We experimentally study the impact of relatively large enterprise loans in Egypt. Larger loans generate small average impacts, but machine learning using psychometric data reveals “ top performers” (those with the highest predicted treatment effects) substantially increase profits, while profits drop for poor performers. The large differences imply that lender credit allocation decisions matter for aggregate income, yet we find existing practice leads to substantial misallocation. We argue that some entrepreneurs are overoptimistic and squander the opportunities presented by larger loans by taking on too much risk, and show the promise of allocations based on entrepreneurial type relative to firm characteristics

Overlap in business aspects serves as a proxy for firm relatedness. Employing an unsupervised topic modelling methodology from machine learning, we characterize the attention allocations of earnings conference call participants (corporate executives, financial analysts, and investors) over the topics discussed. We construct a novel topic similarity measure that captures incremental, difficult-to-observe, and time-varying firm relatedness. However, valuable information from topic peers is not incorporated into stock price in a timely fashion. A long-short strategy based on the returns of topic peers yields a monthly alpha of approximately 69 basis points. Furthermore, return predictability stems primarily from similar business models, customer management, and influential macroeconomic situations. Return predictability is more pronounced among focal firms with higher information complexities and arbitrage costs. Overall, this study provides a novel approach to automatically summarise firms' business aspects in focus and highlights the asset pricing implications of investors' underreactions to non-obvious and dynamic firm relatedness hidden in earnings conference calls

We employ a novel machine learning technique to classify analysts’ forecast revisions into five types based on how the revision weighs publicly available signals. We label these forecast types as quant, sundry, contrarian, herder, and independent forecasts. Our tests reveal that a greater diversity of forecast types within the consensus is associated with increased consensus dispersion and improved consensus accuracy. Additionally, consensus diversity is associated with an improved information environment for firms, as reflected in reduced earnings announcement information asymmetry and volatility, higher earnings response coefficients, and faster price formation. Our study sheds light on how analysts revise their forecasts and documents capital market benefits associated with different analyst forecasting approaches

We train a machine learning method on a class of informed trades to develop a new measure of informed trading, informed trading intensity (ITI). ITI increases before earnings, mergers and acquisitions, and news announcements, and has implications for return reversal and asset pricing. ITI is effective because it captures nonlinearities and interactions between informed trading, volume, and volatility. This data‐driven approach can shed light on the economics of informed trading, including impatient informed trading, commonality in informed trading, and models of informed trading. Overall, learning from informed trading data can generate an effective informed trading measure

Previous research on calculative intermediaries shows how these effectively challenge, distort, and disrupt accounting practices in ways that policy‐makers might not anticipate. The promises of surveillance capitalism—with its attendant data architectures, datafication processes, and technological sophistication—are different, supposing more accurate ways of reading individuals and greater calculative certainty overall. Yet there is little empirical research to explore how surveillance capitalism manifests itself at the organizational level, either conceptually or operationally. As a result, it remains uncertain whether such specters of omniscience are as haunting in reality as they appear in theory. We explore these themes by way of an ethnographic study into credit scoring in China, showing how intermediary organizations developed a multiplicity of credit scoring models based on machine learning and big data that differed both from original expectations and from each other. These different “renditions” of credit scoring suggest that the data architectures of surveillance capitalism are just as much subject to challenge and adaptation by intermediary organizations as calculative practices, such as accounting, are in more analog environments

We show how random subspace methods can be adapted to estimating local projections with many controls. Random subspace methods have their roots in the machine learning literature and are implemented by averaging over regressions estimated over different combinations of subsets of these controls. We document three key results: (i) Our approach can successfully recover the impulse response functions across Monte Carlo experiments representative of different macroeconomic settings and identification schemes. (ii) Our results suggest that random subspace methods are more accurate than other dimension reduction methods if the underlying large dataset has a factor structure similar to typical macroeconomic datasets such as FRED-MD. (iii) Our approach leads to differences in the estimated impulse response functions relative to benchmark methods when applied to two widely studied empirical applications

While hypothesis testing is a highly formalized activity, hypothesis generation remains largely informal. We propose a systematic procedure to generate novel hypotheses about human behavior, which uses the capacity of machine learning algorithms to notice patterns people might not. We illustrate the procedure with a concrete application: judge decisions about whom to jail. We begin with a striking fact: the defendant’s face alone matters greatly for the judge’s jailing decision. In fact, an algorithm given only the pixels in the defendant’s mug shot accounts for up to half of the predictable variation. We develop a procedure that allows human subjects to interact with this black-box algorithm to produce hypotheses about what in the face influences judge decisions. The procedure generates hypotheses that are both interpretable and novel: they are not explained by demographics (e.g., race) or existing psychology research, nor are they already known (even if tacitly) to people or experts. Though these results are specific, our procedure is general. It provides a way to produce novel, interpretable hypotheses from any high-dimensional data set (e.g., cell phones, satellites, online behavior, news headlines, corporate filings, and high-frequency time series). A central tenet of our article is that hypothesis generation is a valuable activity, and we hope this encourages future work in this largely “prescientific” stage of science

What happens when patients suddenly stop their medications? We study the health consequences of drug interruptions caused by large, abrupt, and arbitrary changes in price. Medicare's prescription drug benefit as-if-randomly assigns 65-year-olds a drug budget as a function of their birth month, beyond which out-of-pocket costs suddenly increase. Those facing smaller budgets consume fewer drugs and die more: mortality increases 0.0164 percentage points per month (13.9%) for each $100 per month budget decrease (24.4%). This estimate is robust to a range of falsification checks and lies in the 97.8th percentile of 544 placebo estimates from similar populations that lack the same idiosyncratic budget policy. Several facts help make sense of this large effect. First, patients stop taking drugs that are both high value and suspected to cause life-threatening withdrawal syndromes when stopped. Second, using machine learning, we identify patients at the highest risk of drug-preventable adverse events. Contrary to the predictions of standard economic models, high-risk patients (e.g., those most likely to have a heart attack) cut back more than low-risk patients on exactly those drugs that would benefit them the most (e.g., statins). Finally, patients appear unaware of these risks. In a survey of 65-year-olds, only one-third believe that stopping their drugs for up to a month could have any serious consequences. We conclude that far from curbing waste, cost sharing is itself highly inefficient, resulting in missed opportunities to buy health at very low cost ($11,321 per life-year