Output Gap, Monetary Policy Trade-Offs and Financial Frictions Francesco Furlanetto Norges Bank Paolo Gelain Norges Bank Marzie Taheri Sanjani International Monetary Fund Seminar at Narodowy Bank Polski - 3 May 214 Usual disclaimer applies Furlanetto-Gelain-Taheri () FGT (214) 1 / 37
Motivation Large interest in macroeconomic models with financial frictions Propagation of standard disturbances (Bernanke and Gertler, 1989, Bernanke, Gertler and Gilchrist, 1999) Financial shocks (Christiano, Motto and Rostagno, 214, Jermann and Quadrini, 212) Banking in macroeconomic models (Gerali et al., 21, Gertler and Karadi, 211, Gertler and Kiyotaki, 211) Unconventional monetary policies (Gertler and Karadi, 211) Macroprudential policies (Angelini et al., 211, Quint and Rabanal, 213, Gelain and Ilbas, 214) Much less attention on the policy implications for standard monetary policy objectives Output gap Optimal monetary policy (trade-offs between different policy objectives) Furlanetto-Gelain-Taheri () FGT (214) 2 / 37
What do we do? We estimate a standard DSGE model with financial frictions (BGG, 1999, CMR, 214) over the period 1964-29 using U.S. data Our contribution: we analyze the implications of financial frictions (financial accelerator) and financial shocks (stock market shocks and spread shocks) for Output gap No papers in the DSGE literature. Optimal monetary policy and trade-offs Other papers use calibrated models driven by few shocks We use an estimated model driven by several shocks Furlanetto-Gelain-Taheri () FGT (214) 3 / 37
Why is this interesting I? Output gap. Borio, Disyatat and Juselius (213): financial factors are important for the output gap "Finance neutral" vs. "inflation neutral" output gap Claim: large and positive output gap in the pre-great Recession period Furlanetto-Gelain-Taheri () FGT (214) 4 / 37
Why is this interesting II? Trade-offs for monetary policy Central Banks have dual mandate: real economy and inflation stabilization Underlying assumption: trade-off between the two objectives (loss function approach) With financial frictions: more intermediate targets to trade-off price inflation wage inflation output gap spread (external finance premium) Furlanetto-Gelain-Taheri () FGT (214) 5 / 37
What do we find? Output gap The presence of financial frictions and financial shocks change output gap dynamics Positive output gap in the pre-crisis period Monetary policy trade-offs Estimated model with a Taylor rule: large trade-offs between different objectives Counterfactual with optimal Ramsey monetary policy It is optimal to stabilize price and wage inflation at the cost of some fluctuations in the output gap...and somewhat large fluctuations in the premium Furlanetto-Gelain-Taheri () FGT (214) 6 / 37
Plan of the rest of the talk Overview of the model Results: output gap Results: monetary policy trade-offs Conclusions Furlanetto-Gelain-Taheri () FGT (214) 7 / 37
The model: structure 1 Standard New Keynesian DSGE model with nominal and real rigidities (Christiano, Eichenbaum and Evans, 25, and Smets and Wouters, 27) 2 Financial accelerator mechanism (Bernanke, Gertler and Gilchrist, 1999, Del Negro and Schorfheide, 213, CMR, 214) 3 Financial shocks: stock market (net worth) shocks and spread (risk) shocks 4 Our model perfectly "nests" the model by Justiniano, Primiceri and Tambalotti, 213. Furlanetto-Gelain-Taheri () FGT (214) 8 / 37
The model: entrepreneurs Entrepreneurs balance sheet B t = Q t K t N t They have some internal funds N t to buy capital Q t K t but need to borrow the rest B t Borrowing subject to frictions (imperfect information and monitoring costs): External finance is costly S t = E tr k t+1 R t = f ( ) Nt, σ t Q t K t Spread shocks σ t as in Christiano, Motto, and Rostagno (214) Stock market shocks ϑ t as in Gilchrist and Leahy (22) [ N t = ϑ t {R t k Q t 1 K t 1 R t 1 + M t Q t 1 K t 1 N t 1 ] } (Q t 1 K t 1 N t 1 ) Furlanetto-Gelain-Taheri () FGT (214) 9 / 37
