Let's Talk About Tax Cheating: A Freakonomics Quorum

Photo: hoggardb

The Internal Revenue Service presumably never likes tax cheats, but when money is tight there is more pressure put on the I.R.S. to step up enforcement and collect more money. The most recent gambit, reported today, is an amnesty program designed to root out offshore tax havens. We once wrote of another I.R.S. measure that produced $3 billion with one astonishingly simple and creative idea.
With tax day nearly upon us and with President Obama calling for future changes that will probably alter tax-paying behavior a great deal, we thought it might be a good time to think about how people cheat. So we asked a group of people who know a lot about tax cheating (but who surely don’t do it themselves) — Mark A. Luscombe, William A. Raabe, Peggy Richardson, and Joel Slemrod — the following questions:

1. What are the most pervasive forms of tax cheating?
2. What are some new cheating wrinkles you’ll be looking for under the new Obama tax proposals, and what are a few things that should be done about them?

Here are their answers.
William A. Raabe is a tax professor at Ohio State University.

Prediction: tax cheating will increase significantly during the Obama years.
Prescription: make the I.R.S. more digital, and eliminate the annual Form 1040 ritual altogether for most individuals.
Tax cheating to me means only the purposeful understatement of tax, not a simple error due to ignorance of the law, poor math skills, or confusion due to a difficult point of the tax law. And one is not a tax cheat if the taxing jurisdiction is not fairly enforcing a tax provision, e.g., in the case of sales/use tax collections by states for online purchases.
Tax cheating occurs when there is a combination of all three of these conditions: inclination, reward, and opportunity.
Some people seem inclined to take risks that most see as inappropriate, like speeding and not wearing a seat belt, shaving strokes off of a golf score, and stepping out on a significant other. Risky behavior, like tax cheating, is attractive to some, and it seems to be addictive. By either repetition or rationalization, tax cheats seem to find it hard to break the habit; in fact, it gets easier to understate a tax and do it in larger amounts over time. When cash is short, short-changing the tax system becomes more than just a bad habit.
Prediction: in difficult economic times, more taxpayers will give in to temptations to cheat on their taxes.
The rewards derived from tax cheating are a function of the amounts of the understated income and the overstated deductions, and the tax rates that would apply to them. Tax cheating increases when the corresponding tax rates go up, and the prescription for significantly higher tax rates during the Obama years will result in greater transgressions by tax cheats; the rewards for taking such risks have increased. For most business taxpayers it is easier to overstate deductions than it is to hide income, as there are more requirements that third parties report income items directly to the I.R.S. (e.g., credit card payments, sales of realty, payments to contractors) than there are reporting requirements relating to deductions (e.g., for business supplies, advertising, and cost of sales).
Prediction: tax-cheating behavior will increase significantly during the Obama years as tax rates increase.
Opportunities to cheat on one’s taxes tend to arise when structural changes in the tax law occur. The Obama team seems to favor the use of tax credits against the federal income tax to carry out the stimulus, health, energy, and education agendas that it is committed to. Tax credits, though, have been easy for tax cheats to use. The Earned Income Tax Credit, a federal welfare system mainly for low-income families, is notorious for attracting improper behavior, often via false taxpayer names and ID numbers. Education credits can be overstated when the taxpayer self-reports qualifying expenditures for supplies and travel. Oversight of energy credits (say for “green” home improvements) is almost nil. So the increased use of a credit of this sort by itself over the next few years will trigger new tax cheating.
Prediction: the tax-cut devices chosen by the Obama team will prove to be vulnerable to a new wave of tax cheating.
Tax cheating will never be eliminated, but such leakage of tax revenues could be better managed with a more effective use of technology by the Treasury.
Tax auditors need better training to monitor taxpayers’ financial software, as well as that used to coordinate their own reviews of returns. Matching of documents to track business transactions, including payments for services and for sales of securities, should be expanded.
The banking industry should be encouraged to move even more rapidly toward a cashless economy, focusing on direct deposits and debit card usage, as cash-oriented businesses are the most likely to underreport their tax liabilities.
The clunky annual Form 1040 filing system must be overhauled; most taxpayers need not even file a return, as the Treasury already has all of the information necessary to compute the tax, with Forms W-2, 1099, and the like. Only those taxpayers with income and deductions from other sources (like proprietors and partners) should be filing returns at all! For most individuals, the I.R.S. should simply send them a mocked-up tax form by March 15, with a refund check or a voucher for the tax due. Then the government could focus efforts on collecting the correct tax from the taxpayers who remain, and who are the most likely to cheat anyway.

