What roles do credit constraints and inattention play in the under-adoption of high-return technologies? We study this question in the case of energy efficient cookstoves in Nairobi. Using a randomized field experiment with 1,000 households, we find that the technology has very high returns - we estimate an average rate of return of 300% and savings of $120 per year in fuel costs, around one month of income. In spite of this, adoption rates are inefficiently low. By eliciting preferences using an incentive-compatible Becker-DeGroot-Marschak mechanism, we find that average willingness-to-pay (WTP) is only $12. To investigate what drives this puzzling pattern, we cross-randomize access to credit with two interventions designed to increase attention to the costs and the benefits of adoption. Our first main finding is that credit doubles WTP and closes the energy efficiency gap. Second, credit works in part through psychological channels: around one-third of the impact of credit is caused by inattention to future costs. We find no evidence of inattention to energy savings. These findings have implications for second-best regulation of pollution externalities using subsidies versus taxes. In the presence of credit constraints, Pigovian taxation alone may no longer be the optimal policy. Factoring in private benefits and avoided environmental damages, we estimate welfare gains of $700 for every energy efficient stove adopted and used for two years. A subsidy would have a marginal value of public funds of $19 per $1 spent.

There is limited causal evidence on the effects of different public procurement regulations on project quality and value-for-money for projects funded by national governments and foreign aid donors. This paper uses policy and experimental variation to study how two key contracting features—namely, contract bundling and monitoring—affect outcomes of a large economic development project. We leverage an unusual feature of Kenya’s nationwide electrification program: the quasi-random allocation of multilateral funding sources across nearby villages. African Development Bank (AfDB) projects used bundled contracts while the World Bank (WB) employed unbundled contracts together with strengthened inspections. To measure impacts, we collect on-the-ground engineering assessments, power quality data, household surveys, and analyze original contracts. The analysis suggests a stark trade-off: WB procedures delayed construction completion by 16 months relative to AfDB sites but improved construction quality by a sizeable 0.6 standard deviations. To disentangle the effects of contract bundling versus monitoring, we conducted randomized audits that enhanced monitoring. The audits improve household connectivity, network size, and voltage at AfDB sites, but have no impact at WB sites, suggesting monitoring and unbundling contracts may be substitutes. Given the apparent trade-off, we investigate how net benefits depend on policymaker time preferences and infrastructure longevity

We provide descriptive evidence on the challenges in efficiently, effectively, and fairly distributing in-kind electricity transfers to households. We collect panel data from 1200 households eligible for Ghana's COVID-19 electricity relief program. Distributing relief through electricity transfers enabled an immediate response to the crisis. Theoretical efficiency concerns are mitigated because transfers were inframarginal and storable for most households. Transfer receipt may have increased support for the governing party, possibly due to obfuscation of the program's financial burden. However, the program was regressive in design, and implementation challenges – delays, technological hurdles, information constraints, and the targeting of meters rather than households – add to inefficiency and regressivity. Households receiving the least average relief are those who use less electricity, pay a landlord or other intermediary for electricity, or share an electricity meter—characteristics of low-income households. Program implementation challenges were just as important as design features in determining program costs and benefits.