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.

We placed 85,000 retail trades in six retail brokerage accounts from December 2021 to June 2022 to validate the Boehmer et al. algorithm, which uses subpenny trade prices to identify and sign retail trades. The algorithm identifies 35% of our trades as retail, incorrectly signs 28% of identified trades, and yields uninformative order imbalance measures for 30% of stocks. We modify the algorithm by signing trades using the quoted spread midpoints. The quote midpoint method does not affect identification rates but reduces the signing error rates to 5% and provides informative order imbalance measures for all stocks.

Maintaining economic output during the COVID-19 pandemic results in benefits for firm shareholders but comes at a potential cost to public health. Using store-level data, we examine how a CEO's political leaning impacts this trade-off. We document that firms with a Republican-leaning CEO experience a relative increase in store visits compared to firms with a Democratic-leaning CEO. The increase in store visits is associated with higher sales and positive abnormal stock returns. However, we also document higher COVID-19 transmission rates and more employee safety complaints in communities where establishments with higher store traffic are managed by a Republican-leaning CEO.

Using a natural experiment from a retail gasoline antitrust case, we study how asymmetric information sharing affects oligopoly pricing. Empirically, price competition softens when, following case settlement, information sharing shifts from symmetric to asymmetric, with one firm losing access to high-frequency granular rival price data. We provide theory and empirics illustrating how strategic ignorance creates price commitment, leading to higher price-cost margins. Using a structural model, we find substantial profit-enhancing effects of asymmetric information sharing. These results provide a cautionary tale for antitrust agencies regarding the potential unintended consequences of limiting price information sharing among firms.

Working-memory (WM) capacity is a key component of a wide range of cognitive and noncognitive skills—such as fluid IQ, math, reading, and inhibitory control—but can WM training improve these skills? Here, we examine the causal impact of WM training embedded in regular school teaching, using a randomized educational intervention with 6–7-year-old children. We find substantial gains in WM capacity and document positive spillover effects on geometry, fluid IQ, and inhibitory control. Three years later, treated children are 16 percentage points more likely to enter an advanced secondary school track.

We use a series of field experiments in rural Burundi to examine the impact of exposure to conflict on social, risk, and time preferences. We find that conflict affects behavior: individuals exposed to violence display more altruistic behavior towards their neighbors, are more risk-seeking, and have higher discount rates. Large adverse shocks can thus alter savings and investments decisions, and potentially have long-run consequences—even if the shocks themselves are temporary.

Across two large healthcare systems, intrasepsis factors improved postsepsis cardiovascular risk prediction as compared with presepsis cardiovascular risk profiles. Further exploration of sepsis factors that contribute to postsepsis cardiovascular events is warranted for improved mechanistic and predictive models.