A quarterly magazine of the IMF

September 2008, Volume 45, Number 3

Land and Poverty in Reforming East Asia
Martin Ravallion and Dominique van de Walle

Lessons from Vietnam's agrarian reforms

Both China and Vietnam have made enormous progress in the fight against poverty, and the evidence suggests that rural economic growth has played a large role in this success. Using each country's own definition of poverty, with a constant real poverty line over time, China's poverty rate fell from more than 50 percent in 1981 to about 20 percent in 1991 and 5 percent in 2005. In Vietnam, poverty fell from almost 60 percent to 20 percent during 1993-2004.

Land is the most important non-labor asset in any developing rural economy. The institutions determining how land is used are thus at the core of efforts to fight poverty. China and Vietnam both have had major land reform programs. This article examines the role these major agrarian reforms played in the subsequent progress against poverty and searches for lessons for the future.

We also look at what China might learn from Vietnam's different path in its land policies and what other countries can learn from these star performers. Lessons include the importance of the agricultural sector in the early stages of a pro-poor growth process, the potential role of market-oriented reforms in absolute poverty reduction, and the need to address pressures spurring rising inequality as reforms get under way.

Shift from collectivized farming
In the 1980s and 1990s, China and Vietnam undertook truly major reforms to the laws and regulations governing agricultural land. Prior to that, both countries had collectivized their farming, but both came to realize that this system was not performing well.

Although collectivized farming could ensure low inequality within each commune, this came at too high a price in terms of efficiency, because working in large brigades and sharing the output dulls the incentive for effort. The cooperatives and collectives were dismantled and the land was assigned to individual households in the commune, who had to agree to provide an output quota to the government but could keep the rest for consumption or sale. This system clearly had better incentives, and agricultural output rose accordingly in both countries.

After this important step, promarket reforms to agrarian institutions were put in place in both countries, though Vietnam has gone further. China has still not taken Vietnam's radical, and controversial, step of introducing a legal market in land-use rights.

How the reforms worked
The reform processes in China and Vietnam were not solely concerned with efficiency. Highly inequitable outcomes from the agrarian reforms would have met with popular resistance in the short term and potentially derailed future progress against poverty by stifling the economic opportunities of a large share of the population. Future reform prospects in other areas of policymaking would also have been jeopardized by perceived failures in the initial agrarian reforms.

However, policymakers faced a potentially major threat to the reform process. As in many developing countries, the center had to rely heavily on decentralized implementation of these reforms, down to the commune level. This raised concerns about local elites—whose interests are not well served by the central government's aims—taking over the process for their own purposes. Were these concerns justified? Our research focused on Vietnam's agrarian reforms, which we compared with other observations about the process in China. We first studied how land-use rights were allocated on breaking up Vietnam's collectives. Individual households had to be assigned the use rights for virtually the entire agricultural land area of a country in which three-quarters of the workforce depended directly on farming.

We used econometric models of both household consumption and the behavior of local party cadres to assess the administrative allocation of land achieved by the decollectivization process against explicit counterfactual allocations. One of these was an equity counterfactual in which land was allocated equally within communes and the other was the allocation that would have maximized aggregate consumption, which would have been the competitive market allocation under our assumptions. The model relating consumption to landholding (and other explanatory variables) was used to simulate these counterfactuals; Ravallion and van de Walle (2008) describe the methods in detail.

What we found was not consistent with the picture that some commentators painted of an unjust land allocation stemming from the power of local cadres to capture the process. However, the observed allocation was significantly different from what would be expected of a competitive privatization at market-clearing prices. The consumption-efficient allocation would have put greater weight on education and given less weight to household size, labor force, minority groups, and male heads of household. The reform reinforced existing gender inequities in favor of male heads of household at a cost to efficiency. We find no evidence that land allocation unduly favored those with government or semigovernment jobs.

