Trade and Inequality: From Theory to Estimation

Transcription

1 Trade and Inequality: From Theory to Estimation Elhanan Helpman Oleg Itskhoki Marc Muendler Stephen Redding Harvard Princeton UC San Diego Princeton MEF Italia Dipartimento del Tesoro September / 24

2 Neoclassical trade theory (H-O, SF, R) sector-level comparative advantage focus on between effects Motivation New trade theory Krugman: intra-industry trade Melitz: firm-level comparative advantage focus on within effects Trade and inequality Heavily influenced by H-O framework Empirically has limited explanatory power New view of trade and inequality link wages to firm performance within-industry, between-firm 2 / 24

3 This Paper 1 Uses matched employer-employee data from Brazil for increase in inequality show trade liberalization show 2 Documents, for both levels and changes, the importance of inequality: within sector-occupations for workers with similar observables (residual inequality) between firms 3 Structurally estimates a heterogenous-firm model of firm employment, wages and export status extension of HIR (2010) a model of within-sector, between-firm residual inequality reproduces empirical size and exporter wage premia: w ij = α + βh j + δ xj ε ij 3 / 24

4 Related Literature Long and large tradition in labor literature New view empirics: Bernard and Jensen (1995)... Verhoogen (2008) Amity and Davis (2011) AKM (1999) estimation used in trade context New view theory: Feenstra and Hanson (1999)... Yeaple (2005)... Egger and Kreickemeier (2009) HIR (2010)... 4 / 24

5 Brazilian RAIS Data Matched employer-employee data from All workers employed in the formal sector Focus on the manufacturing sector Observe firm, industry and occupation Observe worker education (high school, college degree), demographics (age, sex) and experience (employment history) Over the period as a whole, our sample includes more than 7 million workers and 100,000 firms in every year Trade transactions data from Merged with the matched employer-employee data Observe firm exports and export products and destinations 5 / 24

6 Sectors Twelve aggregate sectors (IBGE) Industry Empl t Rel. mean Fraction Exporter share (%) log wage Exporters Empl t % 2 Non-metallic mineral products Metallic products Machinery, equipment & instr Electrical & telecomm. equip Transport equipment Wood products & furniture Paper, publishing & printing Rubber, tobacco, leather & fur Chemical & pharma. products Apparel & textiles Footwear Food, beverages & alcohol More than 250 disaggregated industries (CNAE) / 24

7 Occupations Five aggregate occupations Occupation Employment Relative mean share (%) log wage 1 Professional and Managerial Skilled White Collar Unskilled White Collar Skilled Blue Collar Unskilled Blue Collar More than 300 disaggregated occupations (CBO) / 24

8 Stylized Facts 7 / 24

9 Within and Between Inequality Sector-occupation bins Level (%) Change (%) A. Main Period Within occupation Within sector Within sector-occupation Within detailed-occupation Within sector detailed-occupation B. Late Period Within detailed-sector detailed-occupation Fact 1 Within sector-occupation component of wage inequality accounts for over 2/3 of both level and growth of wage inequality show within regions 7 / 24

10 Residual Inequality Conditional on worker observables Level (%) Change (%) Residual wage inequality within sector-occupation Fact 2 (i) Residual inequality is at least as important as worker observables for both level and growth of wage inequality (ii) Almost all residual inequality is within sector-occupations 8 / 24

11 Between-firm Inequality Mincer log-wage regression with firm fixed effect: i worker j firm l sector-occupation bin w it = z itϑ lt + ψ jlt + ν it ψ jlt firm fixed effect includes: Returns to unobserved skill (workforce composition) Worker rents (differences in wage for same workers) Match effects Decomposition of within inequality: Observables: var ( z it ˆϑ ) lt Between-firm component: var ( ) ˆψjlt Covariance: cov ( z it ˆϑ lt, ˆψ ) jlt Within-firm component: var (ˆν ) it 9 / 24

