Standard macroeconomic textbooks generally present macroeconomics in two separate bodies: in the long term an economy's performance is essentially influenced by structural characteristics, such as education, R&D, openness to trade, competition or fnancial development. In the short term however, the economy is essentially influenced by the shocks it undergoes and stabilization policies undertaken (fiscal and monetary policy). These two approaches have been considered for long as separate and distinct bodies of research. Stabilization policies for instance are considered to have no signi?ficant impact on the long run performance of an economy.
The point of this paper is to investigate (the relevance of) this dichotomy focusing on the impact, if any, of cyclical ?scal policy on growth. To answer this question, we take a two step approach. First we build a simple model to illustrate how the cyclical component of ?scal policy can affect growth. Second we take the theoretical predictions to the data and provide empirical evidence of a statistically and economically signi?cant impact of stabilizing fiscal policy on growth.
The theoretical part of the paper is based on a model with risk neutral entrepreneurs and lenders. Entrepreneurs can choose a project to invest in among a set of existing projects, the more productive being also the more risky. When states of nature are non veri?able or alternatively if veri?ability is sufficiently costly then entrepreneurs who invest in more productive projects also face a tighter borrowing constraint because higher average productivity implies lower output in bad states and hence a lower ability to pay back liabilities. To put it in a nutshell, when states of nature are non veri?able, pledgeable income is negatively related to average productivity which creates a trade-off for entrepreneurs in their technological choice. The government can then alter this trade-off by imposing state contingent taxes.
Namely a pro-cyclical fi?scal policy, i.e. high taxes in bad states and low taxes in good states, tends to amplify the negative effect of more risky investments on the ability to borrow. As consequence, entrepreneurs optimally choose less risky and less productive projects. On the contrary a counter cyclical ?scal policy, i.e. low taxes in bad states and high taxes in good states, tends to dampen the negative effect of more risky investments on the ability to borrow which prompts entrepreneurs to take more risky bets. Moreover the positive effect of counter-cyclical fi?scal policy on productivity growth increases with the share of investment financed through external capital but decreases with income pledgeability. The second part of the paper is devoted to test empirically these three predictions: (i) counter-cyclical ?scal policy is growth enhancing, (ii) the growth enhancing impact of counter-cyclical ?scal policy should increase with the share of investment ?nanced through external capital, (iii) but decrease with income pledgeability.