Cash Businesses and Tax Evasion

I. INTRODUCTION

According to government reports, most individuals with business income fail to pay all their taxes, although some appear to cheat more than others. Underpayment of tax on business income is commonly attributed to the receipt of cash. The owner of a clothing store, for example, might sell a dress for cash and not report the cash. Underpayment of tax by individuals on business income contributes significantly to the federal tax gap--the difference between what taxpayers owe on legal source income and what they pay.

The government estimates that the annual tax gap equals $345 billion. About $109 billion of this is attributable to underpayment of taxes on business income by individuals. Sole proprietors also underpay Social Security and other payroll and self-employment taxes. Additional underpayments are attributable to individuals who operate businesses as partnerships and small corporations. In the aggregate, small business owners report less than half of their income, and their underreporting (including informal workers such as gardeners) is estimated to comprise about half of the tax gap.

This Article attempts to provide a qualitative picture of tax evasion. Tax evasion is a criminal offense under federal and state statutes. A person who is convicted is subject to a prison sentence, a fine, or both.

This Article proceeds in three additional parts. Part II summarizes the main threads of relevant social science research on small business tax compliance. Part III describes the methodology and results of this interview study. Part IV concludes.

II. TAX COMPLIANCE IN THE CASH BUSINESS SECTOR: EXISTING RESEARCH

A. Overview

The standard economic analysis frames a tax compliance decision as a comparison between (1) the cost of paying tax and (2) the difference between the benefit of avoiding the tax and the cost of the imposition of tax, interest, and penalties, risk-adjusted for the possibility that the government will successfully challenge the tax avoidance.

Through tax avoidance, an individual takes advantage of all legal opportunities to minimize his or her state or federal strategy. But this model does not provide a complete picture of taxpayer compliance or the reasons for variations in taxpayer compliance. Substantial behavioral research, including contributions from sociology and psychology, deepens the analysis, and our research here offers more descriptive detail.

One summary of the behavioral compliance literature lists fourteen factors that may affect tax compliance, including age, gender, education, income level, income source, peer influence, ethics, fairness, complexity, and tax rates. For our purposes, two of these factors, income source and peer influence, are most relevant and are discussed below. We also discuss studies that explore the relationship between tax preparers and tax compliance.

B. Income Source

By far the most important determinant of tax compliance is income source. Taxpayers report cash income less accurately than income subject to third party reporting and/or withholding. As noted in the introduction and accompanying notes, individuals evade business-source income, which is commonly received in cash, at a rate of approximately 50%, although this evasion is not evenly distributed. In contrast, the evasion rate on wage income--which employers report to the government and on which taxes are withheld--is about 1%.

Cash income represents one extreme of an income visibility spectrum while income subject to third-party reporting or withholding occupies the other end. Some studies show, for example, that taxpayers are more likely to report income received in check form than income received in cash. Taxpayer concerns that the government will detect a failure to pay taxes appear closely related to, but not completely dependent on, income source.

The strong relationship between evasion and income source suggests that the primary causal factor that explains evasion is opportunity. Employees whose employers comply with wage reporting rules cannot cheat successfully and so such employees do not cheat. Individual business owners can cheat successfully (because no one reports much of their income to the government and because their income is hard to detect on audit) and, in the aggregate, individual business owners do cheat. The literature on income source, accordingly, applies directly to a study of evasion in the cash business sector: it predicts a high rate of evasion in the sector and identifies the sector as a, if not the, leading source of non-compliance.

C. Peer Influence and Social Norms

A substantial body of research shows that taxpayers who believe their peers evade tax are more likely to evade tax themselves. This correlation does not necessarily translate to the conclusion that the behavior of a taxpayer's peers causes the taxpayer's behavior. For example, peer behavior may be used to defend a prior decision not to comply, or (less plausibly) a noncompliant taxpayer may seek out noncompliant peers.

Some studies do find a causal relationship, however. For example, one paper reports, based on longitudinal survey data, that a taxpayer tends to internalize the taxpaying norms of a group with which the taxpayer strongly identifies. The compliance norms of individuals outside a taxpayer's small circle may also have relevance, but the relationship is less certain. Another study suggests, for example, that mere mention by the government of broad social compliance norms cannot persuade taxpayers to comply.

Another social norm question is whether attitudes toward government in general, such as approval of government policies or the political party in power, have a significant effect on tax compliance. Some studies show no such effect, while others support a link. Several studies indicate that taxpayers' perception of the equity of the tax system affects their compliance behavior.

Research on the effect of norms generally does not focus on the cash business sector, but some of its findings can extend to that sector and have relevance for understanding evasion in that sector. One plausible hypothesis, based on these findings, is that the (correctly) assumed high level of non-compliance within the cash business sector contributes to, and in fact, increases the level of, non-compliance in that sector. Another hypothesis, not inconsistent with the first, is that differing beliefs as to peer behavior account for a significant variation in compliance among those in the cash business sector.

D. Tax Preparer Influence

Taxpayers with business income typically rely on preparers to help in tax filings. These preparers serve as "gatekeepers" who may (or may not) improve the compliance behavior of their clients. A number of studies have examined how taxpayers choose their preparers, with varying results. For example, one set of survey results suggests that taxpayers choose a tax adviser who reflects their attitudes toward compliance; another study suggests that once taxpayers have chosen a tax preparer, they tend to somewhat passively follow the advice of that tax preparer.

Other studies focus on possible connections between tax preparer characteristics and taxpayer compliance. One line of research, for example, indicates that a licensed tax preparer is likely to influence a taxpayer to be more aggressive on ambiguous questions and less aggressive on unambiguous questions. One study concluded from data gathered through experimental cases presented to CPAs and non-CPAs that CPAs took more pro-taxpayer positions in ambiguous situations but advised compliance with the law if the rules were sufficiently clear. Another paper demonstrated, based on randomly selected I.R.S. audit data, that CPA-prepared returns result in fewer audit adjustments compared to non CPA-prepared returns.

