Effect of Weather Risk on Assets and Labour Allocation Decisions Among Smallholder Farm Households in Developing Countries: Evidence From Burkina Faso

Effect of Weather Risk on Assets and Labour Allocation Decisions Among Smallholder Farm Households in Developing Countries: Evidence From Burkina Faso

ABSTRACT

 

Around the developing world, poverty remains an overwhelming rural issue. Rural dwellers rely heavily on agricultural activities and related assets as main source of livelihoods. They also have widely differing capacities to combine off-farm activities in their income generation process. However, diversification makes households better-off only if it implies increased investment in high return portfolios. An important concern is that high yielding investments are either high risk or exhibit severe entry barriers. In the absence of risk markets, liquidity constrained households may pursue low risk investments to self-insure against income volatility which may create poverty. This study explores how this risk induced behaviour may lead to poverty among smallholder farm households in Burkina Faso by analysing the effect of rainfall risk on their asset holdings and attitudes towards low risk off-farm activities. Two theoretical models are developed using stochastic dynamic programming methods to highlight the mechanisms through which risk implies resource diversion towards less risk but also less profitable agricultural investments. Optimal allocation decisions appear to be determined by the relative profitability of available options which is function of the risk. This implies that consumption smoothing is indirectly pursued by smoothing returns to assets and activities. It is also shown that the household more readily enters off-farm activity as volatility of return to farming increases. Empirical analysis uses a two-year panel data collected in 2010 and 2011 in Burkina Faso. Results suggest that households accumulate liquid assets as a long-term strategy to deal with anticipated rainfall outcomes. Liquidity constraints also contribute to shape the composition of households’ portfolio as well as transaction costs. In addition to these factors, off-farm labour supply is influenced by rainfall uncertainty and rainfall shocks. However, the data suggest that productive farm assets (farm equipment and draught animals), and farming in general, have better returns than less productive liquid assets (livestock, poultry and grain stocks) and off-farm activities as the latter are less attractive to wealthier households. The tendency to accumulate increased level of liquid unproductive assets or work more off-farm will therefore result in lower income paths which may explain why poverty persists. Enabling households to accumulate more farm productive assets and work more on farm is necessary to boosting their income generation capacities. This implies addressing the rainfall risk they face. In the short-term, policies may promote groundwater pump irrigation and fodder crop cultivation to mitigate yield volatility and improve returns from livestock breeding. In the medium term, the development of a rural credit market will encourage high yielding investment directly by enabling households to overcome liquidity constraints and indirectly by enhancing ex-post risk management. Regulatory policies that improve returns to riskless options (liquid assets and off-farm work) are needed. This includes provision of veterinary care, establishment of guaranteed minimum wage and efforts to reduce transaction costs. In the long-term, a structural transformation of the farming system that lowers the reliance on rainfed farming through the development of irrigation schemes and the development of agricultural insurance and irrigation schemes may be pursued.

CHAPTER ONE INTRODUCTION

1.1               Background

 

 

The largest share of the world’s poor lives in rural areas. According to IFAD (2016), about 75 per cent of the world’s hungry poor live in rural areas. Majority of them are smallholder farmers who rely on agriculture and related activities as a main source of livelihood. Although recent progress against global poverty across the world has been steady, there are significant differences across regions of the world in terms of the poverty declines. Asia and Pacific showed the highest performance by reducing poverty rate from 50.4 to 19 per cent between 1990 and 2010. On the other hand, over the same period, Sub-Sahara Africa barely achieved a 10 per cent drop from the 60 per cent baseline in the early 1990s.

 

Levels of poverty and hunger vary considerably, not only across regions but also across and within countries. Generally, poverty rates in rural areas are higher compared to those in urban areas. The social development goal of eradicating poverty and hunger will therefore be won or lost in the rural areas, particularly in Sub-Saharan Africa, where the incidence of poverty is highest. This implies equipping the smallholder farmers and other rural dwellers with the resources to face the challenges to sustaining their livelihoods but also to play their role in feeding the urban poor.

 

Learning from the experience of the successful Green Revolution policies from 1960s and 1970s which benefitted smallholders in southern Asia and Latin America, agricultural

 

 

 

1

 

development policies still offer a potential to be explored. According to Diao et al. (2010), there is no viable alternative to the development process in African countries that could bypass the agricultural sector while; Christiaensen et al. (2011) and Valdés & Foster (2010) show that growth in agriculture is more effective for reducing poverty and that this poverty reduction is higher than its contribution to economic growth.

