Asymmetric Labor Market Institutions in the EMU and the Volatility of In ation and Unemployment Di erentials

Transcription

1 Asymmetric Labor Market nstitutions in the EMU and the Volatility of n ation and Unemployment Di erentials Mirko Abbritti (Universidad de Navarra) Andreas Mueller (Columbia Business School) DNB/MF Workshop Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 1 / 37

2 Motivation 1 Sizeable and persistent in ation and output growth di erentials in the Euro Area =) the adjustment mechanism may not be working e ciently 2 Labor market asymmetries in the EMU Unemployment is high and tends to be prolonged over time Labor markets are heavily regulated and real wages rather in exible Labor market institutions are widely heterogeneous across member states Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 2 / 37

3 Evidence EPL Wage Wage Bene t Union Unempl. Flexibility ndexation repl. rate Density Rate (6) France ,2 9,7 9,6 Germany ,2 25, 8, taly ,3 34,9 9,1 Spain ,1 13,9 1,7 Portugal 1.7 n.a ,9 23,5 5,4 Netherlands ,5 23,2 3,7 reland ,5 37,8 4,5 Av. Euro Area 1.2 1, ,6 28,9 7,1 Av. Others.6,4 n.a. 13,6 21,8 5, Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 3 / 37

4 What are the macroeconomic and policy implications of asymmetric labor market structures? Common perception among policy-makers: Structure of labor markets is important for in ation dynamics "The implementation of the reforms of the Lisbon agenda, by easing labour and product rigidities... will also improve the e ectiveness of monetary policy" (Trichet, June 27) Flexible labor markets are crucial for the e cient functioning of the currency union Received wisdom: Need for more exible labour markets in the context of the EU (ECB Monthly Bullettin, May 25) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 4 / 37

5 Related studies The importance of nominal rigidities Benigno (24), Benigno and Lopez Salido (22), Monacelli and Galì (28)... =) Target what is sticky : the CB should give higher weight to in ation in the region with more sticky prices The importance of labor market rigidities Campolmi and Faia (211): n ation volatility di erentials and bene t replacement ratios Andersen and Seneca (21), Poilly and Sahuc (28), Fahr and Smets (211) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 5 / 37

6 Unemployment Rigidities vs. Real Wage Rigidities Unemployment Rigidities (UR): Employment protection legislation, hiring costs, matching technology Quantities are rigid (unemployment), prices (wages) have to adjust Real Wage Rigidities (RWR): Wage norms, wage bargaining, wage staggering, trade unions, wage indexation Prices (wages) are rigid, quantities (unemployment) have to adjust Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 6 / 37

7 Results 1 RWR vs. UR: UR increase the volatility of in ation di erentials, RWR increase the volatility of unemployment di erentials 2 Asymmetries in labor markets should be a concern for policy-makers 3 The importance of the interactions among labor market institutions! Not recognizing these interactions can lead to wrong policy prescriptions Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 7 / 37

8 The Model Two countries, Home and Foreign, of equal size, in a currency union Three key ingredients (based on Blanchard and Galì, 21) 1 Nominal price rigidities in the retail market 2 Hiring costs in the labor market 3 Real wage rigidities 5 shocks: Home and Foreign productivity shocks, Home and Foreign preference shocks, Monetary Policy shock. Simplifying assumptions: Cobb-Douglas preferences (as in Benigno 24, Galí and Monacelli 28) Complete nancial markets and no migration Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 8 / 37

9 Supply Side Hiring Frictions: The separation rate δ is exogenous. Hiring occurs at cost: G t = A t B (x t ) ϕ (1) where x t = h t U t is the labor market tightness. Real Wage Rigidity: Following Hall (25): wt R = w Nash t 1 γ ( w) γ (2) where w Nash t = MRS t + ηre t and γ captures real wage rigidities Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 9 / 37

10 The NK Phillips Curve Nominal price rigidities à la Calvo (1983). The NKPC has apparently the same form as in the NK model standard ˆπ H t = βe t ˆπ H t+1 + λ cmc H t But now the dynamics of marginal costs are substantially a ected by the introduction of real imperfections: cmc H t = u ŵt H,R + αŝ t ât H n o +h ˆx t H β(1 δ)he t ˆβ t,t+1 + at+1 H + ϕˆx t+1 H where the structural parameters u and h depend on labor market frictions and on the degree of real wage stickiness. S t = P t F. Pt H Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 1 / 37

