Global Banks, Financial Shocks and International Business Cycles: Evidence from Estimated Models

Transcription

1 Global Banks, Financial Shocks and Inernaional Business Cycles: Evidence from Esimaed Models Rober Kollmann () ECARES, Universié Libre de Bruxelles and CEPR February 5, 22 This paper akes a wo-counry model wih a global bank o US and Euro Area (EA) daa. The esimaion resuls (based on Bayesian mehods) sugges ha global banking srenghens he posiive inernaional ransmission of real economic disurbances. Shocks ha originae in he banking secor accoun for roughly 2% of he forecas error variance of invesmen, and abou 5% of he forecas variance of US and EA GDP. Bank shocks explain 5%-2% of he fall in US and EA real aciviy, during he Grea Recession. Key words: financial crisis, financial inermediaries, real aciviy, invesmen, Bayesian economerics. () Corresponding auhor. Addresses: R. Kollmann, ECARES, CP 4, Universié Libre de Bruxelles; 5 Av. Franklin Roosevel; B-5 Brussels, Belgium; rober_kollmann@yahoo.com. Mahias Pausian, Moneary Assessmen and Sraegy Division, Bank of England, Threadneedle Sree, London EC2R 8AH, Unied Kingdom; mahias.pausian@bankofengland.co.uk Mahias Pausian conribued o his projec in is early sages; I hank him for his advice. I am also graeful o Chrisoph Thoenissen and Alan Suherland for useful discussions, and o Sco Davis, Seve Davis, Mick Devereux, Domenico Giannone, Kalin Nikolov, Ken Wes and Raf Wouers for helpful commens and suggesions. I hank he Naional Bank of Belgium and he EU Commission for financial suppor (CEPR projec 'Poliics, Economics and Global Governance: The European Dimensions' funded by he EU Commission under is 7h Framework Programme for Research, Conrac Nr )

2 . Inroducion In he years before he recen (27-9) financial crisis, he leverage of many major financial insiuions increased seadily. The crisis revealed he fragiliy of he financial secor, and of many highly indebed non-financial firms and households, and i has riggered he sharpes global recession since he 93s. Before he crisis, srucural macro models largely absraced from financial inermediaries. The crisis has simulaed much research ha includes financial inermediaries in dynamic macro models. Given he global naure of he crisis, ha research has largely focused on open economy models; see, e.g., Davis (2), Gamber and Thoenissen (2), Devereux and Suherland (2), In Veld e al. (2), Kollmann e al. (2), Nguyen (2), Pausian and Sondergaard (2), Perri and Quadrini (2), Perri and Kalemli-Ozcan (2), Ueda (2) and van Wincoop (2) who presen open economy models wih banks. In his class of models, he presence of banks may affec he ransmission of macroeconomic and financial shocks, as he ne worh of banks is a key sae variable for real aciviy. A negaive shock o bank capial is prediced o raise he spread beween banks lending and deposi raes, and o lower lending and real aciviy; hus, shocks in one counry ha lower global banks capial can rigger a worldwide recession. So far, his new macro-banking lieraure has focused on relaively sylized, calibraed models. A key conribuion of his paper is o ake his new class of models o he daa. Specifically, we esimae a wo-counry DSGE model wih a global bank, using US and EA daa, by Bayesian economeric mehods. 2 The esimaion resuls sugges ha global banking srenghens he posiive inernaional ransmission of real economic disurbances. Shocks ha originae in he banking secor accoun for roughly 2% of he forecas error variance of invesmen, and for abou 5% of he forecas variance of US and EA GDP. Surprisingly, he conribuion of banking shocks accoun o he Grea Recession is relaively model: hese shocks accoun for abou 5%-2% of he fall in real aciviy, during he Recession. Closed economy macro models wih banks were, i.a., presened by Aikman and Pausian (26), Van den Heuvel (28), Gerler and Kiyoaki (29), Dib (29), Adrian and Shin (2), de Walque e al. (2), and Challe e al. (2). 2 Some previous papers have esimaed open economy DSGE models, bu hose sudies absraced from banks; see Adolfson (29), Jusiniano and Preson (2), de Walque e al. (25) and Peersman and Jacob (2). 2

3 Secion 2 presens he model ha we esimae. Secion 3 discusses he economeric approach. Secion 4 describes key daa feaures. Secion 5 repors he esimaion resuls. Secion 6 concludes. 2. A wo-counry world wih a global financial inermediary We esimae a wo-counry model ha builds on he heoreical se-up of Kollmann e al. (2). 3 There is a represenaive global bank. In each of he wo counries, referred o as Home and Foreign, here is a represenaive worker, an enrepreneur and a governmen. All agens are infiniely lived. The bank collecs deposis from Home and Foreign workers, and makes loans o Home and Foreign enrepreneurs. The bank faces a collaeral consrain ha ies he maximum amoun of deb ha he bank can issue o he bank s ne worh. There is a final good ha is produced by Home and Foreign enrepreneurs using local labor and capial. The good can freely be raded. I is used for consumpion and for capial accumulaion (by enrepreneurs). All markes are compeiive. Preferences and echnologies have he same srucure in boh counries. The following exposiion hus focuses on he Home counry. Foreign variables are denoed by an aserisk. 2.. Preferences, echnologies, markes The Home worker The Home worker consumes he final good, provides labor o he Home enrepreneur and invess her savings in one-period bank deposis. Her dae budge consrain is: where C + D + T = WN + DR, () W D + C and W are he worker s consumpion and he wage rae, respecively (he final good is used as numéraire). W T is a lump sum ax. N are hours worked. D + is he bank deposi held by he Home worker a he end of period. D R is he gross ineres rae on deposis, beween - and. The worker s expeced life-ime uiliy a dae is: E [ u( C ) u( D ) ( N )], (2) s D N β s + s +Ψ + + s Ψ χ = + s 3 The model here is differen in ha, i.a., governmens, and a larger number of shocks are assumed. 3

