Robust Inference on Average Economic Growth*

We discuss a method to estimate the confidence bounds for average economic growth, which is robust to misspecification of the unit root property of a given time series. We derive asymptotic theory for the consequences of such misspecification. Our empirical method amounts to an implementation of the subsampling procedure advocated in Romano and Wolf (Econometrica, 2001, Vol. 69, p. 1283). Simulation evidence supports the theory and it also indicates the practical relevance of the subsampling method. We use quarterly postwar US industrial production for illustration and we show that non‐robust approaches rather lead to different conclusions on average economic growth than our robust approach.     

On the Functional Form of an Efficiency Index

An input efficiency index gives a numeric assessment of the degree to which a given input combination falls short of being efficient in producing a fixed amount of output. This paper presents a system of axioms which characterise a certain family of efficiency indices containing the well-known Farrell and Färe-Lovell indices. The family of indices satisfying the axioms can be obtained by minimizing a suitable function, called a performance evaluation, and different choices of performance evaluation will result in different indices.     

International Leadership in Productivity at the Aggregate and Industry Level

Since many policies affect specific parts of economies differently, it is useful to decompose GDP per capita differences across countries into differences across smaller and smaller parts of economies. In this paper, we summarize recent contributions in this area and fit them together into a decomposition procedure for GDP per capita differences. The overall finding is that the U.S. is the productivity leader for the most of the economy. Moreover, international productivity differences at the aggregate level of the economies are in most cases translated into differences in the productivity of industries, at least compared to the productivity leader U.S. The variability of productivity differences at the industry level is, however, substantially higher than any differences at the aggregate or sector level. For the manufacturing sector alone the U.S. and Japan share the leadership on the industry level. In contrast, France, U.K., and Germany exhibit almost no leadership in productivity at the industry level. Hence, nation-specific factors appear to be dominant in the comparison of European countries with the U.S. Finally, mix differences do not play a very large role for big countries. For Germany, however, the mix effect can help to reconcile relative high productivity for the market economy and lower productivity at disaggregated levels.     

Wie lässt sich Wohlstand messen?

Die Stiglitz-Kommission hat Empfehlungen zur Weiterentwicklung der statistischen Berichterstattung vorgelegt und das Bruttoinlandsprodukt als nicht ausreichend für die Messung der Wirtschaftsleistung, der Lebensqualit?t und der Nachhaltigkeit bewertet. Tats?chlich gibt es auch in der Bev?lkerung ein Unbehagen in Hinblick auf die amtliche Statistik. Allerdings sind Zufriedenheitsindikatoren als alternative Zielwerte für die Politik kritisch zu beurteilen.     

The structure of German trade with Africa

The structure of Germany’s trade with Africa differs substantially from the general pattern of its foreign trade relations. But there is a further pecullarity in this trade. Each of the African countries concerned has a very specific importance as a trading partner. The following article reviews the characteristic features of African-German trade relations.     

Der Ereignisbegriff in der Publizistikwissenschaft

This article wants to clarify the notion of »event«. It is based on the differentiation of two levels of action–the level of what is covered in the media, and the level of coverage. Action on one level affects the level it happens on, but it also affects the other level. Neither is coverage independent of the occurrence that is covered, nor is the occurrence that is being covered independent of coverage. The notions that are proposed are meant to demonstrate important interdependencies between the two levels of action and to make possible empirical analyses of the relation between coverage and the events covered. This relates to the comparison of the coverage of different media and to the change in the depiction of reality in one medium. The latter is shown for the coverage of German issues in three quality newspapers for the period from 1951 to 1995.     

Why is GDP typically revised upwards?

Quarterly and annual gross domestic output (GDP) data are usually revised, and this revision process may cover a few years. It is also usually found that GDP figures are revised upwards. In this note, I provide a simple explanation for why this happens. Implications for interpreting and forecasting GDP data are less simple.