The impact of trade policy on poverty, food security and inequality in developing countries is at the centre of a crowded international debate on the role of international trade in development. The current Doha Round of WTO negotiations makes development and poverty impacts a top priority. In addition, the Millennium Declaration underscores the importance of international trade in the context of development and the elimination of poverty.
Increased openness to international trade is unlikely, on its own, to lead to major improvements in economic growth, poverty reduction, or food insecurity. Complementary policies and investments in infrastructure, human capital and safety nets are crucial if trade liberalization is to support food security strategies. Nevertheless, participation in international trade does allow countries access to larger markets for their products. At the same time, trade also provides access to larger and less expensive food supplies than does reliance on domestic production alone. International trade can also be a powerful channel for technology transfer that expands the potential for productivity increases in farming, agro-processing and retailing.
For example, with the help of international trade opportunities, China is reallocating land resources and moving out of grains production into labour-intensive production in which it has a comparative advantage. Between 1991 and 2003, the proportion of land allocated to fruit and vegetable production increased from 8 percent to 18 percent, while area planted to grains, cotton and sugar crops fell. Export opportunities and domestic demand are encouraging rapid expansion both in vegetable production and in the supply chains that move the products to consumers. Between 2000 and 2004, fruit and vegetable exports increased from $3.6 to $6.4 billion (FAO, 2006a). China now supplies around three quarters of the Republic of Korea's fresh vegetable imports, and its share of Japan's fresh vegetable imports increased from less than 10 percent in 1989 - 1991 to 37 percent in 2002 - 2004 (Huang and Gale, 2006). China's comparative advantage in vegetable production has attracted foreign investment, reorganized supply chains and linked producers in China with supermarkets in Japan and the Republic of Korea. It takes 21 days for onions to reach Japan from the United States of America, but only 3 days from China (Chen, Chen and Shi, 2003).
But comparative advantage varies tremendously from crop to crop, and China is also a source of demand for other agricultural products. Thailand concluded a free trade agreement with China in 2003 for various agricultural products, and has been running a large trade surplus for these commodities. Thailand exports cassava, longans, durian and prawns to China, while importing apples, pears, mushrooms, carrots and nuts (Hindu, 2005).
Trade reforms affect the economy through both price and income effects. Changes in the relative prices of commodities will affect consumers and producers differently, and this omnipresent tension is at the heart of many food policy debates. In trying to understand the effects of trade liberalization, it is important to consider the interests of all those who are poor or food insecure, not just farmers or just consumers.
Linkages between trade and poverty include several distinct components, with the first linkage at the border. The border price can be affected by domestic trade reforms, e.g. a lowering of tariffs, or liberalization in other countries. From the border, the focus moves to how prices transmit to producers and consumers, and to households in general. The extent to which households and businesses in the economy experience these price changes is quite varied, and depends on the quality of infrastructure and the behaviour of domestic marketing margins, as well as geographic factors. The empirical literature suggests that the degree of price transmission from the border to the local market can vary widely from commodity to commodity, even within a single country.
The initial impact of trade liberalization on households occurs once the local market price changes are determined. Not surprisingly, households that are net sellers of products whose prices rise, in relative terms, benefit in this first round. Net purchasers of such goods lose. However, the empirical literature demonstrates that first round effects can be altered as households adjust consumption and production in response to changing relative prices. In this second round of effects, households modify their consumption basket, adjust working hours and possibly change their occupation. For example, Friedman and Levinsohn (2002) showed that welfare losses from the financial crisis in Indonesia were reduced by about 50 percent after taking account of consumer substitution among goods and services. Wages in labour markets can also be affected, as described below.
The effects of trade reforms on poverty and food security are complex, and will vary depending on social structure (e.g. the proportion of landlessness in the rural population, inequality of land ownership), infrastructure, the specific commodity involved (e.g. whether it is preferentially consumed by the poor or well to do, whether it is being imported or exported in any particular country) and many other factors. This complexity means that analyses of specific countries and commodities are highly valuable, although some generic lessons can be gleaned from the studies described briefly below.
For example, the elimination of all forms of support and protection to agriculture by the OECD countries is expected to increase border prices of temperate-zone agricultural products about 5 to 20 percent (FAO, 2005a). If such reforms are realized (the outcomes of the Doha Round so far indicate limited movement in that direction), higher commodity prices will benefit countries with a comparative advantage in agriculture. While some developing countries in Asia are net exporters of agricultural products, many are net importers. These latter countries are probably benefiting from OECD subsidies. This is not to say that these farm subsidies make an important contribution to poverty reduction, which is a doubtful proposition because most of the funds benefit wealthy farmers in the OECD countries. The large amounts of money spent on these subsidies could be used for better purposes, including well targeted development assistance. The subsidies also create an uneven playing field for poor farmers in developing countries. Finally, these subsidies make progress on trade negotiations in other sectors more difficult.
