Theses Doctoral

Microcredit, temptation spending and health outcomes in Indonesia: A longitudinal evaluation

Rubenstein, Beth L.

This dissertation examined whether or not microcredit, the provision of small loans to people usually excluded from mainstream financial institutions, demonstrably improved health for typical borrowers in Indonesia. The underlying idea behind microcredit is that loans will increase borrowers’ income and lead to positive changes in their lives, including their health. However, microcredit may actually be harmful to borrowers’ health because of stress associated with repayment obligations, extra working hours needed to start a business and tensions caused by shifting power dynamics in the household. Moreover, for some borrowers, a loan may facilitate increased spending on so-called temptation goods that are damaging to health, such as tobacco and processed foods. Previous research has not adequately explored these competing positive and negative pathways linking microcredit and health.
The project consisted of three parts: a systematic review and two empirical papers. The systematic review synthesized the scientific literature related to individual microcredit loans, health-related temptation spending, psychological stress and self-reported health outcomes in adults. The empirical papers estimated the causal effect of microcredit on household expenditures on tobacco and processed foods, and individual psychological distress and self-rated health. Both empirical papers used data from the Indonesia Family Life Survey, a longitudinal study that followed more than 7,000 households over 21 years.
After adjusting for confounding, people living in borrowing households had levels of psychological distress and self-rated health that were similar to people living in non-borrowing households. These predominantly null findings were relatively robust across multiple models. Microcredit households did spend substantially more money on tobacco than non-borrowers. This difference was driven by households with male borrowers. Still, tobacco spending ultimately did not affect health outcomes. Based on empirical evidence from this dissertation along with findings from other studies, policymakers and practitioners should recalibrate their high expectations of microcredit as a socially transformative intervention. At the same time, fears about the unintended health consequences of microcredit may have been exaggerated. Reliance on longitudinal data generated insights into microcredit and health that could not be established from randomized controlled trials.

Geographic Areas


  • thumnail for Rubenstein_columbia_0054D_15182.pdf Rubenstein_columbia_0054D_15182.pdf application/pdf 671 KB Download File

More About This Work

Academic Units
Thesis Advisors
Bates, Lisa M.
Ph.D., Columbia University
Published Here
April 25, 2019