Remarks on Optimal Monetary Policy

IMF News, September 7, 2018

Source details

Canonical URL: Remarks on Optimal Monetary Policy

Other formats

Bibliographic details

Authors: Tobias Adrian
Published: September 7, 2018
DOI: https://doi.org/10.1093/restud/rdx063

Overview and three guiding questions

Speaker and context: By Tobias Adrian, September 7, 2018, Federal Reserve Bank of Boston.
Focus: three stylized facts since the 2008 global financial crisis and tentative answers to three questions:
- How should the conduct of monetary policy reflect the role of financial conditions?
- Should hysteresis feature in monetary policy?
- Does monetary policy affect r*?
Concluding emphasis: directions for monetary policy research going forward.

Financial conditions, Growth at Risk (GaR), and optimal policy

Empirical stylized facts (from Adrian, Grinberg, Liang, Malik (2018)):
- The elasticity of downside risk to GDP (measured by the 5 th quantile of GDP) depends on financial conditions:
- Significantly positive in the short run (up to two years / 8 quarters).
- Significantly negative in the medium run.
- Loose financial conditions forecast a boost to the median of GDP growth in the short run.
- When financial conditions are loose (measured by the 1st or 10 th percentile of FCIs), the term structure of GaR is downward sloping: low short-term downside risk but larger medium-term downside risk.
- When financial conditions are tight (bottom 10th percentile of FCIs), the term structure of GaR is upward sloping: large downside risk in the short term but less risky medium-term outlook.
Distributional implications:
- Conditional mean and conditional volatility of one-year-ahead GDP growth tend to be negatively correlated: easy financial conditions → low conditional volatility and high expected growth; adverse shocks increase volatility and lower expected growth → negatively skewed GDP growth distribution.
Optimal policy insights (from Adrian and Duarte (2018)):
- A standard New Keynesian model with endogenous risk (IS curve featuring endogenous volatility as a function of expected growth; standard Phillips curve) can reproduce these facts.
- The policymaker with a quadratic loss on output gap and inflation deviation obtains an optimal monetary policy rule that also depends on the level of financial conditions (FCI), in addition to output gap and inflation.
- Simulation results: the optimal rule produces a steady-state distribution of output with less negative skewness and lower GDP volatility compared with a standard Taylor rule.
- Policy interpretation: optimal policy balances effects on expected growth and on downside risk; calibrations show a rule that:
- Leans slightly against the buildup of risk when the economy is close to potential (relative to a standard Taylor rule).
- Aggressively loosens when the economy is contracting.
Empirical/practical linkage: these results relate to central banks’ practice of explicitly weighing balance-of-risk judgments in monetary decisions.

Hysteresis and monetary policy

Empirical stylized fact:
- Deep recessions tend to be followed by persistently low growth and revisions of trend productivity (hysteresis), as documented over time and across countries (e.g., Carrera and Saxena (2008, 2017); Blanchard, Cerutti, and Summers (2015)).
- Potential GDP tends to fall in a persistent manner when recessions occur.
Theoretical developments:
- Benigno and Fornaro (2018): a New Keynesian model with endogenous growth can generate hysteresis—multiple equilibria with a high-output/high-growth efficient equilibrium and a low-output/low-growth inefficient equilibrium.
- Garga and Singh (2018): in New Keynesian models with endogenous growth, optimal monetary policy is an inflation-targeting rule that also conditions on a hysteresis gap (distance of current trend growth from efficient trend growth), under both commitment and discretion, conditional on supply and demand shocks.
- The existence of a zero lower bound (ZLB) complicates policy and worsens the costs of hysteresis.
Policy implication: monetary policy frameworks may need to explicitly condition on hysteresis (trend-growth gaps) in addition to conventional output and inflation gaps.

Natural rate r*, efficient rate re, and financial frictions

Empirical observation:
- A measure of global r*, averaging estimates from Holston, Laubach, Williams (2017) across the US, UK, Euro Area, and Japan, fell from over 4% in the early 1960s to less than 1% in the most recent decade.
Conceptual distinction:
- r* (natural rate) is the real short-term interest rate consistent with inflation at target and full employment when sticky prices are the primary friction.
- re (efficient real rate) is the benchmark when additional frictions (e.g., financial frictions) are present; re is the real rate that would prevail in the efficient equilibrium taking those frictions into account.
Interaction with financial frictions:
- Financial frictions that interact with the supply side (demographics, trend productivity) may affect re. Example: endogenous pricing of risk may impact capital accumulation (productive or housing), thereby affecting re in extended models.
- The computation of r* (e.g., Holston, Laubach, Williams (2017)) does not account for financial frictions; it assumes sticky prices are the sole friction.
Policy interpretation: when policymakers lack instruments to address all distortions, monetary policy trades off distortions using available instruments; the wedge between actual and efficient outputs and between inflation and target determines how policy is used to approach efficient levels.

Going forward — research priorities and tentative conclusions

Three stylized facts monetary policy should consider jointly:

1. Financial frictions matter for the distribution of future GDP growth. 2. Recessions are associated with hysteresis of trend growth. 3. The natural rate of interest has fallen markedly.

Research gaps and promising directions:
- The literature has studied these stylized facts largely in isolation; joint analysis is needed.
- A promising avenue is to introduce financial frictions into New Keynesian models with endogenous growth to analyze interactions between endogenous macroeconomic risk and hysteresis, and implications for the evolution of the efficient rate re.
- Borio, Disyatat, Rungcharoenkitkul (2018) model leverage cycles alongside sticky prices and argue leverage cycles lead to a novel estimate of the natural rate that accounts for financial cycle imbalances; this work begins to jointly model hysteresis and financial frictions but more research is needed to draw robust conclusions for optimal monetary policy.

Remarks on Optimal Monetary Policy — By Tobias Adrian, September 7, 2018, Federal Reserve Bank of Boston