Over the past two decades, millions of hectares of land in Africa have been transferred to investors, raising fears of displacement and conflict. This paper estimates the causal impact of large-scale land acquisitions (LSLAs) on local dissent by comparing successfully implemented projects to a control group of exogenously failed deals. Using staggered difference-in-differences estimators across 1,391 geocoded deals, I find that LSLAs cause a sustained increase in civic unrest of 158% relative to the pre-treatment mean. Protest responses are strongest among domestic investors acquiring community or state land for food-crop production, pointing to local dispossession and domestic elite capture. Integrating media, survey, and electoral data consistent with this hypothesis, I document parallel shifts in property-rights media discourse, an erosion of traditional authority, and broader electoral mobilization in affected constituencies.

This paper proposes a synthetic control (SC) framework for the estimation of conditional distributional treatment effects. Identification rests on a parallel trends condition formulated in the parameter space of the semiparametric distribution regression (DR) model, which keeps the counterfactual conditional distribution within the model class. The weights solve a least-squares problem subject to an adding-up constraint, yielding a closed-form estimator. We derive the asymptotic distribution of the counterfactual estimator, with DR estimation error and weight estimation error contributing at the same rate to the asymptotic variance. Moreover, we propose a supremum test for the null of no treatment effect, whose limit is the supremum of a Gaussian process. Simulations illustrate that conditioning on covariates can reveal effects being difficult to detect from the unconditional distribution alone. An application to the 1992 New Jersey minimum wage increase using CPS data finds effects concentrated in the minimum-wage corridor for low-education, low-experience workers.

We generalize the convergence results of an explosive autoregression, pioneered in Anderson (1959), in three ways: First, we demonstrate that the centered least-squares estimator converges geometrically to a ratio of limits, even in settings where the innovations are correlated and not centered around zero. Secondly, we demonstrate that the requirement of independent innovations in Anderson (1959), Theorem 2.3, can be relaxed to $α$-mixing. Third, we provide an autocorrelation-robust feasible test statistic for the explosive parameter under Gaussian ARMA innovations.

This paper presents a systematic framework for analyzing the economic parameter sensitivity of a hierarchical spatial economy within economic geography models. Through the hierarchical reduction approach proposed in this study, the original region-level governing equation is condensed into country-level and alliance-level equations. Based on the reduced governing equation, we formulate the sensitivity of economic variables on each country's population. This approach is applied to the analysis of international trade competition -- covering both trade liberalization and protectionism around Brexit -- among the UK, France, and Germany. We find that both the UK and the EU should focus on reducing domestic transportation costs, whereas tariffs and retaliatory tariffs act as a double-edged sword that can either strengthen or weaken their trade positions.

With the rise of new industries, often new jobs emerge. Evolutionary Economic Geography and in particular Industry Life Cycle perspectives predict that these activities first emerge in a limited number of cities to then diffuse to other locations as job descriptions become more standardized. Here, we focus on a particularly important new industry: software development, an activity that is economically important, quickly changing, and has a pronounced spatial concentration in a small number of global IT hubs. We use an online database of over 60 million questions and answers about problems in software development that yields a longitudinal dataset of 237 software skills. By geo-locating 3 million posting users at regular intervals, we link these skills to cities worldwide. We find that, in spite of its digital nature, the software industry exhibits similar spatial regularities as previously observed in more traditional sectors. First, cities diversify into skills that are related to their existing ones. Second, new skills first emerge in cities with large and diversified software sectors, and later diffuse -- mostly unhindered by geographical distance -- to smaller cities specialized in closely related skills. We find suggestive but limited support for a windows of locational opportunity account: although even brand-new skills still emerge first in cities with strong prior specialization in related skills, concentrations of related activities impact less the emergence of new skills than the diffusion of existing ones.

This paper axiomatizes the bid-ask market maker's quoting rule. A quoting rule maps the maker's state, namely inventory, belief, variance, trade intensity, and informed-trader fraction, to a bid-ask pair. Eight natural axioms, together with six environmental assumptions on the maker's inventory cost, force a unique three-parameter family: the mid-quote is linear in inventory, and the spread decomposes additively into inventory and adverse-selection components. Each of the three parameters is identified from a distinct moment of the observable quoting rule, with the three identifications mutually decoupled. The eight axioms partition into a four-axiom indispensable core, one structural choice, and three modularity extensions. Two structural corollaries follow: the latent inventory cost function is recoverable from the limit order book, and a sharp phase transition separates a functioning regime from a frozen one. A closing meta-theorem identifies four features invariant across all admissible structural primitives within the axiom system. To our knowledge, this is the first forced-uniqueness axiomatization of the quoting rule.

