An Employer-Level Proxy Tax on Fringe Benefits

A. Overview

When nontaxable substitutes for cash are available, an employer and employee working in concert will inevitably shift compensation away from cash to the untaxed form. Goods and services are usually less convenient and less valuable to the employee than cash. Still, as long as the value lost is less than the tax avoided, the employer and employee will continue the shift. Taxing the cash substitute as well as the cash would make the problem go away because the employee would no longer accept the lower value of the cash substitute.

The proposal would impose a nondeductible 50 percent tax on employee meals, entertainment, and recreational and athletic facilities. The tax should be thought of as a tax on employee consumption but collected by the employer. For group benefits like dining or athletic facilities, taxing each employee would be daunting, perhaps impossible. When benefits are only occasional, the employer, which generally has professional bookkeeping, is better at keeping track of the benefit by looking at the cash paid. The tax base, subject to the 50 percent proxy tax, would be reduced by amounts included in the taxable income of the beneficiary, as shown on a Form W-2 or Form 1099 prepared by the employer and by amounts the consumer has paid. The 50 percent proxy tax is equivalent to a 331⁄3 percent tax on income, as we think of it, because it would be imposed on a tax base that does not include the tax itself.

B. Current Law

The income tax allows important opportunities for employees to consume food and recreational resources without having any tax imposed. In principle, gross income includes compensation whether paid in cash or in noncash form, but the system has difficulty picking up noncash benefits.

Under section 61, gross income includes fringe benefits, unless expressly excluded.[1] Section 132, enacted in 1984, codifies exclusions for some benefits, including benefits that the employer provides at no added cost to itself, employee discounts in which the employer gives up only its profits on its inventory, benefits provided by the employer that the employee could have deducted had the employee been paid in cash and used the cash to buy the benefit, and de minimis fringe benefits. Subsection 132(e) defines a de minimis benefit as one ‘‘so small as to make accounting for it unreasonable or administratively impractible.’’ Several specific examples are expressly excluded under the general de minimis category:

Eating facilities. Eating facilities provided on or near the employer’s business premises are considered de minimis exclusions if the employer charges enough to cover the direct cost of food and beverages and the wages of the workers in the facility. The employee payments need not cover the cost of rental of the facility, normal profit for use of money, or overhead costs, and need not charge the fair market value of the meals.[2] If the employer pays an outside contractor to provide the eating facilities, the parties must allocate the contract price between direct costs (which must be paid for by employees) and all overhead and profit (which need not be covered) to determine whether the benefit is deminimis.[3] Employees using the facility because of the noncompensatory business needs of the employer, such as a need to be on call or because of a short lunch period, are deemed to have paid the requisite direct cost of meals.[4]

Occasional supper money. Supper money, provided to the employee on or off the employer’s business premises to allow the employee to continue to work overtime, is also de minimis but only if the supper money is not ‘‘regular or routine.’’[5] There is no standard or definition of regular versus occasional in the tax regulations or instructions.

Other occasional benefits. De minimis expenses can also include occasional cocktail parties, group meals, or picnics for employees and their guests; occasional theater or sporting event tickets, coffee, doughnuts, and soft drinks; and flowers, fruit, books, or similar items under special circumstances[6]

Athletic facilities. Athletic facilities, including golf courses, tennis courts, swimming pools, workout or exercise rooms, and gyms, qualify for the de minimis exception. The employer must own and operate the facility, and substantially all the use must be by employees and their spouses and children.[7]

Some consumption excluded from tax by the working condition exclusion of section 132(d) also strongly resembles de minimis exclusions. Section 132(d) excludes employer-provided benefits that the employee could deduct as a business expense if the employee paid for them. For example, section 162(a)(2) allows a deduction for meals consumed while the employee is away from his home on business. Thus, under section 132(d), the employee may exclude reimbursements received from his employer for meals while away from home on business. The rationale for subsection 132(d) is that there is no need for the paperwork if the deduction is going to mean the expenditure ultimately is not subject to employee tax. The subsection 132(d) exclusion, however, ignores the limitations the employee would face if he paid for the benefit himself. There is, for example, a 2 percent floor on miscellaneous itemized deductions, under which the deductions are allowed for employee expenses only to the extent they exceed 2 percent of the employee’s adjusted gross income.[8] Also, the excess over 2 percent is an itemized deduction. The 2 percent floor and the standard deduction usually take away any benefit from the deduction of unreimbursed employee expenses. Under the working condition exclusion, however, employer reimbursement avoids the 2 percent floor and the standard deduction override, and allows the employee to avoid tax on the meals consumed.