The preparer studies, like the work on taxpayer norms, do not focus specifically on the cash business sector. But the studies have relevance for our description below of the preparer market. Consider the studies suggesting that licensed preparers such as CPAs confine their aggressive advice to ambiguous situations and decline to advise their clients outright on tax evasion strategies. These studies might predict that licensed tax preparers refuse to accept tax-evading cash business taxpayers as clients. Or, the studies might suggest that licensed tax preparers and taxpayers often have a tacit understanding that they will not discuss cash income, so as to permit the tax preparer to avoid the uncomfortable question of whether to participate in what is plainly an evasion scheme. Our research indicates that the latter prediction is more accurate.

E. What We Don't Know About Evasion in the Cash Sector

Notwithstanding an impressive body of work on tax compliance (only hinted at in the brief summary above), we know surprisingly little about tax evasion in the business sector, aside from the consensus that, in the aggregate, owners of small businesses with substantial cash revenue fail to pay about half their taxes. One fundamental problem is that we lack any thick qualitative description of the actions or attitudes of those in the sector. We do not know very much about how taxpayers evade tax, how they view their actions, how they use preparers, or how preparers in the sector view their role and their clients.

We suspect our lack of knowledge has resulted from a disconnect between the quantitative methodological tools used in most prior studies and the complex and norm-driven nature of the particular behavior at issue. The core government tax gap data emerges from the statistical sampling methods used in the work of the so-called National Research Program and its predecessor, the Taxpayer Compliance Measurement Program. But neither these empirical techniques nor others such as surveys or studies that record subjects' reaction to hypothetical situations can provide the missing qualitative description of cash business taxpayer behavior. We suspect the reason for this is that standard survey and hypothetical situation methodology does not work well in areas defined by criminal conduct. The relatively short, fixed questions or limited fact patterns that populate such surveys and hypothetical studies cannot elicit the detailed responses necessary to fully describe the experience of individuals engaged in regular and systematic tax evasion. This results in part due to the limits of the format--subjects are not given an opportunity to add much detail--and in part due to the subjects' predictable concerns that disclosing information about illegal activity like tax evasion may lead to potential financial or even criminal liability.

III. THIS INTERVIEW STUDY

In this Part III we describe our study and its results. Part III.A outlines our methodology; Part II.B sets out our findings about the extent of evasion and what kinds of taxes are evaded. In Parts III.C, D, E and F, we discuss how cash businesses evade taxes by using a cash economy, designing employee controls, finding appropriate tax preparers, and dismissing the risk of audit. In Part III.G we briefly consider the reasons cash businesses give for underreporting decisions.

A. Methodology

The interviews we conducted belong most closely to the field study category of research techniques. Social scientists use field study research, including informal interviews, to study "people acting in the natural courses of their daily lives." Field study research encompasses many strategies, including the informal interview technique relied on to gather the data presented here. Field study research is particularly useful when it is impossible to conduct a randomized study to examine the behavior of interest, or as a means of identifying testable hypotheses for future studies conducted by other means.

Our study consisted of almost 275 interviews with individuals associated with cash businesses, including cash business owners, bankers, and tax preparers. As is customary with field studies, our subjects were not chosen randomly. The business owners we interviewed typically collected a substantial portion of their revenue in cash; the bankers we interviewed had experience in extending loans to cash business owners.

Our interviewee base included both fully licensed Certified Public Accountants and candidates preparing for the final CPA exam; for simplicity, we refer to individuals in both groups as "CPAs." Our interviewees included accountants from the tier of national firms that includes Grant Thornton, BDO Seidman and RSM McGladrey. We also interviewed CPAs from small or midsized regional firms with between one and 200 employees. We did not include CPAs from the largest accounting firms of Deloitte, Ernst & Young, KPMG, or PricewaterhouseCoopers, which each employ between 20,000 and 40,000 people in the US. The largest accounting firms serve relatively few of the small business owners who interested us in this study.

Some of the characteristics of the interviewees are set forth in Table 1 and the accompanying notes.

Individuals who engage in tax evasion face a panoply of overlapping civil and criminal penalties. This makes data collection in this area difficult. Indeed, we speculate in Part II.E. of this Article that it is this fact--the potential liability concerns of respondents--that is responsible for the dearth of interview or qualitative data about cash sector evasion. To assure our subjects that their identity would be kept confidential, we did not memorialize the interviews in contemporary notes. In addition, the interviews were conducted in as conversational manner as possible, designed to elicit information without (overly) raising concerns about potential liability.

Our interviews typically covered several different questions:

1. Who evades taxes? Under what conditions is evasion usually observed?

2. What taxes are evaded? Just income taxes, or sales, payroll and other taxes as well?

3. How are taxes evaded? How do taxpayers who wish to evade find a sympathetic tax preparer? What strategies do taxpayers use to identify what income they will pay tax on?

4. Why does tax evasion occur? Are factors such as tax law complexity or a negative attitude toward government important? Would replacing the income tax with a flat tax based on consumption reduce evasion?

As is true of many field studies and all survey data, our results depend on the subjects' willingness to describe their behavior and on the accuracy of that description. Here, we ask subjects to describe behavior that, at least in theory, can be prosecuted as a felony. One might expect, therefore, that interviewees would tend to underreport this behavior or to tell us about the least offensive violations. Since we had little difficulty eliciting stories of evasion, this tendency to underreport evasion, if present, would generally strengthen our results. On the other hand, it is also possible (though we think less likely) that interviewees may have offered tales of tax cheating because they felt that was what we hoped to hear. We believe that this possibility of demand bias is mitigated by the graphic details offered by interviewees to illustrate their stories.

