Poverty dynamics in rural Kenya and Madagascar

Abstract

DESPITE TWO DECADES OF MARKET-ORIENTED REFORMS throughout much of sub-Saharan Africa, poverty rates increased, and the sense has spread that the “Washington Consensus” approach of getting macroeconomic policy and prices “right” does not suffice to stimulate broadlybased growth and reduce poverty. If the theory that spawned the “getting prices right” strategy is inadequate in explaining persistent poverty and prescribing policies to reduce poverty, then perhaps other theories of economic growth can suggest more appropriate intervention strategies. Prevalent macroeconomic growth theories are characterized by three different hypotheses, which have analogues at the household level. The “convergence” hypothesis, which led to the “getting prices right” strategy, posits that the poor enjoy higher marginal returns to productive assets than do the rich, so capital naturally flows disproportionately to the poor, enabling them to catch up economically. Shocks cause merely temporary setbacks. The “conditional convergence” hypothesis holds that individuals within identifiable groups converge to a group-specific standard of living. Geographic poverty traps represent a type of conditional convergence wherein some groups defined by physical location converge to a standard of living that falls below the poverty line. Members of these groups need targeted assistance to stimulate productivity growth. Even among the poorest, accumulation and recovery from shocks occur, albeit only to low levels. The “poverty traps” hypothesis holds that individual wellbeing depends fundamentally on initial conditions. Two otherwise identical neighbors will have radically different experiences if one starts with sufficient land, livestock and human capital, while the other lacks the minimum initial stocks necessary to accumulate wealth over time, or else suffers a serious shock like illness or loss of livestock. Poorer households earn lower expected returns on their assets than do wealthier households. Regions of locally increasing returns to assets can only exist in the presence of some mechanism that excludes some people from choosing more remunerative livelihoods. Typically, exclusion occurs through restricted access to the credit or insurance necessary to build assets through investment or protect them against loss, respectively, or through socially-exclusionary processes that limit access by certain groups or individuals to preferred employment, credit or land. This research represents a micro-level attempt to empirically test the hypotheses of economic growth by examining risk management, marginal returns on productive assets, and asset dynamics across settings distinguished by different agroecological and market access conditions.