Does a Currency Union Need a Capital Market Union? Joseba Martinez Thomas Philippon NYU, NBER and CEPR Toward a Genuine Economic and Monetary Union Oesterreichische Nationalbank September 215
Motivation Eurozone crisis: sudden stop in a currency union Paul de Grauwe (212) The situation of Spain is reminiscent of the situation of emerging economies that have to borrow in a foreign currency...they can suddenly be confronted with a sudden stopwhen capital inows suddenly stop leading to a liquidity crisis Reversal of intra-emu capital ows very large country specic borrowing spreads: r t r j t, j {spain,germany,greece,...} Policy response: whatever it takes, Banking Union
Our Goal Does new EMU nancial architecture provide enough insurance against: 1. Deleveraging shocks? Salient feature of nancial crisis 2. Other macro shocks? Competitiveness Productivity Departure: optimistic view of banking union Assume credit markets function perfectly Is this enough?
Our model of the EZ economy Two countries, each populated by borrowers and savers Share a currency, trade goods with each other Home bias: consumption tilted towards domestic goods No labor mobility across them New-Keynesian model: Labor market frictions sticky wages Product market imperfections rms earn rents
Our model of the EZ economy Borrowers and savers interact in union-wide capital markets We compare three versions of this economy: Complete markets (perfect insurance) Banking union: agents can save and borrow at a union-wide interest rate Capital markets union: banking union + diversied claims to prots
Preferences and Demographics Two countries: home and foreign A country? Segmented labor markets, home bias in consumption Two types of households i = b,s, borrower and saver, β b < β s, fraction χ of borrowers E t t= β t i U (C i,t,n i,t ), for i = b,s Consumption of home and foreign goods with home bias (α t <.5) [ C i,t = (1 α t ) η 1 ( ) η 1 η C h,i,t 1 η + αt ] η ( ) η 1 η 1 Cf η,i,t
Budget and Borrowing Constraints Impatient agent budget constraint: nominal debt Borrowing constraint: P t C b,t = B t+1 R t + W t N t T t B t B t+1 B t+1 Assume impatient enough (β b low) s.t.: B t+1 = B t+1 Deleveraging shock: B t+1 stochastic
Budget Constraints and Market Clearing Savers: nominal debt and claims to prots S t + W t N t T t + ϕ Clearing bond markets Π t R t 1 χ + (1 ϕ ) Π t 1 χ = P tc s,t + S t+1 (1 χ)s t+1 + (1 χ )S t+1 = χb t+1 + χ B t+1 Clearing stock markets: ϕ h + ϕh = 1 ϕ f + ϕf = 1
Firms Intermediate goods producers Monopolistically competitive Supply intermediates to domestic nal good producer Technology y j,t = a t n j,t Marginal cost = W t (normalize a = 1) Monopolistically competitive: p j,t = µ t W t Rents: d t = (µ t 1)W t N t
Firms Domestic nal good producers Perfectly competitive Buys intermediates from domestic producers and bundles them together [ 1 ] εt C h = c (j) ε t 1 εt 1 εt dj ε t ε t 1 µ t Intermediate producers all produce same quantity and charge same price, p h,t = µ t W t
Consumption Basket Consumers in home and foreign buy home and foreign nal goods ( ph ) η C h = (1 α t ) C P ( pf ) η C f = α t C P CPI (home) P = [(1 α t )(p h ) 1 η + α t (p f ) 1 η] 1 η 1 LOOP: p h = p h, p f = p f
Wages Wages follow a simple Phillips Curve W t = W t 1 (1 + κ (N t N )) Model set up such that in equilibrium W s,t N s,t = W b,t N b,t
Taylor Rule Taylor rule π t = P t /P t 1 (CPI ination) R t = R ss (( Yt Y ss )( )) Y φy (( )( )) t πt π φπ t Yss π ss πss
