Would freeing up world farm trade reduce or increase poverty?

Kym Anderson, John Cockburn, William Martin 28 April 2010

a

A

Trade policy reforms in recent decades have sharply reduced the distortions that were harming agriculture in developing countries. Yet global trade in farm products continues to be far more distorted than trade in nonfarm goods, and model results suggest removing the remaining distortions would put upward pressure on food prices in international markets. Economists argue that such multilateral reform is welfare-improving globally, since its market opening leads to a more efficient location of farm production. It would also improve economic welfare nationally for most countries, the exceptions being countries that have almost no distortionary policies of their own to reform and that are sufficiently dependent on imports of farm products as to suffer from a worsening in their international terms of trade.

Until recently it has also been widely argued that such an increase in farm product prices would reduce poverty in developing countries. This claim is premised on the fact that three-quarters of the world’s poor live in rural areas, with the majority of them depending directly or indirectly on agriculture for their livelihoods (World Bank 2007). However, as the international price of food rose during 2007 and spiked severely in 2008, key development-focused agencies claimed that this too would be harmful to the poor (see the many citations in Swinnen 2010).

How can both an increase and a decrease in international prices of farm products be bad for the poor? One step towards reconciling these two stances is to recall that a rise in those international prices, if associated with multilateral trade liberalisation, is a consequence of phased reductions in national import restrictions that typically reduce domestic prices over time in all but the least protective countries as protection is cut. By contrast, when prices spike as they did in 2007-8 the causes tend to be one-off exogenous shocks to output or demand during periods of low stock levels. In such situations a reasonable estimate of the poverty impact might be obtained by considering just the short-run impact of a change in international prices before farmers have time to respond. In Ivanic and Martin (2008), net buyers of food are made worse off in the short term, while any gains to net sellers of food, for example, come only after farmers have had time to respond.

Whether poverty rises or falls when trade is liberalised via unilateral or multilateral reform is clearly an empirical question. To answer it requires first being aware of the key ways in which economic welfare of an individual or household is affected by a price shock, and then undertaking quantitative analysis that captures those various mechanisms appropriately. The latter requires using economy-wide models with up-to-date price distortion data and ideally detailed household information on the earning and spending profiles of different groups of people, both rural and urban. Also, it is important to allow for supply to adjust to long-run price changes, such as those brought about by trade policy reforms. In a new CEPR Discussion Paper (Anderson, Cockburn and Martin 2010a) we outline the mechanisms at work and then summarise the results of a set of single- and multi-country simulation studies in a bid to address the question of whether agricultural trade liberalisation will increase or reduce poverty.

Elements of the puzzle

Many farm households in developing countries rely on the farm enterprise for most of their income, and in the world’s poorest countries the share of the poor concentrated in such households is very large. The fact that many of the poorest households in the poorest countries are dependent on agriculture means a rise in international farm product prices, other things equal, may reduce poverty in developing countries. But this outcome is not certain for at least six reasons.

  • First, poor households also spend the majority of their income on staple foods, so if food prices rise then this adverse effect on household expenditure of farm households may more than offset the beneficial effect of higher farm earnings, depending on whether the farm household is a net buyer or net seller of food staples (keeping in mind that plenty of poor farmers grow cash crops whose price might not rise when food prices rise). As for poor non-farm households, they are seriously affected by a hike in the consumer price of food but may be affected very little by a rise in the price of non-food farm products.
  • Second, landless farm labourers and the urban unskilled and under-employed also are among the extreme poor, and they would be hurt directly by a rise in consumer prices of staple food since food accounts for the vast majority of their spending.
  • Third, in addition to these direct effects there is also an important indirect effect. It is possible that the product price change may raise the demand for unskilled labour.
  • Fourth, another important indirect effect is how different households are affected by any change in taxation that might accompany the change in trade tax revenue associated with the shock. If, for example, a multilateral trade reform lowers tax revenues, does the government make up the revenue loss through a consumption tax that hits all households, or an income tax that hits only the better-off?
  • Fifth, the domestic producer prices of farm products do not necessarily change to the same extent as international prices, either because policies are retained that insulate the country from changes in world market prices, or because of big differences between domestic and imported goods.
  • And sixth, do developing country governments alter their duties in line with the rest of the world? If countries lower their own protective barriers, domestic prices may fall even if world prices are increasing. It also matters whether the trade policy reform includes the removal of barriers to non-farm goods. Removing assistance to heavy manufacturing, for example, may allow light manufacturing to expand and provide more opportunities for poor unskilled farm workers to get higher-paid work outside agriculture.