4 ux () = ( x σ )/( σ) and / + ( N) ( N η D χ = )/(+ / η), σ >, η >. Ψ > is a consan. Ψ is an exogenous sochasic ase shock ha affecs he worker s labor supply. < β < N is he subjecive discoun facor. Workers, enrepreneurs and he banker have he same subjecive discoun facor. We assume ha deposis provide uiliy o he worker (liquidiy services). This allows us o calibrae he model in such a way ha, in seady sae, he deposi rae is smaller han he lending rae, and ha workers hold deposis while enrepreneurs borrow. The Home worker maximizes (2) subjec o he period-by-budge consrain (). Tha decision problem has hese firs-order condiions: N R Eβu'( C )/ u'( C) + Ψ u'( D )/ u'( C) =, u'( C ) W =Ψ χ '( N ). D D The Home enrepreneur The Home enrepreneur accumulaes physical capial and uses capial and local labor o produce he final good. Home final good oupu, denoed Z, is produced using he α α Cobb-Douglas echnology Z= θ( K) ( N), wih < α<. K is he capial sock used a. Toal facor produciviy (TFP), θ, is an exogenous random variable ha follows an AR() process (see below). The law of moion of he Home capial sock is K+ = ( δ) K+Ξ I, where δ is he depreciaion rae of capial and I is gross invesmen. Ξ > is an exogenous shock o invesmen efficiency (see Fischer, 22, 26; Greenwood e al., 997; Jusiniano e al., 27). Gross invesmen is generaed using he final good. Le Iξ ( I / I) be he amoun of he final good needed o generae I, where I is seady sae invesmen, and ξ is an increasing, sricly convex funcion wih ξ '() =. The Home enrepreneur s period budge consrain is: LR T ξ( I/ I) WN d L θ ( K) ( N) L E E α α Δ + +Ι + + = + +, (3) where L is a one-period bank loan received by he Home enrepreneur in period -. is he gross rae on ha loan, se a -. We assume ha in period, he Home L R enrepreneur defauls by an exogenous amoun Δ on he conraced amoun LR ha L she owes he bank. T is a lump sum ax paid by he enrepreneur. d E is he E 4

5 enrepreneur s dividend income a. The enrepreneur consumes her dividend income. Her expeced lifeime uiliy a is s E E β u( d+ s), Maximizaion of ha life-ime s= uiliy subjec o (3) yields hese firs-order condiions: W ( ) K α = αθ N α, L E E R+ Eβu'( d+ )/ u'( d ) =, (4) Eβ( u'( d )/ u'( d )){ θ αk N + q ( δ)} q =, E E α α where q ξ '( I / I)/ Ξ is he marginal cos of gross invesmen a dae. / The Home governmen A dae, he Home governmen makes exogenous final good purchases G. These purchases are financed using he lump sum axes levied on he Home household, he Home enrepreneur, and by a lump sum ax levied on he global banker (see below): W E B B G= T + T + T, where T is he ax paid by he bank o he Home governmen. The oal ax burden is divided beween hese agens, according o heir shares in seady sae consumpion, i.e. T = λ G for i=w,e,b where i i i agen i s consumpion share in oal counry H consumpion. 4 i λ is a ime-invarian facor ha equals The global bank In period, he global bank receives deposis D + and D + from he Home and Foreign workers, respecively, and makes loans L + and L + o he Home and Foreign W enrepreneurs. Le D D + D and L L + L be worldwide socks of deposis and W loans a he end of period. The bank faces a capial requiremen: her dae capial L W D should no be smaller han a fracion γ of he bank s asses L. + One may W W + + view his as an implici requiremen reflecing marke pressures, or as a legal requiremen. γ is an exogenous random variable. The bank can hold less capial han he W W W required level, bu his is cosly. Le x ( L+ D+ ) γ L+ = ( γ ) W W L+ D+ denoe he 4 E.g. λ H = C/( C+ d E + d B /2), where o occur in counry H). S d s he banker s seady sae consumpion (of which 5% is assumed 5

6 W W bank s excess capial a he end of period. The bank bears a cos L φ ( x / L ) as a funcion of x, where L W is he seady sae sock of loans. φ is a convex funcion ( φ '' ) for which we assume: φ ( ) > for x < ; φ () =. Thus, for x < he bank x incurs a posiive cos. The cos is zero when he bank mees is capial requiremen. (Noe ha he above assumpions imply φ '().) A, he bank also bears an operaing cos Γ ( D + L ), where Γ> is he (consan) real marginal cos of aking deposis and W W + + making loans. The bank s period budge consrain is: where L + D R +Γ ( D + L ) + L φ({ L ( γ ) D }/ L ) + d = W W D W W W W W W B B d is he profi (dividend) generaed by he bank a. by he bank (while Δ+Δ is he bank s oal loan loss). 6 L R + D T T Δ Δ (5) W L W B B +, T B B + T is he oal ax paid The global bank acs compeiively, and hus loan raes and deposi raes are equaed across counries. The banker does no have access o oher asses, and hus she consumes her dividends. Her expeced life-ime uiliy a is: E s= s B β u( d ).The banker maximizes life-ime uiliy subjec o (5). Ruling ou Ponzi schemes, ha problem has hese firs-order condiions: D B B L B B R+ Eβu'( d+ )/ u'( d ) = Γ+ φ', R+ Eβu'( d+ )/ u'( d ) = +Γ+ ( γφ ) ', W W W wih φ' φ'({( γ ) L+ D+ }/ L ). L D These equaions imply R / ( ( ) '}/{ ' + R+ = +Γ+ γ φ Γ+ φ}. Hence, he loan rae spread is a funcion of he required capial raio and of he bank s excess capial, x. A linear approximaion of he bank s Euler equaions around x= yields: ' R L D 2 ( / W ) 2 () () ( / W + R+ Γ γφ x L Γ γφ γφ x L ). Assume ha he bank raises deposis and loans by one uni. This raises he bank s operaing cos by 2Γ unis; i also lowers he bank s excess capial by γ, which raises W W W he penaly L φ ( x / L ) by γφ( x / L ). The bank s firs order condiions imply ha ' L D W he loan rae spread R+ R+ covers he marginal cos 2 Γ γφ( x / L ). Under sric ' '' ' + s

