53 Tax Notes 1523

Tax Notes

DECEMBER 30, 1991

LENGTH: 6851 words

DEPARTMENT: Letters to the Editor (LTE)

CITE: 53 Tax Notes 1523

HEADLINE: 53 Tax Notes 1523 - KAHN DEPRECIATION AND THE MINTAX BASELINE IN ACCOUNTING FOR GOVERNMENT COSTS.

AUTHOR: Johnson, Calvin H.

University of Texas

TEXT:

To the Editor:

By letter to the editor (Tax Notes, Dec. 2, 1991, p. 858), Professor Douglas Kahn defends a kind of accelerated depreciation he has advocated previously. /1/ He argues that Kahn depreciation is 'neutral,' within a tax system that generally fails to tax unrealized appreciation, and that Kahn depreciation discredits the tax expenditure budget. The tax expenditure budget, he argues, unfairly 'tips the political scales' against methods like Kahn depreciation.

I here defend the tax expenditure budget and reject Kahn depreciation. If Professor Kahn defends his depreciation method after all these years, it is worth some attention. Kahn depreciation is bad accounting because it inaccurately describes the income generated by investments. It is nonneutral in at least two senses: It discriminates in favor of high-bracket taxpayers and excludes lower- bracket investors from purchasing depreciable property; and it discriminates against investors holding savings accounts and other investments whose income is fully taxed. The method is also inconsistent with debt. Kahn depreciation violates current law because it allocates costs to carve-out interests and treats what is in fact income as if it were a recovery and a reduction of capital. It is inconsistent with the definition of 'income' as the world understands the term. Even if all this were otherwise, moreover, Kahn depreciation does not work as a baseline to define government costs in a viable system that measures the effect of tax advantages. Government responsibility for government costs is important enough, I argue here, that it needs to be preserved and supported, whatever the outcome of one small debate about depreciation.

I. Kahn Depreciation

A. Real Depreciation. Kahn and I agree enough about the analysis that I can adopt his hypothetical. A 'Dido,' he tells us, is a machine that costs $2,740. /2/ The Dido produces a net cash flow of $1,200 a year for the next three years at year end.

The Dido is like a $2,740 bank account that advertises interest of 15 percent. The investment has an internal rate of return of 15 percent because 15 percent is the discount rate that will make the $1,200 positive cash flows be equal to the amount invested. /3/ The 15-percent interest rate also means that an investor earning 15- percent interest on the Dido-like bank account could withdraw $1,200 a year from the account and be left with zero at the end of three years:

Table A

Dido at Fifteen Percent Interest

________________________________________________________________________________

(1) (2) (3) (4)

Reduction of

Interest Cash (1) for next

Account at 15% Out at year. [i.e. (2)

Year Balance on (1) year end minus (3)]

1 $ 2,740 $ 411 $ 1,200 $ 789

2 1,951 293 1,200 907

3 1,044 156 1,200 1,044

________________________________________________________________________________

The concept of income comes from interest on such accounts as this. The income from the Dido investment is the interest, computed as a percentage (15 percent) of capital invested, as identified by column (2). If there is some vestigial requirement that income be 'separated from capital,' /4/ fear not, the 15-percent interest has been harvested. The interest in the amount identified in column (2) has been received as a part of the $1,200 cash the taxpayer has withdrawn from the Dido. The income has been realized even by a cash method taxpayer.

Just as column (2) identifies the income from the Dido, so column (1) identifies its capital. The investor starts with an investment of $2,740 and reduces that investment by the amounts shown in column (4). To identify the 15-percent interest from the Dido, a schedule of depreciation deductions would allow deductions of the amounts shown in column (4) and so end up with the adjusted basis, i.e., undeducted investment, as identified by column (1). Column (4) represents the economic depreciation of the Dido, that is, it represents the portion of the $1,200 withdrawn from the account that is a recovery of capital and the amount by which the taxpayer's capital in the account goes down.

