71 Tax Notes 377

Tax Notes

APRIL 15, 1996

LENGTH: 9249 words

DEPARTMENT: Special Report (SPR)

CITE: 71 Tax Notes 377

HEADLINE: 71 Tax Notes 377 - INEFFICIENCY DOES NOT DRIVE OUT INEQUITY: MARKET EQUILIBRIUM & TAX SHELTERS.

AUTHOR: Johnson, Calvin H.

University of Texas

SUMMARY:

Calvin H. Johnson is a professor of law at the University of Texas. This article arises from the author's side of a debate with Professor Theodore Sims of George Washington University at the American Bar Association Tax Section meeting in New Orleans on January 20, 1996. The author wishes to thank Professor Erik Jensen for inviting him and moderating the debate and Ted Sims. The author also thanks his colleague, Joseph Dodge, for helpful comments and Paul Williams for research assistance.

Professor Sims, on his side of the debate, argued that limitations on deductions from debt-financed tax shelters should be repealed to promote market equilibrium. With repeal, he argues, the market would drive down the pretax returns from shelters so that the returns reached equilibrium with returns from debt and other generally available investments. Sims would exclude debt owed to the seller of property from the buyer's basis, but he would then repeal section 469 (passive activity losses) and other restrictions on shelters.

Professor Johnson argues that Sim's remedy to exclude seller debt from basis is overbroad. Johnson then argues that market equilibrium will not cure tax shelters if section 469 and other restrictions are repealed. The supply of tax-favored investments is too large and too elastic, Johnson argues, and returns from tax- favored investments have never dropped to equilibrium with returns from debt in high tax brackets. Even if returns did reach equilibrium with debt, he argues, tax shelters would still allow taxpayers to consume early at too cheap a price and to invest in projects that waste capital.

________________________________________________________________________________

Calvin H. Johnson, 1996

________________________________________________________________________________

Table of Contents

I. Two-Party Debt. . . . . . . . . . . . . . . . . . . . . . . .377

A. Tax Float Diagnosis. . . . . . . . . . . . . . . . . . . .377

B. Cure Without a Disease. . . . . . . . . . . . . . . . . . 378

II. Tax Shelters Revanche. . . . . . . . . . . . . . . . . . . . 380

A. What's a Tax Shelter?. . . . . . . . . . . . . . . . . . .380

B. The 1986 Indictment Against Tax Shelters. . . . . . . . . 380

C. Capitalization Doesn't Cure. . . . . . . . . . . . . . . .381

D. Debt Isn't Right Under Partial Capitalization. . . . . . .384

E. Shelters Cost Us/U.S. Too Much. . . . . . . . . . . . . . 385

III. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . .387

________________________________________________________________________________

TEXT: 15 APR 96 In a recent article subtitled, 'Tax Shelter Abuse Reconsidered,' /1/ Professor Theodore Sims diagnoses the tax shelter problem as arising exclusively from debt owed to the seller of depreciable property ('two-party debt'). Two-party debt is part of depreciable basis on the buyer's side, but it is often excluded from tax on the seller's side until paid. Sims's remedy is to exclude all two-party debt from basis, whether the debt is bona fide or not. Professor Sims would then repeal all other anti-tax-shelter legislation. /2/ He argues that capitalization in the market would heal the remaining tax shelter problems if only tax restrictions on debt-financing were lifted.

Asymmetrical treatment of two-party debt is a useful diagnosis of part of the tax shelter problem, this article argues, although it is neither a new nor a patentable idea. Excluding debt from basis, however, hurts some transactions that should not be hurt and leaves untouched some transactions that need to be reached. Tax shelters, best defined, are transactions worth more after tax than they were worth before tax and they are a problem even when they arise from bona fide cash borrowing. Shelters funnel capital into wasteful investments and allow taxpayers in high brackets to avoid tax on their consumption at far too cheap a price. The Tax Reform Act of 1986 was right, on the merits, to try to kill them. We cannot and should not return to the pre-1986 days of freewheeling marketable shelters.

This article explains why market capitalization has never happened, cannot be expected ever to happen, and would not solve the tax shelter problem if it did. Market equilibrium cannot cure tax shelters.

________________________________________________________________________________

I. Two-Party Debt

________________________________________________________________________________

A. Tax Float Diagnosis

I have long argued that inconsistent tax treatment of two-party debt is a powerful explanation of the tax shelter problem. As I argued in 1984,

________________________________________________________________________________

The tax law commonly treats two-party debt

inconsistently. The obligor -- the debtor of a two-party

debt -- gets immediate tax deductions as if unpaid debt

were already paid. However, the obligee -- the other party

to the debt -- does not pay tax on the debt until it is

actually paid. The debt saves tax on one side before the

concomitant tax must be paid on the other. The parties

therefore create an asset at government expense, a 'tax

float,' by creating two-party debt. /3/

________________________________________________________________________________

The value of the tax float asset is shared by the parties to the tax float by bargaining between them, just as partners share partnership assets.

'Tax float' debt is different from borrowing cash from a third party. If a buyer borrows cash from a third party, the cash paid to the seller causes immediate tax to the seller. The third-party lender, moreover, usually has cash to lend only because the lender has paid tax on the cash -- or because some shareholder or creditor of the lender paid tax on the cash before giving it to the lender. With cash borrowing, the lender is just lending its posttax position to the borrower. The borrower just carries over the lender's basis in the cash for the term of the loan. /4/ Cash borrowing does not create a tax float asset vis--vis the government.

The asymmetrical treatment of shelter debt is an important point. The fact that the shelter promoter did not have to include the shelter debt in taxable income or gain when it arose was a necessary (although not sufficient) condition for the most abusive of tax shelters in the pre-1986 days of bountiful shelters. Certainly if the shelter promoter had had to pay immediate 50-70 percent tax (in the old days) on the abusive $l million nonrecourse debt he charged for jojoba bean planting or mining nonexistent coal, he would not have been selling jojoba beans or nonexistent coal shelters. The economist's analyses in the shelter debate before 1986 almost always assumed that the debt behind the tax shelters was bona fide cash borrowing and thus economists missed the two-party debt and the ugly two-thirds of the shelter market.

