Papers

This paper develops an econometric framework for analysing smooth structural change in cointegrated systems following a known intervention time. We consider a vector error-correction model in which the cointegration rank and the pre-intervention cointegrating structure are identified from a stable pre-intervention subsample. After the intervention, both the adjustment coefficients and the cointegrating vectors are allowed to evolve smoothly as functions of rescaled time, which are estimated using kernel-weighted local reduced-rank methods. The analysis is formulated directly in a cointegrated VAR/VECM system, which preserves the treatment of long-run relations and short-run error-correction dynamics. By working with the decomposition $\Pi(\delta)=\alpha(\delta)\beta(\delta)'$, the method separates changes in the equilibrium relation from those in the speed of adjustment. We also provide two tests for the parameter consistency and the post-intervention parameter smoothness respectively. An empirical application to energy market, foreign-exchange, and gold-market index around the 24 February 2022 Russia's invasion of Ukraine illustrates how the proposed approach distinguishes between a discrete regime shift and smooth post-intervention evolution. The results suggest that cointegrating relation among the price of Brent crude oil, the spot exchange rate (USD/EUR), and the Credit Suisse NASDAQ Gold Price Index has smoothly changed after the outbreak of war, instead of a constant long-run conintegration system in the pre-intervention period.

Success under a demanding method does not show that its additional demands are worthwhile. I study this distinction in a model of intergenerational learning with hidden effort. Moderate effort is informative about the value of a less demanding method; high effort produces more successes but no information about that question. I first characterize efficient experimentation with a complete record. I then show that parents who underweight descendants' effort costs can induce earlier retirement from experimentation. On an explicit region of priors, a paternalistic equilibrium stops after one failed experiment, whereas an aligned equilibrium allows two. Both transmit the relevant record truthfully and withstand communication deviations: every sender already obtains its preferred full-record continuation. Nevertheless, paternalism lowers discounted material welfare, raises long-run effort and success, and leaves less informative public evidence. Under stronger paternalism, parents may instead conceal knowledge to induce experimentation. I characterize a pooling continuation, establish its local communication stability, and show that it can occur in a full equilibrium. The results distinguish the efficient limits to learning from the additional loss of evidence caused by disagreement over effort costs.

A good representation of the population of firms and households is essential for large-scale economic models. While there exist good methods to create synthetic populations of households, creating synthetic populations of firms and, crucially, their supply chain links, is typically much harder. Here, we introduce a flexible method to create synthetic supply networks that match both the known properties of firm-level supply networks and the properties of aggregated input-output tables used in macroeconomic models. Our method is fast, and because it uses only publicly available data, it is fully reproducible and can be easily extended.

We argue that speculators preying on the closing rebalances of leveraged exchange-traded funds (LETFs) contributed to the Korean market's extreme volatility in 2026. An LETF's mandated daily rebalance is sized by the day's return, which generates an upward-sloping demand at market close. In response, rational speculators pre-position, enlarge the fund's order, and liquidate into the demand they have induced. Consistent with this mechanism, Korean stocks tracked by LETFs reverse about 75% of their first-day response to pre-open U.S. news by the next close and oscillate for several days thereafter, a pattern absent in every control group. Our quantification implies that self-reinforcing rebalance raised SK Hynix's annualized volatility from 100.0% to 136.7% over nine weeks and cost its products' predominantly retail holders 17.6% of their initial investment. Dispersing the rebalance across the trading day may backfire, whereas a flexible leverage multiple could help.

This paper examines the relationship between borderlands and political violence in Africa. Using spatiotemporal data on conflict events from 1997 to 2024 alongside the innovative OECD Spatial Conflict Dynamics indicator (SCDi), it suggests that borderlands see more conflict than non-borderlands. Political violence tends to decrease very rapidly with the distance to borders in Africa. The share of violent events recorded in borderlands varies over time and mirrors the cycles of violence that affect specific regions. It was very high in the early 2000s due to the wars in the Gulf of Guinea and Great Lakes region, declined for less than a decade, and increased substantially in the 2010s. Finally, the paper shows that when conflicts diffuse across borders, they tend to occupy a greater spatial extent, rather than relocate from one side of the border to the other.

