With major tax reform once again heading the list of legislative priorities, many analysts are hoping that the 108th Congress will make tax law far less a factor in economic decisions than it is today. Making the tax code more eco­nomically neutral is particularly desirable when it comes to the tax treatment of divi­dends paid on investments, since millions of Americans, including many who earn less than $100,000 a year, earn dividend income.

Corporations pay dividends to investors out of after-tax profits, yet these dollars are then subject to another tax: the personal income tax paid by the individuals who receive them. This multiple taxation of dividends discourages investors from finding equity investments that pay a steady stream of income and encourages corporations to retain profits and seek debt financing solely for tax purposes rather than financing their operations with equity.^[1] Indeed, these effects may well mean that the tax treatment of dividends is contributing to some of the economy's worst problems, under­scoring the hope that Congress will make the tax treatment of dividends more economically neutral.

This economic neutrality could be achieved in a number of ways.^[2] For example, corpora­tions could be allowed to deduct dividend payments from their taxes (just as they do with interest payments), or individual investors could be given tax credits on their personal tax returns using the "imputation credit" method. Other countries that have eliminated or reduced the multiple layers of taxation on divi­dends include Australia, France, Italy, Japan, Germany, Canada, and Great Britain.^[3] All of the possible methods have the same goal: reducing the multiple layers of taxation on cor­porate income.

Consider this example. Assuming a marginal tax rate of 35 percent, for every $100 of pre-tax corporate profits, the federal government initially will take $35. It then will tax the remaining funds again, depending on how much of the $65 the firm decides to pay out as dividends to investors. If the entire $65 were distributed directly to individuals as dividends, the full amount would be taxed as ordinary income to individual investors who receive the dividends.^[4] An individual in the 27 percent marginal tax bracket would pay approximately $18 in taxes upon receiving the dividend, leaving only $47 of the original $100 in corporate profits actually paid out to investors. In other words, for every $100 in pre-tax profits, the federal government will take approximately $53-more than half of every dol­lar earned.

Contrary to common perceptions, such double taxation of dividends affects nearly all Americans, not just a select few. As this report will show, most Americans stand to gain from the elimination of the double tax on dividend dollars. The reasons:

• Corporations exist in conjunction with peo­ple. The corporation exists only as a legal entity owned and operated by individuals. By definition, all corporate taxes must affect the income available to the owners and workers of the corporation. Since people, not "corpora­tions," pay taxes, taxing income at both the corporate and personal level results in taxing the same income twice.
• The double taxation of dividend dollars affects all investments. All investment vehi­cles compete for investors' dollars; therefore, any noticeable change in one investment's rate of return must affect the others' relative rate of return.^[5]      Furthermore, all true dividends are subject to this double taxation.       
• Millions of families would benefit from divi­dend tax relief whether or not they directly receive dividends. By the end of 2000, about 42 million workers in the United States with a median annual salary of less than $60,000 owned a 401(k) plan. Similarly, at the start of 2002, some 40 percent of U.S. households with a median annual household income of $55,000 owned an individual retirement account (IRA). Since the same investments held in retirement accounts are also held in non-retirement forms, these families with pri­vate retirement accounts would also share in the benefits of dividend tax relief.
• Many moderate-income workers receive dividends. In 1998, 70 percent of all taxpay­ers directly receiving dividends earned less than $55,000 in wages and salary. In the same year, 90 percent of all single taxpayers directly receiving dividends earned less than $53,000 in wage and salary income, and 70 percent of all married couples directly receiving divi­dends earned less than $76,000 in wage and salary income in 1998. Even after considering issues regarding how the Internal Revenue Ser­vice (IRS) reports some non-dividend income as "dividends," it is unlikely that only the top one or two income groups would benefit from dividend tax relief.

dispelling Myths about dividend income

A common misconception is that dividend tax relief would benefit only a select few Americans. Part of the reason has to do with how the IRS clas­sifies "dividends."

A recent article by William Gale of the Tax Pol­icy Center, for example, notes that "just under half-46 percent (93/201)-of dividends paid out of the corporate sector were subject to double tax­ation in 2000."^[6] The ratio mentioned in these cal­culations refers to two measures of dividends: the IRS measure of dividends ($93 billion) and the National Income and Product Accounts (NIPA) definition of dividends ($201 billion). Interest received from mutual funds is reported as divi­dend income on the IRS 1040 form, and a portion of NIPA dividends includes distributions from S corporations.^[7] Neither item is subject to double taxation, but neither item represents dividends.

