Several emerging market economies during the 1990s, such as Mexico, south-east Asian countries, and Chile, displayed episodes of peaking growth rates along with increasing current account deficits and appreciating currencies, that ended with abrupt reversions in capital flows and recessions.
Previous to the recessions, in all cases there was enhanced optimism about future prospects. Mexico was negotiating its entrance to NAFTA, along with its membership to the OECD. Chile had exhibited a smooth transition to democracy. The mood of investors was increasingly enthusiastic about the prospects of harvesting the benefits of the market reforms of both the previous period and those introduced under democracy. The south-east Asian economies, in turn, had their own reasons for optimism based on their impressive growth record of previous years. In all cases, optimism was grounded on reasonable arguments, but the prospects of future economic growth could not be estimated accurately.
In this paper we show that overoptimistic perceptions regarding the future by domestic private agents –domestic “exuberance”– could have been a cause of the boom-bust cycles observed in some emerging economies during the 1990s. To that end, we develop a multi-sector dynamic stochastic general equilibrium (DSGE) model for a small economy with short-run stickiness in prices and wages, that features expectation driven boom-bust cycles. We show that under standard parametrization, the model is able to closely match most of the stylized facts observed in the boom-bust episodes in emerging markets. In the model, private agents are rational and forward looking. Therefore, their current decisions rely on their assessment on future productivity prospects. An overoptimistic assessment about future productivity makes them to accumulate excess capital and to over increase their consumption, leading to a boom that is accompanied by a current account deficit. When agents realize that productivity will grow by less than expected, they must readjust their investment and consumption profiles, generating a current account reversal and a recession.
Our analytical approach follows closely Christiano, Ilut, Motto, and Rostagno (2007) (hereafter CIMR). Unlike them, we show that overoptimism about productivity trends, rather than transitory productivity gains, are the source of boom-bust cycles in open economies such as the one observed during the 1990s. We show that if productivity changes follow a stationary process, where expected productivity improvements are perceived to be transitory, news about future productivity improvements are not able to replicate the real appreciation of the currency and the current account deterioration along the boom as observed in the data. This result is related to the work of Aguiar and Gopinath (2007), who show that the observed strong counter-cyclicly of the current account in emerging economies can be explained by productivity trend shocks in a standard real business cycle model. In our case, cycles are generated by miss-perceptions about future productivity growth rather than actual changes in this variable.
According to our model, a boom-bust cycle generated by domestic agents overoptimism is observational equivalent to a cycle driven by exogenous fluctuations in foreign financial conditions. Several authors have claimed that swings in external financial conditions were significant factors behind the observed patterns of macroeconomic variables during the 1990s in many emerging markets (Neumeyer and Perri, 2005; Uribe and Yue, 2006; Valdés, 2007). In this sense, our results can be interpreted as a plausible alternative although complementary explanation for the episodes of abrupt current account deterioration in emerging markets during the 1990s.
Among the policy implications, our model show that the trade-offs faced by the monetary policy in a boom-bust cycle driven by expectations are not trivial. If the central bank tries to stabilize output, the result will be a large fall in inflation and a contraction in output in the tradable goods sector. On the other hand, if the central bank targets inflation more strictly, then the boom in activity, the current account deterioration and the exchange rate appreciation will be larger, and the subsequent recession more severe. If we modify the policy rule to include an endogenous response of the interest rate to exchange rate fluctuation, then the perverse effects on the domestic tradable goods sector are only prevented in the short run, but the boom-bust cycle in other variables is amplified.
Expectations driven macroeconomic fluctuations may be drawn back to at least Pigou (1926). Recently, this hypothesis has received renewed attention in modern macroeconomics. Marfán (2005) analyzes boom-bust cycles provoked by excess optimism and concentrates mainly on the role of fiscal policy in an extended Mundell-Fleming context. The optimist-pessimist mood of the private sector in his model is completely exogenous. Beaudry and Portier (2004), Jaimovich and Rebelo (2006, 2007), Mertes (2007) and CIMR present different unique equilibrium rational expectation models where business cycles are generated by changes in expectations regarding productivity prospects. Jaimovich and Rebelo (2006, 2007) discuss which elements are needed in a standard Real Business Cycle (RBC) models to generate the co-movement observed in the data in response to non-materialized productivity shocks. They show that in a closed economy environment, adjust cost in investment and/or labor, variable capital utilization, and weak wealth effects on labor supply are key to replicate the co-movement in the data. In an open economy set up, variable capital utilization is not that crucial. CIMR, using a sticky-price sticky-wages model, emphasize the role played by the monetary policy in generating expectation driven boom-bust cycles. They show that to generate a sizeable output expansion and a boom in stock prices in response to news about increased future productivity, monetary policy has to respond aggressively to the induced fall in inflation. Thus, the boom is amplified by a loose monetary policy. Mertes (2007) shows that an expectation-driven RBC model is able to replicate relevant stylized facts of the Sudden Stop faced by Korea at the end of the 1990s.
The expectations driven business cycle approach in this literature is related to the literature on multiple equilibria and sunspots cycles (Farmer, 1993). It can also be view as complementary to the literature on rational herding and information cascades lead cycles (Banerjee, 1992; Chamley and Gale, 1994; Caplin and Leahy, 1993 and Zeira, 1994). In particular, this strand of literature has emphasized how information may occasionally be aggregated improperly thereby leading to non-fundamental cycles. In this paper, we examining whether the quantitative implications of (rational/non-systematic) aggregate forecast errors can explain the observed pattern of recessions of small open economies within a fully specified dynamic model that features an unique general equilibrium.
The paper has five sections including this introduction. The second section provides a motivation about boom-bust cycles as well as a broad view of such cycles in the 1990s in Chile, Korea and Mexico. It also discusses stylized effects of structural reforms and innovations on future growth. The third section describes in a detailed way the theoretical model, and the calibration of the parameters. The fourth section analyzes the dynamics of the empirical model and discusses the tradeoffs faced by the monetary policy. The fifth and final section summarizes the main findings.
Download
PDF Ebook Overoptimism, Boom-Bust Cycles and Monetary Policy in Small Open Economies
