Papers

Whether artificial intelligence pays off for the firms that build it into their products is hard to establish, because AI innovation is itself hard to observe. The medical technology sector is a rare exception: an AI-enabled device must obtain clearance from a national health authority before it can reach a patient, leaving a dated, firm-attributable record of AI innovation output that can be observed directly rather than proxied. We exploit this setting with a three-stage recursive model estimated on a novel firm-level dataset linking FDA premarket clearances, USPTO patents, Scopus publications, and Orbis financials, tracing the full innovation chain from external collaboration through AI device introduction to firm performance. We find that external AI research collaboration is a robust driver of AI device introduction across firm sizes and estimators, with a larger effect for small firms, consistent with external knowledge ties substituting for limited internal R&D capacity. Decomposing by partner type, the effect is largest for industry and clinical collaborations and smallest for academic ties, consistent with the former being closer to the regulatory and commercialisation process. Firms that bring AI devices to market display higher labour productivity, an effect robust for small firms and the full sample that holds under both sequential and joint maximum-likelihood estimation and accumulates across successive device introductions. Effects on profit margins are present but weaker and do not survive all specifications, a pattern consistent with competitive entry eroding pricing power as AI devices diffuse through the sector.

When social assistance is scarce, should it prioritize households in greatest need or those expected to benefit most? Using panel data from the China Household Finance Survey, we distinguish allocation principles by combining predicted policy gains with entry into China's Minimum Living Standard Guarantee (Dibao). We estimate heterogeneous predicted gains in consumption and education and then recover the conditional priorities revealed by recipient selection. Predicted gains explain little of allocation: a Shapley decomposition attributes 96.9% of the improvement in allocation fit to household priorities and 3.1% to predicted gains. Lower income, lower education of the household head, and elderly presence consistently predict higher priority across supported outcome-value specifications. Holding local program capacity fixed and removing household-priority differences while retaining the full model's estimated outcome values, the resulting ranking overlaps with recipients by only 10.9%, implying 89.1% recipient churn. These findings show that Dibao allocation is more closely aligned with household circumstances associated with poverty and vulnerability than with policy gains predictable from the outcomes and information observed in our data, underscoring the distinction between distributional and impact targeting in social assistance.

An open problem in approval-based multiwinner voting concerns whether we can efficiently compute committees that satisfy both justified representation and Pareto optimality. We answer this question negatively by proving that, on the domain of all profiles, outputting a committee satisfying both axioms is NP-hard. An initial proof was found by ChatGPT Astra. This was then verified and rewritten by the author.

Spatial treatments are interventions assigned to locations potentially distinct from those of the responding units. We study their optimal design under a general model in which a unit's response diminishes with distance to a treated site. Our estimand of interest is an “uncontaminated” effect equal to the average impact of a single intervention site over all hypothetical sites. We propose a novel design based on a Matérn point process which separates treatments by a distance of at least $r$. A larger choice of $r$ reduces bias by separating interventions but increases variance by reducing their numerosity. We choose $r$ to maximize the rate of convergence of a Horvitz-Thompson estimator and prove that this is minimax rate-optimal. We provide weak conditions under which the estimator is asymptotically normal and propose a variance estimator.

Bouchaud and Farmer (2023) argue that self-fulfilling beliefs generate a realistic wealth distribution. Their reported Gini coefficient, quantiles, and Pareto exponent are computed from dynasty-level wealth averaged over 250 dates, not from a cross section. Correcting the estimand strengthens their qualitative result: with $N=1{,}000{,}000$ dynasties, the terminal cross-sectional Gini is 0.883 rather than 0.7, and median wealth is 0.0052 rather than 0.39 of average wealth. The model therefore generates more inequality, but not the reported quantitative fit. Appendix F analyzes a different process and does not establish the claimed Pareto tail or exponent.

The twenty-first century's transformative technology, artificial intelligence, is increasingly constrained by the twentieth century's transformative technology, the electricity grid. Rapid growth in electricity demand from data centers is leading to higher electricity prices, without a compensating supply-side response. We develop a framework linking data-center load growth, available generation capacity, and market-clearing prices to understand this phenomenon. We first analyze a deterministic model to show how differing estimates of demand and supply growth rates affect prices. We then model the expansion of new data centers and their associated electricity demand, together with build-outs of new electricity supply, as stochastic processes,resulting in probabilistic distributions of supply, demand, and prices rather than a single forecast. Finally, we formulate generation expansion as a stochastic control problem in which a revenue-maximizing investor dynamically chooses the intensity of supply-side investments. The analysis highlights a central challenge of the data-center build-out: even when rapid demand growth increases the need for new generation, the uncertainties related to load forecasts, development execution risks, and value cannibalization from overbuilding capacity may weaken incentives to invest at the pace required to keep electricity prices stable.