Regardless of how various types of earned income are classified by the IRS, all real dividends are subject to double taxation. Even when divi­dend-paying equities are held in retirement accounts, they are subject eventually to double taxation.

It can also be argued that the harmful effects of the double taxation of dividends are not simply confined to the real dividends declared on individ­ual tax returns.^[8] In fact, the effects of the tax struc­ture are not necessarily confined solely to equity markets. Individuals choose from an array of investment choices based on the real after-tax rate of return they expect. Since all the various invest­ment vehicles are competing for investors' dollars, any material change in one investment's return has to affect the relative return of others. Therefore, the double taxation of dividends indirectly affects the returns of all investments-even those in tax-deferred retirement accounts.

In 2001, nearly $2.4 trillion in retirement plan assets were invested in mutual funds, with roughly a 50/50 split between IRAs and employer-spon­sored retirement plans such as 401(k) plans.^[9] Nearly $1.5 trillion of these assets were held in domestic equities. While these equities were held in tax-deferred retirement accounts through mutual funds, at the same time, they were also being held by non-mutual fund and non-retire­ment account investors.

Thus, any given equity investment is likely to be held individually, in mutual funds, and in retire­ment accounts simultaneously. Since dividend tax relief means that firms are more likely to increase their dividend payments, the benefit of this type of tax relief will also accrue to individuals holding equities in retirement accounts even though the benefit will not immediately show up on their tax returns.^[10]

Exactly on which tax returns these benefits will appear is another interesting question. By the end of 2000, roughly 42 million U.S. workers with a median annual salary of less than $60,000 partici­pated in a 401(k) plan.^[11] Similarly, approximately 42 million U.S. households with a median annual household income of $55,000 (40 percent) owned some type of IRA.^[12] Data also show that, as of 2002, nearly 53 million U.S. households (just under 50 percent) held some form of equities and about 84 million individuals owned some form of equities.^[13] Given that these equities exist in both retirement and non-retirement accounts, and that the double taxation of dividends affects the relative returns of all investments, it should be clear that eliminating the double taxation of dividends will not benefit only a select few, but rather millions of Americans. 

In the post-retirement group, however, the 80th percentile taxpayer reports dividend income in all three filing groups. Table 6 shows that single filers at the 80th percentile, with an after-tax income of $34,740, report dividend income over $6,000; married filers, with after-tax income of about $64,000, have dividend income over $4,000.

Even given this higher proportion of dividend-receiving taxpayers in the post-retirement group, using only the data in Table 1 through Table 6, it is still rather easy to support a claim that only the wealthy will benefit from dividend tax relief. One could argue, for example, that since 80 percent of all taxpayers receive less than $46 in dividends (Table 2, Panel A), dividend tax relief benefits only the upper 20 percent of taxpayers.

A problem with this logic is that the data in these six tables do not distinguish between those taxpayers who receive dividends and those who do not. Therefore, these data do not accurately describe the characteristics of taxpayers who receive dividend income. For this reason, Tables 7 through 12 present summary statistics and percen­tiles using the same groups of taxpayers but exam­ining the data only for taxpayers with dividend income.

The statistics reported in Tables 7 through 12 paint a very different picture. An examination of all dividend-receiving taxpayers (Table 7, Panel A) shows that their median after-tax income is just under $44,000, with dividend income of $478, interest of $435, and capital gains of $220.^[18]

conclusion

• <_spanarial27_22_>Wages, salaries, tips, etc., from employees? W-2 forms;
• <_spanarial27_22_>Alimony received;
• <_spanarial27_22_>Business income or loss;
• <_spanarial27_22_>Other gains or losses (sales of business prop­erty);
• <_spanarial27_22_>Total taxable IRA distributions;
• <_spanarial27_22_>Total taxable pensions and annuities;
• <_spanarial27_22_>Rental real estate, royalties, partnerships, S corporations, trusts, etc.;
• <_spanarial27_22_>Farm income;
• <_spanarial27_22_>Unemployment compensation; and
• <_spanarial27_22_>Social Security benefits.           

