Optimal Policy for Macro-Financial Stability Gianluca Benigno 1 Huigang Chen 2 Christopher Otrok 3 Alessandro Rebucci 4 Eric R. Young 5 1 LSE and Princeton 2 MarketShare Partners 3 University of Missouri and Federal Reserve Bank of St Louis 4 Johns Hopkins University and IDB 5 University of Virginia 2014 ASSA Meeting BCORY Optimal Policy 2014 ASSA Meeting 1 / 51
Introduction Motivation Global financial crisis proved very costly to resolve Great recession in the US Near-death experience in Europe Strong and volatile capital flows in and out of emerging economies BCORY Optimal Policy 2014 ASSA Meeting 2 / 51
Introduction Motivation Debate on the role of policy for financial stability Consensus before the crisis: intervene only during crises (e.g., Bailouts) Current view: intervene before crises (e.g., Macro-prudential policies) Key questions: When should policy makers intervene? Which policy tools should they use? BCORY Optimal Policy 2014 ASSA Meeting 3 / 51
Introduction This paper Develop a framework to study the optimal stabilization problem before AND during financial crises DSGE model with occasionally binding financial friction: the crisis event is endogenous and nested in a regular business cycle This requires the numerical solution of an optimal policy problem in which policy functions are not differentiable BCORY Optimal Policy 2014 ASSA Meeting 4 / 51
Introduction Related literature Literature on financial frictions Lorenzoni (2008), Bianchi (2011), Bianchi and Mendoza (2010), Jeanne and Korinek (2011) Chang, Cespedes and Velasco (2012), Benigno et al. (2012) Methodology: Klein, Krusell, and Rios-Rull (2009) BCORY Optimal Policy 2014 ASSA Meeting 5 / 51
Introduction Framework Focus on a simple model: Mendoza (2002) Small open economy Two-goods that are consumed and produced A liquidity constraint that limits consumers’ borrowing to a fraction of their total income BCORY Optimal Policy 2014 ASSA Meeting 6 / 51
Introduction Why is there scope for government intervention? There is a pecuniary externality when the constraint binds: Consumers do not take into account the effect of their choices on the price of collateral This affects their ability to borrow Which in turn affects the price of collateral And so and so forth ... Consumers and producers’ decisions can be affected by this externality even when the borrowing constraint is not binding BCORY Optimal Policy 2014 ASSA Meeting 7 / 51
Introduction Main messages Role and design of macroprudential policies depends on the effectiveness of crises management policy (interaction between ex ante and ex post policy interventions is crucial) When price support policies are costly or not effective, macroprudential becomes desirable (A new, intrinsic rationale for macroprudential policies) How credit is allocated matters as much as total size of credit flows BCORY Optimal Policy 2014 ASSA Meeting 8 / 51
Introduction Outline Simplified version of the model Key results Some evidence Conclusions BCORY Optimal Policy 2014 ASSA Meeting 9 / 51
Model Preferences Utility function: ∞ β t log ( C j ) U j ≡ E 0 ∑ � � , (1) t = 0 Consumption basket and price index: � ω � � 1 − ω � C T C N t t C t = (2) ω ω ( 1 − ω ) 1 − ω � 1 − ω � P N P t = . t with P T t = 1. BCORY Optimal Policy 2014 ASSA Meeting 10 / 51
Model Constraints Budget constraint: C T t + P N t C N t + B t + 1 = Y T t + P N t Y N t + ( 1 + r ) B t , (3) where B t + 1 denotes the bond holding at the end of period t , and 1 + r is a given world gross interest rate with β ( 1 + r ) < 1. International Borrowing constraint: B t + 1 � − 1 − φ � � Y T t + P N t Y N . (4) t φ Crisis occurs when constraint binds endogenously. BCORY Optimal Policy 2014 ASSA Meeting 11 / 51
Model Allocations Debt Tradable Consumption −0.6 2 −0.8 1.5 −1 −1.2 1 −1.4 −1.6 0.5 −1.8 −2 0 −1.1 −1 −0.9 −0.8 −0.7 −0.6 −1.1 −1 −0.9 −0.8 −0.7 −0.6 B(t) B(t) Nontradable Price 6 5 4 CE 3 SP UE 2 1 0 −1.1 −1 −0.9 −0.8 −0.7 −0.6 B(t) BCORY Optimal Policy 2014 ASSA Meeting 12 / 51