Mark A. Luscombe is a principal analyst for the tax and accounting group at CCH.

The I.R.S. in recent years has tried to focus its limited audit resources to try to get the largest possible recoveries for the audit resources expended. This has meant a focus on tax cheating through offshore bank accounts and credit card abuse, usually by high-income taxpayers; tax cheating by small businesses with hidden income or personal expenses being deducted as business expenses; and abusive tax-shelter schemes usually designed to create paper losses for tax purposes to offset otherwise taxable income.
Tax cheating can also occur, however, at the lower income levels, and this form of cheating can be much more difficult for the I.R.S. to attack. It is much easier for the I.R.S to recover $10 million dollars in taxes by going after 10 taxpayers who underpaid taxes by $1 million dollars each than to go after 100,000 taxpayers who underpaid taxes by $100 each. Yet some of Obama’s tax proposals are most likely to generate tax cheaters in this second category.
One area likely to be a concern for President Obama and the I.R.S. is Obama’s fondness for refundable tax credits. Refundable tax credits result in checks issued to taxpayers when no taxes are owed. The temptation of getting a check from the government is too attractive for many taxpayers and preparers with a stake in the size of the refunds that they obtain for their clients to resist. The Earned Income Tax Credit, one of our first efforts at refundable credits, has had a history of abuse. A recent effort to refund the telephone excise tax through the income tax system produced many extravagant claims. Under President Obama, we have already expanded the refundable child tax credit and first-time homebuyer credit and added a new refundable Making Work Pay credit and a partially-refundable education credit.
Abuse of refundable credits creates the scenario of a large number of taxpayers cheating by small amounts. To counter potential abuse from the growth of refundable credits, the I.R.S. might have to revise its general approach of paying out claimed refunds and asking questions later. As was done with the telephone excise tax refunds, the I.R.S. will have to better develop systems to screen out questionable refunds before they are paid out. Following the telephone excise tax refund problem, a new penalty was added to the tax code aimed at excessive refund claims.
Another possible avenue of pursuit would be for Congress and the I.R.S. to further develop third-party reporting requirements that might support claims for deductions and credits on the return. Statistics tend to show that tax cheating occurs most often when there is no independent source of information for the I.R.S. to check what is put on the return. There is very little underreporting of wage income due to the Form W-2 reporting system by employers. Form 1099 provides some additional third-party reporting of income and Form 1098 provides some third-party reporting of deductions, but there are considerable gaps on the income side and even more holes on the expense/deduction side. Third-party reporting has recently been added to the tax code to require broker-basis reporting. Third-party reporting also, however, puts an additional burden on the reporting entity, so there is usually a Congressional lobby actively fighting expanded third-party reporting efforts.

Joel Slemrod, a professor of economics at the University of Michigan, was the senior staff economist for tax policy in President Reagan‘s Council of Economic Advisers, and is co-author of Taxing Ourselves: A Citizen’s Guide to the Debate Over Taxes.

