Commitment failure lies at the core of incomplete contract theory, yet its quantitative significance has rarely been assessed. Using detailed plot-level data from rural Pakistan, we find that non-contractible investment is underprovided on tenanted land, even after controlling for the endogeneity of leasing decisions. Our evidence also indicates that moral hazard in investment effort alone cannot explain this inefficiency. Instead, imperfect commitment appears to be the driving mechanism, since even plots taken on fixed rent contracts where all the rent is paid upfront receive lower investment than owner-cultivated plots. We further show that a considerable portion of the variation in tenancy duration, and hence in the security of tenure, is due to heterogeneity across landlords. One interpretation of this finding is that landlord reputation is important in mitigating hold-up.

Can marriage institutions limit marital inefficiency? We study the pervasive custom of watta satta in rural Pakistan, a bride exchange between families coupled with a mutual threat of retaliation. Watta satta can be seen as a mechanism for coordinating the actions of two sets of parents, each wishing to restrain their son-in-law. We find that marital discord, as measured by estrangement, domestic abuse, and wife's mental health, is indeed significantly lower in watta satta versus “conventional” marriage, but only after accounting for selection bias. These benefits cannot be explained by endogamy, a marriage pattern associated with watta satta.

We assess the role of monetary incentives in a mission-oriented organization by randomly assigning workers to one of two bonus schemes, incentivizing either the performance of a microcredit program (bottom line) or the empowerment of clients (mission). We find that the credit bonus improved credit-related outcomes but undermined the social mission, while the social bonus did not harm the bottom line. These results are consistent with a multitasking model with production spillovers or with prosocial behavior. We show that when mission-related rewards are not feasible, organizations that care about both the mission and the bottom line prefer flat wages to incentives.