Publications

Signaling is wasteful. But how wasteful? We study the fraction of surplus dissipated in a separating equilibrium. For isoelastic environments, this waste ratio has a simple formula: β/(β + σ), where β is the benefit elasticity (reward to higher perception) and σ is the elasticity of higher types’ relative cost advantage. The ratio is constant across types and is independent of other parameters, including convexity of cost in the signal. We show that the directional effects of β and σ on waste extend to non-isoelastic environments.

Solar Radiation Modification (SRM) has been proposed as a potential tool to limit increases in global or regional temperatures caused by anthropogenic greenhouse gas emissions. While previous research has extensively examined the climate system’s response to various SRM strategies, as well as their aggregate economic consequences, the regional distribution of economic impacts has received less attention. In this study, we use NorESM2–DIAM—an Earth System Model coupled to a high-resolution integrated assessment model—to assess the economic impacts, measured in GDP per capita, in an idealised SRM scenario where incoming solar radiation is reduced by 1%. Our results suggest that, relative to a baseline without SRM, most countries experience economic gains under SRM, with only a few countries facing negative impacts. Low-income countries tend to see the largest benefits, reducing global economic inequality relative to the baseline. However, reduced damages and lower inequality are accompanied by higher emissions under SRM, potentially leading to additional adverse effects not captured here. These findings highlight potential trade-offs between economic benefits, reduced inequality, and increased emissions relevant for SRM governance.

AI/ML methods are increasingly used in economics to generate binary variables (or labels) via classification algorithms. When these generated variables are included as covariates in regressions, even small misclassification errors can induce large biases in OLS estimators and invalidate standard inference. We study whether the bootstrap can correct this bias and deliver valid inference. We first show that a seemingly natural fixed-label bootstrap, which generates data using estimated labels but relies on a corrupted version in estimation, is generally invalid unless a strong independence condition between the latent true labels and other covariates holds. We then propose a coupled-label bootstrap that jointly resamples the true and imputed labels, and show it is valid without this condition. Two finite-sample adjustments further improve coverage: a variance correction for uncertainty in estimated misclassification rates and a Hessian rotation for near-singular designs. We illustrate the methods in simulations and apply them to investigate the relationship between wages and remote work status.

We document employment preferences of workers at the margin of informality using open-ended questions and discrete choice experiments in Brazil’s largest favela complex. Stated preferences emphasize pay and tangible job benefits rather than meaning or purpose, while primary complaints center on poor management, customers, and inflexible schedules. Workers exhibit high valuations for all formal sector amenities on average—unemployment insurance, parental leave, and termination notice—as well as for learning opportunities, but lower for non-formal sector amenities such as shorter commutes. Valuations vary systematically by employment sector in ways consistent with sorting: formal workers value formal amenities most, the self-employed value them least or not at all, and the informally employed exhibit mixed valuations. These patterns are also consistent with learning and endowment effects, for which we find suggestive evidence.

We study the design of efficient dynamic recommendation systems, such as AI shopping assistants, in which a platform interacts with a user over multiple rounds to identify the most suitable product among those offered by advertisers. Advertisers have multi-dimensional private information: their private value from a purchase and private information about the user’s preferences. In each round, the platform displays recommendations; the user learns product characteristics of the shown items and then chooses whether to purchase, exit without purchasing, or submit a new query. These actions generate a stream of feedback—purchase, exit, and follow-up queries—that is informative about the user’s preferences and can be used both to refine future recommendations and to design contingent transfers. We introduce a class of data-driven dynamic team mechanisms that condition payments on realized user feedback. Our main result shows that data-driven dynamic team mechanisms achieve periodic ex-post implementation of the efficient allocation rule. We then develop variants that guarantee participation and deliver budget surplus, and provide conditions under which these properties can be jointly attained.