B. Cure Without a Disease

While inconsistent treatment of two-party debt is a helpful diagnosis, Sims's remedy, excluding from basis any debt owed to the seller, is not targeted very well to the problem. Commonly Sims would apply his remedy where there is no inconsistency or advantage to the seller.

On the face of it, debt should be included in basis unless there is a good reason for it not to be. As a matter of economic theory, in a perfect income tax, future payments should be included in basis, discounted to their present value at a pretax interest rate. Interest should then be allowed as a deduction, under compound schedules, as the interest is earned. Only in this way is it possible to make the value of property independent of the tax bracket of the taxpayer bidding for the property. /5/ Sims has himself recently argued the point in an elegant piece with lots of calculus. /6/

Debt might reasonably be excluded from basis on the ground that it is sham debt or confetti debt in which the claimed face amount of the debt does not even vaguely resemble the present value of real future payments. /7/ Sham debt, however, is definitely not Sims's point. /8/ If the debt is bona fide, the exclusion of the debt from basis has to rest, on a second-best theory, as a way to compensate for some other imperfection in the system. Taking debt out of basis or disallowing interest deductions might well be justified on the ground that the purchased property is not being treated in the way a perfect income tax would require. /9/ Sims, however, is distinctly not attacking debt on this ground; he spends much of his piece defending the inclusion of third-party debt in basis, even if the property purchased with the debt is generously taxed. Thus Sims's remedy has to be be justified, if it is justified at all, as a way to get at tax floats, because of the undertaxation of the seller. Sims, however, would impose his remedy even when there is no undertaxation of the seller.

Commonly, the seller for two-party debt gets no tax advantage. The seller, for instance, might have a full cash basis for the property and be selling the property at a loss. The transaction thus is not creating basis on the buyer's side before the seller had basis. Seller might also have elected out of section 453 installment reporting and be paying immediate ordinary income tax on her gain at a rate higher than buyer's tax rate. Recent reforms of section 453 have dramatically contracted the scope within which section 453 is available or advantageous. Section 453A, for instance, imposes an interest charge on the seller for the tax deferral for a whole range of large sales reported under the installment method. The seller might have paid interest on the tax deferral under section 453A that left the seller worse off in net present value terms than if seller had paid immediate tax. Still, in all these cases, Sims would take away two-party debt from the buyer's basis. The buyer might be left with no basis at all and thus have to pay tax on the full gross receipts from the property without recognition of the depreciation cost. When there is no advantage to the seller, the remedy is a penalty without a disease to be cured.

Even when the seller gets some advantage from section 453, excluding installment debt from buyer's basis is usually not a very good proxy for repeal of section 453. The seller may be in a low tax bracket or tax-exempt and thus not owe much tax on the sale, even if section 453 were not available. The gain may be capital gain or very small in amount. A few dollars saved on the seller's side might be the reason why the buyer lost $100 million of basis. Repeal of section 453 would not be a penalty, under plausible theory, but Sims's remedy, taking big debts out of basis, is a penalty that does not fit the crime very well. If the target of Sims's remedy is just inconsistent treatment of buyer and seller, why not just repeal what is left of section 453?

1. PILING ON. Sims's penalty on the buyer can ordinarily be expected to be on top of the penalty a seller will be trying to impose on the buyer, even when the seller in fact uses section 453 advantageously. A seller, reporting gain under the deferred installment sale method, will ordinarily be trying to demand extra payments from the buyer. The seller, who was getting a fair market value return on unrealized appreciation, will be hurt by an installment sale reported under section 453. Before the sale, the seller had a totally untaxed capital in the unrealized appreciation, with the prospect of holding indefinitely with tax forgiven upon death. /10/ With the sale, however, the seller has future tax on her capital due when installments are to be paid and has forfeited the section 1014 step-up in basis at death. /11/ The seller will need to increase interest or other payments received to above the former fair value return level so as to cover future tax. The seller thus will be piling a penalty on the buyer on top of the penalty that Sims would impose.

2. TAX ON GROSS RECEIPTS. Sims's remedy, finally, would in principle create a hybrid method of accounting under which a taxpayer would have to accrue income without being able to deduct the related expenses. There is no principled distinction, for instance, between depreciable basis, which is the focus of Sims's article, and accrued expenses. Some of the most abusive pre-1986 shelters rested on accrued expenses. /12/ Including unpaid debt in basis is just a subdivision of the accrual method.

Consistent application of Sims's remedy would leave accrual- method taxpayers with a terrible hybrid method of taxation in which taxpayers had to accrue gross income, including receivables to be paid long in the future, but could not get a deduction for the matching payables because they are two-party debts. Taxpayers would pay tax on gross, rather than net, profits or 'income.' The remedy is, thus, not driven by good theory.

3. STICK TO THE ABUSES. There were certainly abuses in tax shelter debt under pre-1986 law. The worst abuses were the cases where the debt was a 'silk purse from a sow's ear,' that is, the tax savings claimed by reason of the debt were worth much more than the debt cost in pretax, time-value- of-money, economic terms. /13/ Such silk-purse or cost-free debt is a boon to taxpayers, if it passes tax challenge. Taxpayers are indifferent as to the price they are paying for the underlying property and they are even indifferent as to whether there was any property there. The debt itself makes the profit. Excluding silk- purse debt from basis is perfectly reasonable.

When debt was cost-free to the debtor, then we can rest assured the promoter or seller was avoiding tax on the debt by some way or other. Cost-free debt on the investor's side is a supra-confiscatory tax on the recipient's side. The promoter or seller has to have been a cash-method recipient, paying tax only upon payment, a tax-exempt pension fund, or have made a section 453 election or avoided tax in some way. Still, exclusion on the recipient's side may be necessary to the abuse, but it is not a sufficient condition for the abuse. Not every two-party debt is cost-free to the obligor and not every two- party debt has a present value that is misstated.

________________________________________________________________________________

II. Tax Shelters Revanche

________________________________________________________________________________

Having overkilled the problem when two-party debt is at issue, Sims then argues that there is no problem when shelters arise from cash borrowing and he would repeal all other anti-tax-shelter remedies. He would, for instance, repeal section 469 (limiting deduction of passive activity losses), section 465 (limiting deduction from amounts not at risk), section 163(d) (limiting deduction of investment interest), and especially section 265(a)(2) (limiting deduction of interest traced to tax-exempt municipal bond interest). His argument is based on a faith in the free market to end abuses. If debt financing were available for tax-advantaged investments, without government restrictions, then the market left on its own would capitalize away the remaining problems. /14/ This part rejects the argument.