11 The Constrained E cient Equilibrium Proposition The decentralized equilibrium corresponds to the constrained e cient equilibrium if four conditions are satis ed: 1 Monopolistic distortions are eliminated through a production subsidy 2 The Hosios condition holds, i.e. ϕ = η 3 Real wages are fully exible 4 Prices are fully exible! This allows to focus on the ine cient portion of economic uctuations (unemployment and output gaps) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 11 / 37

12 Calibration Strategy Real Wage Rigidity: change the degree of RWR γ from.25 to.75. Unemployment Rigidity: the labor market is sclerotic when the job- nding rate x and the separation rate δ are low (see Blanchard and Gali 21, Elsby et al. 21). Example: Baseline High Low RWR γ =.5 γ =.75 γ =.25 UR x =.45, δ =.7 x =.2, δ =.3 x =.7, δ =.12 Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 12 / 37

13 Labor Market Rigidities and the Phillips Curve The presence of labor market rigidities substantially a ect the NKPC: ˆπ H t = βe t ˆπ H t+1 h ũ H t + h L ũ H t 1 + h F E t ũ H t+1 + h FS E t s t+1 + h S s t γh T ˆT t Focus on the two key parameters: Slope Coe cient h : captures the elasticity of in ation to unemployment changes Trade-o Coe cient γh T : determines to what extent productivity and preference shocks enter as cost push shocks in the Phillips curve Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 13 / 37

14 Slope of PC Slope of PC Labor Market Rigidities and the Slope of the Phillips Curve 7 6 Low RWR Baseline High RWR 7 6 Low UR Baseline High UR Unemployment Rigidities Real Wage Rigidity Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 14 / 37

15 Labor Market Rigidities and the Phillips Curve RWR vs. UR: RWR atten the Phillips curve and introduce a trade-o of monetary policy. UR steepen the Phillips Curve but have a negligible impact on the trade-o coe cient. ntuition: RWR: lower the sensitivity of wages and marginal costs to shocks ŵ t = (1 γ) ŵ t UR: in a sclerotic labor market changes in unemployment lead to large variations in labor market tightness and marginal costs ˆx t = 1 δn fû t (1 δ) (1 x) û t 1 g Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 15 / 37

16 The Adjustment Mechanism Monetary policy: î t = ρ m î t 1 + (1 ρ m ) (ω π ˆπ U t + ω y ỹ U t ) + ε t Focus on in ation and unemployment di erentials as they re ect the e ciency of the adjustment mechanism: d ˆπ t = ˆπ H t ˆπ F t ; dũ t = ũ H t ũ F t where ũ i t is the unemployment gap Two polar cases: Analyse asymmetric shocks in a symmetric CU: the role of rigidities Analyse symmetric shocks in an asymmetric CU: the role of asymmetries Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 16 / 37

17 std std 1. Volatility of Di erentials and Labor Market Structures (a) The Volatility of nflation Differentials UR RWR (b) The Volatility of Unemployment Differentials 1.1 UR RWR ndex of Rigidity ndex of Rigidity Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 17 / 37

18 The Policy Frontier Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 18 / 37

19 2. Di erentials and Asymmetric Labor Market Structures sym. productivity shock monetary policy shock.5.5 π H π F π H π F.5 A. Baseline B. Asym. UR C. Asym. RWR quarter quarter sym. productivity shock monetary policy shock.1.1 u H u F.1.2 u H u F quarter quarter Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 19 / 37

20 std std 2. Volatilities of Di erentials and Asymmetric Labor Market Structures (a) The Volatility of nflation Differentials Asymmetric UR Asymmetric RWR (b) The Volatility of Unemployment Differentials 1.3 Asymmetric UR Asymmetric RWR ndex of Asymmetry ndex of Asymmetry Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 2 / 37

21 3. How important are interaction e ects between labor market institutions? To answer this question, we consider ve currency unions: 1 Symmetric currency union 2 Asymmetric UR 3 Asymmetric RWR - labor markets are identical - di erent UR (H: UR=1; F: UR=) - di erent RWR (H: RWR=1; F: RWR=) 4 RWR+UR (Complements) 5 RWR+UR (Substitutes) - H: RWR=1, UR=1; F: RWR=, UR= - H: RWR=1, UR=; F: RWR=, UR=1 Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 21 / 37

22 3. How important are interaction e ects between labor market institutions? std (u d ) std (π d ) Symmetric CU Asymmetric UR Asymmetric RWR Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 22 / 37

23 3. How important are interaction e ects between labor market institutions? std (u d ) std (π d ) Symmetric CU Asymmetric UR Asymmetric RWR RWR+UR (Complements) RWR+UR (Substitutes) NB: Complements: H: RWR=1, UR=1; F: RWR=, UR= Substitutes: H: RWR=1, UR=; F: RWR=, UR=1 Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 23 / 37

