Abstract
Rising trade tensions, a spate of trade-inhibiting policy measures and the weakening of multilateral institutions have sparked growing concerns about the potential implications of global trade fragmentation. This paper estimates a structural gravity model to assess the impact of geopolitical distance — measured using UN General Assembly voting patterns — on manufacturing trade over the period 2012–2023. Our analysis provides robust evidence that geopolitical distance has emerged as a significant trade barrier in the post–COVID-19 period. We find that a 10% rise in geopolitical distance, like the observed increase in the U.S.-China distance since 2017, is associated with a reduction in trade by about 2%. The effect is driven by advanced economy imports from middle-income countries and is concentrated in upstream sectors associated with strategic goods. Finally, we show that trade reconfiguration takes place primarily through friend-shoring rather than near-shoring.

Adversarial Risk Analysis (ARA) offers a Bayesian decision-theoretic framework for modeling contests between strategic opponents, distinct from classical game theory. This review highlights significant developments in the field since 2020.

We contribute to the international trade literature by extending the analysis of geopolitical tensions into an out-of-sample predictive framework. We use the adaptive nature of the Random Forest to verify whether the rising importance of geopolitics provides a forecasting signal without imposing parametric constraints. We find that it does. Geopolitical distance emerges as a dominant predictor of trade flows, consistently outranking traditional policy variables such as tariffs and RTAs—a result we validate against regularized linear benchmarks. For policymakers, this suggests that forecasting models accounting only for physical distance and traditional cost factors may suffer from systematic bias.

We study how robot adoption and investment in information and communication technologies (ICT) jointly shape sectoral employment across 20 EU countries from 1995 to 2020. Using a cross-sectional regression design that interacts changes in robot adoption with ICT investment, we find that increases in robot adoption are associated with higher employment in sectors that either entered the period without robots or invested little in ICT. By contrast, robot adoption is associated with lower employment in sectors that initially had some robots and high ICT investment. These findings highlight the importance of both initial conditions and complementary technology investment in shaping labour-market outcomes, suggesting that the employment effects of technology are highly context-dependent.

Empirical evidence that geopolitical considerations are already materially affecting trade flows is scant. In this study, we quantify the impact of geopolitical tensions on trade of manufacturing goods over the period 2012-2022 in a structural gravity framework. To capture the influence of geopolitical tensions, we use a measure of geopolitical distance based on the UN General Assembly voting. The econometric analysis offers robust evidence that geopolitical distance has become a trade friction and its impact has steadily increased over time. Our results suggest that a 10% increase in geopolitical distance, like the observed increase in the US-China distance since 2018, is associated with a reduction in trade by about 2%. Our findings also highlight a differential and stronger impact on advanced economies and the emergence of friend-shoring.

Greenwashing refers to the deceptive practice where a company exaggerates or misrepresents the sustainability of its actions or projects. Given the ambiguity surrounding the methodologies behind conventional ratings, we enquire their robustness through the implementation of an alternative comprehensive measure entailing both internally disclosed and externally generated data. We address a notorious critique in greenwashing research—that the entire voluntary CSR approach inadvertently facilitates the diffusion of greenwashing. We contribute to the statistical methodology by breaking down the difference between internal and external perception of sustainability through regression analysis. We claim that only when the presence of CSR Committees is coupled with tangible initiatives boosting sustainability, both external and internal stakeholders are found to positively evaluate the sustainable commitment of a company. 

We investigate the impact of an enhancement in female presence, meant as women in decision-making positions, on a firm’s performance both in financial and sustainability terms. Most research on sustainable finance and its impact on corporate governance rely only on aggregate ESG ratings for their results. Such scores are typically a black-box, with financial providers supplying little information about their methodology. Our analysis not only develops disaggregate scores for each dimension, but also provides motivation for the measurement of gender equality by means of specific indicators, such as the number of female directors, going beyond the bare (S) or (G) rating. ESG ratings and specific indicators of gender equality were retrieved from the well-known Bloomberg provider. Relying on a dataset concerning European companies, we empirically show that an increase in gender equality has a positive effect on a firm’s financial performance and on its share of sustainable investments.

