade-poverty links. It does, however, limit the
ability to draw broader conclusions. Thus an additional study which provides a 15
country cross-section analysis is aimed at this objective, and a global analysis provides
estimates for the world as a whole.

         A few of the main findings follow:

·        The liberalization targets under the DDA have to quite ambitious if the round is to
         have a measurable impact on world markets and hence poverty.
·        Assuming an ambitious DDA, we find the near-term poverty impacts to be mixed;
         some countries experience small poverty rises and others more substantial poverty
         declines. On balance, poverty is reduced under this DDA, and this reduction is
         more pronounced in the longer run.
·        Allowing minimal tariff cuts for just a small percentage of special and sensitive
         products virtually eliminates the global poverty reduction due to the DDA.
·        Deeper cuts in developing country tariffs would make the DDA more poverty
         friendly.
·        Key determinants of the national poverty impacts include: the incomplete
         transmission of world prices to rural households, barriers to the mobility of
         workers between sectors of the economy, as well as the incidence of national tax
         instruments used to replace lost tariff revenue.
·        In order to generate significant poverty reductions in the near term,
         complementary domestic reforms are required to enable households to take
         advantage of new market opportunities made available through the DDA.
·        Sustained long term poverty reductions depend on stimulating economic growth.
         Here, the impact of the DDA (and trade policy more generally) on productivity is
         critical. In order to fully realize their growth potential, trade reforms need to be
         far reaching, addressing barriers to services trade and investment in addition to
         merchandise tariffs.




                                                ii

Introduction and Motivation

        International trade is arguably the most direct economic means by which rich

countries influence poor countries. Exports of manufactures by developing countries

have increased rapidly over the last 30 years, due in part to falling tariffs in the OECD as

well as in developing countries, declining transport costs, increased specialization, and

sustained economic growth.        Whereas manufactures accounted for just 25% of

developing country exports in 1965, this share subsequently tripled to nearly 75% over

the next three decades, while agriculture's share of developing country exports has fallen

from 50% to under 10% (Hertel and Martin, 2000). Increased manufactures trade has

benefited many developing countries, helping them make the transition out of agriculture,

and lifting many out of poverty.

        Some of the poorest developing countries, however, have gained relatively little

from increased manufactures trade. Market access for their most competitive

manufactured export remains highly restricted (apparel), as it does for their key source of

employment and exports, farming, and the problem with agricultural exports is

exacerbated by the massive government subsidies provided to OECD farmers. Turning

to poverty within the poorest countries, developed-country agricultural policies become

even more central.     A majority of the poor are concentrated in rural areas, where

agriculture is usually the main source of economic activity (World Bank, 2004), and in

the poorest developing countries, large shares of households (including most of the very

poorest) depend on self-employment in agriculture for virtually all of their income

(Hertel et al., 2004b).     Together, these facts highlight the potential influence that

multilateral trade policies can have on poverty in developing countries.

         The Doha Development Agenda (DDA) negotiations, sponsored by the World

Trade Organization, experienced a blow in Cancun, Mexico, precisely over the question

of rich country agricultural support and its potential impacts on poverty in developing

countries. The Doha negotiations are now emphasizing the need to better understand the

linkages between trade policies ­ particularly in rich countries ­ and poverty in the

developing world. Moreover, poverty reduction is now widely accepted as a central

focus for development efforts and has become the main mission of the World Bank and

other development institutions. For example, the "Millennium Development Goals"

commit the international community to halve poverty by 2015, and locate several key

means to this goal in international trade.


         With this high level of policy interest, it is hardly surprising that the issue of trade

and developing-country poverty has become a focus of much research activity over the

last several years.         This paper summarizes a research program that offers the first

comprehensive analysis of the national poverty impacts of specific policy reforms

proposed under the auspices of the WTO.1 It combines the results from several strands of

research in a novel way. First, it draws on an intensive analysis of the July 2004 DDA

Framework Agreement, particularly of potential reforms in agriculture, which, as we

shall see, have special significance to the poor. The scenarios analyzed below are built up

from newly available tariff line data on bound and applied tariff rates. Similarly detailed

analysis is undertaken in the case of domestic support for agriculture and export

subsidies, as well as for non-agricultural market access.




1Most of the studies in the program are included in the World Bank Policy Research Working Paper series.
All are available at https://www.gtap.agecon.purdue.edu/poverty


                                                     4

       Second, the research assesses the implications of these alternative Doha scenarios

for world markets. These are established using a state-of-the-art, global modeling

framework which incorporates the most recent econometric evidence on supply and

demand elasticities ­ with particularly close attention paid to food and agriculture

markets which prove crucial in assessing the poverty impacts of the DDA. The outputs of

this part of the project include export and import price changes for each region of the

world, along with changes in export volumes.


       Third, these world trade impacts form the basis for analyzing the poverty impacts

of the DDA on ten individual countries by way of a dozen case studies. These case

studies use a variety of innovative techniques to establish the potential impacts of the

DDA on different household groups and, in some cases, different regions within the

country. The focus countries are: Bangladesh, Brazil (2 studies), Cameroon, China (2

studies), Indonesia, Mexico, Mozambique, Philippines, Russia, and Zambia.


       Some case studies also examine other poverty policies in addition to trade reforms

­ for example education reform or agricultural extension services. Sometimes these are

complementary to the Doha Round in the sense of enhancing its effect, but more often

they are independent. They are explored here as yardsticks against which trade reform

can be measured and as suggestions as to how governments can seek to overcome any

adverse poverty effects from the Round. However, we do not subscribe to the view that

such `complementary policies' are necessary for the Doha Round to be beneficial.




                                           5

1.     Choice of Methodologies


       In organizing the research summarized here, we had two contrasting objectives.

On the one hand, we wanted the studies to be consistent with one another in order to

ensure an accurate global assessment of the DDA, as well as comparability across

studies. On the other hand, research into the poverty impacts of trade reform is new, and

almost the only consensus it has reached is that countries differ. From this perspective,

we wanted both to encourage a variety of approaches at the country level and to exploit

the specific skills and knowledge of case-study authors to gear their country models most

closely to local characteristics and issues.2

       The project, therefore, is a composite in which the global analysis ­ the

methodology for deriving the global findings and passing them over to the national case

studies ­ is unique and consistent with current standards in the field of quantitative trade

policy analysis, while the country case studies display a wide range of methodological

innovations and topical design features. This variety has been fruitful, with different

country studies emphasizing alternative links between trade and poverty and providing a

diversity of insights. Nevertheless, as a check and in order to permit us to draw some

broader conclusions, we have included two more uniform exercises: first, a 15 country

cross-section analysis, in which a common, fully integrated trade-poverty analysis is

provided for a range of developing countries. Second, we have included a global analysis

of aggregate poverty impacts derived by applying simple poverty elasticities to the

predicted outcomes for developing countries in a global simulation of a prospective Doha

agreement.

2 The forthcoming book on Globalization and Poverty, Harrison (2005), adopts the same strategy,
combining a set of cross-country econometric studies with several individual country case studies.


                                                   6

        The prevailing methodology in the research is Computable General Equilibrium

(CGE) analysis. This is the dominant methodology for the ex ante analysis of the

economic consequences of comprehensive trade agreements ­ be they multilateral or

bilateral in nature (Shiells and Francois, 1994). The reason for this dominance is that no

other approach offers the same flexibility for looking at prospective changes in trade

policy, while respecting the fundamental economy-wide consistency requirements such

as balance of payments equilibrium and labor and capital market constraints that are so

important in determining the consequences of comprehensive trade reforms. The CGE

approach has come under substantial criticism (e.g., from Jorgenson, 1984; McKitrick,

1998; Kehoe, 2005) for having insufficient econometric underpinnings, and for not being

adequately validated. Accordingly, in this project we offered a number of econometric-

based analyses that focus on key dimensions of the trade and poverty question, including:

price transmission from the border to households, cropping choices by farm households,

labor market participation decisions, and the intersectoral movement of labor. In addition,

when we assess the global market impacts, we use a CGE model based on the most recent

econometric evidence on supply and demand elasticities and for which some (modest)

validation has been undertaken.

