Who Benefits from State Corporate Tax Cuts? A Local Labor Markets Approach with Heterogeneous Firms

Transcription

1 Who Benefits from State Corporate Tax Cuts? A Local Labor Markets Approach with Heterogeneous Firms Juan Carlos Suárez Serrato Duke University & NBER Owen Zidar Chicago Booth & NBER Harvard March 9, 2015

2 I, like many economists, suspect that our corporate income tax is economically self-defeating hurting workers, not capitalists What can workers do to mitigate their plight? One useful step would be to lobby to eliminate the corporate income tax. That might sound like a giveaway to the rich. It s not. The rich, including Boeing s stockholders, can take their companies & run

3 We relax two crucial assumptions 1 Firms are perfectly competitive If firm owners earn zero profits, they can not bear incidence 2 Firms are perfectly mobile Every firm is marginal in their location decisions Allow for monopolistically competitive & heterogeneously productive firms

4 This Paper Question: What are the welfare effects of cutting corporate taxes in an open economy on workers, firm owners, and landowners? Contributions 1 New evidence on business location and local economic activity 2 New framework for evaluating welfare effects 3 New assessment of corporate taxation in an open economy

5 Who Benefits from State Corporate Tax Cuts? Our Estimate Standard Model Landowners Workers Firm Owners Workers

6 Context and Challenges Empirical: Gravelle 2011, Clausing 2013 Insufficient time series variation in US corporate rates Cross-country variation compares countries with dissimilar institutions Theoretical: Harberger-type general equilibrium with focus on open economy (Gravelle 2010) Computable General Equilibrium Models (Kotlikoff & Summers 1987, Kotlikoff et. al 2013)

7 Outline: 3 Steps 1 Develop spatial equilibrium model with firms Allow workers, firm owners, landowners to bear incidence Map reduced-form effects to parameters governing welfare 2 Reduced-form effects of corporate tax cuts Implement state apportionment system using establishment data Number of establishments increases by roughly 3.5% following a 1% corporate tax cut 3 Estimate incidence and structural elasticities Implement reduced-form incidence expressions Minimize distance between reduced-form expressions and estimates to estimate structural elasticities Evaluate consequences for equity & efficiency of corporate tax policy

8 Broader Contribution: Local Labor Markets with Firms Last few years - important link between workers and location Kline 2010, Moretti 2011, Busso et al 2013, Diamond 2013, Notowidigdo 2013, Suárez Serrato and Wingender 2012 This literature and benchmark models have representative/identical, perfectly competitive firms & no link between firms and location Incidence: Kotlikoff & Summers 1987, Gordon & Hines 2002 Locational: Rosen 1979, Roback 1982 Monopolistically competitive and heterogeneously productive firms

9 Roadmap 1 Model 2 Incidence Expressions, Identification 3 Data and Reduced-Form Analysis of Business Location 4 Incidence and Parameter Estimates 5 Policy Implications

10 Model

11 A Spatial Equilibrium Model with Firms You have to start this conversation with the philosophy that businesses have more choices than they ever have before. And if you don t believe that, you say taxes don t matter. But if you do believe that, which I do, it s one of those things, along with quality of life, quality of education, quality of infrastructure, cost of labor, it s one of those things that matter. Delaware Governor Jack Markell (11/3/2013) 1 1 Low wages aren t what it s about anymore : Delaware s governor on bringing jobs home, The Washington Post 11/3/2013.

12 A Spatial Equilibrium Model with Firms: Outline 1 Setup 2 Worker Location, Labor Supply Moretti (2011), Busso et al (2013) 3 Housing Market Kline (2010), Notowidigdo (2012) 4 Firm Location and Labor Demand Dixit-Stiglitz (1977), Krugman (1979), Melitz (2003) 5 Results: Incidence ẇ(θ), π(θ), ṙ(θ) ε LS (θ) and ε LD (θ), and b(θ)

