Consumption Dynamics, Housing Collateral and Stabilisation Policy A Way Forward for Macro-Prudential Instruments? Effective Macroprudential Instruments - CFCM-MMF-MMPM Conference Jagjit S. Chadha University of Kent Kent. 3th November 24 Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 / 27
Overview One Club monetary policy has not only insuffi cient to prevent booms and busts but may have played a role in nurturing volatility. The newly formed FPC at the BoE has asked for extra instruments countercyclical capital, sectoral capital, leverage ratio. No liquidity or LTVs, yet... See Chadha and Corrado (JBus 22) and Chadha, Corrado and Meaning (BIS WP 66) on welfare and output enhancing role of liquidity because it can reduce the volatility of the external finance premium. MPIs are untested and with little evidence - no established models or data House of Commons Treasury Committee Evidence, see http://www.publications.parliament.uk/pa/cm223/cmselect/cmtreasy/writev/macropru/mpt3.htm Square : consider role of bank behaviour and consumption-housing dynamics. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 2 / 27
Consumption Dynamics Representative agent paradigm has diffi cult dealing with financial frictions and asset prices Kiyotaki-Moore emphasize role of collateral constraints in investment cycles Much current focus on banking, which intermediate between borrowers and savers and endogenise the external finance premium e.g. Christiano et al, (27), Goodfriend and McCallum (27) and Chadha et al (23) with some violation of Modigliani-Miller Natural to consider heterogeneous agents who are either saver or borrowers Use of micro-founded model allows some welfare analysis of monetary-fiscal-financial policies Creditor-Borrower dynamics exacerbate intra and intertemporal volatility. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 3 / 27
Main Results Chadha, Corrado and Corrado (23) explore the link between house prices, borrower consumption, LTVs and the lending rate via collateral constraints NOT wealth or common shock We can match the house price-consumption correlation Higher order consumption processes for household type c.f. aggregate consumption Borrower-payback lending cycle for borrower households has spillovers for savers over time Corollary Consider the nexus of monetary-fiscal-financial policies A more restrictive set of rules for bank lending over the business cycle seem to reduce welfare losses Not clear that our commercial bank can arrive at the best set of intermediation policies for the representative household or creditor and borrower households in the absence of a policy intervention. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 4 / 27
The Lucas Costs of Business Cycles U( c) = c ρ ρ where c = c + c b and ρ is the CRRA parameter. E [U( c)] c ρ ρ ρ 2 c ρ σ 2 c ρ ( σ ) c 2 ρ σ c + σ = c b + 2σ c,c b 2 c 2 c 2 Obviously aggregate business cycle costs are small if covariances are negative But what if this covariation drives the business cycle? Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 5 / 27
Real House Prices and Consumption (UK) Consumption and House Values Borrower Saver All Levels:*.79 -.452.62 Growth:*.453.8.39 LHS: BoE QB and RHS: our calculations of correlation between house prices and consumption from BHPS Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 6 / 27
Model Structure Stylised representation of the sectors in our model. [Many arbitrary choices!] Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 7 / 27
The Household Problem: Creditor Creditor Household Utility Resource constraint max U = E t=β t log C t +χ B log B t C t + (ωm t+ +d t+ +B t+ ) π t+ = ωm t + R D t d t + R B t B t Cash-in-Advance constraint +q t (H t H t ) + Θ t + Π t Tax t C t ωm t Creditor Household Euler Equation and Bond Demand C t = βr t D π t+ C t+ χ B B t = ( ) βr t B Ct π t+ C t+ Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 8 / 27
The Household Problem: Workers (Borrowers) Borrower Household Utility [ ] max U b = E β t log C b t + χ log H t N+ς t t= + ς Resource Constraint C b t + q t (H t H t ) + ( ω) m t+ π t+ + RL t l t π t = l t + ( ω) m t + w t N t τ h q t H t Cash-in-Advance constraint Borrowing constraint C b t ( ω) m t +w t N t l t κ t q t+ π t+ H t Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 9 / 27