A. What's a Tax Shelter?

1. NEGATIVE TAX. A tax shelter is an investment that is worth more after tax than before tax. There is a negative tax or subsidy, better than mere tax exemption or zero rate tax. A tax shelter saves the investor tax that would otherwise be paid on consumed amounts or other outside income. A tax shelter has a higher rate of return because Uncle Sam tries to run a tax system that it would have if Uncle Sam let everybody go.

Negative taxes commonly arise because of a mismatch between income-tax-type treatment of debt and our consumption-type treatment of investments. Debt works right only with investments giving only undeferred ordinary income tax -- for instance, certificates of deposit. Debt-financed in anything better taxed than CDs will yield artificial accounting losses unconnected to any economic loss. /15/ If borrowed cash is used to buy tax-advantaged investments, the cash is included in basis, and interest is deducted, then cash borrowing will produce artificial deductions that no longer describe the economic losses in standard of living. The deductions in excess of true loss are the artificial accounting losses that shelter outside consumption and produce the negative taxes. Even bona fide third- party debt can generate negative tax.

Tax shelters are also investments treated better than consumption tax would allow. Consumption tax theory says that we should have a zero effective tax on investment returns. /16/ Tax shelters go below zero tax on investment returns to yield a negative tax on investment returns. A true consumption tax would disallow interest deductions or tax borrowed principal to prevent negative tax. /17/ If we at least taxed all consumption, from whatever source derived, that would end tax shelters.

The negative tax in a tax shelter, moreover, is tax-rate dependent. The negative tax gives more of subsidy to high-bracket taxpayers than to low-bracket or tax-exempt taxpayers. High-bracket taxpayers then drive up the price of favorably taxed assets and drive out the talented young entrepreneurs simply because their tax rate is too low. When negative tax subsidies are available, you would then have to be in the maximum tax bracket just to own a tax-advantaged investment or compete in a tax-advantaged industry. You would have to be in the top bracket, for instance, to run a Mom and Pop farm or to own a house in a fancy neighborhood.

B. The 1986 Indictment Against Tax Shelters /18/

In the Tax Reform Act of 1986, Congress tried to end most tax shelters. The 1986 act was a grand compromise between suspicious allies that sought to cut both rates and tax shelters. The core premise behind the act was revenue neutrality at the top. Top-bracket taxpayers got a significant tax rate cut, down from 50 percent in 1985 and down from 70 percent in 1981. In exchange, however, they were to give up tax shelters. Tax collected from the highest tax brackets as a whole was supposed to go neither up nor down in amount. /19/ Both in cutting rates and in making taxes less avoidable, the 1986 act improved the economic efficiency of the tax system. The high-tax- rate, pretend taxes that everybody avoids do economic harm without helping the deficit or anybody else.

The 1986 indictment against tax shelters had two counts:

________________________________________________________________________________

Count 1: EXEMPTION OF LUXURY CONSUMPTION. Supposedly,

shelters were artificial accounting losses by which rich

people avoided tax on their luxury consumption with tax

losses that weren't real. Once luxury consumption is freed

from tax, then tax revenue has to come from less optimal

sources.

Count 2: WASTE. Supposedly, tax shelters caused lots

of economic damage by channeling resources into junky

investments: jojoba beans in Costa Rica, windmills in the

Mojave Desert, garden apartments outside of Dallas that

nobody wanted to live in, see-through office buildings put

up just for the tax shelter. We gave up billions of

dollars of tax revenue to buy junk.

________________________________________________________________________________

On both counts, the 1986 indictment was right.

1. SECTION 469. The primary siege gun against shelters in 1986 was the passive activities limitations of section 469. Section 469 defers the deduction of losses from 'passive activities' until the investor totes up the cash at the end of her voyage and shows she had real cash losses. An investor is also allowed to use the losses as the activity goes along to the extent she can show some income from some other passive activity. Section 469 is a kind of capitalization rule, much like deferring expenses under section 263, or carrying over NOLs or capital losses until the taxpayer has some matching income. Section 469 is based on skepticism that the losses reported on shelters are real. Certainly many of the losses caught by the limitation were artificial accounting losses that did not represent diminution of standard of living or ability to pay tax.

It is hard, however, to defend in theory the specific lines that section 469 draws. 'Passive activity' does not do a very good job of separating the artificial losses in tax shelters, which should be capitalized and carried forward, from the real losses, which should be allowed as they happen. The limitations defer at least temporarily the deduction of some very real losses.

On the other side of the coin, losses from activities the taxpayer materially participates in are allowed as they accrue under normal tax rules. 'Material participation,' however, does not do a very good job of ensuring that the reported losses are real. Material participants are people who are ordinarily quite willing to avoid tax with phony accounting deductions if the opportunities are available to be had. Section 469 overkills on some things and lets some shelters slip through -- much as Sims's basis-exclusion remedy would do. In any event, the hunt in 1986 went after a quarry -- tax shelters -- that was well worth the kill.

C. Capitalization Doesn't Cure

The passive activity limitations, Sims argues, are aimed primarily at debt-financed investments in tax-favored investments. He would repeal the limitations and as well as all other restrictions on debt-financed investments into favorably taxed activity. His rationale is that shelters would be self-healed by capitalization if debt-financed investments were only available. /20/ This section responds that full capitalization cannot be expected and that, in any event, capitalization would not cure the disease.

1. DEFINITION OF TERMS. To explain or debate capitalization, it is necessary to define some terms. 'Capitalization' here refers to the drop in pretax return rates because of market reactions to the investor's tax advantages. 'Full capitalization' means that the return from tax-advantaged investments drops into par with normal after-tax returns that a maximum tax bracket taxpayer would get from a savings account or any routine fully taxed investment.

Assume, for example, that fair market (taxable) interest is 10 percent and that the maximum tax rate imposed and intended by the statute is 40 percent. Taxpayers can then get 6 percent from a fully taxed savings account or certificate of deposit, after paying their tax. Muncipal bonds are tax-exempt under section 103. Full capitalization means that the pretax return from municipal bonds will drop to 6 percent before tax. The 40 percent tax-rate investor would pay no tax on the municipal bonds, but still under full capitalization, the investor gets the same 6 percent that she could achieve from any routine bank deposit.