In general, our results provide a qualitative description of evasion in the cash business sector that does not duplicate anything in the existing literature. Where the questions we ask overlap with questions asked in the existing literature, we note that and state whether our findings are consistent. Some of our results suggest hypotheses that could be tested through more conventional means, such as regression analysis or randomized survey data. Where that is the case, we discuss the hypotheses and how they might be tested.

B. Overview: Extent of Evasion and Kinds of Taxes Evaded

Virtually all of our interviewees believed that small businesses fail to report some of their cash income. Their particular comments on evasion are detailed below. Most of our interviewees hailed from urban areas although our interviews in Centerville, in Hawaii, and with conference participants from smaller cities add some diversity to our interviewee base. We did not notice a difference in our main result--that cash businesses evade tax on cash income--based on the size or geographic location of the town or city where a particular interviewee worked.

Interviewees frequently reported that taxpayers' failure to report cash income was linked to sales tax and payroll tax evasion as well as income tax evasion. Payroll taxes consist of Social Security and Medicare Hospital Insurance taxes and are levied at a combined rate of 15.3% on approximately the first $100,000 of wage income. Self-employment taxes are substitutes for payroll taxes; they are levied at the same aggregate rate and fund the same programs. A storekeeper who underreports cash income and uses that income to pay employees unreported wages and/or pay herself unreported self-employment income evades income, payroll and/or self-employment taxes. Nonpayment of self-employment taxes is estimated to comprise $39 billion, or sixteen percent of the gross tax gap, and is widely thought to be associated with underreporting of business income.

Most interviewees also reported a link between sales tax evasion and income and employment tax evasion. For some employees, sales taxes were the primary motivation for underreporting. A storekeeper we interviewed explained it like this:

Storekeeper Actually, I don't gain anything cheating on income tax this year. I have such a big loss on another investment I don't pay tax.

Interviewer: Then why not report?

Storekeeper: Sales tax. Six percent doesn't sound like a lot, but it's thousands every month and it's on the gross.

We often found that interviewees saw income underreporting as a necessary corollary to the more important goals of nonpayment of sales and employment tax. Many cash business owners focused on the sales tax, in particular, as the tax to avoid because it is levied on gross revenue and directly reduces profit margins. Sales tax applies even to unprofitable businesses and individuals with unrelated losses and no tax due. A profitable business with low margins may pay more in sales tax than income tax. More than one small business owner we interviewed stated that replacing the income tax with a consumption or "flat" tax would not affect his behavior, since it would not eliminate employment or sales taxes. The employment and sales tax link also indicates that cooperation between federal and state governments is a promising compliance strategy.

C. Parallel Cash Economies of Small Businesses

1. Summary

We found that many small businesses that evade taxes do so by constructing parallel cash economies. They collect cash revenues, often pay some expenses in cash, and then use the unreported cash they receive for cash purchases, rather than depositing it. Cash businesses that evade taxes must often make do with self-financing strategies, or at least accept that bank loans will be based on their reported income only.

Parallel cash economies are kept secret from the I.R.S. and state tax authorities. But they are not necessarily kept secret from everyone. Business owners face internal control issues relating to the risk that employees will keep some cash proceeds for themselves. They also face the question of how to find a sympathetic tax preparer and whether to confide in the tax preparer.

2. Cash revenue

The first requirement for a parallel cash economy is cash revenue. The storekeeper quoted above explained it like this:

  What I do is, people come in and buy stuff. If they pay with check
  or Visa, I record it. Guy comes in and if I am there and he spends
  over $40 [in cash], it's entered into the computer as an
  invoice-in-progress. End of the day I get a separate print out [of]
  all the invoices-in-progress, and they're erased from memory. I
  take the cash home. Never deposit the cash, ever.

This storekeeper focused on eliminating all records of cash transactions. As we discuss below, he initially neglected one piece of the puzzle: his inputs or expenses. But his careful approach to erasing records of cash revenues is instructive, and it is echoed in other interview results, such as reports of hair salons that erase pencil entries or fail to record walk-in business, or jewelry shop owners who do not enter cash transactions into their bookkeeping systems.

Another consideration revealed by the storekeeper's story is the question of how to segregate non-reportable cash transactions from the rest of the business. Other businesses we heard about employed similar strategies and heuristics. One practitioner told us of a small clothing retailer that opened its doors on Saturdays to certain customers who paid only in cash, which the retailer did not report. Another clothing retailer explained that he had a rule of thumb of reporting only 85% of sales. Several veterinarians we interviewed similarly suggested that they did not report between ten and twenty percent of their revenue. In each case, cash sales comprised the unreported portion.

Some interviewees suggested that checks were almost as good as cash, because they could be used as cash. Several interviewees in the jewelry, antique, and trophy businesses explained that checks received on the sale of merchandise were frequently signed over to suppliers in exchange for fresh inventory. Other interviewees stated that cashing (as opposed to depositing) a check did not leave a paper trail significant enough to present an audit concern.

In contrast, most interviewees reported that credit card receipts were generally reported as taxable revenue. This finding has relevance for the proposal to require credit and debit card issuers to report processed payments to merchants. The proposal would further increase the visibility of merchant card receipts. If most credit-card related receipts are already reported, one might conclude that the proposal is a waste of effort as there may be (relatively) little additional tax to be collected from this source. However, credit card reporting may be a useful method of estimating cash income and in that sense an important weapon against underreporting. Consider a restaurant that reports $150,000 of gross sales but is shown to have $140,000 of credit card receipts. The lopsided ratio of credit transactions to cash transactions suggests that the restaurant is underreporting cash.

