Tax incidence on top-inome earners

Transcription

1 Tax incidence on top-inome earners Denmark, again... Tax incidence on top-inome earners 1 / 19

2 Outline 1 Lecture Recap 2 Empirical Strategy to identify tax incidence 3 Evidence for rent sharing Tax incidence on top-inome earners 2 / 19

3 Hamlet - Prince of Denmark Ay, marry, is t: But to my mind, though I am native here And to the manner born, it is a custom More honourd in the breach than the observance. This heavy-headed revel east and west Makes us traduc d and tax d of other nations; - Act 1, Scene 4 Tax incidence on top-inome earners 3 / 19

4 Competitive labour supply vs. matching frictions model Standard model worker paid their marginal product y consistent with bunching by deliberate labour supply choice (i.e. from below) missing mass is expected under the notch bunching from above is not expected = absent migration response: density above notch unchanged Matching Frictions Bunching at z from above when firm s bargaining power 1 β > 0 Bunching at z from below when workers s bargaining power β > 0 Tax incidence on top-inome earners 4 / 19

5 Panel B: Preferential Tax Scheme to High-Income Foreigners Density Regular tax rate below cutoff Low tax rate above cutoff Notch smooth pre-scheme density z Earnings z

6 Figure 11: Density of earnings normalized at the scheme threshold Yearly Number of Foreigners in Income Bin Bunching=1.3 (.35) Missing mass=.21 (.14) Earnings/threshold

7 Predictions about wages at the scheme s tax rate, τ S, β determines the sharing of rent in the wage band [y S 0, y] as zs = βy + (1 β)y S 0 at the normal tax rate, τ D, earnings are z D When 1 β > 0, i.e. firm has bargaining power, the scheme reduces pretax earnings. When β = 1, i.e. competitive model, z S = z D Tax incidence on top-inome earners 7 / 19

8 Pretax earnings reduced during the scheme Figure: Effects of the Tax Scheme on Pretax Earnings: Repeated Cross-Section Tax incidence on top-inome earners 8 / 19

9 Pretax earnings rise after scheme expired Around 25% of the scheme beneficiaries stayed more than 36 months. Figure: Effects of the Tax Scheme on Pretax Earnings: Panel Data Evidence Tax incidence on top-inome earners 9 / 19

10 Tax incidence on top-inome earners 10 / 19

11 Figure 10: Number of foreigners by income groups over time Control #1:.8 to.9*threshold Control #2:.9 to.99*threshold Treatment: earnings> threshold DD elasticity: Long term: 1.62 (.16) Short term: 1.28 (.15)

12 Wage Posting vs Bargaining Two models are closely connected Suppose that the supply of labor can be written as: Then wage is G(w) = (w b 0 ) ε w = ε 1 + ε p ε b 0 But implications for distribution of wages differ ex post individualistic wage-bargaining suggest considerable variation in wages within firms between workers with different reservation wages ex ante wage-posting suggest no wage variation within firms between workers with different reservation wages C. Landais (LSE) EC426-Public Economics 48 / 116

13 Incidence in markets with frictions Price indeterminacy: critical characteristic of markets with frictions Canonical example of market with friction: labor market Effect of taxes on wages depends on wage setting mechanism (and there are infinity of wage setting processes compatible with equilibrium) Kleven & al. (2012): study foreigner s tax scheme in Denmark and look at incidence of income taxes on high-skilled labor Identify rent sharing in the market C. Landais (LSE) EC426-Public Economics 49 / 116

14 Foreigners Tax Scheme in Denmark: Passed in 1992; applicable to contracts signed after 1 June, 1991 Flat tax of 30% (25% after 1995) in lieu of the regular progressive income tax (top rate of above 60%) Up to 3 years after which taxpayer subject to ordinary income tax Eligibility requirements: No tax liability in Denmark in the 3 years prior to going on the scheme Annual earnings of at least 103,000 Euros (as of 2009, indexed): top.5% of the income distribution C. Landais (LSE) EC426-Public Economics 50 / 116

15 Common features: Migration Individuals have productivity y Reservation wage y 0 for working in Denmark y 0 = z h(1 τ h ) + ν 1 τ = y 0 τ=0 1 τ 1 z h, τ h wage and average tax rate in home country 2 ν net cost of migrating 3 τ average tax rate in Denmark: τ S < τ D Migration: Worker migrates in if and only if y0 y Scheme reduces y0 and causes migration when y S 0 y < y D 0 C. Landais (LSE) EC426-Public Economics 51 / 116

16 Standard Model vs. Matching Model Standard model, workers paid their marginal product y Model with matching frictions Workers and firms expend resources to create a match When matched, surplus band of acceptable wages (y0, y) Pretax wage z splits the surplus: Nash bargaining maximizes: W = (y z) 1 β ((1 τ)z y τ=0 0 ) β β [0, 1], bargaining power of worker Nests standard model: case β = 1 z = βy + (1 β)y 0 C. Landais (LSE) EC426-Public Economics 52 / 116

17 Bargaining theory predictions Pretax wage increases with τ If β < 1, scheme reduces pretax wage z Bunching at z: z S }{{} βy+(1 β)y S 0 < z D }{{} βy+(1 β)y D 0 Bunching from above when bargaining power of firm 1 β > 0 Bunching from below when bargaining power of workers β > 0 C. Landais (LSE) EC426-Public Economics 53 / 116

18 Panel C: Intensive (Bargaining) Responses Conditional on Migration Density Density hole creates bunching from below (increasing in β) Density shift creates bunching from above (increasing in 1- β) post-scheme density absent migration effect z Earnings z

19 Panel D: Intensive and Migration Responses Density new migrants add to bunching migration response post-scheme (empirical) density z Earnings z