EFFECTS OF A RECESSION ON THE TAX BURDENS DISTRIBUTION: DISTRICT OF COLUMBIA HOUSEHOLDS 2007 TO Demas-Rampersad Seminar.

Transcription

1 EFFECTS OF A RECESSION ON THE TAX BURDENS DISTRIBUTION: DISTRICT OF COLUMBIA HOUSEHOLDS 2007 TO 2009 Demas-Rampersad Seminar The University of The West Indies at St. Augustine Trinidad April 24 th, 2014 Betty Alleyne, OCFO betty.alleyne@dc.gov Daren Conrad, University of the West Indies daren.conrad@sta.uwi.edu Kelly Dinkins, OCFO kelly.dinkins@dc.gov Daniel Muhammad, OCFO daniel.muhammad@dc.gov All opinions expressed here are solely those of the authors and not the DC OCFO or the University of the West Indies.

2 About the Study 2

3 Background a) Difficulties in empirical analysis re taxation studies: Securing adequate data; Disassociation of the effects attributable to a tax reflecting changes in other economic forces; and Measuring tax burdens. a) Tax burdens are used as indicators of how well policies meet the goal of equity in raising revenue to run government. c) Understanding tax burdens provide answers to broader economic and socioeconomic questions about the effect of tax policies on the distribution and potential redistribution of income and wealth

4 The Study PURPOSE a) To identify whether there were significant changes in the overall distribution of household tax burdens in the District of Columbia during the recession. b) If so, identify some of the major causes for the redistribution of citywide tax burdens. Significance of Study Unlike many tax burden studies that rely heavily on methodological assumptions, this study conducts a microanalysis of actual household tax liabilities for owner-occupied households over time. 4

5 The U.S. Economy: 2007 to 2009 Aggregate Demand and Aggregate Supply P Aggregate Demand Aggregate Supply Price Level P 2007 P 2009 Y 2009 Y 2007 Y Output, Income

6 Literature Review 6

7 Tax Burdens There is little detailed information on the tax burden which each state imposes on its residents or on how the burden varies with income and macroeconomic business cycles (Phares, 2003). Taxation occupies the important position of distribution and by extension, which is of equal and potentially greater importance, distributive justice. Consequently, any discussion on taxes must address the issue of the distribution of the burdens of social cooperation (Farrelly, 2004). Burden tables even do a bad job of showing the initial incidence of a tax (IRET No. 88, 2004).

8 Tax Burdens The true measure of a tax burden is the change in the individuals economic situation as a result of the tax after all economic adjustments are taken into account (IRET No. 88, 2004).

9 Data Standardization & Matching 9

10 Standardization a) Considered only residential properties receiving homestead exemption (reduced real property assessed value by $67,500 prior to computing the yearly tax liability). a) Standardize name and address fields in both RPT and IIT files. For Example Doe, Jane 1122 nd Street, NW Washington or J. Doe nd Street N.E. Washington, DC c) Standardization scheme was used to parse the address field of each dataset into several tokens. Standardization scheme was used to standardize taxpayer names on the IIT files and homeowner names on the RPT file.

11 Matching a) Assigned match code to make the address and name unique in both files based on locale, match definition, and sensitivity level b) End result of matching: TY 2007: 54,601 records TY 2008: 55,614 records TY 2009: 53,772 records c) Comprehensive Taxpayer Dimension table

12 Data Analysis 12

13 Data a) For this study, household tax burden is defined as the combined total of DC individual income (IIT) and real property (RPT) taxes relative to federal adjusted gross income (AGI) of District of Columbia residents that live in owner-occupied households. b) The matched data set (MDS) contains matched IIT and RPT tax records for over 53,000 owner-occupied households for each of the three years under study. The MDS was generated by matching records by name and address. c) The tax years under investigation are 2007, 2008 and 2009.

14 * How Relevant is the Matched Data Set? *Individual Income Tax 18% of all Individual Income Tax Filers 30% of all Adjusted Gross Income and Individual Income Taxes Paid *Real Property Tax 59% of all Registered Homesteads 70% of all Homestead Value & Property Taxes Paid by Homesteads *Total Over $570 million in Tax Revenue and Over 34% of all DC Individual Income and Residential Real Property Taxes Paid 14

15 * Distribution of Individual Income Tax Filers by Filer Type All District Filers vs. Matched Data Set CY2007 CY2008 CY2009 Filer Type The District Matched Data The District Matched Data The District Matched Data Single 58.6% 51.9% 61.1% 53.1% 61.9% 52.7% Married 15.6% 29.9% 15.9% 29.3% 16.3% 30.0% Head of Household 19.5% 14.3% 19.8% 13.9% 18.6% 13.6% Married Filing Separate 3.2% 3.8% 3.2% 3.7% 3.2% 3.7%

16 8.00% Effective Tax Rates 7.00% 6.00% 5.00% 4.00% 1.53% 5.50% 1.78% 5.30% 2.15% 5.10% 3.00% Income Tax Real Property Tax 16 Total Effective Tax Rates 2007: 7.03% 2008: 7.08% 2009: 7.25%

17 *District of Columbia Ward Map Ward 4 Ward 3 Ward 1 Ward 5 Ward 2 Ward 6 Ward 7 Ward 8

18 * Matched Data Set Statistics By Wards Percentage Point Changes in ETRs, 2007 to 2009 Wards IIT/AGI RPT/AGI TB/AGI Ward % 0.48% 0.46% Ward % 0.75% 0.43% Ward % 0.73% 0.17% Ward % 0.50% 0.26% Ward % 0.25% 0.37% Ward % 0.16% 0.36% Ward % 0.15% 0.23% Ward % 0.11% 0.17% Total -0.3% 0.6% 0.2% 18