This paper provides partial identification and inference for treatment effects in nonparametric sample selection models with endogenous treatment and (weak) sample selection monotonicity. Outcomes are observed only for a non-randomly selected subsample and treatment is endogenous because of noncompliance with assignment. The proposed bounds for intensive margin treatment effects among compliers are sharp and tighter than those of Chen and Flores (2015). For inference, we develop semiparametrically efficient orthogonal moments and a debiased machine learning procedure that permits valid root-$n$ inference under high-dimensional covariates and/or flexible functional forms. Simulation results indicate good finite sample performance. Applications to Job Corps and the Oregon Health Insurance Experiment show that the method can deliver substantially tighter effect bounds and confidence intervals than existing alternatives.

Despite growing concerns over energy security, infrastructure planning and modelling for emerging green fuel supply chains often neglect risks from supply disruptions. Using a stochastic optimisation model of EU hydrogen imports, we show that 'naive' infrastructure planning results in welfare losses of 12 % (24 billion EUR) compared to risk-aware planning that anticipates supply disruptions. Despite requiring higher upfront investments, anticipatory planning achieves welfare levels close to those of an idealised system without disruptions, but entails a markedly different infrastructure configuration. Two complementary resilience strategies emerge: diversification across import corridors and strategic over-investment. This leads to increased intra-European transport capacity, a broader set of import pipelines, and investments in costly shipping terminals for hydrogen carriers. Our results show that incorporating supply risk considerations into infrastructure planning helps prevent the structural vulnerabilities seen in fossil fuel systems when designing future hydrogen supply chains.

Agentic AI systems can behave differently across runs: the same request may produce a different plan, a different tool call, a different code edit, or a different final answer. Such variability arises from several layers that are often conflated. A foundation model is a large pretrained model, usually adaptable to many downstream tasks, that maps an input context to predictions over outputs. In many current agents, that model is embedded in an orchestration loop that plans, calls tools, observes results, and updates state. One explicit intrinsic source of variability in such systems is token generation: the model computes scores over possible next tokens, the scores are converted into probabilities, and a decoder may sample tokens using a pseudo-random number generator. A small sampled token difference can then propagate upward into a different tool call, code path, search query, or agent state. Other sources of variability are extrinsic to token sampling, including changing environments, live data, serving infrastructure, batch effects, and numerical details. By separating these layers, the manuscript clarifies what it means to call agentic AI systems stochastic, when such variability can be reproduced under matched conditions, and why deterministic execution need not imply identical behavior in deployed settings.

We present a Lean 4/Mathlib formalization of the additive representation theory behind Classical Lottery in Action and the Wakker-Debreu-Koopmans (WDK) layer it relies on. Our central result is a machine-checked proof that the cross-pair Thomsen / double-cancellation (hexagon) condition is irreducible from the ordinal axioms of additive conjoint measurement (weak order, restricted solvability, Archimedean condition, and tradeoff consistency). We exhibit an explicit verified counter-model (additiveRealBoolPref) satisfying all ordinal axioms yet failing the cross-pair condition, with every strict standard sequence being an arithmetic progression and hence non-dense. Around this boundary we mechanize the full derivable construction: continuous Debreu/Eilenberg utility from separability, standard-sequence grids, bisection methods from connectedness, and global additive gluing. All public theorems are sorry-free conditional wrappers over this single irreducible structural input. The development is kernel-clean, depending only on standard Lean foundations (propext, Classical.choice, Quot.sound). The companion file ClassicalLotteryInAction.lean formalizes local classical-lottery constructions, average-utility results, matching-frequency lemmas, and ambiguity-attitude statements used by the Management Science paper. This draws a precise, machine-certified line between what additive conjoint measurement can prove and what it must assume.

We consider the fair division problem of indivisible chores and resolve the long-standing open problem for the existence of EFX allocations with additive cost functions. We show that, even for tri-valued additive cost functions, for every $n\geq 4$, there exists an instance with $n$ agents where no EFX allocation exists. Our counterexample only uses three types of chores, which is also tight on the number of types, as an EFX allocation is known to exist for two types of chores. We then consider bi-valued instances. We show that, for every $n\geq 4$, there exists an instance with $n$ agents where every EFX allocation is not Pareto-optimal. This is also the first example showing the incompatibility of EFX and Pareto-optimality when the costs of items are positive: existing examples showing the incompatibility of EFX and Pareto-optimal exploit items with $0$ costs. Our result shows such an example exists even for bi-valued instances. The number of agents $n$ is also tight: for $n\leq 3$, it is known that EFX is compatible with Pareto-optimality. Finally, we also show that an EFX allocation is guaranteed to exist for $n=4$.