This paper introduces a Boltzmann Generator framework for modeling and forecasting the global trade network. Departing from traditional econometric gravity models, the proposed method employs a conditional deep generative architecture that maps macroeconomic and geographic covariates into full probabilistic distributions of trade matrices. The model’s energy-based formulation connects statistical physics and international economics, enabling realistic trade simulations that capture both bilateral dependencies and higher-order network effects. Using trade data for 206 countries between 2001 and 2020, we demonstrate that the Boltzmann model achieves superior in-sample and out-of-sample predictive accuracy compared with established econometric and machine learning benchmarks. Statistical tests confirm the model’s robust predictive dominance across time and country pairs. Beyond forecasting, the framework supports counterfactual policy analysis: simulations of scenarios reveal distinct propagation mechanisms through the global trade network, highlighting China’s central mediating role in transmitting shocks. The Boltzmann approach thus provides an interpretable, and generative alternative for analyzing trade dynamics, quantifying uncertainty, and evaluating policy in complex economic systems.

Understanding the dynamics of structural transformation is a central problem in economics, yet existing approaches, often based on static or discrete-time representations, capture path dependence without formally modeling the continuous-time evolution of national productive structures. We address this limitation by introducing a continuous-time stochastic process framework for the evolution of countries’ export portfolios. Specifically, we model each country’s productive structure as a high-dimensional vector evolving according to a stochastic differential equation (SDE) with a time-dependent conditional drift and a diffusion term, representing systematic structural forces and idiosyncratic shocks, respectively. The drift is estimated using a combination of flow-matching and Schrödinger Bridge techniques, trained on a panel dataset of 95 countries over 2012–2023. We show that the estimated drift is statistically aligned with realized year-over-year structural changes and respects the path-dependent principle of proximity from economic complexity theory. Using regression analyses, we further demonstrate that the direction of the learned drift, measured by its alignment with a complexity-enhancing frontier, significantly predicts subsequent growth in economic complexity, whereas its magnitude does not. The model also enables probabilistic multi-year forecasts, capturing heterogeneous trajectories across countries.

I introduce a "Trade at Risk" framework -along the lines of Adrian et al. (2019)- to quantify how geopolitical tensions shape the probability of trade collapses. By combining structural gravity models with quantile regression on bilateral data (2012–2023), I show that geopolitics is a non-linear friction. I find that geopolitical misalignment acts as a significant tail risk factor, while leaving median trade growth largely unaffected. 

 In this SUERF policy brief we quantify the impact of geopolitical tensions on trade of manufacturing goods over the period 2012-2022. To capture the influence of geopolitical tensions, we augment a state-of-the-artgravity model with a measure of geopolitical distance based on the UN General Assembly voting. The econometric analysis offers robust evidence that geopolitical distance has become a trade friction and its impact has steadily increased over time. Our results suggest that a 10% increase in geopolitical distance, like the observed increase in the US-China distance since 2018, is associated with a reduction in trade by about 2%. Our findings also highlight a differential and stronger impact on advanced economies and the emergence of friend-shoring.

AI is changing the world in ways that are difficult to forecast, but the impact will surely be enormous. Large language models (LLMs) are the most recent AI system that has captured the public eye. The rise of AI and LLMs offers efficiency and assistance, but raises questions of job loss, fairness, and societal norms. Bias is a significant challenge. Educational reforms are needed, and legal frameworks must adapt to address liability and privacy issues. Ultimately, human choices will shape AI influence, highlighting the need for responsible development and regulation to ensure benefits outweigh risks.  

Large language models have burst onto the scene, and may do much to change the way the world operates. They can write, illustrate, and are rapidly adding new capabilities. This panel discussion brings together a group of experts to discuss ways in which these tools might evolve, particularly in the context of how trustworthy they are and the issues surrounding their regulation. 

There has been a rise in trade restrictions since the US-China tariff war and Russia’s invasion of Ukraine. This column explores the impact of geopolitical tensions on trade flows over the last decade. Geopolitical factors have affected global trade only after 2018, mostly driven by deteriorating geopolitical relations between the US and China. Trade between geopolitically aligned countries, or friend-shoring, has increased since 2018, while trade between rivals has decreased. There is little evidence of near-shoring. Global trade is no longer guided by profit-oriented strategies alone–geopolitical alignment is now a force. 

Rising trade tensions and a spate of policies aiming to bring national security concerns to bear in trade relations have sparked growing concern about the potential implications of geo-economic fragmentation. With so far only limited available empirical evidence that geopolitical concerns are already materially affecting trade patterns, this box investigates the role played by geopolitical tensions in shaping international trade over the last decade.

Identifying the medium-term inflation path in the current environment of high inflation, ongoing energy and pandemic-related shocks and the Russian invasion of Ukraine is a diagnostic challenge. In his ECB Blog post Philip R. Lane, Member of the ECB’s Executive Board, describes some of the key analytical issues involved.