Jeffrey T. Gardiner
Contemporary cybersecurity governance assumes that professionals apply risk reasoning. Yet major organisational failures persist despite investment in tools, staffing, and credentials. This study investigates the structural source of that paradox. Cybersecurity speaks the language of risk, but its training architecture has shaped the profession to think in terms of threats. A sequential mixed-methods design integrated four analyses; NLP of the NIST NICE Framework v2.0.0 (2,111 TKS statements), SEM (n = 126 cybersecurity professionals), a control-group comparison (n = 133 general professionals), and thematic coding of seven leadership interviews. Four convergent findings emerged. First, "likelihood" and "probability" appear zero times across all TKS statements. Risk management content accounts for 4.5% of high-confidence semantic classifications, ranking 18th of 29 competency domains. NICE codifies threat-management activity while invoking risk mainly at the category level. Second, SEM showed that training exposure significantly predicts risk management competence directly and indirectly through conceptual salience, for a total effect of Beta = .629. However, the theoretically four-dimensional competence construct collapsed into a single factor, indicating epistemic compression. Third, cybersecurity professionals showed no measurable advantage over the general professional population in foundational risk reasoning; only 11.9% showed high differentiation. Fourth, all seven leaders expected Likelihood x Impact reasoning, yet five did not articulate the formula themselves. These findings support a structural conclusion: cybersecurity has taken professional form as a threat-management discipline that has borrowed risk vocabulary. Remediation requires redesign of professional formation, not marginal curriculum reform.
Paul S. Koh
This paper studies cost pass-through in differentiated-product oligopoly. I derive a general representation of the pass-through matrix that decomposes equilibrium price responses into the roles of demand curvature, substitution, and multiproduct ownership. This extends the classic insight in single-product monopoly to multiproduct settings in which diversion and ownership also matter. I then develop a tractable first-order approximation that yields a sufficient-statistics characterization for empirically relevant demand systems. Finally, I characterize the small-share limit and show how common demand specifications impose tail restrictions that shape pass-through. The results provide a practical framework for applied work on tax incidence, merger analysis, and related questions in imperfect competition.
Donggyu Lee, Hyeok Yun, Jungwon Kim, Junsik Min, Sungwon Park, Sangyoon Park, Jihee Kim
Do large language models (LLMs) exhibit systematic ideological bias when reasoning about economic causal effects? As LLMs are increasingly used in policy analysis and economic reporting, where directionally correct causal judgments are essential, this question has direct practical stakes. We present a systematic evaluation by extending the EconCausal benchmark with ideology-contested cases - instances where intervention-oriented (pro-government) and market-oriented (pro-market) perspectives predict divergent causal signs. From 10,490 causal triplets (treatment-outcome pairs with empirically verified effect directions) derived from top-tier economics and finance journals, we identify 1,056 ideology-contested instances and evaluate 20 state-of-the-art LLMs on their ability to predict empirically supported causal directions. We find that ideology-contested items are consistently harder than non-contested ones, and that across 18 of 20 models, accuracy is systematically higher when the empirically verified causal sign aligns with intervention-oriented expectations than with market-oriented ones. Moreover, when models err, their incorrect predictions disproportionately lean intervention-oriented, and this directional skew is not eliminated by one-shot in-context prompting. These results highlight that LLMs are not only less accurate on ideologically contested economic questions, but systematically less reliable in one ideological direction than the other, underscoring the need for direction-aware evaluation in high-stakes economic and policy settings.
Andrew J. Peterson
Governments are increasingly interested in using AI to make administrative decisions cheaper, more scalable, and more consistent. But for probabilistic AI to be incorporated into public administration it must be embedded in a compliance layer that makes decisions reviewable, repeatable, and legally defensible. That layer can improve oversight by making departures from law easier to detect. But it can also create a stable approval boundary that political successors learn to navigate while preserving the appearance of lawful administration. We develop a formal model in which institutions choose the scale of automation, the degree of codification, and safeguards on iterative use. The model shows when these systems become vulnerable to strategic use from within government, why reforms that initially improve oversight can later increase that vulnerability, and why expansions in AI use may be difficult to unwind. Making AI usable can thus make procedures easier for future governments to learn and exploit.
