The relationship between fragility and poverty remains unexplored due to a number of factors. First, the concept of fragility and its measures have overlooked heterogeneity at the micro-level while focusing on fragility at the macro- or state-level. Second, due to plausible endogeneity in the relationship between fragility and poverty, as well as the lack of viable and strong instruments, it remains difficult to draw causality pathways. This study contributes to the fragility-poverty literature by taking a micro-level approach, proposing individual-level measure of fragility, namely the fragility exposure index, which measures individuals’ perceptions and experiences of fragility. This allows us to test the effect of poverty on the levels of fragility. We address the potential endogeneity of poverty by using a shock-related variable as an instrument, namely sum of prior climatic shocks faced by households in the past 12 months prior to the survey. This is made possible by a three-year panel data set, HORTINLEA household survey, conducted in rural and peri-urban areas of Kenya. The findings suggest that higher poverty rates increase perceptions and experiences of fragility at the micro-level. This strong association holds for different specifications, but more significant results are found using instrumental variable estimation approach. From the three dimensions of fragility, economic inclusion shows significance and strong relationship to poverty, while human security shows strong and significance association to poverty only in the IV-2SLS estimation. As such, our finding ascertains two outcomes: on the one hand, use of instrumental variable approach is a viable option to identify the link between poverty and fragility; on the other hand, there is indeed a strong and significant association between poverty and fragility, whereby better-off households (in terms of less poverty) are less likely to be fragile.