        The majority of the studies in the project and summarized here are based on

comparative static analysis. Thus the authors abstract from the impact of trade reform on

investment and productivity and therefore economic growth. There are two reasons for

this emphasis. First, most of the issues that arise in the popular debate over the poverty

impacts of trade policy are fundamentally comparative static in nature. Concerns about:

the urban poor being adversely affected by higher food prices, the potential loss of jobs




                                             7

by women in the apparel sector, or the poverty impacts on low income farmers in

developing countries are all questions about the redistributive impact of trade policy

reform. To answer them one needs a disaggregated, comparative static framework. Of

course, we are also keenly interested in the potential for economic growth to alleviate

poverty, and five of the studies utilize and dynamic framework that accounts for the

growth effects of changes in investment deriving from trade policy reform. However,

quantifying the impact of trade reform on growth and poverty through channels such as

the effect on productivity or the benefits of increasing the range of available goods

remains a lively topic for current research on which consensus has yet to emerge. Hence

our second reason for using the comparative static approach is to avoid any appearance of

overstating the poverty alleviating benefits of liberalization.

       In the end, it must be said that this project has proven to be a very ambitious

undertaking ­ attempting to bridge micro-based research focusing on the choices and

opportunities facing individual households in developing countries with macro-based

research on the global impacts of multilateral trade policy reform. The payoff to this

exercise must be judged by the insights offered. And it is to these that we now turn.




2.     The Global Impact of the Doha Agenda

       Kym Anderson and Will Martin (2005) take as its starting point the July WTO

Framework Agreement for the Doha Agenda. It explores the issues flowing out of this

document ­ and in particular the annexes dealing with export subsidies, domestic support

and market access in agriculture, as well as market access for non-agricultural goods. It

examines seven different Doha scenarios, of which we adopt one as the core scenario for



                                             8

the project. In constructing this scenario, the authors have taken considerable care to

distinguish those trade reforms that are actually being negotiated under the Doha

Development Agenda from those that have already been agreed to previously. This

distinction is complicated by the fact that virtually all of our policy data bases pre-date

completion of the Uruguay Round Agreement. In fact, the starting point for all of the

analysis in this project is the year 2001 ­ the most recent one for which comprehensive

data are available for tariffs, domestic support and export interventions. Therefore, prior

to constructing the Doha scenario, a "pre-experiment" is undertaken in order to account

for the major developments in trade policy since 2001. These include: tariff reforms

undertaken by newly acceding WTO members ­ most notably China, the phase-in of

remaining Uruguay Round commitments by developing countries, EU enlargement to 25

countries, and the abolition of export quotas on textiles and apparel under the Agreement

on Textiles and Clothing. Thus, even though the full impact of some of these reforms is

yet to be felt, the analysis in this project looks beyond these reforms, envisioning a global

economy in which they have been fully implemented, and focusing on the further impacts

of trade liberalization undertaken in the context of the Doha negotiations.

        The most important finding from Anderson and Martin (2005) is that, unless the

Doha Agenda is considerably more ambitious than the Uruguay Round in terms of depth

of cuts in bound tariffs and domestic support, it will achieve little development stimulus.

The main problem on the market access side is binding overhang. For example, in

agriculture ­ one of the key areas of the Doha Agenda with respect to trade and poverty ­

bound tariffs in developing countries average 48% while applied tariffs average are only

21%. In the case of the least developed countries, the respective figures are 78% and




                                                9

13%! Even in the EU (21% binding vs. 12% applied) and USA (6% binding vs. 3%

applied) there is substantial binding overhang in agriculture. So for many

countries/products, bound tariffs can be cut deeply with no impact on applied protection

and hence international trade.

        In the central Doha scenario featured in our project, agricultural tariffs are cut

using a tiered formula, with marginal cuts changing at 15 and 90 percent bound tariff

rates. The marginal cuts are 45 percent for the lowest agricultural tariffs, 70 percent for

tariffs in the middle range and 75 percent marginal cuts for the highest tariffs.3 For

developing countries, the inflection points are placed at 20, 60 and 120 percent bound

tariff levels in agriculture, with marginal cuts of 35, 40, 50 and 60 percent, respectively.

In non-agriculture, tariffs are subjected to proportional cuts of 50 percent for developed

and 33 percent for developing countries. The Least Developed Countries (LDCs) are not

required to cut tariffs under this central scenario. Box 1 summarizes our central Doha

scenario.

        There is much more to the DDA than just agriculture and non-agricultural market

access ­ for example, trade facilitation, services liberalization and rules on anti-dumping

and regionalism. We focus on the former issues partly because they are quantifiable and

provide a large agenda in themselves. Mainly, however, they are likely to be the major

issues both in terms of effects and in terms of negotiators' need for detailed quantitative

advice. Moreover, the other issues are basically additive to the analysis of market access

for goods, so that as their outcomes and consequences become clear they may be added to

our results to get an overall picture.



3For example, a tariff of, say, 100% is cut by 66.95%: = [15%*0.45 + (90-15)%*0.70 + (100-90)%*0.75].
By applying the cuts at the margin we avoid the discontinuities implied by the July Framework.


                                                   10

       As a consequence of the relatively ambitious tariff cuts we analyze, average

world-wide tariffs for all merchandise trade drop from 4.7% in the baseline to 3.2%. This

masks rather different cuts for countries at different income levels. High income

countries' tariffs fall from 2.9 to 1.6%, middle income countries' tariffs from 7.2 to 6.3%

and low income tariffs (including LDCs which do not cut tariffs at all) from 15.6 to

14.6%. (Anderson and Martin report these cuts on a more detailed basis in their paper.)

       In the case of domestic support, there is also a problem of bound vs. applied

protection, with bindings generally much higher than applied Aggregate Market Support

(AMS). But even more severe is the definition of the AMS itself ­ in particular its

reliance on administered prices as a benchmark. This feature makes it possible for

administrators in some countries to bring programs into WTO compliance with the stroke

of a pen ­ simply by abolishing the administered price! The core Doha scenario assumes

that industrial countries with domestic support in excess of 20 percent of production cut

their bound AMS commitments by 75 percent, while others cut by 60 percent.

Developing countries are assumed to cut their AMS by 40 percent. Even with these

ambitious reductions, only six WTO members would be required to reduce actual

support, based on 2001 notifications: Australia, EU, Iceland, Norway, Thailand and USA.

       Export subsidies are the one area where bold cuts (full elimination) are on the

table, but these have diminished in importance over time. At present, they remain a

significant factor only in the case of the EU (and in the US for dairy products) ­ and the

abolition of export subsidies has been made conditional on equivalent treatment of food

aid and state-trading. Preliminary estimates suggest that reform of the latter two items

will have little impact, but the linking of these features to the WTO negotiations makes




                                             11

the whole process much more complex. Our central Doha scenario assumes export

subsidies are abolished.

        In addition to this central Doha scenario, we also consider an important variant in

which developing countries fully reciprocate the tariff cuts made by developed countries,

thereby eliminating one of the historical pillars of Special and Differential Treatment.

The rationale for considering this alternative, which we label Doha-All, will become clear

when we discuss the results of the global poverty analyses below. Under Doha-All,

average merchandise tariffs in the middle and low income countries drop further ­ to 5.6

and 13.4 percent, respectively. In the case of the low income countries this represents a

larger incremental cut in average tariffs than was achieved in the central Doha scenario

itself. Finally, we also explore on occasions a Full liberalization (Full Lib) scenario in

which all explicit trade policy barriers identified in the database are abolished.

        Assuming that negotiators do indeed honor their initial vision as set forth in Doha

and make significant cuts in agricultural and non-agricultural protection, what impact

might this have on poverty? Will they really put Development squarely into the Doha

Agenda? We turn next to the studies that endeavor to answer this question.

        We begin with the impact of the Doha reforms on world market prices. Thomas

Hertel and Maros Ivanic (2005) utilize a global computable general equilibrium model to

assess the potential impact on world market prices and trade volumes. As established in

Anderson and Martin (2005), agricultural protection is central to any assessment of

global trade reform and the global analysis in this study bears this out. The trade reform

scenarios invariably have the biggest impact on prices and trade volumes for farm and

food products, followed by textiles and apparel. Given the predominance of the poor in




                                              12

rural areas and their heavy reliance on unskilled wages elsewhere, these are the key

industries when it comes to any poverty assessment. The strongest world price increases

are for the heavily subsidized farm products: rice and other grains, cotton, dairy products

and beef. The ranking of the price rises arises from the composition of cuts ­ both across

the three sets of agricultural distortions and across countries. The other important point

made in this paper is that, given the increasingly differentiated nature of traded products,

there is no one "world price" and careful attention must be paid to bilateral patterns of

trade and country-specific price changes.