24 3. How important are interaction e ects between labor market institutions? The importance of the interactions between labor market institutions f institutions are substitutes (low UR/high RWR or viceversa) the e ects reinforce each other f institutions are complements (high UR/high RWR or viceversa) the e ects o set each other Are UR and RWR substitutes or complements? No clear-cut answer... Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 24 / 37

25 Two Extensions 1. Optimal Monetary Policy (linear- quadratic approach) 2. Empirical Test (nteracted Panel Var) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 25 / 37

26 Extension 1. Optimal Monetary Policy Abbritti - Mueller Should monetary policy target unemployment? ˆπ U t α u û U t = Result: No - except for large degrees of RWR, the optimal weight α u is near zero and the welfare gains of MT over T are small 2. Should the CB respond to in ation di erentials? α π ˆπ Ht + (1 α π ) ˆπ Ft = Result: Yes, but only if LM are substitute The in ation target gives higher weight to countries with:! more rigid unemployment! more exible real wages (NB: Opposite to the target what is sticky prescription by Benigno and Woodford) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 26 / 37

27 Extension 2. nteracted Panel Var Abbritti - Weber 21 Four equation panel VAR for 2 OECD countries: A Y i,t = 4 A l Y i,t l=1 l + U i,t where Y i,t = (S it, UR it, NFL i,t, NT i,t ) and U i,t are structural shocks Allowing for varying slope coe cients: a ij l = β 1 + β 2 EPL it + β 3 RWR it for l =, 1,.., L and i = UR it, NFL i,t, j = S it, UR it, NFL i,t, NT i,t ; main indicator RWR: RWR it = BEN it Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 27 / 37

28 High UR Low RWR.6 High UR High RWR.6 Low UR Low RWR.6 Low UR High RWR.6 nflation Unempl. Rate RF to an oil price shock Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 28 / 37

29 Conclusion 1 Di erent labor market institutions have very di erent dynamic e ects: real wage rigidities vs. unemployment rigidities 2 Asymmetric labor market structures increase the volatility of ALL di erentials. 3 Crucial to understand the interactions between institutions. Are RWR and UR substitutes or complements? Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 29 / 37

30 APPENDX Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 3 / 37

31 x The UR ndex x δ u u, δ UR Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 31 / 37

32 Calibration Preferences β φ i ɛ i µ i α Technology A i ϕ i 1 1 Labor market u i x i δ i η i Price and Real Wage rigidities θ i γ i.66.5 Shocks Persistence and Volatility ρ i a σ i a ρ σa σ ε Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 32 / 37

33 π H π F (a) Home inflation.5 A. Baseline B. Low UR 1 C. Low RWR quarter (b) Foreign inflation quarter (c) nflation differential u H u F (d) Home unemployment gap quarter (e) Foreign unemployment gap quarter (f) Unemployment differential π H π F.5 1 u H u F quarter quarter Note: nflation is annualized. Asymmetric Productivity Shock (Home) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 33 / 37

34 Asymmetric Shocks and the Role of Labor Market Rigidities Labor market rigidities have a strong impact on the adjustment mechanism of the currency union URs increase the volatility of in ation di erentials and reduce the volatility of unemployment di erentials. RWRs increase the volatility of unemployment di erentials but have a small e ect on the volatility of in ation di erentials: cmc t = culc t + dmhc t Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 34 / 37

35 Labor Market Rigidities and the Monetary Union The Consequences of Asymmetric Labor Markets! When labor markets are asymmetric, symmetric shocks have di erent e ects on member countries! Asymmetric labor markets unambiguously worsen the adjustment mechanism of the currency union Monetary Policy mplications! Monetary policy expansions create terms of trade depreciations in countries with high RWR/low UR! Can the central bank bene t from trying to use selectively its policy instrument? Should the CB react to in ation di erentials? NB: Results are robust to shocks and preference speci cations Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 35 / 37

36 3. How important are interaction e ects between labor market institutions? f shocks are perfectly correlated... std (u d ) std (π d ) Symmetric CU.. Asymmetric UR Asymmetric RWR RWR+UR (Complements) RWR+UR (Substitutes) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 36 / 37

37 std(π U ) std(π U ) (a) UR and the PolicyFrontierbetween π U and u U UR = UR = (b) RWRand the PolicyFrontierbetween π U and u U γ =.25 γ = std(u U ) std(u U ) Abbritti & Mueller () Asymmetric Labor Markets in EMU DNB/MF Workshop 37 / 37