Project Description


In cooperation with our local partner, the “National Rural Support Programme of Pakistan” (NRSP) we assess the impacts of several new health and accident insurances for poor households in Pakistan and aim at determining the most suitable insurance contract for poor clients. Lessons drawn from this research have important implications for neighboring low-income countries in South Asia.

According to the Pakistan Ministry of Health public health expenditures comprise about 0.6% of the GDP, which proves exceptionally low in comparison to other countries. Moreover, health related expenses are the largest source of risk in the country, with patients’ out-of-pocket expenses reaching 60 percent, according to the Pakistan Ministry of Health. Due to the limited capacity and availability of public providers patients are forced to seek expensive private medical care in some situations. There is no universal health insurance system in Pakistan, but several arrangements that coexist at a time. In total, only 1.9% of the households are estimated to use any kind of formal insurance product. The most vulnerable households are generally not the target group. If social insurance exists in Pakistan, it is often insufficient to absorb the losses. In addition, it does not reach the population working in the informal sector, which represents a majority of workers in Pakistan. This makes health shocks a substantial economic risk for poor households. Indeed, in developing countries like Pakistan shocks such as illnesses, death of family members, natural catastrophes, price fluctuations, unemployment are likely to create real economic threats. Without proper health insurance, households may be forced into ineffective labor supply allocations. Moreover, the absence of health insurance may also incur long-lasting reductions in labor productivity due to the inability to treat health shocks.


Innovative health insurance solutions are designed and implemented for microfinance clients, particularly targeting low-income and vulnerable households. The effects of different program variations, i.e. individual, household and group-based insurance contracts will be evaluated, considering their effects on adult labor input, child labor activities, schooling and health status. Impacts will be estimated both on financial outcomes of direct importance to NRSP (e.g. re-payment, client retention, adverse selection, moral hazard) as well as on household outcomes (i.e. social impact measurement). The large sample size allows for detailed recommendations for different subgroups of clients: small entrepreneurs, farmers, males and females, and different wealth groups. The rigorous evaluation design used (randomized control trial) will therefore teach about which insurance solutions are most suitable for poor clients in Pakistan and permit the development of tailored and targeted solutions.