12 Between-firm Inequality Within sector-occupation bins unconditional firm wage component, ψjlt U conditional firm wage component, ψjlt C Level (%) Change (%) Level (%) Change (%) Between-firm wage inequality Within-firm wage inequality Worker observables 13 2 Covar observables firm effects Fact 3 Between-firm component account for about half of level and the majority of growth of within sector-occupation wage inequality 9 / 24

13 Between-firm Inequality Size and exporter wage premia unconditional firm wage component, ˆψ jt U conditional firm wage component, ˆψ jt C Firm Employment Size (0.010) (0.009) Firm Export Status (0.042) (0.024) Sector Fixed Effects yes yes Within R-squared Observations 91, , 410 Fact 4 Larger firms on average pay higher wages; exporters on average pay higher wages even after controlling for size. The remaining variation in wages is substantial. 10 / 24

14 Structural Model 11 / 24

15 1 Melitz (2003) product market: Model: Extension of HIR R = ΥAy β, Υ {1, Υ x > 1} 2 Heterogeneity in fixed cost of exports: e ε F x 3 Complementarity between productivity and worker ability: y = e θ H γ ā, γ < 1 4 Unobserved heterogeneity and costly screening: e η C (a c) δ δ ā = k k 1 a c 5 DMP frictional labor market and wage bargaining: W = βγ R 1 + βγ H = b (a c) k/δ 11 / 24

16 A firm with idiosyncratic shock {θ, η, ε}: R(θ, η, ε) = κ r Υ 1 β Γ ( e θ ) β Γ H(θ, η, ε) = κ h Υ (1 β)(1 k/δ) Γ W (θ, η, ε) = κ w Υ k(1 β) δγ Model Predictions β(1 γk) (e η) δγ ( e θ ) β(1 k/δ) Γ (e ( e θ ) βk δγ (e η) k δ (1+ β(1 γk) δγ ) β(1 γk)(1 k/δ) η) δγ k δ Market access variable Υ = 1 + I x ( Υx 1 ), Υ x = 1 + τ β 1 β A x A d Selection into exporting ( ) I x = {κ π Υ 1 β Γ (e θ x 1 ) β Γ } β(1 γk) (e η) δγ F x e ε 12 / 24

17 Empirical model of X j = {h j, w j, ι j } j : h j = α h + µ h ι j + u j, Econometric Model w j = α w + µ w ι j + ζu j + v j, ι j = I { z j f } Distributional assumption: (u j, v j, z j ) N (0, Σ), Σ = σu 2 0 σv 2 ρ u σ u ρ v σ v 1 Selection (ρ u, ρ v ) versus Market access (µ h, µ w ) Theoretical restrictions: µ h, µ w > 0 and ζ µw µ h ζ + σ2 v (1+ζ)σ 2 u 13 / 24

18 Estimation Results Parameters α h and α w σ u σ v ζ ρ u ρ v f µ h µ w / 24

19 Model Fit Firm-level moments All firms Non-exp. Exporters Data Mean h Mean w Std deviation h Std deviation w Correlation h & w Fraction of exporters 9.0% Model Mean h Mean w Std deviation h Std deviation w Correlation h & w Fraction of exporters 9.0% 15 / 24

20 Employment and Wage Distributions Log Employment Log Firm Wages 0.4 Data Model /4 1/ Exporters vs Non-exporters 1.2 Exporters vs Non-exporters /4 1/ Data (non-exp) Data (exporters) Model (non-exp) Model (exporters) 16 / 24

21 Worker Wage Distribution Log Worker Wages 1.2 Exporters vs Non-exporters /4 1/ /4 1/ / 24

22 Model Fit Worker wage dispersion Data Model Std deviation non-exporters exporters Gini coefficient /10-ratio / / / 24

23 Model Fit Size and exporter wage premia Data Model Employment premium Exporter premium R-squared / 24

24 Counterfactuals Estimated model h j = α h + µ h ι j + u j, w j = α w + µ w ι j + ζu j + v j, ι j = I { z j f } where (u j, v j, z j ) N (0, Σ) and f = 1 σ ( ( )) α f + log F x log Υ 1 β Γ x 1 µ h + µ w = Υ 1 β Γ x 19 / 24