7 convexiy of he penaly funcion, φ '' >, he marginal value of excess capial is decreasing in excess capial, and hence he spread is an increasing funcion of excess W W W bank capial. Le cr ( L+ D+ )/ L+ be he bank s capial raio a. Noe ha W x/ L cr γ. Hence, When ''() '' L D R + R + 2 Γ+ [ γφ () φ ()] γ γφ () cr. ' L D φ >, he loan rae spread R+ R+ is hence a decreasing funcion of he acual capial raio cr, and an increasing funcion of he required capial raio γ. The curvaure of he bank s penaly funcion φ ''() governs hus he sensiiviy of he loan spread o change in he bank s capial raio. A percenage poin increase in he capial raio lowers he loan spread by 4 γφ ''() percenage poins per annum. '' Forcing variables There are exogenous forcing variables: Home and Foreign TFP ( θ, θ ), invesmen efficiency (, ), Ξ Ξ governmen purchases ( G, G ) N N, labor supply shocks( Ψ, Ψ ), loan defauls (, ) Δ Δ and he required bank capial raio ( γ ). We allow for a large number of non-bank relaed shocks, o give he model he chance o explain he daa, in he absence of banking shocks. The recen empirical esimaes of DSGE models sugges ha many shocks are needed o adequaely explain macro ime daa (Smes and Wouers (27)). Following he empirical DSGE lieraure, we assume ha TFP, invesmen efficiency, governmen purchases and labor supply shocks follow univariae AR() x z processes (we refer o hese shocks are non-bank shocks): ln( x/ z) = ρ ln( z / z) + ε for 7 non-bank shock z, where ε z is a normally disribued innovaion. We allow for correlaion beween all non-bank innovaions. We assume ha loan defaul in each counry depends on lagged defaul, and on GDP in he same counry, while he required bank capial raio depends on world GDP: Δ / Y= ρ Δ / Y+ ϑ ln( Y/ Y) + ε, Δ Δ Δ Δ / Y = ρ Δ / Y + ϑ ln( Y / Y ) + ε, Δ Δ Δ

8 ln( γ / γ) = ρ ln( γ / γ) + ϑ ln( Y / Y ) + ε, γ γ W W γ where ε Δ, ε Δ γ and ε are normal whie noises. Y, Y and Y Y+ Y are Home and W Foreign GDP and world GDP respecively. The defaul innovaions ε Δ and correlaed wih each oher, bu uncorrelaed wih he innovaions o he non-bank forcing γ variables. The innovaion o he required bank capial raio ε is independen of all oher innovaions. We assume ha defaul and he required capial raio are correlaed wih non-bank forcing, via he dependence of defaul on GDP. Allowing for a more general paern of correlaion (e.g. allowing he hree banking innovaions o be correlaed wih each of he non-banking innovaions) would be compuaionally much more burdensome. As discussed below, we direcly esimae he laws of moion of he non-bank forcing variables, using empirical measures of hese variables; by conras, he laws of moion of he banking shocks are esimaed hrough he lens of he DSGE model. ε Δ are Marke clearing Marke clearing for he final good requires: Z + Z = C+ C + d + d + d + Iξ( I / I) + I ξ( I / I ) + G+ G + L φ( { L ( γ ) D }/ L ) +Γ ( L + D ). E E B W W W W W W Model soluion We ake a linear approximaion of he model equaions around a deerminisic seady sae. The soluion of he linearized model is given by s = Λ s +Λ 2ε, where s is a vecor consising of saes, conrols and forcing variables chosen (or realized) in period, expressed as in deviaion from he deerminisic seady sae. ε is he vecor of dae innovaions o he forcing variables. Λ and Λ 2 are marices whose elemens are funcions of he srucural parameers. 5 5 We use Chris Sims MATLAB proc gensys.m o solve he linearized model (Sims (2)). 8

9 3. Economeric approach The esimaion uses empirical informaion on a subse of he variables included in he vecor s. Le z be he vecor of variables used for he esimaion: z =Λ 3 s, where Λ 3 is a selecion marix. The economerician is assumed o observe he vecor z z = z + ω, where ω is a vecor of Gaussian i.i.d. measuremen errors ha has mean zero (measuremen error is independen across variables and independen of he forcing variables). Given he assumpion ha srucural innovaions and measuremen errors are Gaussian, he likelihood funcion of he daa ZT { z } =,.., T can easily be derived. See Hamilon (994, ch.3) and Schmi-Grohé and Uribe (2). Le LZ ( Θ ) denoe he likelihood funcion, where Θ is he vecor of model parameers. The model is esimaed using quarerly daa for he US and he EA, for he period 99q-2q3. The following 2 empirical series are used for esimaion: US and EA GDP, privae consumpion, invesmen, employmen, he sock of US and EA commercial bank loans (deflaed using he GDP deflaor), he loan spread of US commercial banks, and he capial raio of US commercial banks (based on Flow of Funds daa). EA loan spreads are only available for he period since 23q; as prediced by he model, he EA loan spread closely racks he US loan spread (see below). We use he US loan spread as a measure of he global loan spread, and ake he US bank capial raio as a proxy for he capial raio of he global bank. For esimaion, he loan spread and he capial raio are demeaned, while he oher empirical variables are linearly derended in log-form. (EA daa are aken from he ECB s Euro-Area-Wide-Model daa base, and from he ECB Monhly Saisical Bullein. See he Daa Appendix for a more deailed descripion of he daa.) As we use our model feaures shocks, we need o assume ha a leas one of he 2 empirical series used for esimaion is measured wih error (in order o ensure ha he model is non-singular).we esimae he model under he assumpion ha all observables conain measuremen error (assuming ha only banking variables conain measuremen error does no affec he main resuls.) We calibrae parameers whose values are unconroversial and/or pinned down by (banking) regulaions and/or average long run feaures of bank balance shees. Following T 9

