International Prices and Exchange Rates Gita Gopinath

Size: px

Start display at page:

Download "International Prices and Exchange Rates Gita Gopinath"

Dustin Manning
6 years ago
Views:

1 International Prices and Exchange Rates Gita Gopinath Nominal and Real Exchange Rates Exchange-rate pass-through and expenditure switching Currency Wars, Fear of Floating 1 / 72

2 Non-neutrality of Nominal Exchange Rates Apr-11 Oct-11 May-12 Nov-12 Jun-13 Jan-14 Jul-14 Feb-15 Aug-15 Mar-16 Dollar Trade Weighted NER (Index) Euros per Dollar 2 / 72

3 International Spillovers Nominal Rigidities 1 First generation ( Consensus View ): Fleming (1962), Mundell (1963), Dornbusch (1976), Svenson & van Wijnbergen (1989), Obstfeld & Rogoff (1995) 3 / 72

4 International Spillovers Nominal Rigidities 1 First generation ( Consensus View ): Fleming (1962), Mundell (1963), Dornbusch (1976), Svenson & van Wijnbergen (1989), Obstfeld & Rogoff (1995) Prices rigid in the producer s currency (PCP) 3 / 72

5 International Spillovers Nominal Rigidities 1 First generation ( Consensus View ): Fleming (1962), Mundell (1963), Dornbusch (1976), Svenson & van Wijnbergen (1989), Obstfeld & Rogoff (1995) Prices rigid in the producer s currency (PCP) Depreciations (appreciations) are inflationary (deflationary) P M = E h/f Pf f E h/f, P M 3 / 72

6 International Spillovers Nominal Rigidities 1 First generation ( Consensus View ): Fleming (1962), Mundell (1963), Dornbusch (1976), Svenson & van Wijnbergen (1989), Obstfeld & Rogoff (1995) Prices rigid in the producer s currency (PCP) Depreciations (appreciations) are inflationary (deflationary) P M = E h/f Pf f E h/f, P M Depreciations (appreciations) deteriorate (improve) terms of trade. TOT P X P M = P h h E h/f Pf f E h/f, TOT 3 / 72

7 International Spillovers Nominal Rigidities 1 First generation ( Consensus View ): Fleming (1962), Mundell (1963), Dornbusch (1976), Svenson & van Wijnbergen (1989), Obstfeld & Rogoff (1995) Prices rigid in the producer s currency (PCP) Depreciations (appreciations) are inflationary (deflationary) P M = E h/f Pf f E h/f, P M Depreciations (appreciations) deteriorate (improve) terms of trade. TOT P X P M = P h h E h/f Pf f E h/f, TOT Expenditure Switching: Improvement in trade balance. 3 / 72

8 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) 4 / 72

9 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) Prices rigid in the local (destination) currency (LCP) 4 / 72

10 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) Prices rigid in the local (destination) currency (LCP) Depreciations have no impact on inflation P M = P h E h/f, P M 4 / 72

11 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) Prices rigid in the local (destination) currency (LCP) Depreciations have no impact on inflation P M = P h E h/f, P M Depreciations (appreciations) improve (deteriorate) terms of trade. TOT P X P M = P f h E h/f P h f E h/f, TOT 4 / 72

12 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) Prices rigid in the local (destination) currency (LCP) Depreciations have no impact on inflation P M = P h E h/f, P M Depreciations (appreciations) improve (deteriorate) terms of trade. TOT P X P M = P f h E h/f P h f E h/f, TOT No expenditure switching 4 / 72

13 International Spillovers Nominal Rigidities 2 Second generation: Betts and Devereux (2000), Devereux and Engel (2003) Prices rigid in the local (destination) currency (LCP) Depreciations have no impact on inflation P M = P h E h/f, P M Depreciations (appreciations) improve (deteriorate) terms of trade. TOT P X P M = P f h E h/f P h f E h/f, TOT No expenditure switching 3 Handbook of Monetary Economics (2010, Friedman and Woodford), Optimal Monetary Policy in Open Economies, Corsetti, Dedola, Leduc 4 / 72

14 What does micro data tells us? 1 Neither PCP, nor LCP, but pricing in very few currencies Outsized role for dollar 5 / 72

