We propose a semiparametric local projection estimator of nonlinear impulse response functions for a broad class of structural dynamic models relevant for applied macroeconomics, including models with nonlinearly transformed regressors, state dependent coefficients, and nonlinear interactions between shocks and state variables. The estimator is based on a doubly robust moment condition that identifies the average response function as a linear functional of a nonparametric conditional mean, augmented by a density ratio that captures the effect of shifting the shock of interest. We combine this moment condition with cross-fitting that handles serial dependence. The resulting estimator is $\sqrt{T}$-consistent and asymptotically normal. We examine the finite-sample performance of the estimator across a range of nonlinear data generating processes and illustrate its use in two empirical examples.

This paper provides a practical guide to estimating semiparametric and nonparametric fixed-effects panel data models using the mgcv package in R. The focus is implementation: handling fixed effects with unit indicators, first differencing, or penalized unit effects; specifying smooth terms; and conducting cluster-robust inference. Monte Carlo experiments compare \code{mgcv::bam} estimators with linear and fixed-series spline estimators. Simulations suggest that penalized splines adapt to unknown smoothness and estimate functions accurately in the designs studied here. A penalty-adjusted cluster-robust covariance estimator yields tests with near-nominal size for finite-dimensional parameters, and confidence bands provide accurate coverage for centered unknown functions.

We study strategic interaction when players delegate belief formation to predictive machine learning (ML). In a static Bayesian game, each player's ML agent predicts a payoff-relevant outcome variable as a function of the player's type. The ML agent's training sample is endogenous: it is drawn from the outcome distribution generated by players' ML-guided behavior. In Cross-Validation Equilibrium (CVE), each player's ML agent selects a predictive model to minimize expected out-of-sample squared error, given its realized training sample, and each player best-replies to the belief generated by the model her ML agent selected. We analyze CVE and relate it to other equilibrium concepts. We apply CVE to jury voting, speculative betting, and games with linear-quadratic payoffs. E.g., in a team-effort game, endogenous model selection can give rise to multiple equilibria.

We examine the continuity of equilibrium correspondences in infinite-dimensional settings where the commodity spaces are Banach lattices. Economies are modeled as Borel probability measures on a space of characteristics, with aggregate endowments defined via Bochner or Gel'fand integrals. Within this framework, we prove that the equilibrium correspondence is continuous on a dense subset of the domain of economies admitting equilibria, endowed with a suitable Polish topology. These results extend both classical and recent continuity theorems by providing a unified analytical treatment applicable to a substantially broader class of locally convex spaces and encompass models with infinite planning horizons, monopolistic competition, neoclassical economies, financial equilibria, and asymmetric information. Importantly, this study demonstrates that there is no necessity to impose differentiability assumptions that are typically required in regular economies to study equilibrium continuity.

This paper analyses the distinction between educational and skill types of labour mismatch and their association with earnings. Drawing on cross-sectional data for 26 countries from the 1st Cycle of the OECD (2012) Survey of Adult Skills (PIAAC), I examine educational and skill mismatch using a comprehensive set of education- and skill-based indicators, explore heterogeneity across worker characteristics, and investigate the sources of conflicting country-level correlations with earnings through an error components model. The results show that country-level unobserved heterogeneity induces endogeneity bias, with both its direction and magnitude varying across mismatch measures. Once unobserved heterogeneity is controlled for, over-education and over-skilling are associated with wage penalties, whereas under-education and under-skilling are linked to wage premiums. These findings highlight both conceptual and empirical distinctions between educational and skill mismatch and demonstrate the importance of indicator choice in the analysis.

Who is exposed to generative AI in a developing-country labour market? We map three occupational AI-exposure indices to India's redesigned Periodic Labour Force Survey (2025) and document a steep caste gradient among 83,000 employed graduates: graduates from the Scheduled Castes and the Scheduled Tribes are 0.24--0.37 standard deviations less exposed than upper-caste graduates within the same district. Two channels drive the gap: one in four SC and one in three ST graduates work in farm or elementary occupations untouched by AI, and those in white-collar work are underrepresented in managerial, software, and finance occupations. Because exposure commands a wage premium of up to 20 per cent, generative AI stands to widen, not narrow, India's caste earnings gap.

In the fall of 2020, neural-network methods produced a large improvement in chess engines that became freely and widely available. By the end of 2021, the monthly draw rate in classical chess had risen by about four percentage points, but the distribution of player ratings, which are commonly read as measures of playing strength, had changed little. Ratings, however, are a relative measure, built from results against other rated players rather than from an absolute scale of play quality, so an improvement shared broadly across players need not change their ratings. Using 3.9 million rated classical games from March 2015 to November 2023, we document that the increased draw rate remains after conditioning on both players' ratings, holds within repeated same-color matchups, is not a continuation of a pre-existing trend, and persists through the end of the sample. A linear transformation that maps post-Covid ratings to higher pre-Covid equivalents, with a larger gap at lower ratings, accounts for more than 90 percent of the post-minus-pre shift in the fitted draw, White-win, and Black-win probabilities. Players' ratings and ranks, by contrast, show no additional rank reshuffling and no general widening of within-group dispersion relative to the pre-Covid benchmark. We interpret these findings as consistent with adoption across rating levels, with larger rating-equivalent gains for lower-rated players.