Empirical estimates of poverty effects of trade reform

Our approach is a variant on the pioneering work by Hertel and Winters (2005 and 2006). We examine reforms that individual developing countries might implement unilaterally, as well as multilateral trade reform by the rest of the world; and we distinguish between agricultural and other trade policies. In doing so, our study draws on both global and national economy-wide models.

What do the modelling results show? Global linkage model results suggest that removing the world’s price and trade distortions in place in 2004 would reduce international inequality, largely by boosting net farm incomes and raising real wages for unskilled workers in developing countries, and after full adjustment would reduce the number of poor people worldwide by 3% (bottom right of Table 1).

Table 1. Effects of full global merchandise trade liberalisation on the incidence of extreme poverty (<$1 a day) using the Linkage model

 

 
Average unskilled
wage change, reala
(%)
Baseline headcount
Change in number of poor
from baseline levels
 
(% of population)
 
 
 
million
 %
East Asia
4.4
9
 
-17
-10.3
China
2.1
10
 
-5
-4.0
Other East Asia
8.1
9
 
-12
-30.1
South Asia
-1.9
31
 
8
1.8
India
-3.8
34
 
15
4.2
Other South Asia
4.0
29
 
-8
-9.9
Sub Saharan Africa
5.3
41
 
-11
-3.8
Latin America
4.1
9
 
-3
-6.8
Middle East & North Africa
14.3
1
 
-2
-36.4
 
 
 
 
 
 
Developing country total
5.9
18
 
-26
-2.7
Developing excl. China
6.5
21
 
-21
-2.5
 
 
 
 
 
 
East Europe & Central Asia
4.5
1
 
-0
-6.8

 Note: Nominal unskilled wage deflated by the food and clothing CPI. Source: Linkage model simulations from Anderson, Valenzuela and van der Mensbrugghe (2010).

An additional global liberalisation analysis based on the GTAP (Global Trade Analysis Project) model finds even larger reductions in poverty if the poor are exempt from the increased income taxation needed to compensate for the loss of trade tax revenue from removing trade barriers (Table 2). That full global reform reduces the poverty rate by roughly one-quarter in Thailand and Vietnam, for example. Even if the poor are not exempt from the increased income taxation needed to maintain government revenue, it is only in relatively affluent Mexico and Venezuela among that study’s sample of 15 developing countries that the number of poor would not fall (Table 2). One reason those two countries (and India in Table 1) are different is that farmers there receive substantial support (and in Mexico’s case some preferential market access) whose removal would harm poor farmers more than the removal of other distortions around the world help them, at least in the absence of compensation.

Table 2. Effects of full global liberalisation of agricultural and all merchandise trade on the incidence of extreme poverty using the GTAP model (percentage point change using $1 a day poverty line)

 

 
 
Default tax replacement (poor are not exempt)
Alternative tax replacement (poor are exempt)
 
Agriculture-only reform
 
All merchandise reform
All merchandise reform
Asia
 
 
 
 
Bangladesh
-0.3
 
0.3
-5.3
Indonesia
-1.1
 
-0.6
-5.2
Philippines
-1.4
 
-1.0
-6.4
Thailand
-11.2
 
-10.3
-28.1
Vietnam
-0.5
 
-5.7
-23.6
Africa
 
 
 
 
Malawi
-1.6
 
-1.9
-5.6
Mozambique
-1.2
 
-1.0
-4.3
Uganda
-0.0
 
0.1
-6.0
Zambia
-0.0
 
0.1
-2.0
Latin America
 
 
 