15 What does micro data tells us? 1 Neither PCP, nor LCP, but pricing in very few currencies Outsized role for dollar 2 Prices are rigid in their currency of invoicing 5 / 72

16 What does micro data tells us? 1 Neither PCP, nor LCP, but pricing in very few currencies Outsized role for dollar 2 Prices are rigid in their currency of invoicing 3 Conditional on a price change prices not very sensitive to exchange rates Strategic complementarity in pricing Variable desired mark-ups Imported intermediate inputs 5 / 72

17 What does micro data tells us? 1 Neither PCP, nor LCP, but pricing in very few currencies Outsized role for dollar 2 Prices are rigid in their currency of invoicing 3 Conditional on a price change prices not very sensitive to exchange rates Strategic complementarity in pricing Variable desired mark-ups Imported intermediate inputs 4 Dominant Currency Paradigm: / 72

18 Road Map Dominant currencies Model Empirical Evidence 6 / 72

Dominance of dollar invoicing in world trade Covers 55% of imports, 57% of exports. Averages post 1999. Dollar invoicing share: 4.

19 Dominance of dollar invoicing in world trade Covers 55% of imports, 57% of exports. Averages post Dollar invoicing share: 4.7 times its share in world imports, 3.1 times its share in world exports. Euro invoicing share: 1.2 times for imports and exports. Goldberg (2013), Goldberg and Tille (2009), Ito and Chinn (2013) 7 / 72

20 Limited own currency use in most countries Country Imports Exports Country Imports Exports United States Canada Italy* Poland Germany* Iceland Spain* Thailand France* Israel United Kingdom Turkey Australia South Korea Switzerland Brazil Norway Indonesia Sweden India Japan EM share in world imports: 38%, exports: 33% 8 / 72

21 Model: New Keynesian small open economy Building Blocks Sticky Prices and or Sticky Wages (Calvo) Household and Firms Asset markets Monetary policy 9 / 72

22 Model: New Keynesian small open economy Building Blocks Sticky Prices and or Sticky Wages (Calvo) Household and Firms Asset markets Monetary policy Home H trades with U (dominant currency) and R All prices and quantities in U and R are exogenous (constant) 9 / 72

23 Households Utility: U(C t, N t ) = 1 Ct 1 σc κ 1 σ c 1 + ϕ N1+ϕ t 10 / 72

24 Households Utility: U(C t, N t ) = 1 Ct 1 σc κ 1 σ c 1 + ϕ N1+ϕ t Consumption Aggregator: Kimball ( ) 1 Ωi C ih (ω) γ i Υ dω = 1. Ω i ω Ω i γ i C i Strategic complementarities/variable mark-ups (Dornbusch (1988), Krugman (1987)) 10 / 72

25 Demand for a variety ( C ih,t (ω) = γ i 1 ɛ ln P ) σ/ɛ ih(ω) C t P Elasticity of demand σ ih,t = ( σ 1 ɛ ln P ih(ω) P ) Variability of the mark-up σ ih,t σ ih,t 1 Γ ih,t = ( ɛ ) σ 1 + ɛ ln P ih(ω) P 11 / 72

26 Kimball Demand Figure 1: Demand function with real rigidities ε = 0 ε = 1 ε = 5 ε = relative price (P si /P s ) relative demand (Y /Y ) si s 12 / 72

27 Households Households optimize Budget constraint max C t,w t,b U,t+1,B t+1 (s ) E 0 β t U(C t, N t ) t=0 P tc t+e U,t (1+i U,t )B U,t +B t = W tn t+π t+e U,t B U,t+1 + s S Q t(s )B t+1(s )+E U,t ζ t 13 / 72

28 Households Households optimize Budget constraint max C t,w t,b U,t+1,B t+1 (s ) E 0 β t U(C t, N t ) t=0 P tc t+e U,t (1+i U,t )B U,t +B t = W tn t+π t+e U,t B U,t+1 + s S Q t(s )B t+1(s )+E U,t ζ t Consumption Demand ( ) P ih,t (ω) C ih,t (ω) = γ i ψ D t C t, P t P t C t = i Ω i P ih,t (ω)c ih,t (ω)dω 13 / 72

