Definition
Monetary neutrality is the proposition that a change in the nominal money supply has no permanent effect on real variables — output, employment, the real interest rate — and only raises the price level proportionately in the long run. Short-run non-neutrality (money temporarily affects real output due to sticky prices or wages) is broadly accepted in modern macroeconomics; long-run neutrality is more contested empirically.
Key Ideas
- Classical dichotomy: In classical models, real and nominal variables are determined independently; money is neutral in both the short and long run. New Keynesian models accept long-run neutrality but allow short-run non-neutrality via nominal rigidities (sticky prices, sticky wages).
- Super-neutrality: The stronger claim that the growth rate of money also has no real effects (e.g., no Mundell-Tobin inflation-investment channel). Disputed by endogenous growth models.
- Identification dependency of VAR evidence: Cholesky-identified vector autoregressions (VARs) find large, persistent negative output responses to contractionary shocks — suggesting substantial short-run non-neutrality. Uhlig (2004) argues this is driven by the implicit zero contemporaneous restriction on gross domestic product (GDP), not by the data.
- Agnostic evidence (Uhlig 2004): Under Sign Restriction Identification that leaves the output response unconstrained, the peak GDP response lies within ±0.2% with two-thirds posterior probability — consistent with near-neutrality. Monetary policy shocks explain only 5–10% of GDP forecast-error variance at a 5-year horizon (median) vs. ~50% under Cholesky.
- Long-run restriction identification: Blanchard-Quah (1989) and Fisher (2003) identify monetary policy shocks by imposing that they have no permanent output effect — making long-run neutrality an identification assumption rather than a testable hypothesis.
Why It Matters
Whether monetary policy is neutral determines the scope for monetary stabilization. If contractionary shocks cause persistent output losses, central banks face a real trade-off between inflation control and output stabilization. If near-neutrality holds, the case for activist monetary stabilization policy is weaker. The debate over neutrality is therefore directly tied to the choice of identification strategy in Structural VARs (SVARs) — Uhlig (2004) shows the data are consistent with both the conventional large-negative view and near-zero output effects.
Open Questions
- Identification dependency: The answer depends on the identification scheme. Cholesky finds large non-neutrality; sign restrictions find ambiguity; long-run restrictions impose neutrality by construction. No identification-free test exists.
- State dependence: Non-neutrality may be larger near the zero lower bound or during recessions, when nominal rigidities are more binding.
- Separating channels: Without cleanly separating monetary from technology and demand shocks, long-run neutrality is hard to test directly.
Related