Brazil
-2.5
 
-2.2
-10.0
Chile
-4.8
 
-4.6
-12.3
Columbia
-0.7
 
-0.1
-4.1
Mexico
0.8
 
1.1
-0.5
Peru
-0.6
 
-0.8
-5.2
Venezuela
0.2
 
0.9
-2.1
Unweighted averages:
 
 
 
 -Asia
-2.9
 
-3.5
-13.7
 -Africa
-0.7
 
-0.7
-4.5
 -Latin Amer
-1.3
 
-1.0
-5.7
 -All 15 DCs
-1.7
 
-1.7
-8.0
 
 
 
 
 

Source: GTAP model simulations from Hertel and Keeney (2010).

The project also includes ten stand-alone national economy-wide case studies. The detailed results in Table 3 are disaggregated to show the effects of unilateral versus global reform, and agricultural versus all merchandise trade reform. In the case of the Philippines, this reveals that reform by the rest of the world would be poverty reducing for that country, but removing its own policies would add a little to poverty. This is also the case In China, in part because agricultural policies are now slightly assisting farmers there. On average nearly two-thirds of the poverty alleviation in these countries is due to non-farm trade reform, with the important exception of Brazil where agricultural reform is the major contributor to its large pro-poor outcome. The latter result is a consequence of the increase in demand for unskilled labour following liberalisation, which outweighs the poverty impact of removing tariffs protecting Brazil’s poor import-competing farmers. The contribution of own-country reforms to the fall in poverty appears to be equally as important as rest-of-world reform on average (bottom row of Table 3).

Table 3. Impact of reform on the incidence of extreme poverty (percentage point change using national or $1 a day poverty line)

(a) rural poverty

 

 
Base
Agriculture-only reform
All merchandise reform
 
(%)
Unilateral
 
Global
Unilateral
 
Global
 
 
 
 
 
 
 
 
China($2/day)
58
0.3
 
-1.1
0.5
 
-1.4
Indonesia
29
0.1
 
-1.1
-0.1
 
-4.4
Pakistan
38
-1.4
 
-1.5
-7.6
 
-8.6
Philippines
49
0.0
 
-0.3
0.6
 
-0.1
Thailand
30
0.3
 
-1.3
-3.5
 
-4.4
 
 
 
 
 
 
 
 
Mozambique
36
-1.6
 
-1.6
-2.1
 
-3.6
South Africa
17
-0.3
 
-0.7
-1.1
 
-1.4
 
 
 
 
 
 
 
 
Argentina
 
 
 
 
 
 
 
Brazil
 
 
 
 
 
 
 
Nicaragua
63
-0.7
 
-0.4
-1.3
 
-1.3

 

(b) total poverty

 

 
Base
Agriculture-only reform
All merchandise reform
 
(%)
Unilateral
 
Global
Unilateral
 
Global
 
 
 
 
 
 
 
 
China($2/day)
36
0.2
 
-0.6
0.3
 
-0.9
Indonesia
23
-0.0
 
-0.8
-0.1
 
-3.6
Pakistan
31
-1.6
 
-1.8
-5.2
 
-6.4
Philippines
34
0.4
 
-0.1
1.1
 
0.1
Thailand
14
0.1
 
-0.8
-3.4
 
-4.1
 
 
 
 
 
 
 
 
Mozambique
36
-1.3
 
-1.3
-1.7
 
-3.1
South Africa
10
-0.2
 
-0.5
-0.8
 
-1.1
 
 
 
 
 
 
 
 
Argentina
 
 
 
 
 
 
 
Brazil
31
-0.5
 
-2.8
-0.9
 
-3.5
Nicaragua
41
-0.1
 
-0.3
-1.0
 
-0.4
 
Unweighted averages:
 
 
 
 
 
 
 -Asia
28
-0.2
 
(-2.9)-0.8
-1.5
 
(-3.5)-3.0
 -Africa
32
-0.8
 
(-0.7)-0.9
-1.3
 
(-0.7)-2.1
 -Latin Am.
36
-0.3
 
(-1.3)-1.6
-1.0
 
(-1.0)-2.0
 -All 9 DCs
43
-0.4
 
(-1.7)-1.0
-1.3
 
(-1.7)-2.6

Source: Country case studies in Anderson, Cockburn and Martin (2010b) plus (in the case of the unbolded numbers in brackets in the final 4 rows), from Hertel and Keeney (2010) as reported in the last 4 rows of Table 2 above.

