Knowledge that Transforms

Exponential regressions are frequently used when outcomes are non-negative. They are attractive because they are easy to interpret and to estimate, using pseudo maximum likelihood (PML). However, the validity of these methods depends on the correct specification of the conditional expectation, and little is known regarding their properties when the conditional expectation is misspecified. We show that PML estimators of misspecified exponential models provide optimal approximations to the conditional expectation, in a weighted mean squared error sense, and we give conditions under which their Poisson PML estimator identifies average marginal effects.

Can information from a credible messenger shift behavior in an information-saturated environment? In a randomized controlled trial involving twenty-eight million individuals in West Bengal, we find that SMS-delivered video messages containing information about COVID-19 symptoms and health-preserving behaviors recorded by a credible messenger increased adherence to targeted and non-targeted preventive behaviors, measured by two objective measures (symptoms reported to a health worker, and phone usage at home), as well as self-reported behaviors. We find large spillovers onto non-targeted recipients. Credible light-touch messaging can play an important role in crisis response, even when similar information is widely available.

Applying a media market boundary approach to individual survey data, I show that political advertising on television increases the probability that viewers who identify with a party will espouse its positions, prefer its candidates, and turn out to vote. This is true no matter which party sponsored the ad, suggesting that an ad consolidates and motivates the sponsor's partisans while simultaneously engendering a countervailing consolidation and mobilization among supporters of the other party. My results are consistent with agents who judge a source's quality by their priors and highlight the importance of targeting supporters.

We examine the mechanisms driving the aggregate and distributional impacts of Universal Basic Income (UBI) through model analysis of various UBI programs and financing schemes. The main adverse effect is the distortionary tax increase to fund UBI, reducing labor force participation. Secondary channels are a decline in demand for self-insurance, depressing aggregate capital, and a positive income effect that further deters labor force participation. Due to these channels, introducing UBI alongside existing social programs reduces output and average welfare. Partially substituting existing programs with UBI mitigates the adverse effects, increases average welfare, but does not deliver a Pareto improvement.

The paper studies nonstationary high-dimensional vector autoregressions of order k, VAR(k). Additional deterministic terms such as trend or seasonality are allowed. The number of time periods, T, and the number of coordinates, N, are assumed to be large and of the same order. Under this regime the first-order asymptotics of the Johansen likelihood ratio (LR), Pillai–Bartlett, and Hotelling–Lawley tests for cointegration are derived: the test statistics converge to nonrandom integrals. For more refined analysis, the paper proposes and analyzes a modification of the Johansen test. The new test for the absence of cointegration converges to the partial sum of the Airy1 point process. Supporting Monte Carlo simulations indicate that the same behavior persists universally in many situations beyond those considered in our theorems. The paper presents empirical implementations of the approach for the analysis of S&P100 stocks and of cryptocurrencies. The latter example has a strong presence of multiple cointegrating relationships, while the results for the former are consistent with the null of no cointegration.

Standard extreme value theory implies that the distribution of the largest observations of a large cross section is well approximated by a parametric model, governed by a location, scale and shape parameter. The extremes of a panel of independent cross sections are all governed by the same parameters as long as the underlying distribution as well as the size of the cross sections are time invariant. We derive inference about these parameters, and tests of the null hypothesis of time invariance, under asymptotics that do not require the number of extremes or the number of time periods to increase. We further apply Hamiltonian Monte Carlo techniques to estimate the path of time-varying parameters. We illustrate the approach in four examples of U.S. data: damages from weather-related disasters, financial returns, city sizes and firm sizes.

We develop a general equilibrium model of international trade that incorporates imperfect factor mobility, product entry, and external returns to scale into a unified framework. The effects from these microeconomic channels can be summarized by two composite elasticities that govern supply and aggregation. We structurally derive export supply and import demand curves, develop a heteroskedastic estimator, and estimate supply, aggregation, and demand elasticities across international product markets. Employing our estimated model, we evaluate the impact of recent US protectionist policies and highlight the importance of our estimates and general equilibrium effects for tariff passthrough rates and cross-industry employment reallocations.

We present new findings about the relationship between marriage and socioeconomic background in the United States in the late 19th and early 20th centuries. Imputing socioeconomic status of family of origin from first names, we document a socioeconomic gradient for women in the probability of marriage and the socioeconomic status of husbands, which widens over this period. Regional divergence in occupational structure explains half of the divergence in the probability of marriage, and most of the increase in marital sorting. Urbanization and the associated improvement in women's labor market opportunities drive most of these differences.

In an individual decision-making experiment, we investigate the impact of Tax Deferred Accounts (TDAs). We design six treatments to study various channels through which TDAs may affect decisions. Across both student and Mturk samples, we consistently find that TDAs significantly increase retirement wealth compared to environments with only one non-tax advantaged, liquid saving account. This increase is primarily explained by the requirement of making retirement savings decisions precede consumption decisions. Educating participants by providing a tax calculator has minimal effects. Our results highlight the effectiveness of TDAs in enhancing retirement preparedness and the significance of the order of consumption/savings decisions.

Using a staggered installation of sanitary pad vending machines across schools in the Indian state of Kerala, we study the impacts of free monthly access to sanitary pads on girls' educational outcomes. We find that the number of dropouts among female students in the 7th-grade decrease by 24 percentage points and the attendance rate increases by 23 percentage points after the treatment. Our results are mainly driven by girls in backward-caste, rural schools, and public schools, supporting the idea that free distribution of sanitary pads alleviates cost of obtaining sanitary pads.