Section 119 excludes meals furnished to employees’ spouses and dependents on the employer’s business premises for ‘‘convenience of the employer.’’ The exclusion predates the 1984 articulation of the de minimis exclusion in section 132(e), but the administrative problems of taxing employees on meals are similar. The convenience of the employer test requires a business reason for providing the meal, besides compensation, and it is a question of fact.[9]

Under current law, even weak noncompensatory explanations are used to justify the exclusion. In Boyd Gaming Corp. v. Commissioner,[10] for example, the Tax Court concluded that a casino operator’s policy that dealers must stay on the premises for their shift was not a sufficient business reason to justify providing them meals, but the Ninth Circuit said the casino’s policy was a business decision that needed to be respected and it allowed the exclusion. It is likely that if the employer or employee paid taxes on the benefit, the employer would charge the full cost or stop providing the meals.

Under the regulations, the meal qualifies for the exclusion if it is provided so that the employee is on call during the meal or has only a 40-minute period to eat.[11]

That the meals are required by a union or state contract, or that the employee may decline the meal and still keep his job are excluded from consideration in determining whether the meal is furnished for the convenience of the employer.[12] If more than half the employees are furnished the meals for the convenience of the employer, all employees may exclude the meal.[13]

Section 274(n) disallows the deduction of half the expenses of meals and entertainment, but it does not reduce the deduction for meals that are excluded from income because they are considered to be de minimis or provided for the benefit of the employer.[14] The disallowance of half the expense in principle applies to the employer who reimburses the employee for meals while away from home, although it is not clear how well the disallowance is enforced.

Section 4977 provides for a 30 percent proxy tax on an electing employer in lieu of taxing employees. The subsection 132(b) and (c) exclusions are available only on inventory or services offered to customers within the line of business in which the employee works.[15] The employer nonetheless may exclude from taxation the no added cost and qualified discount fringe benefits to employees from other business lines, as long as the employer elects to pay the 30 percent excise tax. The excise tax is paid only when the benefits exceed 1 percent of employer cash payroll.

C. Reasons for Change

1. The damage from exemption. Tax exemption for benefits to an employee inevitably creates dead-weight losses. A dead-weight loss occurs when employees, customers, and employers get value out of a benefit that is less than the cost of the resource. When it is a given that compensation paid in cash must be taxed, the employer and employee will shift from cash compensation to untaxed forms. The shift stops only when the value lost is equal to the tax avoided.

Assume, for example, an employer commits $100 to Executive W because of the value of her services to the employer. If W is in a 35 percent tax bracket and the employer pays her in cash, W must pay $35 tax and will be left with only $65 from the cash compensation. If the employer instead gives W $100 for supper money on an occasional basis when W works overtime, the meal is tax exempt. As noted, occasional supper money provided so the employee can work overtime is tax exempt to the employee. If we assume that the $100 meal is worth $100 to her, she is ahead by the tax avoided.

Table 1. Comparison of Taxed Cash and 

Exempt Supper Money (Without Shift)

A. Cash compensation B. Exempt fringe benefit
Value of compensation costing employer $100 $100 $100
Tax ($35) $0
Net employee benefit $65 $100

Table 1 is not in equilibrium, because Executive W received too little value out of the column A cash compensation and too much value from the column B tax-exempt supper money. To remain competitive, the employer needs to use the most efficient compensation by giving W as much value for every $100 it spends, so employers will shift more $100 amounts away from cash (column A) and over to exempt benefits (column B). Executive W will cooperate in the shift.

