Recapping an article from the May 2002 issue |
|
|
of the Economic Policy Review, Volume 8, Number 1 | View full article | |
|
17 pages / 240 kb | |
Authors: Sydney Ludvigson, Charles Steindel, and |
Disclaimer | |
Index of executive summaries |
Overview The authors evaluate the importance of the so-called wealth channel in the transmission of monetary policy. More specifically, they explore the extent to which monetary policy movements—changes in the federal funds rate, the Federal Reserve's key policy instrument—affect consumer spending by altering the value of household assets. The authors find little evidence of a strong causal link from policy to consumption by way of asset values. They attribute the weakness of the wealth channel to the fact that unexpected changes in the federal funds rate have only a transitory effect on asset values; changes in wealth must be more permanent if they are to influence consumer spending substantially. Background The wealth channel of monetary policy is thought to operate in the following way: Monetary policy movements affect interest rates, which in turn affect the value of household assets such as stocks, bonds, and real estate. The link from policy to consumption is completed when households respond to changes in the value of their assets by altering their spending behavior. Although discussion of the wealth channel has declined as large shifts in monetary policy have become rare, economic trends during the last decade might suggest that this mode of transmission has been operative. Many analysts hold that the strong growth in consumer spending and the economy in the second half of the 1990s was attributable in large part to the stock market boom. If monetary policy movements helped to sustain the market boom by raising stock values, then policy may have contributed to the growth of consumption through the wealth channel. To determine whether the wealth channel is in fact an important means of policy transmission, Ludvigson, Steindel, and Lettau conduct a statistical analysis using three large-scale macroeconomic models and a small structural vector autoregression (VAR) model. Argument and Methodology With each of the three macroeconomic models, the authors simulate the impact of a l00-basis-point cut in the federal funds rate on real GDP growth and on the growth of consumer spending on durable goods and services. Initially, the authors allow the cut in the funds rate to affect the value of household stock holdings. They then repeat the simulation while constraining stock values to follow the path they were on prior to the rate cut. By comparing the two simulations, the authors can measure the contribution that policy-induced changes in stock values make to the growth of GDP and consumer spending. While the size of this contribution varies across the three models, in none of the models do the changes in stock values play a dominant role in the transmission of monetary policy to the real economy. The wealth channel emerges as even less consequential in the experiments the authors undertake using the structural VAR model. In these experiments, the authors track the dynamic response of consumption to an unexpected change, or "shock," in the federal funds rate. Once again, the first, or baseline, simulation of the model allows asset values to respond to the rate change; a second simulation "shuts down" the wealth channel by assuming that asset values are unchanged. In comparing the two simulations of the VAR model, the authors find that the response of consumption when the wealth channel is shut down differs little from its response under the baseline scenario (chart). This result suggests that the changes in asset values brought about by funds rate shocks have little impact on consumption. Thus, it appears that the wealth channel plays no more than a minor role in transmitting the effects of monetary policy to the consumer sector. The authors attribute the similarity of responses in the two simulations to the fact that the effect of a funds rate shock on wealth is quite transitory, typically dying out in less than two years. They point to other research (Lettau and Ludvigson 2001b) indicating that significant changes in consumer spending occur only in response to permanent changes in asset values. In a refinement of their experiment, the authors add commodity prices to their VAR model. The expanded model yields two significant findings: 1) an unexpected rise in the federal funds rate reduces consumption but has little discernible effect on household wealth, and 2) higher commodity prices (and higher consumer prices) substantially lower real asset values at the same time they prompt a quick rise in the federal funds rate (chart). Together, these findings lead the authors to propose a different relationship between policy, household wealth, and consumer spending than that presumed by the wealth channel. First, the direct effects of a federal funds rate change on consumption appear to be more important in transmitting monetary policy to the real economy than do its indirect effects through the medium of household wealth. Second, asset values do not appear to respond to changes in the federal funds rate per se, but rather to the same price pressures that prompt the monetary authorities to enact rate changes. Findings The authors find that the wealth channel contributes little to the transmission of monetary policy—largely, perhaps, because the effect of federal funds rate shocks on asset values is only transitory. The analysis also suggests that if asset values decline when the funds rate rises unexpectedly, they may be responding to the same inflationary pressures that motivated the rate increase, and not to the increase itself. |
|
|
|
|
|
Response of Nondurables and Services Consumption to a Federal Funds Rate Shock, Baseline Model | |
Source: Authors’ vector autoregressions using data described in the appendix.
Notes: The chart shows a twenty-quarter response of variables to a one-standard-deviation (81basis points) innovation in the federal funds rate. The vertical axis represents percent deviations of variables (basis-point deviations of the federal funds rate). The sample period is 1966:1 to 2000:3. |
|
|
|
Impulse Responses, Six-Variable Structural VAR with Commodity Prices Using Nondurables and Services ConsumptionSource: Authors’ vector autoregressions using data described in the appendix.
Notes: The chart shows the twenty-quarter response of variables to a one-standard-deviation (81basis points) innovation in the federal funds rate. The dashed lines represent one-standard-error bands. The sample period is 1966:1 to 2000:3. |
|
Commentary on article by Stephen P. Zeldes4 pages / 50 kb |
|
Disclaimer | |
The views expressed in this article are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. |
|