Knowledge that Transforms

Programs implemented by nongovernmental organizations (NGOs) are often more effective than comparable efforts by other actors, yet relatively little is known about how implementer identity drives final outcomes. By combining a stratified field experiment in India with a triple-difference estimation strategy, we show that a local development NGO’s prior engagement with target communities increases the effectiveness of a technology promotion program implemented in these areas by at least 30%. This “NGO reputation effect” has implications for the generalizability and scalability of evidence from experimental research conducted with local implementation partners.

We obtain a canonical representation for block matrices. The representation facilitates simple computation of the determinant, the matrix inverse, and other powers of a block matrix, as well as the matrix logarithm and the matrix exponential. These results are particularly useful for block covariance and block correlation matrices, where evaluation of the Gaussian log-likelihood and estimation are greatly simplified. We illustrate this with an empirical application using a large panel of daily asset returns. Moreover, the representation paves new ways to model and regularize large covariance/correlation matrices, test block structures in matrices, and estimate regressions with many variables.

Measuring the prevalence of active SARS-CoV-2 infections in the general population is difficult because tests are conducted on a small and nonrandom segment of the population. However, hospitalized patients are tested at very high rates, even those admitted for non-COVID reasons. We show how to use information on testing of non-COVID hospitalized patients to obtain tight bounds on population prevalence, under conditions weaker than those usually used. We apply our approach to the population of test and hospitalization data for Indiana, and we validate our approach. Our bounds could be constructed at relatively low cost, and for other heavily tested populations.

This study proposes an econometric framework to interpret and empirically decompose the difference between instrumental variables (IV) and ordinary least squares (OLS) estimates given by a linear regression model when the true causal effects of the treatment are nonlinear in treatment levels and heterogeneous across covariates. I show that the IV-OLS coefficient gap consists of three estimable components: the difference in weights on the covariates, the difference in weights on the treatment levels, and the difference in identified marginal effects that arises from endogeneity bias. Applications of this framework to return-to-schooling estimates demonstrate the empirical relevance of this distinction in properly interpreting the IV-OLS gap.

We provide a novel decomposition of changing gaps in life expectancy between rich and poor into differential changes in age-specific mortality rates and differences in “survivability.” Declining age-specific mortality rates increases life expectancy, but the gain is small if the likelihood of living to this age is small (ex ante survivability) or if the expected remaining lifetime is short (ex post survivability). Lower survivability of the poor explains half of the recent rise in inequality in the United States and the entire rise in Denmark. Declines in cardiovascular mortality benefited rich and poor, but inequality increased because of differences in lifestyle-related survivability.

This paper develops a novel dataset of weekly economic conditions indices for the 50 U.S. states going back to 1987 based on mixed-frequency dynamic factor models with weekly, monthly, and quarterly variables that cover multiple dimensions of state economies. We find considerable cross-state heterogeneity in the length, depth, and timing of business cycles. We illustrate the usefulness of these state-level indices for quantifying the main contributors to the economic collapse caused by the COVID-19 pandemic and for evaluating the effectiveness of the Paycheck Protection Program. We also propose an aggregate indicator that gauges the overall weakness of the U.S. economy.

Can revenue sharing of resource rents be a source of distributive conflict? Can cohesive institutions avoid such conflicts? We exploit exogenous variation in local government revenues and new data on local democratic institutions in Nigeria to study these questions. We find a strong link between rents and conflict. Conflicts are highly organized and concentrated in districts and time periods with unelected local governments. Once local governments are elected these relationships are much weaker. We argue that elections produce more cohesive institutions that help limit distributional conflict between groups. Throughout, we confirm these findings using individual level survey data.

The COVID-19 pandemic has led to enormous data movements that strongly affect parameters and forecasts from standard Bayesian vector autoregressions (BVARs). To address these issues, we propose BVAR models with outlier-augmented stochastic volatility (SV) that combine transitory and persistent changes in volatility. The resulting density forecasts are much less sensitive to outliers in the data than standard BVARs. Predictive Bayes factors indicate that our outlier-augmented SV model provides the best fit for the pandemic period, as well as for earlier subsamples of high volatility. In historical forecasting, outlier-augmented SV schemes fare at least as well as a conventional SV model.

This paper quantifies preferences for income equality and mobility by generating statistics that are uncorrelated with beliefs and can be interpreted as marginal rates of substitution (MRS). All things being equal, U.S. residents are willing to reduce average income by $2,744 to reduce the 90/10 income inequality ratio one unit, and $1,228 to increase income mobility from the bottom quintile one percentage point. Democrats and Independents have similar preferences for both social variables, while Republicans have an MRS that is about two-thirds that of Democrats and Independents for both income inequality and mobility.

We provide evidence that violence reduces the adoption and use of mobile money in three separate empirical settings in Afghanistan. First, analyzing nationwide mobile money transaction logs, we find that users exposed to violence reduce use of mobile money. Second, using panel survey data from a field experiment, we show that subjects expecting violence are significantly less likely to respond to random inducements to use mobile money. Finally, analyzing nationwide financial survey data, we find that individuals expecting violence hold more cash. Collectively, this evidence suggests that violence can impede the growth of formal financial systems.