13 Equilibrium in the Local Labor Market

14 Equilibrium in the Local Labor Market

15 Equilibrium in the Local Labor Market

16 Model Setup 1 Geography: Small open economy c C 2 Agents: N c households, E c establishments, representative landowner in each location c 3 Market Structure: Monopolistically competitive traded goods market for each variety j Global capital market Local labor market Local housing market 4 Timing: Steady state, exogenous tax shock, new steady state

17 Household Problem max h,x ln A }{{} amenitites + α}{{ ln h} + (1 α) ln X }{{} housing composite good s.t. rh + j J p j x j dj = w where X = ( j J ε PD +1 ε x PD j dj rh is housing expenditures ) ε PD ε PD +1 p j x j is expenditure on variety j Indirect Utility of a Worker: V W nc = a 0 + ln w c α ln r }{{ c + } ln A }{{ nc } Disposable income Amenities Āc+ξnc

18 Local Labor Supply Location choice: Workers choose location with max utility: max c a 0 + ln w c α ln r c + Ā }{{ c +ξ } nc. u c Local Population: N c = P ( V W nc (Log) Local Labor Supply: ) = max{v W c nc } = exp uc σ W c exp u c σ W ln N c (w c, r c ; Ā c ) = 1 σ W ( ln wc α ln r c + Ā c ) + C0 Key Parameter: σ W, dispersion of idiosyncratic preferences ξ nc

19 Housing Market Housing Market: Upward-sloping supply of housing: H S c = (B H c r c ) ηc B H c is housing productivity r c is price of housing With Cobb-Douglas H D c, HM equilibrium given by: ln r c = 1 (ln N c + ln w c ) +C 1 + η c }{{} 1 Housing Demand Key Parameter: η c elasticity of housing supply

20 Local Labor Supply: Key points People move into a local area when wages increase How many people move in depends on: 1 Dispersion of Idiosyncratic Preferences σ W Higher σ W means smaller inflows of people following wage increases 2 Housing Supply Elasticity η c Lower η c means rents get bid up more when people move in Higher σ W and lower η c make ε LS smaller, so LS is more vertical

21 Establishment Production

22 Local Labor Demand: Establishment Production Demand for variety j is y jc = I ( p jc ) ε PD P Establishment j produces its variety with the following technology y jc = γ B jc ljc }{{} kδ jcm 1 γ δ jc B c+ζ jc Firm Value Function V F jc = Taxes {}}{ ln(1 τs b ) (ε PD + 1) Factor Prices {}}{ γ ln w c δ ln ρ + B c } {{ } v c +ζ jc.

23 Location Choice & Local Establishment Shares Fraction of Establishments: ( ) E c = P Vjc F = max{vjc F } = Establishment Growth: c exp vc σ F c exp v c σ F ln E c,t = ln(1 τ b c,t) σ F (ε PD + 1) γ σ F ln w c,t + φ t + 1 σ F B c,t Key Parameter: Dispersion of idiosyncratic productivity σ F Larger σ F means lower responsiveness to tax changes

24 Local Labor Demand Aggregate labor demand for firms in location c: Elasticity of labor demand: L D c = E }{{} c E ζ [l (ζ jc ) c] }{{} Extensive margin Intensive margin ln L D c = γ 1 ln w c }{{} Substitution More elastic ε LD when: + γɛ PD }{{} Scale Higher output elasticity of labor γ Higher product demand elasticity ɛ PD γ σ F }{{} Firm Location ε LD Lower productivity dispersion σ F (i.e. firms more mobile)

25 Result: Local Incidence of State Corporate Taxes (1/2) Let ẇ c (θ) ln wc. Incidence on wages is: ln(1 τ b ) 1 (ε PD +1)σ F ẇ c (θ) = ( ) ( 1 + ηc α σ W γ ɛ PD (1 + η c ) + α }{{} ε LS Smaller wage increase if: σ F ) + 1 } {{ } ε LD 1 Productivity Dispersion σ F is large (i.e. immobile firms) 2 Preferences Dispersion σ W is small (i.e. mobile people) 3 Any other reason why ε LS and ε LD are large