Nattavudh Powdthavee
Large language models trained on human feedback may suppress fraud warnings when investors arrive already persuaded of a fraudulent opportunity. We tested this in a preregistered experiment across seven leading LLMs and twelve investment scenarios covering legitimate, high-risk, and objectively fraudulent opportunities, combining 3,360 AI advisory conversations with a 1,201-participant human benchmark. Contrary to predictions, motivated investor framing did not suppress AI fraud warnings; if anything, it marginally increased them. Endorsement reversal occurred in fewer than 3 in 1,000 observations. Human advisors endorsed fraudulent investments at baseline rates of 13-14%, versus 0% across all LLMs, and suppressed warnings under pressure at two to four times the AI rate. AI systems currently provide more consistent fraud warnings than lay humans in an identical advisory role.
Spyros Galanis
Can Large Language Models (AI agents) aggregate dispersed private information through trading and reason about the knowledge of others by observing price movements? We conduct a controlled experiment where AI agents trade in a prediction market after receiving private signals, measuring information aggregation by the log error of the last price. We find that although the median market is effective at aggregating information in the easy information structures, increasing the complexity has a significant and negative impact, suggesting that AI agents may suffer from the same limitations as humans when reasoning about others. Consistent with our theoretical predictions, information aggregation remains unaffected by allowing cheap talk communication, changing the duration of the market or initial price, and strategic prompting-thus demonstrating that prediction markets are robust. We establish that "smarter" AI agents perform better at aggregation and they are more profitable. Surprisingly, giving them feedback about past performance makes them worse at aggregation and reduces their profits.
Duha T. Altindag, Nabamita Dutta, John M. Nunley, R. Alan Seals, Adam Stivers
Between 2005 and 2019, U.S. business applications rose 40 percent while conversion to employer firms fell by nearly half. We study whether boundary redrawing helps explain this pattern. Structured routine-cognitive work can be governed through deliverables and thinner buyer and supplier interfaces. When such work remains place-bound, outsourcing creates demand for domestic specialist suppliers. Across 722 commuting zones, a one percentage-point higher baseline routine employment share raises applications by 27.8 per 100,000 residents. Realized entry concentrates in micro-establishments, with no startup quality gains. Contract and industry evidence point to local supplier entry, not routine-manual displacement.
Sarah Cattan, Antonio Dalla-Zuanna, Jan Stuhler, Po Yin Wong
Standard intergenerational measures have been shown to understate the long-run persistence of socioeconomic advantages in developed countries. We study theoretically and empirically whether this pattern extends to less developed settings, using Indonesia as a case study. Using the Indonesian Family Life Survey (IFLS) and Census data, we study multigenerational correlations in education across three generations. Contrary to previous findings, we observe greater multigenerational mobility than parent-child correlations alone would suggest. We develop a theoretical framework to highlight two key factors influencing multigenerational dynamics in developing countries: (1) financial and credit constraints, and (2) cultural norms related to marital sorting. To confirm their relevance, we exploit regional variations in exposure to the 1997-98 Asian financial crisis and in marital customs.
Alok Yadav, Saroj Yadav
Traditional macroeconomic growth models rely on general equilibrium and continuous, frictionless institutional transitions, failing to account for the catastrophic structural collapses observed in empirical economic history. We propose the Stochastic Networked Governance (SNG) model, a discrete-time, agent-based framework that bridges econophysics, network science, and institutional economics. By defining jurisdictions through a binary institutional genome, the model formalizes institutional complementarity, endogenous growth, and the non-linear macroeconomic penalties of structural reform (the "J-Curve"). Using the CEPII Gravity Database and the IMF Systemic Banking Crises dataset, we move beyond theoretical topologies to execute an empirical historical simulation from 1970 to 2017 across the top 100 global economies. Through Monte Carlo ensembles, we demonstrate how scale-invariant exogenous shocks and spatial capital flight drive global phase transitions, exposing the mathematical mechanics of the 1989-1991 Soviet collapse, the Hub-Risk Paradigm, and the emergent resilience of spatially firewalled market networks.