29 Households Optimality Conditions Portfolio decisions C σc t P = β(1 + i U,t )E t C σc t E U,t+1 t+1 P t+1 E U,t C σc t = β(1 + i t )E t C σc t+1 P t P t+1 14 / 72

30 Households Optimality Conditions Portfolio decisions C σc t P = β(1 + i U,t )E t C σc t E U,t+1 t+1 P t+1 E U,t C σc t = β(1 + i t )E t C σc t+1 P t P t+1 Wage setting (Calvo) E t s=t [ δw s t Θ t,s N s Ws ϑ(1+ϕ) ϑ ϑ 1 κp scs σ Ns ϕ W t (h) 1+ϑϕ ] Ws ϑϕ = 0, 14 / 72

31 Producers Production Function Y t = e at L 1 α t Xt α 15 / 72

32 Producers Production Function Y t = e at L 1 α t Xt α Intermediate input aggregator: Same as C Labor Aggregator: Standard CES [ 1 L t = 0 ] ϑ/(ϑ 1) L t (h) (ϑ 1)/ϑ dh 15 / 72

33 Producers Production Function Y t = e at L 1 α t Xt α Intermediate input aggregator: Same as C Labor Aggregator: Standard CES [ 1 L t = 0 ] ϑ/(ϑ 1) L t (h) (ϑ 1)/ϑ dh 15 / 72

34 Producers Optimality Conditions Profits Π t (ω) = i,j E j,t P j Hi,t (ω)y j Hi,t (ω) MC t Y t (ω) Y t (ω) = i,j Y j Hi,t (ω) 16 / 72

35 Producers Optimality Conditions Profits Π t (ω) = i,j E j,t P j Hi,t (ω)y j Hi,t (ω) MC t Y t (ω) Y t (ω) = i,j Y j Hi,t (ω) Marginal Cost MC t = 1 α 1 Wt α α (1 α) 1 α e at P α t 16 / 72

36 Producers Optimality Conditions Profits Π t (ω) = i,j E j,t P j Hi,t (ω)y j Hi,t (ω) MC t Y t (ω) Y t (ω) = i,j Y j Hi,t (ω) Marginal Cost Input demand MC t = (1 α) Y t L t = α Y t X t = P t MC t 1 α 1 Wt α α (1 α) 1 α e at W t MC t, L t (h) = X ih,t (ω) = γ i ψ P α t ( ) Wt (h) ϑ L t W t ( ) P ih,t (ω) D t X t P t 16 / 72

37 Producers Pricing equations (Calvo) θij i : fraction prices in producer currency θ j ij : fraction prices in local/destination currency θij u : fraction prices in dominant currency Domestic prices and wages sticky in H currency 17 / 72

38 Producers Pricing equations (Calvo) θij i : fraction prices in producer currency θ j ij : fraction prices in local/destination currency θij u : fraction prices in dominant currency Domestic prices and wages sticky in H currency Reset Prices E t s=t δ s t p Θ t,sy j Hi,s t (ω)(σ Hi,s(ω) 1) ( E j,s P j Hi,t (ω) σ ) Hi,s(ω) σ Hi,s (ω) 1 MCs = 0 17 / 72

39 Interest Rates Monetary Policy: Domestic interest rates i t i = ρ m (i t 1 i ) + (1 ρ m )φ M π t + ɛ M,t 18 / 72

40 Interest Rates Monetary Policy: Domestic interest rates i t i = ρ m (i t 1 i ) + (1 ρ m )φ M π t + ɛ M,t Dollar interest rate i U,t = i t + ψ(e B U,t+1 B 1) 18 / 72

41 Interest Rates Monetary Policy: Domestic interest rates i t i = ρ m (i t 1 i ) + (1 ρ m )φ M π t + ɛ M,t Dollar interest rate i U,t = i t + ψ(e B U,t+1 B 1) Exchange rate U-R ln E R,t ln P t = η (ln E U,t ln P t ) + ɛ R,t 18 / 72

42 Exchange Rate Pass-through Export price pass-through in H currency higher Greater the variability of mark-ups Greater the reliance on imported inputs 19 / 72

