Job training is widely regarded as crucial for protecting workers from automation, yet there is a lack of empirical evidence to support this belief. Using internationally harmonized data from over 90,000 workers across 37 industrialized countries, we construct an individual-level measure of automation risk based on tasks performed at work. Our analysis reveals substantial within-occupation variation in automation risk, overlooked by existing occupation-level measures. To assess whether job training mitigates automation risk, we exploit within-occupation and within-industry variation. Additionally, we employ entropy balancing to re-weight workers without job training based on a rich set of background characteristics, including tested numeracy skills as a proxy for unobserved ability. We find that job training reduces workers’ automation risk by 4.7 percentage points, equivalent to 10 percent of the average automation risk. The training-induced reduction in automation risk accounts for one-fifth of the wage returns to job training. Job training is effective in reducing automation risk and increasing wages across nearly all countries, underscoring the external validity of our findings. Women tend to benefit more from training than men, with the advantage becoming particularly pronounced at older ages.

Our results reveal that higher lending spreads between foreign and home markets redirect real estate backed lending towards foreign markets offering a higher interest rate, which provides evidence for "search for yield" behavior. This re-allocation is found especially for banks with more expertise on the foreign market due to a higher local activity and holds for commercial and residential real estate backed loans. Furthermore, "search for yield" behavior and a resulting increase in foreign real estate backed lending is found when macroprudential regulation is missing or misaligned between a bank’s country of residence and the destination country. When turning to the question of whether the detected search for yield behavior results in more risk, we find that especially better capitalized banks report higher forbearance ratios as they might face less stigma effects compared to low capitalized banks.

We examine effects of government-imposed employment targets on firm behavior. Theoretically, such policies create “polarization," causing low-productivity firms to exit the market while others temporarily distort their employment upward. Dynamically, firms are incentivized to improve productivity to meet targets. Using novel data from East German firms post-privatization, we find that firms with binding employment targets experienced 25% points higher annual employment growth, a 1.1% points higher annual exit probability, and 10% points higher annual productivity growth over the target period. Structural estimates reveal substantial misallocation of labor across firms and that subsidizing productivity growth would yield twice the long term increases in employment.

The Transition Report 2024-25 focuses on industrial policies in the EBRD regions and beyond. Such policies have seen a resurgence, seeking to address market failures such as environmental degradation. However, their track record is mixed. Their growing popularity is shaped primarily by domestic political economy considerations and rising geopolitical tensions. While industrial policies are typically employed by higher-income economies, they are also now used more frequently in economies with less administrative and fiscal capacity to implement them.

Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The ex-post threshold optimization is based upon a loss function accounting for preferences between forecast errors, but comes with two crucial drawbacks: unstable thresholds in recursive estimations and an in-sample overfit at the expense of out-of-sample performance. We propose two alternatives for threshold setting: (i) including preferences in the estimation itself and (ii) setting thresholds ex-ante according to preferences only. We provide simulated and real-world evidence that this simplification results in stable thresholds and improves out-of-sample performance. Our solution is not restricted to binary-choice models, but directly transferable to the signaling approach and all probabilistic early-warning models.

Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The expost threshold optimization is based upon a loss function accounting for preferences between forecast errors, but comes with two crucial drawbacks: unstable thresholds in recursive estimations and an in-sample overfit at the expense of out-of-sample performance. We propose two alternatives for threshold setting: (i) including preferences in the estimation itself and (ii) setting thresholds ex-ante according to preferences only. Given probabilistic model output, it is intuitive that a decision rule is independent of the data or model specification, as thresholds on probabilities represent a willingness to issue a false alarm vis-à-vis missing a crisis. We provide simulated and real-world evidence that this simplification results in stable thresholds and improves out-of-sample performance. Our solution is not restricted to binary-choice models, but directly transferable to the signaling approach and all probabilistic early-warning models.

Can a negative shock to sovereign ratings invoke a vicious cycle of increasing government bond yields and further downgrades, ultimately pushing a country toward default? The narratives of public and political discussions, as well as of some widely cited papers, suggest this possibility. In this paper, we will investigate the possible existence of such a vicious cycle. We find no evidence of a bad long-run equilibrium and cannot confirm a negative feedback loop leading into default as a transitory state for all but the very worst ratings.