Estimation results in a nutshell Data: quarterly US data from 1964 QII to 29 QIV in log differences Stock market index (in growth rate) External finance premium (baa - 1y yield on government bonds) Credit-to-GDP gap Financial accelerator estimated parameters similar to the values estimated by Del Negro and Schorfheide (213) and CMR (214) Financial shocks and macroeconomic fluctuations Relatively important at business cycle frequency (absorb some explanatory power from investment shocks) Very important to explain low-frequency dynamics (absorb some explanatory power from labour supply shocks) Furlanetto-Gelain-Taheri () FGT (214) 1 / 37
Business Cycle vs. Credit Cycle Business and Credit Cycles 3 2 Business Cycle Credit Cycle NBER Recessions 1 1 2 3 4 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 11 / 37
Selected IRFs Investment Shock (M.E.I.).8 3 Networth Shock Spread Shock.6 2.5.5.4 2 1.2 1.5 1.5.2 1.5 2.4 2.5.6.5 3.8 1 3.5 1 2 3 4 1 2 3 4 1 2 3 4 Net Worth Output Credit Furlanetto-Gelain-Taheri () FGT (214) 12 / 37
Acceleration or deceleration.2 Monetary Policy Output Gap Response to shocks Technology.2.2.1 Government.4.4 1 2 3 4 1 2 3 4 1 2 3 4.4 M.E.I..1 Price mk Labor Supply.2.4.1 1 2 3 4 1 2 3 4.2 1 2 3 4 1 Preference Wage mk 1 Inflation Target.5.2.5.4 1 2 3 4 1 2 3 4 JPT Baseline 1 2 3 4 Furlanetto-Gelain-Taheri () FGT (214) 13 / 37
Sources of ineffi ciencies in a model Output is ineffi ciently low in this economy Static Distortions Monopolistic competition in goods and labor market Spread in steady state Dynamic distortions Price and wage stickiness Financial frictions: wedge between expected return to capital and the risk-free rate Furlanetto-Gelain-Taheri () FGT (214) 14 / 37
Output gap...gap from what? In standard NK models common to use actual output in deviation from potential output (second best) Potential output in standard NK models: counterfactual level of output when prices and wages are flexible ineffi cient shocks are turned off (price and wage mark up shocks) Why not effi cient output (first best)? Monetary policy is neutral in the long-run and cannot reduce the static distortion Dynamics: potential output follows effi cient output up to a level difference Furlanetto-Gelain-Taheri () FGT (214) 15 / 37
Output gap: Effi cient vs potential in JPT (213) Furlanetto-Gelain-Taheri () FGT (214) 16 / 37
Potential output with financial frictions Counterfactual level of output that emerges when flexible prices and wages no ineffi cient shocks price mark-up, wage-mark-up, stock market and spread shocks no dynamic distortion due to financial frictions New in this paper: we turn off the financial accelerator and the financial shocks Following JPT (213) the gap is in deviation from potential Furlanetto-Gelain-Taheri () FGT (214) 17 / 37
Output gap in deviation from effi cient and potential Potential vs. Efficient Gap 5 5 1 Actual Potential Actual Efficient 15 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 18 / 37
Actual and potential GDP Actual and Potential GDP 6.3 Log GDP Per Capita Log Potential GDP (JPT) Log Potential GDP (Baseline) NBER Recessions 6.2 6.1 6 5.9 5.8 5.7 5.6 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 19 / 37
Comparison with JPT (213) 5 5 1 Output Gap (JPT) Output Gap (Baseline) NBER Recessions 15 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 2 / 37
What explains the difference? In JPT hours worked unique source of low-frequency movements In JPT labour supply shocks explain 7% out low-frequency output fluctuations In our case extra source of low-frequency movements (i.e. credit-to-gdp gap) Labour supply shocks become irrelevant in favour of financial shocks (risk 54%, net worth 5%) Furlanetto-Gelain-Taheri () FGT (214) 21 / 37
Comparison: Gali, Smets and Wouters (211) Furlanetto-Gelain-Taheri () FGT (214) 22 / 37
Comparison: Christiano, Trabandt and Walentin (211) Furlanetto-Gelain-Taheri () FGT (214) 23 / 37
Financial factors and output gap Borio, Disyatat and Juselius (213): the financial cycle affects the output gap We confirm their intuition in a DSGE set-up Financial shocks and financial frictions imply a lower potential output in the pre-crisis period Important differences due to dominance of financial shocks in explaining low-frequency movements of output Furlanetto-Gelain-Taheri () FGT (214) 24 / 37