Language models increasingly perform empirical analyses end to end, yet existing evaluations assess the written explanation or whether generated code executes, not whether the executed workflow recovers the intended causal estimand. We introduce CausalVerify, an execution-grounded benchmark for end-to-end causal analysis that follows a model from research-context interpretation to estimand recovery. It scores this workflow at four distinct layers: method recognition, design specification, executable implementation, and estimand recovery. It combines 259 real-paper contexts, 100 fixed-seed synthetic scenarios with executable reference estimates, and 23 paper-twin pairs in which a model commits to a design before seeing the data and its executed analysis is scored against a canonical estimator on the same realised dataset. Execution is not correctness. Among 426 model-written workflows that run without error, 15.5% fail verification, and a keyword score of the effect direction stated in the text is a poor proxy for recovery. In a 23-pair, nine-model paired study, replacing a model's committed design with the reference design and its execution conventions raises joint recovery of the point estimate and standard error from 15.0% to 51.5%; yet 48.5% of eligible seeds still fail under the reference design. The direction replicates on six pairs built afterwards under a frozen construction protocol, although on the two newest pairs the gain is confined to models from the family that built the references. Design specification is consequential but not sufficient: a plausible method and runnable code do not guarantee recovery, and even supplying the reference design and its conventions leaves substantial downstream failure. CausalVerify evaluates the executed workflow rather than its surface plausibility, and every reported number is recomputed from frozen artifacts by a single script.

Decentralized finance (DeFi) lending has grown from nonexistent in 2017 to nearly 40 billion US Dollars in deposited funds in May 2022. Using cryptocurrency as collateral, the platforms match speculative margin trading with yield-seeking depositors lending coins pegged to the dollar (stable coins). Depositors receive claims guaranteed by a basket of collateral, akin to new stable coins. We develop a framework requiring only knowledge of aggregate deposits and borrowings to measure overall system risks to lenders and borrowers. Using evidence from major protocols, the measures identify an increase in system fragility beyond prudent levels around mid 2021, with a potential loss of peg for extreme variations in coin prices. Overall, the model offers an easily implementable aggregate risk metric capturing the perspectives of synthetic investors and offers early warning signals as the industry is moving from deposits guaranteed by collateral to fiat money.

We study voluntary disclosure when investors observe firm reports through noisy information intermediaries such as auditors, analysts, rating agencies, or data providers. Any processing noise overturns the standard prediction of a unique partial-disclosure equilibrium. With low disclosure costs, the model unravels to full disclosure despite positive costs. With higher costs, the game admits two threshold equilibria featuring different disclosure probabilities. We characterize how the cost threshold for unraveling and the equilibrium set respond to changes in noise and fundamental uncertainty. In settings with high disclosure, both uncertainty and processing noise reduce disclosure, while higher certification costs can counterintuitively increase it. Endogenizing disclosure costs as optimal fees shows how profit-maximizing intermediaries select among equilibria, potentially generating a high-fee, high-disclosure regime. Extensions with bounded support, uncertain information endowment, endogenous noise, and competing information sources apply the insights to general information environments. The results caution against interpreting greater frictions as necessarily reducing disclosure.

Heavy-tailed and skewed outcomes are common in the randomized experiments and observational studies used to estimate heterogeneous treatment effects, yet the mean-squared-error criterion that guides splitting in honest causal trees is sensitive to the extreme values they generate. Building on the causal forest framework (Athey and Imbens, 2016; Wager and Athey, 2018), we introduce the Median Squared Deviation (MSD) criterion, which replaces the leafwise difference in means in the honest splitting objective with the Hodges–Lehmann location estimator while leaving honest leaf estimation and forest inference unchanged. Two further median-based rules, the Median Absolute Deviation (MAD) and the Least Median of Squares (LMS), serve as robust baselines. We evaluate the criteria in a simulation study covering precision, bias, and confidence interval coverage. MSD restricts its robustness to split selection and lowers the error of conditional average treatment effect estimates under heavy-tailed and skewed outcomes. Further, we re-visit two empirical applications: the first analyzes the electoral effects of a Mexican conditional cash transfer program on precinct-level observations, while the second application studies antiretroviral treatments in HIV-positive adults.

