Using new narrative measures of exogenous variation in marginal tax rates associated with postwar tax reforms in the US, this study estimates short run tax elasticities of reported income of around 1.2 based on time series from 1946 to 2012. Elasticities are larger in the top 1% of the income distribution but are also positive and statistically significant for other income groups. Previous time series studies of tax returns data have found little evidence for income responses to taxes outside the top of the income distribution. The different results in this study arise because of additional efforts to account for dynamics, expectations and especially the endogeneity of tax policy decisions. Marginal rate cuts lead to increases in real GDP and declines in unemployment. This study also presents evidence that the responses are to marginal tax rates rather than average tax rates. Counterfactual tax cuts targeting the top 1% alone have short-run positive effects on economic activity and incomes outside of the top 1%, but increase inequality in pre-tax incomes. Cuts for taxpayers outside of the top 1% also lead to increases in incomes and economic activity, but with a longer delay.

Applied macroeconomists often compute confidence intervals for impulse responses using local projections, that is, direct linear regressions of future outcomes on current covariates. This paper proves that local projection inference robustly handles two issues that commonly arise in applications: highly persistent data and the estimation of impulse responses at long horizons. We consider local projections that control for lags of the variables in the regression. We show that lag‐augmented local projections with normal critical values are asymptotically valid uniformly over (i) both stationary and non‐stationary data, and also over (ii) a wide range of response horizons. Moreover, lag augmentation obviates the need to correct standard errors for serial correlation in the regression residuals. Hence, local projection inference is arguably both simpler than previously thought and more robust than standard autoregressive inference, whose validity is known to depend sensitively on the persistence of the data and on the length of the horizon.

We apply classical statistical decision theory to a large class of treatment choice problems with partial identification. We show that, in a general class of problems with Gaussian likelihood, all decision rules are admissible; it is maximin-welfare optimal to ignore all data; and, for severe enough partial identification, there are infinitely many minimax-regret optimal decision rules, all of which sometimes randomize the policy recommendation. We uniquely characterize the minimax-regret optimal rule that least frequently randomizes, and show that, in some cases, it can outperform other minimax-regret optimal rules in terms of what we call profiled regret. We analyse the implications of our results in the aggregation of experimental estimates for policy adoption, extrapolation of Local Average Treatment Effects, and policy making in the presence of omitted variable bias.

This article provides an axiomatic characterization of quasi-hyperbolic discounting and a more general class of semi-hyperbolic preferences. We impose consistency restrictions directly on the intertemporal trade-offs by relying on what we call “annuity compensations.” Our axiomatization leads naturally to an experimental design that disentangles discounting from the elasticity of intertemporal substitution. In a pilot experiment we use the partial identification approach to estimate bounds for the distributions of discount factors in the subject pool. Consistent with previous studies, we find evidence for both present and future bias.

We consider impulse response inference in a locally misspecified vector autoregression (VAR) model. The conventional local projection (LP) confidence interval has correct coverage even when the misspecification is so large that it can be detected with probability approaching 1. This result follows from a “double robustness” property analogous to that of popular partially linear regression estimators. By contrast, the conventional VAR confidence interval with short‐to‐moderate lag length can severely undercover for misspecification that is small, difficult to detect statistically, and cannot be ruled out based on economic theory. The VAR confidence interval has robust coverage if, and only if, the lag length is so large that the interval is as wide as the LP interval.

Different agents need to make a prediction. They observe identical data, but have different models: they predict using different explanatory variables. We study which agent believes they have the best predictive ability—as measured by the smallest subjective posterior mean squared prediction error—and show how it depends on the sample size. With small samples, we present results suggesting it is an agent using a low-dimensional model. With large samples, it is generally an agent with a high-dimensional model, possibly including irrelevant variables, but never excluding relevant ones. We apply our results to characterize the winning model in an auction of productive assets, to argue that entrepreneurs and investors with simple models will be overrepresented in new sectors, and to understand the proliferation of “factors” that explain the cross-sectional variation of expected stock returns in the asset-pricing literature.

We propose a new framework for optimal experimentation, which we term the “A/B testing problem.” Our model departs from the existing literature by allowing for fat tails. Our key insight is that the optimal strategy depends on whether most gains accrue from typical innovations or from rare, unpredictable large successes. If the tails of the unobserved distribution of innovation quality are not too fat, the standard approach of using a few high-powered “big” experiments is optimal. However, if the distribution is very fat tailed, a “lean” strategy of trying more ideas, each with possibly smaller sample sizes, is preferred. Our theoretical results, along with an empirical analysis of Microsoft Bing’s EXP platform, suggest that simple changes to business practices could increase innovation productivity.