With cash, we can presume the employee will get at least the $65 value out of the take-home salary. Cash allows each employee to maximize preferences. The meals, however, eventually become less valuable. The first $100 might well be worth its cost, but a second, third, or nth $100 meal fades in value and is no longer worth its cost. The executive will tire of the $100 meal eventually. The equilibrium point when the parties can do no better either by shifting from meals to cash or vice versa is shown by Table 2 below. The small plus sign in Table 2 represents a smidgen, too small to mention and small enough that it indicates it is time to stop shifting from column A to B.

Table 2. Comparison of Taxed Cash and 

Exempt Supper Money in Equilibrium

A. Cash compensation B. Exempt fringe benefit
Value of compensation costing employer $100 $100 +
Tax ($35) 0
Net employee benefit $65 $65+

The employer may work in consultation with valued employees, or the employer may anticipate W’s desires. It is also in the employer’s self-interest working alone to reach the equilibrium shown in Table 2 because the employer wants to give the executive the maximum value from its $100 costs. The process of shifting from Table 1 to Table 2 is not a perfectly oiled machine. The parties might misestimate the optimal shift and have meals worth less than $65 and then have to pull back to more cash. Still, given the common interest of the employer and employee to move to Table 2, on a whole range of possible exempt benefits, the equilibrium position in Table 2 seems as inevitable as water running down hill. As long as cash compensation is taxed, the two parties will shift compensation from a taxable to a tax-exempt form until there is value lost by way of lower utility equal to the tax avoided.

Table 2 shows the rational, maximizing value for the parties in partnership, but it does harm to the sum of human happiness. The difference between the $100 cost and the $65 value, or $35, is dead-weight loss. The $100 the employer spent in either column represent real resources pulled from some other use. The $100 spent will circulate around the economy and have secondary and tertiary effects on economic activity, but the dead-weight loss shown in column B of Table 2 represents a loss of value. The amounts are significant. If the utility from all resources could be transformed magically from $65 to $100, there would be an immediate 54 percent improvement in the value of all goods.

Column B also has the detriment of not providing federal revenue for the national needs. In equilibrium the shift to column B provides a small benefit to Executive W, but it causes the government to lose $35 tax it would collect on column A. Having failed to collect $35 tax from a good-source, high-tax-bracket consumption, the federal government will need to raise $35 tax from an inferior source. The tax system needs trillions of dollars in revenue in the coming years and column B does not help. Indeed, raising tax rates on other transactions can be expected to induce more shifts from taxable to nontaxed columns because the increased tax on column A types of transactions induces shifts over to column B with greater dead-weight loss. A tax on column B exempt benefits is a better tax than raising tax rates more generally for the same revenue because it will reduce the harm that tax does.

If the $100 cost in the right-hand column were subject to $35 in tax, the problem would go away. Executive W would accept the supper money in lieu of cash only if she received at least a dollar’s worth of value for a dollar’s cost. If the meal is worth less than the money spent on it, the employer alone or employer and employee working in concert would shift back and use cash compensation instead.

Indeed the need for revenue justifies taxation on column B, even for inframarginal meals when the shift has not yet produced a $35 dead weight loss. A tax of 35 percent on the $100 spent for a benefit in column B would foreclose the shifts that create a situation in which most taxpayers and most meals are consumed from after-tax income. If an employee gets the exemption, there is an equitable case for taxing Executive W on the consumption, just like other taxpayers.

Taxing the meal in column B can be expected to cut back resources devoted to meals. If Executive W cannot exclude the $100, she is likely to order a cheaper meal. Indeed, taxpayers usually buy their meals with the after-tax amounts ($65) they get from cash compensation, and buy meals with the take-home pay (here $65) rather than the gross pay (here $100) derived from their labor. If Executive W has to pay for her meal with after-tax dollars, she would insist on the same value from her money paid for her dinner as she gets from her dinners on the other nights.

The dead-weight loss in equilibrium, shown in Table 2, occurs even for meals eaten in a business context. There is dead-weight loss on tax-excluded meals spent by the executive while away from home on business, on supper money, and on meals provided for some collateral business reason. Even in a business context, the employer has the incentive to give the employee compensation because of the private ‘‘tax efficiency’’ of inframarginal meals. Thus, there is a need to reduce the dead-weight loss even for meals that are currently excluded by current law because they are eaten away from home or in some business context.

