Die Dis cus si on Pape rs die nen einer mög lichst schnel len Ver brei tung von neue ren For schungs arbei ten des ZEW. Die Bei trä ge lie gen in allei ni ger Ver ant wor tung der Auto ren und stel len nicht not wen di ger wei se die Mei nung des ZEW dar.Dis cus si on Papers are inten ded to make results of ZEW research prompt ly avai la ble to other eco no mists in order to encou ra ge dis cus si on and sug gesti ons for revi si ons. The aut hors are sole ly respon si ble for the con tents which do not neces sa ri ly repre sent the opi ni on of the ZEW.Download this ZEW Discussion Paper from our ftp server:ftp://ftp.zew.de/pub/zew-docs/dp/dp08007.pdf
Non-Technical SummaryThis paper analyses and extends two seminal macroeconometric studies by Gali (1999) and King et al. (1991; henceforth KPSW) We show in this paper that this finding extends to the six-variable case as well. Therefore, we augment the six-variable model with hours worked to create a seven-variable one and impose the identification restriction of Gali in order to estimate technology shocks together with labor-supply, inflation and real-interest-rate shocks. This augmentation yields several advantages for macroeconometric analysis. First, the dynamic effects of more structural shocks on more macroeconomic variables can be investigated than in a bivariate model.Second, we can check whether the so-called omitted variables bias exists, and if it does, how it affects the identification of macroeconomic mechanisms. Third, we can distinguish between transitory and permanent shocks due to the presence of cointegration in contrast to the bivariate model.The sample we use is taken from KPSW and Gali (1999) and covers the period 1954:1 to 1988:4. Thus, our findings are almost directly comparable to the findings in these papers as well as to those in Carlsson (2001, 2005).When augmenting the bivariate model to a seven-variable model, we consider several subsets. Our main finding is that the identification restriction in the bivariate model leads to a robust estimation of technology shocks in every subset model as well as in the sevenvariable model. A second finding is that the estimation of inflation and real-interest-rate shocks is not greatly affected by the type of restriction used to identify technology shocks.Third, we obtain that the KPSW identification scheme is very sensitive to the presence of nominal variables, whereas the Gali restriction for identifying technology shocks is robust to the inclusion of nominal variables in the VAR. Fourth, the estimated technology measures of both three-variable and six-variable KPSW models contain a significant labor-supply component; i.e., they are mingled with nontechnology and demand measures.We furthermore investigate the driving forces of business cycles. Whereas Gali (1999) compares the comovements of the business cycle components of macroeconomic variables and KPSW employ forecast error variance decompositions for a similar investigation, we find using historical variance decompositions for the same purpo...