Credit Supply and the Housing Boom

Transcription

1 Credi Supply and he Housing Boom Alejandro Jusiniano Federal Reserve Bank of Chicago & PSE Andrea Tambaloi Federal Reserve Bank of New York Giorgio Primiceri Norhwesern Universiy & NBER Workshop on Counerfacual Analysis for Policy Evaluaion USC, Nov Views expressed are solely hose of he auhors and need no represen hose of he FRB of Chicago, New York or any oher person associaed wih he Federal Reserve Sysem

2 The US economy in he 2000s: Four sylized facs 1 Decline in morgage raes

3 1. Real morgage raes

4 The US economy in he 2000s: Four sylized facs 1 Decline in morgage raes 2 Unprecedened boom bus cycle in house prices

5 2. Real house price

6 The US economy in he 2000s: Four sylized facs 1 Decline in morgage raes 2 Unprecedened boom bus cycle in house prices 3 Massive HH deb accumulaion, and hen deleveraging

7 3. Household deb

8 The US economy in he 2000s: Four sylized facs 1 Decline in morgage raes 2 Unprecedened boom bus cycle in house prices 3 Massive HH deb accumulaion, and hen deleveraging 4 Deb o collaeral raio consan, and hen spikes

9 4. Deb o collaeral raio

10 The US economy in he 2000s: Four sylized facs 1 Decline in morgage raes 2 Unprecedened boom bus cycle in house prices 3 Massive HH deb accumulaion, and hen deleveraging 4 Deb o collaeral raio consan, and hen spikes

11 This Talk and Preview of Resuls Sylized model wih housing and wo consrains Collaeral/borrowing consrain: demand for credi Lending consrain: supply of credi

12 This Talk and Preview of Resuls Sylized model wih housing and wo consrains Collaeral/borrowing consrain: demand for credi Lending consrain: supply of credi Loosening of lending consrain explains large fracion of he boom in house prices and deb also lower morgage raes and consan deb o collaeral raio

13 This Talk and Preview of Resuls Sylized model wih housing and wo consrains Collaeral/borrowing consrain: demand for credi Lending consrain: supply of credi Loosening of lending consrain explains large fracion of he boom in house prices and deb also lower morgage raes and consan deb o collaeral raio Loosening of borrowing consrain alone no an imporan driving force of he boom, small effecs on prices a odds wih he behavior of morgage raes, deb o collaeral raio

14 This Talk and Preview of Resuls Sylized model wih housing and wo consrains Collaeral/borrowing consrain: demand for credi Lending consrain: supply of credi Loosening of lending consrain explains large fracion of he boom in house prices and deb also lower morgage raes and consan deb o collaeral raio Loosening of borrowing consrain alone no an imporan driving force of he boom, small effecs on prices a odds wih he behavior of morgage raes, deb o collaeral raio Key: ineracion beween borrowing and lending consrains driven by expansion in credi supply

15 Some lieraure Imporance of borrowing consrains in he boom bus of he 2000s Boom: Favilukis, Ludvigson, Van Nieuwerburgh (2013), Boz and Mendoza (2012), Garriga, Manuelli and Perala Alva (2012), Midrigan and Philippon (2011) Bus: Guerrieri and Lorenzoni (2012), Eggersson and Krugman (2012), Hall (2012) We focus on barriers o lending and heir ineracion wih collaeral consrains Consrains on composiion of balance shee of inermediaries Gerler and Kiyoaki (2010), Adrian and Shin (2010), Adrian and Boyarchenko (2012 and 2013), Dewacher and Wouers (2012), He and Krishnamury (2013), Brunnermeier and Sannikov (2014), ec We focus on he link beween he availabiliy of credi, household deb and house prices

16 Ouline Model Graphical Analysis Parameerizaion Quaniaive resuls Loosening of lending consrain Relaxaion of borrowing consrain Direcions for fuure research

17 Simples model

18 Simples model Build on Kiyoaki and Moore (1997) Iacoviello (2005) Campbell and Hercowiz (2006) 2 groups of households Paien Lenders Impaien Borrowers