Even net exporters with a comparative advantage in agriculture will need to compete with land surplus countries throughout the world (e.g. in South America) in order to gain lasting benefits from the hypothetical elimination of OECD support, as technological progress may contribute to a continued downward trend in international commodity prices even if major trade reforms occur. To further complicate matters, net agricultural exporting countries are net importers of many commodities, and conversely, net agricultural importing countries are net exporters of many commodities.
Thus, elimination of OECD support and protection will have complex effects, and it is hardly a panacea for poor farmers and consumers in developing countries. Indeed, the review of the empirical evidence on trade liberalization in FAO (2005a; chapter 4) suggests that the largest gains to developing countries tend to come from their own trade-liberalizing measures and domestic reforms. Thus, the discussion below will focus on understanding the effects of domestic trade reforms.
Studies of the rice sector in Viet Nam (Minot and Goletti, 2000) indicate that higher rice prices help to alleviate poverty and food insecurity in this particular context.
Liberalization of export quotas allows farmers greater access to foreign consumers and serves to raise rice prices within the country, as the supply available for domestic consumption is reduced. Viet Nam has a comparative advantage in rice production because of its large river deltas and is thus a net exporter of rice (in fact, all four quartiles in the income distribution are net producers of rice), so higher prices have a positive effect on aggregate household income, similar to the effect of a positive terms of trade shock (this is in addition to the increased income from the removal of market distortions). As a result, export liberalization was estimated to reduce the number of poor by a net 5 percent. The positive effect would have been larger except that, even in rice-exporting Viet Nam, there are large numbers of poor who do not produce a rice surplus, even in response to the higher prices, and are thus hurt by the reforms.
As noted earlier, "second round effects" are important in understanding the effects of changes in trade policies. For example, the Viet Nam study took into account the fact that higher rice prices stimulate rice production and cause some households that are net buyers of rice before the reforms to become net sellers of rice after the reforms. Another "second round effect" that can cause net consumers of food to benefit from higher food prices works through labour markets. Higher agricultural prices, by stimulating the demand for unskilled labour in rural areas, can result in a long run increase in rural wages, thereby benefiting wage labour households in addition to self-employed farmers.
Ravallion (1990), using a dynamic econometric model of wage determination and data from the 1950s to the 1970s, concludes that the average landless poor household in Bangladesh loses from an increase in the rice price in the short run (because of higher consumption expenditures), but gains slightly in the long run (after five years or more). This is because in the long run, as wages adjust, the increase in household income (dominated by unskilled wage labour) is large enough to exceed the increase in household expenditures on rice. However, this study used relatively old data, when rice farming was a larger sector of the economy and thus had a more profound impact on labour markets. Rashid (2002), using co-integration techniques and updating the data used by Ravallion (1990), found that since the mid-1970s, rice prices in Bangladesh no longer have a significant effect on agricultural wages. Thus, if higher rice prices no longer induce higher rural wages in Bangladesh, where agriculture is a larger share of the economy and rice dominates the agricultural sector to a greater extent than in most other Asian countries, it seems that higher grain prices are unlikely to stimulate the rural labour market in economies that have a more diversified range of employment opportunities.
Higher agricultural prices can help to reduce poverty in some circumstances, as shown in the case of Viet Nam above. But high food prices may not have the same benefits in food importing countries with different social structures, especially when those higher prices are the result of trade barriers that distort geographical patterns of comparative advantage. For example, in the Philippines there is a large class of landless labourers because of greater inequality in land distribution than in Viet Nam. In addition, corn is also an important crop in the Philippines, and corn farmers are the poorest class of farmers in the country. Thus, the poorest twenty percent of the income distribution buys more rice than it sells, and this is especially pronounced for the bottom ten percent (Balisacan, 2000; by contrast, in Viet Nam, the poorest quarter of the population is a net producer of rice). Thus, in terms of first round effects, higher rice prices are likely to hurt the poor in this context. This negative distributional effect of high rice prices is compounded because the high rice prices result from import restrictions that distort relative prices, lead to a misallocation of resources in the economy and reduce the level of national income (known as "deadweight losses" in economics jargon). In the case of Viet Nam, the reforms that led to higher rice prices removed distortions, thus reducing deadweight losses. An analysis of Indonesia by Warr (2005) also concludes that import barriers to rice increase poverty in that country.
