Should multivariate early warning systems for banking crises pool across regions?

Most extant work on prediction of banking crises has utilised global samples, which are in turn dominated by observations from middle-income countries, and rely on a single estimator. However, economic and financial structures as well as the pattern of shocks may differ substantially across regions, while a range of specifications is desirable to check robustness. Accordingly, in this paper we test the implicit pooling assumption in earlier multivariate work on Early Warning Systems using both logit and binary recursive tree specifications separately for crises in Asia and Latin America, as well as the pooled sample. Results suggest markedly different crisis determinants across regions, implying global samples are inappropriate.     

Foreign Investors in The Netherlands: Heterogeneous Employment and Productivity Effects

Recent academic research suggests that the effects of inward foreign direct investment for productivity and employment may vary across foreign investors with different characteristics. Using a newly created dataset that integrates micro-level information from The Netherlands Foreign Investment Agency and Statistics Netherlands, this paper explores the role of several of these characteristics, including e.g. the country of origin, activities and tasks, age and strategic role of foreign affiliates in The Netherlands in the 2000–2007 period. Our analysis shows that important employment and productivity differences exist for nearly all these dimensions.     

UK World War I and interwar data for business cycle and growth analysis

This article contributes new time series for studying the UK economy during World War I and the interwar period. The time series are per capita hours worked and average capital income, labor income, and consumption tax rates. Uninterrupted time series of these variables are provided for an annual sample that runs from 1913 to 1938. We highlight the usefulness of these time series with several empirical applications. The per capita hours worked data are used in a growth accounting exercise to measure the contributions of capital, labor, and productivity to output growth. The average tax rates are employed in a Bayesian model averaging experiment to reevaluate the Benjamin and Kochin (J Political Econ 87:441–478, 1979) regression.