Kobayashi-sensei's Economic Research Picks

Part Four: Collateral Constraints and Nominal Debt in the Formation of Monetary Policy - Should monetary policy respond to movements in asset prices?

Faculty Fellow

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: In the New Keynesian-type model currently being developed, the major theme is how to minimize inefficiency due to price stickiness (staggeredness in price changes) through monetary policy. This time, I would like to introduce "House Prices, Borrowing Constraints, and Monetary Policy in the Business Cycle" (American Economic Review 95[3]: 739-64) by Matteo Iacoviello (2005). This paper discusses what happens if distortions such as collateral constraints on debts and the fact that debt contracts are set in nominal terms (i.e. they cannot be set in real terms) are considered in addition to price stickiness.

Econo-kun's photoEcono-kun: What are the major features of the paper?

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: It theoretically discusses how two distortions; (1) "collateral constraints" and (2) "nominal debt" affect macroeconomic variables, and makes a comparison with United States data using simulations. The major contribution of the paper is its development of a model for the relationship between asset prices (land prices) and other variables such as production and inflation, incorporating these two distortions, and presenting desirable monetary policy theoretically (quantitatively) under these distortions.

In this paper, collateral constraints are modeled as follows.As discussed in Kiyotaki and Moore (1997), two types of households (and firms) with different time preference rates are assumed: "patient" and "impatient." As the impatient households are not satisfied with the market interest rates, they consume faster than the patient households. Therefore, they cannot save internal funds and borrow as much as possible to smooth consumption. When borrowing money, there are constraints for households to provide real estate as collateral (there is little discussion regarding micro-foundation in terms of why collateral is required). Also, the paper analyzes cases where debt contracts are set only in nominal terms or where contracts can be set in real terms, and considers the differences between the cases.

In order to embed these collateral constraints within the New Keynesian-type price stickiness model (Calvo-pricing model), the paper assumes that all goods produced by firms are sold to mass retailers in a perfectly competitive market. It also assumes that retailers process and differentiate the goods at no cost, and that the retailers are under monopolistic competition with few chances to modify the retail price, i.e. under the constraints of the Calvo-pricing model. In other words, upstream firms/households produce goods under collateral constraints and the market for the goods is perfectly competitive, but a downstream retail market has New Keynesian-type price stickiness (like a structure where a Kiyotaki-Moore model in the upstream is combined with a New Keynesian model in the downstream). Then, Iacoviello estimates the parameters to make a series of data generated by the model match the actual U.S. economic data series (using a methodology mentioned in Christiano, Eichenbaum, and Evans [JPE 2005]).

Econo-kun's photoEcono-kun: And what do we find in the simulation?

Kobayashi Keiichiro's photoKOBAYASHI Keiichiro: The paper provides three major results. The first is that in the U.S. economic data the collateral effects in this model can explain why real spending shows a positive response to the rise in asset prices. The second point is that in cases where debt contracts are set only in nominal terms, real spending produces a hump-shaped response to an inflation shock; thus the simulated result is closer to the actual data. This is due to the debt deflation effect (amplifying inflation shocks to constrain firm/household spending because debts are fixed in nominal terms). The third point is the finding that even if the monetary policy responds to asset prices, it has almost no effect (in terms of output and inflation stabilization). The implication that the monetary policy should not respond to asset prices (a response could in fact be harmful to the economy) is also suggested by prior research by Bernanke and Gertler (2001), and others. This paper indicates that if the monetary policy responds to asset prices then its effect is slightly improved, but at a possibly negligible level.

A minor problem in this model is the assumption that borrowing by firms/households (a subject of collateral constraints) is used to smooth purchases of land and consumption (by firms/households). Although firms borrow money to employ labor and purchase raw materials in an actual economy, there are no collateral constraints on the purchase of inputs in this model.

If firms need to finance money for a labor force and the purchase of raw materials and collateral constraints are put on such borrowing, then the simulation results of the model may differ to some degree. For example, the results may imply that the optimal monetary policy is to respond to the movements of assets prices to some degree.

May 11, 2006
  • Kiyotaki, Nobuhiro, and John Moore. 1997. "Credit Cycles." Journal of Political Economy 105, pp. 211-248.
  • Ben S. Bernanke & Mark Gertler, 2001. "Should Central Banks Respond to Movements in Asset Prices?," American Economic Review, American Economic Association, vol. 91(2), pages 253-257.
  • Christiano, Lawrence, Martin Eichenbaum, and Charles Evans. "Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy." Journal of Political Economy 113 (February 2005)

May 11, 2006

Article(s) by this author