We estimate the causal price elasticity of gas demand on Ethereum mainnet (L1) and Arbitrum One (L2), a quantity necessary for calibrating fee mechanism simulations, evaluating resource pricing reforms, and explaining observed usage patterns. A two-way fixed effects panel regression instrumented by each wallet's own lagged base fee removes the congestion-driven endogeneity that causes naive regressions to substantially underestimate demand sensitivity. On Ethereum mainnet (full year 2025), the pooled IV elasticity is -0.006***, near-inelastic: a 10% fee increase reduces total gas demand by approximately 0.06%. On Arbitrum One (October 2025--April 2026), the pooled IV elasticity is -0.036**. Both chains are inelastic in the aggregate, with L2 measurably more responsive than L1. A per-resource decomposition of L2 demand reveals elasticities ranging from modestly elastic computation (-0.027*) to -0.27*** for refunds, with storage growth (-0.15***) and calldata (-0.06*) in between. Behavioral clustering identifies always-on protocol wallets as near-inelastic and high-volume operators as substantially more responsive, with cluster-level elasticities up to roughly 6x the pooled estimate. These results establish an empirical foundation for downstream simulations and for evaluating fee mechanism designs.

Artificial intelligence (AI) systems increasingly assist human experts, but the consequences of AI assistance on productivity can be heterogeneous. Caplin, Deming, S. Li, Martin, Marx, Weidmann, and Ye (2025b) provide evidence that two characteristics, ability and belief calibration, help to determine the returns to AI assistance. This note shows that their results replicate to a setting where professional radiologists analyze chest X-rays with access to state-of-the-art machine learning predictions. I leverage the public Collab-CXR data repository described by Moehring, Kutwal, Huang, Banerjee, Jacobi, Eber, Mendoza, Chung, Dayan, Gupta, Bui, Truong, Pareek, Langlotz, Lungren, Agarwal, Rajpurkar, and Salz (2025) and first analyzed for human-AI collaboration by Agarwal, Moehring, Rajpurkar, and Salz (2023). To faithfully reproduce the analysis in Caplin, Deming, S. Li, Martin, Marx, Weidmann, and Ye (2025b), I use the radiologist assessments from the repeated-case designs, which include 68 radiologists and 11,420 paired radiologist-patient-pathology observations. The results of this replication support the external validity of their core findings: lower baseline ability and higher calibration predict larger incremental value from AI.

Large language models (LLMs) are increasingly used for tasks once reserved for trained researchers, including hypothesis generation, specification choice, and drafting conclusions. We argue that the reliability of AI-assisted research depends not only on model capability, but also on how cognitive labour is structured between humans and machines. We study this problem through Human-in-the-Loop Economic Research (HLER), a decision architecture based on pre-commitment, decision sequencing, accountability, and attention allocation. In a pre-specified 2*4 factorial experiment with 280 complete research runs across four datasets, an unconstrained multi-agent baseline produced critical failures in 72% of runs. Using the same underlying model, the same agent decomposition, and identical prompts for the shared reasoning agents, HLER reduced the failure rate to 16% by imposing three architectural commitments: LLMs reason but do not execute data work, data and estimation are handled deterministically, and three human decision gates bind the workflow. Fisher's exact test rejects equality of failure rates at p<0.001. Reliability gains were largest on the least publicly represented dataset, a Qing-dynasty population register, consistent with a task-based production model with Frechet-distributed output quality. An 80-run ablation suggests that deterministic computation and human gates contribute independently, with exploratory evidence of complementarity. We interpret HLER as a research harness rather than an autonomous AI scientist: it sharply reduces failures, makes residual weaknesses more visible, and prevents unreliable claims from being advanced as publication-ready outputs.

As the EU Carbon Border Adjustment Mechanism (CBAM) approaches, the global semiconductor value chain faces growing structural tensions between regulatory transparency and data sovereignty. This article proposes a RegTech reference architecture using the International Data Spaces (IDSA) framework to orchestrate trustworthy environmental telemetry across the semiconductor-petrochemical nexus. The framework distinguishes the mandatory CBAM requirements from voluntary Science Based Targets initiative (SBTi) frameworks, while addressing the additive complexities of the Safe-and-Sustainable-by-Design (SSbD) framework. Moving beyond standard linear technology stacks, we introduce a prospective roadmapping methodology that transforms upstream physical vulnerabilities into circular, negative feedback loops. Focusing on the Taipei and Penang technology corridor, the article details how sovereign data exchange enables Digital Product Passports (DPPs) to drive Global Business Services (GBSs) capability demands. Finally, we discuss the integration of Agentic AI for autonomous compliance and FinTech green financing, providing a scalable blueprint for global industrial clusters to achieve sovereign, sustainable, and transparent value chains.