This paper examines how political violence in West Africa is distributed between urban agglomerations and their surrounding areas using spatially disaggregated conflict and population data. Covering seventeen countries from 2012 to mid-2025, the analysis shows that political violence remains clearly tied to cities and their peripheries with the majority of violence occurring within and around urban agglomerations. Large cities record the highest number of incidents within their boundaries, while smaller towns, which are more frequently situated near borders, typically experience higher levels of violence outside their boundaries. However, urban-centric violence has also declined since 2012, indicating a progressive ruralization of conflict. These findings provide important regional context to the urbanization of conflict narrative by showing that West African insurgent movements use both urban and rural environments strategically. They also underscore the ongoing vulnerability of peripheral populations and the need for mobile and community-based security approaches outside major cities. Finally, they highlight the importance of strengthening border infrastructure, transport networks, and civic institutions as tools of both development and security.

We study how advertised products (e.g., Top Picks, Recommended, Featured) affect consumer choice on digital platforms and retail interfaces by extending the Luce (1959) (or multinomial logit) model. A consumer either focuses on the advertised items or considers the full menu, then chooses among the considered alternatives according to the Luce/logit rule. We characterize this model and show that its underlying primitives are uniquely identified from choice data. We also study a managerially important advertisement-design problem, in which a platform or retailer chooses the advertised subset to maximize expected profit, and we derive implementable design rules. We then introduce a richer framework in which advertising can influence both attention and preference. For this more general model, we provide a characterization and show how choice data can be used to separate the attention effect from the preference effect.

We develop a fully nonlinear structural vector autoregressive framework in which the contemporaneous structural mapping may be nonlinear and non-additive. Identification is achieved by exploiting variation in the conditional distributions of the mutually independent structural shocks induced by an observed exogenous variable. Specifically, a general contrastive learning framework that makes use of this variation together with the assumed exponential-family structure is employed to recover the shocks. Existing independent innovation analysis results identify such shocks only up to arbitrary componentwise invertible transformations, which is generally insufficient for structural econometric analysis. We strengthen this result by imposing a structured exponential-family specification for the conditional shock distributions. With the imposed sufficient statistics, the remaining ambiguity is reduced to a one-parameter transformed-scale map for each shock. We then show that, under a logistic specification used for the natural parameters, the identification is further strengthened up to permutation and componentwise sign changes. Once the shocks have been recovered, the fully nonlinear structural vector autoregression can be estimated using feed-forward neural networks, motivated by their universal approximation capabilities. The empirical application studies asymmetries in the responses of U.S. industrial production to the real oil price shock. We find modest asymmetries with respect to the sign of the shock and state of the economy. The accompanying R package iiasvar implements the introduced methods.

We consider a variant of the Assignment Game of Shapley and Shubik (1971), where agents do not observe the assignment or the surplus division of other matched pairs. We propose a set-valued solution concept (Self-Stabilizing Set) and characterize the largest such set. This leads to the formulation of the Stable Payoff Guarantee (SPG) set, which assumes that agents receive at least their payoff guarantees and that this is common knowledge. Our main result is that, for generic surplus matrices, the SPG set consists only of the efficient assignment. The associated payoff vectors are given by the smallest interval that contains the worker-optimal and firm-optimal stable payoffs.

Digital State Capacity is the ability of governments to deploy ICT infrastructure and information systems to implement policy. This paper introduces a new measure of government ICT capacity based on an observable stock of deployable public-sector network infrastructure: public IPv4 address space held by government organisations. These address holdings are key inputs into digital administration because they support internet-facing systems, networked information exchange, and coordination across agencies and functions. The core panel covers approximately 150,000 country-entity records classified as government across more than 150 countries from 2019 to 2024 and can be disaggregated by administrative level and government function. In the 2019 to 2024 Admin-1 panel, government IP holdings are observed in 1,681 subnational regions across all years. We validate the measure at the crosscountry and subnational levels and apply it to government tasks related to corruption control and vaccination rollout. In illustrative country-year analysis, higher Digital State Capacity is associated with higher-quality governance and publicservice outcomes in the expected directions, including lower measured corruption and higher vaccination coverage. These associations are descriptive; they demonstrate the empirical relevance of the measure and are not causal estimates.

