Measurement of Economic Variables

Abstract

The basic concern of any empirical work is to employ statistical data that correspond to the notion of the theoretical variables in the model. The problems and economic consequences connected with the measurement of selected economic variables are the focus of this thesis. It consists of three chapters that in succession analyze the issues associated with the measurement of economic growth, multi-factor productivity and capital input into production. The first chapter looks into the differences among the growth rates of GDP per capita based on data from the three most commonly used databases, namely International Financial Statistics, World Development Indicators and Penn World Table. Using a wide international dataset, we find significant differences in the growth rates that are mainly due to the adjustment for cross-country comparability of GDP per capita levels. Importantly, these differences are correlated with the level of development. We replicate six recent studies of growth determinants and find their results sensitive to the choice of data. The second chapter analyses the sensitivity of calculated multi-factor productivity (MFP) growth to assumptions of growth accounting, concentrating on the measurement of quantity, composition and the respective shares of labor and capital inputs, and...