The question of already-perceived income visibility also has relevance for any proposal to expand third-party reporting to banks. Our research suggests that cash business owners regard bank deposits and withdrawals as more visible than cash, but less visible than credit card transactions. If most bank transactions are already reported for tax purposes, a third-party reporting requirement for banks may not make sense as a tool intended to increase reporting of income deposited into bank accounts. But as with merchant card payment data, information about bank transactions other benefits may have other uses. For example, a total income measurement formula or audit filter might compare reported bank transactions to total reported income to help determine whether a taxpayer had underreported.

We found that businesses express their preference for cash (or checks) in different ways. Some state it explicitly, as in the case of the clothing retailer described above who accepts only cash on preferred-customer Saturdays. In other cases it is an industry norm. The tax preparers and businesspeople we spoke to in the jewelry and construction businesses, for example, suggested that many jewelers and contractors offer a 20% discount for cash transactions. A study involving researchers posing as potential consumers might confirm this data point. For example, some researchers might offer certain businesses cash and the prices they negotiated could be compared with prices negotiated by a control group that did not offer cash.

3. Cash business expenses

The strategies described in the preceding paragraphs relate to the revenue side of a parallel cash business. But the careful businessperson must consider inputs or expenses as well. Our interviewees generally considered overstating deductions an inferior strategy relative to misreporting income. "Never do anything with deductions," one business owner told us. Several accountants offered the maxim, "If you are going to cheat, cheat on the income side or cheat on the deduction side, but not both." In fact, a number of our interviewees reported the opposite problem: that understatement of income made even accurately reported deductions seem too large.

Consider the following problem faced by Storekeeper:

  Last year I gave my accountant all recorded sales and all costs.
  Accountant says to me, Donald, the way you've done it, you've lost
  $80,000. You can't show a loss of $80,000, it's impossible; the
  I.R.S.'ll be all over you, you can't live where you live and show
  that.

Another interviewee, a semi-retired accountant who had specialized in the cash business sector, highlighted the state sales tax audit risk that can result from failing to show a profit:

Interviewer Studies show that small cash businesses report about 50% of their gross income. Is that your experience?

Accountant 50%? No. I'd say 33%.

Interviewer: Do your clients ever get caught?

Accountant: Oh, yes, the SBE comes in and says you bought too many goods to have grossed what you grossed. Then they recalculate profit.

To avoid this problem, some businesses try to pay for inventory and other expenses in cash, and then not report the expenditure. The "cheat on the gross income" approach has two advantages to a "cheat on the deduction" approach. First, understating income reduces sales and employment taxes. Overstating deductions has no effect on those taxes. Second, paying employees and suppliers in cash provides employees or suppliers with tax-free income of their own. Several restaurateurs explained to us that they would not have the right bartenders, waitresses or entertainment unless cash was used to lubricate the process. Paying non-reported cash also makes workers complicit in the evasion scheme, and therefore less likely to report it.

Other interviewees reported sometimes elaborate schemes involving the purchase and sale of certain inventory. One preparer told us about an owner of 100 vending machines who bought the goods--potato chips, M&Ms and such--for thirty of his most profitable machines from a wholesaler such as Sam's or Costco for cash. He segregated the cash from those machines as nonreportable cash. We heard similar stories about a clothing retailer that bought and sold half its inventory in cash and about jewelry and antiques businesses that bought inventory from individuals and estates for cash and sold it for cash. Several interviewees described flower businesses that bought seeds for cash, paid employees in cash and sold their wares at farmers' markets for cash.

We did hear some reports of taxpayers overstating their deductions. In many of these cases, taxpayers lacked the ability to minimize taxes by not reporting cash income. A few taxpayers explained that if their deductions were low in a particular category, their tax preparers would decrease their tax bill by plugging in a higher number that conformed to the national averages published by the government.

Our findings as to the relative importance of understated receipts (rather than overstated deductions) in tax evasion is not entirely consistent with other literature. The latest government study estimates that understated receipts comprise about 55% of the tax gap in this area; the remainder is comprised of overstated deductions. This estimate is consistent with our finding that understated receipts are the largest source of underreporting, but gives a larger role to overstated deductions than does our study. It is possible that the sole proprietors who report significant overstated deductions do not receive significant cash receipts and consequently lack the opportunity to underreport receipts.

4. Cash spending

The next piece of the parallel cash economy involves spending the cash--for the cash business owners and preparers we spoke to generally agreed that one must avoid depositing it. We heard three strategies from our interviewees: spend it; hoard it; and invest it in the business.

The spending strategies typically involved purchases of personal property such as jewelry, rugs, antiques, clothing, and furniture. One preparer remarked that cash business taxpayers often have homes whose modest exteriors belie their expensive contents--clothing, jewelry, artwork, rugs, and furniture which may be valued at two or three times the value of the structure and land. Many interviewees also reported spending cash at restaurants and hotels.

Such spending, however, may not absorb a business's free unreported cash flow. Business owners typically reported spending between one and five thousand dollars a month on personal property purchases and hotel and restaurant expenses. Surplus cash can build up, and many of our interviewees reported that business owners who significantly underreport income often eventually purchase investments, housing and cars, boats or airplanes that are inconsistent with reported income. Many of these items carry paper trails such as brokerage account records, property transfer recordings or property tax assessments, or vehicle titles.

Many of our interviewees told us of hoarding strategies, often describing safe deposit boxes full of cash. Some explained that heirs would rush to clean out such safe deposit boxes when the depositor died. Many also reported investment in the business, saying that business owners purchased additional inventory or made capital improvements with surplus cash.

Our finding that business owners find themselves with assets that are inconsistent with tax records suggests the following two-part research strategy. First, some portion of audits of taxpayers with business income might be designed to specifically look for these assets; the same strategy might be used with respect to the National Research Program audits. Depending on the results of this first step, a pilot program might "data mine" records of asset purchases and check those purchases against reported income of the purchasing taxpayers. Expensive purchases by taxpayers with low reported income might be added as a factor in determining audit.