  <_spanarial27_22_>Aside from those already mentioned, using the data in the SOI Public Use Tax File for this pur­pose presents two major shortcomings. It should be noted that the amount reported as ?wages, sala­ries, tips, etc.? on the W-2 form does not include payroll taxes or any wages and salary contributed to employer-sponsored tax-advantaged savings plans: cafeteria plans, 401(k) plans, etc. Also, ?div­idend income? on the IRS form 1040 does include cash distributions that are actually interest pay­ments, not dividends.^[22]

  <_spanarial27_22_>Next, the author combined the total tax owed or refunded with the above types of income to create a variable for ?after-tax income.? The after-tax income variable basically includes all income other than that earned from investments. In other words, it excludes income from dividends, capital gains, and interest. Therefore, the calculated after-tax income serves as a broad measure of taxpayers? ?permanent? income.^[23]

  <_spanarial27_22_>Summary statistics (mean, median, and stan­dard deviation) and percentiles are calculated for all returns as well as for those returns reporting dividend income. As reported above, when all returns in the 1998 SOI file are considered, more than half of all taxpayers do not report dividend income. Therefore, focusing on all returns alone does not provide a clear picture of the income (and other characteristics) of those individuals reporting dividend income.

  <_spanarial27_22_>Consequently, the author examined filers in every filing status (single, married filing jointly, married filing separately, head of household, or widower) for all ages in two separate categories? those receiving dividends and those not receiving dividends. Additionally, the data in both categories are partitioned based on filing status for both pre-retirement age and post?retirement age taxpay­ers.

  <_spanarial27_22_>In regard to filing status, the author first com­bined all filers and then examined single and mar­ried filers separately (married filing separately and married filing jointly were combined as married filers). Finally, while it is not a perfect classifica­tion, all returns including Social Security benefits are categorized as ?post-retirement age,? and all returns not reporting such benefits are categorized as ?pre-retirement age.?

   

  
  
  

  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  The statistics reported in Tables 7 through 12 paint a very different picture. An examination of all dividend-receiving taxpayers (Table 7, Panel A) shows that their median after-tax income is just under $44,000, with dividend income of $478, interest of $435, and capital gains of $220.^[18]

  <_spanarial27_3b_22_>As seen on Panel A of Table 8, the 70th percen­tile taxpayer has an after-tax income of about $68,000 and dividend income of $1,650. These figures show that 70 percent of all taxpayers who received dividends in 1998 earned less than $68,000 in after-tax income.^[19] Just as telling is the fact that the median wage and salary income for this 70th percentile taxpayer is just over $55,000 in 1998.

  <_spanarial27_3b_22_>Similarly, single taxpayers receiving dividends have a median after-tax income of $18,718, divi­dend income of $515, interest of $317, and capital gains of $338 (Table 7, Panel B). Even single tax­payers at the 90th<_spanarial27_3b_22_>percentile, reporting dividend income of over $7,300 (Table 8, Panel B), have an after-tax income of just over $63,000 and wage and salary income of about $53,000.

  <_spanarial27_3b_22_>Panel C of Table 7 shows that the median reported after-tax income of married taxpayers receiving dividends was about $64,300, with a median dividend income of $481. Married taxpay­ers at the 70th percentile (Table 8, Panel C) with an after-tax income of $89,339 report dividend income of $1,671 and wage and salary income of just over $76,000.

  <_spanarial27_3b_22_>Further partitioning the data on the dividend-receiving taxpayers into pre- and post-retirement ages produces similar results. As Table 9 shows, the median pre-retirement age single filers (Panel B), married filers (Panel C), and combined single and married filers (Panel A) all report about $300 in dividends. These results are quite different from those on Table 3, where all pre-retirement age tax­payers?even those not receiving dividends?were found to have $0 in dividends at the median.

  <_spanarial27_3b_22_>When only post-retirement age, dividend-receiving taxpayers are considered, the median taxpayers? (single, married, and combined) divi­dend income is $2,406, with an after-tax income of $35,544 (Table 11, Panel A). The median single taxpayer in this group (Panel B) has a lower after-tax income of $21,844 and a higher dividend income of $3,184. Panel C shows that the median married taxpayer in this group has an after-tax income of $44,921 and a dividend income of just under $2,000.