Global Business Services (GBS) have emerged as a "living laboratory" for the Twin Transition of Green and Digital Transformation, as multinational corporations (MNCs) face increasing pressure to harmonize digital efficiency with environmental stewardship. Aiming to derive a socio-technical framework, this paper synthesizes Technology Roadmapping (TRM) with the International Telecommunication Union (ITU) ICT-centric innovation ecosystem toolkit. A bibliometric analysis of research clusters reveals an evolutionary shift from basic process automation toward "Sustainable Intelligence," identifying the GBS unit as a central "operational airlock" that mediates between landscape pressures -- such as the EU's dual mandate and Carbon Border Adjustment Mechanisms -- and niche innovations in AI-native workflows. The study further maps these clusters onto a stakeholder engagement canvas, highlighting how resilient "Middle Power" hubs in Poland, Portugal, and Malaysia are bypassing the middle-income trap to provide a "third way" for global value chains amidst a bifurcated geopolitical cloud. The results offer a data-driven design approach for leaders and entrepreneurial support networks to orchestrate talent and supply chain flows, thereby enriching the conceptual understanding of Industry 5.0 and the role of GBS as a primary mechanism for navigating a volatile, multipolar digital economy.

This study investigates semiparametric efficient estimation of causal and structural parameters in a semi-supervised setting. In our setting, unlabeled auxiliary regressors are available in addition to labeled observations consisting of outcomes and regressors. Our goal is to construct estimators of causal and structural parameters whose asymptotic variances are smaller than those of estimators constructed using only labeled data. We refer to this framework as prediction-powered causal inference (PPCI). We first derive the efficient influence function and the efficiency bound, which imply that the use of auxiliary regressors can attain a smaller asymptotic variance than the efficiency bound attainable from labeled observations alone. Then, by combining the efficient influence function with the debiased machine learning (DML) framework, we propose methods that we call DML-PPCI. If we construct an estimating-equation estimator, we refer to the method as EE-DML-PPCI; if we construct a targeted-learning estimator, we refer to the method as TMLE-DML-PPCI. The asymptotic variances of both estimators match our derived efficiency bound. In the construction of the estimators, estimation of the efficient influence function plays an important role. In our study, the efficient influence function is also a Neyman orthogonal score, which depends on the Riesz representer and the regression function. For Riesz representer estimation, we develop semi-supervised generalized Riesz regression with convergence rate guarantees.

This paper studies sovereign stress avalanches and network amplification in Latin American credit markets using monthly J.P. Morgan EMBI Global Diversified spreads for eleven sovereigns over 2007-2026. Country stress events are defined as positive log-spread innovations exceeding country-specific volatility thresholds, and regional avalanches count the number of stressed countries in each month. The empirical design combines finite-sample power-law diagnostics, threshold robustness checks, a country-level reshuffling placebo, and rolling correlation, partial-correlation, and minimum-spanning-tree networks. Avalanche sizes are heavy-tailed, with an estimated exponent of 1.77, while spread changes and inter-event times lie in a heavy-tail boundary regime. The placebo shows synchronization far above independent stress timing, with p-values below 0.001. Large avalanches coincide with denser and more spectrally amplifying raw-correlation networks, but not after partial-correlation filtering, indicating common-factor co-movement rather than conditional regional propagation. Network metrics describe contemporaneous stress regimes rather than early-warning signals. The results provide a finite-size criticality framework for monitoring sovereign fragility in emerging markets.

Frontier AI systems are bridging the gap between intelligence and utility by shifting from conversational assistants to autonomous agents that execute tasks end to end. Using production data from Perplexity's Search and Computer products, we study this transition by examining how AI agents accelerate and reshape knowledge work. Three key empirical findings emerge. First, using sessions with near-identical initial query pairs as natural experiments for the same underlying task attempted with both products, Computer performs 26 minutes of autonomous work per user session, versus 33 seconds for Search. Computer automates task decomposition and execution that Search users might otherwise manually orchestrate and implement. As a result, Computer shifts follow-up query distribution toward higher-order work such as verification and extension. Autonomy also increases execution quality, with per-query dissatisfaction rates 55% lower on Computer than on Search. Second, due to its autonomy advantage, Computer reduces completion time from 269 to 36 minutes on matched tasks, lowering estimated time and cost by 87% and 94%, respectively, compared to humans equipped with Search alone. Third, Computer changes the scope of work that users attempt: Computer queries more often cross occupational boundaries, require higher-order cognition, draw on broader expertise, take the form of composite tasks that bundle interdependent subtasks into a single query, and unlock work activities that are essentially absent from Search usage among the same users. Together, the evidence indicates that AI agents accelerate workflows, enhance output quality, reduce costs, and expand the breadth and depth of automated work.