Banque Nationale De Belgique Sa · EURONEXT:BNB
Banque Nationale De Belgique Sa — Investor Presentation (December 2019)
30 slides from Banque Nationale De Belgique Sa's investor presentation, December 2019
Slide 1: Money and Banking in a New Keynesian Model
Money and Banking in a New Keynesian Model Monika Piazzesi Ciaran Rogers Martin Schneider Stanford Stanford Stanford Belgium, December 2019
Slide 2: Motivation
Motivation Standard New Keynesian model I central bank controls short rate in household stochastic discount factor I short rate = return on savings & investment This paper: New Keynesian model with banking sector I central bank controls interest rate on fed funds or reserves I households do not hold
Slide 3: Corridor system with scarce reserves
Corridor system with scarce reserves monetary policy targets fedfunds rate, sets reserve rate trading desk supplies reserves elastically to meet target banks’ cost of liquidity > 0, rises if central bank tightens Reserve Supply iF Reserve Demand M i Reserves
Slide 4: Floor system with ample reserves
Floor system with ample reserves monetary policy sets reserve rate & quantity of reserves banks’ cost of liquidity zero, remains zero if central bank tightens Reserve Supply Reserve Demand i M , iF Reserves
Slide 5: Implications
Implications Standard NK model I interest rate is all that matters, plumbing & quantities not important Banking & short rate disconnect: plumbing & quantities matter I floor system: interest rate policy only affects banks’ cost of safety - higher reserve rate, cheaper safe collateral to back inside
Slide 6: Plan for talk
Plan for talk Transmission in minimal model with disconnect I Households make payments with CBDC (no banks) Introduce banks that provide inside money for payments I Government supplies ample reserves (floor system) I Scarce reserves (corridor system)
Slide 7: Literature
Literature NK models with financial frictions & banking Bernanke-Gertler-Gilchrist 99, Cúrdia-Woodford 10, Gertler-Karadi 11, Gertler-Kiyotaki-Queralto 11, Christiano-Motto-Rostagno 12, Del Negro-Eggertson-Ferrero-Kiyotaki 17, Diba-Loisel 17, Arce-Nuño-Thaler-Thomas 19 Convenience yields on bonds
Slide 8: Minimal model with short rate disconnect (no banks)
Minimal model with short rate disconnect (no banks) Representative household I utility separable in labor + CES bundle of consumption & money I σ = IES for bundles, η = interest elasticity of money demand I for now, separable in consumption & money: η = σ I later consider complementarity: η < σ Firm
Slide 9: Linear dynamics
Linear dynamics Steady state with zero inflation Standard NK Phillips curve & Euler equation, κ = λ ϕ + σ1 ∆p̂t = β∆p̂t +1 + κ ŷt ŷt = ŷt +1 − σ itS − ∆p̂t +1 − δ Households’ money demand η S dˆt − p̂t = ŷt − i t − it D − δ − r D δ − rD
Slide 10: Why money does not matter in the standard NK model
Why money does not matter in the standard NK model ∆p̂t = β∆p̂t +1 + κ ŷt ŷt = ŷt +1 − σ itS − ∆p̂t +1 − δ η S dˆt − p̂t = ŷt − i − i D − δ − r D δ − rD t t Solve for (p̂t , ŷt , itS , itD , dˆt ) given initial condition p̂0 Standard model I add 2 policy rules: Taylor rule for itS , peg for itD
Slide 11: Why money matters in CBDC model
Why money matters in CBDC model ∆p̂t = β∆p̂t +1 + κ ŷt ŷt = ŷt +1 − σ it − ∆p̂t +1 − δ S η S dˆt − p̂t = ŷt − i − i D − δ − r D δ − rD t t CBDC model I adds 2 policy rules: interest on money itD , quantity of money Dt I short rate itD endogenous, satisfies Euler equation → disconnect: policy rat
Slide 12: Disconnect and role of money with banks
Disconnect and role of money with banks standard model: policy rate = short rate, money does not matter CBDC model: policy rate 6= short rate, money matters banking model with floor system works like CBDC model I rules for reserve rate itM , quantity of reserves Mt I short rate disconnect: household
Slide 13: Interest rate policy
Interest rate policy Standard model: short rate itS = policy rate Transmission of interest rate policy + real return − output, policy rate −→ on savings −→ inflation Money supplied elastically to implement itS , fix itD = 0
Slide 14: Interest rate policy
Interest rate policy CBDC model: convenience yield is endogenous wedge δ − rD itS − δ = itD − r D + p̂t + ŷt − dˆt η convenience yield, increasing in policy rate velocity = spending / money Transmission of interest rate policy + real return − output, policy rate −→ on savings −→ inflation +↑ +↓ con
Slide 15: Local determinacy with interest rate peg
Local determinacy with interest rate peg Standard model: many bounded solutions to difference equation When do we get multiple bounded equilibrium paths? Taylor principle violated low real rate on savings Euler equation % & high inflation ←− high output Phillips curve Taylor principle: policy reacts
Slide 16: Taylor rule: local determinacy with separable utility
When do we get local determinacy with separable utility? Taylor rule itD = r D + φπ ∆p̂t + φy ŷt + υt Money supply rule Dt = µDt −1 + Pt G , µ ≤ 1 I choose µ, G , r D to achieve zero inflation in steady state I with µ = 1, G = 0 → constant money supply, nominal anchor I with µ < 1, no nominal ancho
Slide 17: Cost channel
Cost channel Consumption & money complements in utility I nonseparable utility with η < σ I higher cost of liquidity itS − itD makes shopping less attractive → reduce consumption, increase leisure/decrease labor → lower output, higher inflation Effect of higher policy rate on cost of liquidity itS −
Slide 18: IRFs to monetary policy shock: standard model
IRFs to 25 bp monetary policy shock: standard model % deviations from SS price level output money % deviations from SS % deviations from SS 0 0 -0.08 -0.1 -2 -0.09 -0.2 -0.3 -4 -0.1 -0.4 0 4 0 4 0 4 inflation policy rate conv. yield 0 0.4 0.4 -0.1 % p.a. % p.a. % p.a. -0.2 0.2 0.2 -0.3 0 0 0 4 0 4 0
Slide 19: IRFs to monetary policy shock: standard vs CBDC
IRFs to 25 bp monetary policy shock: standard vs CBDC % deviations from SS price level output money % deviations from SS % deviations from SS -0.06 0 0 -0.07 -0.1 -2 -0.08 -0.2 -0.09 -0.3 -4 -0.1 -0.4 0 4 0 4 0 4 inflation policy rate conv. yield 0.6 0 0.6 standard CDBC 0.4 -0.1 % p.a. % p.a. % p.a.