43 Exchange Rate Pass-through Export price pass-through in H currency higher Greater the variability of mark-ups Greater the reliance on imported inputs Import price pass-through in H currency lower Greater the variability of mark-ups 19 / 72

44 Some Analytics Exchange Rate Pass-through: Fully flexible prices Export Prices µ Hi = µ Hi (p Hi e i p i i ) p Hi,t = µ Hi,t + mc t 20 / 72

45 Some Analytics Exchange Rate Pass-through: Fully flexible prices Export Prices µ Hi = µ Hi (p Hi e i p i i ) Log-differentiating: p Hi,t = µ Hi,t + mc t p Hi,t = Γ mc t + Γ ( ) p i 1 + Γ i,t + e i,t 20 / 72

46 Some Analytics Exchange Rate Pass-through: Fully flexible prices Export Prices µ Hi = µ Hi (p Hi e i p i i ) Log-differentiating: p Hi,t = µ Hi,t + mc t p Hi,t = Γ mc t + Γ ( ) p i 1 + Γ i,t + e i,t mc t = (1 α) w t + α p t a t 20 / 72

47 Some Analytics Exchange Rate Pass-through: Fully flexible prices Export Prices µ Hi = µ Hi (p Hi e i p i i ) Log-differentiating: p Hi,t = µ Hi,t + mc t p Hi,t = Γ mc t + Γ ( ) p i 1 + Γ i,t + e i,t mc t = (1 α) w t + α p t a t mc t = 1 α α ( w t+ γ i mc i 1 i,t + e i,t ) a t 1 αγ H 1 αγ H 1 αγ H i U,R 20 / 72

48 Some Analytics Exchange Rate Pass-through: Fully flexible prices Export Prices p Hi,t = [ ] 1 αγ i + Γ e i,t 1 + Γ 1 αγ H + 1 αγ j e j,t 1 + Γ 1 αγ H α w t Γ 1 αγ H 1 + Γ 1 a t 1 αγ H where j i, for i, j {U, R} 2. If Γ = 0, α = 0 or γ H = 1, 100% PT into destination currency 21 / 72

49 Some Analytics Exchange Rate Pass-through: Fully flexible prices Import Prices p ih,t = [ Γ αγ Hγ i 1 + Γ 1 αγ H + Γ αγ H γ j ] e i,t e j,t 1 + Γ 1 αγ H + Γ 1 + Γ γ 1 α H w t Γ 1 αγ H 1 + Γ γ H a t where j i, for i, j {U, R} / 72

50 Some Analytics Exchange Rate Pass-through: Fully rigid prices PCP, θ H HU = 1 and θ H HR = 1 p Hi,t = 0 e i,t + 0 e j i,t, p ih,t = 1 e i,t + 0 e j i,t, i tot Hi,t = p Hi,t p ih,t = 1 e i,t i LCP, θ U HU = 1 and θ R HR = 1 p Hi,t = 1 e i,t + 0 e j i,t p ih,t = 0 e i,t + 0 e j i,t i tot Hi,t = p Hi,t p ih,t = 1 e i,t i DCP, θhu U = 1 and θhr U = 1. p Hi,t = 1 e U,t + 0 e i U,t p ih,t = 1 e U,t + 0 e i U,t i tot Hi,t = p Hi,t p ih,t = 0 i where tot Hi is the terms of trade between regions H and i 23 / 72

51 Parameter Value Household Preferences Discount factor β 0.99 Risk aversion σ c 2.00 Frisch elasticity of N ϕ Disutility of labor κ 1.00 Production Interm share α 2/3 Demand Elasticity σ 2.00 Super-elasticity ɛ 1.00 Rigidities Wage δ w 0.85 Price δ p 0.75 Monetary Rule Inertia ρ m 0.50 Inflation sensitivity φ M 1.50 Shock persistence ρ εi 0.50 Note: other parameter values as reported in the text. Table : Parameter Values 24 / 72

52 Impulse Response to Monetary Expansion Log-linearization # DCP PCP LCP (a) Inflation # DCP PCP LCP (b) Exchange Rates 25 / 72

53 Impulse Response to Monetary Expansion Log-linearization # DCP PCP LCP (c) Terms of Trade # DCP PCP LCP (d) Export Quantity 26 / 72