10 much of he recen lieraure on he esimaion of DSGE models, we follow a Bayesian approach o esimae he remaining parameers (e.g., Orok (2), Smes and Wouers (27)). Le p( Θ ) be a prior densiy of Θ. According o Bayes law, he poserior densiy of Θ is p( Θ ZT) = L( ZT Θ) p( Θ )/ L( ZT), where LZ ( T) LZ ( T Θ) p( Θ) dθ is he marginal daa likelihood of he model. For each model varian discussed below, we repor he mode and sandard deviaion of he poserior parameer densiy, and he marginal likelihood (a measure of model fi). 3.. Calibraed parameers Calibraed echnology, preferences and governmen purchases parameers The elasiciy of final good oupu wih respec o capial is se a α=.3, which implies a 3% labor share, consisen wih US and EA daa. As is sandard in he DSGE lieraure, he (quarerly) depreciaion rae of physical capial is se a δ=.25 (consisen wih he esimaes of, e.g., Chrisiano and Eichenbaum (992)). We consider a baseline specificaion in which all agens have log uiliy, σ =, and labor supply is infiniely elasic, η= ; he same values of σ, η have widely been used in macro model (Hansen and Rogerson (995)), and hey are especially useful in he model here as hey imply ha N Ψ = WC / holds (from worker firs order condiions), which allows direc esimaion of N he labor supply shock Ψ. Our empirical measure of Home exogenous demand G is he sum of US governmen consumpion and US o counries oher han he EA. Foreign exogenous demand is defined analogously, using EA governmen consumpion and nex expors (o hird counries). During 99-2, hese measure of exogenous demand represened 4.2% of US GDP and 2.2% or EA GDP, respecively. We se he seady sae GDP raio in boh counries a he average of hose shares, 7.8%. Calibraed banking parameers The mean value of he required bank capial raio is se a γ =.7%; his corresponds o he average capial raio of US commercial banks during he sample period 99-2 (based on Flow of Funds daa). We ake as our measure of he US lending rae spread he series Commercial and Indusrial Loan Raes Spreads over inended federal funds rae

11 published by he FRB (Survey of Terms of Business Lending, able E.2). The average real loan rae and he Federal Fund rae (aken as a measure of banks marginal funding cos) were 3.44% and.279% p.a., respecively (real raes are compued using CPI inflaion). Hence, he mean loan rae spread was 2.6% p.a.. We se he seady sae deposi and loan raes in he model a hese average sample raes. (We use US rae informaion o calibrae he ineres raes, as EA raes are only available for During ha period, he mean EA lending spread (2.%) was close o he value assumed in he calibraion. The EA spread closely racked he US spread, in 23-2; correlaion:.9). 6 Using he ineres rae on shor erm Cerificaes of Deposi as a measure of banks marginal funding cos yields a spread series ha has a mean of.929% p.a. and a.75 correlaion wih he spread used in he subsequen analysis. Our baseline spread measure racks informaion on lending spreads provided by he US Senior Loan Officer Opinion Survey (SLOOS) on Bank Lending Pracices he correlaions beween our spread measure and he SLOOS Ne percenage of banks increasing spreads of loan raes over banks' cos of funds o large and middle-marke firms is.39; he correlaion wih he Ne percenage of banks increasing spreads of loan raes over banks' cos of funds o small firms is.46. Given our calibraion of seady sae ineres raes, we hus se he quarerly L subjecive discoun facor a β =.9956 (as β R =, from he enrepreneur s Euler D L equaion). The bank s Euler equaions imply R β = Γ+ φ ' and R β =+Γ+ ( γφ ) '. These wo condiions pin down he marginal operaing cos Γ and he seady sae slope of he bank s penaly funcion φ ' ( Γ=.25%, φ ' =.28%). As he calibraed bank capial raio maches he average empirical raio, we assume ha excess bank capial is zero in seady sae, W W L ( γ) = D. We se he seady sae raio of he sock of loans o annual GDP a 7% (which corresponds o he mean raio across he US and EA in 99-2). 7 This calibraion pins down he worker s 6 Our measure of he EA spread is he MFI loan rae minus he EONIA rae. 7 Mean raio of sock of oal bank credi by US Commercial banks o annual GDP: 53%. Mean raio of EA MFI loans o privae secor o annual GDP: 87%.

12 D preference parameer Ψ =.24 (ha affecs he demand for deposis), and he seady N sae value of he labor supply parameer Ψ = Calibraed shock processes. As he non-bank forcing variables can be measured direcly, we fi (by OLS) AR() equaions o he forcing variables (insead of esimaing heir laws of moion hrough he lens of he DSGE model), and we use he esimaed auoregressive parameers and innovaion sandard deviaions o calibrae he non-bank shock processes in he model. Empirically, he non-bank forcing variables are highly correlaed. I hus seems imporan o allow for correlaion beween he non-bank shocks, in he calibraion. To ensure ha he cross-correlaions of he non-bank forcing variables in he model mach empirical correlaions, we se he correlaions of he non-bank shock innovaions equal o he empirical (uncondiional) correlaions of he forcing variables. (In conras o he approach here, he empirical DSGE lieraure assumes ha all shocks are uncorrelaed, and i esimaes all shock processes via he full DSGE model; applying ha convenional esimaion sraegy approach o he model here leaves he key empirical resuls unaffeced.) The esimaed ime series parameers of he non-bank forcing variables are repored in Table (where a descripion of he empirical measures used in esimaion can also be found). TFP, invesmen efficiency (measured as he raio of he CPI o he invesmen price index), he labor supply shock (measured as he raio of wage earnings o consumpion) and exogenous demand (G) are all highly persisen (AR coefficiens in he range.8-.98). US innovaions o invesmen efficiency, he labor supply shock and exogenous demand are more volaile han he corresponding EA innovaions. Exogenous demand is negaively correlaed across he US and EA (-.3), he oher forcing variable are posiively correlaed across he US and EA. By conras, we esimae he laws of moion of he loan defaul and of he required bank capial raio, hrough he lens of he DSGE model. The moivaion for his is ha 8 In seady sae, he raio of he capial sock o annual GDP is 2.27, while he consumpions of he worker, he banker and he enrepreneur represen 55.3%,.% and 4.% of GDP, respecively. 2

