Regulatory Arbitrage and Systemic Liquidity Crises

Transcription

1 Regulatory Arbitrage and Systemic Liquidity Crises Stephan Luck & Paul Schempp Princeton University and MPI for Research on Collective Goods Federal Reserve Bank of Atlanta The Role of Liquidity in the Financial System Nov 19, / 20

2 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility 1 / 20

3 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen 1 / 20

4 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? 1 / 20

5 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? This paper: 1. Contagion in absence of contractual linkages 1 / 20

6 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? This paper: 1. Contagion in absence of contractual linkages Mechanism: pecuniary channel Systemic runs and deterioration of funding conditions 1 / 20

7 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? This paper: 1. Contagion in absence of contractual linkages Mechanism: pecuniary channel Systemic runs and deterioration of funding conditions 2. Regulatory arbitrage is excessive 1 / 20

8 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? This paper: 1. Contagion in absence of contractual linkages Mechanism: pecuniary channel Systemic runs and deterioration of funding conditions 2. Regulatory arbitrage is excessive 3. Policy implications Macroprudential regulation, central bank interventions 1 / 20

9 Motivation/research question Regulatory arbitrage and the financial crisis of Linkages regulated banks shadow banks identified as source of fragility Regulatory response: Restrict explicit and implicit linkages Volcker, Vickers, Liikanen Implemented and proposed reforms effective? This paper: 1. Contagion in absence of contractual linkages Mechanism: pecuniary channel Systemic runs and deterioration of funding conditions 2. Regulatory arbitrage is excessive 3. Policy implications Macroprudential regulation, central bank interventions 1 / 20

10 Table of contents 1 A banking model with systemic runs 2 Banks, shadow banks, and systemic liquidity crises 3 Policy implications 1 / 20

11 Table of contents 1 A banking model with systemic runs 2 Banks, shadow banks, and systemic liquidity crises 3 Policy implications 1 / 20

12 Setup: primitives Modified Diamond & Dybvig 83 Three dates, t = 0, 1, 2 One good, can be used for consumption and investment Three types of agents: Depositors Intermediaries Investors Three types of technologies 2 / 20

13 Setup: technologies t = 0 t = 1 t = 2 Productive technology -1 0 R Shirking technology -1 0 R Shirk + B 3 / 20

14 Setup: technologies t = 0 t = 1 t = 2 Productive technology -1 0 R Shirking technology -1 0 R Shirk + B Storage in t = Storage in t = / 20

15 Setup: technologies t = 0 t = 1 t = 2 Productive technology -1 0 R Shirking technology -1 0 R Shirk + B R > 1 > R shirk + B B not pledgeable Storage in t = Storage in t = / 20

16 Setup: technologies t = 0 t = 1 t = 2 Productive technology -1 0 R Shirking technology -1 0 R Shirk + B R > 1 > R shirk + B B not pledgeable Storage in t = Storage in t = / 20

17 Setup: depositors Continuum of depositors, endowment of 1 unit each A fraction π is impatient, utility u(c 1) A fraction 1 π is patient, utility u(c 2) Types are initially unknown 1st key friction: privately revealed in t = 1 4 / 20

18 Setup: depositors Continuum of depositors, endowment of 1 unit each A fraction π is impatient, utility u(c 1) A fraction 1 π is patient, utility u(c 2) Types are initially unknown 1st key friction: privately revealed in t = 1 CRRA utility Expected utility EU = U(c 1, c 2) = πu(c 1) + (1 π)u(c 2) 4 / 20

19 Setup: depositors Continuum of depositors, endowment of 1 unit each A fraction π is impatient, utility u(c 1) A fraction 1 π is patient, utility u(c 2) Types are initially unknown 1st key friction: privately revealed in t = 1 CRRA utility Expected utility EU = U(c 1, c 2) = πu(c 1) + (1 π)u(c 2) Intermediary is required for investment 4 / 20

20 Setup: depositors Continuum of depositors, endowment of 1 unit each A fraction π is impatient, utility u(c 1) A fraction 1 π is patient, utility u(c 2) Types are initially unknown 1st key friction: privately revealed in t = 1 CRRA utility Expected utility EU = U(c 1, c 2) = πu(c 1) + (1 π)u(c 2) Intermediary is required for investment 4 / 20

21 Setup: intermediaries Continuum of intermediaries, competitive Endowment that may be invested in intermediation Required return ρ > R skin-in-the-game costly adverse selection, leverage-ratchet, non-pecuniary benefits, risk anomalies 5 / 20

22 Setup: investors Continuum of investors, no market power Endowment A at date t = 1, market liquidity Required rate of return µ [1, R] Assumption R/µ A πr/µ market provides sufficient liquidity in normal times, but liquidity is scarce in crisis 6 / 20