We study why organizations conduct interim performance reviews when monetary rewards are limited. An interim review creates incentive capacity by allowing future work and career opportunities to serve as rewards for past performance. Optimal review policies map a continuum of performance outcomes into a simple incentive ladder: termination, tough or easy continuation, and, for exceptional performance, an early maximal reward with no further work. Review can even sustain high effort when terminal compensation alone cannot. Its timing balances two forces: waiting improves the information revealed by performance, but leaves less future work available to motivate the agent.

We show that ordinary business language — "maximize profitability" — induces profit-oriented ambiguity resolution: LLMs systematically dismiss ambiguous signals of potential safety violations to serve business objectives. In 3,600 controlled trials across eight reasoning-capable LLMs, adding a profit mandate to otherwise identical prompts increases risk-dismissing judgments by 6.8 percentage points (p < 0.0001), suppresses board escalation recommendations by 13.9pp (p < 0.0001), and shifts severity assessments downward (p < 0.0001). The mandate never instructs models to downplay risks; instead, chain-of-thought traces reveal motivated reasoning: models acknowledge concerns, then invoke profit logic to justify dismissing them. We characterize these findings as the Profit Alignment Problem: when AI systems are given ordinary business objectives, they develop systematic strategies for suppressing inconvenient information that no designer intended or specified.

This paper studies the intensity of priority violations in matching markets. An agent is said to have very justified envy if she prefers another agent's match to her own and her priority advantage for the object exceeds $k$ ranks. I show that unless $k$ is prohibitively large, Pareto efficiency is incompatible with the elimination of very justified envy. A simple generalization of Deferred Acceptance and Immediate Acceptance is shown to eliminate very justified envy for any $k$. In simulations of random matching markets, I find that modest relaxations to justified envy can yield significant improvements in assignment ranks.

A/B tests are standard in firm decision making. In the standard pipeline, experimental data is converted to a deployment decision by applying a t-test of the difference in means (the lift) and deploying the treatment if lift is positive and statistically significant. This common workflow answers the wrong question. We argue that firms need a decision rule for economic payoffs in the future deployment environment, not a test of equality in the experimental sample. We develop an ambiguity-averse decision framework in which each arm is evaluated by its ambiguity-penalized value over distributions close to the experimental outcome distribution. The resulting rule has a simple closed form thanks to the Donsker-Varadhan representation and it requires only the outcome data from a standard A/B test plus one interpretable parameter governing trust in the experiment. Our rule is thus no more difficult to implement than a t-test. A mean-variance approximation shows how the rule penalizes variability, while a connection to utility maximization shows it to be a certainty equivalent. We are able to perform a real-world evaluation of our proposed rule in the context of digital marketing using an archive of 552 advertising experiments from an anonymous US-based online platform. The proposed rule substantially reduces regret relative to conventional hypothesis testing. The results show that economically conservative, distribution-aware deployment rules can outperform statistical-significance rules in digital experimentation.

Generative AI has become an everyday advisor, and the systems people consult are live and interactive, not pre-scripted. We ask whether access to such a system changes behavior under risk. In an incentivized experiment (N = 158), participants made lottery choices with an optional decision aid presented as a conventional pre-written tool, a live one-shot AI, or a live interactive AI they could query, with information format held equivalent across conditions. Risk preferences are elicited via DOSE. We find no evidence that access to a live AI advisor changes risk aversion.

We develop a filter for time series, defined at each time $t$ as the minimizer of a discounted convex combination of observed and expected losses. The filter can be estimated by simulation to an arbitrary level of accuracy in $O(1)$ flops at each time point $t$ and can be run for all values $t=1,...,T$ in parallel. These methods are applied to robustly compute a preaveraged price process from the more than 1.5 million trades made on a single financial asset in a single day where the noise's variance is infinite. It yields a flat "volatility signature" plot, down to the 1 second level, so the microstructure noise no longer biases the volatility estimate. This is not true when linear methods are employed.

