1/2010

Financial accelerator and investment in a small open economy in a currency union

It was not long ago when one had to appeal to either highly adverse systemic events like the Great Depression of the 1930s in the US and the collapse of the Nordic economies in the late 1980s and early 1990s, or to the experiences of many emerging market economies, in order to motivate research on financial factors in business fluctuations and to induce someone to write a paper for publication in a refereed international journal on this topic. Ever since the publication by Kydland and Prescott of the their article in 1982[1] and all the way up until the onset of the most recent financial crisis and global recession, mainstream macroeconomic theory that relied on dynamic general equilibrium representative agent modelling methodology saw no deeper reason to develop the approach to business cycle modelling so as to incorporate a genuine role for financial market frictions and imperfections. In fact, these models adopt the assumptions underlying the Modigliani-Miller theorem, which implies that the financial structure is both indeterminate and irrelevant to real economic outcomes. For years, existing business cycle models of the mainstream type appeared to be successful in accounting for the prime features of normal business cycle fluctuations, both qualitatively and quantitatively. Thus Occam's razor would imply that there is no need to further complicate the mainstream business cycle models.

Now, the things have changed quite a bit. It is no longer necessary to make references to old ghosts to motivate research on the interaction of financial factors and business cycle fluctuations. Over the past few years, the US and much of the industrialized world have experienced the worst financial crisis of the post WWII period. The global recession that followed the financial crisis also appears to have been the most severe of this era. Currently we getting optimistic signals that the financial sector has stabilized and the real economy has stopped contracting. However, the signals, in particular on the real economy, still appear to relatively weak, and the path to recovery remains highly uncertain. By arguing that things have changed and that the macroeconomic profession currently yearns for models of aggregate economic activity incorporating credit market frictions, we are not denying the existence of a robust literature in this area, which has been developing over the last several decades. In fact Bernanke, Gertler and Gilchrist (BGG, 1999)[2] provided an excellent survey of much of the earlier work a decade ago, and the literature has continued to expand since that survey. Much of this work continues to be relevant also to the current situation. But this literature obviously did not – and could not – anticipate all the key phenomena that have been observed during the most recent crisis. A literature that builds on and extends the earlier work in addressing these issues is cropping up with surprising speed. Although most of this new literature is in preliminary working paper form, it will not take long before the contributions start to appear in refereed journals.

There is more than one way to incorporate financial frictions into a standard business cycle model of the New Keynesian variety. Borrowing and collateral constraints are two possibilities that have been used quite frequently. The financial accelerator mechanism of BGG provides yet another approach to modelling financial frictions in a model of business cycle fluctuations. BGG develop a dynamic general equilibrium model that is intended to help clarify the role of credit market frictions in business fluctuations, in both qualitative and quantitative terms. The financial accelerator in their framework derives from endogenous developments in credit markets that work to amplify and propagate shocks to the macroeconomy. More specifically, the key mechanism involves the link between the external finance premium (difference between cost of funds raised externally and the opportunity cost of the firm’s internal funds) and the net worth of potential borrowers (defined as borrowers' liquid assets plus collateral value of illiquid assets less outstanding obligations)[3]. On the other hand, to endogenously motivate the existence of an external finance premium, BGG postulate a simple agency problem that introduces a conflict of interest between a borrower and his respective lenders. The financial contract is then designed to minimize the expected agency costs.

Although the financial accelerator mechanism introduced by BGG is theoretically simple and very interesting, including in particular a linear relationship between the demand for capital goods and entrepreneurial net worth that facilitates aggregation, much empirical work in needed to establish and quantify the strength of this mechanism. In particular, as the procyclicality of entrepreneurial net worth, and hence countercyclicality, of the external finance premium lies at the core of the mechanism amplifying the effects of business cycle shocks, hard evidence is needed to establish the importance of this core to business cycle fluctuations and to quantify its strength. In her forthcoming paper Financial market disturbances as sources of business cycle fluctuations in Finland Hanna Freystätter presents an analysis with the aim of providing empirical evidence of the strength of the financial accelerator mechanism and the role of financial market shocks for business cycle fluctuations in the small open economy of Finland. The latter extension is highly relevant, since, as she also notes, the 2007–08 financial crises has also shown the need for new sources of shocks stemming from the financial market itself and for assessing both the qualitative and quantitative importance of financial market disturbances in understanding macroeconomic dynamics.

To address the issue in her research agenda, Freystätter constructs a New Keynesian DSGE model that incorporates the financial accelerator mechanism of BGG as well as a rich set of stochastic shocks hitting the economy. Two of the most interesting shocks are domestic credit market shocks reflecting exogenous sources of entrepreneurial wealth destruction and exogenous shocks to finance premiums that capture exogenous changes in domestic financial intermediation. Here Freystätter follows Gilchrist, Ortiz and Zakrajsek (2009)[4] and calls this latter shock a credit supply shock. The model is then subjected to the data (quarterly over the period 1995–2008) and estimated using Bayesian Maximum Likelihood methods. As she notes, from the point of view of investigating the potential role of financial markets for aggregate fluctuations, the estimation period is particularly interesting and relevant. Around the turn of the century, the Finnish economy, along with many others, experienced a stock market boom, and the estimation period also covers the start of the financial market turmoil in 2007. Furthermore, her analysis takes into account an important feature of the small open economy of Finland: as part of the euro area, the Finnish economy cannot rely on the two important channels that help a standard small open economy to adjust to economic shocks, namely the nominal exchange rate and the policy rate set independently by the central bank.

In contrast to much of the existing literature, Freystätter studies a small open economy where shocks originating in international markets play an important role. By extending existing model, she is able to evaluate, via her estimation and simulation exercise, the relative importance of domestic and international shocks to the aggregate fluctuations of the domestic economy. According to her results, domestic financial market shocks emerge as key drivers of recent business cycle fluctuations in Finland, even after allowing for several sources of domestic and international shocks. Freystätter also notes that the results are obtained without using any financial market data in the estimation, in contrast to Gilchrist, Ortiz and Zakrajsek (2009), who construct and use a highly sophisticated measure of credit spread in their model estimation. This makes it possible for Freystätter to assess the performance of the model by investigating the match between the relevant part of the model outcome and financial market data. One further important aspect of her exercise is the exclusion of investment-technology-specific shocks from the analysis. That this is important is revealed in the empirical DSGE literature, which has often argued that investment-specific shocks are among the most important driving forces of economic fluctuations. However, as argued in the background literature[5], investment-technology-specific shocks can actually hide unmodelled frictions in the capital accumulation process.

The evidence presented by Freystätter suggests that there is an operative financial accelerator mechanism in Finland. The estimate of the parameter governing the strength of the financial accelerator mechanism is of the right sign and close to values obtained in the relevant international reference literature using estimated DSGE models to study the quantitative significance of the financial accelerator for the aggregate economy. Hence, she concludes that the financial accelerator mechanism acts as an amplifying mechanism for many disturbances hitting the Finnish economy. Furthermore, according to her main results, disturbances stemming from the financial market itself contributed significantly to the cyclical fluctuations of the Finnish economy between 1995 and 2008. On the basis of her results, Freystätter is able to argue that domestic financial market shocks impinging upon the creation of entrepreneurial wealth and the demand for capital are key drivers behind aggregate investment dynamics in Finland. These shocks consequently explain particular business cycle episodes in Finland, such as the boom and bust of the stock market late 1990s and early 2000s, the subsequent early millennium slowdown and, more recently, the sudden reversal of investment activities in 2008 due to the global financial crises.

Jouko Vilmunen