Kobayashi-sensei's Economic Research Picks

Part Two: A New Understanding of Debt Deflation Theory

KOBAYASHI Keiichiro Faculty Fellow

Econo-kun Econo-kun is in his second year of the master's program at a private university, studying hard to become an economist.

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: This time, I would like to focus on "'Sudden Stops' in an Equilibrium Business Cycle Model with Credit Constraints: A Fisherian Deflation of Tobin's q" by Enrique G. Mendoza (2004).

In the 1994 Mexican currency crisis (the so-called Tequila Crisis) as well as in those that hit Asia, Russia, Brazil, and Argentina in subsequent years, an abrupt withdrawal of foreign capital sent the otherwise sound national economy into recession. This phenomenon, also known as the "capital account crisis" or the "21st century-type crisis," has been dubbed "Sudden Stop" by Professor Guillermo Calvo at the University of Maryland. Mendoza (2004) argues that the Sudden Stop occurs when normal shocks observable under normal business cycles, such as changes in productivity and world interest rates, are combined with certain specific conditions triggering a sharp fall in asset prices (asset deflation). This concept attempts to explain a currency crisis by using a real business cycle model in a small open economy.

Econo-kun's photoEcono-kun: Is there any particular reason why you have picked deflation as a theme twice in a row?

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: Deflation is tied to my recent research interests. I believe that it is possible to develop an integrated theoretical framework that can explain, in a dynamic general equilibrium model, unusually large recessions that occur outside the bounds of normal business cycles. These include the U.S. Great Depression of the 1930s and Japan's protracted recession in the 1990s, as well as a series of Sudden Stops that hit developing countries. (These large recessions are called "great depressions" by Patrick Kehoe, Edward Prescott, and others.) Borrowing constraints (or collateral constraints) and changes in asset prices are the key factors. For instance, when asset prices fall and companies come under more stringent collateral constraints, capital investments decrease and employment declines. This mechanism can explain, to some extent, the U.S. Great Depression and Japan's prolonged recession. (This idea is discussed in my discussion paper, "Borrowing Constraints and Protracted Recessions.")

Mendoza (2004) further emphasizes the importance of collateral constraints as a factor explaining not only large recessions in major economies such as Japan and the U.S. but also Sudden Stops in small economies such as Mexico.

Econo-kun's photoEcono-kun: Could you give a brief overview of Mendoza's paper?

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: Mendoza points to a mechanism in which when borrowing constraints become tighter on falls in asset prices, economic agents are forced into a fire sale of assets and this, in turn, further drives down asset prices. Mendoza's model assumes that households borrow from abroad to smooth consumption but, in doing so, they are required to offer assets (physical capital) held in their country as collateral. In the small open economy assumed in the model, even a modest decline in asset prices triggered by a minor incident results in tighter collateral constraints when the debt ratio exceeds a certain level. Bound by collateral constraints, households are not able to increase borrowing. Thus, they instead try to smooth consumption by selling (essentially a fire sale of) their physical capital, which results in further falls in asset prices. This further tightens collateral constraints, which leads to a lesser amount of overseas funds available for households, then to more fire sales. Mendoza notes that a Sudden Stop - in which asset prices collapse while both consumption and capital investments fell sharply - is consequently triggered by this vicious cycle.

This mechanism is almost identical to the one identified by Irving Fisher's debt deflation theory, which was developed in 1933 to explain the U.S. Great Depression. Mendoza has succeeded in reproducing Fisher's debt-deflation theory in dynamic general equilibrium modeling.

Along with collateral constraints, Mendoza also introduced the concept of borrowing constraints that limit companies' access to working capital financing. Specifically, it is assumed that companies need to rely on external debt to finance part of their raw materials costs and capital investments, but the amount of such working capital loans cannot exceed an amount equal to their gross sales net of wage costs. Without such borrowing constraints, that is, if it only placed greater collateral constraints on households, Mendoza's model would produce the following unrealistic outcomes:
1) Output, factor inputs, capital utilization, and working capital do not respond quickly when a Sudden Stop starts.
2) Capital utilization rises in the midst of depression caused by the Sudden Stop.

The presence of borrowing constraints on working capital financing prevents these problems, and realistic outcomes can be derived.

Econo-kun's photoEcono-kun: How has the debt deflation theory been viewed thus far?

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: About 50 years after Fisher proposed the debt deflation theory, Ben Bernanke and Mark Gertler translated it in a theoretical model. However, this "financial accelerator" theory is designed to explain the mechanism of how an unexpected transfer of income from debtors to creditors, caused by deflation, inhibits capital investment and other growth initiatives, thereby further aggravating the depression. This is to explain the one-way (and probably temporary) phenomenon of deflation escalating into depression, which is different from the kind of vicious cycle mechanism cited by Fisher, in which falling asset prices and decreasing demand affect each other reciprocally (that is, when borrowing constraints become tighter as a result of deflation, the fire sale of assets intensifies, which in turn causes further falls in asset prices). Fisher highlighted the mechanism in which debtors' efforts to repay their debt accelerate asset deflation. However, the theory developed by Bernanke and Gertler did not specify the cause of deflation; deflation was simply defined as an exogenous shock, i.e. an event caused by external forces. Therefore, their theory is unsatisfactory as their model cannot be considered a true translation of Fisher's debt deflation theory. In comparison, Mendoza succeeded in illustrating the reciprocally vicious cycle between deflation in asset prices and decreases in borrowing (decreases in demand) by incorporating collateral constraints into his model. This, indeed, is the very first model to truly reproduce Fisher's debt deflation theory, exactly as he intended it, in the world of dynamic general equilibrium modeling.

So, there is a possibility that abnormal phenomena that occurred in "large recessions" can be explained by taking into consideration the roles played by collateral constraints. Japan's deflation from the mid-1990s onward may be theorized in the same way.

Econo-kun's photoEcono-kun: It would be incredibly significant if someone could come up with a theoretical explanation for this.

December 27, 2005

December 27, 2005

Article(s) by this author