11 February 2004 • The Working-Day Effect in the Austrian Economy • Marcus Scheiblecker

In 2004, there will be three working days more in Austria than in 2003, due to the leap year and the particular situation of holidays in the calendar. An analysis of quarterly GDP data using a seasonal time series model suggests, that three additional working days will boost annual output by 0.1 percent.

The assessment of the current business situation requires an adjustment of economic data for seasonal factors and the changing number of working days. The simplest method of calculating the working-days effect, namely dividing output by the number of working days, is not the adequate approach, since it attributes the same weight to each working day and, moreover, does not examine whether the varying number of working days is of relevance at all for the level of output.

Using a seasonal time series model, WIFO has estimated the seasonal and the working days component in a consistent way. These estimates have been carried out for overall quarterly GDP as well as for its sectoral components. For the more detailed approach, a total effect of around Euro 56 million per additional working day was derived, for aggregate GDP an amount of Euro 40 million. Thus, the impact of the three additional working days in 2004 would in both cases be some 0.1 percent of annual GDP. In recent periods, the positive effect was greatest in the first quarters of 2000 and 2001, when the number of working days was 1.22 above the average, raising quarterly GDP by about 0.1 percent, respectively.

However, this result should be interpreted with caution, since the underlying approach assumes the working-days effect to be independent from the cyclical component. The effect obtained is an average of values that can vary in positive and negative direction, depending on the phase of the business cycle.

The method used differs from the one applied by the Deutsche Bundesbank in calculating the working-days effect for Germany. The latter yields an output increase of 0.6 percent of GDP for the four additional working days occurring in 2004.

In general, quarterly GDP forms a weak base for calculating a working-days effect, since its calculation is designed for the monitoring of the business cycle and a proper identification can only be done on a monthly basis. Moreover, the quarterly national accounts are established in some instances upon sectoral data which themselves do not include a working-days effect.

The working-days effect obtained does not allow straightforward conclusions on the potential impact of changing a public holiday into a working day or of a day's strike. Thus, even in sectors for which no working-day effect can be identified, e.g., because of shift work on weekends, a strike day may well reduce output and vice versa.

Vienna, 10 February 2004.

For further information, please refer to Marcus Scheiblecker, phone (1) 798 26 01, ext. 245, E-Mail-address Marcus.Scheiblecker@wifo.ac.at