When in the lifecycle does a large gender wage gap emerge? South Korea has the largest aggregate gender pay gap (29%) in the OECD. Among recent college graduates the conditional gap is only 3.3% over 2008-2019, falling from 3.4% to 2.5%. Correcting for differential selection into employment with semiparametric and machine learning methods leaves it unchanged. Nonparametric and semiparametric bounds contain the point estimates. Selection-corrected quantile regression shows that the remaining entry-level gap is concentrated at the top of the distribution. In Korean panel data the corrected gap widens severalfold across the prime-age workforce, where the selection correction becomes first order. Korea's gender wage disparity is mostly generated after labor market entry.

Artificial intelligence (AI) is increasingly integrated into medical decision-making, yet its liability implications remain complex, particularly when physicians differ in diagnostic skills and their quality is unobservable. This paper develops a principal-agent model in which a social planner designs medical liability to regulate a physician with private quality information who chooses between a standard treatment, a personalized judgment-based treatment, or following an imperfect AI recommendation. Our analysis yields several novel insights. First, we show that the optimal mechanism under asymmetric information is surprisingly simple: a uniform, one-size-fits-all liability level for all physician types who deviate from the standard of care. Despite physician heterogeneity, this simple policy often achieves the full-information first-best outcome, particularly when standard care is reliable or AI is highly accurate. Second, the relationship between AI accuracy and optimal liability is non-monotonic. Contrary to common intuition, better AI does not always imply more relaxed liability. As AI accuracy increases, the optimal liability either decreases monotonically or follows an inverted-U pattern, depending on the uncertainty of the standard treatment. Third, asymmetric information does not universally reduce social welfare. Welfare loss arises only when standard care is unreliable and AI accuracy is too low; even then, its magnitude follows an inverted U-shape, initially increasing as AI complicates the regulatory problem, but declining as more accurate AI helps mitigate it. Finally, we find that information asymmetry is a double-edged sword in the presence of AI, and greater transparency does not benefit all stakeholders equally.

Recursive nonlinear impulse responses require an estimated innovation law whenever the impact shock is normalized by innovation ranks and future innovations are integrated out. The closest semiparametric recursive construction in the literature estimates the relevant innovation quantile functions smoothly and discusses a direct empirical-residual implementation without developing its complete first-order inference theory. We tackle this gap in a finite-dimensional nonlinear structural autoregression with unrestricted continuous marginal innovation distributions and a fixed normal-rank shock. Our estimator replaces each innovation quantile function with the empirical quantile of generated structural residuals and iterates the same structural transition. For any fixed collection of responses, we establish a joint \sqrt{T} asymptotic linear representation with four components: direct transition estimation, the effect of transition estimation on residual order statistics, ordinary innovation-quantile estimation, and the shifted impact quantile. After projection through the recursion, the quantile terms admit a residual-rank-and-spacing representation, yielding feasible inference without innovation-density estimation or quantile smoothing. We then characterize the propagated bias from smoothing, establish validity of a full recursive residual bootstrap, and derive the additional covariance contribution from a finite number of simulated paths, providing bandwidth-free inference for the empirical-residual version of the same normal-rank response used in the smooth recursive construction.

Prediction-based methods, including Large Language Models (LLMs) and other machine learning techniques, are often used to construct measures of political phenomena that are difficult to quantify directly, such as policy positions in manifestos or emotions expressed on social media. In many applications, these prediction-generated measures are used as explanatory variables in regression models, even though they are measured with error. This leads to biased estimates. In this paper, we propose a simple solution to these biases: instrumental variables constructed from multiple measures created on independent splits of the original data. This approach is theoretically valid, easy to implement, and does not require new data. Through simulations, we show that this approach recovers estimates close to the true values, even in relatively small samples, while the standard approach can produce substantial bias in practice. We illustrate the method by revisiting two applications: whether gendered speech affects legislative outcomes in the German Parliament, and whether political risk influences poverty alleviation programs in China.