        Finally, Hertel and Ivanic outline the methodology for transmitting the generated

price and volume changes to the national case studies. This turns out to be quite complex,

and represents an important innovation in the linking of global economic outcomes with

national impacts.




3.      Price Transmission

        Our analysis of the country case studies is structured around the conceptual

framework laid out by Winters (2002) and Winters, McCulloch and McKay (2004). This

begins with the question of price transmission ­ namely how much of the world price

shock is transmitted to producers and consumers?

        With a majority of the poor in most countries located in rural areas ­ often poorly

served by transportation and communication infrastructure, it is important to ask whether

developments in global markets will really have an impact on these households. Of

course, this is an empirical question, subject to econometric investigation, and this is

precisely what Alessandro Nicita (2004) does in the case of Mexico. He shows that



                                             13

indeed world prices are differentially transmitted to the regions of the country, depending

on their distance from the border and the nature of the commodity in question. He begins

his analysis by examining the extent of "pass-through" from international prices to

domestic prices at the border. Here, he finds that for manufactured goods, about two-

thirds of the international price change passes through to the domestic market, whereas

the comparable figure for agriculture is just one-quarter.

       Nicita's econometric estimates also show that the transmission of world market

price changes diminishes with distance from the border. In addition, urban areas are more

sensitive to border prices changes, when compared to rural areas. Therefore, he concludes

that in the more remote, rural regions of Mexico, very little of the international price

changes will be felt ­ particularly in the case of agricultural products. As a consequence,

the impact of the Doha scenarios ­ which have only modest impacts on world prices,

anyway ­ are negligible in rural Mexico, except in the North, near the US border, where

rural households see some small gains. Urban consumers face higher food prices and a

small decline in unskilled wages as the privileged Mexican position in the US market is

eroded by MFN tariff cuts. Thus the urban poor experience small losses.

       Nicita (2005) also explores the impact of complementary domestic reforms that

might permit rural producers to respond to improved world market conditions without

incurring additional costs (e.g., a productivity gain or the employment of surplus labor).

This enhances the welfare outcome for rural households in all regions excepting the

South. Rural households in the South benefit from Doha only when the reforms are

accompanied by enhanced price transmission ­ e.g., through improved transport and




                                             14

market infrastructure. So there is an important interaction between price transmission and

the distribution of gains from global trade reforms.

        One of the poorest countries in the world, which also has very poor infrastructure

and is plagued by high domestic marketing costs, is Mozambique. In fact, recent work by

Arndt et al. (2000) estimates producer-consumer margins as high as 300% (for cassava).

The biggest margins reported in their study are for food products, which tend to dominate

both the consumption and production bundles of the poor. So the existence and behavior

of these margins is critically important for any poverty study. Channing Arndt (2005)

explores this issue in the context of the Doha Round scenarios for Mozambique. As with

the Mexico study, the combination of these marketing margins with modest world price

changes means that the impact on household welfare in Mozambique is quite small.

Indeed, about one-third of rural households are unaffected by the Doha scenario. The

largest rural losses are about one percent of income, with some households experiencing

modest gains. The dispersion among urban households is larger, due to the presence of

smaller marketing margins. Overall the impact of multilateral trade reform on

Mozambique is adverse, as preferences are eroded and prices of imports rise.




4.      The Disaggregated Impact on Households

        Moving beyond the question of price transmission, we come to the issue of

household level impacts of ­ and household responses to ­ the price changes ensuing

from trade reforms. The simplest way of exploring this link is to focus on a single

commodity. This is the approach taken by Jorge Balat and Guido Porto (2005) on the

impact of trade reform on cotton producers in Zambia. They note that the critical factor in



                                             15

this case is the share of household income generated by cotton production. To a first-

order approximation, the real income impact of a change in the price of cotton may be

obtained by multiplying this income share by the percentage change in cotton price. This

leads them to focus on the evolution of cotton income shares amongst the poor in

Zambia. Since cotton is grown in significant quantities only in three provinces, this is

where they focus attention.

       One of the striking things about world cotton markets in the late 1990s was the

collapse in world prices. Between 1996 and 1998, cotton prices in Zambia fell by 20

percent. Therefore, it is surprising that cotton's share in income among the poor rose

sharply in the Eastern and Southern Provinces over this same period. Indeed, amongst the

poorest households in the Eastern Province, the increase was nearly five-fold ­ even as

the income share fell for wealthier households. While there are many factors that may

bear on this change, the authors argue that the most likely reason was the reform of the

cotton marketing board system and the implementation of an out-grower scheme which

proved effective in getting seed and fertilizer into the hands of credit-constrained, small

scale producers. This increase in the cotton share boosts the potential benefits from

multilateral agricultural reforms, since one of the main consequences of such reform

would be to raise cotton prices.

       Despite the increase in cotton income shares over this period, the income impact

on the poor of higher cotton prices ­ the authors assume a 12% price rise, based on

several independent studies of world cotton markets ­ is still relatively modest (of the

order of one percent of real income, on average) because the average income share is

about 8%. This brings them to a discussion of complementary domestic reforms. In




                                            16

particular, they cite evidence from other research they have conducted in Zambia which

finds that access to extension services can boost productivity by more than eight percent,

resulting in an aggregate gain of more than nine percent, when combined with higher

cotton prices.

       But the largest poverty reduction benefits appear to arise when subsistence

households switch to cotton production in the wake of increased demand for exports.

Here, a careful matching of subsistence and cotton-producing households shows that, all

else constant, subsistence producers could boost their incomes by nearly 20 percent if

they switched to cotton production. Such a switch would be greatly facilitated by

continued improvement of the out-grower schemes and strong demand for cotton exports.

When combined with improved extension services and higher cotton prices, the switch

from subsistence production to cotton could boost incomes of some of the poorest

households in Zambia by nearly one-third. In sum, Balat and Porto conclude that trade

reform alone is not sufficient to raise a large number of poor out of poverty in Zambia,

but that when the market opportunities presented by trade reforms are combined with

complementary domestic reforms, significant headway in the fight against poverty is

possible.

       Of course, global trade reforms do not simply alter one single commodity price:

rather they potentially affect all prices in the economy ­ including the prices of non-

tradeable commodities and services as well as wages and returns to land and capital. So

we turn next to a study that seeks to account for the full range of price impacts at a highly

disaggregated level. The unusual thing about Joaquim Ferriera Filho and Mark

Horridge's (2005) paper is the very large number of individuals considered in their




                                             17

analysis ­ 264,000 adults who are members of 112,000 households spread across the 27

regions of Brazil. The authors argue that the regional dimension of their study is critical,

given the tremendous disparities in income and poverty incidence across regions. The

proportion of poor households ranges from about 14% in parts of the Southeast, to nearly

60% in the North (Amapa). When combined with large variations in industrial

composition across regions, there is a recipe for great differences in poverty impacts due

to trade reform.

       Ferriera Filho and Horridge find that the Doha scenarios benefit agriculture at the

expense of industry. This is no surprise, as virtually all previous studies of global

agricultural trade reform have concluded that Brazil would be a substantial beneficiary

from such a development. However, the real question is: Which households within Brazil

will benefit? Many believe that all of the benefits will go to large farmers, thereby

worsening the income distribution in Brazil. The research reported in this paper argues

that, when one takes account of the additional employment generated by the expansion of

agriculture and related industries in many of the poorer states of Brazil, the largest

gainers are actually the households most heavily reliant on low-skill labor. As a

consequence, the income distribution in Brazil improves under the Doha scenario. This is

a very important finding. It is a point that has been previously emphasized in more highly

aggregate research on trade and poverty reported in Harrison et al. (2003).

       As a percentage of initial poverty, the estimated national decline in this paper is

modest (less than one percent), but it still amounts to a large number of persons. Under

the Doha scenarios, poverty falls by about 236,000, and it declines by about twice that

amount in the case of the Full-Lib scenario. The declines in poverty are fueled by the




                                              18

growth in agricultural activity ­ Brazilian farm and food exports expand strongly in the

wake of trade reform ­ and the subsequent increase in demand for the lowest skill

workers, 41% of whom still work in the farm sector.

        Of course these wage gains hinge on the existence of an operational labor market.