19 Simples model Build on Kiyoaki and Moore (1997) Iacoviello (2005) Campbell and Hercowiz (2006) 2 groups of households Paien Lenders Impaien Borrowers No producion income is exogenous

20 Simples model Build on Kiyoaki and Moore (1997) Iacoviello (2005) Campbell and Hercowiz (2006) 2 groups of households Paien Lenders Impaien Borrowers No producion income is exogenous Fixed supply of (new) houses

21 The problem of he borrowers max E 0 b 0 uc b, vh b, c b, p h b, 1 1 h b, R 1 D b, 1 y b, D b,

22 The problem of he borrowers max E 0 b 0 uc b, vh b, c b, p h b, 1 1 h b, R 1 D b, 1 y b, D b, Borrowing is limied by a collaeral (borrowing) consrain D b, p h b, 1

23 The problem of he borrowers max E 0 b 0 uc b, vh b, c b, p h b, 1 1 h b, R 1 D b, 1 y b, D b, Borrowing is limied by a collaeral (borrowing) consrain D b, p h b, 1 Muliplier on he borrowing consrain: μ 0

24 The problem of he borrowers max E 0 b 0 uc b, vh b, c b, p h b, 1 1 h b, R 1 D b, 1 y b, D b, Borrowing is limied by a collaeral (borrowing) consrain D b, p h b, 1 Muliplier on he borrowing consrain: μ 0 Maximum Loan o Value (LTV): p D h b, b, 1

25 The problem of he lenders ( β l > β b ) max E 0 l 0 uc l, vh l, c l, p h l, 1 1 h l, R 1 D l, 1 y l, D l,

26 The problem of he lenders ( β l > β b ) max E 0 l 0 uc l, vh l, c l, p h l, 1 1 h l, R 1 D l, 1 y l, D l, Morgage lending is limied by a lending consrain D l, L

27 The lending consrain D l, L In reduced form capures all facors hampering he free flow of funds from he savers o morgage financing

28 The lending consrain D l, L In reduced form capures all facors hampering he free flow of funds from he savers o morgage financing Implici or explici, regulaory, insiuional and echnological consrains on morgage lending Money marke funds, pension funds and insurance companies are resriced by regulaions o holding only he safes securiies Leverage resricion, regulaory or risk capial requiremens on financial inermediaries

29 The lending consrain D l, L In reduced form capures all facors hampering he free flow of funds from he savers o morgage financing Implici or explici, regulaory, insiuional and echnological consrains on morgage lending Money marke funds, pension funds and insurance companies are resriced by regulaions o holding only he safes securiies Leverage resricion, regulaory or risk capial requiremens on financial inermediaries Can be derived as limiing case of model wih financial inermediaries and cos of issuing equiy

30 Two addiional simplifying assumpions

31 Two addiional simplifying assumpions Rigid demand for houses by he lenders Implicaions Borrowers are marginal buyers of houses

32 Two addiional simplifying assumpions Implicaions Borrowers are marginal buyers of houses 1, 1,, 1, 1 ) '( ) '( 1 1 c h b b b b p mrs c u c u p Rigid demand for houses by he lenders

33 Two addiional simplifying assumpions Implicaions Borrowers are marginal buyers of houses 1, 1,, 1, 1 ) '( ) '( 1 1 c h b b b b p mrs c u c u p Rigid demand for houses by he lenders

34 Two addiional simplifying assumpions Implicaions Borrowers are marginal buyers of houses 1, 1,, 1, 1 ) '( ) '( 1 1 c h b b b b p mrs c u c u p Rigid demand for houses by he lenders ighness of borrowing consrain

35 Two addiional simplifying assumpions Implicaions Borrowers are marginal buyers of houses 1, 1,, 1, 1 ) '( ) '( 1 1 c h b b b b p mrs c u c u p Rigid demand for houses by he lenders Max. Loan o Value Raio (LTV)