We introduce a framework for preference-robust decision making when preferences over risk are modelled through generalised distortion risk measures. Unlike distributional robustness, our approach addresses ambiguity in the risk functional itself. We construct ambiguity sets on distortion (weight) functions using the Wasserstein distance and Bregman divergences, and derive closed-form expressions for the worst- and best-case distortion risk measures. We further extend the framework to rank-dependent utility, yielding preference-robust behavioural models. In particular, rank-dependent utility appears as a robustification of the expected utility model, yielding a novel way to address the Allais paradox.

We revisit micro real rigidities as a source of monetary non-neutrality in a menu-cost model with variable markups, using firm-level evidence to pin down key primitives. We embed a non-CES demand system in a quantitative monetary model and use firm-dynamics evidence to identify demand curvature and firm-level productivity and demand processes. The calibrated model matches untargeted micro pricing moments, the markup distribution, and cost pass-through, while generating comparable non-neutrality. The key innovation is an empirically supported placement of idiosyncratic demand shocks that shifts residual demand, thereby moving desired markups and prices under non-CES demand. The broader implication is that this calibrated model provides a portable framework linking monetary economics with trade and IO evidence.

Hybrid platforms disadvantage third-party sellers through the platform fee and self-preferencing, and regulators have worried that constraining either instrument may intensify the other. We model a platform that chooses both instruments and find the opposite: single-instrument regulation is effective because the instruments are strategic complements, and regulating either instrument curbs the other. We also find that the two instruments achieve what monopolization achieves, higher prices and reduced consumer welfare, while passing every conventional antitrust test.

We present a simple Gaussian approximation to the finite-sample distribution of the classical ridge regression estimator. Our approximation captures the fact that, in finite samples, the ridge regression estimator trades off bias and variance to reduce estimation and prediction error. Our approximation is based on nonstandard asymptotics where $i)$ we let the estimator's regularization parameter grow proportionally to the sample size; and $ii)$ we treat the population regression coefficients as local to the reference vector that defines the estimator's direction of shrinkage. In contrast to other asymptotic approximations in the literature, we allow for general forms of heteroskedasticity and autocorrelation in the data generating process (at the cost of considering a low-dimensional model where the number of covariates is not allowed to grow with the sample size). We use our simple Gaussian approximation to propose two new strategies to select the regularization parameter for the ridge regression estimator. The suggested strategies select the regularization parameter to minimize either average or worst-case excess prediction risk, where risk is computed using our suggested Gaussian approximation.

We study prior-independent auction design when bidder values are independently and identically distributed and the seller knows only a scale-invariant shape restriction on their distribution, but neither the distribution nor the scale of values. We show that the maximin problem over a broad class of dominant-strategy incentive-compatible mechanisms reduces without loss to scale-free mechanisms. For any $n\ge 2$ monotone-hazard-rate bidders, the second-price auction without a reserve is maximin optimal over this class, including randomized mechanisms that may allocate to a lower bidder. We derive its exact guarantee for every $n$ and the sharp exponential rate at which its loss relative to the Bayesian optimum vanishes. Many familiar auctions are standard: they allocate only to a highest bidder, although incentive compatibility does not require this. For two regular bidders, we solve the standard problem exactly: its optimal mechanism mixes the second-price auction with a relative-markup auction and achieves a worst-case ratio of approximately $0.524413$. We construct a nonstandard mechanism that sometimes allocates to the lower bidder and achieves approximately $0.524829$, proving that standardness is strictly costly. The contrast is driven by tail restrictions: monotone hazard rate makes lower-rank allocation unhelpful, whereas regularity permits it to improve worst-case revenue.