Such a market may not exist in some cases and the potential consequences of factor

market failure are explored in considerable depth in Marijke Kuiper and Frank van

Tongeren (2005). These authors approach this problem by employing a village-level

model of a community in Jiangxi Province in China. They capture the heterogeneity of

household types by grouping them according to their factor endowments. In particular,

they distinguish whether or not households have access to draught power and whether or

not they have family members involved in temporary migration outside the province.

After a detailed analysis of circumstances in this village, they conclude that the markets

for labor, land and capital are imperfect, thereby preventing households from simply

taking wages and rental payments as given when making decisions about consumption

and production. This "non-separability" complicates the household's decision process

and can result in some striking results in the wake of trade reforms as the authors show.

        In the case of Doha reforms, the real income gains for the village are quite modest

­ about 1.2% of income ­ and relatively evenly spread across the different household

groups. However, in the case of full liberalization, the aggregate gains are four times as

large, and also much more unevenly spread across households, with the gains to

households with draught power nearly twice as large as those for the other household

groups. This reflects the intensification of production in agriculture engendered by higher

prices for rice and other farm products.




                                             19

5.      Labor Markets

        The main resource with which the poor are endowed is their own labor. Whether

they are self-employed farmers, providers of services, or wage earners, their income is

closely tied to conditions in the labor market. This point surfaces clearly in the Brazil and

China studies discussed above, both of which emphasize the importance of labor markets

as a mechanism for transmitting favorable developments in the world marketplace, as

well as elsewhere in the domestic economy, to impoverished households. We now turn to

studies which focus primarily on the labor markets in Brazil and China, as well as one

focusing on a third country ­ Indonesia. The first of these is Maurizio Bussolo, Johan Lay

and Dominique van der Mensbrugghe (2005) on Brazil. Their focus is specifically on the

link between the farm and non-farm labor forces. They model the decision to move out of

agriculture based on an econometric model that predicts the likelihood of a given

individual changing sectors, based on the historical evidence in Brazil. The other

important feature of this paper is that they set their analysis in the context of a 2001-2015

baseline for the Brazilian economy. This permits us to view the impacts of trade reform

in the context of ongoing changes in the economy, labor markets and poverty.

        In their baseline projection, Bussolo, Lay and van der Mensbrugghe find that the

poverty headcount falls by almost 14%. The majority of this decline is due to poverty

reduction in agriculture -- a sector which grows considerably faster than the non-farm

economy under their business as usual (BaU) forecast. The majority of this poverty

reduction is due to factor price changes (e.g., higher wages), but a significant portion is

due to the exit of labor from the relatively low wage agricultural sector to higher wage,




                                             20

non-farm jobs. This intersectoral movement is particularly important to the poorest farm

households.

        Having established this baseline scenario, the authors analyze the implications of

alternative trade reforms for poverty ­ and in particular for the different labor force

groups: the "movers" who move from agriculture to non-agriculture over the course of

the baseline, the "stayers" who remain in agriculture, and the "stayers" in non-

agriculture. The largest percentage point reduction in poverty over the baseline is for the

"movers" who experience a 22.4 percentage point reduction in their headcount (down

from 53.4% to 31%). This is the poorest of the three groups, and it is also the group that

experiences the greatest incremental poverty reduction, above and beyond the baseline, as

a result of the Doha trade reforms. Overall, the authors find quite modest poverty gains

from the Doha scenarios (just 3% of the baseline change over the 2001-2015 period). Full

liberalization generates estimates of national poverty reduction that are three times as

large as the Doha reductions ­ but still modest in the context of projected baseline

changes. This underscores the fact that trade reforms taken alone are a relatively small

piece of the overall poverty reduction puzzle.

        Fan Zhai and Thomas Hertel (2005) take a deeper look at the Doha reforms

through the lens of a labor-focused CGE model of China ­ and the scope for enhancing

these outcomes through complementary education reforms. Like the Bussolo et al. paper,

this paper emphasizes the farm/non-farm labor market linkage which the authors argue is

partly a function of educational attainment and therefore susceptible to change through

educational policy. They also emphasize the link between rural and urban labor markets

in China ­ through the temporary migration of workers. (Permanent migration is still




                                            21

restricted in that country.) In their analysis of multilateral trade reforms, the authors find

that poverty falls across all of their household categories: by 1.3% in the case of Doha

and 2.7% in the case of full liberalization. Inequality also declines slightly under these

scenarios.

        Zhai and Hertel cite econometric evidence that suggests that an additional year of

education boosts an individual's chances of obtaining an off-farm job in China by 14%.

Educational attainment is also important for workers seeking to meet the needs of an

increasingly integrated global marketplace. Yet education expenditures per pupil in the

rural areas lag significantly behind their urban counterparts in China. So the authors

explore the implications of accompanying trade reform with additional educational

investments in rural areas to enhance rural labor mobility, productivity and income. In

particular, they boost expenditures per pupil enrolled in mandatory education by 16% to

reach the comparable urban level. This increment is assumed to be financed in part by

public funds, raised through additional taxation, and in part through increased private

contributions taken out of rural households' disposable income. This combination of

educational and trade reforms has a much stronger impact on poverty alleviation, with the

number of poor (living below $2/day) falling by 13.4%. This scenario also has a

favorable impact on rural-urban income inequality.

        The final paper focusing on labor markets that we discuss is a case study of

Indonesia by Anne-Sophie Robilliard and Sherman Robinson (2005). Instead of focusing

on the farm/non-farm or rural/urban movement of labor, these authors draw a sharp

distinction between the formal and informal labor markets. The formal sector offers high

wages, but few opportunities for employment. The informal sector, by contrast, has a




                                              22

flexible wage which is assumed to clear the market. Robilliard and Robinson explicitly

model each individual's decision to participate in one or the other of these labor markets.

In this way, they are able to predict which types of individuals will lose their job when

formal sector employment contracts, and which will be hired when employment expands.

These changes in employment represent an important determinant of the welfare impacts

on households of any change in a country's pattern of trade, production and employment.


         Robilliard and Robinson explore the poverty impacts of multilateral trade reform

under three alternative labor market closures: fixed aggregate employment and flexible

wages, fixed, sector-specific labor (no change in employment by sector), and fixed real

wages and variable aggregate employment (i.e. changes in unemployment are permitted).

They focus on the full liberalization scenario for this sensitivity analysis and find that the

largest reduction in poverty comes from the fixed employment scenario ­ about 1.4

million people are lifted out of poverty. The proportional reduction is slightly higher in

the rural areas and more favorable to the poorest of the poor as well, so that the national

Gini index falls in this closure. When labor is not permitted to move across sectors, the

poverty reduction is much smaller ­ only 900 thousand: because the economy is not

permitted to fully adjust to the new world prices, efficiency gains are blunted and the

national rise in per capita income is muted.


         The third case, in which wages are fixed and the unemployment rate is permitted

to fall in the wake of increasing labor demand, presents a particularly interesting contrast

in this paper. With increasing aggregate employment, national per capita income rises

more than in the first case with fixed employment and flexible wages. The authors point

out that the poverty outcome depends critically on who gets the new jobs. If the new jobs



                                            23

go to individuals from non-poor households, i.e. families with other wage earners or other

sources of income, the unemployment specification could worsen income inequality since

the pool of unemployed workers prevents unskilled wages from rising and, without the

benefit of higher wages, the poverty reduction would muted. In order to quantify this

outcome, the authors have estimated the likelihood that each type of unemployed

individual will obtain one of the newly available jobs. There is a considerable uncertainty

associated with these estimates, and the authors reflect this by reporting their results in

terms of the mean and standard deviation of a monte carlo simulation for each

closure/scenario. While the mean poverty reduction under the unemployment closure is

larger than that under the standard labor market specification, the standard deviations

suggest that the two are not significantly different in a statistical sense.




6.      Interactions with Tax Policies


        An important theme in many of the papers in our program is the potential for

interactions between the Doha scenarios and domestic policies to alter the poverty

outcomes obtained from multilateral trade reform. Does multilateral trade liberalization

lessen the distortions introduced by domestic commodity and factor market policies, or

does it exacerbate them? To what extent can complementary reforms of domestic policies

enhance the degree of poverty reduction? When trade liberalization results in reduced tax

revenues, how will this shortfall be made up? Two papers focus squarely on the question

of tax replacement.4




4Of course, some assumption about tax replacement is required in every analysis of trade policy reforms.