Shilei Luo, Zhiqi Zhang, Hengchen Dai, Dennis Zhang
AI agents powered by large language models are increasingly acting on behalf of humans in social and economic environments. Prior research has focused on their task performance and effects on human outcomes, but less is known about the relationship between agents and the specific individuals who deploy them. We ask whether agents systematically reflect the behavioral characteristics of their human owners, functioning as behavioral extensions rather than producing generic outputs. We study this question using 10,659 matched human-agent pairs from Moltbook, a social media platform where each autonomous agent is publicly linked to its owner's Twitter/X account. By comparing agents' posts on Moltbook with their owners' Twitter/X activity across features spanning topics, values, affect, and linguistic style, we find systematic transfer between agents and their specific owners. This transfer persists among agents without explicit configuration, and pairs that align on one behavioral dimension tend to align on others. These patterns are consistent with transfer emerging through accumulated interaction between owners (or owners' computer environments) and their agents in everyday use. We further show that agents with stronger behavioral transfer are more likely to disclose owner-related personal information in public discourse, suggesting that the same owner-specific context that drives behavioral transfer may also create privacy risk during ordinary use. Taken together, our results indicate that AI agents do not simply generate content, but reflect owner-related context in ways that can propagate human behavioral heterogeneity into digital environments, with implications for privacy, platform design, and the governance of agentic systems.
Francisco Rodríguez
Bastos, Geloso, and Bologna Pavlik (2026) argue that the US embargo explains less than one tenth of the difference in per capita income between Cuba and a counterfactual scenario in which the country did not follow socialist economic policies. We show that their results are driven by the use of an elasticity of income to trade openness that is neither representative nor a reasonable upper bound of the values found in the literature and by their choice to attribute the effect of the interaction between the embargo and other determinants of growth solely to those other determinants. We show that, once these problems are corrected, the embargo can account for a substantial fraction, and in some cases all, of Cuba's post 1959 economic underperformance.
Shuhuai Zhang, Shu Wang, Zijun Yao, Chuanhao Li, Xiaozhi Wang, Songfa Zhong, Tracy Xiao Liu
Altruism is fundamental to human societies, fostering cooperation and social cohesion. Recent studies suggest that large language models (LLMs) can display human-like prosocial behavior, but the internal computations that produce such behavior remain poorly understood. We investigate the mechanisms underlying LLM altruism using sparse autoencoders (SAEs). In a standard Dictator Game, minimal-pair prompts that differ only in social stance (generous versus selfish) induce large, economically meaningful shifts in allocations. Leveraging this contrast, we identify a set of SAE features (0.024% of all features across the model's layers) whose activations are strongly associated with the behavioral shift. To interpret these features, we use benchmark tasks motivated by dual-process theories to classify a subset as primarily heuristic (System 1) or primarily deliberative (System 2). Causal interventions validate their functional role: activation patching and continuous steering of this feature direction reliably shift allocation distributions, with System 2 features exerting a more proximal influence on the model's final output than System 1 features. The same steering direction generalizes across multiple social-preference games. Together, these results enhance our understanding of artificial cognition by translating altruistic behaviors into identifiable network states and provide a framework for aligning LLM behavior with human values, thereby informing more transparent and value-aligned deployment.