One may be away from home for business reasons or meet for business reasons, but eating is always consumption. Eating is inherently consumption, and it is the opposite of business. A certain level of calories a day is needed to keep working, even at a desk job, but the minimum calorie requirement is quite cheap to satisfy. You can get 1,000 calories from peanut butter, for example, for 85 cents.16 Getting 1,000 calories for a price greater than 85 cents is for the employee’s pleasure and consumption and not for the noncompensatory reason that the employer has in providing the meal. The costs above 85 cents per meal are discretionary payments in which the dead-weight loss in equilibrium will reign. Business is also never the sole reason for eating because one would eat even if not engaged in business.

Consumption is the primary component of the tax base, even within an income tax that also taxes savings. Consumption tax norms are becoming more persuasive.

Given the dead-weight loss on meals, it is fair to presume that while one may meet for business reasons or have to be someplace for business reasons, one eats for personal reasons.

2. Noncash. The income tax is primarily a tax on cash and cash equivalents. It is commonly argued that no taxpayer should have to pay tax on noncash benefits because of concerns over the liquidity to pay tax. Tax on cash is easier to administer than tax on noncash because the cash value of noncash goods and services must be determined.

A tax on cash only, however, would not be a viable tax in either theory or in practice. Theoretically, the only reason that cash has value is because it buys goods and services.

In practice, moreover, taxing only cash would make tax far too easy to avoid and would induce the economy to move from efficient cash transactions to noncash barters. Barters are terribly inefficient: No buyer ever has quite the goods or services you need. Tax only on cash would increase the economic damage that the tax system does, because it would induce many transactions across the economy in which taxpayer moved from cash payments to noncash transfers with dead-weight losses like those in Table 2.

In any event, a proxy tax on employers’ costs, as is proposed here, would tax the employer’s cash, not the value of the benefits. Cash is always worth accounting for because people keep track of it.

3. Collecting tax at the employer level. Despite the theory that noncash substitutes for cash need to be taxed to prevent waste, valuing the benefit at the individual employee level is often administratively impossible. The rationale for the deminimis exclusion that tax should not attempt to tax items when the revenue involved does not justify that the administrative effort is sound. Still, it is common that weighting of revenue and administrative effort depends on whether an awkward remedy at the individual employee level or a remedy at the easier employer level is used to collect the revenue.

For example, employee golf courses, swimming pools, gyms, and dining facilities are not inexpensive. Any employer might well spend millions of dollars in capital costs and annual expenses on those facilities. The deadweight loss can be expected to be materially judged on the employer and federal levels. The amount at stake makes it worth the administrative effort.

Still, if we were to judge the administrative effort of tax on these facilities by the value the employee gets out of them, the administrative costs would be very high. Are we, for instance, to place an IRS agent at the end of the cafeteria line to say, ‘‘Lemon meringue pie today, Johnson? We will charge that to your tax — also to your calorie counter.’’ Are we to charge employees for swimming pools by the lap or by the time spent in the pool? Does the duffer who spends more time on 18 holes at the golf course get more value or less value out of the course than the pro that goes through quickly? If we are to charge for access, how are we to determine comparable costs? Trying to tax each employee according to value would commonly require data that would be used only for tax.

By contrast, a proxy tax on the employer’s cost would be imposed from accounting data — cost — collected at the same time the employer accounts for its costs. No employer ignores cash costs and stays in business in a competitive market, so employers have and can isolate the data they need for a proxy tax. Costs of meals, entertainment, and facilities would continue to be deductible and it is a very rare employer who loses track of deductible costs.

Even for the occasional cash benefits, a proxy tax on the employer would be easier than an employee tax. Collecting tax from the employer will reduce the administrative burdens because it relies on the professional bookkeepers of the employer. The employer’s bookkeeper must clear the reimbursement payments and they can record the tax due at the same time. Employers must keep books for both tax and nontax reasons. They can also spread the cost of professional accounting over a much larger tax base.