Because the share of agriculture in the Philippine economy is relatively low, and because rice occupies a lower share of crop area harvested in the Philippines than in Bangladesh, it is unlikely that higher rice prices raise wages for unskilled labourers in the Philippine context. Furthermore, the main substitute crops that Philippine farmers are likely to plant instead of rice are corn and vegetables. Corn has about the same labour intensity as rice, but vegetables are substantially more labour-intensive. Thus, favouring rice at the expense of vegetables may reduce labour demand and worsen the impact on the poor (Dawe, Moya and Casiwan, 2006).
The contrasting examples of Viet Nam and the Philippines, and even Bangladesh at different points in time, show why it is not possible to come up with simple "one size fits all" rules about the effects of food prices or trade liberalization on poverty, and highlight the importance of detailed situation specific analysis.
Price protection for agriculture (e.g. import restrictions on competing products, output price support, input subsidies) can be justified on efficiency grounds if there is a technology that is profitable at long-run prices but has not been adopted by farmers who are risk averse or do not have sufficient knowledge of the technology. In these circumstances, subsidies can spur farmers to take risks or acquire knowledge that will lead to adoption. But understanding when such market failures exist and are causing serious resource misallocation is not necessarily easy, as there are market failures in many parts of the economy. Furthermore, tilting the terms of trade in favour of agriculture simultaneously tilts the terms of trade against other sectors, which then presumably discourages adoption of promising new technologies in those sectors.
In the absence of a key profitable technology waiting to be adopted, price protection for agriculture (or any other sector) simply redistributes income from one group to another. Such redistribution can have implications for both equity and growth. It obviously affects equity, but, at least in the Asian context, it is not quite so simple as redistributing income from relatively well off urban dwellers to relatively poor rural residents. The complexity arises because of the large pools of functionally landless labourers in countries such as India, Indonesia, Bangladesh and the Philippines. These landless labourers are usually the poorest of the poor, do most of the production work and are net consumers of food. Thus, many poor people in rural areas would be hurt by higher food prices caused by protection. Unfortunately, nearly all of the popular trade liberalization debate that is carried out in a global context focuses on farmers, to the exclusion of the landless poor in rural areas.
Price protection also has implications for hunger and food security. For example, Block et al. (2004) found that increased rice prices in Indonesia during the financial crisis caused mothers in rural Central Java to reduce the intake of nutritious foods in order to balance the food budget, while maintaining consumption of rice at relatively constant levels. This decision to sacrifice consumption of nutritious foods led to increased prevalence of anaemia in both mothers and children. In addition, mothers buffered the caloric intake of their children, resulting in increased maternal wasting. The negative effects on anaemia were less, however, for mothers who had higher levels of nutrition knowledge and thus reduced the consumption of nutritious foods to a lesser extent (Block, 2004). This shows the importance of complementary reforms to increase knowledge that allow households to make more informed decisions.
Price policy can also have implications for economy-wide growth if it redistributes income in favour of those with a higher propensity to consume domestically produced goods (as opposed to imports), thus raising aggregate domestic demand. But again the existence of landless labourers complicates the matter beyond a simple comparison of rural and urban or agriculture and industry, because the rural landless most likely have an even greater marginal propensity to consume out of domestic production than do farmers, who in turn are less likely to consume imports than urban dwellers.
Price protection may also have the potential to lock farmers into production of a crop with little growth potential, thus discouraging productivity growth in the economy, which is ultimately the only sustainable way to alleviate poverty and food insecurity. Indeed, the Philippines and Sri Lanka, the two developing countries where protection of rice has been the most pronounced for a long period of time, have had the lowest growth rates of agricultural output per worker among major rice-producing Asian countries since 1980.
A further consideration is that high food prices caused by protection for grains may reduce the competitiveness of unskilled labour via a wage good argument. This could slow growth and hinder movement of the economy towards sectors in which it has a medium-term comparative advantage, which for many developing countries may be in the realm of labour-intensive manufactures (or crops). Of course, there are many factors that affect economic growth, but it is interesting to note that the most dynamic developing countries in Asia seem to have followed a policy of low staple food prices, while the more stagnant economies have suffered from relatively high food prices.
For example, China, Thailand and Viet Nam have all avoided price protection for rice for decades, during which time all experienced quite rapid growth. Indonesia did the same (on average) during its period of rapid growth that came to an abrupt end with the onset of the Asian financial crisis. During the past few years in Indonesia, rice prices have been above import parity levels, although clearly this is not the main reason for the slowdown in economic growth. On the other hand, real GDP per capita in the Philippines has been stagnant for the past 20 years, during which time rice prices have been highly protected and well above levels in neighbouring developing countries. Sri Lanka has also had high levels of protection for its rice sector during this time. Its growth performance has been substantially better than that of the Philippines, with a doubling of per capita income from 1980 to 2002. This performance was not nearly as strong, however, as that of China, Thailand, Viet Nam and, before the crisis, Indonesia.
