Upload
hayley
View
42
Download
3
Tags:
Embed Size (px)
DESCRIPTION
State Tax Policy and Entrepreneurship. Donald Bruce, Xiaowen Liu, and Matthew Murray Center for Business and Economic Research and Department of Economics The University of Tennessee, Knoxville Conference on Subnational Government Competition The University of Tennessee April 25, 2014. - PowerPoint PPT Presentation
Citation preview
Donald Bruce, Xiaowen Liu, and Matthew MurrayCenter for Business and Economic Research and
Department of EconomicsThe University of Tennessee, Knoxville
Conference on Subnational Government CompetitionThe University of Tennessee
April 25, 2014
State Tax Policy and Entrepreneurship
• States have a long history of using income and sales tax policy to compete for mobile entrepreneurs and/or encourage new ones– 2012: Kansas removes income tax on pass-
through income– 2014: Missouri considers 25% deduction of pass-
through income– Other states considering income tax repeal, with
small business promotion as one selling point• Empirical literature has not generally
supported this at the state level
Policy Background
• Intensive-margin indicators of success–Most prior work is on extensive margin– Policy makers care more about success
• Dynamic panel regression–Most prior work uses fixed effects–Underlying trends matter
• Expanded specification and time period– 1978-2009
Contributions
• Nonfarm Proprietors’ Income (NFPI)– Per capita– As a % of total state personal income– As a % of national NFPI
• Nonfarm Proprietors’ Employment (NFPE)– As a % of total state employment– As a % of national NFPE
• Nonfarm Proprietors’ Productivity– NFPI/NFPE
Measures of Performance
• Past performance can predict future• Arellano-Bond (1991) estimator
– Y is the outcome of interest– X includes state policy characteristics– Z includes state economic/demographic
factors– is a state fixed effect– is a year fixed effect– is a well-behaved error
Empirical Approach
• Addresses time series issues in panel data
• Model transformed to first-difference– Removes state fixed effect
• Inclusion of lagged Y raises endogeneity concern; external instruments not required
• Arellano-Bover (1995) / Blundell-Bond (1998) approach used as an alternative
AB 101
• Policy– Sales tax rate– Top marginal personal income tax (PIT) rate– Top marginal corporate income tax (CIT) rate– Sales factor weight in CIT apportionment– Per capita state government expenditures– Tax Amnesty programs
• Economic/Demographic– Unemployment rate– % of population aged 65 and older– Crime rate– % female– Agricultural, Manufacturing % of GSP– Nonfarm job growth– Population density
Independent Variables
NFPI and NFPE, 1978-2009
$0
$200,000,000
$400,000,000
$600,000,000
$800,000,000
$1,000,000,000
$1,200,000,000
$1,400,000,000
0
5,000,000
10,000,000
15,000,000
20,000,000
25,000,000
30,000,000
35,000,000
40,000,000
NFPI (left)NFPE (right)
NFPI/E Shares and Productivity
0.000
0.050
0.100
0.150
0.200
0.250
0.000
5.000
10.000
15.000
20.000
25.000
30.000
35.000
40.000
45.000NFPE Share (left)NFPI Share (left)NFProductivity (right)
Summary StatisticsVariable 1978 2009
NFPI ($1,000) 9,666,381 17,900,000
NFPI per capita 2,189 2,745
NFPI as a share of total income (%) 8.36 7.06
State share of national NFPI (%) 2 2
NFPE (1,000) 256,572 708,628
NFPE as a share of total employment (%) 12.21 20.05
State share of national NFPE 2 2
Nonfarm proprietors’ productivity 35,784 23,247
Summary StatisticsVariable 1978 2009Sales tax rate 3.54 5.07Top PIT rate 6.90 5.47Top CIT rate 5.95 6.56Expenditures per capita 2.88 5.55Sales factor weight 32.3 57.2Unemployment rate 5.62 8.45Age > 64 10.72 13.16Crime rate 4.39 2.94Female percentage 51.02 50.58Nonfarm job growth 5.72 -4.18Agriculture share of GSP 3.90 1.61Manufacturing share of GSP 20.91 11.10Population density 153 193Amnesty 0 0.14