Borrower Consumption Plans Borrower Consumption Goods Demand Residential Goods Demand β β 2 Rt Ct+ b π L t+ Ct+2 b π = ν t t+2π t+ χ q t = βh t Ct+ b π ( + τ h ) q t+ J ( ) t+ ( with J ( ) = κ t βr ) t L β +.is the shadow price of housing Ct+ b Ct+2 b π t+2 demand in terms of consumption units. Labor Supply: N ς t C b t C b t+2 π t+2 = w t Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 / 27
Bank Balance Sheet Bank lending constraint and critical threshold for house price: ( l j,t δκ t q t H t q t = C t+ b π t+ (+τ h ) χ β + q t+j ( )) Ex ante profits: Π j,t = s( κt ) δκ t q t H t φ(κ t )dκ t + l j,t Rj,t L φ(κ t )dκ t s( κ t ) R D j,td j,t +R M t rr j,t Bank Balance Sheet: r j,t +l j,t = d j,t Price of Risk (where lambda is the failure rate priced into loans): ˆR t L = ˆR M t + λ [ˆl t δ (ˆκ )] t + ˆq t + Ĥ t δ Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 / 27
Macroprudential Instruments Definition time varying (state-contingent) requirements placed on financial intermediaries by the monetary authorities that constrain the scale or set of available tools E.g. capital requirements or leverage ratios or LTVs or liquidity covers We consider three that emerge from this model: λ - the perception of the risk of loan default - benchmark 2% 2 δ - the fraction collateral capital that can be seized in the event of default - can be interpreted as firesale value -.5 3 κ - the loan to value ratio limit to which banks can lend -.6 A fourth is not considered here i.e. reserve requirements, rr, see Chadha and Corrado (22). Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 2 / 27
Real Sector and Monetary-Fiscal Policy Rules Phillips Curve: ˆπ t = β ˆπ t+ + η w (ŵt + ˆX ψ,t ) ηa A t + η y Ŷ t where η w = ϱ, η a = ϱγ, η y = γ Aggregate Supply: where ϑ ya = +ς +ς γ and ϑ yc = Monetary Policy: ϱγ. Ŷ t = ϑ ya A t ϑ yc [Ĉ b t+ + ˆπ t+ ] γ +ς γ. ˆR M t = ρ ˆR M t +( ρ) [α πe t ˆπ t + α y ŷ t ] +ξ t Fiscal Policy: B Y ˆB t = B ( ) Y RB ˆR t B + ˆB t ˆπ t + G Y Ĝt T Y ˆT t m Y ( ˆm t + ˆπ t ) Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 3 / 27
Welfare Criterion Welfare function: U t = C Y (log C t + χ B log B t ) + C b Y ( ) log C b t + χ log H t N+ς t + ς + G Y (log G t) Second-order Taylor series expansion of U t around steady-state (C t, C b t, H t ): (U t U) 2 ( +t.i.p.+o ) ( +ς γ Ŷ 2 t C Y Ĉ t 2 + C (Ĉ ) b b 2 ) Y t + G Y Ĝ t 2 Y C χ B ˆB t 2 C b Y χĥ2 t ψ η ( a ˆπ2 3) E β t (U t U) = ( t= 2 E β t ϕy σ 2 Y +ϕ C σ2 C +ϕ C σ 2 +ϕ b C b G σ 2 G +ϕ t= B σ 2 B +ϕ H σ2 H +ϕ π σ2 π +t.i.p+o ( a 3) ) t+ Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 4 / 27
Solution Technique AE t y t+ = By t + Cε t t, (dynamic system) y t = Πs t (state space form) s t = Ms t + Ge t, E t y t+k E t y t+k = Π (E t s t+k E t s t+k ) = M k Ge t. (Impulse Responses) vec (V ss ) = (I M M) vec ( GV ee G T ). (Variances) L = ϕ Y σ 2 Y + ϕ C σ 2 C + ϕ C b σ 2 C b + ϕ G σ 2 G + ϕ B σ 2 B + ϕ H σ 2 H + ϕ π σ 2 π (Loss function) Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 5 / 27
Simulated Data.6 Consumption Saver Consumption Borrower House Price Aggregate Consumption.4.2.2.4.6 2 3 4 5 6 7 8 9 Quarters creditors and savers are negatively correlated but average consumption is closely tied to house prices Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 6 / 27
unbundled Supply Shock - borrowers increase consumption now but pay Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 7 / 27 Productivity Consumption Savers 2 Consumption Borrowers.5 Lending Rate.5.5.5 5 5 2 5 5 2 5 5 2. Deposit Rate, Policy Rate.5 Lending.5 House Prices.5.5.5.5. 5 5 2 Employment.5 5 5 2 Quarters after shock Inflation 5 5 2 Output.6..2.5.4.2.3..4 5 5 2.5 5 5 2.2 5 5 2 Benchmark Lax Regime Restrictive Regime
Loan to Value.4 Consumption Savers Consumption Borrowers.5 Lending Rate.2.5.5.2.4 5 5 2.5 5 5 2.5 5 5 2.5 Deposit Rate, Policy Rate Lending.5 House Prices..5.5.5.5.6 5 5 2 Employment. 5 5 2 Quarters after shock Inflation.3 5 5 2 Output.4.5.2.2..5.2 5 5 2. 5 5 2. 5 5 2 Benchmark Lax Regime Restrictive Regime unbundled demand shock - under lax regime is amplified Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 8 / 27
classical monetary policy shock - attenuated by offsetting sectors Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 9 / 27 Monetary Policy.5 Consumption Savers Consumption Borrowers.6 Lending Rate.5.4.2.5 5 5 2.5 5 5 2.2 5 5 2.3 Deposit Rate, Policy Rate.2 Lending.2 House Prices.2..2.2.4.4..3 5 5 2 Employment.6.2 5 5 2 Quarters after shock Inflation.6.5 5 5 2 Output.2...2.5.4. 5 5 2.6 5 5 2.5 5 5 2 Benchmark Lax Regime Restrictive Regime