                                               24

          Christian Arnault Emini, John Cockburn and Bernard Decaluwe (2005) focus on

the case of Cameroon. They examine the poverty impacts of the central Doha scenario

paying particular attention to the structure of the domestic tax system and the different

options available for replacement of the lost tariff revenue. They view the Value-Added

Tax (VAT) as the most likely tax replacement tool in Cameroon. This tax has a very

heterogeneous impact on sectors, with effective rates ranging from zero in the case of

agriculture, to 13% in the case of petroleum refining. When they combine this tax

replacement tool with the Doha scenario, they find that poverty falls slightly, by about

22,000 people, in Cameroon, as does inequality. Of course, with relatively small tariff

cuts under the Doha scenario, tax replacement is not all that central in this scenario.


          In the case of full liberalization, tax replacement becomes much more important

and the authors consider three alternative tax scenarios in conjunction with these tariff

cuts. In every case, poverty rises, but the size of the poverty increase, as well as its

causes, vary with the choice of replacement tax. When they utilize a non-distorting

production tax, 106,000 people are estimated to be lifted out of poverty, but 193,000

formerly non-poor fall into poverty, resulting in a net poverty increase of 87,000 people.

This occurs despite an increase in aggregate welfare in Cameroon, so it is clearly a

consequence of the pattern of imports and exports in this country. When trade reform is

coupled with an increase in consumption taxes, the poverty rise is much larger ­ nearly

half a million people. This impact is lessened somewhat (a 300,000 increase) by the use

of the value-added tax to replace the forgone tariff revenue. Clearly in the case of


The "standard assumption" used is one of replacement of lost tariff revenue with an equi-proportional
(distribution neutral) income tax. While not a realistic assumption in most cases, it facilitates the
comparability of results across regions. In those cases where country authors in our program emphasized
the treatment of the domestic tax system, they were encouraged to explore the impacts of replacing the lost
tariff revenue with the most likely instrument (usually the value-added tax).


                                                     25

Cameroon, the choice of tax instrument used to replace the lost tariff revenue can be as

important as the type of trade liberalization (full liberalization vs. Doha reforms only).


        A second paper analyzing tax replacement is Caesar Cororaton, John Cockburn

and Erwin Corong (2005) on the Philippines. This is an interesting case since the

agriculture sector has evolved from net exporter to net importer over the past three

decades. As a relatively recent net food importer there is widespread concern in the

Philippines that trade reforms will jeopardize food security. However, in their analysis of

the Doha scenarios, the authors find that the national poverty headcount is barely

affected. There is a small rise in poverty among the self-employed households ­

particularly those in rural areas, while poverty amongst salaried urban workers falls.

Unlike many of the focus economies in this volume, the Doha reforms are not favorable

to Philippine agriculture, and this effect is more pronounced under full liberalization.

Because of the relatively high protection for Philippine agriculture presently, full

liberalization results in a contraction of the agricultural sector and an increase in rural

poverty. This is offset by a reduction in poverty amongst the urban population, where

wages rise. As a consequence, there is a small decline in the national poverty headcount.

However, when the authors switch from the VAT to a uniform income tax for purposes of

tariff replacement, poverty rises under the full liberalization case. Once again, the pattern

of exemptions in the indirect tax system favors the poor, and its use for purposes of tax

replacement is a critical piece of the poverty puzzle.




                                             26

7.      Cross-Country Comparisons


        With their differences in factor market closures, elasticities of substitution,

methodologies for grouping households and modeling labor markets, etc., the country

case studies described up to this point have been non-comparable. This makes it difficult

to generalize on the basis of cross-country comparisons. Therefore we included two

cross-country studies in the program: first, Maros Ivanic (2005) which provides a cross-

country comparison for 15 countries ­ each of which is treated in a symmetric manner.

While this approach is somewhat stylized, and therefore less definitive for any given

country, each of the focus country data bases has been built up from the same types of

individual household surveys as the single country case studies. Another virtue of this

paper is that is offers a fully integrated, global/national/micro modeling approach. In

particular, Ivanic has augmented the GTAP global CGE model with reconciled data on

140 disaggregated household groups for each of the 15 focus countries. His grouping is

based on income specialization, e.g., agriculture-specialized households rely almost

entirely on agricultural self-employment for their income, and similarly for a wage-

specialized stratum, etc. Because his is a global framework, he can simulate all of the

trade reform scenarios directly in his model, which also facilitates further decomposition

of the elements of trade reform and their poverty impacts.


        Ivanic's findings with respect to the poverty impacts of the Doha Agenda are

particularly interesting. Specifically, he finds that the Doha trade reform scenarios are not

as poverty-friendly as the global liberalization scenario. Now, if Doha represented the

same mix of policy reforms as full liberalization, we would expect both simulations to

have the same pattern of poverty reduction but with larger cuts under Full-Lib because of


                                              27

its deeper cuts in protection (e.g., 100% vs. 33%). However, this is not the case, and, in a

decomposition analysis, Ivanic shows why.

        The Doha Agenda as outlined by Anderson and Martin (2005) has a variety of

different elements, and these have conflicting impacts on poverty. The removal of export

subsidies in the EU and the USA tends to raise poverty in most of the developing

countries in Ivanic's sample ­ even while reducing poverty amongst the agricultural

households in these poorer countries. This is hardly surprising in light of earlier studies

highlighting the vulnerability of low-income, net food importing countries to higher

world prices for these products (e.g., Valdes and McCalla, 2004).5 Since these export

subsidies are fully removed under the Doha scenario, this impact is fully realized under

that partial reform. On the other hand, Ivanic finds that cuts in developing country tariffs

as a group have a very favorable impact on national poverty in the focus countries.6 Yet

there is very little reform of developing country tariffs under Doha ­ firstly due to limited

reciprocity (part of Special and Differential Treatment), and secondly due to the

extensive binding overhang in developing countries. Thus, while developing country

tariff cuts are among the most poverty-friendly elements of global trade reform, very little

of the beneficial impact of these reforms is felt under the Doha scenario. When

combined, these facts explain why Doha is less poverty-friendly than the comprehensive

reform scenario. It accentuates those aspects of reform that adversely affect poverty

(export subsidies), while largely omitting those aspects that benefit the poor.

5 Dimaranan, Hertel and Keeney (2004) demonstrate that many developing countries have become much
more heavily dependent on imports of subsidized crops from OECD countries over the past 40 years.
Removing these subsidies will obviously have an adverse effect in the near term.
6In Ivanic's analysis most of these gains come from improved market access to other developing countries.
This is due to the relatively high optimal tariff in the underlying GTAP model, which makes unilateral
reform relatively unattractive (see Hertel and Ivanic, 2005). An implication of this is that to reap the
benefits developing countries must liberalize together ­ i.e. that the multilateral aspect of reform is
important.


                                                    28

       This suggests that deeper cuts in developing country tariffs under the Doha

scenario might have a beneficial impact on the poverty outcome. This is explored under

the alternative scenario, Doha-All, in which developing countries fully reciprocate the

developed country reductions in tariff bindings. Ivanic shows that Doha-All does indeed

have a more favorable poverty outcome than the base Doha scenario.


       An additional finding from Ivanic's cross-section analysis pertains to the common

assumption that "a rising tide lifts all boats", i.e., that poverty rises and falls in concert

with changes in national per capita income. Ivanic shows that this is not always the case

in the near term. The reason is that trade reform generates uneven gains in the economy.

One sector gains and another loses, so it matters greatly where the poverty is

concentrated. If most of the poor reside in agriculture, and agriculture is hurt by trade

reform, poverty may rise, even if real national income rises. This is the case in Malawi,

where 40% of the population is specialized in agricultural self-employment.




8.     Effects on Productivity and Economic Growth


       Sustained reductions in poverty require economic growth, which leads naturally

to the question of how a prospective Doha Development Agenda might affect the growth

rates of countries currently experiencing the highest levels of poverty. This is a

challenging area of research ­ worthy of an entire volume in its own right ­ but the

research program contains two country case studies and a global analysis oriented

towards this theme.