26 Result: Local Incidence of State Corporate Taxes (2/2) Rental Costs: ṙ c (θ) = ( 1+ε LS 1+η c ) ẇ c Smaller rent increases if housing supply is very elastic Firm Profits: π c (θ) = 1 δ(ε PD + 1) }{{} + γ(ε PD + 1)ẇ }{{ c } Reducing Capital Wedge Higher Labor Costs Mechanical effects vs. higher production costs

27 Welfare Effects of Corporate Tax Cut Stakeholder Benefit Statistic Workers Disposable Income ẇ c αṙ c Landowners Housing Costs ṙ c Firm Owners After-tax Profit 1 δ(ε PD + 1) + γ(ε PD ( + 1)ẇ c ) = 1 + γ(ε PD + 1) ẇ }{{} c δ γ Labor cost factor Net Markup

28 Empirical Implementation and Identification

29 Structural Form of the Model where A = 1 α 1 0 σ W σ W 1 AY c,t = BZ c,t + e c,t ε LD 1+η 1 1+η 1 0, B = γ σ F 0 1 ε LD σ F (ε PD +1) 0 1 σ F (ε PD +1) Y c,t = [ ln w c,t ln N c,t ln r c,t ln E c,t ] Z c,t = [ ln(1 τ b c,t) ] e c,t is a structural error term

30 Exact Reduced Form of the Model Y c,t = A 1 B }{{} β Business Tax Z c,t + A 1 e c,t where β Business Tax is a vector of reduced-form effects of business tax changes: β W ẇ ẇε LS 1+ε LS β Business Tax = β N β R = β E µ 1 σ F 1+η ẇ γ σ F. ẇ

31 4 Reduced-Form Equations of the Model Effects on establishments, pop., wages, & rental cost growth over 10 years b ln w c,t = (ẇ(θ)) ln(1 τ }{{} c,t) + φ 1 t + uc,t 1 β W ( ) ln N c,t = ε LS ẇ(θ) ln(1 τc,t) b + φ 2 t + uc,t 2 }{{} β N ( ) 1 + ε LS ln r c,t = ẇ(θ) ln(1 τ 1 + η c,t) b + φ 3 t + uc,t 3 c }{{} β R ( 1 ln E c,t = σ F (ε PD + 1) γ ) σ F ẇ(θ) } {{ } β E ln(1 τ b c,t) + φ 4 t + u 4 c,t

32 Identification of Local Incidence on Welfare Reduced forms: ẇ = β W, Ṅ = β N = ε LS = βn β W Labor Demand ε LD = γ(ɛ PD + 1) γ 1 σ F Establishment Location ln D ln(1 t) = βe + γ β W σ F β W = β N β W β E + γ σ F β W γ(ɛ PD + 1) + γ σ F + 1 = γ(εpd + 1) = ( β N β E ) β W + 1

33 Identification of Local Incidence on Welfare Stakeholder Benefit Statistic Workers Disposable Income ˆβ W α ˆβ R Landowners Housing Costs ˆβ R ( ) Firm Owners After-tax Profit 1 + ˆβ N ˆβ E + 1 ( ˆβ ˆβ W δ W γ )

34 Benefits of the incidence formulae This framework enables us to: 1 Accommodate the conventional view 2 Transparently evaluate the sensitivity of our incidence estimates 3 Use data to govern relative factor mobility 4 Conduct inference and compare results to existing estimates

35 Data and Institutional Details of State Corporate Taxes

36 Non-Tax Data 1 Annual Data Number of establishments from County Business Patterns Population from BEA 2 Decadal Data Wage and rental cost indexes from Censuses and 2009 ACS Adjust for changes in composition of observable characteristics 3 Geographical Level Focus on county groups called consistent PUMAs [490 localities] 4 Bartik: Construct Bartik shock to predict labor demand: Bartik c,t = Ind EmpShare Ind,t 1,c Emp Ind,t,National