Generative AI and large language models can produce realistic predictions of human behavior from rich, unstructured inputs with little to no task-specific training data. Recent work uses these ``digital twin'' predictions to supplement human responses in surveys and experiments. We study the special case of using AI-generated predictions to reduce variance in randomized experiments. We argue that doing so requires no new estimators and that researchers can simply include AI predictions as covariates in standard regression adjustment, analogous to adjusting for a prognostic score. A benefit of this approach is a ``do no harm'' property whereby the adjusted estimator reverts to the unadjusted difference in means when predictions are uninformative. Other methods, such as variants of prediction-powered inference, do not have this guarantee. We provide implementation guidance, including how to obtain continuous scores from discrete LLM outputs and how to use LLMs to featurize unstructured inputs as auxiliary covariates. We demonstrate these ideas in simulations and three empirical applications: a survey mega-study, an email marketing A/B test, and a large-scale technology platform experiment. Overall, efficiency gains are real if modest, with greater benefits in studies that contain substantial text and other unstructured data. We also confirm the do no harm property empirically. Given these gains and limited costs, we recommend adjusting for AI-generated predictions as a regular empirical practice.

We study the problem of auditing a black-box algorithmic decision-maker from observable inputs and outputs alone. Our main result is an exact decomposition: under precisely characterized conditions, the cumulative \emph{regret} of a dynamic policy equals the sum of per-period covariances between the cost vector and the policy's decision. This extends the single-period identity of Aldridge~(2026) to the full multi-period setting of stochastic dynamic programming. We prove the identity holds exactly under i.i.d. costs and mean-unbiased Markov policies, derive closed-form bias corrections for non-stationary and time-varying cases, and establish the discounted-horizon analog. A Bellman recursion for the covariance regret functional connects the result to standard reinforcement learning algorithms; for rolling-window policies, the estimation-error bias is $O(d/w)$. The decomposition has direct implications for algorithmic auditing in strategic environments: in platform mechanism design, it provides a welfare-based audit metric without access to the agent's private type; in repeated games, covariance reduction is a sufficient condition for policy improvement; in procurement and ad auctions, the bias correction quantifies welfare loss from strategic misreporting. The associated trajectory estimator is consistent, asymptotically normal with HAC variance, and computable in $O(T \cdot nd)$ time. This makes the proposed approach a tractable, model-free audit tool for platform mechanisms, algorithmic portfolio strategies, and any sequential decision system subject to external performance review.

We adopt the canonical polyadic (CP) decomposition to model high-dimensional tensor time series. Our primary goal is to identify and estimate the factor loadings in the CP decomposition. We propose a one-pass estimation procedure through standard eigen-analysis for a matrix constructed based on the serial dependence structure of the data. The asymptotic properties of the proposed estimator are established under a general setting as long as the factor loading vectors are linearly independent, allowing the factors to be correlated and the factor loading vectors to be not nearly orthogonal. The procedure adapts to the sparsity of the factor loading vectors, accommodates weak factors, and demonstrates strong performance across a wide range of scenarios. To further reduce estimation errors, we also introduce an iterative algorithm based on a novel double projection approach. We theoretically justify the improved convergence rate of the iterative estimator, and derive the associated limiting distribution. A consistent estimator of the asymptotic variance is also provided, which plays a key role in the related inference problems. All results are validated through extensive simulations and two real data applications.

This paper considers a semiparametric difference-in-differences (DID) framework for identifying and estimating treatment effects on the treated (ATT) when outcomes are missing not at random (MNAR), and a fully observed shadow variable is available. The shadow variable is assumed to be associated with the outcome evolution but independent of the missingness process, conditional on covariates and the possibly unobserved outcome evolution. We establish the identification conditions, derive the corresponding identification results and estimation algorithm, and evaluate the finite-sample performance of the proposed estimator through simulation studies and a real data application.