  <_spanarial27_3b_22_>The three panels in Table 10 and Table 12 list the percentiles for the three categories of filers in the pre-retirement and post-retirement age groups, respectively. The 80th percentile taxpayer in the pre-retirement age group (Table 10, Panel A) has an after-tax income of about $98,000 and divi­dend income of $1,643. When this is compared with the 80th percentile taxpayer in the post-retirement age group (Table 12, Panel A), who has after-tax income of just under $60,000 and divi­dend income of about $10,000, it appears that dividends make up a larger percentage of retired people?s income. Similar results for both single and married taxpayers in the post-retirement age group (Table 12, Panels B and C) confirm this finding.

  <_spanarial27_3b_22_>Thus, many Americans who invest in the busi­ness of America are not ?rich,? but rather single, married, and retired Americans who make under $100,000 per year. All of these Americans would benefit from dividend tax relief.

  conclusion

  <_spanarial27_22_>The double taxation of dividends in the U.S. tax code distorts economic activity because it encour­ages the retention of corporate profits and debt financing and discourages dividend payments and equity financing, solely for tax purposes.

  <_spanarial27_22_>Eliminating the double taxation of dividends would be a positive step toward tax neutrality. Mil­lions of families would benefit from dividend tax relief, including the many moderate-income work­ers who receive dividends. As of 2000, more than 42 million American workers owned a 401(k) plan, and nearly 40 percent of all U.S. households owned an IRA. Data show that, as of 2002, nearly 53 million U.S. households (just under 50 per­cent) and about 84 million individuals owned some form of equities. Additionally, in 1998, 70 percent of all taxpayers receiving dividends earned less than $55,000 in wages and salary.

  <_spanarial27_22_>All true dividends are subject to double taxa­tion. Since all investments compete for investors? dollars, removing the 35 percent layer of corporate dividend taxes would affect other investments as well. Even the fact that many Americans hold equities in tax-deferred accounts will not, in the long run, diminish the impact of eliminating the double taxation of dividends. Dividend tax relief may be criticized as providing a tax break for ?the wealthy,? but the IRS?s own data clearly suggest otherwise.

  <_spanarial27_22_>Norbert J. Michel is a Policy Analyst in the Cen­ter for Data Analysis at The Heritage Foundation.

  <_spanarial27_22_> Appendix
  Methodology

  <_spanarial27_22_>One issue surrounding the use of tax return data is the definition of taxpayers? income. For exam­ple, individual returns vary from reporting only wage and salary income to reporting up to 15 dif­ferent types of income.^[20] For statistics derived from the Internal Revenue Service?s data in its 1998 Public Use Tax File from the Statistics of Income Division (SOI), the author has used a broad definition of income that includes virtually all non-investment income.^[21] The measure is cal­culated for each observation in the SOI 1998 Pub­lic Use Tax File and includes the following types of income: 
• <_spanarial27_22_>Wages, salaries, tips, etc., from employees? W-2 forms;
• <_spanarial27_22_>Alimony received;
• <_spanarial27_22_>Business income or loss;
• <_spanarial27_22_>Other gains or losses (sales of business prop­erty);
• <_spanarial27_22_>Total taxable IRA distributions;
• <_spanarial27_22_>Total taxable pensions and annuities;
• <_spanarial27_22_>Rental real estate, royalties, partnerships, S corporations, trusts, etc.;
• <_spanarial27_22_>Farm income;
• <_spanarial27_22_>Unemployment compensation; and
• <_spanarial27_22_>Social Security benefits.           

  <_spanarial27_22_>Aside from those already mentioned, using the data in the SOI Public Use Tax File for this pur­pose presents two major shortcomings. It should be noted that the amount reported as ?wages, sala­ries, tips, etc.? on the W-2 form does not include payroll taxes or any wages and salary contributed to employer-sponsored tax-advantaged savings plans: cafeteria plans, 401(k) plans, etc. Also, ?div­idend income? on the IRS form 1040 does include cash distributions that are actually interest pay­ments, not dividends.^[22]

  <_spanarial27_22_>Next, the author combined the total tax owed or refunded with the above types of income to create a variable for ?after-tax income.? The after-tax income variable basically includes all income other than that earned from investments. In other words, it excludes income from dividends, capital gains, and interest. Therefore, the calculated after-tax income serves as a broad measure of taxpayers? ?permanent? income.^[23]

  <_spanarial27_22_>Summary statistics (mean, median, and stan­dard deviation) and percentiles are calculated for all returns as well as for those returns reporting dividend income. As reported above, when all returns in the 1998 SOI file are considered, more than half of all taxpayers do not report dividend income. Therefore, focusing on all returns alone does not provide a clear picture of the income (and other characteristics) of those individuals reporting dividend income.