13 daa on EA bank losses are only available for a shor ime span (23-); also, i is no clear how o direcly measure he implici capial requiremen reflecing marke pressures. 4. Daa plos and business cycle saisics. Figure -4 plo key macro/financial series. Figure shows linearly derended log bank loans, loan spreads (% p.a. no demeaned or derended) and loan loss raes (wrie-downs, as annualized fracions of he sock of loans), for US commercial banks and EA Moneary and Financial Insiuions (MFIs). (EA loan losses and loan spreads are only available for 23q-2q3). Loans grew srongly in he years before 28, and hen collapsed sharply, in he US and EA. Loan loss raes in he US have likewise increased srongly since 27, especially in he US (he EA loan loss rae series exhibis sizable shor-erm movemens). US and EA loan rae spreads have risen sharply since he sar of he crisis; he EA lending spread closely racks he US spread. Panel (a) of Figure 2 plos he capial raio of US commercial banks (CB), compued from Flow of Funds (FoF) daa. The capial raio exhibis relaively mild flucuaions, and mosly says in he range beween 9.5% and 2.5%. Panel (b) of he Figure plos he CB-FoF capial raio, ogeher wih he loan spread (boh series are sandardized). Excep for he period of he financial crisis (when he capial raio and he spread were above heir mean values), he wo series are negaively correlaed, as prediced by he model. While he loan rae spread rose sharply,during he crisis (as menioned above), he bank capial raio has had a fla rend since abou 25. I has been argued ha his may parly reflec accouning discreion, which has allowed banks o oversae he value of heir asses in he crisis (e.g., Huizinga and Laeven, 29). The correlaion beween he CB-FoF capial raio and he lending spread was -.46 during he period 99-27, bu close o zero (-.6) over he whole sample period (99-2). Panel (b) of he Figure also plos he (sandardized) ne percenage of banks increasing spreads of loan raes over cos of funds o large and middle marke firms (from Senior Loan Officers Opinion Survey, SLOOS). Tha series closely racks our loan spread series (correlaion.39, during 99-2). The SLOOS series is negaively correlaed wih he CB-FoF bank capial raio (-.47 for 99-27; -.2 for 99-2). Figure 3 plos he marke value of equiy o asses (a book values), for financial corporaions included in he Dow Jones sock prices index US-Banks, as repored by 3

14 Daasream. 9 Tha series exhibis a seady fall since abou 2. The marke-based capial raio fell during he firs year or he financial crisis, 28 (from.8% in 28q o 3.8% in 29q), bu recovered afer ha (o 7.8% in 2q3). Is correlaion wih he loan spread is -.5, during he sample period. Figure 4 plos linearly derended (log) GPD, privae consumpion, invesmen and employmen. In he second half of 28, hese variables conraced sharply, in he US and EA. The fall in US and EA oupu was similar (-6%) beween 27q4 and 29q4. Consumpion and invesmen fell much more sharply in he US han in he EA; e.g. US invesmen was 34% below rend in 29q2, while EA invesmen was merely 8% below rend, in he same quarer. Table 2 repors momens of HP filered key macro and banking variables, for he US and he EA (99q-2q3). Oupu volailiy is very similar in he US (.2%) and he EA (.4%). Consumpion is less volaile han GDP, while invesmen is markedly more volaile. US invesmen is almos wice as volaile as EA invesmen. In boh counries, loans are more volaile han oupu, while he loan spread is counercyclical. The variables considered in he Table are posiively correlaed across he US and EA. 5. Model esimaes 5.. Esimaed parameers We esimae he following behavioral parameers, using he Bayesian approach: (i) sensiiviy of he (annualized) loan rae spread o he bank capial raio, 4 γφ ''; (ii) curvaure of he cos of invesmen, ξ '' ; (iii) parameers of he laws of moion of loan defauls and he required bank capial raio; (iv) sandard deviaions of measuremen errors of he 2 empirical variables used for esimaion. 9 The US-Banks index only include he major financial insiuions, while Flow of Funds daa cover all firms in a given secor. The average marke-value based capial raio was 2.7% during This greaer raio reflecs he fac ha he marke value of equiy is generally greaer han is book value. Leverage measures based on book-value equiy (also available from Daasream) are much closer o FoFbased leverage measures, 7% (see Kollmann and Zeugner (2) for furher comparisons beween FoF and marke-value based capial raios. 4