23 Setup: investors Continuum of investors, no market power Endowment A at date t = 1, market liquidity Required rate of return µ [1, R] Assumption R/µ A πr/µ market provides sufficient liquidity in normal times, but liquidity is scarce in crisis 2nd key friction: investors born in t = 0 (Holmstrom and Tirole 1998) 6 / 20

24 Setup: investors Continuum of investors, no market power Endowment A at date t = 1, market liquidity Required rate of return µ [1, R] Assumption R/µ A πr/µ market provides sufficient liquidity in normal times, but liquidity is scarce in crisis 2nd key friction: investors born in t = 0 (Holmstrom and Tirole 1998) 6 / 20

25 Depositors Intermediaries Liquid Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

26 Depositors Intermediaries Liquid Illiquid Demand Deposits Disciplining short-term debt Illiquid Investment Technologies 7 / 20

27 Depositors Intermediaries Liquid Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

28 Depositors Intermediaries Liquid Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

29 Depositors Intermediaries Wholesale Funding Market Liquid Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

30 Depositors Intermediaries Wholesale Funding Market Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

31 Depositors Intermediaries Wholesale Funding Market Illiquid Demand Deposits Illiquid Investment Technologies 7 / 20

32 Depositors Intermediaries Wholesale Funding Market Illiquid Demand Deposits Secured Wholesale Funding Illiquid Investment Technologies 7 / 20

33 Bank runs and systemic runs Intermediaries are financing illiquid asset by demand deposits (c 1, c 2 ) Mass of intermediaries that experience a run: α [0, 1] c 1 > R/µ: runs on single intermediaries always possible 8 / 20

34 Bank runs and systemic runs Intermediaries are financing illiquid asset by demand deposits (c 1, c 2 ) Mass of intermediaries that experience a run: α [0, 1] c 1 > R/µ: runs on single intermediaries always possible (boring) When are systemic runs possible? (interesting) 8 / 20

35 Bank runs and systemic runs Intermediaries are financing illiquid asset by demand deposits (c 1, c 2 ) Mass of intermediaries that experience a run: α [0, 1] c 1 > R/µ: runs on single intermediaries always possible (boring) When are systemic runs possible? (interesting) Systemic runs: runs on some intermediaries that affect other intermediaries 8 / 20

36 Bank runs and systemic runs Intermediaries are financing illiquid asset by demand deposits (c 1, c 2 ) Mass of intermediaries that experience a run: α [0, 1] c 1 > R/µ: runs on single intermediaries always possible (boring) When are systemic runs possible? (interesting) Systemic runs: runs on some intermediaries that affect other intermediaries Proposition A run on α intermediaries is systemic if α > ᾱ, where ᾱ = A πc 1. R/µ πc1 8 / 20

37 Bank runs and systemic runs Intermediaries are financing illiquid asset by demand deposits (c 1, c 2 ) Mass of intermediaries that experience a run: α [0, 1] c 1 > R/µ: runs on single intermediaries always possible (boring) When are systemic runs possible? (interesting) Systemic runs: runs on some intermediaries that affect other intermediaries Proposition A run on α intermediaries is systemic if α > ᾱ, where ᾱ = A πc 1. R/µ πc1 8 / 20

38 Systemic runs ( proof ) L units of the asset can be sold at p = R/µ ( fundamental value ) if LR/µ A Otherwise, the market clears via cash-in-the-market pricing (compare Allen and Gale 94) Cash-in-the-market price is given by p = A/L 9 / 20

39 Systemic runs ( proof ) L units of the asset can be sold at p = R/µ ( fundamental value ) if LR/µ A Otherwise, the market clears via cash-in-the-market pricing (compare Allen and Gale 94) Cash-in-the-market price is given by p = A/L Possible because we assumed R/µ > A 9 / 20

40 Systemic runs ( proof ) L units of the asset can be sold at p = R/µ ( fundamental value ) if LR/µ A Otherwise, the market clears via cash-in-the-market pricing (compare Allen and Gale 94) Cash-in-the-market price is given by p = A/L Possible because we assumed R/µ > A 9 / 20

41 Systemic runs ( proof ) α intermediaries experience a run and sell all assets 1 α intermediaries need to pay out πc1 each If p = R/µ, the amount of assets sold in aggregate is: α + (1 α)πc1 /(R/µ) 10 / 20

42 Systemic runs ( proof ) α intermediaries experience a run and sell all assets 1 α intermediaries need to pay out πc 1 each If p = R/µ, the amount of assets sold in aggregate is: Not compatible with p = R/µ if: Given our assumptions, ᾱ < 1 α + (1 α)πc 1 /(R/µ) R/µ[α + (1 α)πc 1 /(R/µ)] > A α > A πc 1 R/µ πc 1 ᾱ