5. Self-financing

Self-financing is the final piece of the cash economy. Income that is not reported on a cash business owner's tax return generally cannot support bank loans, whether for business use or for home purchase or other personal use. Bankers we spoke to explained that they relied on tax returns to support loan applications because tax returns provide verifiable information. Accordingly, a small business owner who fails to fully report income for tax purposes sacrifices the capacity of the unreported income to support bank financing and must make do with savings or other self-financing strategies.

We heard some stories of loan applicants submitting fictional tax returns to support applications, but bankers also explained that they routinely checked tax returns by sending adjusted gross income or taxable income figures to the I.R.S. Under one pilot program involving California mortgage bankers and the I.R.S., the government informed a lender by return fax within forty-eight hours whether submitted adjusted gross income figures substantially matched government records. Our interviewees reported that the availability of I.R.S. verification significantly reduced borrowers' use of false tax returns in loan applications.

D. Employee Controls

Employees can be limiting factors for tax evasion by cash business taxpayers because of the risks of employee cheating, blackmail and whistle-blowing. However, our cash business owner interviewees seemed more concerned with the possibility that their employees would cheat the business out of revenue, just as the owners cheated the government out of taxes. One tavern owner told us this story:

  An accountant and I were watching a bartender serve drinks. First
  drink to a customer, the bartender collects $6 and puts $5 in the
  cash register. Second drink collects $6 and puts $3 in the cash
  register. Third drink, the bartender pockets the $6. At this point,
  I go up to the bartender and ask him, what's the matter, aren't we
  partners any more?

Many cash business owners limit underreporting to cash receipts received by the business owner and perhaps by family members or certain trusted employees. "Never let employees in on it," one owner of a service business stated. "If we see an employee gets cash, we make sure to march him right over to the bookkeeper; we make a big deal out of it. Only the cash my partner and I receive comes home." Another interviewee told us about a retail food market that "found their margins were lower than the historic norm. They checked for skimming, food theft etc. to no avail. Finally, they counted the number of registers and found that there was an extra register." An employee had bought his own register and was pocketing the revenues from the sales rung up on that register.

Some interviewees also reported that their accounting systems deterred employee cheating. The storekeeper quoted above believed that his convoluted system was not susceptible to employee cheating: "[Employees] don't [cheat] if I'm not there. I want to be able to come back and balance the books, and when you take cash out of my system, it's so cockamamie, that you never can figure out anything." In other cases, business owners reported that the automatic and somewhat mysterious workings of computerized bookkeeping deterred employee cheating--and sometimes owner cheating. One professional commented that bookkeeping software made it harder to hide cash, saying that "if I could subvert the computer software, so could my help."

The role of employees as a limiting factor for cash business tax evasion might generate ideas of interest to policymakers. For example, the task of policing employees to ensure that they don't cheat the business is easier to carry out if the owner is frequently on site. This may mean that owners of chain stores are less likely to evade taxes, because they would be less able to stop employees from following their cheating example.

A common pattern for an expanding small business is to leave the owners managing the "original" store, with trusted employees and family members managing the next store or two, and at some point to rely on "outsiders" to manage succeeding stores. If in fact owners evade and managers do not, there might be large discrepancies between reported profits of the original store and the last-added store. The presence of these discrepancies might be a factor toward audit; or, once a chain has been audited, a factor in the decision to devote greater resources toward that audit.

The uneasy relationship between owners and employees might also support expansion of the federal whistleblower program and adoption of similar programs at the state level. The federal whistleblower program, although underpublicized, is considered a moderate success. As is often the case with tax compliance enforcement, success has been measured by the amount of new taxes brought in. But our study suggests that the prospect of an employee blowing the whistle may deter evasion in the first instance. An expanded and effectively publicized whistleblowing program that encouraged employees to inquire about how sales are recorded for tax and other purposes might present a double-threat to cash businesses that evade tax through off-the-books purchases and sales. Employers would worry that employees would profit from their knowledge by whistleblowing, or by theft.

E. Tax Preparers for Cash Businesses

1. Overview

Available tax preparation and accountancy services vary in sophistication and cost. CPAs at the largest accounting firms--Deloitte, Ernst & Young, KPMG and PricewaterhouseCoopers--occupy the high-sophistication, high--cost end of the spectrum. CPAs at national firms such as BDO Seidman, Grant Thornton, Moss Adams offer barebones, discount service to taxpayers with simple returns and low or moderate income.

Business owners we interviewed described their ideal preparer as someone who "understands cash businesses" and "will be comfortable with me" and "creative." Most of them found their preparer through a referral from a friend, family member or colleague. In some cases, business owners described their preparers as belonging to their social network. The storekeeper, for example, said of his accountant, "He's cool; he's a buddy of mine."

Even business owners without a good lead from a trusted source reported no trouble finding the right kind of preparer. One business owner who once asked an accountant to sign a false tax return to present to a lender stated that she had gone through four or five accountants in the past few decades. "My approach is that I tell them what I need and say that if they are not comfortable with me or with being creative, we won't work out," she says, explaining that most accountants she encountered did not refuse or express discomfort.

Our field study collected data about the tax preparers who most often serve cash businesses--CPAs at national, mid-sized and small firms; enrolled agents; and bookkeepers. In our interviews, we investigated whether tax preparers know about their clients' tax evasion behavior and whether they recommend tax evasion strategies. The interviews we conducted in Centerville provide a good example of a segmented tax preparer market including preparers willing to ignore their clients' evasion and preparers willing to assist it.

2. Centerville preparers

Centerville is a medium-sized city with a mixed economy of agriculture and tourism. Its preparer market, like most preparer markets, consists of the mix of CPAs, accountants and bookkeepers, enrolled agents, and mass preparers described above.