15 Prior disribuions The priors on he esimaed are shown in column (). We se he mean of he prior disribuion a 4 γφ '' a.2, which implies ha a percenage poin increase in he bank capial raio lowers he loan spread by 2 basis poins p.a. This corresponds of he esimae of he sensiiviy of he loan spread (w.r.. bank capial) based on aggregae spreads and bank capial daa provided by Kollmann e al. (2), and is also in he range of esimaes based on micro-banking daa (e.g., Hubbard e al. (22), Sanos and Winon (29)). The mean of he prior disribuion of he curvaure of he invesmen cos is se a, which implies ha a % increase in invesmen raises he price of invesmen by % [add: references]. The sandard deviaions of he prior disribuions of hese parameers are se a half of he mean of he disribuion, which (for he gamma prior disribuion assumed here) implies ha a wide range of parameer values around he mean has nonnegligible mass. The prior disribuions of he sandard deviaions of innovaions o Home and Foreign loan defaul (normalized by seady sae GDP) and o he benchmark (required) bank capial raio have a mean of.5% and a sandard deviaion of.%. The prior disribuions of he AR() correlaion coefficiens of he banking shocks, and of he crosscounry correlaions of defaul have a mean.5 and sandard deviaion.. The response coefficiens of loan defauls and he required capial raio has a zero prior mean. The prior disribuion of he sandard deviaion of measuremen error has mean equal o /4 of he sandard deviaion of he corresponding empirical series, and a sandard deviaion ha is se a /5 of he mean. Poserior esimaes Columns (2) and (3) of Table 3 repor he mode of he poserior parameer disribuion, and he sandard deviaion of he poserior. The daa are informaive abou he esimaed parameers: in almos all cases, he poseriors have lower sandard deviaions han he The prior disribuion of he sandard deviaions of shock innovaions is invered gamma (IG). The IG has faer ails han he normal disribuion. Hence, a.% sandard deviaion of he prior is no very resricive. The empirical resuls are no very sensiive o he choice of priors. The mode of he poserior disribuion is he parameer vecor Θ ha maximizes he poserior disribuion; he sandard deviaions of he poserior repored here are based on a Normal approximaion of he poserior disribuion; see, e.g., Canova (27), p.34. 5

16 priors, and he poserior esimaes (modes) differ noiceably from he priors. The poserior esimae sugges ha a percenage increase in he bank capial raio is accompanied by a.45 basis poin in he annualized loan rae spread. The poserior esimaes sugges ha US loan defaul innovaion are less volaile han EA loan defaul innovaions (poserior sd..59% and.5%, respecively). The required bank capial raio undergoes sizable flucuaions (poserior sd.:.54%). US and EA loan defaul shocks are posiively correlaed (.44). Surprisingly, he responses coefficiens of loan defaul and of he required bank capial raio o GDP are slighly posiive. The poserior esimaes of he sandard deviaion of measuremen error are mosly markedly smaller han he priors. Business cycle momens implied by poserior esimaes Table 4 implies business cycle saisics (of HP filered heoreical variables) implied by he poserior parameer esimaes. The repored momens perain o counry, which we ake as he heoreical counerpar of he US (he prediced momens for he EA are similar). Column () allows for all srucural shocks, and measuremen error. In Columns (2)-(7), only one ype of shocks is considered, wihou measuremen error (he model is no re-esimaed). Column (8) repors empirical momens (from Table ). The model wih all shocks generaes momens ha are broadly in he range of he empirical momens. The prediced sandard deviaion of GDP,.6% is larger han he empirical sandard deviaion,.2% The model (wih all shocks) capures he fac ha invesmen and loans are more volaile han GDP, bu i over-predics he volailiy of consumpion (prediced relaive sandard deviaion:.2). The model also capures he cross-correlaions of he variables wih domesic GDP i correcly predics ha he loan spread is counercyclical. Finally, he model correcly predics ha he variables considered in he Table are posiively correlaed across he US and he EA he prediced cross-counry correlaion of GDP is.44, which is close o he empirical correlaion (.56). The model varians wih jus one ype of shock show ha TFP shocks and Labour supply shocks are he main drivers of GDP flucuaions (prediced sd. of GDP wih jus hese shocks:.87% and.8%, respecively). The oher shocks induce much smaller 6

17 flucuaions in GDP (defaul shocks and shocks o he required bank capial raio induce GDP sandard deviaions of.28% and.3%, respecively). Wih jus TFP shocks, jus invesmen efficiency shocks, and lus labor supply shocks GDP is negaively correlaed across counries. By conras, governmen purchases shocks and he banking shocks induce posiive cross-counry oupu correlaions. Wih jus defaul shocks, GDP, invesmen and employmen ha are (almos) perfecly correlaed across he wo counries. This is due o he fac ha a defaul by Home enrepreneurs (say) lowers he bank s capial, which riggers a rise in he world-wide loan spread and, in response o his, loans, invesmen and GDP fall in boh counries. Forecas variance decomposiion Table x [o be added] decomposes he forecas error variance of GDP, consumpion, invesmen, employmen, loans, and spreads. Banking shocks (i.e. he defaul shocks and he shocks o he required bank capial raio) accoun for abou 5% of he forecas error variance of counry and counry 2 GDP, and for abou 2% of he forecas error variance of invesmen, a horizons ranging beween and quarers. Slighly less han half of each counry s GDP forecas variance is accouned for by foreign defaul shocks. The banking shocks accoun for 99% of he forecas error variance of he loan spread and of counry 2 loans, and for 65% of he forecas error variance of counry loans (a all horizons). Decomposing hisorical ime series A decomposiion of he hisorical ime series ino conribuions of he differen shocks yields a picure ha is consisen wih he forecas error variance decomposiions. Banking shocks accoun for a small componen of he hisorical ime series on GDP and invesmen. Figure 5 shows decomposes he hisorical series ino he conribuions of he banking shocks and of US and EA non-bank shocks. The banking shocks accoun for abou %-2% of he fall in GDP during he financial crisis. The conribuion of banking shocks o he decline in EA invesmen is more sizable close o 5%. 7

18 Does global banking maer for he (inernaional) ransmission of shocks? While bank-specific shocks only play a relaively modes role for flucuaions in GDP, he exisence of (global) banks maers for he ransmission of shocks o GDP. Hence, banks do no maer primarily as a source of disurbance, bu because hey affec he ransmission mechanism. When we (essenially) eliminae he bank, by eliminaing he bank-specific shocks, by seing he seady sae loan spread a a very small number, and by seing he curvaure of he bank s penaly funcion φ '' very close o zero, so ha he loan spread is (essenially) consan (and close o zero), hen he model here behaves like a sandard inernaional RBC model wih a bonds-only-srucure (Kollmann (996)). The cross-counry correlaion of GDP and invesmen drop o -.8 and -.7 (compared o.32 and.59 in he baseline model wih banks). These prediced correlaions are obained by swiching off he bank, in he baseline mode, wihou re-esimaing he nonbanking parameers. Re-esimaing a model varian in which he bank is a veil yields much lower prediced cross-counry oupu correlaions (-.4). The marginal likelihood of he baseline model (wih banking fricion) is , while he marginal likelihood of he model wihou a bank is Thus, he model wih a global bank is overwhelmingly preferred o he model wihou bank. To be added: Resuls are robus o esimaing he ime series parameers of all forcing variables and also oher behavioral parameers (risk aversion, labor supply elasiciy ec.). 6. Conclusion Shocks originaing in he banking sysem were no a major source of flucuaions in US and EA GDP (bu hese shocks maer more for invesmen). However banking has a noiceable effec on he ransmission of oher macro shocks. Global banking leads o more synchronized naional business cycles. 2 We compue he marginal likelihood using a Laplace approximaion. 8