AB Results
NFPI NFPE
Per CapitaAs a Share of
Total Income
State Share of National NFPI
As a Share of Total
Employment
State Share of
National NFPE
Productivity
Sales tax rate –0.009 0.009 –0.017 –0.019 0.004 –0.052 (0.016) (0.042) (0.014) (0.016) (0.024) (0.193)Top PIT rate 0.006 0.028 –0.001 0.001 –0.005 –0.030 (0.011) (0.029) (0.006) (0.012) (0.004) (0.138)Top CIT rate –0.009 –0.020 0.007 –0.004 –0.003 –0.173 (0.019) (0.047) (0.047) (0.012) (0.005) (0.188)Expend. per capita 0.051 0.104 0.044 0.063 0.0004 0.760*** (0.045) (0.077) (0.032) (0.050) (0.016) (0.186)Sales factor weight –0.001* –0.001 0.0002 0.001 –0.0002 –0.011** (0.001) (0.001) (0.001) (0.001) (0.0002) (0.005)
• Main result echoes Bruce & Deskins (2012): taxes generally don’t matter– Higher sales factor weight lower NFPI per capita and lower NFP
productivity– Higher state gov’t. expend. per capita higher NFP productivity
• Other controls matter– Higher unemployment higher NFPI/E shares– Older population lower NFPE share; higher NFP productivity– Higher NF job growth higher NFPI/E shares; lower NFP
productivity– Lower crime rate higher NFPI per capita– Higher pop. Density higher NFPE share– Lower GSP shares in manufacturing or agriculture higher NFPI
• Lags always matter (dynamic specification is important)
Discussion
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins
Sales tax rate 0.065
Top PIT rate –0.046
Top CIT rate –0.025
Expend. per capita –0.028
Sales factor weight –0.001
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins Arellano-Bond
Sales tax rate 0.065 –0.019
Top PIT rate –0.046 0.001
Top CIT rate –0.025 –0.004
Expend. per capita –0.028 0.062
Sales factor weight –0.001 0.001
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins Arellano-Bond
Simple Replication
Sales tax rate 0.065 –0.019 0.043
Top PIT rate –0.046 0.001 –0.015
Top CIT rate –0.025 –0.004 –0.079**
Expend. per capita –0.028 0.062 0.051
Sales factor weight –0.001 0.001 0.003
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins Arellano-Bond
Simple Replication
Simple Replication and
Different Controls
Sales tax rate 0.065 –0.019 0.043 0.019
Top PIT rate –0.046 0.001 –0.015 0.023
Top CIT rate –0.025 –0.004 –0.079** –0.086***
Expend. per capita –0.028 0.062 0.051 0.096
Sales factor weight –0.001 0.001 0.003 0.004***
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins Arellano-Bond
Simple Replication
Simple Replication and
Different Controls
Simple Replication and Different Time
Sales tax rate 0.065 –0.019 0.043 0.019 –0.216***
Top PIT rate –0.046 0.001 –0.015 0.023 –0.010
Top CIT rate –0.025 –0.004 –0.079** –0.086*** –0.062**
Expend. per capita –0.028 0.062 0.051 0.096 0.038
Sales factor weight –0.001 0.001 0.003 0.004*** 0.002
Comparison to Bruce & Deskins (2012)
Employment stockBruce & Deskins Arellano-Bond
Simple Replication
Simple Replication and
Different Controls
Simple Replication and Different Time
Simple Replication and Arellano-Bond
Sales tax rate 0.065 –0.019 0.043 0.019 –0.216*** –0.046
Top PIT rate –0.046 0.001 –0.015 0.023 –0.010 0.117***
Top CIT rate –0.025 –0.004 –0.079** –0.086*** –0.062** 0.002
Expend. per capita –0.028 0.062 0.051 0.096 0.038 0.003
Sales factor weight –0.001 0.001 0.003 0.004*** 0.002 0.008**
• Simply adding (1) variables or (2) years of data or (3) estimating an AB model would have generated misleading significance
• The combination of these three updates is important
• The additional years of data (2003-2009) are enough to drive the importance of a dynamic estimation approach
Discussion