We build a general equilibrium model in which firms endogenously choose whether to target prices or quantities. We characterize how these choices of organizational targets depend on firms' uncertainty about microeconomic and macroeconomic factors. In equilibrium, the transmission of both nominal and real shocks hinges on firms' organizational targets. For example, under otherwise identical microfoundations, money is neutral under quantity targets and non-neutral under price targets. We further characterize how targets shape firms' strategic interactions and prove that the macroeconomic uncertainty that arises from each choice of targets reinforces incentives to choose that target. That is, choices of organizational targets are strategic complements. For this reason, monetary policy aimed at stabilization can backfire by inducing a regime shift that renders it ineffective. A simple quantification suggests that incentives over organizational targets can vary markedly at business-cycle frequencies and help explain the state-dependent pass-through of monetary shocks to prices and output.

Bilateral bargaining under incomplete information provides a controlled testbed for evaluating large language model (LLM) agent capabilities. Bilateral trade demands individual rationality, strategic surplus maximization, and cooperation to realize gains from trade. We develop a structured bargaining environment in which LLMs negotiate via tool calls within an event-driven simulator, separating binding offers from natural-language messages to enable automated evaluation. The environment serves two purposes: as a benchmark for frontier models and as a training environment for open-weight models via reinforcement learning. In benchmark experiments, a round-robin tournament among five frontier models (15,000 negotiations) reveals that effective strategies implement price discrimination through sequential offers. Aggressive anchoring, calibrated concession, and temporal patience are associated with both the highest surplus share and the highest deal rate. Accommodating strategies that concede quickly disable price discrimination in the buyer role, yielding the lowest surplus capture and deal completion. Strategically competent models scale their behavior proportionally to item value, maintaining consistent performance across price tiers; weaker models perform well only when wide zones of possible agreement compensate for suboptimal strategies. In training experiments, we fine-tune Qwen3 (8B, 14B) via supervised fine-tuning (SFT) followed by Group Relative Policy Optimization (GRPO) against a fixed frontier opponent. The two stages optimize competing objectives: SFT approximately doubles surplus share but reduces deal rates, while RL recovers deal rates but erodes surplus gains—a tension traceable to the reward structure. SFT also compresses surplus variation across price tiers, and this compression generalizes to opponents unseen during training, suggesting that behavioral cloning instills proportional strategies rather than memorized price points.

This paper proposes a semi-endogenous growth theory that incorporates technology vintages and the endogenous evolution of multiple technological paradigms through innovation. It provides a characterization of both balanced growth equilibrium and transitional dynamics in an environment where new technologies continuously emerge. From a positive perspective, the model rationalizes two distinct empirical patterns. Using two centuries of US patent data, I first document that the age profile of patents has a pronounced hump shape: most contemporary patents build upon technologies that are between 50 and 100 years old. Second, this age profile has remained stable throughout the past century. From a normative standpoint, the theory underscores a misallocation of research effort induced by the tendency among profit-maximizing firms to overinvest in further developing mature technologies. This yields a suboptimally slow development of emerging technologies. According to a calibrated version of the model, correcting such misallocation could generate welfare gains of 7%.

This paper proposes a semi-endogenous growth theory that incorporates technology vintages and the endogenous evolution of multiple technological paradigms through innovation. It provides a characterization of both balanced growth equilibrium and transitional dynamics in an environment where new technologies continuously emerge. From a positive perspective, the model rationalizes two distinct empirical patterns. Using two centuries of US patent data, I first document that the age profile of patents has a pronounced hump shape: most contemporary patents build upon technologies that are between 50 and 100 years old. Second, this age profile has remained stable throughout the past century. From a normative standpoint, the theory underscores a misallocation of research effort induced by the tendency among profit-maximizing firms to overinvest in further developing mature technologies. This yields a suboptimally slow development of emerging technologies. According to a calibrated version of the model, correcting such misallocation could generate welfare gains of 7%.

