By Raghav Gaiha; Samuel Annim

Introduction
In less than three years, the poverty headcount ratio in Lao People's Democratic Republic is expected to have reached a low level of 23 per cent, following the MDG's 25-year road map of halving poverty by 2015. Estimates based on the 1992/93 and 2007/08 Lao Expenditure and Consumption Surveys (LECS) showed that poverty had dropped by about 18 per cent over the 15-year period, 1992/93 to 2007/08. In view of the complex channels through which poverty reduction is achieved, simplistic extrapolation of past experience could be misleading. An attempt is made here to elaborate these channels with agriculture at the centre stage.

Our analysis, based on a recent household survey (LECS IV, 2007/08), throws new light on the determinants of household poverty in Lao PDR, with emphasis on returns on all crops as well as glutinous rice, producer prices and access to credit3. Although in the past few years, the service and industrial sectors have contributed more to GDP growth in Lao PDR, most of the poor households still depend on agriculture. Three measures of poverty are considered: consumption expenditure (food and non-food) poverty, food expenditure poverty and dollar poverty based on the World Bank's poverty line of $1.25 (PPP, 2005). The rationale for using different measures of poverty is to examine variation (if any) in the drivers of poverty and identify correlates of poverty that cut across all three measures. In fact, sources of these measures of poverty differ but there are a few common correlates as well.

Lao's economy in the last few years - transitioning to a market- oriented policy regime - has performed well compared to other developing economies in the South-East Asia region. With a GNI per capita of US$ 1,050 (World Bank, 2011), it is currently classified as a lower middle income country.

In the context of MDG1, Lao PDR is on track in terms of an annual GDP growth rate pegged at a minimum of 8 per cent per annum and an inflation rate lower than GDP growth rate. However, given the hikes in food prices between 2007 and 2008, and between 2009 and 2010, vulnerability of large segments of the population to poverty is high. Real interest rate volatility is also worrying as it dampens investment. The government has failed to maintain steady real interest rates over the period 2000 and 2011. Unlike gross fixed capital formation, which has steadily increased over the period, foreign direct investment increased sharply between 2005 and 2007, fell slightly the year after, and since 2008 has been fairly stable. In sum, although Lao's macroeconomic outlook is promising, there are a few disturbing symptoms.

The last few decades have seen significant changes in agricultural markets on account of several factors such as reduced state intervention and deregulation, changing food basket, growing urbanization and emergence of supermarkets, and globalization of agricultural trade. Increasing intensification and diversification of agriculture are dependent on integration with markets (relatively small domestic markets and much larger and rapidly growing regional and global markets) and ultimately on the development of value chains. But agroecological characteristics and cross-border trade matter a great deal.

Agricultural households operate both as producers and consumers. While many of them are 'net buyers' of food and sellers of labour, some are producers and others perform more than one activity. All of them need access to markets in order to get high returns on their resources of land, labour and capital, including human capital. Rural households with assets, non-farm income, and occupying favourable locations (irrigated regions with good infrastructure) have better access to markets than those who do not have one or more of these characteristics. When output prices rise, the latter are not able to take advantage, either because they do not have enough surpluses to sell or easy access to markets. A major priority, therefore, is how to make agricultural activities profitable for smallholders and those located in unfavourable agro-climatic regions.

Incidence of poverty
Just under a third (27.6 per cent) of the Lao PDR population is poor in terms of consumption expenditure, using 159,611.9 kip as the national poverty line. Using the 'dollar' poverty line, more than a third of the Lao PDR population is poor. The incidence of food poverty (food expenditure <2,100 calories per capita per day) is slightly lower and shows that about 1 out of 4 households is poor in terms of food expenditure (Ministry of Planning and Investment, Department of Statistics, 2010).

Main findings
One of our major findings is that higher returns on all crops reduce poverty. This finding applies to all three poverty indices. A related finding is that higher returns on glutinous rice also reduce poverty.