The de minimis exclusion of current law has some difficult administrable rules because it is not a trivial job to determine what is deminimis. Supper money, presents, and cocktail parties are said to be excluded only if they are occasional. There is no legal definition or standard of ‘‘occasional,’’ however, and it seems clear that the line is unenforceable. In a world with trivial audit rates, taxpayers can take the position that eating only three meals a day is occasional, and mostly get away with it. If facilities are provided by a contractor, there is a great deal of hard accounting work to determine what is the direct cost of food and on-site labor, whereas a contractor will charge for all costs, including offsite employees, overhead, and profit.17 Taxing the employer at fixed rates on all costs would obviate the need to determine whether the benefit is deminimis.

4. Why the 50 percent tax rate?

a. Tax-exclusive tax base. It is proposed here that employers pay an excise tax of 50 percent on their cost of meals, entertainment, and facilities. The tax would be a ‘‘tax-exclusive’’ tax — that is, the tax itself would not be included in the amount of the tax base. We ordinarily think of tax rates on a tax-inclusive basis — that is, tax imposed on gross pay and not on take-home pay (for example, $35 in column A). A 50 percent tax on the cost transferred to the employee is equivalent to a 331⁄3 percent tax on gross pay.18 One could avoid the terrible public relations effect of a 50 percent tax, when normal maximum tax rates are 35 percent, but only with a gross-up formula that few could be expected to understand.

Similarly, a tax imposed on a tax-exclusive basis on take-home pay is really a lower-rate tax under the logic we ordinarily use to compute income tax. Thus, the 30 percent excise tax in section 4733 is imposed only on what the employee gets, normally $70 from a $100 gross amount, and 30 percent of 70 percent is only a 23 percent tax, as we normally compute tax.19 For employees in tax brackets higher than 23 percent, fringe benefits exempt from tax cause dead-weight loss even though the 30 percent excise tax applies. If one does not understand the difference between tax-exclusive and tax-inclusive tax bases, the taxes sound high.

b. Maximum high rate. To take away all dead-weight loss on shifts to noncash benefits, it is necessary to set the employer’s proxy tax so that it is equivalent to the highest tax rate applied to individuals — 53.8 percent under current tax-inclusive 35 percent rates, and 65.6 percent if maximum rates go back up to 39.6 percent. Nonetheless, a 50 percent proxy tax on the employer’s cost will reduce most of the dead-weight loss on the consumption of employer-provided meals and entertainment. A 50 percent tax is a nice round number.

For employees who have taxable income of less than roughly $200,000, the 50 percent tax is higher than the 28 percent maximum tax they pay on cash salary.[20] The proposal is intended to level the playing field, not to discriminate against meals or entertainment.

It would be possible, but complicated, to estimate an employee’s tax from the employer’s records. The employer knows how much taxable income the employee gets, and the W-4 gives some incomplete evidence of outside income. The employer might also take a survey of employees every year and base the overall withholding on the survey results. The survey results would not be accurate but they would be approximate.

The simplest remedy, however, is to rely on the self-interested economics of the parties to solve the difficult issue of adjusting to employee rates. If the proxy tax is too high, tax-efficient compensation would shift away from column B tax benefits and back to column A cash. Casinos, for example, might charge full cost to the dealers and entertainers who use their dining facilities. If the casino collects enough at the cash register to cover its true costs, including indirect cost, it will not pay any tax.

Employees will then pay for meals at the casino on the same basis that they pay for meals at home: If the lemon meringue pie is worth the price, they will buy it, and if it is not, they won’t. Forcing compensation back into cash to avoid dead-weight loss is always a good idea.

5. Skepticism as to value? In general, it is fair to assume that when an employer provides food, and the cost of the food is taxed, the employer will get value out of the food at least equal to its costs. There are contexts in which a meal or other fringe benefit is not a substitute for a cash benefit because the employer does not try to maximize the benefit to the employee. The food taster to the king, for instance, will probably get to sample very fine dishes, but the risks of the job, when the king is unpopular, offset the tastiness. The food taster is getting paid to eat, he has no choice of food, and has no chance to refuse the food. The interest of the king is adverse to the employee. An employer adverse to the employee regarding the quality of the food, however, is rare.

