Banque nationale de Belgique SA · EURONEXT:BNB
Banque nationale de Belgique SA — Technical Presentation (September 2015)
56 slides from Banque nationale de Belgique SA's technical presentation, September 2015
Slide 1: Clearing Up the Fiscal Multiplier Morass
Clearing Up the Fiscal Multiplier Morass Eric M. Leeper, Nora Traum & Todd B. Walker Indiana University, NCSU & Indiana University National Bank of Belgium, September 2015
Slide 2: The Morass
The Morass I Coenen et al. (2012) . . . a robust finding across all models that fiscal policy can have sizeable output multipliers. I Cogan et al. (2010) . . . multipliers are less than one.. . . The impact in the first year is very small. And as the government purchases decline . . . the multiplier
Slide 3: Objective
Objective I To clear up the morass by I bringing more fiscal detail into Smets-Wouter style model I permitting government spending to be useful I considering alternative monetary/fiscal regimes I Regime M: active monetary & passive fiscal policies I Regime F: passive monetary & active fiscal policie
Slide 4: What We Find
What We Find I Posterior estimates for U.S. data 1955q1–2014q2 1. High degrees of nominal rigidities & habit formation 2. Government spending complements private consumption 3. Impact output multipliers ≈ 1.4 in both regimes 4. Long-run output multipliers: 0.7 in M; 1.9 in F (in present value) 5. Co
Slide 5: Prior Predictive Analysis
Prior Predictive Analysis I Key model features 1. Labor & capital 2. External habits over total consumption—private plus government I G may complement or substitute for C 3. “Saver” and “non-saver” households 4. Monopolistic competition in goods & labor sectors 5. Price & wage rigidities; real rigid
Slide 6: Prior Predictive Technique
Prior Predictive Technique I Four steps 1. Given DSGE, Aj , parameters, θAj , posit prior p(θAj |Aj ) 2. Linearize DSGEs to obtain {Aj }; generate ex ante predictive densities for observables, yT , from p(yT |Aj ) = ΘA p(θAj |Aj )p(yT |θAj Aj )dθAj R j 3. Specify vector of interest, ω, a function of
Slide 7: Multiplier Definition
Multiplier Definition I We compute present-value multipliers using model-generated impulse responses Pk Q k −1 Et j=0 i=0 (1 + rt+i ) ∆Yt+j Present Value Multiplier(k) = Pk Q k −1 Et j=0 i=0 (1 + rt+i ) ∆Gt+j rt : model-implied real interest rate Gt : change in government spending—exogenous & endoge
Slide 8: Prior Predictive Questions
Prior Predictive Questions I Use the prior predictive analysis to answer 1. Given a model, what is Prob P V ∆Y ∆G >1 ? 2. Given a model, what range of possible multipliers fall within a 90 percent probability band? I Model definition includes policy regime I Calibrate some parameters I β, α, δ, π̄,
Slide 9: Policy Regimes
Policy Regimes Active monetary/passive fiscal policy I Central bank aggressively targets inflation I Fiscal authority adjusts taxes & expenditures to stabilize debt I The conventional policy regime, though fiscal behavior is usually implicit Passive monetary/active fiscal policy I Fiscal authority s
Slide 10: How Likely Are Large Multipliers?
How Likely Are Large Multipliers? Prob P V ∆Y ∆G >1 Impact 10 qtrs 10 years Model 1: RBC Real Frictions 0.00 0.00 0.00 Model 2: New Keynesian Sticky Prices & Wages Regime M 0.12 0.00 0.01 Regime F, short debt 1.00 0.94 0.92 Regime F, long debt 0.96 0.68 0.68 Model 3: New Keynesian Nonsavers Regime M
Slide 11: How Big Are Output Multipliers in Regime M?
How Big Are Output Multipliers in Regime M? 2 1.5 1 0.5 0 −0.5 0 10 20 30 40 Government spending in utility unrestricted, steady-state taxes only, long debt (black lines) [mean & 90% credible set]
Slide 12: How Big Are Output Multipliers in Regime M?