Similarly under full capitalization, the pretax returns from every investment drop so that the investor gets only 6 percent in the end. Life insurance, property held until death, ACRS equipment (with ITC pre-1986), qualified pension plans, and other effectively zero-tax investments will return only 6 percent.

Full capitalization also means that investments that bear some tax, but under favorable terms, will also return only 6 percent after tax. Assume, for instance, that accelerated depreciation means that tax on investments in depreciable property will take away only one- seventh of annual pretax income, even for investors in the highest tax bracket. Full capitalization would mean that the pretax return from that accelerated depreciation will drop to 7 percent, so that when tax takes away one-seventh, the after-tax return will drop to 6 percent. Full capitalization means that returns reach a constant sea level, under which every investment in the economy is at equilibrium or par and gives the same 6 percent return, no matter whether the investment started as tax-favored or disfavored.

When tax favors are engineered to accomplish something, capitalization also represents the process by which the benefit of the exemption passes over to the municipality or other activity. Under full capitalization, for instance, municipalities issuing bonds get to borrow paying 6 percent, when taxable rates are at 10 percent. Muncipalities save 4 percent interest, which is the point the exemption is supposed to serve. With full capitalization, issuers have captured the benefit of the exemption away from the maximum tax bracket investor. Full capitization means that the investor as middleman keeps nothing but the normal return from the investment and passes over the rest of the tax benefits to the activity.

Full capitalization also means that taxpayers in tax brackets lower than the maximum tax bracket cannot longer rationally own tax- favored investments. A taxpayer facing a 30 percent tax rate, for instance, would do better investing in taxable deposits paying 10 percent, than in municipal bonds. The 30 percent tax-rate investor can get a 7 percent return from fully taxed sources, but only 6 percent from tax-favored sources.

The amount by which the pretax return rates on tax-favored investments drop in reaction to the tax advantages is sometimes called an 'implicit tax.' /21/ Under full capitalization, the implicit tax rises to equal the maximum statutory tax rate. /22/ The implicit tax, it should be noted, unlike real taxes, does not help the federal deficit or go to the federal government. Investors who are in a tax bracket that is lower than the implicit tax should just pay their taxes at their low statutory rates instead of going into tax-advantaged investment that carry an implicit tax. The implicit tax is sometimes the intended purpose of the tax advantage, for instance, for municipal bonds or accelerated depreciation, but sometimes the implicit tax is a harmful side effect of the tax advantage, for instance, when implicit tax drives taxpayers with less than the maximum tax rate from ownership of homes or farms.

'Partial capitalization' means that the return from tax- advantaged investments drops some, but not enough as to reach 6 percent. Under partial capitalization, there is some investor with a tax rate that is lower than the maximum rate, the 'marginal investor,' who gets an equal return from the fully taxed savings account and from tax-exempts. For that investor the implicit tax and the statutory tax rate are the same. Higher tax rate, 'infra-marginal investors,' however, get to keep some of the tax benefit under partial capitalization. With partial capitalization, the implicit tax rate does not take away the advantage for the highest statutory tax rate.

Assume, for example, that fair market taxable interest is still at 10 percent and municipal bond interest rates drop to 8.5 percent. The marginal (15 percent) tax bracket investor then gets an 8.5 percent return both from muni-bonds and certificates of deposit; that investor is indifferent between implicit tax and paying real tax. For the infra-marginal investor, however, the implicit tax is lower than the statutory tax rate. Under partial capitalization, the full benefit of exemption did not pass over to the municipality, i.e., it was not capitalized. Under the example, the 40 percent bracket investor saved tax equal to 4 percent interest, kept 2.5 percent as windfall and passed over 1.5 percent to the municipality.

2. WHY PARTIAL? Partial (rather than full capitalization) happens because total supply of tax-advantaged investments is too large to be absorbed by highest-tax-bracket investors alone. As long as issuer or promoter needs (or could use) one investor in less than the highest tax bracket, the return must be set above 6 percent to attract the last or marginal investor. To attract 30 percent tax bracket investors, for example, the investment must give 7 percent, since they can get 7 percent from their CDs. As the supply of tax- advantaged investments gets larger and larger, the return must be set higher and higher to attract demand for tax advantage from investors in lower and lower tiers. The tax rate of the marginal investor determines the implicit tax on the whole. As long as the marginal investor, allowed aboard the train, has a tax rate less than the maximum tax rate, then full capitalization is impossible.

3. FULL CAPITALIZATION HAS NEVER HAPPENED. Graph 1, following, shows the implicit tax rate on 10-year municipal bonds over a 25-year period. Tax-exempt muncipal bonds have had an implicit tax, but at a rate considerably below the maximum statutory tax rate. The benefits of tax exemption are partially capitalized but not fully capitalized:

Graph 1 shows that implicit rates on municipal bonds are noisy, but have not reached full capitalization. Longer-term bonds show a smaller implicit tax, because the bonds must compete with step-up in basis at death and other 'free' tax exemptions. Other compilations, looking at longer-term bonds, have found implicit taxes as low as 7 percent. /23/

The implicit tax, shown on Graph 1, acts a thermometer measuring the health of the entire tax system. The implicit tax, at any time, shows what fraction of the tax benefits is passed on to the municipality or other activity intended to be subsidized and what fraction is kept as windfall by the middleman highest-bracket investor. The municipal bond market is the most efficient market for tax advantages because the market has many bidders, many offerors, and a cadre of professionals that follow it. Any publicly available information will be reflected in price of bonds and the returns they give within 15 minutes or so. Municipals are also fluidly in competition with all other investments at the margin, so implicit tax on municipals reflects maximum implicit tax throughout the tax system. The implict tax is also the highest tax a capital investor needs to bear. Full capitalization has not happened anywhere in the system if it has not happened in the municipal bond market.

