Boivin-Giannoni (2006) Has Monetary Policy Become More Effective?

Summary

Boivin and Giannoni ask why identified-VAR studies find that monetary-policy shocks have had a smaller effect on the U.S. economy since the early 1980s, and whether this means monetary policy has become less powerful. They combine a structural VAR with an estimated dynamic general-equilibrium (DSGE) model. First, an identified VAR estimated separately over the pre- and post-1980 periods confirms both the reduced effect of policy shocks and a stronger systematic response of policy to the economy in the later period. Second, they estimate a fully specified sticky-price model (with habit formation) by minimum-distance matching of the model and VAR impulse responses in each period. Third, counterfactual experiments swap the estimated monetary-policy rule and the private-sector parameters across periods to decompose the change. The main finding: changes in the systematic component of monetary policy — a more stabilizing rule after 1980 — largely account for the reduced effect of exogenous policy shocks, so there is little evidence that monetary policy has become less powerful. (Circulated in 2001–2002 as "Has Monetary Policy Become Less Powerful?"; published in The Review of Economics and Statistics 88(3): 445–462.)

Key Claims

• The stylized fact. Recursively identified VARs show that the dynamic responses of output, inflation, and the federal funds rate to an identified monetary-policy shock are smaller after 1980 than before.
• Systematic policy strengthened. The estimated policy reaction function responds more strongly to output and inflation in the post-1980 period — a more aggressive, stabilizing systematic rule.
• VAR + estimated DSGE. A small optimizing sticky-price model with habit formation is estimated by minimum-distance matching of its impulse responses to the VAR responses, separately for each subperiod, and reproduces the observed dynamics in both.
• Counterfactual decomposition. Using the structural estimates, the authors ask what the later-period responses would have been under the earlier policy rule (and vice versa), isolating the contribution of the policy rule versus the private-sector (structural) parameters.
• Main result. The shift in the systematic monetary-policy rule accounts for most of the reduced measured effect of exogenous interest-rate shocks; the monetary transmission mechanism itself is essentially unchanged. A more stabilizing rule leaves less scope for — and smaller measured effects of — policy surprises, not a weaker economy-to-policy linkage.

Concepts Introduced or Extended

• Monetary Policy Shocks
• Great Moderation
• Structural Identification

Entities Mentioned

• Jean Boivin
• Marc P. Giannoni

Quotes

"Changes in the systematic elements of monetary policy are consistent with a more stabilizing monetary policy in the post-1980 period and largely account for the reduced effect of unexpected exogenous interest rate shocks. Consequently, there is little evidence that monetary policy has become less powerful."

My Take

This is a careful "good policy, not good luck" argument for the Great Moderation, and its methodological move is what makes it stick: rather than read the smaller shock responses as weaker policy, it separates the exogenous shock from the systematic rule by estimating a structural model that can be re-simulated under counterfactual rules. The finding — that a more aggressive systematic response shrinks the measured impact of policy surprises while leaving the transmission mechanism intact — reframes the whole VAR fact. It complements the reduced-form time-variation evidence (Cogley–Sargent, Primiceri, Sims–Zha) with a structural interpretation, and it pairs naturally with the monetary-policy-shock identification literature: the same object those SVARs isolate (the exogenous shock) is precisely the one whose measured effect depends on the systematic rule around it.