                                             29

        Nabil Annabi, Bazlul Khandker, Selim Raiham, John Cockburn and Bernard

Decaluwe (2005) writing on Bangladesh focus on the growth question by emphasizing

the impact of trade reform on capital accumulation. It begins with a short run analysis in

which they find that Bangladesh experiences an aggregate loss, as well as a small rise in

poverty under the Doha scenario. Previous authors have attributed these adverse

consequences to two factors. First, Bangladesh is a net agricultural importer and suffers

from higher world prices agricultural products. Second, as a Least Developed Country

(LDC), Bangladesh currently enjoys tariff free access into many of the rich country

markets. When tariffs in these markets fall, she is expected to suffer from "preference

erosion", i.e. the value of these tariff preferences diminishes. Our analysis suggests that

the first explanation is the relevant one ­ with the main losses associated with imports of

cotton, wheat and oilseeds. We do not find evidence of preference erosion adversely

affecting the terms of trade for Bangladesh. This is because the apparel exports displaced

by erosion from the EU are absorbed in the North American market, where, de facto,

most apparel exports from Bangladesh do not enjoy preferential market access and so

Bangladesh benefits from the tariff cuts. The terms of trade losses facing Bangladesh

under Doha are magnified under full liberalization. In addition to the above, to pay for

additional imports, Bangladesh must expand the volume of her textile and apparel exports

­ which account for nearly eighty percent of export revenues. This tends to depress their

prices.


        However, these short run losses are transitory and Annabi et al. estimate that after

2 ­ 3 years, the economy will be better off under Full-Lib than under the business as

usual scenario. The reason is that the cost of investment goods will fall and increased




                                             30

investment will flow to the more competitive sectors, thereby stimulating additional

growth. They estimate that in the long run (15 years) GDP will be 1.44% higher and

poverty 6.1% lower under the Full-Lib scenario. A closer look at these results reveals that

most of the stimulus for the increased investment and economic growth comes from the

reduction in Bangladesh's own tariffs, which would be missing under the Doha scenario.


        The authors also briefly explore an issue that has received quite a bit of discussion

recently in the context of the WTO: remittances from overseas workers. They formally

explore the implications of a fifty percent increase in the flow of remittances to

Bangladesh ­ and specifically to those households currently receiving these transfers. In

return, the domestic labor supply is reduced. This development has a favorable impact on

poverty, reducing it by 0.8% in the short run and 4.0% in the long run. To the extent that

rich countries are concerned about the impact on Bangladesh of higher food prices and

preference erosion, a policy which permitted increased temporary migration appears to be

a good way to offset some of these negative effects, as the benefits of increased

remittances dominate the short run costs of trade liberalization.


        Thomas Rutherford, David Tarr and Oleksandr Shepotylo (2005) explore one of

the key trade/growth linkages in the case of Russia. They focus particularly on the

potential for international trade and foreign direct investment in the services sector to

bring new varieties of goods and new technologies to Russia, thereby enhancing her

productivity, generating economic growth, and lifting households out of poverty. The role

of services sector reforms ­ an important aspect of future WTO agreements ­ is often

neglected in analyses of trade and poverty. Yet, as Mattoo et al. (2001) demonstrate, such

reforms ­ particularly in telecommunications and financial services, can boost long run



                                            31

growth rates. The paper on Russia begins by analyzing the Doha scenario explored by

other authors. The impact of this scenario is mixed, but most of the households

experience a small welfare loss. The Full-Lib scenario shifts the distribution of welfare

impacts in the positive direction, so that most Russian households now gain and poverty

falls, but again the changes are quite modest.


        The authors then turn to domestic reforms in the services sectors ­ a part of the

economy that the Doha Agenda is not expected to affect to any great degree, but an area

which is currently receiving a great deal of attention in the context of Russia's WTO

accession negotiations. The authors show that the liberalization of barriers to Foreign

Direct Investment (FDI) greatly enhances the potential welfare gains. The main vehicle

for this enhancement is the provision of new varieties of services, which improve

productivity, not only in the services sector, but also in services-using sectors as well.

Indeed, the added productivity boost from the elimination of services FDI barriers alone

is sufficient to generate per capita income increase of 5.3%, ensuring that virtually all

Russian households benefit from the reform. There are two lessons to be drawn from this

work. First, productivity growth is essential for generating widespread gains from trade

reforms, and second, one way of obtaining such growth is through ambitious services

sector reforms, such as those that have been a part of recent WTO accession negotiations

­ most notably in China, but also now in Russia.7


        Finally, the program includes an integrated, global analysis of the potential for

multilateral trade reforms to reduce poverty in the long run (2015). Kym Anderson, Will

Martin and Dominique van der Mensbrugghe (2005) utilize the latest version of the


7Similarly dominant welfare effects from services reforms have been found in the case of China's WTO
accession agreement (Walmsley et al., 2005).


                                                 32

World Bank's Linkage model, along with the GTAP data set to project the growth path of

the global economy from 2001 to 2015. They find that trade reforms have a modest

impact on capital accumulation and thereby boost the projected global gains from

multilateral trade reform by about one-quarter. However, they devote most of their

attention to the potential impacts of increased trade on productivity growth. (It should be

noted, however, that the authors focus entirely on productivity growth associated with

increased manufactures exports ­ not services trade or investment as with the Russia

study).


       There is now a rapidly growing literature on the impacts of trade and trade policy

reforms on productivity and Anderson, Martin and van der Mensbrugghe draw on this in

their paper. When they incorporate the additional impact of openness on labor

productivity, they find a substantial boost to the global gains (40% larger gains in 2015)

with a disproportionate share accruing to the South and East Asia developing economies.

The poverty impacts of these alternative scenarios are elicited by first estimating the

income gains to the poorest households and then applying to this an estimated elasticity

of poverty reduction with respect to income growth at the poverty line. Instead of using

real per capita income for the region as a whole, they use the unskilled wage rate,

deflated by an index of food and clothing prices, reflecting the dual facts that the main

endowment of the poor is their own labor, and they spend the bulk of their income on

non-durable goods. Another critical assumption is that the poor do not pay taxes, so that

any increase in tax rates required to offset forgone tariff revenues does not affect them.


       Applying these estimates of earnings at the poverty line to the poverty elasticity

of income in each region ­ which varies depending on the regional distribution of income



                                              33

­ the authors predict the extent of poverty reduction in developing countries. Of course,

this depends on the poverty line. It also depends on the baseline poverty projections,

which decline considerably between 2001 and 20015. For $1/day poverty, the estimated

reduction in 2015, in the absence of additional productivity gains, is 2.5 million for Doha

and 31.9 million for Full-Lib. When applied to current (2001) poverty levels, the authors'

calculations result in poverty reductions of 9.7 million and 80.5 million under Doha and

Full-Lib, respectively. The 2015 poverty reductions are increased to 4.3 million and 43.5

million, for Doha and Full-Lib respectively, when productivity gains are factored in. For

$2/day poverty, the reduction in number of poor is larger, but the percentage reduction is

smaller (see Table 1).8


         Based on the Doha/Full-Lib comparison, it is clear that the (rather ambitious)

Doha scenarios capture only a relatively small portion of the total poverty reduction

possible under trade reforms. When the authors consider the Doha-All scenario, they find

that implementing deeper cuts in the developing countries enhances the poverty outcome,

nearly doubling the poverty reduction obtained under the central Doha scenario. This

finding reinforces Ivanic's (2005) conclusions with respect to the beneficial poverty

impacts of developing country tariff cuts under the Doha Development Agenda. It is also

hardly surprising in light of the increasing importance of South-South trade and the

relatively high level of developing country tariffs as reported by Anderson and Martin

(2005).


8 These estimates of poverty reduction are considerably smaller than earlier predictions using the World
Bank's Linkage model. The difference is due to the fact that these estimates are based on the most recent
(version 6) GTAP data base, which is further updated to account for EU enlargement as well as the WTO
accession of China and others. These recent trade reforms have reduced the overall level of protection
worldwide ­ thereby lessening the gains from reform. In addition, the version 6 data base has a complete
treatment of preferential tariffs ­ including the EU's 2001 Everything But Arms initiative, which means
that gains to the least developed countries from trade reform are considerably reduced.


                                                     34

        Another important finding from the Anderson, Martin and van der Mensbrugghe

paper relates to sensitive agricultural products, as well as special products in developing

countries. Industrial countries have proposed that certain sensitive products be exempt

from steep tariff reductions, instead being liberalized through a combination of quota

expansion and tariff reduction. Anderson and Martin (2005) suggest that a cut in bound

tariffs might be most effective, and they consider the case in which these commodities,

limited to 2% of industrial country tariff lines in agriculture, face a modest 15% cut in

bound tariffs. In the case of developing countries, an additional category of exemptions is

provided for in the Framework Agreement. These "special products", identified "based

on criteria of food security, livelihood security and rural development needs," will be

eligible for more flexible treatment as well (WTO, 2004). Allowing for this additional

category, the scenario outlined by Anderson and Martin permits developing countries to

exempt 4% of agricultural tariff lines from the tiered cuts, facing instead just a 15% cut in

bound tariffs.