In an income tax, investments are made and continued with 'hard money' after-tax amounts. An investor has not really lost the costs invested in a Dido or a bank account, but just converted the costs from cash to another form of wealth. The taxpayer gets tax recognition for her investment costs only when she sells or withdraws her investment. Within an income tax, the ability to make or continue an investment with 'soft-money' deducted capital is an extraordinary privilege. The ability to deduct capital that still generates income is a privilege as valuable as not having to pay tax on the subsequent income. /5/ The capital of the Dido identified in column (1) is maintained as undeducted basis, as appropriately as it is that we disallow the deduction of money left in the bank.

A benefit of taxing interest, as identified by column (2), is that then and only then, the investor's pretax return goes down by the amount implied by the statutory tax rates. If we taxed the Dido interest at 10 percent, for instance, the investor's return would go down from 15 to 13.5 percent (i.e., to 90 percent of pretax interest):

________________________________________________________________________________

Table B

Ten-Percent Tax Rate

________________________________________________________________________________

(5) (6B) (7B) (8B)

Tax on

Cash Out Interest Present

at year end (10% of After-Tax Value of

[=Col(3) col. (2) Cash [(5) (7B) at

Year of Table A] Table A less (6B)] 13.5%

1 $ 1,200 $ 41 $ 1,159 $ 1,021

2 1,200 29 1,171 909

3 1,200 16 1,184 810

_____

Sum $ 2,740

________________________________________________________________________________

Column (8B) shows that an investor paying $2,740 for this investment would get a return of 13.5 percent after tax, which is exactly what a 10-percent bracket investor can expect from a 15- percent pretax investment. The investor is getting a 13.5 percent after tax return and a 10-percent effective tax rate.

A bonus of the taxing real interest is that the Dido is of equal value to a high- and a low-bracket taxpayer. Thus, if Congress enacted a 60-percent tax on investment income, it would expect the investor's return to go down from 15 percent before tax to six percent after tax (i.e., to 40-percent of pretax interest). A 60- percent bracket taxpayer who can expect six percent after tax from 15-percent investments would also pay $2,740 for the Dido:

________________________________________________________________________________

Table C

Sixty-Percent Tax Rate

________________________________________________________________________________

(5) (6C) (7C) (8C)

Tax on

Cash Out Interest Present

at year end [60% of After-Tax Value of

[=Col(3) col. (2) Cash [(5) (7C) at

Year of Table A] Table A) less (6C)] 6%

1 $ 1,200 $ 247 $ 953 $ 899

2 1,200 176 1,024 912

3 1,200 94 1,106 929

___

Sum $ 2,740

________________________________________________________________________________

Column (8C) shows that the investment gives six percent after tax. Come prince or come pauper, both will pay the same $2,740.

Finally, if the Dido is entirely debt-financed, at 15-percent interest, taxing the interest identified by column (2) of Table A will prevent sheltering deductions that allow the taxpayer to keep unrelated income without tax. If we assume the investor buys the Dido with debt and agrees to pay the creditor 15 percent over three years with level payments, the investor will pay $1,200 per year at year end. Since the interest on the debt will be the same as the interest from the Dido, the tax accounting will describe the taxpayer's wash position if and only if the interest taxed from the Dido is the same interest allowed on the debt. Allowing faster recovery of capital then identified by column (4) of Table A, will allow sheltering deductions against unrelated income.

Economic depreciation identified by column (4), in sum, is pretty elegant stuff. It is neutral among taxpayers of different tax brackets and neutral among investments that bear full tax on their income. It is consistent with our treatment of debt.