36 Two addiional simplifying assumpions Implicaions Borrowers are marginal buyers of houses 1, 1,, 1, 1 ) '( ) '( 1 1 c h b b b b p mrs c u c u p Rigid demand for houses by he lenders Linear uiliy in consumpion

37 Two addiional simplifying assumpions Rigid demand for houses by he lenders Linear uiliy in consumpion p b 1 mrs h, c 1 p 1 Implicaions Borrowers are marginal buyers of houses Variaion in house prices only due o variaion in discouning hrough Exogenous changes in maximum LTV Endogenous changes in ighness of borrowing consrain

38 Ineracion of borrowing and lending consrains Borrowing consrain : D p h b, b, 1 (1) Lending consrain : - D l, L D b, L (2) Marke clearing : D b, D l, 0

39 Ineracion of borrowing and lending consrains Borrowing consrain : D p h b, b, 1 (1) Lending consrain : - D l, L D b, L (2) Marke clearing : D b, D l, 0

40 Ineracion of borrowing and lending consrains Borrowing consrain : D p b, h b, 1 (1) Lending consrain : - D l, L D b, L (2) Marke clearing : D b, D l, 0 Which consrain binds, (1), (2) or boh is exogenous L and endogenous p b mrs 1 1 p 1

41 Graphical Analysis 1. Demand and Supply 2. Locus of equilibrium house price and ineres rae as vary L Boh cases keep θ consan hree regions depending on which consrains bind

42 Demand of funds and he borrowing consrain R 1/β b Demand of funds 1/β l D b

43 Case 1: Lending consrain only binds R 1/β b Demand of funds 1/β l Supply of funds L 1 D b

44 Case 1: Lending consrain only binds R 1/β b Demand of funds - R = 1/β b - Borrowing consrain no binding - p b mrs 1 p 1 1/β l Supply of funds L 1 D b

45 Case 2: Boh consrains bind R 1/β b Demand of funds -Noe: equilibrium is unique 1/β l Supply of funds L 2 ph b D b

46 Case 3: Borrowing consrain only binds R 1/β b Demand of funds 1/β l Supply of funds L 3 D b

47 Case 3: Borrowing consrain only binds R 1/β b Demand of funds - R = 1/β l - Collaeral consrain binding (μ>0) b p mrs p 1 1/β l Supply of funds L 3 D b

48 Graphical Analysis 1. Demand and Supply 2. Locus of equilibrium house price and ineres rae as vary L keeping θ consan

49 Equilibrium price as vary p p( ) p p h b p ( ) h b L

50 Equilibrium price as vary p p( ) Lending Consrain p Only Binds p h b p ( ) h b L

51 Equilibrium price as vary p p( ) p Lending and Borrowing Consrain Bind p h b p ( ) h b L

52 Equilibrium price as vary p p( ) Borrowing Consrain p Only Binds p h b p ( ) h b L

53 Equilibrium Ineres Rae as vary R 1 b 1 l p h _ b b p ( ) h L

54 Ouline Model Parameerizaion Quaniaive resuls Expansion in credi supply Loosening of collaeral requiremens

55 Calibraion and Quaniaive Experimens Calibrae parameers o mach Micro daa: Survey of Consumer Finances Triennial deailed survey daa of US households balance shee

56 Calibraion and Quaniaive Experimens Calibrae parameers o mach Micro daa: Survey of Consumer Finances Triennial deailed survey daa of US households balance shee Exend he model o allow for amorizaion of deb

57 Quarerly calibraion Parameer Value Source/Targe Discoun facor borrower (β b ) % real morgage rae Discoun facor lender (β l ) % decline in real morgage raes ~ Krusell and Smih (1998) ~ Carroll e al. (2013) Depreciaion (δ) Fixed Asse Tables Maximum LTV (θ) 0.80 Amorizaion (ρ) Median LTV of new or recenly refinanced morgages of liquidiy consrained HHs in he SCF Evidence from Duca e al. (2012) Collaeral consrain close o binding Morgage-o-RE raio of liquidiy consrained HHs in he SCF (43%)