How Big Are Output Multipliers in Regime M? 2 1.5 1 0.5 0 −0.5 0 10 20 30 40 Rule-of-thumb agents, everything responds to debt, long debt (red lines) [mean & 90% credible set]
Slide 13: How Big Are Output Multipliers in Regime M?
How Big Are Output Multipliers in Regime M? 2 1.5 1 0.5 0 −0.5 0 10 20 30 40 Rule-of-thumb agents, steady-state taxes only, long debt (blue lines) [mean & 90% credible set]
Slide 14: How Big Are Output Multipliers in Regime F?
How Big Are Output Multipliers in Regime F? 2.5 2 1.5 1 0.5 0 0 10 20 30 40 Government spending in utility unrestricted, steady-state taxes only, long debt (black lines) [mean & 90% credible set]
Slide 15: How Big Are Output Multipliers in Regime F?
How Big Are Output Multipliers in Regime F? 2.5 2 1.5 1 0.5 0 0 10 20 30 40 Rule-of-thumb agents, everything responds to debt, long debt (red lines) [mean & 90% credible set]
Slide 16: How Big Are Output Multipliers in Regime F?
How Big Are Output Multipliers in Regime F? 2.5 2 1.5 1 0.5 0 0 10 20 30 40 Rule-of-thumb agents, steady-state taxes only, long debt (blue lines) [mean & 90% credible set]
Slide 17: What We Learn From the Prior Predictive
What We Learn From the Prior Predictive 1. In M, rule-of-thumb agents deliver tighter prior distributions for multipliers 2. In M, when all instruments stabilize debt, multipliers are uniformly smaller than when only steady-state taxes exist and transfers and spending stabilize 3. In F, rule-of-thum
Slide 18: Estimation
Estimation I Estimate model suggested by prior predictive analysis I Eight U.S. observables, quarterly 1955q1–2014q2 1. Consumption (growth rate) 2. Investment (growth rate) 3. Federal government consumption (growth rate) 4. Real wages (growth rate) 5. Real market value of federal government debt (g
Slide 19: Priors for Estimation
Priors for Estimation I Including government consumption & debt in observables. . . I more spectral power at low frequencies I greater persistence in data compared to typical monetary policy models I Some priors more diffuse than in much literature I data want high values for stickiness & habits I P
Slide 20: Priors
Priors Habits Wage Stickiness Price Stickiness 4 4 4 Smets-Wouters Leeper-Traum-Walker 2 2 2 0 0 0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 MP to π G to B Z to B 2 4 4 F M 1 2 2 0 0 0 0 0.5 1 1.5 2 2.5 3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 -0.2 -0.1 0 0.1 0.2 0.3 0.4 G in Utility 0.3 0.2
Slide 21: Posterior Estimates of Critical Parameters
Posterior Estimates of Critical Parameters 1955q1:2014q2 Parameter Regime M Regime F mean 90% C.S. mean 90% C.S. Preference and HHs θ, habit formation 1.00 [0.99, 1.00] 0.99 [0.98, 1.00] αG , G in utility -0.24 [-0.41, -0.07] -0.20 [-0.38, -0.02] Frictions & Production ωp , price stickiness 0.92 [0.
Slide 22: Phillips Curves Are Flat
Phillips Curves Are Flat 1955q1:2014q2 Parameter Regime M Regime F mean 90% C.S. mean 90% C.S. Inflation Phillips curve slope 0.0071 [0.0036, 0.0104] 0.0028 [0.0016, 0.0039] Wage Phillips curve slope 0.0005 [0.0002, 0.0008] 0.0013 [0.0006, 0.0019] I Important elements in transmission mechanism of G
Slide 23: Some Model Comparisons
Some Model Comparisons Log Marginal Data Density 1955q1 1955q1 1982q1 –2014q2 –1979q4 –2007q4 Regime M -2557 -1122 -956 Regime F -2549 -1125 -968 I No strong evidence in favor of one regime over other I Weak preference for F in 1955q1–2014q2 I Weak preference for M in 1982q1–2007q4 I Will study mult
Slide 24: Data Are Informative About Multipliers in M
Data Are Informative About Multipliers in M Output Multiplier in Regime M: P V ∆Y ∆G Impact 10 qtrs 10 years Prior 0.80 0.56 0.49 [-0.58,2.12] [-0.38,1.55] [-0.41,1.53] Posterior 1955q1–2014q2 1.36 0.90 0.70 [1.17,1.55] [0.72,1.09] [0.45,0.94] 1955q1–1979q4 1.41 0.67 0.36 [1.15,1.68] [0.50,0.83] [0.