        Of course it goes without saying that both special and sensitive products

invariably have the highest tariffs, so that exempting them can make a big difference in

the results. Indeed, the authors find that merely introducing these modest exemptions for

a maximum of 2% of the industrial tariff lines in agriculture (4% for developing

countries) virtually eliminates the poverty impacts of a Doha agreement. Therefore, in

order to have a significant poverty impact, the Doha Agenda must not only have

ambitious numerical targets, it must also seek to limit ­ indeed eliminate ­ the use of

sensitive and special product exemptions. 9



9We are not arguing here that individual developing countries could not improve their poverty outcomes


                                                 35

9.       Summary and Conclusions

         Assessing the impact of multilateral trade liberalization on poverty is a

challenging assignment. As Winters (2002, p. 43) notes: "Tracing the links between trade

and poverty is going to be a detailed and frustrating task, for much of what one wishes to

know is just unknown. It will also become obvious that most of the links are very case

specific." The research reported here is an attempt to make known a few more of these

"unknown" linkages. As such, the approach has been heterogeneous and opportunistic,

calling on experts in this field to undertake in-depth studies in countries for which

appropriate data and analytical infrastructure are available. All of this research capacity

has been directed towards the analysis of the trade policy question which is central in

many policy makers' minds today ­ namely: What are the likely poverty impacts of a

successful Doha Development Agenda? And what elements could be added to enhance

this outcome?

         As noted previously, the approach taken by the project ensures consistency of

methods in the global analysis of the multilateral trade reform scenarios ­ and also in the

methodology for incorporating these results into the national analyses. However, at the

country level, different authors have had the liberty to take a variety of approaches

depending both on the particular circumstances facing their countries and their own

analytical interests. This is why we have two studies of the Brazilian economy ­ one of

which focuses on near term impacts across heterogeneous individuals, households and


by exempting a few special products from liberalization. But given the multitude of products and countries
we cannot identify such cases here. Besides, we find it implausible that developing countries could leave
the exemptions door open in any significant fashion without industrialized countries also squeezing their
sensitive products through the same opening.


                                                  36

regions in Brazil, and one of which focuses on longer terms impacts ­ particularly in light

of the barriers to inter-sectoral labor mobility. In the case of China, we have one study

which focuses on market failure at the village level, and another which focuses on labor

mobility at the national level. Similarly there are differences in methodology taken across

countries, with a mix of partial and general equilibrium approaches, and static and

dynamic frameworks. The base years differ across studies, and even the poverty lines

chosen are not uniform across all studies. Their findings, therefore, are not strictly

comparable. Finally, since the choice of countries for the program was made on the basis

of pre-existing work that laid a foundation for the current research, this is not a random

sample of developing countries. With these qualifications in mind, let us take an

overview of the findings.

       Table 1 summarizes the poverty results from each of the national studies (sub-

national studies are not reported here) for both the Doha and Full-Lib scenarios,

distinguished by length of run for the analysis. The long term studies factor in the impact

of trade policy on investment and capital accumulation ­ and in the case of the global

analysis, productivity as well, whereas the short term studies do not. The national poverty

changes are reported in two different ways ­ first as the change in number of persons in

poverty, and second as the percentage change in the poverty headcount. Thus a negative

number in Table 1 means that the number of poor has fallen as a result of multilateral

trade reform, while a positive number indicates that the number of poor has risen.

       Table 1 suggests several tentative conclusions. First, the near term analyses are

mixed in their outcomes, with poverty rising in some cases and falling in others. We view

this diversity as correct and a strength of our country-based approach. Even setting aside




                                             37

the methodological differences between studies, the case-specificity alluded to above

leads us to expect differences between countries' interests in the DDA, and in the various

case studies the authors explain exactly why this is so.

        The largest poverty reductions in Table 1, both in absolute and relative terms, are

in countries with agricultural export potential to the markets that liberalize most (i.e., East

Asia and Europe). The strong poverty reduction in Brazil is driven by increased

agricultural production, which tends to be concentrated in regions with relatively higher

poverty incidence. In China, the poverty reduction is fueled by increased agricultural

exports to the highly protected agricultural markets of East Asia. On the other hand, the

poverty increases tend to be in countries which are net importers of agricultural products

(e.g., Bangladesh), and which may presently benefit from preferential market access (e.g.,

Mozambique). Thus the strongest difference between countries concerns their exposure

to the shocks generated by the DDA. Even holding this constant, however, poverty

impacts can vary with, for example, the degree of transmission of world prices to rural

households, the barriers to the mobility of workers between sectors of the economy, and

the incidence of national tax instruments used to replace lost tariff revenue. Taken as a

whole, the number of countries where poverty declines under the Doha scenario is about

the same as the number of countries where it falls, although looking at the absolute

number of poor, we see that poverty declines in several of the most populous countries

(Brazil, China and Indonesia) and therefore declines overall in this non-random sample of

countries.

        Turning to the long run results, we see that all of the studies that consider the

impact of trade on capital accumulation and/or productivity predict a reduction in poverty




                                             38

(with the exception of Doha-Bangladesh, where there is no long run measurable impact).

Trade stimulates investment, investment stimulates growth and growth reduces poverty.

When productivity impacts are also considered (bottom row), this effect is even stronger.

This short run/long run distinction is particularly striking in the case of the Full-Lib

scenarios for Bangladesh, where the short run impacts of trade reform translate into a rise

in headcount poverty, while the long run impacts of trade reform suggest a substantial

decline.

        In addition to the quantitative summary reported in Table 1, the research program

has generated a number of additional insights. First, the liberalization targets under the

DDA have to be ambitious if the round is to have a measurable impact on world markets

and hence poverty. Second, assuming an ambitious DDA, we have seen that the near-

term poverty impacts are likely to be mixed.


        The analysis suggests, however, that countries can enhance the impact on poverty,

by pursuing complementary domestic reforms to enable households to take advantage of

market opportunities created by the DDA.10 These include improved infrastructure and

the reform of domestic marketing institutions to improve price transmission to rural

areas, rural education reform to enhance labor mobility between the farm and non-farm

sectors, and extension outreach to permit farmers to take advantage of new export

opportunities opened up by the DDA.


        Of course, sustained poverty reduction depends on stimulating economic growth.

Here, the impact of the DDA on productivity is critical. Empirical evidence suggests that

increased merchandise trade will likely bring with it productivity gains through


10The volume edited by Harrison reaches a similar conclusion, based on a set of ex post analyses of trade
reform and poverty.


                                                 39

disciplinary effects of import competition on domestic firms as well as, possibly,

learning-by-doing on the export side. To fully realize potential productivity gains,

however, trade reforms need to be far reaching and should include reducing barriers to

services trade and investment in addition to merchandise tariffs which lie mainly or

wholly outside the DDA. Thus even if the DDA is very successful, a major agenda of

unilateral reform and further rounds of multilateral talks remains. Only through such

comprehensive reforms can long term growth and poverty reduction be ensured.




                                         40

References

Anderson, Kym and Will Martin. 2005. "Scenarios for Global Trade Reform," Chapter 2
       in Hertel and Winters (eds.), Poverty and the WTO: Impacts of the Doha
       Development Agenda (forthcoming). New York: Palgrave Macmillan.

Anderson, Kym, William J. Martin and Dominique van der Mensbrugghe. 2005. "Global
       Impacts of the Doha Scenarios on Poverty," Policy Research Working Paper
       3735, The World Bank.

Arndt, Channing. 2005. "The Doha Trade Round and Mozambique," Policy Research
       Working Paper 3717, The World Bank.

Arndt, Channing, Henning Tarp Jensen, Sherman Robinson, and Finn Tarp. 2000.
       "Marketing Margins and Agricultural Technology in Mozambique." Journal of
       Development Studies 37(1):121-37.

Balat, Jorge F. and Guido G. Porto. 2005. "The WTO Doha Round, Cotton Sector
       Dynamics and Poverty Trends in Zambia," Policy Research Working Paper 3697,
       The World Bank.

Cororaton, Caesar B., John Cockburn, and Erwin Corong. 2005. "Doha Scenarios, Trade
       Reforms, and Poverty in the Philippines: A CGE Analysis," Policy Research
       Working Paper 3738, The World Bank.