We study common-value auctions in which bidders have horizontally differentiated preferences. In a specific two-bidder parameterization, winning conveys good news about the object's value to the winner, a phenomenon we call the winner's bliss in contrast to the conventional winner's curse. Additional implications also differ from the conventional analysis. When bidders' preferences are horizontally differentiated, seller revenue is reduced with information disclosure, and advantageous selection sustains bilateral trade under asymmetric information.

We show analytically and via simulation that cross-sectional aggregation can substantially attenuate regime-switching signals in time-series data, making regime switches harder to detect. Building on this, we develop regime-switching models and an estimation algorithm which allow for autoregressive dynamics and grouped heterogeneity. We apply the approach to a U.S. macroeconomic dataset of 94 series, covering components of real gross domestic product, industrial production, capacity utilization, employment, and hours worked. The estimates give sharper business cycle classifications than those typically found in the literature. Monte Carlo simulations show that the computation is practical for datasets with a few hundred time series.

We develop a covariate-assisted approach to partially identified parameters that are solutions to an under-identified system of linear equations with known coefficients. Examples include bounds on treatment effects, models of unemployment with state dependence, choice-theoretic models of IV, and random utility models. The boundary (i.e., support function) of the proposed identified set is represented as an average of intersections of regression functions, aggregated over the covariate distribution. We show that the boundary is a regular parameter, propose asymptotic theory, and demonstrate using an empirical application to Jobs First.

We provide a strategic foundation for information: in any given game with incomplete information we define strategic quotients as information representations that are sufficient for players to compute best-responses to other players. We prove 1/ existence and essential uniqueness of a minimal strategic quotient called the Strategic Type Space (STS) in which a type is given by an interim correlated rationalizability hierarchy and represents a set of beliefs over other players' types and nature that rationalize this hierarchy and 2/ that the minimal STS has a recursive structure that is captured by a finite automaton.

Online hate speech is a global challenge amplified by engagement8-driven social media algorithms. This paper develops an epidemiological model of hate speech propagation capturing the strategic interaction between a profit-maximizing platform and a welfare-maximizing government. The platform's profit depends on the prevalence of hate speech and on its own algorithmic reactivity, creating a feedback loop between the epidemic and economic incentives. The government sets an optimal tax on amplification to internalize the social costs, balancing the benefit of tax revenue against the deadweight loss of taxation. The Stackelberg equilibrium is characterised analytically and solved numerically. The optimal tax reduces hate speech prevalence, eliminates bistability and lowers victim harm.

A well-known model in game theory, the Volunteer's Dilemma describes a group of $n$ players who decide whether to volunteer for a collective benefit at a personal cost, or to abstain and risk forfeiting the benefit altogether. A quantum version of this dilemma, developed within the Eisert-Wilkens-Lewenstein framework, allows each player to manipulate one qubit of a shared entangled state, leading to symmetric Nash equilibria with higher expected payoffs than in the classical game. Existing analyses, however, assume maximal entanglement. Within the same framework, we introduce a generalized Quantum Volunteer's Dilemma with a tunable entanglement parameter $γ$ and study the extent to which equilibrium behavior depends on the level of entanglement. We derive explicit conditions relating $γ$, the number of players, and the players' strategies under which symmetric Nash equilibria exist, focusing on two canonical strategy profiles: one for $2\leq n\leq 9$, and one for even $n$. We find that maximal entanglement is not required to sustain symmetric equilibria. Instead, equilibrium behavior persists above a threshold value, which we compute analytically in both cases. We also demonstrate that the threshold value directly depends on system size. This characterization is directly relevant for implementations on resource-constrained quantum devices, where entanglement is inherently limited.

Urban spatial structure is commonly evaluated through the spatial distribution of homes and jobs or through aggregate commuting outcomes. Yet these approaches do not reveal how the opportunities created by urban form are selectively transformed into actual residence-workplace connections. This study introduces opportunity-normalized residence-workplace matching by comparing observed commuting distance distributions with opportunity-based distributions constructed from all within-city residence-workplace pairs weighted by residential and employment mass. Using Output Area-level data for nine British cities, we show that realized pairings are systematically more concentrated at shorter distances than the urban opportunity structure alone would predict. After normalization, matching intensity declines with distance in a recurrent but heterogeneous pattern that is approximately linear in log-log space in many cities and can be summarized by a city-specific distance-decay coefficient. London further reveals that this regularity is scale-sensitive: a comparatively flattened citywide pattern separates into consistently negative but heterogeneous relationships across employment-centered subsystems. Supplementary evidence from New York and Chicago shows similar attenuation patterns. These findings identify realized residence-workplace matching as a distinct layer of urban structure and suggest that, in complex metropolitan systems, meaningful spatial regularities may reside in coherent matching fields rather than in aggregate city boundaries.