________________________________________________________________________________

[GRAPH 1 OMITTED]

________________________________________________________________________________

4. AIN'T GOING TO HAPPEN. Full capitalization should also not be expected ever to happen. The implicit tax is determined by the laws of supply and demand and the supply swamps the demand. Implicit tax rates less than the maximum tax rate happen because the supply of alternative tax shuns is too large to be absorbed by the demand from maximum tax bracket investors. Graph 1 is itself good empirical evidence that the supply of tax advantages swamps the demand in the top bracket. Even beyond the empirical evidence, there are a number of reasons to think that the the supply of tax-advantaged arrangements is so large and elastic that full capitalization can never happen even in theory.

5. LARGE SUPPLY. The existing supply of tax-favored investment is large. The list of tax-exempt or nearly tax-exempt investments available to high-bracket investors is long, including, for instance, state and local bonds, owner-occupied houses, appreciating property held until death, and life insurance. The list also includes those investments that are effectively tax-exempt because the investment may be made with tax-deducted capital, including, for instance, qualified pension plans, research and development, development of business goodwill and other intangibles, oil drilling, coal mining, and farm crops. Before 1986, depreciable property was effectively tax-exempt because the investment tax credit and ACRS depreciation in combination were worth about as much as the tax that would be collected from income on the machine. While ACRS has been cut back, the effective tax rate on depreciable property is still low. Galper & Steuerle estimated, before the 1986 Tax Reform Act, that 80 percent of individual capital was in effectively zero-rate investments. /24/ Congress shows no immediate signs of a dramatic cutback in the tax advantages for houses or gain held until death or life insurance or research or oil or municipal bonds or any of the effectively tax-exempt investments.

6. SELF-CREATED SUPPLY. The supply of tax-exempt or near tax- exempt investments can also expand cheaply and elastically. Tax exemption sometimes comes from mere legal or financial arrangements. Investors, for instance, can use tax-exempting vehicles such as qualified pension plans or life insurance to make their underlying investments if the implicit tax on other investments gets large. An investor can invest in any industry or economic activity through a qualified pension plan or life insurance. There is no need to invest through a high-taxed vehicle like debt or capital stock. Advantages of tax exemption achieved by the vehicle cannot be capitalized away by the ultimate industry or activity because the ultimate recipient of the investment does not even know the source of the capital and does not know whether the investor has a tax advantage or not. The activity or industry cannot price discriminate. Returns set for the fully taxable investor (10 percent) must be accorded to the tax- exempt vehicle as well. Since the vehicle can be created by lawyers and accountants, there are also few effective limitations on the supply of the vehicle.

Sometimes tax advantages just come from the lawyer's or accountant's word processor. /25/ There are no limitations on such supply, at least in the short run, until Congress and the courts catch up with the abuse and decide that it needs to be limited. By that time the word processors will probably be off to some other scheme.

7. CAPITALIZATION DRIVES UP PRETAX RATES. In some cases, the capitalization process itself increases the pretax rate of return, rather than driving it down. Holding property until death, for instance, will give the investor a tax forgiveness on the built-in gain on the property at death. If implicit taxes rise on other forms of tax-advantaged investments, then holding until death becomes relatively more attractive. As capital flows into property with good potential for holding until death, existing capital owners get more unrealized appreciation they can hold until death. Their pretax returns go up, not down. Capitalization takes away inequity only when it is driving the pretax rate of return down to 6 percent, not when it pushes the appreciation above the going interest rates.

8. COUNTER-ARBITRAGEURS. Low- and zero-tax rate institutions (like university endowments) can profitably buy taxable bonds and sell tax-exempt bonds short, whenever the implicit tax gets significant. They thus push down implicit taxes and prevent capitalization. /26/ There can be no inevitable law of capitalization when for a large group of investors 'equilibrium' occurs only when there is no implicit tax.

9. CAPITAL FLOWS AROUND TRIVIA. There are restrictions on debt- financed investments in municipal bonds, the subject of Graph 1. Sims attributes the failure of municipal bonds to reach full capitalization to the restrictions on debt-financing. /27/ Section 265(a)(2) prevents deduction of interest to buy municipals, but it does not prevent capital investment in equity format rather than debt format. If the supply of tax-advantaged investments were not so large, equity would flow around section 265(a)(2) like the tide around a bean pole. Section 265(a)(2) also sometimes relies on a tracing rule and so it is avoidable by tracing borrowing into and through a general business account. /28/ The tide (demand for tax advantage in the top bracket) is absorbed by other investments or transactions so that it cannot make the implicit tax high enough for full capitalization.

In a world of a large and elastic supply of tax advantages, repealing section 265(a)(2) would strip high-bracket compensation out of the tax base, tier by tier from the top, without having a material impact on the implicit tax. A taxpayer with $l million salary income, for instance, tiring of tax, could then just borrow enough to have a $1 million interest deduction and use the borrowed proceeds to buy municipal bonds. The interest deduction would shelter the salary and the taxpayer would report tax-exempt bond income instead. Because of some implicit tax, the bonds would give less than $l million, but in the absence of full capitalization, the implicit tax would not be so bad as the explicit tax Congress was trying to collect. The tax base would disappear on high-bracket consumption, tier by tier from the top down, and once luxury consumption was freed from tax, then Congress would have to collect all its revenue from less optimal sources. Repeal of section 265(a)(2) would increase the demand for shelter, because salary would be eligible for shelter, but if nothing were done about limiting supply or access to shelter, the huge and elastic supply would soon swamp even the expanded demand. The 1986 act tried to limit the supply and access to tax shelter, which is the only way that it is possible to maintain a tax on the optimal, high- tax-bracket sources.

D. Debt Isn't Right Under Partial Capitalization

In a world of only partial capitalization, debt does not work right. Debt financing becomes more valuable

than equity financing and debt-financers can pay more for property. With partial capitalization, original issue discount does not cure the problems of debt and does not even fix the cost-free debt problem.

1. DEBT MATTERS TO THE PURCHASE PRICE. When generally available after-tax discount rates on investments are higher than full capitalization returns (6 percent in our continuing example), the net present value of ACRS equipment (or other advantage) is higher for debt financing than for equity. Graph 2 shows the net present value for debt and equity with general discount rates set by partial capitalization.

