On the permanent effect of an aggregate demand shock: Evidence from the G-7 countries Omar H.M.N. Bashar School of Accounting, Economics and Finance, Deakin University, Australia abstract article info Article history: Accepted 5 February 2011 JEL classifications: C32 E31 E32 Keywords: Structural VAR Blanchard–Quah Decomposition G-7 Country This paper extends the work of Cover, Enders and Hueng (2006) to examine the idea that an aggregate demand shock may have permanent effect on the output level by indirectly shifting the aggregate supply curve. We utilize the bivariate SVAR modeling and adopt an identification scheme, which allows for the possibility that a shift in the aggregate demand curve may induce the long-run aggregate supply curve to shift. We have shown that aggregate supply shocks are positively affected by the demand shocks in each of the G-7 countries. It is found that a one-time positive aggregate demand shock increases the output level permanently in these industrialized economies. We have also shown that our decomposition strategy can help resolve anomalies in the responses of inflation to a positive aggregate supply shock observed in a simple Blanchard– Quah decomposition. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Do aggregate demand shocks affect the output level permanently? Conventional macroeconomic analysis suggests ‘No’, and as a matter of fact, many previous prominent empirical studies were based on the assumption that an aggregate demand shock has only a temporary effect on the aggregate output level. The classic paper by Blanchard and Quah (1989) first applied this assumption within the SVAR framework in order to isolate the demand shocks from supply shocks. To date we can find a huge amount of literature that applied the Blanchard–Quah (BQ, henceforth) approach in identifying macroeco- nomic shocks. The main assumption in this approach is that the long run aggregate supply curve is vertical, and a shift in the aggregate demand curve will increase the inflation (or the price level) proportionately, but will not alter the output level. The BQ technique further assumes that the shocks are uncorrelat- ed. Recently Cover et al. (2006) questioned this assumption. Using the US data, they showed that this assumption leads to a complete isolation in the dynamics of inflation and output. With demand and supply shocks being uncorrelated, changes in output were found to be driven mainly by the supply shocks and inflation by the demand shocks. Cover et al. (2006) showed that this finding can be reversed if we allow the supply shock to be affected by the demand shock. In the current paper, we move forward a step further and basically address the question: can a demand shock affect the supply shock, which in turn, may cause a permanent effect on the output level? There indeed exist a few significant literatures that draw attention to the role of demand in affecting innovation in technology in the production process and point to the fact that changes in demand that raise the level of output in the short-run can, through a number of channels, exert a permanent influence on the supply side. For example, Stadler (1990) states that “….changes in the utilization of factor inputs when demand changes can result in reorganization and the acquisition of new skills; or a higher level of output may make innovation more profitable and result in the allocation of more resources to R&D”. With the use of theoretical models with endogenous technology, this study essentially shows that “Changes in the supply side of the economy are not independent of changes on the demand side”. The empirical work of Utterback (1974) also points to the same idea, which shows that majority of the innovations in industry takes place in response to market demand conditions. Lucas (1972, 1973) shows how an unanticipated change in money (an aggregate demand shock) could affect output. This effect occurs because firms misperceive the aggregate demand shock for the relative demand shock. However, this effect should not be persistent, as firms would have perfect information about the demand shock over time. Blanchard (1987) surveyed the literature on the effects of money on output and observed that if the initial misperception about the shock led firms or workers to change a state variable, which affect their decision in subsequent periods, then the initial changes in money or demand would have persistent effect. For instance, if the initial misperception leads the firms to invest in productivity enhancing technology, then we may expect a shift in the long run aggregate supply curve, which in turn will change the output level permanently. Economic Modelling 28 (2011) 1374–1382 E-mail address: omar.bashar@deakin.edu.au. 0264-9993/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.econmod.2011.02.013 Contents lists available at ScienceDirect Economic Modelling journal homepage: www.elsevier.com/locate/ecmod