Estimating the first stage of an instrumental variables (IV) model with the least absolute shrinkage and selection operator (LASSO) requires choosing a dictionary of technical instruments and a penalty level. First-order asymptotic theory offers no guidance on these choices, as any consistent implementation yields a structural parameter estimator with the same limiting distribution. In finite samples, however, these choices can have a substantial impact on the resulting structural parameter estimate. Working in a model with a single endogenous regressor and homoskedastic Gaussian errors, we use first- and second-order Stein identities to derive the approximate mean squared error (AMSE) of the instrumental-variables LASSO (IV-LASSO) estimator, which can be consistently estimated and used to rank a prespecified list of dictionary-penalty candidates. The AMSE reveals a bias-variance trade-off: more complex first-stage fits better approximate the conditional mean of the endogenous variable but are also more correlated with the structural errors, with complexity measured by the degrees of freedom of the LASSO fit. The weight on this bias rises with the endogeneity of the regressor, a quantity that neither plug-in nor cross-validation penalty rules take into account. Despite the AMSE being derived in a Gaussian model, penalty selection by minimizing the feasible AMSE criterion delivers up to a one-third lower mean squared error compared to cross-validation and plug-in penalty rules in Gaussian and non-Gaussian simulation designs calibrated to the data of Gilchrist and Sands (2016).

We review the evidence on subsidized job training programs in industrial countries. We provide critical, narrative summaries of randomized studies of major job training in the US; of two "judge randomization" studies in the US and Denmark; and of previous meta-analyses. We then perform a meta-analysis of 56 randomized trials of job training in the US since 1973, the most exhaustive meta-analysis in this niche. Here, average impacts are positive but modest: employment rises 1.7 percentage points in years 3-5, and annual pre-tax earnings by roughly \$800 per person offered training. At an average cost of \$13,598 per participant, programs roughly break even under (debatable) assumptions about how long benefits persist. They recoup about three-quarters of their cost for government through higher tax revenue and lower benefit use. Evaluations of the largest federal efforts–JTPA, the Workforce Investment Act, and Job Corps–return small or null effects. One cluster of programs stands out: "sector programs," which screen applicants intensively, design curricula with employers, and track local skill demand. They raise earnings roughly ten times as much. They have proven difficult to replicate.

Sign restrictions on the slopes of supply and demand curves are often used to identify historical decompositions in structural vector autoregressions. I show that the identifying power of these restrictions depends on both reduced-form parameters and realised forecast errors. Consequently, unlike many other structural objects, the strength of identification cannot be assessed from reduced-form parameters alone. Empirically, identified sets for historical decompositions of US inflation are typically largely uninformative, both in aggregate and in most expenditure categories. Existing inflation decompositions are therefore sensitive to auxiliary assumptions used to select among observationally equivalent models.

We develop an expected shortfall factor model (ESFM) to estimate and price common variation in the severity of lower-tail losses in large panels of asset returns. Mean factor models describe common variation in average returns, while quantile factor models describe common movements in tail thresholds. ESFM instead captures common variation in the average severity of losses below those thresholds. The model combines observed risk exposures with latent common factors. We estimate ESFM using an orthogonalized two-step procedure under which first-stage quantile estimation error has no first-order effect on the ES coefficient estimates. We establish nonasymptotic error bounds for the ES coefficients, a finite-sample Gaussian approximation, and consistent selection of the number of latent factors. Applied to a large panel of equities, ESFM uncovers common factors that react sharply to market stress and contain information not captured by mean and quantile factors. Average returns increase across portfolios sorted on ESFM exposure; high-minus-low portfolios earn annualized returns of 8.0%–11.7% and Fama–French five-factor alphas of 10.3%–15.0%. These spreads remain positive and statistically significant after conditioning separately and jointly on mean- and quantile-factor exposures. Tail-by-tail spanning tests show that ESFM factors retain significant alphas after controlling for standard traded factors and the corresponding mean and quantile factors. Adding ESFM to these benchmark factor sets increases the maximum attainable Sharpe ratio. These findings identify common loss severity as a distinct and priced dimension of downside risk.