Model Government Policy Three policy instruments: Macroprudential tool (e.g., Capital control): τ B t Price support tool (e.g., Real exchange rate targeting): τ N t or τ T t Balanced budget: Non distortionary taxation T t = τ N t P N t C N or T t = τ B t B t + 1 t Distortionary taxation τ B t B t + 1 = τ N t P N t C N t Ramsey approach: maximizes agents’ utility subject to resource constraint, FOCs of competitive equilibrium, and government budget constraint conditional on policy tools available BCORY Optimal Policy 2014 ASSA Meeting 13 / 51
Key results Comparing different tools R1: Macroprudential (Capital control) with lump-sum transfers/taxes achieves SP (Korinek, 2010; Bianchi, 2011) R2: Price support (Real exchange rate) with lump-sum transfers/taxes achieves UE If costless , price support policy dominates macroprudential policy BCORY Optimal Policy 2014 ASSA Meeting 14 / 51
Key results Mechanism Relative price determination: � C T � = ( 1 − ω ) � � t P N 1 + τ N . (5) t t ω C N t When the constraint does not bind τ N t is neutral (Euler equation and resource constraint determines tradable consumption) When the constraint binds, τ N t can affect the price of collateral, and hence the consumption of tradable goods BCORY Optimal Policy 2014 ASSA Meeting 15 / 51
Key results Costly price support (Distortionary financing) Taxes 0.04 τ B t 0.03 τ N t 0.02 t ) t , τ N 0.01 ( τ B 0 −0.01 −0.02 −1 −0.95 −0.9 −0.85 −0.8 −0.75 −0.7 −0.65 −0.6 −0.55 −0.5 B t BCORY Optimal Policy 2014 ASSA Meeting 16 / 51
Key results Competitive Equilibrium (CE) and Optimal Policy (OP) Debt Tradable Consumption −0.6 1.2 −0.7 1 −0.8 0.8 −0.9 0.6 −1 0.4 −1.1 0.2 −1.1 −1 −0.9 −0.8 −0.7 −0.6 −1.1 −1 −0.9 −0.8 −0.7 −0.6 B(t) B(t) Nontradable Price 3 2.5 2 CE OP 1.5 1 0.5 −1.1 −1 −0.9 −0.8 −0.7 −0.6 B(t) BCORY Optimal Policy 2014 ASSA Meeting 17 / 51
Key results Welfare Gains and Crisis Probabilities Ergodic Averages Debt to Income Prob. of Crisis Welfare Gain CE − 29.2% 6.7% NA SP − 28.4% 1.2% 0.41% UE NA 0.0% 33.8% OP − 30.5% 4.9% 1.10% Welfare gains from OP are quite large The economy with OP borrows more than the CE and macroprudential policies remain desirable BCORY Optimal Policy 2014 ASSA Meeting 18 / 51
Key results Production Production economy: the same externality has effect on consumption and production choices. Logic of the results extend to production economy: effectiveness of ex-post policies determine optimal design of ex ante policies Price support policies tend to dominate macro prudential policies BCORY Optimal Policy 2014 ASSA Meeting 19 / 51
Some evidence What do countries do? Consider Brazil and Mexico before and after Lehmann’s collapse: they had balance sheet mismatches in the corporate sector and used unconventional policy tools before and after the crisis BCORY Optimal Policy 2014 ASSA Meeting 20 / 51
Some evidence Brazil and Mexico used a multiplicity of tools before and during crises Capital Controls Index Nominal Interest Foreign Reserves 230 Reserves (Billions of Dollars) 140 210 Nominal interest rate index (Sept 07=100) Mexico 120 190 100 170 150 80 130 60 110 Brazil 40 90 20 70 0 0 0.2 0.4 0.6 Brazil Mexico Outflows 2006 Inflows 2006 Brazil Mexico Inflows 2008 Outflows 2008 Reserve Requirements Real Interest Rate Real Exchange Rate 0.10 105 130 Reserve requirements indedx (Jan 07=100) 0.09 0.08 120 100 0.07 Real exchange rate (Jan 2007=100) 110 Real interest rate 0.06 95 100 0.05 0.04 90 90 0.03 80 0.02 85 0.01 70 0.00 80 60 Brazil Mexico Brazil Brazil Mexico BCORY Optimal Policy 2014 ASSA Meeting 21 / 51
Conclusions Caveats and areas for future research General results extends to cases when the constraint depends on asset prices or is forward looking Simple framework in which there is no policy trade off: with multiple distortions there is an intrinsic rationale for macroprudential policies even if price support is costless (e.g., price and financial stability) BCORY Optimal Policy 2014 ASSA Meeting 22 / 51
Conclusions Conclusions Study optimal stabilization policy in an environment in which financial crises are nested in regular cycles Role and design of macroprudential policies depends on the effectiveness of crisis management policies When price support policies are costly, there is an intrinsic rationale for macropurdential policies Where credit goes is as important as how much credit flows ... BCORY Optimal Policy 2014 ASSA Meeting 23 / 51
Conclusions THANK YOU BCORY Optimal Policy 2014 ASSA Meeting 24 / 51
Recommend
More recommend