A is a CPA whose firm counts among its clients many of the largest businesses and wealthiest families in town, including one or two publicly traded corporations and their founding families. According to A, his clients do not cheat on their taxes:

Interviewer Studies have shown that cash business owners significantly underreport income. What do you do when you have a client that fits that profile?

A: I get rid of them. Maybe warn them once.

Interviewer: Does that happen often?

A: Maybe twice in ten years.

Interviewer: That's all? Can they be underreporting income without you knowing about it?

A: Not really. You can just look at their books and see that things don't add up.

A believes that other Centerville tax preparers share his aversion to tax evasion. But the other preparers tell a different story. B is a CPA in a smaller firm with many small business clients. B guffaws when he hears A's perspective on evasion. "He thinks everyone is paying their taxes?" asks B incredulously, "I know clients of A's that cheat." B also states that he knows clients that have dropped A as an accountant because A is hard to work with. A continues to succeed, B explains, because his clients mostly consist of taxpayers who do not receive cash, or who are comfortable hiding income from their accountant as well as from the I.R.S.

B speaks knowingly about the ways business owners misreport and the segments of the local community where misreporting is most common. He acknowledges that many of his clients probably follow these practices. B is not interested in ferreting out these clients, but says he will not actively help a client create false books. "If a client is that clueless and wants that kind of thing done," says B, they have to go elsewhere. B also says that he refused to prepare taxes for two drug dealers who wanted to launder their illegal profits through a local business.

According to B, C, an enrolled agent with a shady reputation, picked up the drug dealers as clients. B also explains that C prepares taxes for the town bookies. Separately, A mentions C as a tax preparer whose clients underreport. A calls C a "nightmare" for clients, explaining that C has 120 audits a year compared to A's rate of an audit every two years or so.

The Centerville results are consistent with the other data from our study. Licensed CPA preparers who did not specialize in cash businesses were most likely to make it clear to clients that they would not tolerate unreported income. For example, one CPA told us she asked prospective clients detailed questions about cash flows, lifestyle and funding of activities. Frequently, the prospective client would not come back. The CPA reported that she accepts only ten new clients a year and has not seen her ethical standards hurt her practice, although not many of her clients were cash businesses. She said that "her license is too important to risk for a few extra bucks." The majority of cash business preparers were people like B, who suspected his clients of tax evasion and refused to help his clients cheat, but did not investigate further or make serious attempt to limit the cheating. A much smaller set of preparers seemed to go much farther and actively assist tax evasion.

3. "Don't ask, don't tell" preparers

In both our Centerville interviews and our other interviews, we consistently encountered preparers like B who maintained a distance from the details of their clients' recordkeeping, maintaining attitudes of "I didn't hear that" or "You didn't tell me that" while describing their clients as "sharp" or "street smart." One sole proprietor CPA told us that he has three clients in the restaurant industry and that he practices a "don't ask, don't tell" philosophy. He looks at checking account deposit slips, charge card income, and cash disbursements as well as payroll and other expenses and prepares the returns accordingly. He never advises about any cash business activity.

Similarly, another CPA called it "human nature" to cheat and said that his clients "hide whatever they can." He said that he doesn't ask questions unless he sees funds going through a client's checking account. If he does see such funds, he encourages clients to report that revenue since the I.R.S. might discover it. He also explains to his clients that they need to show enough income to cover their cost of living.

Our interviewees reported that many taxpayers gathered information about tax evasion tactics from colleagues, friends and family. "These guys don't need a crooked accountant" said one tax preparer. "They talk to each other all day and learn more than they would ever get from a few hours with an accountant." Some preparers reported that taxpayers calculated an appropriate net income number given fixed data points such as business rent or mortgage costs and personal living expenses. Other tax preparers cited tax preparation software such as TurboTax as a source of information for taxpayers. They stated that they believed clients were running simulations with such programs in an effort to back into a target tax amount. Many tax preparers told us that they maintained files that attempted to shield them from liability by noting that information had been provided by their clients.

From a distance, it is easy to criticize the behavior of "don't ask, don't tell" preparers. There is some research, noted earlier in this Article, that suggests that clients tend to passively follow the advice of preparers. Accordingly, active attempts on the part of preparers to ferret out and limit tax cheating might reduce evasion of many clients. On the other hand, the market for preparers in this sector is highly competitive, and there is no evidence that "don't ask, don't tell" preparers receive any form of super-normal return for their efforts. A preparer who made it clear to her clients that she would not tolerate evasion, and took active steps (for which she could not charge) to ferret out evasion, would undoubtedly lose clients to other preparers, or to electronic software. Given the competitive equilibrium, it is even possible that the optimal behavior for the individual preparer from the perspective of the society as a whole is to maintain a light touch, to nudge clients toward compliance without losing them to the involved preparers described below.

4. Involved preparers

Other preparers were more involved in their clients' tax evasion. The storekeeper quoted above is one taxpayer whose accountant helped him maintain a false set of books that will pass muster on audit:

Interviewer: Worry about getting caught?

Storekeeper: I do, that's why, the first ten years I was totally cheating. Honest, at that time everything was just pulled out of thin air, kind of just looked at what I wanted to pay, just made up figures.... Now I really got it under control, can back it up.

Interviewer: Suppose you're audited.

Storekeeper: I did get audited actually, but it was only over a couple of items. Bottom line is my accountant makes up all this backup information. So when they ask him a question, bang it's there. He goes over all the deposits, makes them reconciled with the sales.

Interviewer: What do you tell your accountant?

Storekeeper: I tell him everything.

Interviewer: Think that's typical?

Storekeeper: No, 70%, 80% don't, they just do scams on their own. That's bad news. [My accountant] tells stories, all of a sudden he finds out his client who reported 100K really grossed 500K. Nothing he can do to come up with backup now.