This paper proposes a Structural Matrix Autoregressive (SMAR) model for the joint analysis of asset returns, realized volatility, and trading volume in a large-dimensional setting. This framework simultaneously captures dynamic spillovers across financial variables and cross-sectional dependence across assets while preserving a parsimonious parameterization relative to conventional vector autoregressive models. The model is estimated on daily data for the constituents of the Dow Jones Industrial Average over the period 2021-2025 and is structurally identified through restrictions consistent with the Mixture of Distributions Hypothesis and efficient market theory. The empirical findings indicate that volatility is the primary driver of trading activity, suggesting that informational shocks are predominantly incorporated into markets through price variability. Forecast error variance decompositions further reveal that, although internal shocks dominate short-term volume dynamics, cross-asset spillovers account for more than 50% of trading volume variation at longer horizons. Finally, an event-study analysis around FOMC announcements supports the proposed decomposition by identifying significant increases in the informative component of trading activity on announcement days followed by rapid mean reversion.

This study investigates a statistical property of Lagrange multipliers in constrained Maximum Likelihood Estimation (MLE) and Least Squares (LS) problems from the perspective of numerical optimization. Building on large-sample theory, we show that the associated Lagrange multipliers converge to zero as the sample size increases, provided the distribution is correctly specified in MLE or the residuals are normally distributed in LS. Although this asymptotic behavior has long been recognized in statistics, it has received little explicit attention in numerical optimization and has rarely been exploited in algorithmic design. Importantly, the insight extends beyond classical low-dimensional settings: even in modern high-dimensional applications, such as deep learning, where the number of parameters may exceed the sample size, the same reasoning applies provided the generalization performance is good. This observation has two main implications. First, many constrained optimization algorithms, including the Augmented Lagrangian Method, Sequential Quadratic Programming, and Interior Point methods, require initial values for the multipliers, and choosing zero is statistically justified. Numerical experiments for constrained regressions and dynamic discrete choice model estimations support this implication by showing that initializing multipliers at zero usually lead to stable and efficient performance. Second, penalty-based approaches that convert constrained problems into unconstrained ones can perform well when the true multipliers are small. This helps explain why penalty-based methods often perform well in practice.

Traditional inference on the coefficient in an instrumental variables regression does not retain size when the instrument set is weak. With constant treatment effects or one instrument, the Anderson and Rubin (1949) AR test, the Klieibergen (2002)-Moreira (2003) LM test, and the Moreira CLR test provide robust alternatives which retain validity. Under treatment effect heterogeneity, no valid inference procedure exists in the overidentified setting. This paper develops the TSLS likelihood ratio (TLR) statistic, for performing inference on the TSLS estimand. When combined with a two-step procedure in the spirit of Berger and Boos (1994), it retains uniform validity across both the weak- and strong-instrument regimes. The procedure retains power with small choices of first-step level, hence the test can be constructed to numerically coincide with the Wald test in the strong-instrument limit.

How efficiently do markets update beliefs when public information arrives in rapid sequence? We use a real-time prediction market setting that combines binary payoffs, precisely observed public signals, and high-frequency market data, allowing us to compare market price changes with changes in a benchmark probability implied by publicly available information. We first show that prices are informative and become more accurate as resolution approaches. During the event, prices respond rapidly to public signals and move in the expected direction. However, directional responsiveness is not the same as efficient updating. Relative to an out-of-sample benchmark probability model, a one-minute change in the benchmark probability is associated with only about a 0.64-for-one contemporaneous change in market prices. The missing adjustment predicts future price drift over the following several minutes, including drift net of subsequent changes in the benchmark probability. We then study the mechanisms underlying this gradual adjustment. Salient public signals are incorporated relatively quickly in liquid markets, but the same signals generate substantially greater underreaction when liquidity is low. Underreaction gaps associated with salient states also predict stronger subsequent drift. The evidence therefore points to gradual price discovery shaped by the interaction between attention and trading frictions. The results contribute to the literatures on prediction markets, market efficiency, and behavioral finance. More broadly, they show that markets can aggregate public information quickly without necessarily incorporating it fully on impact. Market-implied probabilities are often directionally correct, yet adjustment remains incomplete and predictably depends on liquidity and salience.