37 Three Types of Firm Taxes 1 Partnership and S-corps: τ INC personal income tax rate Synthetic changes as in Zidar (2013) using NBER s TAXSIM 2 Single-state C-corps: τ c corporate income tax rate Digitized corporate tax rates from Book of the States 3 Multi-state C-corps: τ A apportioned corporate income tax rate Depends on corporate rate, apportionment, and activity weights τ A i = s τ c s ω is ( where ω is = θ w s W is ) } W {{ } payroll ( + θs ρ R is ) } R {{ } property ( + θs x X is ) } X {{ } sales

38 Nike Apportionment Example

39 Nike Apportionment Example

40 Nike Apportionment Example (2/2) Suppose Nike earns $2 M of profit in every state Their tax liability differs based on how profits are apportioned State I. Using Payroll II. Using Sales Apportioned Profit ($M) OR 80 2 IL 10 2 AL 10 2 Corporate Tax Liability ($M) OR with τor c = 50% 40 1 IL with τil c = 10% AL with τal c = 0% 0 0 Total Tax Liability ($M) 41 3

41

42 Number of Corporate Tax Rate Changes by State: 77-12

43 Number of Corporate Tax Rate Changes by Region: 77-12

44 Gradual Shift Towards Sales Apportionment Average Sales Weight Year

45 Using Variation from Apportionment Goolsbee and Maydew (Journal of Public Economics, 2000) Use variation in payroll burden τ c s θ w s Find that reducing payroll weight from 33% to 25% increases manufacturing employment by 1% This paper τ A i = s τ c s ω is ( where ω is = θ w s W is ) ( + θs ρ R is } W {{ } } R {{ } } X {{ } payroll property sales Use RefUSA data to construct ω is for each firm i ) ( + θs x X is Take average of all local establishments to obtain τ A )

46 Average Business Tax Rate Use data on shares of establishments to calculate the average business tax in a conpsuma: ln(1 τ b ) c,t fc,t SC ln(1 τ c ) c,t + fc,t MC ln(1 τ A ) c,t }{{} Corporate + f P c,t ln(1 τ INC ) c,t }{{} Personal Calculate shares fc,t SC, fc,t MC, fc,t P using County Business Patterns and RefUSA data

47 Reduced-form Effects on Business Location (and Local Economic Activity)

48 Business Taxes & Establishment Location Specification ln E c,t ln E c,t 10 = β[ln(1 τ b c,t) ln(1 τ b c,t 10)] + D s,tψ s,t + u c,t LHS: Growth in number of establishments RHS: Growth in net-of-business tax rate D s,t is a vector of year dummies and state dummies for industrial Midwest in the 1980s

49 Validity of Business Tax Variation Potential for bias due to: Concomitant changes in corporate tax base, esp. tax credits Concomitant changes in spending Concurrent changes in productivity Prior economic conditions

50 Business Taxes & Establishment Growth Establishment Growth (1) (2) (3) (4) (5) (6) ln Net-of-Business-Tax Rate 4.07** 4.14** 4.06** 3.35** 3.91** 3.24** (1.82) (1.80) (1.83) (1.43) (1.78) (1.41) State ITC (0.32) (0.30) ln Gov. Expend./Capita (0.01) (0.01) Bartik 0.59*** 0.57*** (0.19) (0.18) Change in Other States Taxes -4.66*** -4.18*** (1.60) (1.43) Fixed Effects Year Year Year Year Year Year Observations 1,470 1,470 1,470 1,470 1,470 1,470 R-squared Tax changes & growth are over 10 years. *** p<0.01, ** p<0.05, * p<0.1 Robust standard errors clustered by state in parentheses Binscatter

51 Cumulative Effects of Business Tax Cuts on Est. Growth Percent F-test all leads are 0 has p-value= 0.92 F-test all lags are 0 has p-value= Year Cumulative Effect no leads Long Difference Point Estimate Cumulative Effect w/ leads 95 % Confidence Interval

52 Additional Validity Tests of Business Location Estimate Synthetic controls for states that change taxes Specifications over shorter durations that flexibly control for measures of prior economic conditions No detectable responsiveness of other state tax rates Bottom Line: The approx. 3.5% effect on establishment growth over ten years is robust and economically sensible Event Study Population Binscatter Population Binscatter Wage Binscatter Rental Cost