Another business owner described her visit to a new tax preparer and his tactic of backing into the tax payment amount:

  I visited this guy in a random office on the advice of a friend. I
  didn't know how good he'd be. So we are talking and he seems sort
  of personable and I kind of flirt and say "Look, I don't want to
  end up owing any money," and he says, "Don't worry, you'll never
  owe any money."
  
  So I go see him and he types everything into the computer and it's
  like click click click and that's the exciting part, it's like
  waiting for the envelope to be opening, to see how much I owe. And
  he screws up his eyebrows and says "I think you're going to have to
  pay some money to the state this year," and I scream out "I can't,
  if I have to pay them I can't pay you," obviously not very serious
  about not paying him but serious about not paying the state, and he
  immediately starts click click clicking and a minute later he says,
  "Okay, you don't owe anything."

We heard stories about preparers who advised clients about industry averages, profit margins, and other typical practices. One common tip business owners reported receiving from preparers, also confirmed by conversations with preparers, was to report low values for end-of-year inventory using the lower of cost or market (LCM) method. This increases the cost of goods sold and lowers gross profit and taxable income, although it presents the disadvantage of starting the next year with a low inventory value. One accountant, describing other preparers, stated that this kind of creative information might only be shared for a higher fee, such as $2,000 per return rather than $200.

One tax preparer offered an economic theory of at least some of these involved preparers. He stated that he saw tax preparers with less than ten years' experience, including ex-I.R.S, agents and tax managers of national and Big Four firms who leave to start their own practices, facilitating tax cheating by small cash businesses. "These individuals have a large nut--$300,000 home, wife and kids, BMW, country club dues, etcetera," he said "and since they are not established, they are willing to be more 'flexible' with their clients' reporting positions."

Some interviewees, including both CPAs and business owners, reported that non-CPAs were more willing to actively assist their misreporting, but this observation was not uniform. (Our interviewee base did not include non-CPA preparers.). But other business owners we interviewed used CPAs who became involved in the details of failing to fully report cash business clients' taxable income. CPAs may have less of a tendency to actively assist a tax evasion strategy, but we found that some belong in the involved preparer category.

With only a few exceptions, preparers did not describe themselves as doing anything to knowingly aid evasion. Our stories of involved preparers came from clients or other preparers. Presumably, clients feel they benefit from the help these preparers provide. Non-involved preparers feel otherwise. They view the aid these preparers give to clients who evade taxes as morally reprehensible and feel these preparers tarnish the reputation of others in the profession. Many believe the involved preparer segment of the profession tends to reduce the willingness and ability of other preparers to ferret out (or at least not actively aid) tax evasion.

Our study suggests that enforcement might be usefully directed at the involved preparer segment of the cash business market. We found that while most cash business owners could evade without preparer help, some, like the Storekeeper described above, could not. This finding is consistent with evidence outside the non-cash business sector that a small number of dishonest preparers are associated with a disproportionate amount of tax fraud.

There is no ambiguity as to the social utility, morality or legality of the actions of these preparers. This differentiates the involved preparers from the "don't ask, don't tell" preparers described above. Enforcement actions against involved preparers would not require new law or any new interpretations of professional norms, and would be supported by the great majority of preparers. An interesting social experiment would be to mount a sting operation among involved cash business preparers, effectively publicize the consequences of the operation (civil or criminal penalties for the affected preparers), and look at the effect of that operation on taxes paid by other members of the cash business community targeted by publicity.

F. The Risk of Getting Caught

As noted above, the standard economic model treats compliance as a simple cost-benefit decision in which taxpayers weigh the gains from evasion against the likelihood of detection and the penalties that accompany detection. The perceived likelihood of getting caught is, unsurprisingly, a key determinate of evasion. The likelihood of getting caught is, in turn, a function of the odds of getting audited, and the chance that unreported income will be uncovered on that audit.

The overall federal audit rate on individual returns dropped to a little over one-half of one percent in 2001 and 2002 but has risen since then. That overall figure is misleading, however. It includes as audits notices sent to taxpayers who have omitted interest or other income. Since interest income, for example, is automatically reported to the government, the omission of such income is generally inadvertent and the amounts omitted are generally small. On the other hand, the odds of audit are determined by a (secret) government regression, the so-called DIF, which is designed to maximize audit revenue. As noted earlier, the rate of underreporting in the cash business sector is absolutely high and high relative to virtually any other sector of the economy. Due to the lack of paper records and other factors, the cash business sector is difficult to audit. Still, given the absolute and relative levels of evasion in this sector one might imagine that the audit rate would be relatively high for cash business taxpayers and that taxpayers in this sector would perceive the risk of detection and penalty as relatively high.

In fact, cash business owners seemed surprisingly unconcerned about audit risk and penalties. The storekeeper's statement quoted above, provides one example; the fact that cash business owners were willing to speak so candidly to us is another. Our results suggest that one reason for this attitude is that audits and audit-related penalties were surprisingly uncommon among our subjects. The experience of one preparer in our sample is illustrative. He guessed that his clients were typical of cash business owners and hid a substantial portion of their income. Yet in a typical year only a handful of his 300 or so clients were audited and while audits produced additional payments they did not lead to civil penalties. He had never had a client threatened with criminal penalties.

The government typically realizes at least four or five times the cost of the audit from back taxes, interest and penalties levied against taxpayers. In addition, audits deter evasion. Most studies show that the latter, general deterrence, effect of audits overwhelms the direct revenue effect. One recent study estimates the general deterrence effect is over ten times the direct revenue effect, and that a doubling of audit funding at the federal level would increase taxes by as much as 60 times the cost of the additional audits. This ratio overestimates the true benefits of audits, in part by omitting the costs of time and professional fees borne by those audited.

