Abstract
This paper investigates the effects of changes in public investment on industrial output, nominal industry wages and household consumption, to identify the channel through which public spending affects aggregate demand and boost industrial output and productivity. I use both Fixed effects and Instrument Variables regression on a panel for 20 select OECD countries for the period 1996-2015. Our main results from fixed effect regression show public investment have a positive effect on industrial output and nominal industry wages. In particular, we find a 1% increase in annual public investment increases industrial production 0.027%. The effect is higher with Instrumental variables regression increasing production by 0.345%, perhaps showing public investment to be endogenous. The effect on nominal wages is about 0.027% (at 1% significant level) on fixed effects regression but negative at -0.05% with instrumental variables regression but not statistically significant even at 10% level. The effect on household consumption on the other hand is negative with fixed effects with about -0.014% (at 1% significant level) in all our regressions. Our IV results however are not statistically different from zero even at 10% significant level. Household unlike Industrial output and wages seem to react to contemporaneous public investment and so we assume that household being a fast-moving variable, it may react faster from anticipatory effects. The results though not conclusive suggests that households react more to a negative wealth effect in expectations of higher future taxes than to increases in real interest rates’ intertemporal substitution effects. All our results on seems to suggest that public investment has only a transitory effect on industrial output, wages and household consumption and further research on its effect on total factor productivity might reveal its long-term role in industrial development. We conclude that the reaction of industrial output, nominal industry wages and household consumption, to public investment shocks, agree with the studies that predict procyclical wages interpret the cyclicality of employment as a consequence of shifts in labour demand by firms.
This paper investigates the effects of changes in public investment on industrial output, nominal industry wages and household consumption, to identify the channel through which public spending affects aggregate demand and boost industrial output and productivity. I use both Fixed effects and Instrument Variables regression on a panel for 20 select OECD countries for the period 1996-2015. Our main results from fixed effect regression show public investment have a positive effect on industrial output and nominal industry wages. In particular, we find a 1% increase in annual public investment increases industrial production 0.027%. The effect is higher with Instrumental variables regression increasing production by 0.345%, perhaps showing public investment to be endogenous. The effect on nominal wages is about 0.027% (at 1% significant level) on fixed effects regression but negative at -0.05% with instrumental variables regression but not statistically significant even at 10% level. The effect on household consumption on the other hand is negative with fixed effects with about -0.014% (at 1% significant level) in all our regressions. Our IV results however are not statistically different from zero even at 10% significant level. Household unlike Industrial output and wages seem to react to contemporaneous public investment and so we assume that household being a fast-moving variable, it may react faster from anticipatory effects. The results though not conclusive suggests that households react more to a negative wealth effect in expectations of higher future taxes than to increases in real interest rates’ intertemporal substitution effects. All our results on seems to suggest that public investment has only a transitory effect on industrial output, wages and household consumption and further research on its effect on total factor productivity might reveal its long-term role in industrial development. We conclude that the reaction of industrial output, nominal industry wages and household consumption, to public investment shocks, agree with the studies that predict procyclical wages interpret the cyclicality of employment as a consequence of shifts in labour demand by firms.