Vietnam's reform to privatize land-use rights achieved a more equitable outcome than one would have expected from a fully efficient allocation that would have been achieved through free markets. There were both gainers and losers relative to the efficient market allocation, but the gains tended to favor the poor. This can be seen in Chart 1, which plots the estimated loss incurred by the sampled farm-households—judged relative to market allocation of land—against initial consumption. The lower left quadrant shows greater net gains (negative losses) for the poor.

Our results point to an effort by the authorities to protect the poorest and reduce overall inequality, at the expense of aggregate consumption. The solution that was reached entailed an equity-efficiency trade-off, indicating that both objectives were valued positively. Our (more casual) observations and other evidence suggest that the same was probably true in China's decollectivization process.

Reform had winners and losers
A farming household in China cannot sell its land and use the money to finance a new nonfarm enterprise, or move to the city to take up work. Land is not a marketable asset, but is allocated, and at times reallocated, by the local administration, sometimes with the involvement of the village assembly. There have been long-standing concerns about the possible inefficiencies of this system; notably, whether better farmers get enough land and whether the threat of losing one's (limited) rights over the land deters farm investment and dulls the incentive for off-farm work and entrepreneurship.

In response to these concerns, the Chinese government has recently tried to discourage land reallocations. There has also been more widespread use of land rental arrangements among farmers; although these are typically informal (oral) understandings among friends and family, they have probably helped make the allocation of land more efficient, in lieu of a legal market in land-use rights.

Vietnam took a different route than China. The new land law introduced in 1993 attempted to foster free transactions in land-use rights. But there was much debate. Some observers believed that this reform would allow a closer approximation of the efficient allocation, but at the expense of equity. The prospect of renewed class differentiation—the re-emergence of a rural proletariat—has fueled much debate about the wisdom of Vietnam's efforts at liberalizing land markets. This has been a concern in China as well, and has arguably been the main factor stalling market-oriented reforms to land laws.

A long-standing view in some quarters is that even from an equal starting point, the market mechanism will generate excess inequality. However, the same features that helped ensure an equitable allocation at the time of decollectivization—including relatively high and equal human capital—may well have operated to moderate any un-equalizing forces generated by the emerging market economy. And the fact that other policy reforms, including more open external policies, were creating new opportunities for diversification and growth is clearly relevant to the outcomes of these reforms.

What does our research for Vietnam suggest? We find signs that, after legal reforms to introduce a market in land-use rights, land was reallocated in a way that attenuated the inefficiencies of the initial administrative assignment of land. Households that started with an inefficiently low (high) amount of annual cropland tended to increase (decrease) their holdings over time. The adjustment was not rapid; in the aggregate, only one-third of the initial proportionate gap between the actual allocation and the efficient allocation was eliminated within five years. And there was continuing local government intervention in some regions. But it seems that the market mechanism did start to take hold.

The market worked more rapidly for some types of households than others. Overall, the transition process favored those who initially had too little land. The speed of market adjustment was also affected by location and demographic shocks, and the new market-driven process favored households with long-term roots in the community, better education, and with more land in other (non-annual cropland) categories.

Did these efficiency gains from introducing land markets come at a cost to the poor? One should not be surprised to find a higher incidence of landlessness. Many farmers will no doubt benefit from the new opportunities to use their limited wealth in other ways, including spending on consumer durables and housing. But there will also be losers from such a reform. Welfare losses can occur for those who were previously landless, who receive lower wages than without the reform, and for farmers who find that other benefits provided by the cooperatives were retrenched once their role in land allocation was removed.

Our analysis of the survey data for Vietnam—spanning a decade after legal reforms to introduce markets in land-use rights—confirms the expected rise in the landlessness rate among many of the poor. Similarly, it was the initially poor who saw the highest pace of urbanization over time. Even so, the postreform landlessness rate tends to be higher for the rural non-poor in Vietnam as a whole, as can be seen from Chart 2, which plots the average landlessness rate against household consumption per capita for both 1993 and 2004. Landlessness rises with living standards. Between 1993 and 2004, it fell for the very poorest as well as for the rich, and rose slightly for those in the middle of the distribution. The empirical analysis suggests that, by and large, it is not the currently poor who took up the new opportunities for selling (or buying) land and acquiring land titles, but the relatively well off. Access to formal credit appears to have improved overall (and displaced informal credit), though more markedly for better-off households. Among equally poor households, the landless are less likely to receive credit from formal sources, including the targeted antipoverty programs.