53 Estimating Incidence Using Reduced-Form Estimates

54 Identification of Local Incidence on Welfare Stakeholder Benefit Statistic Workers Disposable Income ˆβW α ˆβ R Landowners Housing Costs ˆβR ( ) Firm Owners After-tax Profit 1 + ˆβN ˆβ E + 1 ( ˆβ ˆβ W δ W γ ) Housing expenditure share α =.3 from Consumer Expenditure Survey Output Elasticity of Capital δ =.9γ from BEA

55 Economic Incidence Estimates Using RF Effects A. Incidence B. Share of Incidence (1) (2) (3) (4) (1) (2) (3) (4) Workers 1.10* 1.38** 1.10* *** 0.31*** 0.28*** 0.37 (0.59) (0.59) (0.59) (0.52) (0.09) (0.10) (0.09) (0.43) Landowners (1.43) (1.42) (1.43) (1.36) (0.19) (0.18) (0.19) (0.48) Firm Owners 1.63* 1.79** 1.63* *** 0.40*** 0.42*** 0.45*** (0.90) (0.80) (0.90) (1.40) (0.12) (0.11) (0.12) (0.13) Controls State FX N Y N N N Y N N ln Gov./Capita N N Y N N N Y N Bartik N N N Y N N N Y

56 Robustness: Economic Incidence Estimates Panel (a) Incidence (1) (2) (3) (4) (5) (6) (7) (8) (9) Workers 1.26* 1.43** 1.39** 1.38** 1.54** ** 1.23** 1.23** (0.67) (0.58) (0.59) (0.60) (0.63) (0.69) (0.59) (0.52) (0.52) Landowners * (1.57) (1.48) (1.44) (1.40) (1.74) (1.61) (1.42) (1.45) (1.45) Firm Owners 1.81** 2.17** 1.80** 1.80** 2.24** ** 1.65** 1.65** (0.85) (0.89) (0.80) (0.82) (0.92) (1.10) (0.79) (0.87) (0.86) Panel (b) Shares of Incidence (1) (2) (3) (4) (5) (6) (7) (8) (9) Worker 0.25*** 0.24*** 0.31*** 0.31*** 0.25*** 0.33* 0.31*** 0.33** 0.33** (0.10) (0.05) (0.10) (0.10) (0.08) (0.19) (0.10) (0.15) (0.15) Landowner 0.38** 0.40*** 0.30* 0.29* 0.39*** (0.16) (0.10) (0.18) (0.17) (0.12) (0.31) (0.18) (0.26) (0.26) Firmowner 0.37*** 0.36*** 0.40*** 0.40*** 0.36*** 0.44*** 0.40*** 0.45*** 0.45*** (0.09) (0.07) (0.10) (0.10) (0.08) (0.13) (0.11) (0.13) (0.13) Controls Political Y N N N N N N N N Tax Rule N Y N N N N N N N τ s N N Y N N N N N N τ s N N N Y N N N N N τ i N N N N Y N N N N τ i N N N N N Y N N N G/capita N N N N N N Y N Y Corp Rev to GDP N N N N N N N Y Y

57 Estimating Structural Parameters that Govern Incidence

58 Structural Estimation 4 Parameters of interest 4 Simultaneous equations with the following outcomes: 1 Establishment Growth 2 Population Growth 3 Wage Growth 4 Rental Cost Growth RF effects of Taxes on 4 Outcomes to estimate σ F, σ W, η Enhance precision with supplement labor demand (Bartik) Shocks 1 RF effects of Both Shocks on 4 Outcomes σ F, σ W, η 2 RF effects of Both Shocks on 4 Outcomes σ F, σ W, η, ε PD

59 Parameters θ 1. Estimated Parameters 1 Productivity Dispersion σ F 2 Preference Dispersion σ W 3 Housing Supply Elasticity η 4 Product Demand Elasticity ε PD 2. Calibrated Parameters Housing expenditure share α =.3 from Consumer Expenditure Survey Output Elasticity of Labor γ [.1,.3] from IRS, BEA Output Elasticity of Capital δ =.9γ from BEA residual of L, M