While agricultural protectionism has many difficulties associated with it, there may be circumstances in which it is justified. For example, what if the agricultural sector of a particular country is not competitive at world prices for any crops? The theory of comparative advantage states that every country must have a comparative advantage in something, given appropriate exchange rates. But that comparative advantage need not be found in the agricultural sector. If in fact a country has no comparative advantage in agriculture, how does one go about moving all the workers into other sectors? Agriculture is the single largest employer in most developing countries, so this presents a major problem. Clearly the process will take decades at a minimum, so what are agricultural workers to do in the meantime? If labour markets are imperfect, and farmers and agricultural workers are simply thrown out of work without alternative forms of employment, then even aggregate GDP will not necessarily increase in response to liberalization (never mind the distributional consequences). These implications for rural to urban migration must be taken into account for the developing countries, which have a much larger proportion of their labour force in agriculture than do the developed countries.
But if protection is necessary, some types of protection will likely help more than others. For example, a positive across the board uniform agricultural tariff will allow comparative advantage within agriculture to be maintained while preserving some absolute advantage as the labour force is slowly being transferred out of agriculture.
One could also make an argument for preferential protection of labour-intensive crops such as vegetables in order to support the income of rural landless labourers, although it would be important to couple this with increased educational opportunities for such households so that they have a viable long-term strategy for exiting poverty. Unfortunately, agricultural protection in actual practice is often skewed in favour of commodities produced by relatively well to do farmers (relative to other farmers, that is, not well to do in some absolute sense).
Many people fear that reduced protection will cause a catastrophic decline in domestic production and expose nations to food insecurity. However, farmers and the agricultural sector can adapt to change. To take an example outside of Asia, there were concerns that maize production in Mexico would suffer greatly from lower prices and trade liberalization under the North American Free Trade Agreement (NAFTA). In fact, maize production in Mexico has not fallen in the wake of these price declines. Local labour and land markets were able to redistribute land away from large commercial producers towards smaller subsistence farmers as land rents dropped (Taylor, Yunez-Naude and Dyer, 2003). Subsistence producers, who expanded cultivated area, bolstered maize production in the wake of the price drops.
Price stability is also a concern when opening to external markets. However, trade can play an important role in ensuring price stability, as in the case of Bangladesh during the "flood of the century" in 1998 that caused a drop in rice production. The government eliminated tariffs on rice imports during the crisis, and the increased supplies brought in from India by private traders kept domestic prices stable (Dorosh, 2001). In addition to trade, government storage also has a role to play in ensuring price stability, but the costs of food reserves are often much higher than the benefits. Tariffs that vary in response to changing world prices and domestic harvests (either according to a predetermined schedule or on an ad hoc basis) may be a more cost-effective mechanism for stabilizing domestic prices. This is a more feasible option today than it was in the 1970s, because the world rice market is more stable as the result of a greater prevalence of irrigation and pest-resistant modern varieties (Dawe, 2002). Such tariffs should be imposed only for key commodities, confined to a low level and used to moderate price fluctuations, not provide protection for an extended length of time.
Research on and experience with agricultural trade liberalization around the globe allows us to draw some useful generic lessons (FAO, 2005a):
Reforms can be conducive to poverty reduction and improved food security if they are carefully designed and implemented within an explicit pro-poor strategy. Every situation is different, and it is important to conduct detailed analysis of specific situations in order for trade to benefit the poor as much as possible.
The sequencing of reforms is important. Appropriate output incentives should be assured before (or at the same time as) input prices are raised, even at the cost of maintaining some well targeted input subsidies during a transitional adjustment period. In addition, whenever possible, reforms should not be too abrupt. Phased reforms are likely to result in less backsliding, and they also allow society to study unanticipated outcomes that emerge during implementation.
Finding mechanisms to encourage and assist the private sector to fill the gaps left by dismantling state agricultural marketing institutions is vital, and new policies may be needed to ensure price stability of important commodities.
Improving rural infrastructure is important for successful policy reform in most countries, but it is particularly needed in low-income areas, along with support for productive investments by small farmers. Without such investments it is difficult for such farmers to respond to price incentives.
Policies to encourage the development of rural non-farm employment are also important for the rural poor. These can include the development of microfinance, simplification of regulatory regimes, infrastructure improvement and special incentives for rural industrialization in poor areas.
As complementary policies to facilitate adjustment of the kind mentioned above can take time to bear fruit, transitional compensatory measures, targeted on lower-income groups, may be needed.
 Technology is being used
in a broad sense to mean any change in farming practices or inputs.|
 An undervalued exchange rate would be similar, but not identical, in its effects because most agricultural output is tradable, at least in theory. However, government trade policies often transform tradable goods into non-tradables. Furthermore, there are many tradables outside the agricultural sector that benefit from an undervalued exchange rate.