54 Impulse Response to Monetary Expansion Log-linearization # DCP PCP LCP (e) Import Price # DCP PCP LCP (f) Import Quantity 27 / 72

55 Impulse Response to Monetary Expansion Log-linearization # DCP PCP LCP (g) Mark-up # DCP PCP LCP (h) Trade 28 / 72

56 Colombia , Source: DIAN/DANE (customs), SIREM (agency supervising large private firms Commodity Currency, Free float since September 1999 Share of mining output in exports is 58.4%, Manufacturing, 36.9% Currency composition of exports: USD: 98.4% Weighted (by income) average imported input share: 38% for manufacturers, 44% for manuf exporters Focus on manufacturing 29 / 72

57 Colombia Table : Currency Distribution All Exports Manufactures US Dollar 98.28% 98.39% Euro 0.72% 0.70% Colombian Peso 0.67% 0.52% Venezuelan Boĺıvar 0.27% 0.33% Sterling Pound 0.02% 0.01% Mexican Peso 0.01% 0.01% Other currencies 0.03% 0.03% 30 / 72

58 Colombia Terms of Trade and Nominal Exchange Rate q3 2008q1 2010q3 2013q1 2015q3 TIME ER TOT TOT P X P M, Corr(TOT, E p/$ ) = / 72

59 Colombia: Stability of Terms of Trade q3 2008q1 2010q3 2013q1 2015q3 TIME ER TOT (Manuf) TOT Data Cov(ER,TOT ) Var(ER) / 72

60 Parameter Value Measured Domestic Input Share γ H 0.60 Export Invoicing Shares to U θhu U 1.00 to R θhr, U θhr R 0.93,0.07 Shocks commodity prices σ ζ, ρ ζ 0.13, 0.74 Strategic complementarities ε 2 Elasticity of Demand σ 2 Estimated Import Invoicing Shares from U θuh U 1 from R θrh, U θrh R 0.86, 0.14 Oil share ζ 1 Shocks productivity shocks σ a, ρ a 0.01,0.9 e R η, σ R 1.5, Note: other parameter values as reported in the text. Table : Parameter Values 33 / 72

61 Pass-through, U Data p t = α + 8 k=0 β k e t k + Z t + ɛ t firm*industry*country FE, quarter*year clusters P HU PT UH / 72

62 Pass-through, U Data Vs. DCP P HU (Est) P HU (Data) P UH (Est) P UH (Data) / 72

63 Pass-through, U Data Vs. PCP P HU (PCP) P HU (Data) P UH (PCP) P UH (Data) / 72

64 Pass-through, U Data Vs. LCP P HU (LCP) P HU (Data) P UH (LCP) P UH (Data) / 72

65 Pass-through, R Data PT HR PT RH / 72

66 DCP P HR (Est) P HR (Data) P RH (Est) P RH (Data) / 72

67 PCP P HR (PCP) P HR (Data) P RH (PCP) P RH (Data) / 72

68 LCP P HR (LCP) P HR (Data) P RH (PCP) P RH (Data) / 72

69 Table : ERPT (Non-Dollarized Economies) (1) (2) (3) (4) p HR p HR p RH p RH Data e R 0.697*** * 0.742*** 0.301*** (0.115) (0.0464) (0.126) (0.0791) e U 0.660*** 0.540*** (0.0473) (0.0662) DCP e R e U PCP e R e U LCP e R e U / 72

70 Conclusion Most trade is invoiced in very few currencies. Dominant currency paradigm pricing in a dominant currency pricing complementarities imported input use in production Data rejects PCP/LCP in favor of DCP. Implications MP has limited impact on exports and terms of trade. TB adjusts mainly through exports not imports Adverse shock in emerging markets reduces world trade Inflation and imports sensitivity to DC exchange rates far exceeds the share of the dominant currency country in trade 43 / 72

Dominant Currency Paradigm

Dominant Currency Paradigm A New Model for the Small Open Economy Camila Casas Banco de la República Gita Gopinath Harvard Federico Díez Federal Reserve Bank of Boston Pierre-Olivier Gourinchas UC Berkeley