The underlying property in Graph 2, depreciable property, depreciated under section 168 faster than the economic depreciation that would maintain the statutory tax rate, but not so fast as to give the investment a zero effective tax rate. Debt financing yields a higher price for the property in Graph 2 because, under partial capitalization, equity investment has a higher opportunity cost than debt does. For an equity investment, the taxpayer has a discount rate of 8 percent after-tax return, under the assumptions. Debt costs only 6 percent after tax. Since debt is cheaper, the debt financier can bid more for the property. A given, tax-advantaged investment, moreover, is always better the higher the tax rate avoided. Tax shelters are more valuable, the higher the tax that Congress is trying to collect.

2. IMPLICATIONS. Graph 2 shows that high-bracket taxpayers, with debt, will pay the most for tax-advantaged investments and drive lower-bracket taxpayers out of the market. All things being equal, the smart, entrepreneurial taxpayers are excluded from ownership of the equipment because their tax bracket is too low.

3. OID DOESN'T FIX. Under conditions of partial capitalization, moreover, the original issue discount remedy does not cure the worst tax shelter abuses. With partial capitalization, OID does not fix the 'silk purses' problem, that is, the problem in which the time-value of the tax savings from a debt can exceed the time-value detriment of the debt itself. When they have 'silk-purses' debt, debtors will thereafter try to increase debt. When that happens, as noted, taxpayers are indifferent to the value of the underlying property or even whether it is real. Graph 3, on the next page, shows the terminal value of zero-coupon debt, less the tax savings generated by the debt. Graph 3 shows that value can be negative when general discounts rates do not drop to full capitalization.

________________________________________________________________________________

[GRAPHS 2 AND 3 OMITTED]

________________________________________________________________________________

E. Shelters Cost Us/U.S. Too Much

Full capitalization would not cure the tax shelter problem, even if we assumed that full capitalization could happen. Under full capitalization, even the low-value tax shelters defer tax at a rate that is cheaper than the cost the federal government bears from the tax deferral. Shelter investors then make investments and take early consumption, paying a price that is lower than the social cost of the deferral. The too-cheap tax deferral means that, even under conditions of full capitalization, tax shelters generate both inequity and inefficiency.

A shelter at its minimum value defers tax to a following year. Full capitalization means that zero-tax investments drop to 6 percent return, in our continuing example (10 percent pretax interest and 40 percent maximum tax rate). Even with the drop in return, however, deferral of tax costs the investor only 6 percent. The tax deferral, however, will cost the federal govern- ment something on the order of 9 percent. The federal government must pay 10 percent pretax to finance the deficit caused by the deferral and collects relatively modest tax on its debt. Debt is held by low- tax or tax-exempt investors. The federal cost for tax deferral is 10 percent pretax, less something like 1 percent tax collected on the interest, for a net cost to Uncle Sam of 9 percent.

The taxpayer making the decisions as to whether to defer tax is in a classical position in which waste is encouraged. The taxpayer thinks the cost of deferral is 6 percent, and will make decisions that would be justified with a 6 percent cost. The true federal cost of the deferral, however, is 9 percent. The private decisionmaker who goes into the shelter will make wasteful decisions, imposing 9 percent cost on the rest of us while paying only 6 percent interest herself.

1. HERBAL ROLLOVER EXAMPLE. Assume, for instance, a minimum value shelter that just defers tax for a year. Assume Taxpayer A, in a 40 percent tax bracket, makes and consumes $l million salary. Tax would be $400,000 but Taxpayer A, instead of paying tax, buys an 'herbal rollover' shelter, borrowing $l million cash at 10 percent interest to plant herbs in some warm country at year end. Planting costs are deductible when made. /29/ The shelter generates $l million deduction, at no net out of pocket cost to A since everything was borrowed. No tax is due in year 1.

In year 2, the shelter turns around and Taxpayer A must pay her original $400,000 tax, plus 6 percent interest. Full capitalization means the return rate from the herb planting is only 6 percent, in the 10-percent-interest economy. All of the gross receipts from the harvest are taxable, because the taxpayer has expensed her basis the year before. The accounting is as follows:

________________________________________________________________________________

1. Gross Receipts (after fees and $ 1,060,000

transaction costs) (6 percent of $ 1

million)

2. Tax on 1. at 40 percent (424,000)

3. Repay Borrowed Principal (1,000,000)

4. Pay Interest at 10 percent 100,000

less deduction (40,000)

----------

60,000 (60,000)

SUM: Pay Deferred Tax + ----------

6 percent ($ 424,000)

________________________________________________________________________________

Under the hypothetical, deferring tax cost the taxpayer 6 percent interest, or $24,000.

Meanwhile in the example, the federal government was expecting $400,000 from Taxpayer A on her $1 million consumed income to pay the Marines. Because Taxpayer A has avoided tax in year 1 by planting herbs in a warm country, the federal government must borrow the $400,000 to pay the Marines, promising 10 percent interest or $40,000 to its financiers. The U.S. collects $4,000 tax at 10 percent weighted-average tax rate from low- and zero-tax institutions that lent the deferred tax. The net cost of the deferral to the federal government was $36,000. In sum, the cost of deferral to Taxpayer A is $24,000 (6 percent), whereas the net cost to U.S./us was $36,000 (9 percent).

Taxpayer A wasted capital in the example. She consumed her $400,000 tax early, at a cost to her of only 6 percent, whereas her early consumption cost U.S./us 9 percent. Taxpayer A put real resources into herbs paying 6 percent, whereas the real cost of capital was 9 percent. Taxpayer A has put her taxes into projects paying less than real interest cost and on early consumption that does carry its costs. We must now borrow capital from the Japanese, at the margin, and it is too precious to waste.

Full capitalization in the example is fairer than no capitalization. If the harvest was valuable enough to give 10 percent pretax, then the accounting for year 2, just shown, would have yielded a gross of $1,100,000 and Taxpayer A would have had to pay $400,000 tax in year 2. Paying $400,000 in year two would represent a zero-interest loan. Capital gain and lower tax rates in year 2 or various kinds of tax fraud or avoidance could make the shelter even better and give the investor less than zero interest on the deferral. Thus, 6 percent interest on deferral of Taxpayer A's tax is better than it could be. It is just not good enough.