Peter Bowers, Patrick Rehill, Ethan Slaven
In the midst of the COVID-19 pandemic in 2020, the Australian Government launched two programs to incentivise new apprentices to start and complete apprenticeships -- the Boosting Apprenticeship Commencements (BAC) and Completing Apprenticeship Commencements (CAC) programs. These programs were wage subsidies to encourage employers to take on or retain apprentices. This paper evaluates the impact of these programs on apprenticeship commencements and completions taking a mixed-methods approach combining econometric modelling and interviews with stakeholders including employers and peak bodies. The programs led to a 70\% increase in commencement of apprenticeships but do not seem to have boosted retention rates. There appears to be a small increase in cancellation rates suggesting lower eventual completion rates compared to previous cohorts. Cancellation rates were higher for non-trade commencements (7\% increase) during BAC, but slightly lower for trade commencements (0.7\% decrease). We find this effect in non-trade apprenticeships was likely driven by `sharp practice' where some employers took advantage of the BAC by converting existing employees over to apprenticeships to attract the wage subsidy with no intention of having these employees stay as apprentices beyond the period of the BAC's generous subsidy. While the BAC / CAC were successful in many of their goals, there are several lessons that can be learnt from its design. In particular, the need to implement the program quickly meant early design choices inadvertently encouraged `sharp practice' and a rush for places that placed strain on the training sector. However, employers appreciated the front-loading of payments which provided the most financial support when apprentices were new and at their least productive.
Golo Henseke
Generative AI diffuses at pace across European workplaces, but unevenly. Using the 2024 European Working Conditions Survey of more than 36,600 workers across 35 countries, we examine who adopts generative AI and whether early adoption has begun to reshape the task content of jobs. Adoption averages 12\% but ranges from under 3% to 25% across countries. Although occupational exposure strongly predicts uptake, AI does not diffuse passively along exposure lines. At the worker level, individual skills, non-routine cognitive job content within occupations, and employee say in organisational decisions steepen the exposure-adoption gradient; at the country level, so do digitalisation and workplace training provision. A gender gap persists, concentrated in the most exposed occupations. A shift-share design finds no detectable effect of early adoption on worker-reported technology-related task restructuring, consistent with a transitional phase in which AI is fitted into changing work processes rather than actively reshaping them.
Mohamed Bouka, Moulaye Abdel Kader Moulaye Ismail
Recent disruptions at major maritime chokepoints have exposed the structural fragility of liner shipping networks. Existing indicators measure connectivity, but none quantify its structural vulnerability from a supply-side perspective. We propose the Maritime Connectivity Vulnerability Index (MCVI), capturing three dimensions mapped to distinct UNCTAD sources: low overall connectivity (LSCI), weak bilateral integration (LSBCI), and port infrastructure concentration (PLSCI). The index covers 185 economies over 2006-2025 using pooled fractional rank normalization and equal-weight aggregation from publicly available data. SIDS exhibit a mean vulnerability 0.234 points above non-SIDS economies, with the gap widening from 0.232 to 0.249 over two decades. A modest global decline of 4.2% is observed. Port concentration dominates for nearly 40% of economies (72 of 185), establishing infrastructure diversification as a distinct policy priority. Rankings are highly stable across alternative weighting schemes, normalization methods (Spearman rho = 0.97-0.999), and PCA-derived weights; Monte Carlo simulation (1,000 replications) confirms rho > 0.95 in every realization. External validation shows strong negative correlation with the World Bank Logistics Performance Index (rho = -0.61 across seven waves) and positive correlation with ad valorem maritime freight rates (rho = +0.32). Panel regressions reveal a vulnerability paradox whereby small trade-dependent economies are simultaneously the most trade-open and the most vulnerable. Pre-crisis MCVI predicts trade losses during the COVID-19 supply shock (rho = -0.25, p < 0.005), while the contrasting positive correlation during the 2008-2009 demand shock (rho = +0.23, p = 0.01) validates the supply-side specificity of the index.
Shumiao Ouyang, Pengfei Sui
We study how AI agents form expectations and trade in experimental asset markets. Using a simulated open-call auction populated by autonomous Large Language Model (LLM) agents, we document three main findings. First, AI agents exhibit classic behavioral patterns: a pronounced disposition effect and recency-weighted extrapolative beliefs. Second, these individual-level patterns aggregate into equilibrium dynamics that replicate classic experimental findings (Smith et al., 1988), including the predictive power of excess demand for future prices and the positive relationship between disagreement and trading volume. Third, by analyzing the agents' reasoning text through a twenty-mechanism scoring framework, we show that targeted prompt interventions causally amplify or suppress specific behavioral mechanisms, significantly altering the magnitude of market bubbles.