House Prices Cons um ption Savers Consumption Borrowers.5 Lending Rate.5.5.5.5.5 5 5 2 5 5 2 5 5 2. Deposit Rate, Policy Rate.6 Lending Hous e Pric es.5.4.5.2.5. 5 5 2 Employ m ent.2.5 5 5 2 Quarters after shock Inflation.5.6 5 5 2 Output.5.4.5.2..5 5 5 2.5 5 5 2.2 5 5 2 Benchmark Lax Regime Restrictive Regime temporary impact on output growth but not permanent impact on level so subsequent negative adjustment Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 2 / 27
Cost Push 3 Consum ption Savers Consum ption Borrowers Lending Rate 2 2 3 2 5 5 2 4 5 5 2 3 5 5 2 Deposit Rate, Policy Rate Lending House Prices.4.2.2 2 2.4.5 5 5 2 Em ploy m ent 3 5 5 2 Quarters after shock Inflation 3.5 5 5 2 Output.5.4.2.2.5.5 5 5 2.4 5 5 2 5 5 2 Benchmark Lax Regime Restrictive Regime may want relaxed MPIs in case of cost-push shocks Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 2 / 27
Government Spending. Consumption Savers.5 Consumption Borrowers.6 Lending Rate.5..4.5.2.5. 5 5 2.5 5 5 2.2 5 5 2 5 x Deposit Rate, Policy Rate 3.4 Lending.2 Hous e Pric es.2 5.2.2 5.3 5 5 2 Employment.4 5 5 2 Quarters after shock 5 x Inflation 3.4.2 5 5 2 Output.2.. 5. 5 5 2 5 5 5 2. 5 5 2 Benchmark Lax Regime Restrictive Regime small output response. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 22 / 27
Model Moments Table 4. Moments Rel stdev (Y ) Corr (Y ) Benchmark Policy C.8.99 C 2.5 -.6 C b 3.8.9 G.9.7 l.9.4 R L.7.43 R M, R D.59.24 EFP.69.24 q 2.5.45 π.6. All shocks and standard calibration. Note procyclical EFP. High volatility of households consumption cf aggregate Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 23 / 27
Optimal Policy Parameters T 5. O P P Default Rate Optimal Policy Parameters f y f T f m α q λ = Fiscal Policy.2.3. Monetary Policy. λ = 5 Fiscal Policy.6.27. Monetary Policy.3 Higher perception of default imply more active fiscal policy (more countercyclical) and a greater role for monetary policy to offset house prices (collateral). Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 24 / 27
Welfare Losses Table 6. Welfare Losses Loss Lax Regime λ =, δ =.9, κ =.9 Benchmark Policy.628 Monetary Policy.586 Fiscal Policy.63 Restrictive Regime λ = 5, δ =.2, κ =.4 Benchmark Policy.53 Monetary Policy.529 Fiscal Policy.543 A given monetary-fiscal nexus leads to lower losses under more restrictive MPIs Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 25 / 27
Main Results When we split up consumption into saver and borrower households - we clearly locate consumption cycles Borrowers use houses as collateral and are able to smooth shocks and experience similar levels of volatility as savers Savers observe standard Euler equations but also face spillovers as they fund borrower demand The bank perception of risk, increases consumption variance, as the EFP becomes more sensitive to the loans function When banks do not price risk, welfare can be enhanced by more active fiscal policy Monetary policy can stabilise inflation (price of money) but fiscal policy has a role in pricking excessive bank lending (quantities) In general MPIs seem to lead to a more effi cient outcome in terms of volatility frontier. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 26 / 27
Bibliography Chadha, J., Pathology of a Heart Attack - the LIBOR Market, In Press, Quantitative Finance, 24. Chadha, J. and L. Corrado, Macro-prudential policy on liquidity: What does a DSGE model tell us?, Journal of Economics and Business, Volume 64, Issue, 22, Pp37-62. Chadha, J., L. Corrado and S Holly, A Note on Money and the Conduct of Monetary Policy, In Press, Macroeconomic Dynamics, 24. Chadha, J, L. Corrado and G. Corrado, 24, Stabilisation Policy in a Model of Consumption, Housing Collateral and Bank Lending, mimeo. Chadha, J, Corrado L. and J. Meaning, Reserves, Liquidity and Money: An Assessment of Balance sheet Policies, Are central bank balance sheets in Asia too large?, BIS Working Paper 66, 22, vol. 66, pp 294-347 Chadha, J., P. Turner and F. Zampolli, The Ties that Bind: Non-Conventional Monetary Policy and Debt Management, Oxford Review of Economic Policy, 23. Chadha J. and A. Waters, Applying a Macro-Finance Yield Curve to UK Quantitative Easing, In Press, Journal of Banking and Finance, 24. Chadha at al. (Kent.) Macro-Prudential Policies 3th November 24 27 / 27