Cross-country disparities in collateral technologies alone can account for large capital flows among mature economies, and allow the most advanced country to run a permanent trade deficit. When the collateral technology advantage is in creating negative beta (super safe) financial assets backed by positive beta assets, a Global Collateral Cycle emerges, with pro-cyclical gross and net flows and increased global asset price volatility. The supply of super safe assets is necessarily curtailed in downturns, providing a complementary (supply) channel to the flight to safety (demand) channel for explaining why US safe asset prices rise during crises.

As AI systems shift from directing users to content toward consuming it directly, publishers need a new revenue model: charging AI crawlers for content access. This model, called pay-per-crawl, must solve a problem of mechanism selection at scale: content is too heterogeneous for a fixed pricing framework. Different sub-types warrant not only different price levels but different pricing rules based on different unstructured features, and there are too many to enumerate or design by hand. We propose the LM Tree, an adaptive pricing agent that grows a segmentation tree over the content library, using LLMs to discover what distinguishes high-value from low-value items and apply those attributes at scale, from binary purchase feedback alone. We evaluate the LM Tree on real content from a major German technology publisher, using 8,939 articles and 80,451 buyer queries with willingness-to-pay calibrated from actual AI crawler traffic. The LM Tree achieves a 65% revenue gain over a single static price and a 47% gain over two-category pricing, outperforming even the publisher’s own 8-segment editorial taxonomy by 40%—recovering content distinctions the publisher’s own categories miss.

I estimate the effect of trade on local labor market concentration and its implications for wages using employer-employee linked data and tariff shocks from Brazil’s trade liberalization. Trade increased concentration by 7%, an effect driven by firm exit and worker flows to surviving import-competing firms. Increased concentration reduced wage take-home shares—estimated at 50 cents on the dollar pre-shock—enough to offset small wage gains from reallocation, but did not meaningfully reduce wages on net. Most of the wage declines attributed to Brazil’s trade liberalization resulted instead from reductions in the marginal revenue product of labor. Incorporating informality reveals substantial regional heterogeneity.

This paper studies how new varieties enter markets and become locally available. We provide causal evidence of demand externalities that operate in two steps. First, information about new varieties diffuses directly through real-world social ties among consumers. Second, early purchases generate an indirect spillover to firms: local retailers learn from 'pioneer' consumers which new varieties are most likely to succeed and adjust their product offerings accordingly. We study this process in the context of direct-to-consumer imports. Using customs records on individuals' purchases matched to population-wide social networks, international migrant links, and retailer catchment areas, we document economically meaningful demand externalities. Product-specific demand shocks abroad transmit through migrant networks and shift which varieties consumers purchase. Leveraging these shocks as a plausibly exogenous source of local demand variation, we show strong peer effects: prior purchases by close neighbors, coworkers, or friends increase an individual’s likelihood of purchasing the same variety, especially for premium and visible goods. We leverage this result to identify an indirect spillover from consumers to firms: retailers are more likely to add a variety when it becomes popular among consumers in their catchment area. Combining the instrument with linked consumer--retailer data and a self-conducted retailer survey, we show that this response reflects learning about latent demand for varieties not yet stocked locally. Together, social diffusion and retailer learning generate demand multipliers that reshape local product availability and expand access to global variety.

Here we provide our solutions to the First Proof questions. We also discuss the best responses from publicly available AI systems that we were able to obtain in our experiments prior to the release of the problems on February 5, 2026. We hope this discussion will help readers with the relevant domain expertise to assess such responses.

This paper examines the impact of early childcare on academic achievement for children in grade 5 and grade 9, based on a 2003 policy expansion that created quasi-random variation in slot availability for children aged 1–2. Starting childcare one year earlier increases math scores by 9.7% of a standard deviation (SD) in grade 9. Children whose mothers do not hold a high school diploma benefit by a significant 28% of a SD at grade 9, reducing the math achievement gap from children of higher-educated mothers by about one third. We also present evidence of strong improvements for children of immigrants.