  <_spanarial27_22_>Consequently, the author examined filers in every filing status (single, married filing jointly, married filing separately, head of household, or widower) for all ages in two separate categories? those receiving dividends and those not receiving dividends. Additionally, the data in both categories are partitioned based on filing status for both pre-retirement age and post?retirement age taxpay­ers.

  <_spanarial27_22_>In regard to filing status, the author first com­bined all filers and then examined single and mar­ried filers separately (married filing separately and married filing jointly were combined as married filers). Finally, while it is not a perfect classifica­tion, all returns including Social Security benefits are categorized as ?post-retirement age,? and all returns not reporting such benefits are categorized as ?pre-retirement age.?

  Footnotes

  ^[1]A corporation's after-tax profits can be distributed either as dividends or as capital gains. In either case, these after-tax dollars are taxed again at the personal level. In contrast, when corporations take on debt, they receive a tax deduction for their interest payments. Consequently, interest payments are made out of pre-tax dollars rather than after-tax dollars. These interest payments, therefore, are not subject to the same multiple levels of taxation as dividend payments.
  ^[2]While it has been reported that President Bush's economic proposals will include some form of dividend tax relief, the mechanics of the dividend tax cut are not clear as of this writing. See Bob Davis, Greg Hitt, and John McKinnon, "Bush to Propose Broad Tax Cuts in a $600 Billion 'Jobs' Package," The Wall Street Journal, January 6, 2003, p. A1, and Mike Allen and Dana Milbank, "President to Seek Dividend Tax Cut," The Washington Post, Jan­uary 3, 2003, p. A1.
  ^[3]For a thorough discussion of the methods for eliminating the multiple taxation of dividends, see Deborah Thomas and Keith Sellers, "Eliminate the Double Tax on Dividends," Journal of Accountancy, November 1994.
  ^[4]Any amount retained by the firm is liable to be taxed eventually as a capital gain at the personal level.
  ^[5] http://www.taxpolicycenter.org/taxfacts.
  ^[7]For differences between IRS and NIPA dividends, see Thae S. Park, "Comparison of BEA Estimates of Personal Income and IRS Estimates of Adjusted Gross Income," Survey of Current Business, November 2002, pp. 13-20. For more on NIPA dividends, see Bureau of Economic Analysis, "Annual Revision of the National Income and Product Accounts: Annual Estimates, 1999-2001, and Quarterly Estimates, 1999:I-2002:I," Survey of Current Business, August 2002, Table 8.25, at http://www.bea.doc.gov/bea/ARTICLES/2002/08August/0802AnnualRevision.pdf. 
  ^[8]For more on the effects of the double taxation of dividends, see James M. Poterba, "Tax Policy and Corporate Saving," Brookings Papers on Economic Activity<_spanarial27_22_>, Vol. 2 (1987), pp. 455-515. 
  ^[9]These employer-sponsored plans include 401(k), 403(b), 457, and other defined contribution plans, as well as state and local government employee retirement funds and private defined benefit plans. See Investment Company Institute, 2002 Mutual Fund Fact Book, at http://www.ici.org/facts_figures/factbook_toc.html.