2. CLIENTELE EFFECT WILL ALWAYS CAUSE WASTE. The situation in the example in which taxpayers borrowed at too low a cost, compared to the social cost, will always happen if interest is deductible because of the 'clientele' effect. The 'clientele' effect segregates investors according to their tax bracket and leaves only low- or zero-bracket investors as the holders of federal debt. Debt, including federal debt, is a high-tax investment: investors must make the investment with undeductible capital and must pay ordinary income on the interest annually whether or not paid. High-tax-bracket investors eschew debt as an investment and go into better taxed investments instead. The investors left in federal debt are then always the lower-bracket investors. The lower bracket investors are driven out of the tax-advantaged investments by the implicit tax, even under partial capitalization, but their zero- or low-tax rate status protects them from the worst consequences of the taxation of federal debt. If tax were the only consideration in the world, the segregation into clienteles would be absolute, separated by an iron curtain at the marginal investor's tax rate (i.e., the implicit tax). All investors with a tax rate higher than the implicit tax would buy tax-advantaged investments and all investors with a tax rate lower than the implicit tax would buy debt, including federal debt. The clientele effect means that the federal government's cost will always be higher than the shelter investor's cost and that the shelter investor will neither pay nor judge his actions by the full cost by which the tax deferral hurts us all.

3. SECTION 163(h). The same clientele phenomenon that makes shelters per se wasteful also justifies the disallowance of a general deduction of personal or consumption interest. If interest on borrowing to consume were deductible, a taxpayer could borrow $1 million to consume early. The cost to the taxpayer would be 6 percent, in our example, whereas the public cost would be 9 percent. Section 163(h), disallowing deduction of personal interest, is thus sound legislation when it works, even under an income tax.

________________________________________________________________________________

III. Conclusion

________________________________________________________________________________

Professor Ted Sims has argued that we need to take the two-party or seller-provided debt out of the buyer's basis for depreciable basis and that we should then repeal all other restrictions on tax shelters, including the passive-activity-loss restrictions. If restrictions on debt-financing are lifted, he argues, then market equilibrium will cure the harm from tax shelters.

Market equilibrium will not heal tax shelters, this article has argued. Equilibrium between the returns from tax-favored investments and from debt has never happened and cannot be expected to happen. The supply of tax-favored investments is too large and too elastic. In absence of equilibrium, the interest deduction on debt-financed investments in tax-favored assets does not work right. In absence of equilibrium, debt-financed, high-bracket investors bid up the price of tax-favored investors and drive out all lower-bracket competitors. In absence of equilibrium, debt can become cost-free after tax: the tax savings generated by the debt is more valuable than the debt itself costs in real or present value terms.

A tax shelter is an investment with a negative tax: the return from a tax shelter is higher because of the government's attempts to tax than the return would be in absence of tax. Tax shelters, if available, would allow taxpayers to consume early at too cheap a price. Tax shelters, when available, encourage investors to go into projects that give too low a real rate of return. Even if tax shelters gave a return that reached equilibrium with returns from debt, shelters would still allow early consumption at a cost lower than it costs the government and shelters would funnel capital into projects that waste capital. We are a capital-poor country now that must borrow capital from abroad. Capital is too expensive to put into wasteful projects and early consumption.

________________________________________________________________________________

FOOTNOTES:

________________________________________________________________________________

/1/ 'Debt, Accelerated Depreciation, and the Tale of a Teakettle: Tax Shelter Abuse Reconsidered,' 42 U.C.L.A. L. Rev. 263, 271 (1994),

/2/ Shelter restrictions Sims would repeal include section 469 (passive activity losses), section 465 (at-risk limitations), section 163(d) (investment interest) and section 265(a)(2) (interest on debt traced to municipal bonds).

/3/ Calvin Johnson, 'A New Way to Look at the Tax Shelter Problem [Letter to David Brockway, Chief of Staff of Joint Commitee on Taxation],' Tax Notes, May 14, 1984, pp. 765-766. See also Calvin Johnson, 'Front End of the Crane Rule,' Tax Notes, Apr. 30, 1990, pp. 593, 599; Calvin Johnson, Statement on Relief from Imputed Interest Rules of Section 1274, in Tax Treatment of Imputed Interest on Deferred Payment Sales of Property, Hearings before the House Ways and Means Committee, 99th Cong., 1st Sess. at 168 (April 24, 1985); Calvin Johnson, Statement on Installment Sales, in Tax Simplification, Hearings on S.1063 before the Subcommittee on Taxation and Debt Management General of Sen. Finance Committee 96th Cong., 1st Sess. at 187 (1979).

The argument is just an obvious, nonpatentable, extension of section 404(a)(5), which prevents an employer from deducting a debt for compensation before the employee includes it in income. Tax floats are a problem, like the target of section 404(a)(5), even when the parties are not related as employer and employee.

/4/ Cash borrowing is much like section 1015 giving a donee the donor's basis or section 362 giving a corporation basis for property contributed to it by its shareholders. The carryover of basis is only temporary, however, because the borrower returns the basis when he or she repays the cash.

/5/ See Emil Sunley, 'Observation on the Appropriate Tax Treatment of Future Costs,' Tax Notes, Feb. 20, 1984, p. 719 (arguing that future costs discounted at pretax value should be included in basis); Donald Keifer, 'The Tax Treatment of Reverse Investment,' Tax Notes, Mar. 4, 1985, p. 925 (supporting Sunley with helpful tables); William Klein, 'Tax Accounting for Future Obligations: Basic Principles,' Tax Notes, Aug. 10, 1987, p. 623 (supporting Sunley with helpful spreadsheets). The theory consistently, however, requires that property acquired with the debt be depreciated under a slow, 'economic depreciation' that allows a deduction only as the investment value of the property goes down. Paul Samuelson, 'Tax Deductibility of Economic Depreciation to Insure Invariant Valuation,' 72 J. Pol. Econ. 604 (1964).

/6/ Theodore Sims, 'Environmental 'Remediation' Expenses and a Natural Interpretation of the Capitalization Requirement,' 47 Nat. Tax J. 703 (1994).