Roughly fifty thousand families in England appeal their school assignments each year through a centralized and understudied procedure that improves the placement of about one in five claimants. This paper shows that appeals are not an institutional afterthought: they change how parents report preferences to the education authority and the quality of placements it generates. Once the appeal stage is incorporated, Immediate Acceptance (IA) becomes less manipulable and can support equilibrium outcomes that Pareto dominate those of Deferred Acceptance (DA) under truth-telling, reversing the DA-IA welfare comparison. In a preregistered experiment, we find results consistent with the theoretical predictions: appeals increase truth-telling in IA, leaving DA essentially unchanged, and widen the IA-DA efficiency gap. Around 60% of the IA efficiency gain appears before any appeal is upheld, because subjects anticipating the possibility of a potential appeal rank schools differently.

Although social mobility is widely viewed as desirable, we document an 'inverse-U' shaped relationship with well-being: mobility is positively associated with well-being when low, but the association diminishes as mobility increases and eventually becomes negative. To explain this, we build a model in which agents are embedded in a social comparison network and suffer from falling behind their peers. The model highlights two channels that can cause welfare to be decreasing in mobility at high levels. First, when mobility is high, further increases widen already significant gaps between agents from similar parental backgrounds but of different abilities. Second, when mobility is high, high-ability agents are already relatively well off. So further increases in social mobility increase inequality. We find evidence supportive of both channels in US data. The model also predicts, and data supports, that social integration – where relationships form more along ability than background lines – allows societies to benefit from mobility for longer. We interpret this as social mobility being multidimensional with both economic and social dimensions that are complementary.

Heterogeneous-agent New Keynesian (HANK) models characterize how entire cross-sectional distributions respond to structural shocks. Traditional representative-agent models are routinely disciplined by impulse responses from aggregate vector autoregressions (VARs). HANK models have no comparable established empirical benchmark because they make predictions not only about aggregates, but also about distributions of micro-level data. We propose a Bayesian benchmark that jointly models macroeconomic aggregates and several marginal distributions from repeated cross sections, including distributions observed in different surveys. Our approach can use both standard structural VAR identification approaches on macroeconomic aggregates and identification restrictions imposed on micro-level data. The model delivers a joint posterior of the distributional effects of shocks, without the need for household panel data or a separate first-stage density estimate.

Sports betting moves money continuously from a large public to a small number of firms. The most influential account of that flow, due to Levitt (2004), holds that books price away from the market-clearing point to exploit predictable public biases. It computes profit from two numbers (the probability that a side wins the proposition and the fraction of handle it attracts), treating the share on a side as independent of the outcome. Here we relax that assumption and derive a profit-bias identity: profit is affine and increasing in the expected share of handle on the losing side, with Levitt's expression as the special case of independence. The identity resolves the book's margin into exactly three channels: its hold, the product of its price shading and the public's lean, and the covariance between bet share and outcome. A lean is thus worthless without shading, and shading is worthless without a lean. Under a public-belief model, the profit driver is the public's Bayes error: the probability of a representative bettor selecting the losing side. We provide necessary and sufficient conditions for a "Goldilocks Zone": prices at which book and bettor both profit. Testing these predictions on 1,139 Major League Baseball games, we find that the apparent dependence between bet share and outcome is a Simpson's paradox: present when games are pooled, but absent once they are separated by which side the book favored. The public leans heavily toward favorites, but we detect no matching shading, and the realized margin is indistinguishable from the hold.

This study investigates the spatial relationships among housing-market appreciation, housing affordability, and automobile commuting to Toronto across 23 census metropolitan areas and municipalities in southern Ontario from 2016 to 2021. Utilizing Statistics Canada Census data, the analysis employed linear and quadratic ordinary least-squares regression, Global and Local Moran's I, and residual spatial diagnostics to assess geographic patterns in average dwelling values, owner shelter-cost burden, and lone-driver commuting. The study found that average dwelling values increased significantly throughout the region, maintaining a strong nonlinear relationship with distance from Toronto, which indicates a persistent metropolitan accessibility gradient. Commuter volumes also demonstrated pronounced nonlinear distance gradients; however, housing appreciation was not significantly correlated with percentage changes in commuting ($R^2$ = .045, p = .331), suggesting that the expansion of the housing market did not result in proportional growth in Toronto-bound automobile travel. The shelter-cost burden exhibited weaker distance relationships and greater residual spatial dependence, indicating localized affordability influences not fully explained by distance or dwelling values. Overall, the findings suggest an increasingly extensive regional housing market with spatially heterogeneous affordability and commuting outcomes, highlighting the importance of integrating housing, transportation, and regional planning.