19 DATA APPENDIX To be added REFERENCES Adrian, T., A. Esrella and M. Shin (2), Moneary Cycles, financial cycles and he business cycle. Federal Reserve Bank of New York Saff Repor No 42. Aikman, D. and M. Pausian (26). Bank capial, asse prices and moneary policy. Bank of England Working Paper No. 35. Canova, F. (27), Mehods for Applied Macroeconomic Research. Princeon Universiy Press Davis, S. and K. Huang, (2) "Opimal Moneary Policy under Financial Secor Risk," Globalizaion and Moneary Policy Insiue Working Paper No. 85, June 2. Davis, S. (2) "The Adverse Feedback Loop and he Effecs of Risk in boh he Real and Financial Secors," Globalizaion and Moneary Policy Insiue Working Paper No. 66. Devereux, M. and A. Suherland (2) Evaluaing inernaional financial inegraion under leverage consrains. European Economic Review. 55, 3, p Gerler, M. and N. Kiyoaki (29). Financial Inermediaion and Credi Policy in Business Cycle Analsysis. De Walque, G, O. Pierrard and A. Rouabah (2). Financial (In)Sabiliy, Supervison and Liquidiy Injecion: A Dynamic General Equilibrium Approach. Economic Journal, Vol. 2(549), pp De Walque, G., F. Smes and R. Wouers (25). An Esimaed Two-Counry DSGE Model for he Euro Area and he US Economy. Working Paper. Dib, A. (2). Banks, Credi Marke Fricions, and Business Cycles. Working Papers -24, Bank of Canada. 9

20 Dib, A. (2). Capial Requiremen and Financial Fricions in Banking: Macroeconomic Implicaions. Working Papers -26, Bank of Canada. Fisher, J (22)."Technology shocks maer," Working Paper Series WP-2-4, Federal Reserve Bank of Chicago. Fisher, J. (26). The Dynamic Effecs of Neural and Invesmen-Specific Technology Shocks. Journal of Poliical Economy, Vol 4 No. 3, pp Hubbard, R., K. Kuner, and D. Palia (22). Are here bank effecs in borrowers' coss of funds? evidence from a mached sample of borrowers and banks. Journal of Business 75 (4), Huizinga, H. and L. Laeven (29). "Accouning discreion of banks during a financial crisis," CEPR Discussion Papers 738, C.E.P.R. Discussion Papers. Jusiniano, A, and B. Preson (2). Moneary Policy and Uncerainy in an Empirical Small Open Economy Model, Journal of Applied Economerics, 25(), pp Jusiniano, A, G. Primiceri and A. Tambaloi (28). Invesmen Shocks and Business Cycles. Journal of Moneary Economics, 57(2), March 2, pp Kollmann, R., Z. Enders and G. Müller (2). "Global banking and inernaional business cycles," European Economic Review, Vol. 55(3), pp Kollmann, R. and S. Zeugner (2) "Leverage as a Predicor of Real Aciviy and Volailiy". CEPR Discussion Paper Nguyen, H., "Inernaional Crisis Transmission and Asymmeric Recoveries", Working Paper, World Bank. Peersmann, G. and J. Punnoose (2) "Dissecing he Dynamics of he US Trade Balance in an Esimaed Equilibrium Model", Working Paper Gen Universiy. Perri, F. and V. Quadrini (2), "Inernaional Recessions", Working Paper, Universiy of Minnesoa and USC. Sanos, J. A. and A. Winon (28). Bank loans, bonds, and informaion monopolies across he business cycle. Journal of Finance 63 (3), 35{359. 2

21 Van den Heuvel, S. (28). 'The Welfare Cos of Bank Capial Requiremens,'' Journal of Moneary Economics, vol.55, pp Van Wincoop, E. (2), "Inernaional Conagion hrough Leveraged Financial Insiuions", Working Paper, Universiy of Virginia. 2

22 Figure. Bank loans, loan spreads, loan loss raes Bank loans (derended) Loan spreads p.a US EA US EA 22

23 Fig. 2 US Commercial Bank Capial Raios and Loan Rae Spreads (a) Capial Raio US Commercial Bank Capial Raio 99q 2q3 (b) Capial Raio, Loan Rae Spread US CB capial raio (FoF) (sandardized) Loan Spread (sandardized) Ne % of banks increasing spreads (from Loan Officer Survey) (sandardized)

24 Fig. 3 US Banks, Marke Value Based Capial Raio (Dow Jones US Banks Index) and Loan Rae Spreads (a) Capial Raio US CB Marke Value Capial Raio (b) Capial Raio, Loan Rae Spread (sandardized) US CB Marke Value Capial Raio (sandardized) Loan Rae Spread p.a. (sandardized)

