Vulnerability to Poverty

Life in developing countries often is marked by poverty and risk. Prominent studies such as Voices of the Poor: Crying Out for Change (Narayan et al., 2000) acquaint their readers with the daily struggle faced by poor households that try to escape their hardship in the face of recurring illness, natural disasters and other hazards. In the absence of well-functioning credit and insurance markets — a distressing reality in most developing countries -exposure to these adverse events renders households highly vulnerable. This can affect not only households who were poor to start with and who might face utter destitution as a result of adverse shocks, but also non-poor households who can fall back into poverty, and thus also need to be considered in an analysis of vulnerability to poverty.

Until today there have been few databases that satisfy the theoretical requirements for studying vulnerability to poverty as indicated in Chapter 2. Among existing datasets the Ethiopian Rural Household Survey used, for example, by Dercon et al. (2002; 2005) is prominent and provides a panel dataset well suited to vulnerability analysis. The first wave of data collection was carried out in 1989 in six farming villages in central and southern Ethiopia, with a focus on the crisis and recovery in the 1980s. An expansion of the survey to 15 villages across the country in early 1994 yielded a sample of 1477 households. Additional rounds of the survey were carried out in late 1994, 1995, 1997, 1999, and 2004. In the 1995 wave, a general shock module was implemented, which was further improved in later rounds (Dercon et al., 2005).

While the conceptual work on vulnerability reviewed in Chapter 2 is already quite advanced, the actual measurement of vulnerability to poverty of households is lagging behind. In fact, most empirical assessments of vulnerability have used very simplified methods to assess vulnerability to poverty, such as basing them on a single cross-section or vulnerability scores, or related short-cut procedures. Often these assessments are based on rather crude assessments of shocks experienced and risks faced by households. This is not due to the lack of dedication to the task or the fieldwork or econometric skills of the researchers involved. As will be discussed in more detail below, this is largely due to the fact that the measurement of vulnerability to poverty is extremely demanding as far as data needs are concerned. In particular, adequately capturing shocks and risks and their impact on vulnerability is very challenging.

In recent years, poverty mapping has gained popularity among researchers and policy makers as a powerful tool for identifying pockets of poverty and providing detailed information on regional differences in well-being (Demombynes et al, 2004; Elbers et al, 2007). Poverty mapping has the advantage of combining detailed household data from relatively small surveys that are not representative at the local level with census data which contains only a few key variables but provides complete coverage of households (Elbers et al., 2003). Such maps can be very informative for researchers and policy makers, since they provide poverty estimates at a local level.

The 2008 price crisis in the world markets for fuel, chemical fertilizer and agricultural commodities came as a shock to both producers and consumers. For farmers in Thailand and in Vietnam, this situation generated opportunities and risks. For agricultural households with a food surplus, higher commodity prices will lead to higher incomes provided price expectations hold and the price ratio of factor to commodity price remains favourable. For food deficit households, however, rising prices reduce their real income and tend to increase the risk of falling into poverty.

Thailand and Vietnam are two emerging market economies where agriculture still plays an important role even though its contribution to their GDP has been reduced from 16 per cent and 40 per cent in 1985 to around 12 per cent and 22 per cent in 2008, respectively.¹ In the rural areas, however, agriculture is still the major source of income and employment. Agriculture in Thailand and Vietnam has differences as well as similarities. One of the main differences relates to the historical perspective. In Vietnam, prior to the introduction of the doi moi policy, performance of the agricultural sector was strongly influenced by the centrally planned economic system; the policy change towards a market-based pricing system of agricultural commodities can be seen as a starting point for a period of sustained growth in output and productivity. Today, Vietnam has become a major player in world food markets, and the country now ranks third among the world’s leading rice exporters. However, Vietnam’s economic policy reform has also introduced risks into the agricultural sector and the rural areas. The process of liberalization and rapid integration into the world economy with less trade protection and reduced subsidies has exposed the domestic markets to the fluctuations of the international markets. In contrast, such risks are not so severe in Thailand, as the agricultural sector has benefited from a long history of commercialization and market orientation. Thailand is now the top exporter for a number of agricultural raw materials and processed food products. The country has a well developed agribusiness sector, with some large multinational co-operations.

This research examines financial decision-making not from the perspective of the ordinary but from the extraordinary. Shocks are extraordinary. Events such as excessive rainfall or drought, illness or death of a household member have the potential to dramatically impair household income streams. Shocks to income are shocks to consumption unless households manage to mitigate or avoid the reduction in consumption through the choice of an appropriate smoothing mechanism. Affected households take such measures under the commonly employed assumption that households prefer smooth over fluctuating consumption (due to risk aversion). It is the purpose of this research to analyse household financial decision making after the occurrence of a shock.

The Village Fund (VF) programme in Thailand is one of the largest microfinance programmes in the world. It aims at improving access to finance and income in rural areas. These are worthwhile objectives for policy as finance is often limited in rural areas and incomes are low. In this sense, the introduction of a programme that sets up an additional fund of one million baht, roughly US$28,000, per village leading to a significant increase of loanable funds is welcome. Indeed, the rural population seems to be highly sympathetic to the 2001 government’s decision to start the village fund programme, as election results continuously show.