According to the latest I.R.S. “tax gap” study — for tax year 2001 — tax that should have been paid but was not totaled $345 billion. Of the $345 billion, the I.R.S. expected that it would eventually recover $55 billion, resulting in a “net tax gap” of $290 billion, which is 13.7 percent of the tax that should have been reported.
About two-thirds of all underreporting of income happens on the individual income tax. Of that, business income — as opposed to wages or investment income — accounts for about two-thirds.
Most striking is the huge variation in the rate of misreporting by type of income. Only 1 percent of wages and salaries and only 4 percent of taxable interest and dividends is underreported. In sharp contrast, 57 percent of non-farm proprietor income is not reported.
Why the disparity? Payments of wages and salaries, interest, and dividends must all be reported to the I.R.S. by those who pay them; moreover, wages and salaries are subject to employer withholding. Most self-employment business income is subject to neither information reporting nor withholding, so the I.R.S. knows little absent a thorough audit.
The I.R.S. estimates that the net misreporting rate is 53.9 percent, 8.5 percent, and 4.5 percent for income types subject to “little or no,” “some,” and “substantial” information reporting, respectively, and is just 1.2 percent for those amounts subject to both withholding and substantial information reporting.
That noncompliance rates are lower when there are information reports and employer withholding provides compelling support for the “deterrence” model of tax evasion — that people remit what they owe because they’re afraid of being caught and penalized. To be sure, some people pay what they owe out of a sense of duty or a predilection for honesty. Both duty and calculation may matter. In a recent survey, 96 percent of people mostly or completely agreed that “It is every American’s civic duty to pay their fair share of taxes”; but also 62 percent said that “fear of an audit” had a great deal or somewhat of an influence on whether they report and pay their taxes “honestly.”
High-income people apparently evade more dollars, and for the most part evade more as a percentage of their true incomes. This is in part, but not entirely, due to the fact that they tend to receive the kind of income that is more difficult for the I.R.S. to monitor.
Tax noncompliance depends on the extent and effectiveness of I.R.S. enforcement. Among industrialized countries, the United States has one of the lowest ratios of administrative costs to revenue collections — less than half the average — and the lowest ratio of full-time staff of the national revenue body to the labor force. Unlike many industrialized countries, the United States does not utilize withholding at source for the collection of personal income tax on dividends, interest, independent personal services, or rents. However, the United States maintains one of the most substantial programs of information reporting.
What will surprise many readers is that the United States is in the minority of industrialized countries that require individuals to file tax returns. About half of them operate a system in which most wage earners need not file due to exact, cumulative employer withholding; and in other countries, the tax authority sends taxpayers a “pre-populated” tax return based on the information they receive from third parties, requiring the taxpayer only to verify that the information is correct. There’s no tax filing deadline to dread.
Eliminating the tax gap completely is not a reasonable policy goal, no more so than completely eliminating all robberies; the benefits would not outweigh the costs in resources and intrusiveness. Decreasing the tax gap is very reasonable, and would allow reductions in the tax burden for honest taxpayers. Expanded and smarter auditing would make a dent, as would enlarging the scope of third-party reporting of business transactions — but all come with downsides.
Simplifying the tax system is unlikely to have much of an impact on noncompliance; it’s worth doing for other reasons, but doesn’t seem to be in the cards. One out-of-the-box idea is for the I.R.S. to reward or reimburse people who come through an audit relatively unscathed. It might be worth a try.

Peggy Richardson, co-counsel at the law firm Baron & Budd, served as I.R.S. commissioner from 1993-1997.

Probably the most pervasive forms of “tax cheating” are not the large tax-shelter scams that have garnered headlines in recent years, but less “sexy” and more mundane forms of “cheating,” such as underreporting of income and claiming excessive deductions by large numbers of taxpayers. For example, people who are self-employed often fail to report all of their income properly. Since their income is not subject to withholding (employees’ wages are subject to withholding, but amounts paid to self-employed individuals or independent contractors are not) and payments to self-employed individuals are frequently not reported on information returns, those payments are often not reported or underreported on individuals’ tax returns.
Although it may not sound like much tax revenue would be involved in the underreporting or omission of income by self-employed individuals, including small business owners, given the numbers of people who are self-employed or who may earn a little extra income in addition to their wages earned as employees, the total amount of lost revenue has been estimated to be in the billions of dollars. In light of the current economic conditions, and with more people working part-time in self-employment situations, more misreporting and underreporting can be expected.
Claiming excessive deductions is also a form of “tax cheating” that presents compliance concerns; again, it’s not “sexy,” but it’s costly to the government. For example, if every individual tax return were to have deductions that were overstated (for example, inflated amounts claimed as gifts to charities) by just $25, that amount might seem trivial or innocuous individually, but since there are over 100 million individual returns filed annually, the total revenue lost from those overstated deductions would amount to over $2.5 billion.
One concern about potential for tax noncompliance under some of the current tax proposals relates to the various proposals for refundable tax credits. Because tax-credit computations can be very complex (the Earned Income Tax Credit is a prime example of a credit requiring a complex calculation), and the dollars that potentially could be refunded could be significant under some of the proposals, without stepped up enforcement by the Internal Revenue Service, these additional credits could present a significant opportunity for underreporting or out-and-out cheating. Additional compliance resources for the I.R.S. to conduct audits and match information that is reported to the I.R.S. will be essential to curbing additional tax-compliance problems arising from some of the current proposals.