25 Figure 4. Macro daa.5.5 gdp US EA consumpion US EA.2 invesmen.4 hours US EA US EA 25

26 Figure 5. Hisorical Decomposiions Hisorical decomposiion of US GDP Hisorical decomposiion of EA GDP Daa Bank Shocks NonBk Shks US NonBk Shks EA Daa Bank Shocks NonBk Shks US NonBk Shks EA Hisorical decomposiion of US Invesmen.8 Hisorical decomposiion of EA Invesmen Daa Bank Shocks NonBk Shks US NonBk Shks EA.8..2 Daa Bank Shocks NonBk Shks US NonBk Shks EA Hisorical decomposiion of US Consumpion.3 Hisorical decomposiion of EA Consumpion Daa Bank Shocks NonBk Shks US NonBk Shks EA Daa Bank Shocks NonBk Shks US NonBk Shks EA

27 Table 5 cd. Hisorical decomposiion of US Loans Hisorical decomposiion of EA Loans Daa Bank Shocks NonBk Shks US NonBk Shks EA Daa Bank Shocks NonBk Shks US NonBk Shks EA x 3 Hisorical decomposiion of US Loan Spread Hisorical decomposiion of US Bank Capial Raio Daa Bank Shocks NonBk Shks US NonBk Shks EA.2.5 Daa Bank Shocks NonBk Shks US NonBk Shks EA

28 Table. Esimaed parameers of exogenous processes (99q-2q3) (a) Auocorrelaions US TFP EA TFP US Ieff EA Ieff US G EA G US LS EA LS (b) Sandard deviaions of innovaions (diagonal) and cross-correlaions (off-diagonal elemens) US TFP EA TFP US Ieff EA Ieff US G EA G US LS EA LS US TFP.48% EA TFP.48% US Ieff.64% EA Ieff.3% US G 2.6% EA G.43% US LS.8%.47 EA LS.65% Noe: The Table repors he ime series parameers if of linearly derended logged forcing variables. TFP: oal facor produciviy; Ieff: invesmen effiency; G: governmen consumpion plus ne expors o hird counries; LS: Labor supply shock Log TFP is esimaed as ln( Y).7ln( N) where Y and N are GDP and employmen, respecively. Our esimae of invesmen efficiency is he raio of he CPI o he invesmen deflaor. N Our esimae of he labor supply shock is Ψ = WC /, where W is wage earnings per employee, while C is per capia consumpion. US G is he sum of governmen consumpion and of US ne expors o counries oher han he EA (EA G is defined analogously). 28

29 Table 2. Hisorical business cycle saisics, 99q-2q3 US EA Sandard deviaion (in%) GDP (Y).2.4 Relaive sandard deviaions (sd(x)/sd(gdp)) Consumpion Invesmen Employmen.3.62 Loans Loan spread (p.a.).7.33 Correlaion wih domesic GDP Consumpion Invesmen Employmen Loans Loan spread (p.a.) Cross-counry correlaions GDP.56 Consumpion.39 Invesmen.45 Employmen.53 Loans.64 Loan spread (p.a.).79 Noe: momens of HP filered series are shown (GDP, consumpion, invesmen, employmen and loans were logged before applying he filer). Sample period: 99q- 2q3 (excep for EA loan spread: 997q3-2q3). 29

30 Table 3. Prior and poserior disribuion of parameers baseline specificaion Prior disribuion Poserior disribuion Parameer Mean Sd Disrib Mode Sd () (2) (3) (4) (5) Behavioral parameers 4 γφ ''.2. G.45. ξ ''..5 G.2. Banking shocks Sandard deviaiions (%) and cross-correlaions σ Δ.5. IG.59. σ Δ.5. IG.5. γ σ.5. IG.54. Corr ε ε Δ Δ (, ).5. B.44. AR coefficiens ρ Δ.5. B.63. ρ Δ.5. B.54. γ ρ.5. B.79. Responses o GDP ϑ Δ.. N.2. γ ϑ.. N.6. Sandard deviaions (%) of measuremen errors GDP US.79.6 IG.39.4 GDP EA.54. IG.35.4 I US IG I EA IG.89.2 C US.78.6 IG.68.8 C EA.44.9 IG.37.5 N US.47.9 IG.29.4 N EA.45.9 IG.28.4 Loan rae spread.3. IG.2. Loans US.85.7 IG.83.7 Loans EA..22 IG.44.5 Bank cap. raio.2.4 IG.34. Noes: Columns () and (2) shows he mean and sandard deviaions of he prior disribuion for model parameers. In Column (3), B,G,IG,N indicae he disribuion of he prior (B: Bea; G: Gamma; IG: Invered Gamma; N: Normal). Column (4) repors he mode of he poserior disribuion (i.e. he parameer vecor Θ ha maximizes he poserior disribuion); Column (5) repors sandard deviaions of he poserior disribuion (based on a Normal approximaion of he poserior disribuion; see Canova (27, p.34)). 3

31 Table 4. Counry ( US ) business cycle saisics implied by poserior mode of model parameers baseline esimaion Jus shocks o: All shocks TFP Inv.Eff. G LabS Defaul Reqd.BkCap DATA () (2) (3) (4) (5) (6) (7) (8) Sandard deviaion (in%) GDP (Y) Relaive sandard deviaions (sd(x)/sd(gdp)) Consumpion Invesmen Employmen Loans Loan spread Correlaion wih domesic GDP Consumpion Invesmen Employmen Loans Loan spread Cross-counry correlaions GDP Consumpion Invesmen Employmen Loans Noe: The Table shows momens of HP filered model variables, for he mode poserior esimae of he model parameers. The momens perain o counry, which we ake as he heoreical counerpar of he US. Column () allows for all srucural shocks, and measuremen error. In Columns (2)-(7), only one ype of shocks is considered, wihou measuremen error (he model is no re-esimaed). Column (8) repors US empirical momens (from Table ). Col. (2): jus TFP shocks ( θ, θ ) ; Col. (3): jus shocks o invesmen efficiency ( Ξ, Ξ ) ; Col. (4): jus shocks o governmen purchases ( G, G ) ; Col.(5): jus labor N N supply shocks ( Ψ, Ψ ); Col.(6): jus loan defaul shocks ( Δ, Δ ) ; Col. (7): jus shock o benchmark bank capial raio ( γ ). 3