58 Calibraion and Quaniaive Experimens Compare wo experimens Experimen 1: gradual loosening of lending consrain Experimen 2: gradual relaxaion of borrowing consrain in a model wihou lending consrains

59 Experimen 1: Loosening of lending consrain Saring in 2000, gradual relaxaion of morgage lending consrain

60 Experimen 1: Loosening of lending consrain Saring in 2000, gradual relaxaion of morgage lending consrain Specacular growh in securiizaion markes (Brunnermeir 2009) pension and money marke funds gain access o morgage lending reduces banks capial requiremens and dependence on liquidiy for morgage lending (Louskina and Srahan 2009)

61 Experimen 1: Loosening of lending consrain Saring in 2000, gradual relaxaion of morgage lending consrain Specacular growh in securiizaion markes (Brunnermeir 2009) pension and money marke funds gain access o morgage lending reduces banks capial requiremens and dependence on liquidiy for morgage lending (Louskina and Srahan 2009) Regulaory arbirage (Acharya and Richardson, 2009) SIVs allow banks o move asses off balance shees, no coun hem owards risk weighed capial (Acharya and Schnabel, 2009) morgage companies of BHC (Demanyak and Louskina, 2013)

62 Experimen 1: Loosening of lending consrain Saring in 2000, gradual relaxaion of morgage lending consrain Specacular growh in securiizaion markes (Brunnermeir 2009) pension and money marke funds gain access o morgage lending reduces banks capial requiremens and dependence on liquidiy for morgage lending (Louskina and Srahan 2009) Regulaory arbirage (Acharya and Richardson, 2009) SIVs allow banks o move asses off balance shees, no coun hem owards risk weighed capial (Acharya and Schnabel, 2009) morgage companies of BHC (Demanyak and Louskina, 2013) Foreign capial inflows: Banking and Savings Glu (Bernanke e al. 2011) Saving Glu: Governmen bonds & Agency MBSs Banking Glu: European purchases of privae label ABS/MBSs

63 Experimen 1: Loosening of lending consrain Saring in 2000, gradual relaxaion of morgage lending consrain Specacular growh in securiizaion markes (Brunnermeir 2009) pension and money marke funds gain access o morgage lending reduces banks capial requiremens and dependence on liquidiy for morgage lending (Louskina and Srahan 2009) Regulaory arbirage (Acharya and Richardson, 2009) SIVs allow banks o move asses off balance shees, no coun hem owards risk weighed capial (Acharya and Schnabel, 2009) morgage companies of BHC (Demanyak and Louskina, 2013) Foreign capial inflows: Banking and Savings Glu (Bernanke e al. 2011) Saving Glu: Governmen bonds & Agency MBSs Banking Glu: European purchases of privae label ABS/MBSs Experimen imed o complee he ransiion in 2006

64 Experimen 1: Loosening of lending consrain L relaive o income 1990s: only lending consrain binds

65 Experimen 1: Loosening of lending consrain L relaive o income : lending and borrowing consrains bind

66 Experimen 1: Loosening of lending consrain

67 Experimen 2: Loosening of collaeral consrain Sandard model wihou lending consrains

68 Experimen 2: Loosening of collaeral consrain Sandard model wihou lending consrains Simulae he effecs of a gradual relaxaion of borrowing consrain Designed o mach he increase in deb of experimen 1 requires gradually increasing θ from 0.8 o 1.02

69 Experimen 2: Loosening of collaeral consrain (θ)

70 Experimen 2: Loosening of collaeral consrain Implicaions of loosening θ in a model wihou lending consrains a odds wih daa house prices barely move ineres raes do no fall deb o collaeral raio no consan Same in larger scale models (JPT, 2014)

71 Conclusions Loosening of lending consrain implies ouward shif in supply of credi and can explain large fracion of boom in house prices and deb decline in morgage raes consan deb o collaeral raio

72 Conclusions Loosening of lending consrain implies ouward shif in supply of credi and can explain large fracion of boom in house prices and deb decline in morgage raes consan deb o collaeral raio Relaxaion of borrowing consrain alone no an imporan driving force a odds wih sylized facs