Village characteristics matter too. We examined the effect of average village level producer prices of all crops and of glutinous rice, and presence of a credit bank in the village. On the effect of all crops' and glutinous rice prices, our analysis consistently showed a poverty reducing effect for all three poverty measures. Or, in other words, the higher the price of all crops or of glutinous rice, the higher the revenue and the lower the risk of poverty. Presence of a credit bank in a village consistently reduced poverty regardless of the measure used. To illustrate, households residing in villages with easier access to credit were 26 per cent less likely to be poor than those in villages without access to a credit bank. This evidence reinforces the case for financial inclusion in the current wave of financial sector reforms in Lao PDR.

Other household characteristics that consistently showed a poverty reducing effect are head of household's level of education, number of adults employed in non-agricultural activities in the household, and landholdings. For example in the context of food poverty, households with medium landholdings (between 2-5 ha) were 35 per cent less likely to be poor while those with larger landholdings (>5 ha) were 65 per cent less likely to be poor than those with small holdings (<2 ha).

Households in the Southern region (the provinces of Saravane, Sekong, Champasack and Attapeu) were less likely to be poor than those in Vientiane. Across the three poverty measures, the HmongIuMien group was less likely to be poor compared to the Lao-Tai. Both midland and the upland areas were more likely to be poor than the lowland areas. In fact, households in the midland and upland areas were, respectively, 45 per cent and 44 per cent more likely to be poor than their counterparts in the lowland areas.

Policy perspective
Our analysis suggests that larger landholdings, higher education of household heads and more persons employed in non-agricultural activities all have poverty reducing effects. Moreover, higher returns on all crops as well as on glutinous rice further reduce poverty. Finally, a few village level characteristics matter. Specifically, higher prices of all crops and glutinous rice, and easier access to credit in the village have considerable potential for reducing poverty.

From a broader policy perspective of accomplishing the goal of halving poverty by 2015, a few areas of policy concern are delineated. Of considerable importance is better integration of Lao PDR agriculture with sources of demand outside the country. While there is substantial cross-border trade of paddy with Thailand and reimport of milled rice, and of glutinous rice with Viet Nam, the export potential is far from fully utilized. From this perspective, the Greater Mekong Subregion (GMS) alone represents a huge market with rapidly growing incomes and effective demand for products in which Lao PDR has actual or potential comparative advantage (e.g. temperate and off-season fruits and vegetables). So the potential for demand-pull stimulus to diversification and increase in aggregate production needs careful assessment. However, an over-optimistic assessment of such trade prospects must not overlook the emerging concerns for biosecurity, food safety and sanitary phytosanitary (SPS) measures and Lao PDR's limited capacity in addressing them. Public-private partnerships must be forged for better enforcement and compliance.

Both generic and product-specific issues must be addressed. Briefly, the generic issues relate to: (i) poor physical infrastructure, (ii) poor quality of output for sale (in terms of grade standards), and (iii) limited technical assistance and dysfunctional extension services.

Some of the constraints for specific products that deserve close scrutiny include: (i) the rice value chain is constrained by lack of promotion of non-glutinous rice varieties (with high export demand) and that of glutinous rice by inefficient milling; and (ii) coffee production and exports are hampered by the absence of a grading system, standards and certification, and by poor post-harvest handling and processing.

Supply chains or value chains offer both opportunities and obstacles. The main obstacle is that the value chain imposes prohibitive costs on smallholders who often live in areas with poor infrastructure, weak credit and input support, and outdated technology. An imperative for the government is to provide public goods (infrastructure, food safety standards and a favourable environment for enforcing contracts) and facilitate collaboration with forward-looking private players in providing inputs and transferring technology to smallholders. Providing support to farmers' associations would also help improve quality and marketing of produce.

To conclude, an accelerated transition to a more market-oriented policy regime may promote not just a more efficient agriculture but also a more equitable outcome.

(References available upon request)