25 Variation in Fixed Export Cost % 40% 60% 80% 100% Exporter Employment Share 20 / 24

26 Variation in Variable Trade Cost % 40% 60% 80% 100% Exporter Employment Share 21 / 24

27 Bounds on Inequality Reduced-form model: h j = α h + µ h ι j + u j, w j = α w + µ w ι j + ζu j + v j, ι j = I { z j f } Subject to market access µ h, µ w 0 and selection ρ u, ρ v 0 size and exporter wage premia: w j = α + βh j + δι j + ɛ j 22 / 24

28 Bounds on Inequality Reduced-form model: h j = α h + µ h ι j + u j, w j = α w + µ w ι j + ζu j + v j, ι j = I { z j f } Subject to market access µ h, µ w 0 and selection ρ u, ρ v 0 size and exporter wage premia: w j = α + βh j + δι j + ɛ j Estimated coefficients GMM µ h µ w ρ u ρ v objective Theoretical model No market access effect (µ w = 0) No selection effect (ρ v = 0) No trade effects (µ h = µ w = 0) / 24

29 Bounds on Inequality 1994 over Peak over autarky autarky Panel A: Inequality bounds Theoretical model 7.6% 7.8% Lower bound (no market access, µ w = 0) 3.3% 5.0% Upper bound (no selection, ρ v = 0) 8.3% 8.8% 23 / 24

30 Bounds on Inequality and extensions 1994 over Peak over autarky autarky Panel A: Inequality bounds Theoretical model 7.6% 7.8% Lower bound (no market access, µ w = 0) 3.3% 5.0% Upper bound (no selection, ρ v = 0) 8.3% 8.8% Panel B: Robustness and extensions Sector-region heterogeneity 8.8% Multi-destination A 14.0% Multi-destination B 15.9% 23 / 24

31 Conclusions Neoclassical trade theory emphasizes wage inequality between occupations and industries In contrast, new theories of firm heterogeneity and trade point to wage dispersion within occupations and industries Using matched employer-employee data for Brazil, we show: Much of the increase in wage inequality since the mid-1980s has occurred within sector-occupations Increased within-group wage inequality Increased wage dispersion between firms Between-firm wage dispersion related to trade participation Develop a framework for the structural estimation of a model with firm heterogeneity and wage dispersion across firms Use this framework to quantify the effect of trade on wage dispersion 24 / 24

32 25 / 24

33 Wage Inequality Mean log wage (U.S. dollars) Year Variance log wage (U.S. dollars) Mean log wage (left scale) Variance log wage (right scale) Back to slides 26 / 24

34 Panel A: Firm Export Participation Trade Openness Share of Firms Year Share of Employment Share Exporters Exp Employ Share Back to slides 27 / 24

35 Regional Robustness overall inequality residual inequality Level Change Level Change Within sector-occupation Within sector-occupation, São Paulo Within sector-occupation-state Within sector-occupation-meso Back to slides 28 / 24

36 Likelihood Function L(Θ X j ) = j P{(h j, w j, ι j ) Θ} 1 φ ( ) 1 û j φ (ˆv [ ( ) j Φ σ u σ v P{(h j, w j, ι j ) Θ} = )] 1 ιj [ f ρ u û j ρ v ˆv j 1 ρ 2 u ρ 2 v 1 Φ ( f ρ u û j ρ v ˆv j 1 ρ 2 u ρ 2 v )] ιj û j = h j α h µ h ι j σ u, ˆv j = (w j α w µ w ι j ) ζ(h j α h µ h ι j ) σ v back to slides 29 / 24