Economic Policy Review Executive Summary

Assessing Changes in the Monetary Transmission Mechanism:
A VAR Approach

Recapping an article from the May 2002 issue
View full article PDF
of the Economic Policy Review, Volume 8, Number 1 15 pages / 185 kb

Disclaimer

Authors: Jean Boivin and Marc Giannoni

Index of executive summaries

Overview
The authors argue that a change in monetary policy transmission explains much of the increased stability of output and inflation since the beginning of the 1980s. Their statistical analysis shows that monetary policy "shocks"—unexpected movements in the federal funds rate, the Federal Reserve's key policy instrument—have a weaker effect on output and inflation than in the past. While this finding could be interpreted to mean that monetary policy is now less potent than in the past, it could also indicate that monetary policy is now aimed more directly at minimizing the variability of inflation and output.

Background
In the past twenty years, variability in quarterly output growth has fallen 30percent, while variability in inflation has fallen more than 40percent. Researchers have advanced different explanations for the increased stability of the economy. Some have argued that shocks or disturbances in the economy—monetary policy shocks, for example, or shocks to productivity, fiscal policy, and foreign economies—have become smaller and less frequent, while others contend that the transmission of these shocks to the economy has changed. Boivin and Giannoni begin their analysis by testing these alternative explanations.

Argument and Methodology
The authors use a reduced-form vector autoregression (VAR) model to determine whether the decreased variability of output and inflation since the beginning of the 1980s is attributable primarily to smaller and less frequent shocks or to a change in the transmission, or propagation, of these shocks. They find that both factors have played a stabilizing role. Most notably, a change in the propagation mechanism accounts for roughly 40percent of the decreased variance in output and 60percent of the decreased variance in inflation.

Since the authors find that the propagation of shocks has changed significantly, they then investigate whether there has been a change in particular in the transmission of monetary policy shocks to the rest of the economy. Using "impulse response" analysis to track the economy's response to unexpected federal funds rate movements in three periods (1963-79, 1980-97, and 1984-97), Boivin and Giannoni determine that both output and inflation reacted to interest rate fluctuations in a much less pronounced and persistent way after 1980 than they did in the preceding period (chart).

This diminished response, the authors note, can be interpreted in different ways. It may mean that the functioning of the economy has changed in a way that insulates output and inflation from monetary policy movements. If so, then it would appear that monetary policy is no longer as effective as it once was in influencing the economy. Alternatively, the authors' finding could mean that the conduct of monetary policy has changed. Specifically, the monetary authorities may now be responding more systematically to economic conditions with the objective of minimizing the variability of inflation and output.

To evaluate these competing interpretations, Boivin and Giannoni undertake some counterfactual experiments that allow them to isolate the individual contributions of monetary policy and "the rest of the economy" to the observed change in the transmission process. They find that a change in the systematic behavior of monetary policy appears to account for much of the diminished response of output and inflation to federal funds rate movements in the post-1980 and post-1984 samples (chart).

Findings
Boivin and Giannoni conclude that the increased stability of the U.S. economy in the last two decades stems in large part from a change in the propagation of shocks. This change in turn is very closely linked to the reduced effect of monetary policy shocks on output and inflation. The authors offer some evidence that a change in the conduct of monetary policy accounts for the reduced effect of the monetary policy shocks, but the limitations of their model make it difficult to show a strong causal relationship.

In other recent work, however, the authors address similar issues using a sophisticated structural model. In Boivin and Giannoni 2002, they are able to demonstrate that the emergence, in the 1980s, of a monetary policy geared more directly toward stabilizing output and inflation explains most of the economy's diminished response to monetary policy shocks. This finding leads the authors to reject the notion that U.S. monetary policy has lost its effectiveness in the last two decades.


An unexpected 1-percentage-point increase in the federal funds rate elicits a much weaker response from output and inflation after 1980 than before.

Impulse Responses to a Monetary Shock over Different Samples

Chart 1 - Impulse Responses to a Monetary Shock over Different Samples

Source: Authors' calculations.


The chart below tracks the response of output and inflation to a 1-percentage-point federal funds rate increase in the pre-1980 period, in the post-1984 period, and in two counterfactual scenarios. Comparing the pre-1980 period (black dashed line) with a counterfactual scenario in which monetary policy shifts to its post-1984 behavior while the "rest of the economy" is held to its 1980 values (solid blue line), we can see that the response of output and inflation becomes much less variable with the change in monetary policy. Moreover, the close correspondence between this counterfactual scenario and the post-1984 period (blue dashed line) suggests that the change in policy explains an important part of the output and inflation response in the later period.

VAR-Based Counterfactual Analysis:
Pre-1980 Sample versus Post-1984 Sample

Chart 3 - VAR-Based Counterfactual Analysis:  Pre-1980 Sample versus Post-1984 Sample

Source: Authors’ calculations.

Commentary on article by Mark W. WatsonPDF4 pages / 58 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.