  ^[10]There is also reason to believe that dividend tax relief will lead to greater investment and economic growth-outcomes that would benefit all working Americans (even those who do not invest). See Ervin L. Black, Joseph Legoria, and Keith F. Sell­ers, "Capital Investment Effects of Dividend Imputation," Journal of the American Taxation Association<_spanarial27_22_>, Vol. 22, No. 2 (2000), pp. 40-59; James M. Poterba, "Tax Policy and Corporate Saving," Brookings Papers on Economic Activity, No. 2 (1987), pp. 455-515; Peter Birch Sorensen, "Changing Views of the Corporate Income Tax," National Tax Journal, Vol. 48, No. 2 (June 1995), pp. 279-294; and James M. Poterba and Lawrence H. Summers, "New Evidence That Taxes Affect the Valuation of Dividends," Journal of Finance, Vol. 39, No. 5 (December 1984), pp. 1397-1415.
  ^[11]See Investment Company Institute, "401(k) Plan Asset Allocation, Account Balances and Loan Activity in 2000," Perspec­tive, Vol. 7, No. 5 (November 2001), at http://www.ici.org/pdf/per06-01.pdf. 
  ^[12]See Investment Company Institute, "Fundamentals," ICI Research in Brief, Vol. 11, No. 3 (September 2002), at http://www.ici.org/pdf/fm-v11n3.pdf. 
  ^[13]See Investment Company Institute and Securities Industry Association, Equity Ownership in America, 2002<_spanarial27_22_>, September 2002, at http://www.ici.org/pdf/rpt_02_equity_owners.pdf. 
  ^[14]See John McKinnon, "Tax Cut on Dividends Paid to Individuals Gains Support," The Wall Street Journal<_spanarial27_22_>, December 4, 2002, p. A1. 
  ^[15]Again, it should be noted that the IRS classifies interest income from mutual funds as dividends. This classification discrepancy will be addressed further in a forthcoming CDA report. Nonetheless, the data reported in this paper are representative of the income diversity among investors in the U.S. and how many Americans will be affected in the long run by eliminating the double taxation of dividends.
  ^[16]See Appendix, infra. 
  ^[17]For complete details on how after-tax income is defined, see Appendix, infra.
  ^[18]Despite the problems with the IRS classification of dividends, the fact that this median taxpayer has capital gain income of $220 increases the likelihood that this dividend income does include real dividends.
  ^[19]To reflect "normal" earnings more accurately, the measure used for after-tax income excludes investment income. When income from dividends, capital gains, and interest is added back into the taxpayer's income, the total income is $73,624. For more on using this measure of after-tax income, see Appendix, infra. 
  ^[20]See IRS Form 1040.
  ^[21]Results similar to those reported above were found when adjusted gross income (AGI) was used instead of this mea­sure of income. These results are available from the author.^{<_spanarial27_22_>
  [22]<_spanarial27_22_>In spite of these shortcomings, the statistics presented in this report reflect the income diversity among groups of investors in America today, and therefore how many Americans are affected by the double taxation of dividends. Again, since all investments are competing for investors' dollars, any significant change in one investment choice will have an effect on the others. Both of the major shortcomings will be addressed in a forthcoming CDA Report.
  [23]}The idea is that income earned from working an hourly wage job and/or running a business is representative of normal or "permanent" income, while investment-related income, which can be transitory, depends largely on how much permanent income is earned. The permanent income hypothesis was developed by Milton Friedman in the 1950s. See Milton Fried­man, "The Permanent Income Hypothesis: Comment," American Economic Review, Vol. 48 (December 1958), pp. 990-991.

