We study the use and economic impact of artificial intelligence (AI) technologies among U.S. firms. We propose a new measure of firm-level AI investments, using a unique combination of detailed worker resume and job postings datasets. Our measure reveals a stark increase in AI investments across sectors in the last decade. AI-investing firms see higher growth in sales, employment, and market valuations.
We examine innovation after the Great Depression using data on a century’s worth of U.S. patents and a difference-in-differences design that exploits regional variation in the severity of the economic crisis. Harder-hit areas experienced large and persistent declines in independent patenting, which lasted for the next 70 years. This decline was larger for young and inexperienced inventors and lower-quality patents. In contrast, large firms had relative innovation increases, especially for inventors with the largest declines in independent patenting.
When an inventory manager attempts to construct probabilistic models of demand based on past data, demand samples are almost never available: only sales data can be used. This limitation, referred to as demand censoring, introduces an exploration-exploitation trade-off as the ordering decisions impact the information collected. Much of the literature has sought to understand how operational decisions should be modified to incorporate this trade-off. We ask an even more basic question: when does the exploration-exploitation trade-off matter?
Bitcoin provides its users with transaction-processing services which are similar to those of traditional payment systems. This article models the novel economic structure implied by Bitcoin’s innovative decentralized design, which allows the payment system to be reliably operated by unrelated parties called miners. We find that this decentralized design protects users from monopoly pricing. Competition among service providers within the platform and free entry imply no entity can profitably affect the level of fees paid by users.
We study the performance of many traditional and novel, text-based variables for in-sample and out-of-sample forecasting of oil spot, futures, and energy company stock returns, and changes in oil volatility, production, and inventories. After controlling for small-sample biases, we find evidence of in-sample predictability. Our text measures, derived using energy news articles, hold their own against traditional variables.
We follow a representative panel of millions of consumers in the U.S. from 2007 to 2017 and document several facts on the long-term effects of the Great Recession. There were about six million foreclosures in the ten-year period after Lehman's collapse. Owners of multiple homes accounted for 25% of these foreclosures, while comprising only 13% of the market. Foreclosures displaced homeowners, with most of them moving at least once. Only a quarter of foreclosed households regained homeownership, taking an average four years to do so.
We study qualitative information disclosure by mutual funds when investors learn from these disclosures in addition to past performance. We show theoretically that fund managers with specialized strategies optimally choose to disclose detailed strategy descriptions, while managers with standardized strategies provide generic descriptions. Generic descriptions lead to errors in benchmarking by investors and thus higher volatility in capital flows.
Loss aversion can occur in riskless and risky choices. We present novel evidence on both in a non-student sample (660 randomly selected customers of a car manufacturer). We measure loss aversion in riskless choice in endowment effect experiments within and between subjects and find similar levels of average loss aversion in both. The subjects of the within study also participate in a simple lottery choice task which arguably measures loss aversion in risky choices. We find substantial heterogeneity in both measures of loss aversion.
In modern equity markets, participants have a choice of many exchanges at which to trade. Exchanges typically operate as electronic limit order books under a "price-time" priority rule and, in turn, can be modeled as multi-class FIFO queueing systems. A market with multiple exchanges can be thought of as a decentralized, parallel queueing system. Heterogeneous traders that submit limit orders select the exchange, i.e., the queue, in which to place their orders by trading off financial considerations against anticipated delays until their orders may fill.