Trade-offs in NK models Small scale NK models: Divine Coincidence (Blanchard and Galí, 27) There is no trade-off between output gap stabilization and inflation stabilization Exception 1: Cost push shocks Exception 2: Real wage rigidities Medium-scale NK model (Smets and Wouters, 27) Capital accumulation and habit persistence all shocks generate trade-offs Furlanetto-Gelain-Taheri () FGT (214) 25 / 37
Trade-offs in NK models Justiniano, Primiceri and Tambalotti (213): Trade-offs are present Optimal monetary policy: trade-offs are small price inflation, wage inflation and the output gap can be roughly stabilized at the same time Contribution: does "trinity" hold also in a New Keynesian model with financial frictions where trade-offs are more complicated? Furlanetto-Gelain-Taheri () FGT (214) 26 / 37
Why trinity in JPT (213)? Main reason: wage mark-up shocks are not a fundamental driver in macroeconomic fluctuations at high frequency: measurement error at low frequency: labor supply shocks Additional reason: Price inflation is largely explained by inflation targeting shocks (absent under optimal policy) Furlanetto-Gelain-Taheri () FGT (214) 27 / 37
Optimal monetary policy with financial frictions Several previous contributions: Faia and Monacelli (27), Curdia and Woodford (29), De Fiore and Tristani (29),,Carlstrom, Fuerst, and Paustian (29), Fendoglu (213), Huang and Davis (213), Kolasa and Lombardo (214) calibrated models driven by few shocks Contribution: use an estimated model driven by several shocks Furlanetto-Gelain-Taheri () FGT (214) 28 / 37
Monetary policy trade-offs and financial frictions 5 5 1 (a): GDP Gaps Optimal Potential Actual Potential 1965 197 1975 198 1985 199 1995 2 25 21 (b): Price Inflation 3 (c): Wage Inflation 2 1 2 1 197 198 199 2 (d): External Finance Premium 197 198 199 2.5.5 1965 197 1975 198 1985 199 1995 2 25 Furlanetto-Gelain-Taheri () FGT (214) 29 / 37
Fluctuations that could have been avoided 5 5 1 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 3 / 37
Price mk M.E.I. Government Technology Optimal monetary policy and financial frictions.2.4.6.4.2.3.2.1.1.5.5 Output Gap.5.1.15 1 2 3 4 1 2 3 4 15 1 5 15 1 2 3 4 1 2 3 4 5 1.15.1.5 Infla tio n 1 2 3 4 x 1 3 1 2 3 4 x 1 3 1 2 3 4 1 2 3 4 Taylor.6.4.2 8 6 4 2 Wage Inflation 1 2 3 4 x 1 3 x 1 3 2 4 6 x 1 3 1 5 1 2 3 4 1 2 3 4 1 2 3 4 Ramsey.5.5.1.5.2.1 1 5 5 Spread 1 2 3 4 1 2 3 4 1 2 3 4 x 1 3 1 2 3 4 Furlanetto-Gelain-Taheri () FGT (214) 31 / 37
I nflation target Wage mk Preference Labour s upply Optimal monetary policy and financial frictions.2.2.4 Output Gap 1 2 3 4 x 1 Infla 3 tio n 1 5 5 1 2 3 4 15 x Wage Inflation 1 3 1 5 1 2 3 4.6.4.2.2 Spread 1 2 3 4 1.5.2.4.4.2 1 2 3 4 1 2 3 4 1 2 3 4.4.2 15 1 5.1.5 1 2 3 4 x 1 3 1 2 3 4 1 2 3 4 Taylor.3.2.1.1.5.1.5 1 2 3 4 Ramsey.2.1 1 2 3 4.2.2.4.6.8 1 2 3 4 1 2 3 4 1 2 3 4 5 1 x 1 3 1 2 3 4 Furlanetto-Gelain-Taheri () FGT (214) 32 / 37
Net worth Risk Optimal monetary policy and financial frictions.5 1.4.2 Output Gap.2.4 1 2 3 4.1.1 1 2 3 4 Inflation.2.4 1 2 3 4 1 2 3 4 Wage Inflation x 1 3 1 5 5 1 2 3 4 1 2 3 4.1.5.5.5 Spread 1 2 3 4 1 2 3 4 Taylor Ramsey Furlanetto-Gelain-Taheri () FGT (214) 33 / 37
Conclusion We have looked at the policy implications of an estimated DSGE model with financial frictions Financial frictions and financial shocks matter for the output gap Positive output gap in the pre-great Recession period Trade-offs between different monetary policy objectives are substantial......but optimal monetary policy is very effective in stabilizing nominal variables at the cost of some fluctuations in the output gap and somewhat large fluctuations in the premium Furlanetto-Gelain-Taheri () FGT (214) 34 / 37
Robustness - Standard financial observables and labour supply shocks on 1 8 6 Low Frequency Movements Driven by Labour Supply Shocks Output Gap (JPT) Output Gap (JPT+CMR) NBER Recessions 4 2 2 4 6 8 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 35 / 37
Robustness - Standard financial observables and labour supply shocks off 12 1 Low Frequency Movements Driven by Wage Mark up Shocks Output Gap (JPT) Output Gap (JPT+CMR) NBER Recessions 8 6 4 2 2 4 6 8 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 36 / 37
Robustness - Alternative series of net worth in level 1 5 5 1 1965 197 1975 198 1985 199 1995 2 25 21 Furlanetto-Gelain-Taheri () FGT (214) 37 / 37