60 4 Reduced-Form Equations of the Model Effects on establishments, pop., wages, & rental cost growth over 10 years ( 1 ln E c,t = σ F (ε PD + 1) γ ) σ F ẇ(θ) ln(1 τc,t) b + φ 1 t + uc,t 1 }{{} ( ) ln N c,t = ε LS ẇ(θ) ln(1 τc,t) b + φ 2 t + uc,t 2 }{{} β N b ln w c,t = (ẇ(θ)) ln(1 τ }{{} c,t) + φ 3 t + uc,t 3 β W β E ( ) 1 + ε LS ln r c,t = ẇ(θ) ln(1 τ 1 + η c,t) b + φ 4 t + uc,t 4 c }{{} β R

61 Estimating Structural Parameters 1. Reduced Form: Estimate reduced form ˆb and covariance ˆV 2. Recover Structural Parameters via Classical Minimum Distance: ˆθ = arg min θ Θ [ˆb m(θ)] ˆV 1 [ˆb m(θ)] Results: Establishments Population Wage Rent Business Tax Predicted Moments Empirical Moments 4.074** (1.80) (1.46) (0.94) (1.42) χ 2 (1) Stat χ 2 P-Value 0.979

62 Enhancing precision with supplemental LD shocks Effects on establishments, pop., wages, & rental cost growth over 10 years ln E c,t = b 1 ln(1 τc,t) b + b 5 Bartik c,t + φ 1 t + ũc,t 1 ln N c,t = b 2 ln(1 τc,t) b + b 6 Bartik c,t + φ 2 t + ũc,t 2 ln w c,t = b 3 ln(1 τc,t) b + b 7 Bartik c,t + φ 3 t + ũc,t 3 ln r c,t = b 4 ln(1 τc,t) b + b 8 Bartik c,t + φ 4 t + ũc,t 4

63 8 Moments from Tax and Bartik Shocks Bartik and Tax Shock (γ =.15, ε PD = 2.5) Establishments Population Wage Rent Business Tax Predicted Moments Empirical Moments 3.354** (1.41) (1.27) (0.83) (1.35) Bartik Predicted Moments Empirical Moments 0.595*** 0.445** 0.557*** 0.702*** (0.19) (0.18) (0.08) (0.27) χ 2 (2) Stat χ 2 P-Value Note: ˆσ F = 0.17 (0.10), ˆσ W = 0.77 (0.31), ˆη = 2.47(5.10)

64 Structural Elasticities Using Estimated Parameters Panel (b) Demand and Supply Elasticities (1) (2) (3) Tax Only Tax & Bartik Output Elasticity γ Elasticity of Product Demand ε PD (10.337) Labor Mobility ** 1.379** 1 σ W (1.636) (0.535) (0.578) Elasticity of ** 1.163* Labor Supply (1.305) (0.541) (0.659) Micro Elasticity of Labor Demand (1.551) Macro Elasticity *** *** of Labor Demand (0.850) (0.510) (26.914)

65 Economic Incidence Using Estimated Parameters Incidence Shares of Incidence (1) (2) (3) (4) (5) (6) Calibrated Parameters Tax Only Tax & Bartik Tax Only Tax & Bartik Output Elasticity γ Elasticity of Product Demand ε PD (10.337) (10.337) Estimated Parameters Wages ẇ 1.438* 1.211** (0.798) (0.592) (0.708) Landowners ṙ (1.329) (1.241) (1.298) (0.251) (0.338) (0.463) Workers ẇ αṙ 1.090** 0.994*** 0.847** 0.348*** 0.375*** 0.372** (0.476) (0.316) (0.419) (0.105) (0.145) (0.152) Firm Owners π 0.879*** 0.930*** 0.908* (0.180) (0.133) (0.512) (0.191) (0.220) (0.405)

66 Firm Owner s Share of Incidence for Calibrated Values of γ and ε PD Elasticity of Product Demand: ε PD * Baseline Output Elasticity of Labor: γ Share to Firm Owners