19 * Average AGI by Ward, 2007 and 2009 Wards 2007 Average AGI 2009 Average AGI % Change Ward 1 $124,713 $121, % Ward 2 $263,598 $213, % Ward 3 $394,046 $300, % Ward 4 $143,341 $130, % Ward 5 $64,250 $66, % Ward 6 $82,103 $92, % Ward 7 $53,729 $56, % Ward 8 $47,028 $49, % Total $186,511 $156, % 19

20 * Matched Data Set Statistics By Income Level Total Percent Changes, 2007 to 2009 Income Levels AGI IIT Liability AV RPT Liability Total Tax Burden LT $50, % -17.8% 14.9% 22.2% 8.5% $50,000 - $75, % -5.8% 16.2% 24.6% 6.6% $75,000 - $100, % -4.0% 12.5% 21.7% 5.0% $100,000 - $200, % -1.7% 6.6% 16.8% 4.0% $200,000 - $500, % -4.0% 5.8% 15.9% 0.6% GT $500, % -27.5% 4.5% 13.1% -24.0% Total -13.0% -18.2% 10.1% 19.4% -10.0% 20

21 * Total Percentage Point Change in Effective Tax Rates By Income Level 2007 to 2009 Income Levels IIT/AGI RPT/AGI TB/AGI LT $50, % 4.9% 5.5% $50,000 - $75, % 0.6% 0.4% $75,000 - $100, % 0.5% 0.3% $100,000 - $200, % 0.4% 0.3% $200,000 - $500, % 0.3% 0.2% GT $500, % 0.2% -0.4% Total -0.3% 0.6% 0.2% 21

22 * Effective Tax Rates for the Matched Data Set by Deciles - TY 2007 vs. TY2009

23 * Share of Taxes Paid for the Matched Data Set by Deciles - TY 2007 vs. TY2009

24 * Percentage Point Change in Shares of Total Taxes Paid for the Entire MDS by Deciles, TY2007 vs. TY2009

25 * Percentage Point Change in Shares of Taxes Paid for Capital Gains Filers by Deciles, TY2007 vs. TY2009

26 * Percentage Point Change in Shares of Taxes Paid for Business Income Filers by Deciles, TY2007 vs. TY2009

27 Some of the burden of taxes has shifted from very high income to lower income taxpayers over the study period. The Share of Total Taxes Paid by Income Group- MDS 100% 90% 80% 70% 60% 50% 40% 30% 20% 38.7% 25.5% 74.5% 61.3% Income LT $500k Income GT $500k The Share of IIT Paid by Income Group- MDS The Share of RPT Paid by Income Group- MDS 100% 80% 60% 40% 47.4% 52.6% 33.4% 66.6% 100% 80% 60% 40% 13.3% 9.9% 86.7% 90.1% 20% Income LT $500k Income GT $500k 20% Income LT $500k Income GT $500k

28 Wage Losses Among Common Filers in 2007 and 2009 Filers with some wage/salary losses in 2009 Income Groups N Share % Chg in W/S Share of WS $1-30k 23, % -40.2% 7.3% $30-50k 13, % -34.2% 10.1% $50-70k 7, % -34.7% 8.0% $70-90k 4, % -34.1% 6.3% $90-110k 2, % -33.6% 5.1% GT $110 9, % -50.6% 63.2% Total 59, % -45.0% 100.0% 59,571 filers with some wage losses paid $149.4 million less in taxes in These filers account for about 20% of all AGI wage: mean $86,311, median $38,569 6,964 filers with total wage losses paid $20.3 million less in taxes in These filers account for about 2% of all AGI wage: mean $48,074, median $22,432 Filers with total loss of all wage/salaries in 2009 Income Groups N Share % Chg in W/S Share of WS $1-30k 4, % 100.0% 12.6% $30-50k 1, % 100.0% 12.3% $50-70k % 100.0% 11.0% $70-90k % 100.0% 9.6% $90-110k % 100.0% 6.6% GT $ % 100.0% 47.9% Total 6, % 100.0% 100.0%

29 Some of the city s tax burden shifted from households with income GT $500k to households with income LT $500k households. The lower effective tax rates for high income households appears to be caused by major declines in AGI (capital gain losses). IIT effective tax rates in DC East went up, and IIT effective tax rates in the DC West went down. RPT effective tax rates in DC West went up more than the RPT effective tax rates in DC East. The biggest change to the DC economy in 2009 was the 13% decline in IIT revenue (-$158 million). This appears to be correlated to a 40% decline in the S&P 500 in 2008 and a 61% ($1.9 bill.) decrease in DC capital gains in Even though total AGI and total tax revenue declined, wages/salaries and business income increased for several categories of taxpayers.

30 Tax reform requires political fortitude. Existing beneficiaries could see a loss of income and real property values while others may see a gain in income and wealth ISSUES THAT NEED CONSIDERATION Should the tax system be income based or consumption based? Should we have high tax rates and liberal exemptions or should we have low tax rates levied on broad bases? Do we want horizontal equity or do we want wide differentiation? How much redistribution should be reasonably expected?

31 *Conclusions *The recession caused a significant redistribution in the share of taxes paid. The burden shifted away from high income filers to the middle class and low income filers. *Real property tax policy (and not market values) are the cause of higher real property effective tax rates in 2008 and *Unlike most tax burden studies, this analysis of actual microlevel tax data for households allows for a high degree of insight into underlying changing dynamics that caused a significant redistribution of the city s tax burden during the most recent recession. 31

32 CGE Models Computable general equilibrium (CGE) models have been used since the early 1970s to analyse the economic effects of changes in taxation (see Shoven and Whalley, 1992; Greenaway et al., 1993). Estimation of burdens using the CGE