Ben Bernanke's Speech to the Kansas City Fed
The Monetary Transmission Mechanism Since the Mid-1980s
The dramatic changes in mortgage finance that I have described appear to have significantly affected the role of housing in the monetary transmission mechanism. Importantly, the easing of some traditional institutional and regulatory frictions seems to have reduced the sensitivity of residential construction to monetary policy, so that housing is no longer so central to monetary transmission as it was.In particular, in the absence of Reg Q ceilings on deposit rates and with a much-reduced role for deposits as a source of housing finance, the availability of mortgage credit today is generally less dependent on conditions in short-term money markets, where the central bank operates most directly.
Most estimates suggest that, because of the reduced sensitivity of housing to short-term interest rates, the response of the economy to a given change in the federal funds rate is modestly smaller and more balanced across sectors than in the past.9 These results are embodied in the Federal Reserve's large econometric model of the economy, which implies that only about 14 percent of the overall response of output to monetary policy is now attributable to movements in residential investment, in contrast to the model's estimate of 25 percent or so under what I have called the New Deal system.
The econometric findings seem consistent with the reduced synchronization of the housing cycle and the business cycle during the present decade. In all but one recession during the period from 1960 to 1999, declines in residential investment accounted for at least 40 percent of the decline in overall real GDP, and the sole exception--the 1970 recession--was preceded by a substantial decline in housing activity before the official start of the downturn. In contrast, residential investment boosted overall real GDP growth during the 2001 recession. More recently, the sharp slowdown in housing has been accompanied, at least thus far, by relatively good performance in other sectors. That said, the current episode demonstrates that pronounced housing cycles are not a thing of the past.
My discussion so far has focused primarily on the role of variations in housing finance and residential construction in monetary transmission. But, of course, housing may have indirect effects on economic activity, most notably by influencing consumer spending. With regard to household consumption, perhaps the most significant effect of recent developments in mortgage finance is that home equity, which was once a highly illiquid asset, has become instead quite liquid, the result of the development of home equity lines of credit and the relatively low cost of cash-out refinancing. Economic theory suggests that the greater liquidity of home equity should allow households to better smooth consumption over time. This smoothing in turn should reduce the dependence of their spending on current income, which, by limiting the power of conventional multiplier effects, should tend to increase macroeconomic stability and reduce the effects of a given change in the short-term interest rate. These inferences are supported by some empirical evidence.
On the other hand, the increased liquidity of home equity may lead consumer spending to respond more than in past years to changes in the values of their homes; some evidence does suggest that the correlation of consumption and house prices is higher in countries, like the United States, that have more sophisticated mortgage markets (Calza, Monacelli, and Stracca, 2007). Whether the development of home equity loans and easier mortgage refinancing has increased the magnitude of the real estate wealth effect--and if so, by how much--is a much-debated question that I will leave to another occasion.
I hope this exploration of the history of housing finance has persuaded you that institutional factors can matter quite a bit in determining the influence of monetary policy on housing and the role of housing in the business cycle. Certainly, recent developments have added yet further evidence in support of that proposition. The interaction of housing, housing finance, and economic activity has for years been of central importance for understanding the behavior of the economy, and it will continue to be central to our thinking as we try to anticipate economic and financial developments.
In closing, I would like to express my particular appreciation for an individual who I count as a friend, as I know many of you do: Edward Gramlich. Ned was scheduled to be on the program but his illness prevented him from making the trip. As many of you know, Ned has been a research leader in the topics we are discussing this weekend, and he has just finished a very interesting book on subprime mortgage markets. We will miss not only Ned's insights over the course of this conference but his warmth and wit as well. Ned and his wife Ruth will be in the thoughts of all of us."