Standard real business cycle theory predicts that consumption should be smoother than output. Agents focus on permanent rather than temporary income when making decisions, so they react to an increase in income by saving part of it for future consumption. This prediction holds for developed countries, where the volatility of consumption is in general smaller than the volatility of output. In emerging economies however consumption is more volatile than income (23% at annual frequencies and 40% at quarterly frequencies), a phenomenon known as the Consumption Volatility Puzzle. In this paper we explore whether the introduction of political frictions into a standard real business cycle model can qualitatively and quantitatively explain this puzzle.
We are motivated by three observations drawn from analyzing stylized facts in emerging versus developed economies. First that there is a positive correlation between the variability of private and public consumption. Second that the relative volatility of consumption is positively correlated with the degree of political polarization (the dispersion in political preferences among the population). Finally, emerging economies tend to be more polarized. Figure 1 illustrates the last two points.
Our main hypothesis is that fluctuations in economic variables are not only caused by innovations to productivity but also by shifts in ideology regarding the size of government. Countries that are more polarized exhibit larger swings in the level of spending. Because consumption responds more than output to the resulting changes in taxation these countries will tend to have a larger relative volatility of consumption. We elaborate this argument in a dynamic political economy model and provide intuition by looking at an example economy for which analytical solutions can be computed. We then calibrate the more general environment to a set of emerging and developed economies in order to quantify how much of the variability in relative consumption volatility can be explained by the model by varying their degree of political polarization.
Our setup embeds Persson and Svensson’s (1989) political economy model of public-goods provision in a neoclassical growth framework. Political parties that disagree on the size of the government alternate in power. Left-wing parties place more weight on public spending than right-wing parties and hence tax income at a higher rate in order to finance a larger level of expenditures. This introduces an additional source of volatility for economic variables triggered by changes in government policy which can be interpreted as political shocks. In contrast to TFP shocks, a political shock affects consumption immediately through changes in agents’ disposable income, while the response of output (caused by changes in investment) is delayed and muted.
As a result, consumption volatility can be larger than output volatility in the presence of political shocks. Our mechanism is related to the earlier work of Dotsey (1990) and Baxter and King (1993) who study the effects of exogenous government expenditure shocks on macroeconomic activity. A main departure is that public policy is endogenous in our model. We are able to generate the stylized fact that government spending is more volatile in emerging economies than in developed ones (the average volatility is three times as large, as shown in Table 1) and to provide a reasonable channel by which this happens.
Download
Partisan cycles and the consumption volatility puzzle
