سیاست های پولی و بازار مسکن ایالات متحده: یک تجزیه و تحلیل VAR با اعمال محدودیت نشانه ای

This article examines the impact of monetary policy shocks on the US housing market using an identification procedure similar to the one suggested by Uhlig [Uhlig, H., 2005. What are the effects of monetary policy on output? Results from an agnostic identification procedure. Journal of Monetary Economics 52, 381–419]. The identification procedure imposes sign restrictions on the response of some variables for a certain period. No restrictions are placed on the response of the housing variable. Overall, the results indicate that housing starts and residential investment respond negatively to contractionary monetary policy shocks. However, the magnitude of the impact is sensitive to the selection of the horizon for which the restrictions hold. Moreover, a comparison of the results with those obtained from a conventional Choleski decomposition, suggests that the impact of monetary policy on the housing market is much less certain under the sign restrictions approach.

There is plenty of evidence in the literature suggesting that the housing market is linked to aggregate economic activity in the US (e.g. Iacoviello, 2005). Housing investment is an important indicator of household wealth (Case et al., 2005), and at the same time is one of the most volatile sectors of the US economy (Bernanke and Gertler, 1995). The importance of the housing market is clearly visible in Fig. 1, which plots the number of new privately owned housing units starts each year (solid line), and recessions as determined by the National Bureau of Economic Research (shaded areas). It is common for a large drop in housing starts to precede a recession.1This article examines the impact of monetary policy innovations on the US housing market. In a recent testimony before the US Congress, the Chairman of the Federal Reserve Ben S. Bernanke argued that “given the substantial gains in house prices and the high levels of home construction activity over the past several years, prices and construction could decelerate more rapidly than currently seems likely. Slower growth in home equity, in turn, might lead households to boost their saving and trim their spending relative to current income by more than is now anticipated”, Federal Reserve Board (2006). The events in the housing market have a significant effect on household wealth, and thus have a big influence over household behavior.2 Therefore, a comprehensive analysis of the impact of monetary policy on the housing market is necessary in order to understand the impact of monetary policy on the whole economy. There is an extensive literature on the impact of monetary policy on different sectors of the US economy (for some examples in the housing sector, see Ahearne et al., 2005 and Falk, 1986). Vector autoregressive (VAR) models are commonly used to measure the impact of monetary policy innovations. VARs provide a plausible assessment of the response of macroeconomic variables to monetary policy shocks without requiring a complete structural model of the economy. However, in order to use VARs the researcher must identify the monetary policy shock. The use of different identification schemes can alter the results significantly. For instance, McMillin (2001) shows that the choice between contemporaneous restrictions (as in Bernanke and Blinder, 1992, Bernanke and Mihov, 1998a, Christiano et al., 1998 and Strongin, 1995) and long-run restrictions (as in Fackler and McMillin, 1998) can deliver important differences in the magnitude and timing of the response of macroeconomic variables to monetary policy shocks. We follow Uhlig (2005) in applying what he calls an “agnostic” identification procedure. The identification scheme imposes sign restrictions on the response of some of the variables for a certain period, while leaving the response of the main variable of interest open. We assume that a contractionary monetary policy shock does not lead to an increase in prices, non-borrowed reserves and real GDP, or decreases in the federal funds rate for a number of periods after the shock. We impose no restrictions on the response of the measure of housing activity. Hence, the response of this variable is left “agnostically” open by the identification procedure. The methodology proposed by Uhlig (2005) is a way to impose explicit theoretical restrictions in the estimation and, at the same time, to leave the main question of interest open. This contrasts with other identification methods used in the literature that impose informal restrictions in order to get “reasonable results”. For example, if using certain specification, the results show that prices increase after a contractionary monetary policy shock, then the model is re-specified to get a more “reasonable result”. These results are therefore influenced by a priori theorizing of what is a “reasonable result”. Moreover, if the results are not consistent with economic theory it is common to call the result a puzzle, as in Sims (1992). In our approach we want to be explicit about the restrictions that we are imposing and not depend on informal restrictions (see Canova and de Nicoló, 2002 and Faust, 1998 for more on this topic). Several authors have previously used other identification schemes to study the impact of monetary policy shocks and money supply shocks on the housing market using VARs. Lastrapes (2002) studied the effect of money supply shocks on the housing market using two identification procedures. First, he assumed that money supply shocks were neutral in the long-run (long-run restrictions as in Blanchard and Quah, 1989). Second, he assumed a block-recursive structure in which housing variables do not affect monetary policy contemporaneously. The results suggest that money supply shocks have a positive impact on different measures of house sales. The results are robust to the use of different identification schemes (see Lastrapes and Potts, 2006 for more on the impact of money supply shocks in the housing market). Wheeler and Chowdhury, 1993 and Hasan and Taghavi, 2002 used a recursive structure with the monetary policy variable before residential investment in the ordering to study the impact of macroeconomic variables in the housing market. Results based on variance decompositions and historical decompositions suggest that monetary policy has important effects on residential investment. Although these studies have provided valuable insights about the impact of monetary policy shocks and money supply shocks on the housing market, these studies imposed restrictions on the response of the housing variable to monetary policy shocks. We impose explicit restrictions on other variables, but not on the response of the housing variable. For comparison purposes we also present the results when the conventional Choleski decomposition is used to identify the monetary policy shock. The remainder of the article is organized as follows. Section 2 discusses the data and the identification procedure. Section 3 presents the impulse response functions when sign restrictions are imposed, while Section 4 reports the impulse response functions when the Choleski decomposition is used to identify the monetary policy shock. Section 5 concludes.

This article has examined the relationship between monetary policy and the US housing market using an identification procedure similar to the one suggested by Uhlig (2005). The identification scheme imposes sign restrictions on the response of some of the variables for a number of periods after a monetary policy shock. We assume that shocks to monetary policy do not lead to increases in the GDP deflator, house prices, a commodity price index, non-borrowed reserves and real GDP, or decreases in the federal funds rate for a certain period. No restrictions are placed on the response of the housing market activity variable (housing starts or residential investment). The response of the housing variable is left “agnostically” open by the identification procedure. Overall, the results suggest that contractionary monetary policy shocks have a negative impact on housing starts and residential investment. The response is, in general, similar across US regions. However, the impact of monetary policy on housing seems to be stronger in the Midwest. When the horizon for which the restrictions are imposed is expanded (to 9, 12 or 24 periods), the impact of a monetary policy shock lasts for a longer period (especially for residential investment). A comparison with the results obtained using the conventional Choleski decomposition reveals important differences. In the estimation with sign restrictions the response of the housing activity variable to a monetary policy shock is smaller and lasts for a shorter period than in the case with the Choleski decomposition. Moreover, there are cases under the sign restrictions for which we fail to find a significant response of the housing variable to the monetary policy shock. Once we become “agnostics” about the impact of monetary policy on the housing market, the results suggest that the impact is much less certain.