Many technology firms now build software products in the open, inviting outside developers to contribute alongside their employees on platforms like GitHub. Does this openness in product development pay off? Analyzing 977 U.S. high-tech firms from 2001 to 2025, this study finds that open-source adoption raised firms' gross margins by 4-5% on average. These gains flow largely through higher labor productivity, as firms integrate external contributors' diverse knowledge into internal workflows, broadening the organizational knowledge base without a commensurate rise in labor costs. However, the payoff emerges only when outside volunteers supply a meaningful share of the work (around 35% in this sample), and hinges on the firm's resource configuration. While there are multiple pathways to profitability, pairing open-source product development with sustained internal R&D is a core condition present in all high-profitability configurations. For managers, the lesson is to treat open source as a knowledge-integration strategy matched to the firm's resources, not merely a means of sourcing external effort at lower cost. For policymakers, the findings underscore the broad economic value of open-source ecosystems and the need for policies that foster skilled volunteer communities and public-private open-source infrastructure.

Agentic AI is gaining acceptance in asset management, but governance has not kept pace: 88% of surveyed finance professionals report no operational governance framework for agentic AI, and only 24 of 75 large U.S. money managers disclosing AI use in Form ADV filings report a formal governance policy. We argue this gap is architectural: governance built for static validation does not survive continuously retrained agentic policies. We propose a four-layer framework (Policy, Engineering, Composition, Systemic) grounded in two distinct kinds of evidence, kept explicitly separate: two calibrated synthetic illustrations (a regret-covariance drift monitor; a crowding simulation showing joint drawdown risk rising from 39.2% to 79.3%), and three real, documented cases (a deployed LLM-embedding trading strategy, a \$45 billion discretionary fund's forced-deleveraging blowup, and a tribunal ruling holding an airline liable for its chatbot). The synthetic examples demonstrate computability from observable data; the cases demonstrate that the failure modes are not hypothetical. We provide a 90-day implementation sequence spanning trading and payments/customer-facing systems.

Creator economy platforms face a strategic dilemma: allocating traffic to established stars for immediate ad revenue versus nurturing emerging creators to build a follower base for future monetization. We develop a continuous-time dynamic optimization model to characterize the optimal traffic allocation policy for a platform managing heterogeneous creators with dual revenue streams (advertising and direct follower contributions). We characterize the optimal policy analytically, revealing a “most-valuable-creator-first" rule driven by a forward-looking activation set. Under Bass diffusion dynamics, this policy exhibits a sophisticated “conditional reversal" strategy, where the platform temporarily prioritizes lagging creators to capitalize on word-of-mouth effects. Regarding the ecosystem structure, we find the optimal policy acts as a selective gatekeeper. Unlike myopic policies that lead to a harsh “winner-take-all" market, or naive fairness-driven heuristics that foster inefficient “indiscriminate growth," the optimal policy imposes a strict capability threshold that screens which creators receive platform traffic. Furthermore, it enforces disciplined growth for successful entrants, capping their follower bases at an optimal ceiling to prevent over-investment. Finally, we demonstrate that simple heuristics can lead to significant revenue losses (up to 25%) and propose a practical “follower-growth adjusted" heuristic that achieves near-optimal performance by leveraging observed growth momentum.

As artificial intelligence transforms labor markets, understanding what makes workers adaptable has become increasingly important. Existing approaches typically characterize human capital using occupations, educational credentials, or predefined skill taxonomies, providing limited insight into how the structure of workers' skill portfolios shapes resilience to technological change. We develop an agnostic network based framework that reconstructs the hierarchy and diversity of skills directly from observed patterns of skill co occurrence. Using longitudinal data on 2.4 million United States workers and 16,753 distinct skills from LinkedIn, we introduce three complementary measures of skill complexity: specialisation, capturing productive depth; diversity, capturing adaptive breadth; and the diversity frontier, measuring the highest attainable diversity conditional on a worker's level of specialisation. We show that these dimensions predict distinct career outcomes. Specialisation is most strongly associated with sorting into higher wage occupations, whereas diversity is associated with broader skill accumulation and occupational mobility. Workers closest to the diversity frontier are significantly more likely to acquire new skills, receive promotions, transition into occupations with lower exposure to automation than workers with comparable levels of specialisation but narrower skill portfolios. These findings distinguish productive from adaptive capital and demonstrate that workers' adaptive capacity depends not simply on possessing specialised expertise or broad capabilities, but on combining both. More broadly, our framework provides a data driven approach for measuring workforce resilience and identifying reskilling pathways, offering new tools for understanding human capital in rapidly changing labor markets.