43 Systemic runs ( proof ) α intermediaries experience a run and sell all assets 1 α intermediaries need to pay out πc 1 each If p = R/µ, the amount of assets sold in aggregate is: Not compatible with p = R/µ if: Given our assumptions, ᾱ < 1 α + (1 α)πc 1 /(R/µ) R/µ[α + (1 α)πc 1 /(R/µ)] > A α > A πc 1 R/µ πc 1 ᾱ 10 / 20

44 Systemic runs ( proof ) α intermediaries experience a run and sell all assets 1 α intermediaries need to pay out πc 1 each If p = R/µ, the amount of assets sold in aggregate is: Not compatible with p = R/µ if: Given our assumptions, ᾱ < 1 α + (1 α)πc 1 /(R/µ) R/µ[α + (1 α)πc 1 /(R/µ)] > A α > A πc 1 R/µ πc 1 ᾱ 10 / 20

45 Systemic runs: an illustration Consu mers A L Consu mers A L Consu mers Market for Wholesale Funding A L A L Consu mers A L Consu mers 11 / 20

46 Systemic runs: an illustration Consu mers A L Consu mers A L Consu mers Market for Wholesale Funding A L A L Consu mers A L Consu mers 11 / 20

47 Systemic runs: an illustration Consu mers A L Consu mers A L Consu mers Market for Wholesale Funding A L A L Consu mers A L Consu mers 11 / 20

48 Systemic runs: an illustration Consu mers A L Consu mers A L Consu mers Market for Wholesale Funding A L A L Consu mers A L Consu mers 11 / 20

49 Systemic runs: an illustration Consu mers A L Consu mers A L Consu mers Market for Wholesale Funding A L A L Consu mers A L Consu mers 11 / 20

50 Systemic runs: an illustration Fire sale price p Consu mers A L Consu mers A L R/µ Consu mers Market for Wholesale Funding A L A L A Consu mers A L Consu mers πc 1 ᾱ 1 α: fraction of intermediaries subject to a run Interaction of two frictions 1st friction: Types are private information (Diamond and Dybvig 1983) 2nd friction: Cannot contract with investors in t = 0 (Holmstrom and Tirole 1998) 11 / 20

51 Systemic runs: deterioration of funding conditions Fire sale price p R/µ Fire-sale price is given by { R/µ if α ᾱ p(α) = A (1 α)πc1 if α > ᾱ. α For α > ᾱ, market liquidity becomes scarce: depressed asset prices, or deteriorated conditions of wholesale funding A πc 1 ᾱ 1 α: fraction of intermediaries subject to a run 12 / 20

52 Table of contents 1 A banking model with systemic runs 2 Banks, shadow banks, and systemic liquidity crises 3 Policy implications 12 / 20

53 Depositors Illiquid Investment Technologies 13 / 20

54 Depositors Shadow Banks Illiquid Quasi Deposits Illiquid Investment Technologies 13 / 20

55 Depositors Shadow Banks Illiquid Quasi Deposits Disciplining short-term debt Illiquid Investment Technologies 13 / 20

56 Depositors Regulated Banks Shadow Banks Illiquid Insured Deposits Illiquid Quasi Deposits Equity Disciplining short-term debt Illiquid Investment Technologies 13 / 20

57 Depositors Regulated Banks Shadow Banks Illiquid Insured Deposits Illiquid Quasi Deposits Equity Diligence induced via skin-in-the-game Disciplining short-term debt Illiquid Investment Technologies 13 / 20

58 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Illiquid Quasi Deposits Equity Illiquid Investment Technologies 13 / 20

59 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Liquid Illiquid Insured Deposits Equity Liquid Illiquid Quasi Deposits Illiquid Investment Technologies 13 / 20

60 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Illiquid Quasi Deposits Equity Illiquid Investment Technologies 13 / 20

61 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Illiquid Quasi Deposits Equity Illiquid Investment Technologies 13 / 20

62 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Illiquid Quasi Deposits Secured Wholesale Funding Equity Illiquid Investment Technologies 13 / 20

63 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Illiquid Quasi Deposits Secured Wholesale Funding Equity Illiquid Investment Technologies 13 / 20

64 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Illiquid Quasi Deposits Secured Wholesale Funding Equity No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

65 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Illiquid Quasi Deposits Equity No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

66 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Fire sale Illiquid Quasi Deposits Equity No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

67 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Fire sale Cash-in-the-market pricing Illiquid Quasi Deposits Equity No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

68 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Fire sale Cash-in-the-market pricing Illiquid Quasi Deposits Equity No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

69 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Equity Fire sale Cash-in-the-market pricing Deterioration of funding conditions Illiquid Quasi Deposits No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