Except in cases like the king’s food taster, we can assume that the parties are trying to give the best benefit to the employee that time and the budget will allow. A cheap meal might not be all that pleasurable, but the taxed amount would not be high. Even if one must eat or meet in a specified place for business reasons, one can generally assume that one eats for personal consumption.

For many of the meals currently excluded, there is no reason to think that the employee will get less than full value. For example, with meals away from home and for supper money, the employee gets access to any restaurant in town and a full menu. Public restaurants stay in business only because they give customers full value for their money. Even for an in-house cafeteria or sandwiches brought into the conference room, the food may be relatively inexpensive, but the employer has no reason to make it less attractive or wholesome than its cost, as long as both meals and cash are taxed equally. Current law will treat a meal as provided for the convenience of the employer, even if the employee can refuse the meal, but of course if the meal were provided for the employer without benefit to the employee (as for instance with the king’s taster), the employee would not be allowed to refuse the meal.

Except when the employer is adverse to the employee as to the benefit from the meal, the best remedy for over taxation is again the private adjustment. If tax on the cost of a meal leads to over taxation of the meal, the employer and employee need to reduce the meal’s taxable cost. The easiest way to reduce the taxable amount in situations like employer-owned facilities is to charge more. The employer can reduce the cost of the meal to reduce its tax, or it must improve the quality and charge more. Given the seriousness of the dead-weight loss and the need for revenue, business reasons such as a short lunch period or the need to be on call can not justify the exemption of the meal. If the employer wants to give value to the employee, that is sufficient reason to tax the meal. In the end, the proposed tax would be a tax on the cost of the meal — the resources that went into the meal and not on the meal’s value.

D. Explanation of the Provision

The proposal would impose a 50 percent tax on the employer’s cost of meals, entertainment, theater and sport tickets, and dining and athletic facilities. The tax would be imposed on cost computed on a full absorption basis, allocating all of the employer’s overhead costs to some activity, including the fringe benefit. Only net cost would be subject to tax, so that employee, family, or customer payments for a meal, athletic facility, or ticket would reduce the amount subject to the 50 percent tax.

The full absorption principles of section 263A, including allocating interest to facilities under the cost-avoided method, would be used to determine the employer’s cost. The tax base would be reduced by amounts charged the employee or customer and also by amounts included in employee income as evidenced by the W-2 statements for employees and Form 1099 for nonemployees.

The tax would be imposed even when it is customers, suppliers, or independent contractors who eat the meal or use the employee gym, tennis court, or swimming pool. The employer is trying to please customers, suppliers, and independent contractors and there are the same dead-weight loss problems when nonemployees consume the meal as when employees do. There would be no reason for the employer to determine whether the facility was used by employees, retirees, or customers.

Charging more to nonemployees, however, would reduce the employee’s 50 percent tax.

The tax would not be imposed on meals, tickets, and dining and athletic facilities when the employer is adverse to the employee or customer regarding pleasure to be derived from the meal or other noncash benefit. The only example I know of that fits that model is the king’s food taster, and I suspect there are no others. Even a restaurant review writer gets the full choice of menu in his profession. Any meal exempt from the 50 percent tax on this ‘‘king’s taster ’’ grounds would have to be a meal that the employee could not refuse and keep the job. With

only a tiny exemption for the king’s taster, the strong general rule would be the cost of all meals and entertainment is subject to the 50 percent tax. The law would incorporate the maxim that one may meet for business reasons or eat in circumstances dictated by business, but one eats for personal consumption.

The cost of the meal, ticket, or facility would be deductible to the employer if the cash is deductible given the context. The loss of deduction of half the cost of a meal, now provided by section 274(n), would be repealed insofar as the amount is subject to the 50 percent tax. The 50 percent tax, however, is a proxy for an employee tax, collected by the employer, and the employee tax itself would not be deductible. It is expected that the employer will pass on the tax to the employee.

