Job Market Paper
Human Decision-Making with Machine Prediction: Evidence from Predictive Maintenance in Trucking (with Maggie Yellen)
In this paper, we study the role of predictive artificial intelligence (AI) in human decision-making. Using a rich decision-level data set from the maintenance of heavy-duty trucks, we document how the repair decision-making of expert technicians changes with the introduction of an AI tool designed to predict the risk of truck breakdowns. We develop and estimate a dynamic discrete choice model of technician decision-making. The resulting estimates show that technicians with the AI tool exhibit a substantially better ability to predict breakdown risk than those without the tool. This improvement in predictive ability translates into better outcomes: The AI tool reduces the total costs that technicians incur by $240-$480 per truck per year. This brings the technician close to the efficient frontier; only 15% more cost savings could be achieved by further improvements in the quality of decision-making. The AI tool enables these cost savings by helping technicians avoid costly, unnecessary repairs.
Working Papers
Long-Term Relationships in the US Truckload Freight Industry (with Thi Mai Anh Nguyen)
Conditionally accepted at AEJ: Microeconomics
This paper provides evidence on relational contracting in the US truckload freight industry. In this setting, shippers and carriers engage in repeated interactions under contracts that fix prices but leave scope for inefficient opportunism. We describe empirically the strategies of shippers and the responses of carriers. We show that shippers use the threat of relationship termination to deter carriers from short-term opportunism. Carriers respond to the resultant dynamic incentives, behaving more cooperatively when their potential future rents are higher. While shippers and carriers often interact on multiple lanes, we show that separate relational contracts appear to govern transactions on each lane.
Long-Term Relationships and the Spot Market: Evidence from US Trucking (with Thi Mai Anh Nguyen)
Long-term informal relationships play an important role in the economy, capitalizing on match-specific efficiency gains and mitigating incentive problems. However, the prevalence of long-term relationships can also lead to thinner, less efficient spot markets. We develop an empirical framework to quantify the market-level tradeoff between long-term relationships and the spot market. We apply this framework to an economically important setting—the US truckload freight industry—exploiting detailed transaction-level data for estimation. At the relationship level, we find that long-term relationships have large intrinsic benefits over spot transactions. At the market level, we find a strong link between the thickness and efficiency of the spot market. Overall, the current institution performs fairly well against our first-best benchmarks, achieving 44% of the relationship-level first-best surplus and even more of the market-level first-best surplus. The findings motivate two counterfactuals: (i) a centralized spot market for optimal spot market efficiency and (ii) index pricing for optimal gains from individual long-term relationships. The former results in substantial welfare loss, and the latter leads to welfare gains during periods of high demand.
Research in Progress
Which Workers Benefit from AI? Estimating Heterogenous Effects on Productivity
This extension of my job market paper aims to explore heterogeneity in how technicians utilize a predictive AI tool in making engine repair decisions for heavy-duty trucks. By combining data on technician characteristics with rich data on repair decisions, this study seeks to address two pivotal questions: First, how might the quality of technicians’ decision-making vary with experience? Second, how does the introduction of a predictive AI tool differentially affect the quality of decision-making for technicians with different experience levels? The first question speaks to the returns to experience in this context. The second speaks to whether predictive AI tools act as complements to or substitutes for such experience. The findings aim to offer insights into the distributional impacts of predictive AI on professional human decision-makers, as well as potential effects on incentives for these decision-makers to invest in experience (i.e., human capital).
Long-term Relationships and Supply Chain Resilience (with Thi Mai Anh Nguyen)
Recent supply chain disruptions have highlighted the vulnerability of the goods economy to upheaval in freight transportation markets. In the US, the trucking industry may represent a particular susceptibility, both because of its singularly central role (72% of all domestic shipments are transported by truck) and because of its peculiar market institutions. As described in our first two papers, long-term relationships, rather than a centralized spot market, are the key means of arranging trucking transactions. This likely affects the ability of the industry—and thus, the US goods economy as a whole—to adjust to shocks. If transactions in this industry were arranged through a spot market, we would expect price signals to effect a rapid adjustment to shocks. However, in a world where transactions are actually arranged through a decentralized network of informal long-term relationships with prices that are (at least in the short-run) fixed, this may not be true. With this motivation in mind, this study analyzes—at the micro level—how shocks affect relationship stability and—at the macro level—how such shocks are transmitted through relationship networks.