Fluctuation of payroll taxes harm state fiscal austerity

Conclusion: This paper develops a general theory of state-dependent fiscal multipliers for a broad range of spending and taxation policies. The framework accounts for empirically relevant goods market frictions by incorporating idle productive capacity and unsatisfied households’ demand into an otherwise standard general equilibrium setup. Our key novel finding is that cyclicality of fiscal multipliers is pinned down by the source of economic fluctuations, and we provide model-free econometric evidence that strongly supports our predictions. Crucially, we establish that multipliers associated with fiscal instruments which stimulate aggregate demand, such as government spending and consumption tax cuts, are countercyclical under demand-driven fluctuations and procyclical under supply-driven fluctuations. On the other hand, multipliers associated with interventions that stimulate aggregate supply, such as reductions in taxes on firms’ payroll, sales and households’ labor income, are countercyclical under supply-driven fluctuations and procyclical under demand-driven fluctuations. In addition, our theoretical results establish a relevant role for fiscal austerity, implemented by a reduction in government consumption in severe supply-driven recessions and demand-driven booms, provided elasticities of labor supply and labor demand are sufficiently low. Further, we develop and estimate a novel econometric specification that allows us to perform model-free evaluation of both spending and taxation multipliers in recessionary and expansionary episodes, conditional on those being either demand- or supply-driven in nature. Our empirical results detect substantial state dependence, conditional on the source of fluctuations, which is in line with the predictions of our theory. Such findings offer a resolution to the debate on state dependence of fiscal policy, on both empirical and theoretical grounds, and they provide guidance for the conduct of fiscal policy in the different phases of the business cycle. Our analysis opens fruitful avenues for future research. First, in our dynamic framework, current changes in fiscal policy determine the future path of goods market tightness and thus constrain the effectiveness of policy in the future. Our framework can therefore be extended to study the intertemporal trade-offs and the path dependence of fiscal policy. Second, by extending the model to include heterogeneity in the goods markets, one can study how composition of spending and taxation policies may generate of fiscal policy across sectors and the socially optimal distribution of such policies. We plan to investigate these issues in future research.