Optimal transport provides a common language for allocation, equilibrium, computation, and inference. Its primal problem assigns mass or agents, while its dual variables admit economic interpretations as utilities, prices, and scarcity rents. This review explains why this combination has proved unusually effective in economics. We first present the core results, including Kantorovich duality, integrality, cyclical monotonicity, the canonical distance and quadratic costs, and entropic regularization. We then trace the field's development from planning and operations research to modern analysis, statistics, and computation. The economic literature is organized around two roles for transport: as a model of matching, trade, hedonic equilibrium, and aggregate assignment; and as a tool for coupling distributions, measuring discrepancies, constructing multivariate ranks, solving inverse problems, and certifying economic conclusions. We conclude by examining extensions beyond transferable utility, one-to-one matching, static allocation, and unconstrained transport, together with open questions involving learning and identification across multiple markets.

I propose a novel, tractable model of pro-social norms in large communities with slightly altruistic agents. Agents participate in the norm to influence others to participate in the norm, influencing further agents. A continuum of equilibria sustain the pro-social norm, varying in the distribution of how much agents are influenced by their observations. If agents' effective altruism $\alpha$ is larger than their impatience $1-\delta$, equilibria exist that are robust to a population of bad actors. Greater strategic homogeneity increases the robustness of the norm. Robustness is not improved by increasing the number of observers of each action beyond 1.

The relationship between investment policy associated with species growth profile is essential in seeking the most appropriate strategy for a policymaker. Balancing maximum profit with the sustainability of species remains a central issue in both ecological and economic contexts. This study presents a comparative analysis of two interest principles, annuity and compounding, within the framework of optimal control. Various investment policies are examined from the perspective of capital theory, incorporating concepts such as future value, accumulation function, and force of interest, each contributing to the formulation of an optimal control strategy. Our analysis is based on a one dimensional logistic model incorporating linear harvesting. Key parameters include the species growth rate and interest rate, and optimality is evaluated with respect to these variables. Using Pontryagins Maximum Principle, we derive the optimal harvesting policies under both discounting laws and characterize the resulting steady state equilibria. The principal finding indicates that for species with low intrinsic growth rates and low annual interest rates, the annuity law of interest yields optimal outcomes. Conversely, for any annual interest rate, species exhibiting moderate or high growth rates maximize profit under the compound law of interest. The study also addresses maximum net revenue and optimal strategies for varying growth rates under each interest law.

This paper develops fixed-smoothing (fixed-b, fixed-K) inference methods for time-series quantile regression that are robust to heteroskedasticity and autocorrelation. Our approach is uniformly valid over quantile levels and accounts for dependence both over time and across quantiles. It enables the construction of uniform confidence bands, Wald, and Sup-t tests for joint hypotheses, and tests of shape restrictions, providing a unified framework for assessing heterogeneity in quantile effects. A key challenge is that, under weak dependence, uniform inference for quantile regression processes is generally non-pivotal because the limiting distributions depend on the long-run covariance structure across quantiles. To address this issue, we develop two complementary approaches. The uniform-in-$\tau$ method estimates the covariance structure and simulates the non-pivotal limiting distribution. For certain tests involving a finite collection of quantile levels, the stack-Wald method delivers pivotal fixed-smoothing inference. We establish the asymptotic validity of both approaches. Simulation results show that the proposed methods substantially improve size control relative to existing HAC-based procedures while maintaining good power. An application to predictive quantile regressions for stock returns reveals substantial heterogeneity in predictive effects across both quantiles and forecast horizons.

Why China Succeeds offers what no existing book does: a single, unified explanatory framework that traces China's rise from its ancient civilizational foundations to its twenty first century technological ascent. Rather than treating China's success as ideologically exceptional or historically accidental, it reveals the universal institutional logic (regulated social mobility, functional land and taxation systems, and formidable state capability) that has sustained Chinese governance across millennia and powered its modern economic miracle. The book directly challenges Western development orthodoxy by demonstrating that institutions can be built through growth, not before it. For policymakers, scholars, and development practitioners worldwide seeking transferable lessons from the world's most consequential development story, this book provides the most historically grounded, analytically rigorous, and forward-looking account available.