This paper describes a process for generating academic papers using large language models (LLMs) and demonstrates this process’s efficacy by producing hundreds of complete papers on stock return predictability, a topic well-suited for our illustration. After mining over 30,000 potential return predictors from accounting data, we generate template reports for 95 signals passing rigorous criteria from the Novy-Marx and Velikov (2024) Assaying Anomalies protocol. These templates detail signal performance predicting returns using a wide array of tests and benchmark performance against more than 200 documented anomalies. Finally, for each template we use state-of-the-art LLMs to generate multiple complete versions of academic papers with distinct theoretical justifications for the observed return predictability, incorporating citations to literature supporting their respective claims. This experiment illustrates the potential of artificial intelligence (AI) for enhancing financial research efficiency, but also serves as a cautionary tale, illustrating how it can be abused to industrialize hypothesizing after results are known (HARKing).

We provide evidence that equity investors are slow to process information about how current oil price changes affect future earnings announcements. Stock prices respond to lagged quarterly oil price changes when firms start announcing earnings in the next quarter. A cross-sectional equity trading strategy that exploits this predictability yields an annualized Sharpe ratio of 0.50. Our oil-response forecast strategy earns especially high returns after large absolute oil price changes, in recessions or bear markets, and during peak earnings season. The predictability we document is consistent with limited attention, is not driven by risk factor exposure, and survives several robustness tests.

We study the after-trading-cost performance of anomalies and the effectiveness of transaction cost mitigation techniques. Introducing a buy/hold spread, with more stringent requirements for establishing positions than for maintaining them, is the most effective cost mitigation technique. Most anomalies with less than 50% turnover per month generate significant net spreads when designed to mitigate transaction costs; few with higher turnover do. The extent to which new capital reduces strategy profitability is inversely related to turnover, and strategies based on size, value, and profitability have the greatest capacity to support new capital. Transaction costs always reduce strategy profitability, increasing data-snooping concerns.

Trade credit is an important source of firm financing, yet its rich informational content pertaining to payment timeliness is under-explored in asset pricing. Using an extensive data set from a leading private information exchange on business payment performance, we study the effects of trade credit payment timeliness on stock returns. We document two distinct channels through which trade credit payment behavior impacts future stock returns — slow diffusion of information and risk stemming from a customer firm’s vertical bargaining power position in the supply chain. Consistent with our first channel, a sudden delay in a firm’s payment to its suppliers predicts significantly lower future returns for its stock. Consistent with our second channel, firms that pay their bills moderately late on a consistent basis relative to terms earn significantly higher stock returns.