Negative shocks will obviously have lasting effects if they force households to liquidate savings-as this will make them less able to deal with future shocks. The persistence of negative effects from past agricultural and weather shocks on consumption has been well documented (e.g., Dercon (2004) Dercon, Hoddinott, and Woldehanna (2005)). Households cutting back on production inputs in response to ‘bad’ years could even comprise a poverty trap mechanism. On the other hand, permitting shocks to compromise prospects for future consumption threatens a household in a different way: if such shocks occur with regularity, each event can slowly drag the household (deeper) into poverty. Clearly, the inability of a household to maintain a subsistence level of consumption represents an immediate catastrophe. Very poor households face the unenviable task of choosing between present consumption and future production. In the wake of such a shock, we think of households as reallocating and transacting their many types of assets-including livestock, land, labour and cash-with the dual objectives of maintaining minimum levels of current consumption and protecting prospects for future consumption. Smallholder farm households face a myriad of potential negative income shocks, including weather shocks. Curiously, we also observe that households with poultry and small livestock reduce their fertilizer use even more than those without. Importantly, households with cattle and access to the national fertilizer subsidy program show significantly smaller reductions in the use of mineral fertilizer. Our findings suggest that rainfall shocks negatively affect the use of some maize inputs. Crucially, we allow the magnitude of these input adjustments to differ by household liquidity indicators, as measured by chickens, small livestock (sheep, goats, pigs), cattle, off-farm income and access to fertilizer subsidies.
We estimate semi-elasticities, with respect to these shocks, of four maize inputs: basal fertilizer, top dressing fertilizer, improved maize seed and area planted to maize. Using nationally representative, household-level panel survey data from rural Zambia, we test for differential effects-by household liquidity level-of rainfall shocks on input investments in own-farm production. This study posits that access to liquidity sources plays a role in determining a household’s investments in crop inputs following low-harvest years. If such shocks increase in frequency, the ability to recover quickly becomes critical. If households cope with ‘bad’ harvests by cutting back on production inputs in the following seasons, they are likely to be slowing down their economic recovery. Despite the current focus on asset smoothing, very few studies consider how seasonal crop input decisions are affected by income shocks.