The Federal Planning Bureau (FPB) is an independent public agency. It draws up studies and projections on economic, social and environmental policy issues and on the integration of these policies within a context of sustainable development.
Users of Supply and Use Tables (SUT) and Input-Output Tables (IOT) compiled in different national accounts (NA) vintages face a problem of consistency of their data due to revisions in the NA. This paper describes the methodology that has been followed to compile a consistent time series of Belgian SUT and IOT for the period 1995-2007, in line with the NA published in November 2010.
This Working Paper presents the different methodologies currently used to construct a volume index of capital services and analyzes the effects of methodological changes on capital services and total factor productivity estimates for Belgium over the period 1970-2004. The measurement of capital services is realized in two steps. First, productive capital stocks have to be estimated for each type of asset. Two methodologies are generally used: the geometric and the hyperbolic profile. Secondly, these stocks are aggregated, using the user costs of capital (exante or ex-post approach) as weights to derive an overall index. For the economy as a whole and the entire period, under an ex-post approach, the volume indices of capital services estimated with a hyperbolic age-efficiency profile grow at a higher rate than the indices estimated with a geometric profile. This general conclusion is, however, not observed in every sector. Under an ex-ante approach, the different volume indices are quite similar for the whole economy, even if the indices grow generally at a slightly higher rate in the case of a geometric pattern. A higher growth rate of the volume indices generates a higher capital contribution and, consequently, a lower TFP contribution. Over long periods of time, the different TFP estimates are relatively similar. Over shorter periods, the different methodologies generate more significant variations in the TFP contribution.