Slide 20: IRFs to monetary policy shock: standard vs CBDC
IRFs to 25 bp monetary policy shock: standard vs CBDC % deviations from SS price level output money % deviations from SS % deviations from SS -0.06 0 0 -0.07 -0.1 -2 -0.08 -0.2 -0.09 -0.3 -4 -0.1 -0.4 0 4 8 12 16 20 0 4 8 12 16 20 0 4 8 12 16 20 inflation policy rate conv. yield 0.6 0 0.6 standard C
Slide 21: NK Model with Banks
NK Model with Banks central bank provides ample reserves (”floor system”) I reserves are special as collateral, not needed for liquidity I monetary policy targets reserve rate Reserve Supply Reserve Demand i M , iF Reserves
Slide 22: Banking sector
Banking sector Balance sheet Assets Liabilities M Reserves Money D A Other assets Equity Shareholders maximize present value of cash flows Mt −1 1 + itM−1 − Mt + At −1 1 + itA−1 − At − Dt −1 1 + itD−1 + Dt Costless adjustment of equity Leverage constraint: Dt ≤ ` (Mt + ρAt ) I ρ < 1 other assets are
Slide 23: Bank optimization: perfect competition
Bank optimization: perfect competition Nominal rate of return on equity = itS I banks equate returns on assets & liabilities to cost of capital itS I γt = multiplier on leverage constraint Optimal portfolio choice: assets valued as collateral itS = itM + `γt 1 + itS itS = itA + ρ`γt 1 + itS Optimal
Slide 24: Bank market power
Bank market power Many monopolistically competitive banks Households care about CES bundle of deposit varieties Z 1 1− η1 1− η1 Dt = Dti b b I ηb = elasticity of substitution between bank accounts ⇒ Constant markup over marginal cost ηb 1 S itS − itD = it − itM ηb − 1 `
Slide 25: Equilibrium with ample reserves
Equilibrium with ample reserves Government: floor system with ample reserves I path or rule for supply of reserves Mt I path or rule for interest rate on reserves itM Market clearing for reserves & other bank assets I exogenous path Ar of real assets, so At = Pt Ar t t I stands in for borrowing by f
Slide 26: Dynamics with ample reserves
Dynamics with ample reserves Interest rate pass-through: reserve rate to short rate δ − rM itS − δ = itM − r M + p̂t + ŷt − dˆt η I reserves back inside money, inherit convenience yield of deposits Money supply M ρA dˆt = m̂t + ât M + ρA M + ρA I reserves a separate policy instrument: QE stimulate
Slide 27: Banking with scarce reserves
Banking with scarce reserves Banks manage liquidity I deposit outflow/inflow λ̃Dt to/from other banks I iid liquidity shock λ̃ has mean zero, cdf G with bounded support I satisfy leverage constraint after deposit inflow/outflow I borrow/lend in competitive fed funds market at rate i F Assets valued
Slide 28: Dynamics with scarce reserves
Dynamics with scarce reserves Interest rate pass-through: fed funds rate to short rate δ − rM itS − δ = itF − r M + p̂t + ŷt − dˆt η Inside money in reserveless limit: share of reserves in bank assets → 0 η ε η dˆt = ât + p̂t + ŷt − F itF − r F η+ε η+ε r I ε = function of bank technology paramete
Slide 29: 25bp increase in policy rate: corridor vs floor systems
25bp increase in policy rate: corridor vs floor systems price level output money % deviations from SS % deviations from SS % deviations from SS -0.05 0 0 -0.1 -0.1 -0.15 -0.5 -0.2 -0.2 -0.3 -1 -0.25 0 4 0 4 0 4 inflation policy rate conv. yield 0 0.6 0.08 corridor 0.06 floor % p.a. % p.a. % p.a. -0.
Slide 30: Conclusion
Conclusion Disconnect between policy rate and short rate I convenience yield is endogenous wedge, changes transmission I less scope for multiple equilibria, even without Taylor principle I policy weaker if more nominal rigidities in balance sheets Bank models vs CBDC model I same basic transmission
Banque Nationale De Belgique — Investor Presentation (December 2019) on 6ix