 

The role of productive farm assets is more often emphasized as the way to overcome rural poverty (Dawson et al., 2016; Finan et al., 2005; Jayne, Mather, & Mghenyi, 2010; Mckay, 2009; Quisumbing & Baulch, 2013). The getting in and out of poverty may therefore occur because households are unable or unwilling to accumulate assets as argued by Hoddinott (2006). From a much broader perspective, rural poverty is seen as the consequences of the lack of assets and limited economic opportunities (Beverly et al., 2008 and Stein & Horn, 2012). However, rural households have also shown widely differing capacities to combine non-farm sources in their income generation process (Barrett et al., 2001; Haggblade et al., 2010; Lanjouw & Feder, 2001).

 

Several studies have shown that access to assets, particularly when it is combined with diversity of activities, is the key to boosting income of rural households (Barrett et al., 2001). Furthermore, a large body of the literature has shown that, beyond income, well- being of households in the rural settings are closely related their asset holdings (Carter & Lybbert, 2012; Dercon & Christiaensen, 2011; Zimmerman & Carter, 2003) as well as their to access non-farm sources of income (Bezu et al., 2012; Reardon et al., 1992). Addressing rural poverty, and more generally improving rural households’ well-being, requires policies that support personal initiatives which enable rural people to access and

 

accumulate assets, and undertake higher return activities. This does not only enable movement out of poverty but also enhances resilience to shocks, the main determinant of falling back into poverty.

 

Focus should therefore be on the assets that rural households accumulate, the activities they undertake and how they invest in those assets and activities for their livelihood within their communities. According to Beverly et al. (2008) and Sherraden et al. (2004), in the developed countries, policy-makers are increasingly interested in exploring asset- accumulation strategies to reduce poverty that were outside the realm of means-tested policies (cash and food transfers). This argument is emphasized by Pender et al. (2012) who point out that rural development researchers and practitioners have recognized that investing in a broad range of assets is critical for long-term improvement of well-being of rural farm households and to a large extent, economic growth and prosperity in rural communities.

 

Nevertheless, rural development policies have often focused either on enhancing farm productivity or attracting external capital into rural areas (Mikulcak et al., 2015), but in a less sustainable way. Former category of policies is largely involved in promoting access to improved technologies (mainly improved seed and fertilizers) through subsidization or free distribution; but beneficiaries are often left without enough capacity to take over. Among other reasons, the cost and risk associated with such technologies may be mentioned. The latter category of policies, the exogenous rural development strategies, also shows low performance in terms of poverty reduction as it is often restricted to the creation low unskilled paid work. As indicated by Haggblade et al. (2010), for the poor to take

 

advantage of the rural non-farm growth, this process must induce more than simple low productivity employment.

 

Although the role of the external capital must not be neglected in efforts to achieve rural development goals, all development strategies should start from harnessing local resources to achieve high impact and sustainability. Indeed, the success of any development strategy will depend on the extent to which the social planner understands and uses information about the local context and the structure of the local economy. Those are determined by the endowment of households in different types of assets, the interrelationships among those assets and the households’ decision making with respect to the allocation of those assets to their income generation process. The prevalence of high levels of poverty, despite a continuously positive economic growth in Sub-Sahara Africa, suggests that growth has not been inclusive enough. This implies that there is a need to implement more pro-poor growth policies for the post-2015 development agenda. Such policies must target the rural population, enable them to acquire and build wealth. The question of why some households can take advantage of high return assets and activities, and accumulate wealth over time while others are unable to do so is therefore relevant to investigate, particularly in communities where poverty is high and persistent.

 

The literature discusses two key factors, often treated separately and sometimes combined, to address the issue of why households behave differently in asset accumulation and labour allocation. The first thread focuses on the role of liquidity constraints in explaining differences in asset holdings (Carroll & Summers, 1991; Carter & Zimmerman, 2000; Dercon, 1998) and occupational choices (Dercon, 1998; Dercon & Krishnan, 1996), and

 

therefore poverty. According to this thread of the literature, differences in assets and activity choice over time reflect how liquidity constraints interact with differences in initial endowment to determine who can afford each type of asset and activity. High return assets and activities usually exhibit high entry constraints unlike those with lows yields, implying that initially less endowed households may find themselves restricted to low yielding assets and activities while wealthier households can enter high yielding assets and activities. This ultimately creates large inequalities.

 

The second thread in the literature emphasises the role of risk as the key explanation of the persistence of poverty in developing countries which are generally characterized by the absence of risk markets risk (Deaton, 1991; Dercon, 2005; Dercon & Christiaensen, 2011; Kochar, 1995; Morduch, 1994; Rijkers & Soderbom, 2013; Townsend, 1995) as the prominent explanation of the persistence of poverty in developing countries which are generally characterized by the absence of risk markets. One way risk contributes to creating rural poverty is directly by being responsible for large income fluctuations (Morduch, 1994). More indirectly, the lack of insurance (either formal risk markets or informal risk insurance) causes households to pursue costly risk-reducing strategies in terms of forgone consumption and returns (Deaton, 1991; Dercon & Christiaensen, 2011; Kochar, 1995; Rosenzweig & Binswanger, 1993; Rosenzweig & Wolpin, 1993). In other words, to smooth consumption or income, households may intentionally pursue lower income generation strategies, provided they are relatively stable. Such attitudes are particularly more valuable when, in addition to the missing formal risk markets, informal risk insurance mechanisms are also not functioning effectively.