This paper studies specification testing in dynamic linear models in the presence of omitted variables. The null hypothesis of interest is weak exogeneity: shocks have zero conditional expectation given their own past and the past of omitted variables. Existing tests based on quadratic forms of serial cross-correlations suffer from size distortions because their variance incorporates symmetric dependence in both directions, including causality from past shocks to present omitted variables (inverse causality). This paper proposes an asymmetric Portmanteau test that isolates violations of weak exogeneity from inverse causality, is asymptotically normal under the null, and does not require a parametric specification of the joint dynamics. An empirical application examines the Economic Policy Uncertainty shock series and rejects its weak exogeneity. Addressing this failure by controlling for omitted variables changes the estimated inflation response from negative to positive, suggesting a supply-side shock interpretation.

Generative AI raises short-term productivity by completing tasks that learners would otherwise practice on their own. Whether this substitution erodes frontier skill, the skill behind top-tail non-AI-aided performance, is an open question of rising stakes. The sharper question is whether selection mechanisms can screen apart two coexisting types: substitute-users, who use AI in place of deliberate practice, and complement-users, who use it to accelerate skill development. In elite programming, the International Collegiate Programming Contest (ICPC) and the International Olympiad in Informatics (IOI) prohibit AI under proctoring and admit entrants through qualification rounds, whereas online Codeforces (CF) contests are unproctored and open to all. From CF histories we build an AI-prompt signature (more first-attempt acceptances, fewer attempts and retries) consistent with AI-assisted practice. Three patterns triangulate institutional screening. First, CF practice shifted toward this signature across cohorts over two AI rollouts. Second, in open CF contests a stronger signature predicts smaller rating gains for users with no ICPC/IOI affiliation, but not for those who qualified for the AI-prohibited contests. Third, inside the AI-prohibited ICPC environment, a shift toward AI-style practice predicts higher non-AI-aided scores for AI-era entrants. The same practice input carries opposite signs depending on whether the environment screens for it. The contrast points to two levers: how AI is integrated into training, since within the screened pool AI-style practice coincides with stronger non-AI-aided performance; and the design of AI-prohibited evaluation gates as a type-separating institution. Both extend beyond programming to credentialing systems (medical and legal boards, professional certification) that certify skill in a workforce increasingly shaped by AI.

Subnational studies of aid effectiveness often rely on repeated cross-sections or nighttime lights, making it difficult to separate local treatment effects from baseline differences and potentially favoring infrastructure-heavy projects. We address these limitations by studying World Bank and Chinese development projects in Africa with a balanced panel of 2,166 DHS clusters across 35 countries from 2002 to 2013. Geocoded AidData projects are linked to satellite-imputed International Wealth Index estimates, a household-centered measure of material living standards. We compare a conventional two-way fixed effects (TWFE) event-study with the switcher--stayer estimator of de Chaisemartin and D'Haultfoeuille (dCdH), which avoids contaminated comparisons under staggered treatment timing. Pre-treatment diagnostics show that project placement is frequently selective: clusters that later receive projects often begin from weaker relative positions before treatment onset. Consequently, TWFE often implies larger post-treatment gains than the preferred staggered-treatment design supports. Under dCdH, the evidence becomes more selective and sector-specific. For the World Bank, positive evidence is strongest in Health, while Education shows positive but less cleanly identified gains. For China, Water Supply and Sanitation and Other Social Infrastructure and Services show positive associations with local wealth, although residual selection concerns remain. By contrast, Chinese Energy Generation and Supply appears strongly positive under TWFE but falls close to zero under dCdH. Overall, the results do not support a donor-wide claim that either the World Bank or China uniformly improves local wealth. Instead, estimated effects are concentrated in a limited set of donor--sector panels and depend strongly on how treatment timing, selection, and outcome measurement are handled.

This paper studies interdependent durations as equilibrium outcomes of a synchronization game, a continuous-time stopping game in which the incentive to stop increases when other players stop. We allow the payoffs to vary with both common shocks and observed and unobserved agent characteristics. The common shocks follow a spectrally negative Lévy process, a semiparametric process that includes Brownian motion as a special case but may also have jumps. We show that equilibrium outcomes can be represented as interdependent hitting times and use this to establish the game's nonparametric identification from data on stopping times and covariates. We develop maximum simulated likelihood and method of simulated moments estimators and evaluate their finite-sample and computational performance in Monte Carlo experiments. The results provide a tractable framework for identifying and estimating synchronization games from interdependent duration data.