The tax on the employer’s cost of the meal is a tax on resources used and not on value received by the employee. The employer’s tax is a proxy tax for tax on employee consumption, but no attempt would be made or allowed to prove the employee’s value from the meal or the athletic facility. Not liking the food or the push ups would be no defense.

The 50 percent tax would also be imposed on nonprofit organizations exempt from tax when the employee or suppliers are getting meals, tickets, or access to dining or athletic facilities. Amounts contributed to charity have been deducted from tax. When the amount provides private benefit, the beneficiary should pay tax on the benefit, just as beneficiaries pay tax on salary.

There would be, however, an exemption for meals provided for the poor when feeding the poor is included in the tax-exempt organization’s mission. One might easily rationalize that rule as if amounts are included in income of the employee or customer. The poor pay no tax on their income because the standard deduction and personal exemptions exempt income to the destitute. We might treat all feeding of the poor as if it were taxed to the poor, but at a zero tax rate.

The proposal would repeal the current-law 30 percent tax of section 4977, which allows an employer to elect to give a no-additional-cost and employee discount exemption on inventory offered to employees not working on the product line the inventory is made or sold on.

The proposal would also repeal the section 162(a)(2) deduction for meals away from home. Dead-weight loss occurs on meals away from home. It may well be that meals away from home are sometimes more expensive than meals at home, but with peanut butter at 85 cents for 1,000 calories, the money spent above 85 cents is discretionary. Meals more expensive than the minimum 85 cents are personal costs, generally disallowed by section 262.

[1]Section 61(a)(1).

[2]Section 132(e)(2); reg. section 1.132-7.

[3]Reg. section 1.132-7(b)(3).

[4]Section 132(e)(2) (post (B); reg. section 1.119-1(a)(2)(ii)).

[5]Reg. section 1.132-6(d)(2). Supper money was first excluded by O.D. 514, 2 C.B. 90 (1920), but see, for example, TAM 9148001, 91 TNT 244-19, saying that O.D. 514 is no longer valid and holding that supper money that is more than occasional is taxable.

[6]Reg. section 1.132-6(e)(1).

[7]Section 132(j)(4); reg. section 1.132-1(e)(1989). Athletic facilities are excluded by a special rule and not by the de minimis exclusion, but I have categorized the facilities as de minimis because the problem covers both.

[8]Section 67.’

[9]Reg. section 1.110-1(a)(a) (substantial nonbusiness reason).

[10]T.C. Memo. 1997-445, Doc 97-27456, 97 TNT 190-12, rev’d. 177 F.3d 1096 (9th Cir. 1999), Doc 1999-17361, 1999 TNT 92-13.

[11]Reg. section 1.119(1)(2)(ii)(a) and (b).

[12] Section 119(b)(1) and (2).

[13]Section 119(b)(4) enacted by Internal Revenue Service Restructuring and Reform Act of 1998, P.L. 105-206. Section 5002 replaces prior law that the casinos must withhold unless ‘‘substantially all,’’ that is, more than 90 percent of employees have meals for noncompensatory business reasons. TAM 98290001. The liberalization of the rules for on-premises eating facilities was the result of lobbying by the casino industry for meals provided to dealers and dancers. Amy Hamilton, ‘‘IRS Reform’s Flying Circus — Tales of One Last-Minute Change,’’ Tax Notes, July 13, 1998, p. 145, Doc 98-22146, or 98 TNT 132-4.

[14]Section 274(n)(2)(A) and (B).

[15]Section 132(b).

[16]Figures are available at, but if the figures are inaccurate by half, the point that calories are cheap holds true.

[17]Indeed, the border for other exclusions are hard to administer: section 132(c) allows an exemption for employee discounts if the discount does not exceed the employer’s gross profit margin on the inventory sold; gross profit margin is ordinarily a trade secret, however, and employees then have no idea whether the discount exceeds the allowable amount. Those rules are apparently not written to be enforceable.



Previously published by the University of Texas – School of Law, April 2009

Andrews & Kurth Centennial Professor of Law, University of Texas at Austin - School of Law, USA