Slide 25: Data Are Informative About Multipliers in F
Data Are Informative About Multipliers in F Output Multiplier in F: P V ∆Y ∆G Impact 10 qtrs 10 years Prior 1.34 1.17 1.31 [-0.13,2.75] [0.22,2.12] [0.30,2.23] Posterior 1955q1–2014q2 1.51 1.58 1.87 [1.33,1.70] [1.39,1.76] [1.65,2.09] 1955q1–1979q4 1.42 1.13 1.33 [1.15,1.69] [0.91,1.34] [1.13,1.53]
Slide 26: Dynamic Responses Across Regimes
Dynamic Responses Across Regimes Output Multiplier Consumption Multiplier Investment Multiplier 1.5 0.4 -0.5 1 -1 0.5 0.2 -1.5 0 0 -2 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 8 2 6 0 5 4 -2 2 0 0 -4 10 20 30 40 50 6
Slide 27: Dynamic Responses Across Regimes
Dynamic Responses Across Regimes Output Multiplier Consumption Multiplier Investment Multiplier 0.6 2 0 0.4 1 -1 0.2 0 0 -2 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 15 2 10 10 0 5 5 -2 0 0 -4 10 20 30 40 50 60 70 80 10
Slide 28: Transmission Mechanism in Both Regimes
Transmission Mechanism in Both Regimes I Labor market I labor rises—more weakly in M than in F I real wages—flat in M; rise strongly in F I strong increase in demand in F; less so in M I Complementarity of G with C I produces persistently higher C in both regimes I higher C strongly crowds out I in
Slide 29: Long-Run Fiscal Dynamics Across Regimes
Long-Run Fiscal Dynamics Across Regimes Market Value Debt/Output 5 4 3 2 1 0 100 200 300 400 500 600 700 800 900 1000 Primary Surplus 5 0 -5 100 200 300 400 500 600 700 800 900 1000 Regime M & Regime F Spending reversals & fiscal overshooting in regime M
Slide 30: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 1 0.2 -0.5 0.5 0.1 -1 0 0 -1.5 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 5 5 0 -2 0 0 10 20 30 40 50 60 70 8
Slide 31: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -0.5 0 0.5 -1 -0.2 0 -1.5 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 5 5 0 -2 0 0 10 20 30 40 50 60 70
Slide 32: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -0.5 0 0.5 -1 -0.2 0 -1.5 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 5 5 0 -2 0 0 10 20 30 40 50 60 70
Slide 33: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -1 0 0.5 -0.2 -2 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 4 5 2 5 0 -2 0 0 10 20 30 40 50 60 70 80
Slide 34: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -1 0 0.5 -0.2 -2 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 4 5 2 5 0 -2 0 0 10 20 30 40 50 60 70 80
Slide 35: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -1 0 0.5 -0.2 -2 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 4 5 2 5 0 -2 0 0 10 20 30 40 50 60 70 80
Slide 36: Digging Deeper: Counterfactuals in Regime M
Digging Deeper: Counterfactuals in Regime M Output Multiplier Consumption Multiplier Investment Multiplier 0 0.2 1 -1 0 0.5 -0.2 -2 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 4 5 2 5 0 -2 0 0 10 20 30 40 50 60 70 80
Slide 37: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 2 0.4 0.2 1 0.2 0.1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 10 0 5 5 -1 0 0 10 20 30 40 50 60 70 80 1
Slide 38: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.5 2 2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 50 -20 20 -40 0 0 10 20 30 40 50 60 70 80 10 20
Slide 39: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.5 2 2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 50 -20 20 -40 0 0 10 20 30 40 50 60 70 80 10 20
Slide 40: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.6 2 0.4 2 0.2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 50 20 -20 -40 0 0 10 20 30 40 50 60 70
Slide 41: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.6 2 0.4 2 0.2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 50 20 -20 -40 0 0 10 20 30 40 50 60 70