/7/ Levin v. Commissioner, 832 F.2d 403, 407 (7th Cir. 1987) (Easterbrook, J.) (debt payable in super-inflationary currency became confetti, not worth collecting, by the time it was due); Agro Science Co. v. Commissioner, T.C. Memo. 1989-687, 89 TNT 262-8, aff'd without opinion 934 F.2d 573, 91 TNT 73-10 (5th Cir. 1991) ($600,000 research and development deduction claimed on basis of debt in hyper-inflating currency worth under 1/1000th of claimed debt). Confetti debt, worth a trivial fraction of the claimed face, can arise from contingencies, including nonrecourse features (see, e.g., Calvin Johnson, 'Front End of the Crane Rule,' Tax Notes, Apr. 30, 1990, p. 593 (arguing that nonrecourse liability is not a cost cognizable by tax, unless the down payment is larger than the value of the tax savings) and from time-value errors. See, e.g., Calvin Johnson, 'Silk Purses from a Sow's Ear: Cost Free Liabilities Under the Income Tax,' 3 Amer. J. of Tax Policy 231 (1984) (arguing that debt saved more tax than it really cost, even when subject to OID remedy of sections 1272-1274 because of general availability of tax-advantaged investments).

/8/ Sims mocks the 'confetti debt' argument as 'picturesque.' Sims, supra note 1, at 333 n.287.

/9/ See part II and especially infra note 15.

/10/ Section 1014.

/11/ Section 691(a)(4).

/12/ For example, liabilities to pay R&D expenses to research the use of shark's blood as a cure for cancer, accrued expenses to pay planting expenses for jojoba beans in Costa Rica, accruals to pay long future interest computed under Rule of 78s so that first-year interest deduction exceeds principal, accrued expenses for promoter's fees sufficient to shelter all subscribers' income but not payable for 100 years.

/13/ See, e.g., Calvin Johnson, 'Silk Purses from a Sow's Ear: Cost Free Liabilities Under the Income Tax,' supra note 7, (tax savings can be worth more than debt in time-value terms); Calvin Johnson, 'Front End of the Crane Rule,' supra note 7 (nonrecourse liability is 'silk-purse' debt when down payment is too small).

/14/ Theodore Sims, supra note 1, at 303-304

/15/ As to expensed investments, see William Andrews, 'A Consumption-Type or Cash Flow Personal Income Tax,' 87 Harv. L. Rev. 1113, 1121, 1136-1139 (1974); Calvin Johnson, 'Tax Shelter Gain: The Mismatch of Debt and Supply Side Depreciation,' 61 Texas Law Rev. 1013 (1983). As to accelerated depreciation, see Calvin Johnson, 'Soft Money Investing under the Income Tax,' 1989 Ill. L. Rev. 1019, 1070.

Sims argues that the tax value of artificial deductions are offset by the drop in pretax returns called 'capitalization' (see Sims, supra note 1, at 303), but he would not deny the artificiality of the tax deductions in the first place (Id. at 290). Whether capitalization cures the artificial shelter losses is the point of this article and the debate.

/16/ See William Andrews, supra note 15, at 1121, 1123-1128, 1167.

/17/ Id. at 1153-1155, 1158.

/18/ Section B draws heavily from Calvin Johnson, 'What's a Tax Shelter?' Tax Notes, Aug. 14, 1995, pp. 879, 880-881.

/19/ See, e.g., Jeffrey Birnbaum and Allan S. Murray, Showdown at Gucci Gulch: Lawmakers, Lobbyists, and the Unlikely Triumph of Tax Reform 23-31, 204-208, 218-222 (1987). Senator Bill Bradley of New Jersey was the orginating sponsor of the revenue neutrality idea. David Brockway, 'Comprehensive Gift and Estate Tax Reform,' Tax Notes, Aug. 14, 1995, pp. 1089, 1093.

/20/ Theodore Sims, supra note 1, at 271, 303. Prior literature arguing that capitalization will heal the problems of shelters if only the restrictions on shelters were removed, include Frederick Hickman, 'Interest, Depreciation and Indexing,' 5 Va. Tax Rev. 773, 774, 788 (1986) (failure to allow interest deduction on debt-financed investment in municipal bonds is deliberate sabotage of municipal bond program); Lawrence Zelenak, 'When Good Preferences Go Bad: A Critical Analysis of the Anti-Tax Shelter Provisions of the Tax Reform Act of 1986,' 67 Tex. L. Rev. 499 (1988); Joseph Bankman, 'The Case Against Passive Investments: A Critical Appraisal of the Passive Loss Restrictions,' 42 Stan. L. Rev. 15 (1989).

/21/ Myron Scholes and Mark Wolfson, Tax and Business Strategies 84 (1992).

/22/ If there is a double tax passed through to the investors, for instance, if the corporate income is passed back to the investors, or if there is a fool's tax on inflationary gains, then full capitalization would mean that the implicit tax from tax-exempt investment would rise not just to equal the nominal statutory tax rate, but also to equal the real effective tax rate.

/23/ Arak & Guenter, 'The Market for Tax-Exempt Issues: Why Are the Yields So High?' 36 Nat. Tax. J. (1983).

/24/ Harvey Galper and Gene Steuerle, 'Tax Incentives for Savings,' 2 Stat. of Income Bull. 1, 4 (Spring 1984).

/25/ See, e.g., Joseph Stiglitz, 'General Theory of Tax Avoidance,' 38 Nat. Tax J. 325 (1986) (explaining four strategies of avoiding tax without diminution of estate or consumption).

/26/ Myron Scholes and Mark Wolfson, supra note 21, at 120.

/27/ Theodore Sims, supra note 1, 271.

/28/ See, e.g., Wisconsin Cheeseman, Inc. v. United States, 388 F.2d 420 (7th Cir. 1968) (taxpayer's interest expense could not be traced to municipal bond holdings); Ball v. Commissioner, 54 T.C. 1200 (1970) (accord). See generally, 1 B. Bittker and L. Lokken, Fed. Taxation of Income, Estates and Gifts par. 31.3 (1.) at 31-58 (2d ed. 1990). But see, Dept. of Treasury, General Explanations of the Administration's Revenue Proposals (March 1996) (proposing to require pro rata disallowance of tax-exempt interest expense for all corporations).

/29/ Section 180; Treasury reg. section 1.162-12 (1972). There are, in fact, now a number of restrictions that prevent taxpayers who have no mud on their boots from deducting farming expenses (see, e.g., section 263A(d)(1)(B), (d)(3)(B) (agribusiness must use full absorption inventory accounting), section 461(i)(1) (prepaid farm supplies), section 469 (passive activity losses), but it is assumed that these have been repealed under Sims's changes.

________________________________________________________________________________