Policy implications

It would be surprising if all the cases examined in this study came to the same conclusion, but there is a high degree of similarity in the most important sign: they estimate that national poverty will reduce if barriers to global merchandise trade are removed.

These empirical findings have a number of policy implications. First and foremost, the winners from trade reform would include poorer countries and the poorest individuals within countries. Nevertheless, it is also clear that even among the extreme poor, some would lose. This illustrates the merit of compensatory policies, ideally ones that focus not on private goods but rather on public goods with high rates of return, such as investments in agricultural research.

Second, while there are more-direct and hence more-efficient domestic policy instruments that could meet the poverty and hunger Millennium Development Goals, these alternatives are generally more of a drain on treasury finances. This is a problem for governments of low-income countries that still rely heavily on trade tax revenue, but it could be eased by expanded aid-for-trade funding as part of official development assistance programs.

Finally, our finding from most of the national case studies that domestic reform on its own can not only boost aggregate national income but also reduce poverty and inequality suggests developing countries should not hold back on domestic reforms while negotiations in the World Trade Organisation’s Doha Round and other international accords continue. It also suggests that developing countries have little to gain, and potentially much to lose from a poverty alleviating perspective, from negotiating exemptions or delays in national reforms in the framework of WTO multilateral agreements.

Disclaimer: A product of a World Bank research project on Distortions to Agricultural Incentives, the views expressed here are the authors’ alone and not necessarily those of the World Bank and its Executive Directors, nor the countries they represent, nor of the institutions funding this research.

References

Anderson, K, J Cockburn and W Martin (2010a), “Would Freeing Up World Trade Reduce Poverty and Inequality? The Vexed Role of Agricultural Distortions”, CEPR Discussion Paper 7749, March.
Anderson, K, E Valenzuela and D van der Mensbrugghe (2010), “Global Poverty Effects of Agricultural and Trade Policies Using the Linkage Model”, Chapter 2 in Anderson, K, J Cockburn and W. Martin (eds) (2010b), Agricultural Distortions, Inequality and Poverty, Washington DC: World Bank.
Hertel, TW and R Keeney (2010), “Inequality and Poverty Impacts of Trade-related Policies Using the GTAP Model”, Chapter 4 Anderson, K, J Cockburn and W. Martin (eds) (2010b), Agricultural Distortions, Inequality and Poverty, Washington DC: World Bank.
Hertel, TW and LA Winters (2005), “Estimating the Poverty Impacts of a Prospective Doha Development Agenda”, The World Economy 28(8):1057-1071, August.
Hertel, TW and LA Winters (eds.) (2006), Poverty and the WTO: Impacts of the Doha Development Agenda, London: Palgrave Macmillan and Washington DC: World Bank.
Ivanic, M and W Martin (2008), “Implications of Higher Global Food Prices for Poverty in Low-Income Countries”, Agricultural Economics 39:405-416.
Swinnen, JFM (2010), “The Right Price of Food: Reflections on the Political Economy of Policy Analysis and Communication”, LICOS Discussion Paper 259/2010, Katholieke Universiteit Leuven, April.
World Bank (2007), World Development Report 2008: Agriculture for Development, Washington DC: World Bank.

 

a

A

Topics:  International trade Poverty and income inequality

Tags:  protectionism, Global poverty, agricultural trade

George Gollin Professor of Economics and Foundation Executive Director of the Centre for International Economic Studies (CIES) at the University of Adelaide and CEPR Research Fellow

Co-director of the Poverty and Economic Policy (PEP) research network and Professor at Laval University in Québec

Senior Research Fellow, IFPRI

Events

CEPR Policy Research