Market clearing in goods and equilibrium Market clearing in goods N t = Y t p h,t = (1 α t ) + α t ( ph P t ( ph,t P t ) η ((1 χ)c s,t + χc b,t) ) η ( (1 χ)c s,t + χc b,t) + Gt Equilibrium: allocations {C s,t,c b,t,n t,c s,t,c b,t,n t } and prices {p h,t,p f,t,w t,w t,r t } such that: 1. Borrowing constraints satised 2. Savers on their Euler 3. Markets clear Complete markets (λ is Lagrange multiplier on savers' budget constraint): λ t = λ t
Experiments Deleveraging shock: persistent reduction in domestic borrowing limit Technology shock: home bias parameter Competitiveness shock: markups IRFs comparing response to each shock in linear approximation around deterministic steady state Bond economy: banking union, all prots stay domestic Diversied stocks: capital markets union, rents shared equally between savers (ϕ = ϕ =.5) Complete markets: λ = λ
Deleveraging shock -.9 B.1 n.1 n * -.95-1 1 -.1 n+n *.2 -.1 c s.2 c s * -1 -.2 -.2 2 c b.5 c b *.2 R -2 -.5 -.2 Bond Economy Diversified Stocks Complete Markets
Deleveraging shock.1 y.1 y * -.4 s -.6 -.1 -.2 s * -.1 p h -.8.1 p h * -.4 -.5.5 -.6 -.1.1 Π.1 Π * -.1 -.1 Bond Economy Diversified Stocks Complete Markets
Deleveraging shock - no accomodation -.9 B.5 n.2 n * -.95-1.5 -.5 n+n *.2 -.2 c s.2 c s *.1 -.5 -.2 2 c b.5 c b *.2 R -2 -.5 -.2 Bond Economy Low Response
Deleveraging shock - no accomodation.5 y.2 y * -.65 s -.7 -.5 -.2 -.2 s * -.75 p h.2 p h * -.25 -.2 -.3 -.4 -.2.5 Π.2 Π * -.5 -.2 Bond Economy Low Response
Implications Banking union don't need to worry about deleveraging shocks? Better risk sharing among savers doesn't improve welfare of borrowers Monetary policy not very eective (as we know) Policies that redistribute towards borrowers?
Debt Restructuring Now supose that borrowers can default η = amount of deleveraging achieved by default Ex-post ecient: need to cut spending less But who bears the cost of default? domestic savers? foreign savers? fraction ω Example: banks make loans to households, bank equity is held by foreign savers capital market integration of bank equity
Impulse response with default, ω =.5 5 B 5 def 1 n 5 5 5 1 1 n * 1 n+n * 1 pc s 1 1 1 1 p * c s * 1 pc b 2 p * c b * 1 1 2 η=1 η=.5 η=
Other types of shocks When does the frictionless bond economy provide enough risk-sharing? Shocks don't change relative wealth "too much" distort relative prices "too much" These caveats do not apply to technology or competitiveness shocks
Impulse response to preference shock (α increase) -.9 α * n.4 n * -.95 -.2.2-1 1 -.4 n+n *.5 c s.5 c s * -1 -.5 -.5 -.2 c b.6 c b * 1 R -.4.4 -.6.2-1 Bond Economy Diversified Stocks Complete Markets
Impulse response to preference shock (α increase) -.2 y.6 y * s -.4.4-1 -.6 2 s *.2-2 p h.4 p h * 1 -.2.2 -.4 -.2 Π.6 Π * -.4.4 -.6.2 Bond Economy Diversified Stocks Complete Markets
Impulse response to markup shock (µ increase) 1 µ n.2 n *.95-1.1.9 n+n * -2 1 c s.5 c s * -.5-1 -1 -.5-1 c b.2 c b * 1-3 -2 R -1.5-2.2-2 -.2-2.4 Bond Economy Diversified Stocks Complete Markets
Impulse response to markup shock (µ increase) y.2 y * 2 s -.1.1 -.2 2 s * 1-2 p h.2 p h *.5.1-2 1.6 Π.2 Π * 1.4.1 1.2 Bond Economy Diversified Stocks Complete Markets
Conclusions Perfect banking union emulates full insurance with respect to deleveraging shocks Sharing of other types of shocks requires more capital markets integration Capital union improves on banking union in case of productivity shocks Debt restructuring can be ex-post ecient Integration of bank equity ownership