70 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Equity Fire sale Cash-in-the-market pricing Deterioration of funding conditions Illiquid Quasi Deposits No Runs Fragility: Runs on Shadow Banks Illiquid Investment Technologies 13 / 20

71 Depositors Regulated Banks Wholesale Funding Market Shadow Banks Illiquid Insured Deposits Secured Wholesale Funding Equity Contagion w/o contractual linkages No Runs Fire sale Cash-in-the-market pricing Deterioration of funding conditions Illiquid Investment Technologies Illiquid Quasi Deposits Fragility: Runs on Shadow Banks 13 / 20

72 Result 1: Systemic Runs Contagion from the shadow banking sector to regulated banks, even without contractual linkages Deterioration of funding conditions for regulated banks Even without aggregate risk and classic bank runs Banks may turn illiquid and insolvent Deposit insurance may become tested and costly 14 / 20

73 Fire sales and deterioration of funding conditions Fire-sale price / Funding conditions Fundamental Value Eq. fire sale price Equilibrium size 1 Relative size of the shadow banking sector 15 / 20

74 Fire sales and deterioration of funding conditions Fire-sale price / Funding conditions Fundamental Value Optimal fire sale price Eq. fire sale price Social optimum Equilibrium size 1 Relative size of the shadow banking sector 15 / 20

75 Result 2 Shadow banking sector is too large in equilibrium Atomistic agents and incomplete markets; pecuniary externality has effect on welfare (Lorenzoni 08) Excessive regulatory arbitrage Low fire-sale price Probability of runs in the shadow banking sector is too high 16 / 20

76 Determinants of composition of financial system Consumers face the following trade-off a) Deposit at regular bank Low interest due to regulatory requirements No risk b) Deposit at shadow bank Higher interest as no regulatory cost Face prospect of panic-based run Coordination with sunspots á la Gertler and Kiyotaki (2015) 17 / 20

77 Social optimum and equilibrium EU sb EU b DI(σ) σ σ σ 1 σ Equilibrium Size Size of SB-sector 18 / 20

78 Social optimum and equilibrium EU sb σeu sb + (1 σ)eu b EU b DI(σ) σ σ σ 1 σ Social optimum Equilibrium Size Size of SB-sector 18 / 20

79 Table of contents 1 A banking model with systemic runs 2 Banks, shadow banks, and systemic liquidity crises 3 Policy implications 18 / 20

80 Policy Implications Addressing regulatory arbitrage Some regulatory arbitrage may be efficient But: regulatory arbitrage has negative externalities Possible? Macroprudential liquidity regulation Basel III? Restricting wholesale funding: shields banks from turmoil originating outside the banking sector But: allocative inefficiency and growth of shadow banking sector Welfare effects are ambiguous, may backfire Ring-fencing/shielding banks is no end in itself 19 / 20

81 Policy Implications Addressing regulatory arbitrage Some regulatory arbitrage may be efficient But: regulatory arbitrage has negative externalities Possible? Macroprudential liquidity regulation Basel III? Restricting wholesale funding: shields banks from turmoil originating outside the banking sector But: allocative inefficiency and growth of shadow banking sector Welfare effects are ambiguous, may backfire Ring-fencing/shielding banks is no end in itself Lender of last resort, Market maker of last resort Shields banking sector, but also change of composition of financial system Market maker of last resort: crowding out of regulated banks 19 / 20

82 Policy Implications Addressing regulatory arbitrage Some regulatory arbitrage may be efficient But: regulatory arbitrage has negative externalities Possible? Macroprudential liquidity regulation Basel III? Restricting wholesale funding: shields banks from turmoil originating outside the banking sector But: allocative inefficiency and growth of shadow banking sector Welfare effects are ambiguous, may backfire Ring-fencing/shielding banks is no end in itself Lender of last resort, Market maker of last resort Shields banking sector, but also change of composition of financial system Market maker of last resort: crowding out of regulated banks 19 / 20

83 Last slide Shadow banking remains important (FSB 2014) 25% of total (global) financial assets 50% of assets held by the (global) banking system and 120% of GDP on average Systemic runs: shadow banking impacts regulated banking even without contractual linkages 20 / 20

84 Last slide Shadow banking remains important (FSB 2014) 25% of total (global) financial assets 50% of assets held by the (global) banking system and 120% of GDP on average Systemic runs: shadow banking impacts regulated banking even without contractual linkages Shadow banking sector is too large in equilibrium 20 / 20

85 Last slide Shadow banking remains important (FSB 2014) 25% of total (global) financial assets 50% of assets held by the (global) banking system and 120% of GDP on average Systemic runs: shadow banking impacts regulated banking even without contractual linkages Shadow banking sector is too large in equilibrium 20 / 20