Future studies might test our finding that cash business owners do not perceive a significant risk of audit, detection and the application of significant penalties. For example, one might survey random samples of cash business owners and employees. Cash business owners are known to systematically evade taxes while employees are known to pay virtually all taxes due on wage income, so the audit risk is certainly higher for cash business owners than for employees. A finding, consistent with our study, that the cash business group does not perceive a much greater risk than the employee group, suggests that we should try to increase the perception of cash business audit risk. For example, we might profitably put more resources into cash business audits, or at least take steps (consistent with individual taxpayer privacy) to better publicize current audits in the area.

G. Why Do Small Cash Businesses Evade Taxes?

Our interview results suggest that the reasons for cash business tax evasion are predominantly norms and opportunity--not complexity or morality or opposition to government policy. In general, the business owners and preparers that we interviewed reported that they cheated on their taxes because (1) people they know and trust who are in the same position cheat on their taxes and (2) there is a very low likelihood that they will get caught. However, a few interviewees said they believed that cash business tax evasion was roughly equivalent to a sensible government subsidy for small businesses.

1. Family and friends

Misreporting of income is such a common practice among our interviewees and their business circle that most of them found themselves at a loss at first when asked when and why they decided to underreport. "Honestly, when I saw how much tax [the I.R.S.] would take from me, I never even thought of paying it," stated one business owner. Tax preparers similarly considered failure to report cash income as an inevitable response to self-interest and opportunity.

Further questioning, however, often revealed that taxpayers learned evasion behavior from family and friends. The storekeeper quoted above, for example, had this exchange with the interviewer:

Interviewer: How did you first decide to non-report?

Storekeeper: I grew up knowing that; learned it from the swap meet where I used to sell stuff. They gave you an envelope to list sales and put in sales tax--nobody put in anything. If you did, the guy who ran it would just take it. My dad did it [i.e., non-reported] big time but always told me, "Pay the f--g taxes, there's plenty there for everyone." He was smoking but telling me not to smoke.

The same storekeeper explained that he has advised friends and associates about how to evade taxes safely. "I tell people everything," he told us, "Like never, ever deposit the cash." Other business owners we interviewed showed a similar readiness to advise others. One related, "When I saw how stupidly [my cousin] was taking the cash, in front of employees, I almost died. Lucky I got to him in time." And tax preparers confirmed that tax evasion tactics are shared wisdom among cash business owners.

The possibility of norm absorption from family and friends is highlighted by the story of two cash business owners (whom we call the Vs) who had left their longtime jobs with large employers to start their own business. They did not misreport income, but they knew that fellow small business owners in their town evaded taxes and seemed embarrassed to admit compliance. They described their preparer as follows:

Ms. V: He's a very straight, Republican kind of guy. Good, but not the kind of guy you'd hire to do things for you.

Mr. V: We could probably do better.

Ms. V: But you know that's okay, the I.R.S. is not anyone I would want to be in trouble with. He's competent; I think he's doing a good job.

Perhaps the Vs comply because they do not come from a circle of friends and family who practice tax noncompliance, but rather from a background as employees of large employers who presumably practiced tax compliance. If so, it is possible that the Vs will in time adopt the norms of their new sector.

The Vs' story suggests that new entrants into the cash sector may exhibit different behavior than those who have spent many years in the sector, and have thus learned how others underreport. It might be possible to test this hypothesis through examination of tax filings or other longitudinal research.

2. Complexity, morality and opportunity

Neither cash business owners nor preparers cited tax rule complexity as a contributing factor in their decision to evade. This may suggest that efforts to reduce the tax gap through simplification are misguided, at least for cash business owners.

In addition, for the most part, business owners and preparers did not claim that their decision to evade was morally sound. Most business owners or tax preparers we interviewed did not defend evasion on grounds of resentment or anger at the government in general or the tax system in particular. "Why do people cheat?" asked one preparer rhetorically, "I'll tell you why. You'd rather have $10 than $5. What you think of the tax isn't relevant."

Nevertheless, some interviewees described their tax evasion as sound government policy. One considered tax evasion by cash businesses a subsidy "that equates to direct subsidies to farmers or bail-outs to various international businesses." He believed small businesses deserved such a subsidy and explained that "a small business person is key to a healthy economy." Several other interviewees shared this view. Some also cited revenue pressures, arguing that full payment of taxes would wipe out many small businesses' profits.

Our interview results here generally agree with other research cited above: Tax cheating follows opportunity, not complexity or immorality, and it is shaped by peer influence. Perceptions of tax system equities are sometimes salient, but are generally less important to compliance decisions.

IV. CONCLUSION

Cash business owners rely on parallel cash economies to underreport receipts and thereby evade income, employment and sales taxes. Many preparers in this sector adopt a "don't ask, don't tell" attitude toward their clients reported receipts. A small minority of preparers, however, actively aid in their clients' evasion. Evasion seems best explained by opportunity, including the low-perceived likelihood of detection and penalty, and by peer norms. The perceived equity of the tax system has less importance, and the complexity of the tax law does not appear to play a significant role.

Our study generates a number of testable hypotheses, including hypotheses as to factors that might trigger productive audits and factors that might increase the revenue from audits. We also generate a number of policy ideas. These include prescriptions relating to the frequency of, and publicity accompanying, audits in this area, and to the policing of the small segment of the involved preparers in this area. Perhaps the greatest contribution of this study, however, is the qualitative picture it provides of taxpayer and preparer behavior in the cash business sector, which can aid the development of additional tax compliance hypotheses, communication strategies, and policy prescriptions.

TABLE 1: Summary of Field Study Interviewees 

                  CPAs   Business    Bankers
                         Owners

New York City       20          52         0
Chicago             58           8         0
San Francisco/      27          14        10
 San Jose Area
Kansas City         30           0         0
Honolulu             3           0        22
Centerville          6           6         0
Los Angeles          2          12         0
Other                3           0         0
TOTAL              149          92        32

COPYRIGHT 2009 Stanford Law School