We find little sign that rising landlessness has undermined the gains to the poor from the relatively equitable assignment of land-use rights achieved at the time of Vietnam's decollectivization. Even in the South's Mekong Delta—where there are signs of class differentiation—poverty has been falling among the landless, albeit at a lower rate than for those with land. However, we find no sign that this pattern is emerging elsewhere in Vietnam; indeed, as a rule, the landless are enjoying rates of poverty reduction similar to (or even higher than) people with land.

On the whole, rising rural landlessness appears to have been a positive factor in Vietnam's process of poverty reduction, as farm households take up new opportunities, notably in the labor market. This does not imply that any policy effort to encourage landlessness will be poverty-reducing; it is one thing to give people the opportunity to sell their land to take up more rewarding opportunities, but quite another to compel such changes by forcing farmers off their land. Policies should focus instead on making land markets work better for poor people and on complementary efforts to enhance nonfarm opportunities, notably for the landless rural poor, who tend to have less access to credit for financing investments in nonfarm enterprises.

What are the lessons?
At the start of this article we noted both similarities and dissimilarities between Vietnam's agrarian reforms and those of China. There are historical and contextual factors to consider in understanding the difference in agrarian-reform policies. For example, China had a more deeply rooted tradition of collectivized farming and (in contrast to Vietnam) had largely succeeded in displacing the peasant-family economy. This alone made for a more rapid transition in Vietnam.

While acknowledging that these differences between the two countries had an important influence on the policies chosen, China should not ignore the lessons from the experience of its neighbor. Vietnam's more radical approach of introducing a land market did not have the dire consequences predicted by those who favored the Chinese model of administrative land allocation. Starting from a relatively equitable allocation of land, introducing free exchange did not end in peril and poverty for the rural population, though (as in any major policy reform) there are both losers and gainers. Vietnam's experience also reminds us that the efficiency gains do not happen overnight, and may well take many years to be realized. But gains can be expected, including gains for the poor.

There are also lessons for other countries today. As many developing countries strive to raise farm output in the wake of the dramatic increase in food prices, they should pay close attention to the reforms that may be needed to ensure that individual farmers can respond to market incentives. The reforms will be specific to each country, but countries that still have the kinds of land policies that Vietnam has been so successful in dismantling could benefit by studying that experience.

There are broader lessons for other countries, going beyond the specific reforms to land policies undertaken by China and Vietnam. Their experiences reaffirm that reforms can work when they are sensitive to the context, including concerns about equity in the process and its outcomes. Their experiences also confirm that poor people respond to market incentives when given the opportunity. And the importance of strong state institutions (including at the local level), and a leadership committed to poverty reduction, is confirmed by the experience of both China and Vietnam.

These are generic lessons. We would also point to a more specific lesson—namely, the role played by the evidently high priority both countries gave to agriculture and rural development in the early stages of the reform process. This quickly benefited the poorest segment of society and laid the foundation for the success of later reforms. The high priority given to this sector also helped attenuate the pressures for rising inequality (though Vietnam has been more successful than China in avoiding rising inequality). Yet many low-income, primarily rural, developing countries think they can jump-start their economies by rapidly developing a modern, relatively capital-intensive, manufacturing sector. So they largely ignore their agricultural sectors. This approach has had disappointing outcomes, particularly in countries with high initial inequality in human resource development; indeed, it may even increase poverty through the financing methods (notably the heavy taxation of agriculture) and price distortions that are needed.

Poor, primarily rural, economies cannot reasonably hope to bypass the key steps in actively promoting agricultural and rural development that China and Vietnam took from the early stages of their reform processes. That is an important message for the many low-income countries today that would like to emulate the successes that China and Vietnam have had against absolute poverty.