 

Finally, little effort has been made to explain how these features (liquidity constraint and missing risk market) may interact to determine the long-run social equilibrium. According to Zimmerman & Carter (2003), when liquidity constrained households have to endogenously generate income given a stochastic random shock, the cost of accumulation can be considerably higher than suggested by above mentioned authors. That is, the liquidity constraint and missing risk markets exacerbate the risk induced poverty. This suggests that risk management strategies must also reflect the presence of proactive behaviour. On the one hand, the ex-ante asset and activity portfolios decisions reflect the household’s risk-returns preferences, given one’s asset endowment. Risk averse households will more readily trade high risk for lower stable returns. On the other hand, the inter-temporal consumption smoothing may still involve consumption reduction when households face downside risk. Therefore, risk reducing strategies are not only costly for poor households, they may also create poverty traps.

 

With respect to asset accumulation, the empirical literature has, however, extensively approached the problem of households’ behavioural response to risk by analysing ex-post shock coping strategies; presumably because if risk predetermines the structure of the income generation strategy, then this should be reflected in the assets’ response to risk realization. Moreover, the ex-post approach to understanding the effect of risk on households’ asset holdings has also focused on identifying the potential of specific asset in providing insurance against income fluctuation, omitting the potential portfolio effects (Deaton, 1991; Fafchamps et al., 1998; Kazianga & Udry, 2006; Mogues, 2011). Few exceptions are Jalan & Ravallion (2001); Kazarosian (1997); Rosenzweig & Binswanger

 

(1993) and Zimmerman & Carter (2003) who have analysed the effect of risk induced behaviour on agricultural assets’ portfolio decisions of farm households.

 

With respect to livelihood activities, the literature has also extensively discussed off-farm work either as an ex-post response to farm income shortfalls or as a result of differences in initial asset endowment. Few attempts to unfold the proactive labour allocation in response to potential farm risk are found in Ito & Kurosaki (2009); Menon (2009); Mishra & Goodwin (1997); Rijkers & Soderbom (2013) and Rose (2001). Despite the extensive literature on off-farm work decisions, household surveys usually do not investigate in greater detail the issue of off-farm labour; thereby limiting the scope of the empirical literature on cross-sector labour allocation by smallholder farm households. Indeed, the largest body of the empirical literature is restricted to the decision to participate in off-farm activities. However, unlike households’ fixed assets, labour may be reallocated with less cost, at least as regards explicit cost, in response to change in the rural environment. Labour may therefore be even more responsive to weather risk than any other resource of the households. Extending the framework of risk induced behavior to analyse households’ labour allocation may provide a better understanding about the effect of risk on households’ resource allocation.

 

The challenge that remains therefore is to investigate the household’s consumption- investment decision in ways that reflect the trade-offs that a rural household would actually face. Assets are chosen with regards to their risk-return structure and their liquidity. The latter may be enough to justify the presence of precaution as far as consumption is

 

concerned, while the former reflects the potential causal effect of risk on creating and perpetuating poverty. Inter-sector labour mobility may also reflect these key features.

 

1.2               Problem Statement

 

 

1.2.1          State of Poverty in Burkina Faso

 

 

Despite decades of continuous growth, more than 5 per cent on average since 2000, poverty persists in Burkina Faso. The latest Households Living Standard Survey (INSD, 2016), showed that in 2014, about 40.1 per cent of the total population lived below the poverty line which is less 1 USD per day. Though this indicates a drop of about 5 points in the average poverty rate over the two decades (45.725 per cent), poverty level remains far from the target of the commitment of reducing poverty (less than 38 per cent in 2020)1. Figure

1.1 presents the evolution of the proportion of people living below the poverty line over the last two decades.

 

Figure 1.1 reveals that poverty in Burkina Faso is overwhelmingly a rural issue. Although rural poverty declined in 2014, there is still more than 47 per cent of the rural dwellers who live below the poverty line. Between 1994 and 2009, rural poverty was consistently more than 50 per cent, with a peak of 52.8 per cent in 2009. At the same time, around 13.7 per cent of urban dwellers were reported poor in 2014 while it has been on average less than 20 per cent; despite a transitory spike in 2009. Furthermore, this average rural poverty ignores significant disparities across the different regions of the country. For instance, in

 

Please note: This content is from UGSPACE.UG.EDU.GH and it’s free and you can get the download link for complete work below.

 

References

• ugspace.ug.edu.gh

 

DOWNLOAD PDF LINK : http://ugspace.ug.edu.gh/handle/123456789/23238

Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution License 4.0