Dimaranan, Betina, Thomas Hertel and Roman Keeney, 2004. "OECD Domestic Support
       and the Developing Countries," in B. Kuha-Gasnobis (ed.) The WTO, Developing
       Countries and the Doha Development Agenda, New York: Palgrave MacMillan.

Emini, Christian Arnault, John Cockburn and Bernard Decaluwé. 2005. "The Poverty
       Impacts of the Doha Round in Cameroon: The Role of Tax Policy," Policy
       Research Working Paper 3746, The World Bank.

Ferreira Filho, Joaquim Bento and Mark Horridge. 2005. "The Doha Round, Poverty and
       Regional Inequality in Brazil," Policy Research Working Paper 3701, The World
       Bank.

Francois, J.F. and C.R. Shiells, "Applied General Equilibrium Models of North American
       Free Trade: An Introduction," in Modelling Trade Policy: Applied General
       Equilibrium Assessments of North American Free Trade. Cambridge University
       Press: Cambridge, (1994).

Harrison, Ann (editor), 2005. Globalization and Poverty, Chicago: University of Chicago
       Press.




                                           41

Harrison, G.W. Rutherford, T.F. and Tarr, D.G., and Gurgel, A;, 2003. "Regional,
        Multilateral, and Unilateral Trade Policies of MERCOSUR for Growth and
        Poverty Reduction in Brazil", Policy Research Working Paper 3051, The World
        Bank.

Hertel, Thomas W. and Maros Ivanic . 2005. "Assessing the World Market Impacts of
        Multilateral Trade Reforms," Chapter 3 in Hertel and Winters (eds.), Poverty and
        the WTO: Impacts of the Doha Development Agenda (forthcoming). New York:
        Palgrave Macmillan.

Hertel, Thomas W., Maros Ivanic, Paul V. Preckel, and John Cranfield, 2004. "The
        Earnings Effects of Multilateral Trade Liberalization: Implications for Poverty",
        World Bank Economic Review 18(2):205-236.

Hertel, Thomas W. and Will Martin, 2000. "Liberalizing Agriculture and Manufactures
        in a Millennium Round: Implications for Developing Countries," World Economy
        23: 455-470.

Hertel, Thomas W. and L. Alan Winters, 2005. Poverty and the WTO: Impacts of the
        Doha        Development       Agenda,      World        Bank;       also      on
        https://www.gtap.agecon.purdue.edu/poverty

Jorgensen, D.W., 1984.      "Econometric Methods for Applied General Equilibrium
        Analysis". In: Scarf, H.E., Shoven, J.B. (Eds.), Applied General Equilibrium
        Analysis. Cambridge: Cambridge University Press.

Kehoe, Timothy J.2005. "An Evaluation of the Performance of Applied General
        Equilibrium Models of the Impact of NAFTA," in Timothy J. Kehoe, T. N.
        Srinivasan, and John Whalley, editors, Frontiers in Applied General Equilibrium
        Modeling: Essays in Honor of Herbert Scarf, Cambridge University Press, 2005,
        341-77.
Kuiper, Marijke and Frank van Tongeren. 2005. "Growing Together or Growing Apart?
        A Village Level Study of the Impact of the Doha Round on Rural China," Policy
        Research Working Paper 3696, The World Bank.

Mattoo, A., R. Rathindran, A. Subramamian, 2001. "Measuring Services Trade
        Liberalization and its Impact on Economic Growth", Policy Research Working
        Paper No. 2655, Washington, DC: The World Bank.

McKitrick, R.R., 1998. "The Econometric Critique of Computable General Equilibrium
        Modeling: The Role of Functional Forms". Economic Modeling. 15(4), 543-573.

Nicita, Alessandro. 2004. "Who Benefited from Trade Liberalization in Mexico?
        Measuring the Effects on Household Welfare." World Bank Working Papers
        3265.




                                           42

Nicita, Alessandro. 2005. "Multilateral Trade Liberalization and Mexican Households:
        The Effect of the Doha Development Agenda," Policy Research Working Paper
        3707, The World Bank.

Robilliard, Anne-Sophie and Sherman Robinson. 2005. "Social Impact of a WTO
        Agreement in Indonesia," Policy Research Working Paper 3746, The World
        Bank.

Valdes, A. and A. McCalla, 2004. "Where the Interests of Developing Countries
        Converge and Diverge", Chapter 7 in M. Ingco and L.A. Winters (eds.),
        Agriculture in the New Trade Agenda, Cambridge University Press.

Walmsley, T., T. W. Hertel, and E.I. Ianchovichina 2005. "Assessing the Impact of
        China's WTO Accession on Investment",                Pacific Economic Review
        (forthcoming).

Winters, L.A. (2000) "Trade, Trade Policy, and Poverty: What are the Links?" World
        Economy 25(9), 1339-67.

Winters, L.A., N. McCulloch, and A. McKay. (2004) "Trade Liberalisation and Poverty:
        The Evidence So Far." Journal of Economic Literature XLII(March), 72-115.

World Bank. 2004. Global Economic Prospects 2004 ­ Realizing the Development
        Promise of the Doha Agenda. Washington, D.C.: The World Bank.

World Trade Organization (WTO). 2004. Doha Work Programme: Decision Adopted by
        the General Council on 1 August, 2004, Geneva: World Trade Organization.

Zhai, Fan and Thomas Hertel. 2005. "Impacts of The Doha Development Agenda On
        China: The Role Of Complementary Education Reforms," Policy Research
        Working Paper 3702, The World Bank.

Zhai, Fan and Thomas Hertel. 2005. "Impacts of The Doha Development Agenda On
        China: The Role Of Complementary Education Reforms," Policy Research
        Working Paper 3702, The World Bank.




                                           43

Box 1.1 Elements of the DDA Scenario based on July Framework Agreement

· Agriculture:
  - Market access: 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: marginal rates (35, 40, 50, 60%) change at 20, 60, 120% tariffs
    LDCs: no cuts to tariffs
 - Aggregate Measure of Support: apply tiered formula:
    For developed: marginal rates of 60% (AMS less than 20%) and 75%
    For developing: marginal rate of 40%
    LDCs: no cuts to domestic subsidies
 - Export subsidies abolished

· Non-Agriculture Market Access: 50% cuts in tariffs (33% developing, 0% LDC)




                                           44

Table 1. Poverty Impacts of a Prospective Doha Development Agenda


                                                          Change in Poverty Headcount

Country (Chapter No.)               Near Term: Fixed Capital                  Long Term: Investment Impacts

                                    Doha                  Full-Lib               Doha                  Full-Lib
                                1,000          %      1,000         %        1,000          %        1,000      %
Bangladesh (15)                    38        0.3      1,354        1.1           0          0       -5,758     -4.6

Brazil (7)                       -236       -0.4        -482      -0.8

Brazil (9)                                                                   -380        -1.1       -1,030     -2.9

Cameroon (12)                     -22       -0.4        303        4.8

China (10)                     -4,590       -1.1     -8,271       -2.0     -5,378        -1.3      -11,170     -2.7

Indonesia (11)                    -48       -0.1     -1,384       -3.5

Mexico (4)                          4        0.0        127        1.0

Mozambique (5)                     27        0.3          60       0.6

Philippines (13)                   12        0.0          -7       0.0

Russia (16)                       209        0.9        -122      -0.5

All Developing (17)

$1/day: 2001*                  -7,000               -66,300                -9,700                  -80,500
       2015**                  -1,700       -0.3    -23,800       -3.8     -2,500        -0.4      -31,900     -5.1

$2/day:2001                    -8,700               103,900               -12,600                 123,200
       2015                    -4,100       -0.2     52,300       -2.7     -6,200        -0.3      -65,600     -3.3

Productivity Effects Added***

$1/day: 2001                                                               -20,400               -126,500

       2015                                                                 -4,300       -0.6      -43,500     -6.5

$2/day: 2001                                                               -29,600               -193,200

       2015                                                                -12,100       -0.6      -94,700     -4.9
     *Based on percentage changes in 2015, but applied to 2001 poverty headcount.

     **Computed for the year 2015 when the total number of poor is projected to be significantly lower.

     ***Productivity gains from increased openness to trade apply to both manufactures and agriculture.
     (Earlier versions only assumed productivity gains in manufactures ­ see Chapter 17 for details.)




                                                         45