67 Two Additional Considerations 1 Regional Heterogeneity We document average effects, but regions can vary (e.g., housing market elasticities η c ) equity and efficiency impacts vary Everything is bigger in Texas, including the efficiency costs of business location incentives 2 Accounting for (small) Government Spending Changes Quantify 3 scenarios: cutting services, infrastructure, both Expenditure shares on services exceed those on infrastructure, so worker amenities hit more Shared impact even for infrastructure only case (lower productivity lower wages) This reinforces conclusion that firm owners enjoy substantial portion of benefit Incidence accounting for government spending changes

68 Government Revenues, Behavioral Responses, & Efficiency Q: Given these behavioral responses, why do we observe low state corporate taxes? Fiscal externalities, not mobility may explain why states have low rates Amenable feature of state corporate tax system

69 Revenue-Maximizing Corporate Tax Rate 1 If states wanted to maximize corporate tax revenues, the maximal tax rate would be: τ c = 1 π c + Ė c 2 However, this rate doesn t account for fiscal externalities from other taxes (or from other spending) τ c = 1 π c + Ėc + (revshare pers c /revshare C c )(ẇ c + Ṅc), 3 Depends on size of location (e.g. states versus cities). It is likely that more local smaller σ F smaller t 4 Depends on policy design: source based versus destination based

70 Corporate Rates vs Revmax Rate w/ Fiscal Externalities Corporate Tax Rate in OH HI MO SC SD IA ID NM MI OK NE CT AZ IN KYGA VA CO MN RI ME WVVT MD LA OR WI UT MS PA KS NY NC MT AL AR ND Revenue-Maximizing Rate with Fiscal Externalities FL NJ DE CA MA TN IL

71 Rates, Fiscal Externalities, and Apportionment Corporate Tax Rate in RI NJ DE CA MA OH WV VT MD NMIDCT KS AZ MTNC HI AL AR MO ND TN OK KY VA FL MSUT MN IN PA NH AK Revenue-Maximizing Rate with Fiscal Externalities and Sales Apportionment

72 Revenue-Maximizing Corporate Tax Rates Sales Apport. Corporate Revenue Max. Corp. Rate State Weight θs x Tax Rate τ s τs τs τs /(1 θs x ) Kansas Indiana U.S. Avg U.S. Med U.S. Min U.S. Max

73 Conclusion Conventional view: corporate taxation in an open economy hurts workers since shareholders can take their companies and run 1 New Measure of Local Business Taxes 2 New Reduced Form-Effects 3 New Tractable Spatial Equilibrium Framework with Firms New Assessment: in terms of equity and efficiency, corporate taxation in an open economy may not be as bad as we thought

74 APPENDIX

75 Incidence Estimates Accounting for Government Spending Back (1) (2) (3) (4) Assumptions for Analysis Value of Government Services N Y N Y Value for Infrastructure N N Y Y Change in Funds None Services Infrastructure Proportional Incidence Landowners Firm Owners Workers Share of Incidence Landowners 18% 23% 19% 23% Firm Owners 45% 59% 41% 57% Workers 38% 18% 40% 21%

76 Cumulative Effects of Business Tax Cuts on Pop. Growth Percent Year Cumulative Effect no leads Long Difference Point Estimate Cumulative Effect w/ leads 95 % Confidence Interval

77 10 Yr Effect of Business Tax Cut on Establishment Growth 10 Year Log Change in Establishments Slope= 3.94 (1.61) Year Log Change in Net of Business Tax Rate Back

78 10 Yr Effect of Business Tax Cut on Population Growth 10 Year Log Change in Population Slope= 3.94 (1.57) Year Log Change in Net of Business Tax Rate Back

79 10 Yr Effect of Business Tax Cut on Wage Growth 10 Year Log Change in Composition Adj. Wages Slope= 1.78 (.72) Year Log Change in Net of Business Tax Rate Back

80 10 Yr Effect of Business Tax Cut on Rental Cost Growth 10 Year Log Change in Composition Adj Rents Slope= 2.16 (1.4) Year Log Change in Net of Business Tax Rate Back