Modern economic panel data sets are often high-dimensional: they contain information on a wide variety of control variables whose number may even exceed the sample size. Nevertheless, the literature on econometric methods for high-dimensional panels is quite limited. In this paper, we study high-dimensional panel models with interactive fixed effects where the regression function has an additive structure, i.e., each covariate enters the model via an unknown nonlinear component function. We develop estimation methodology and theory in this additive framework which substantially extends previous work on the high-dimensional linear case by Ruecker et al. (2025). In the theoretical part of the paper, we derive the convergence rate of our estimator for both the small-T and the large-T panel case. The theory is complemented by comprehensive Monte Carlo experiments and an empirical application.

Monderer and Shapley (1996) showed that a game is a potential game precisely when the players' second-order cross-differences agree pair by pair. This paper asks what can be built from them when the agreement fails. The resulting MS-potential, assembled from their common-interest part, is unique up to separable payoff terms and exists precisely when a higher-order MS-condition holds; on exact potential games it recovers the potential up to the players' main effects. A least-squares construction extends the MS-potential to all finite games and induces the \MS-decomposition: every game splits into a common-interest MS-potential game and a residual absorbing every player's individualistic effects. Everything read off the second differences is invariant under the transformations that leave strategic content untouched, relabelling, non-strategic translation and action duplication, where the CMOP decomposition is not; only the least-squares extension fails, since it averages and centres. The central result is an identity: when all players have equally many actions, an augmentation of the MS-potential coincides, up to the additive constant, with the potential of Candogan et al. (2011). If action counts are unequal they diverge, and no bound on that divergence is established here. Both rest on the same uniform weighting of the players' actions.

We ask a representative sample to write prompts seeking spending and investing advice from LLMs, then simulate the lifetime effects of following the advice under realistic asset and labor market conditions. Applying this method to GPT-5.2, we find following the advice would move respondents toward life cycle theory: broader participation in diversified equity funds, age-declining equity shares, and larger savings buffers. Recommendations vary systematically by gender, prior AI experience, and financial literacy. For gender, two-thirds of recommended equity-share differences arise from men and women writing different prompts (demand), while one-third arise from gender labels attached to otherwise identical prompts (supply).

Per-impression auctions have long served as the allocation mechanism in online advertising. We argue that in “hands-off-the-wheel” (HOTW) markets, where advertisers declare budgets and “return-on-investment” (ROI) targets and the exchange's ML models predict click values, auctions are no longer necessary: all information required for optimal pricing is already known to the exchange, which occupies the position of a monopolist pricing against a downward-sloping demand curve. A HOTW market is a Fisher market, whose competitive equilibrium can be computed via the convex program of Eisenberg and Gale, yielding market-clearing prices and allocations satisfying all budget and ROI constraints simultaneously. This competitive-equilibrium price is revenue-optimal for the exchange among all uniform-price mechanisms: avoiding the demand reduction problem in typical uniform-price multi-unit auctions. The resulting competitive equilibrium is moreover outcome-equivalent to sequential first-price auctions with pacing, with pacing multipliers computable ex-ante by the exchange. This one-shot approach replaces millions of individual auctions with one convex program, which is not only operationally simpler than dynamically evolving bidding strategies, but revenue-optimal for the exchange, while delivering the same equilibrium outcome.

We develop a unified modeling framework that connects two distinct types of bubbles defined in the literature: the rational bubbles (aka P-bubbles), and the local martingale bubbles (aka Q-bubbles). We show that the local martingale bubble model includes the classical rational bubble as a special case. We relate both types of bubbles to an equity's risk premium via a novel decomposition.