73 Conclusions Loosening of lending consrain implies ouward shif in supply of credi and can explain large fracion of boom in house prices and deb decline in morgage raes consan deb o collaeral raio Relaxaion of borrowing consrain alone no an imporan driving force a odds wih sylized facs Ineracion of borrowing and lending consrain generaes rich dynamics for deb and house prices E.g. Loosening of borrowing consrain can produce decline in house prices

74 Direcions for fuure research Robusness wih a larger scale model Curvaure Producion of houses (JPT, JIE 2014) Micro evidence on morgage ineres raes during he boom Macroprudenial policy and lending consrains

75 THANKS FOR LISTENING

76 The Evoluion of our Research Agenda Household Leveraging and Deleveraging (forhcoming) Loosening (and reversal) of borrowing consrain alone canno explain housing boom (and bus) Medium scale DSGE, borrowing consrains only, closed economy

77 The Evoluion of our Research Agenda Household Leveraging and Deleveraging (forhcoming) Loosening (and reversal) of borrowing consrain alone canno explain housing boom (and bus) Medium scale DSGE, borrowing consrains only, closed economy The Effecs of he Banking and Saving Glu (JIE, 2014) Massive inflows of gross and ne foreign capial can explain 1/3 of increase in house prices and rise in HH deb during boom Large scale DSGE, borrowing consrains only, large open economy

78 The Evoluion of our Research Agenda Household Leveraging and Deleveraging (forhcoming) Loosening (and reversal) of borrowing consrain alone canno explain housing boom (and bus) Medium scale DSGE, borrowing consrains only, closed economy The Effecs of he Banking and Saving Glu (JIE, 2014) Massive inflows of gross and ne foreign capial can explain 1/3 of increase in house prices and rise in HH deb during boom Large scale DSGE, borrowing consrains only, large open economy Credi Supply and he Housing Boom Shif focus from consrains on borrowing (demand) o lending (supply) Sylized model wih lending consrains

79 Conclusions Wha if collaeral requiremens depend on some of he facors shifing also he supply of funds? No a propery of he sandard model of collaeralized deb Even in his case, our resuls sugges ha credi supply has o shif more han demand o accoun for he boom phase of he cycle

80 The sory in words The apparen safey of he financial secor s collecive balance shee was aribuable o he fac ha he bigges global banks had amassed vas quaniies of AAA raed ( safe ) ranches backed by residenial morgages. These asses had hisorically been safer han similarly raed corporae loans. This was he principal reason behind heir lower risk charge (by a facor of five) under he Basel capial requiremens ha were in place for European banks, for allowing he US commercial banks o park hese in off balance shee vehicles wih lile capial, and leing invesmen banks use inernal models for risk managemen ha largely ignored he ail risk of a secular housing collapse.

81 Risk weighed capial raio In he Unied Saes, deposiory insiuions are subjec o riskbased capial guidelines issued by he Fed. These guidelines are used o evaluae capial adequacy based primarily on he perceived credi risk associaed wih balance shee asses, as well as cerain off balance shee exposures such as unfunded loan commimens, leers of credi, and derivaives and foreign exchange conracs. The risk based capial guidelines are supplemened by a leverage raio requiremen To be adequaely (well) capialized under federal bank regulaory agency definiions, a bank holding company mus have a Tier 1 capial raio of a leas 4% (6%), a combined Tier 1 and Tier 2 capial raio of a leas 8% (10%), and a leverage raio of a leas 4% (5%)

82 Non agency MBSs (Mayer)

83 Share of securiized morgages (Krainer and Laderman, 2011)

84 Securiizaion raes (Simkovic, 2013)

85 Securiizaion over ime Value of ousanding RMBSs relaive o GDP percen

86 Morgage spreads (1 year ARM minus he FFR)

87 Senior Loan Officer Opinion Survey 2007-Q1

88 Experimen 2: Loosening of collaeral requiremens

89 Experimen 2: Loosening of collaeral requiremens (ρ)