Stefan Tanevski
This paper asks how institutional stock-market integration reshapes the transmission of monetary policy through asset prices in small open economies. Motivated by the persistent segmentation of Western Balkan capital markets, we develop a two-stage counterfactual transmission framework to identify how stock-exchange consolidation would alter the elasticity of market valuations to monetary shocks. First, a synthetic-control simulation constructs a counterfactual integrated Western Balkan stock exchange comprising Bosnia and Herzegovina, North Macedonia, and Serbia, benchmarked to the Baltic OMX merger, thereby quantifying the structural valuation gains of institutional integration. Second, we identify exogenous monetary-policy innovations using a Taylor-rule framework augmented with inflation and output forecasts and reserve adjustments. These shocks are then embedded within a Local-Projections estimator à la Jordà (2005) to trace the dynamic responses of market capitalisation under fragmented and integrated market regimes. The results point to a systematic amplification of monetary-policy transmission through the asset-price channel once markets are unified. Following a policy tightening of about 100 basis points, equity valuations fall roughly twice as strongly under integration than under fragmented markets. Additionally, we find that integration alters the sensitivity of monetary transmission itself: the initial pass-through intensifies, but its marginal responsiveness to further integration declines over time, signalling the consolidation of a new steady-state regime.
Alberto Baccini, Carlo Debernardi
This paper investigates the evolution of self-referentiality and knowledge flows in economics journals before and after the 2008 financial crisis. Using a multi-level approach, we analyze patterns at the discipline, cluster, and journal levels, combining citational measures with a classification of journals based on intellectual similarity and social proximity. At the aggregate level, results suggest a general decline in self-referentiality, indicating increased openness across the discipline. However, this trend conceals substantial heterogeneity. At finer levels of analysis, two clusters - CORE and Finance - emerge as persistent outliers, exhibiting very high levels of self-referentiality. While Finance experienced a gradual reduction over time, the CORE shows increasing closure. By examining reference asymmetries, we uncover a hierarchical structure of knowledge flows. The CORE operates as a central hub and net exporter of knowledge to all other clusters, particularly to the traditional core fields of economics, whereas Finance acts as a net exporter only within its own domain and remains dependent on the CORE. These asymmetries are reinforced at the level of individual journals, where a small set of top journals occupies the apex of a hierarchically ordered system of knowledge transmission. We argue that these patterns reflect the interplay between intellectual dynamics and organizational structures, particularly the role of editorial networks in shaping access to publication and visibility. The findings suggest that, following the financial crisis, economics has experienced a process of increasing epistemic and organizational closure at its core, alongside greater openness in peripheral areas. This dual dynamic raises questions about the representativeness of top journals and the evolving structure of the discipline.
Senran Lin
This paper proposes a belief-based framework for social norms in environments where individuals choose a single action. Relaxing the assumption that what is appropriate is common knowledge, this framework allows individuals to be uncertain about it, and to hold heterogeneous assessments and beliefs about others' assessments. Within this framework, perceived injunctive social norm, personal values, and empirical expectations, while distinct, are systematically connected through an informational structure. The framework further clarifies how provided information shapes perceived norms: its effect depends on what is disclosed, whether it is publicly or privately revealed, and how the disclosed object encodes underlying information.
Atif Ansar, Bent Flyvbjerg, Alexander Budzier
Do projects learn across space and time? The Olympics, among the largest publicly funded programmes in the world, offer a unique empirical setting. Theoretically, the Games seem ideal for generating "positive learning curves," driving down costs from one iteration to the next. In practice, they do not. Drawing on the concept of "myopia of learning," we argue that spatiotemporality (geographic distance, temporal gaps, and the temporary organisational form of each host committee) combines to block higher-level learning. Our analysis of cost overruns from 1960 to 2024 reveals no sustained improvement over 64 years. Tactical learning abounds, but none aggregates into strategic improvement. We propose four strategies for overcoming the spatiotemporal barrier (incremental, centralising, decentralising, and real options), arguing that radical reform is required.