Slide 42: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.6 2 0.4 2 0.2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 60 40 -20 20 20 -40 0 0 10 20 30 40 50
Slide 43: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 4 0.6 2 0.4 2 0.2 1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 40 0 60 40 -20 20 20 -40 0 0 10 20 30 40 50
Slide 44: Steady-State Counterfactuals in Regime F
Steady-State Counterfactuals in Regime F I Special case of regime M is Ricardian equivalence I in estimates, it breaks down because G used to stabilize debt I There are no cases in regime F consistent with Ricardian equivalence I government debt important state variable I To understand role of debt
Slide 45: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 2 0.4 0.2 1 0.2 0.1 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 10 10 0 5 5 -1 0 0 10 20 30 40 50 60 70 80 1
Slide 46: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 0.5 0.4 2 1 0.2 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 0 10 10 -1 5 5 -2 0 0 -3 10 20 30 40 50 60 70 80
Slide 47: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 0.5 0.4 2 1 0.2 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 0 10 10 -1 5 5 -2 0 0 -3 10 20 30 40 50 60 70 80
Slide 48: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 0.5 0.4 2 0.2 1 0 0 0 -0.2 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 10 10 0 5 5 -2 0 0 10 20 30 40 50 60 70 8
Slide 49: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 0.5 0.4 2 0.2 1 0 0 0 -0.2 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 10 10 0 5 5 -2 0 0 10 20 30 40 50 60 70 8
Slide 50: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 3 0.8 0.6 2 0.4 0.4 1 0.2 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 15 15 0 10 10 -2 5 5 -4 0 0 10 20 30
Slide 51: Digging Deeper: Counterfactuals in Regime F
Digging Deeper: Counterfactuals in Regime F Output Multiplier Consumption Multiplier Investment Multiplier 3 0.8 0.6 2 0.4 0.4 1 0.2 0 0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 Nominal Interest Rate Inflation Real Interest Rate 2 15 15 0 10 10 -2 5 5 -4 0 0 10 20 30
Slide 52: Fiscal Financing Across Regimes
Fiscal Financing Across Regimes I Define some variables ex-post bond return: r̂tB ≡ (βρ/eγ )P̂tB − P̂t−1 B − π̂t exogenous G: Ĝxt endogenous G: Ĝet ∞ G X ∆ PV surpluses from G shock: ξt ≡ −(1 − β) Et β j Ĝxt+j S j=0 I Intertemporal equilibrium condition (debt valuation) ∞ ∞ βρ b̂t−1 = − γ P̂tB +
Slide 53: Fiscal Financing Across Regimes
Fiscal Financing Across Regimes I Obtain sources of financing for innovation in ξt βρ B B ξt = b̂t−1 + P̂ −P̂t−1 γ t − π̂t e |{z} surprise | {z } surprise inflation bond prices ∞ X ∞ X + Et β j r̂t+j B − (1 − β)Et β j Ŝt+j e j=1 j=0 | {z } | {z } expected future expected future real returns endogen
Slide 54: Fiscal Financing Across Regimes
Fiscal Financing Across Regimes Percentage of Government Spending Financing, ξt , Due to Specification P̂tB π̂t P V (r̂ B ) P V (Ẑ) P V (Ĝe ) P V (Ŝ e ) Posterior Estimates 1955q1–2014q2 Regime M 11.1 0.4 -6.3 -42.9 113.5 94.8 [8.2, 13.6] [0.3, 0.6] [-8.7, -3.7] [-78.3, -6.5] [77.9, 149.4] [91.8,
Slide 55: Wrap Up
Wrap Up I Paper uses prior predictive as a model specification device I ensures results are not hardwired I Despite relatively diffuse priors, data highly informative I posterior credible sets much smaller than prior sets I diffuse priors land posterior in regions of parameter space that yield fresh
Slide 56: Clearing Up the Morass
Clearing Up the Morass
Banque Nationale De Belgique — Technical Presentation (September 2015) on 6ix