The tax gap

Alan Reynolds
|
Posted: Nov 09, 2006 12:00 AM

A USA Today headline said, "Democratic House Means Changes in Tax Agenda." If so, such changes will be small and subtle. Congressional Democrats might want to reverse the tax cuts of 2003, but they know they can't override a presidential veto. There is no point in appearing eager to raise taxes before the next presidential election, since doing so would just make it easier to lose in 2008. Since most tax cuts expire in 2010 anyway, all the pro-tax Democrats need to do is to lay low, not show their hand and stall for time.

The new chairman of the House Budget Committee, South Carolina's John Spratt, cleverly advised CNBC that Democrats would focus on narrowing "the tax gap." That refers to the difference between taxes owed and taxes actually paid -- as a result of understating income, overstating deductions and credits, not making required payments or failing to file a tax return.

Who could object to closing the tax gap? Certainly not the IRS, which deploys "tax gap" estimates as a reason to keep increasing the agency's $10.6 billion budget. If the House Democrats are simply hoping to fund an even more aggressive and more intrusive IRS, however, the effort may not prove as popular as it sounds.

The Treasury's Office of Tax Policy already issued a "Strategy for Reducing the Tax Gap" in late September. That paper estimated that 71 percent of the gap was due to the individual income tax, 17 percent to payroll tax, 9 percent to the corporate tax and 2 percent to the estate tax. More than 80 percent of the overall gap is caused by underreporting income (particularly small business income) or overstating tax deductions and credits. About 7 percent of the gap is from failing to file tax returns, while another 10 percent is largely due to employers and the self-employed failing to withhold payroll and/or income taxes.

Small business is estimated to account for 32 percent of the tax gap. But there is no reason to assume that the offending taxpayers typically have high incomes. After all, auditing small business accounted for 37 percent of the IRS enforcement expenditure in 2006, and the number of high-income audits more than doubled from 2001 to 2004.

Businesses too tiny to merit such expensive auditing seem more likely to slip by. There may, however, be inherently difficult problems of auditing the huge number of Subchapter S corporations, partnerships and limited liability companies now choosing to be taxed as individuals rather than corporations. If so, the best solution to that problem may be to reduce the corporate tax rate, as nearly all other major nations have done.

Since nearly the entire tax gap is within the individual income tax, rather than corporate tax, we can get a good idea of what makes that gap widen or narrow by looking the "AGI Gap" -- the difference between Adjusted Gross Income (AGI) as reported on individual tax returns and AGI as measured as by the Bureau of Economic Analysis (BEA).

The BEA estimates, based on carefully collected personal income data, are much larger than the amount of AGI that shows up on tax returns. Until very recently, that gap narrowed when the highest, most punitive tax rates were reduced and widened when top tax rates were increased. The AGI gap narrowed from 13.5 percent in 1984 to 9.6 percent in 1988, as the top tax rate fell from 50 percent to 28 percent. With tax rates on extra income sharply reduced, the risk-reward ratio tilted toward honesty. That is one reason tax receipts in 1988-89 turned out much larger than expected.

The AGI gap widened to 12.3 percent when tax rates on salaries were increased in 1993, but narrowed to 10.9 percent when the tax rate on capital gains was reduced in 1997. Surprisingly, the gap suddenly rose again to 14.4 percent in 2003, despite a reduction in tax rates late that year. I believe the AGI gap has narrowed since 2003, as evidenced by unexpectedly large gains in tax revenue.

In any case, such variability of the AGI gap is just one of many reasons why we cannot rely on income tax returns to estimate changes in income distribution, as I explain in the fifth chapter of "Income and Wealth" from Greenwood Press.

Democrats are instinctively inclined to believe the AGI gap is mainly the result of cheating by the very rich. But the way the AGI gap is calculated makes that quite unlikely. Underreporting of capital gains is widely thought to be the largest source of tax evasion among wealthy investors. Yet unreported capital gains are not part of the AGI gap (although they are more than 2 percent of the broader tax gap) because capital gains are not counted in personal income. Stock options are a famously big source of one-time riches among executives and others, but gains from the most popular type of stock options are conspicuously reported on W2 forms -- making evasion nearly suicidal.

At the lower end of the income scale, a 2005 IRS study found the highest rates of underreporting happened with alimony income (21 percent to 24 percent), unemployment benefits (8 percent to 15 percent) and tax credits (17 percent to 23 percent). Only the misreported tax credits, notably the refundable Earned Income Tax Credit, amounted to much loss of tax revenue.

Aside from a few recipients of alimony, however, those collecting but not reporting unemployment benefits or the EITC usually have low incomes. The same is true of tips and other income received in cash. The IRS estimates that underreporting of wages, salaries and tips accounts for only about 4 percent of the tax gap, but they can't possibly know.

Tax audits rarely discover income received in cash. Many illegal immigrants work in the underground cash economy to avoid detection. Although they frequently earn too little to owe income tax, their efforts to remain invisible to federal authorities means they must also avoid Social Security taxes (and benefits).

The main thing the House Democrats need to learn about the tax gap is that legal tax avoidance and illegal tax evasion (among those with enough income to matter to the IRS) are most enticing when marginal tax rates are high and least tempting when such tax rates are low. Attempting to narrow the tax gap is not necessarily objectionable, so long as it isn't accomplished by trampling all over the privacy of honest taxpayers. But the only proven way to maximize taxpayer compliance, as we learned in 1988 and 1997, is to keep the highest tax rates as low as possible.