  Footnotes

  ^[1]A corporation's after-tax profits can be distributed either as dividends or as capital gains. In either case, these after-tax dollars are taxed again at the personal level. In contrast, when corporations take on debt, they receive a tax deduction for their interest payments. Consequently, interest payments are made out of pre-tax dollars rather than after-tax dollars. These interest payments, therefore, are not subject to the same multiple levels of taxation as dividend payments.
  ^[2]While it has been reported that President Bush's economic proposals will include some form of dividend tax relief, the mechanics of the dividend tax cut are not clear as of this writing. See Bob Davis, Greg Hitt, and John McKinnon, "Bush to Propose Broad Tax Cuts in a $600 Billion 'Jobs' Package," The Wall Street Journal, January 6, 2003, p. A1, and Mike Allen and Dana Milbank, "President to Seek Dividend Tax Cut," The Washington Post, Jan­uary 3, 2003, p. A1.
  ^[3]For a thorough discussion of the methods for eliminating the multiple taxation of dividends, see Deborah Thomas and Keith Sellers, "Eliminate the Double Tax on Dividends," Journal of Accountancy, November 1994.
  ^[4]Any amount retained by the firm is liable to be taxed eventually as a capital gain at the personal level.
  ^[5] http://www.taxpolicycenter.org/taxfacts.
  ^[7]For differences between IRS and NIPA dividends, see Thae S. Park, "Comparison of BEA Estimates of Personal Income and IRS Estimates of Adjusted Gross Income," Survey of Current Business, November 2002, pp. 13-20. For more on NIPA dividends, see Bureau of Economic Analysis, "Annual Revision of the National Income and Product Accounts: Annual Estimates, 1999-2001, and Quarterly Estimates, 1999:I-2002:I," Survey of Current Business, August 2002, Table 8.25, at http://www.bea.doc.gov/bea/ARTICLES/2002/08August/0802AnnualRevision.pdf. 
  ^[8]For more on the effects of the double taxation of dividends, see James M. Poterba, "Tax Policy and Corporate Saving," Brookings Papers on Economic Activity<_spanarial27_22_>, Vol. 2 (1987), pp. 455-515. 
  ^[9]These employer-sponsored plans include 401(k), 403(b), 457, and other defined contribution plans, as well as state and local government employee retirement funds and private defined benefit plans. See Investment Company Institute, 2002 Mutual Fund Fact Book, at http://www.ici.org/facts_figures/factbook_toc.html.
  ^[10]There is also reason to believe that dividend tax relief will lead to greater investment and economic growth-outcomes that would benefit all working Americans (even those who do not invest). See Ervin L. Black, Joseph Legoria, and Keith F. Sell­ers, "Capital Investment Effects of Dividend Imputation," Journal of the American Taxation Association<_spanarial27_22_>, Vol. 22, No. 2 (2000), pp. 40-59; James M. Poterba, "Tax Policy and Corporate Saving," Brookings Papers on Economic Activity, No. 2 (1987), pp. 455-515; Peter Birch Sorensen, "Changing Views of the Corporate Income Tax," National Tax Journal, Vol. 48, No. 2 (June 1995), pp. 279-294; and James M. Poterba and Lawrence H. Summers, "New Evidence That Taxes Affect the Valuation of Dividends," Journal of Finance, Vol. 39, No. 5 (December 1984), pp. 1397-1415.
  ^[11]See Investment Company Institute, "401(k) Plan Asset Allocation, Account Balances and Loan Activity in 2000," Perspec­tive, Vol. 7, No. 5 (November 2001), at http://www.ici.org/pdf/per06-01.pdf. 
  ^[12]See Investment Company Institute, "Fundamentals," ICI Research in Brief, Vol. 11, No. 3 (September 2002), at http://www.ici.org/pdf/fm-v11n3.pdf. 
  ^[13]See Investment Company Institute and Securities Industry Association, Equity Ownership in America, 2002<_spanarial27_22_>, September 2002, at http://www.ici.org/pdf/rpt_02_equity_owners.pdf. 
  ^[14]See John McKinnon, "Tax Cut on Dividends Paid to Individuals Gains Support," The Wall Street Journal<_spanarial27_22_>, December 4, 2002, p. A1. 
  ^[15]Again, it should be noted that the IRS classifies interest income from mutual funds as dividends. This classification discrepancy will be addressed further in a forthcoming CDA report. Nonetheless, the data reported in this paper are representative of the income diversity among investors in the U.S. and how many Americans will be affected in the long run by eliminating the double taxation of dividends.
  ^[16]See Appendix, infra. 
  ^[17]For complete details on how after-tax income is defined, see Appendix, infra.
  ^[18]Despite the problems with the IRS classification of dividends, the fact that this median taxpayer has capital gain income of $220 increases the likelihood that this dividend income does include real dividends.
  ^[19]To reflect "normal" earnings more accurately, the measure used for after-tax income excludes investment income. When income from dividends, capital gains, and interest is added back into the taxpayer's income, the total income is $73,624. For more on using this measure of after-tax income, see Appendix, infra. 
  ^[20]See IRS Form 1040.
  ^[21]Results similar to those reported above were found when adjusted gross income (AGI) was used instead of this mea­sure of income. These results are available from the author.^{<_spanarial27_22_>
  [22]<_spanarial27_22_>In spite of these shortcomings, the statistics presented in this report reflect the income diversity among groups of investors in America today, and therefore how many Americans are affected by the double taxation of dividends. Again, since all investments are competing for investors' dollars, any significant change in one investment choice will have an effect on the others. Both of the major shortcomings will be addressed in a forthcoming CDA Report.
  [23]}The idea is that income earned from working an hourly wage job and/or running a business is representative of normal or "permanent" income, while investment-related income, which can be transitory, depends largely on how much permanent income is earned. The permanent income hypothesis was developed by Milton Friedman in the 1950s. See Milton Fried­man, "The Permanent Income Hypothesis: Comment," American Economic Review, Vol. 48 (December 1958), pp. 990-991.