B. Kahn Depreciation. Kahn depreciation, by contrast, does not identify the interest from the Dido nor the capital investment that is generating that interest. Kahn depreciation would subdivide the Dido chronologically into carved-out interests of one year for each of the $1,200 cash flows. The total cost of $2,740 would be allocated to each one-year carve-out, according to the present value of the $1,200 cash flow at 15 percent, as of the time of purchase. As each year goes by, the taxpayer would be permitted to deduct the basis allocated to the carved out interest. Table D computes an income figure using Kahn depreciation and then shows that income as an interest rate on Kahn's adjusted basis.

________________________________________________________________________________

Table D

Kahn Depreciation Shown as Interest on Capital

________________________________________________________________________________

(1) (2) (3) (4) (5)

Kahn Adjusted

depre- Basis at

ciation yr start Interest

Year [(2)/ Cash Income [2,740- percent

('n') (1.15)n] Flow [(2)-(1)] (2)-(3)] [(3)/(4)]

1 $ 1,045 $ 1,200 $ 155 $ 2,740 5.7%

2 905 1,200 295 1,695 17.4%

3 790 1,200 410 790 51.9%

________________________________________________________________________________

Kahn depreciation identifies interest from the Dido, Table D shows, that starts at 5.7 percent and rises to 52 percent of the identified capital remaining in the Dido. Interest rates do fluctuate, so there is nothing sacred about constant interest assumptions. But, the Dido hypothetical has abstracted out any details that might explain the drama of this change in interest rate. Kahn constructed the hypothetical, moreover, using present values that assumed constant interest.

Whatever Kahn depreciation is doing, its income figures (column (3)) are not identifying a real interest return caused by the investment (column (4)). His interest is too low in early years. He is calling some of his $1,200 a reduction of investment when the amount is interest income instead. His income accounts and his net investment accounts are too low in early years for the tax accounting to capture the process by which an investment generates a return. If the function of tax accounting is to identify investment income, without favor or subsidy, /6/ then Kahn's accounts are not doing it. This is bad accounting because it misdescribes.

By its logic, the method would also allow depreciation of perpetuities such as corporate stock, savings accounts, or farmland. Assume, for instance, that the Dido were farmland that generated a harvest of $1,200 per year (again net of all costs) forever. Kahn would hold that the farmer buying land was just buying harvests and he would allocate the total cost of the land among the harvests. Kahn depreciation would allow a depreciation deduction for the land as each harvest goes by. The life of the Dido asset does not change the value allocated to its early cash flows under the logic of the method. Thus in the first three years, Kahn would allow deductions of the cost of the farmland at the same $1045, $905, $790 schedule that applied to the three-year Dido. The argument would also allow a taxpayer to deduct amounts on deposit in her bank account, at least if it were possible in advance to identify the present value of withdrawals.

Under current law, an infinite-life asset like the land is not depreciable. After a year goes by, the investor has not lost anything. The investor has not recovered or reduced her capital because of the cash flows because she still has her capital. The farmer still has the land. The depositor still has the savings account to generate interest. Our understanding of what 'income' is, for the purposes of the income tax, comes by analogy from the harvest off the land and the interest from a savings account. Kahn depreciation has misidentified some of the income, calling it return of capital instead. The income identified by his depreciation accounts is not 'income' as the world knows it.

Kahn depreciation would also cause the hypothetical 60-percent bracket taxpayer to pay more for the Dido than the 10-percent bracket taxpayer will. Table E & F show the present value of the after tax cash flows using the logic of Tables B and C, except that Kahn depreciation is substituted for economic depreciation in the calculation.

________________________________________________________________________________

Table E

________________________________________________________________________________

Ten-Percent Tax Rate Investor; Kahn Depreciation

________________________________________________________________________________

(5) (6E) (7E) (8E)

Tax on

Kahn

Cash Out Income Present

at year end [10% of After-Tax Value of

[=Col(3) col. (3) of Cash [(5) (7E) of

Year of Table D] Table D) less (6E)] 13.5%

1 $ 1,200 $ 16 $ 1,184 $ 1,043

2 1,200 30 1,170 908

3 1,200 41 1,159 793

_____

Sum $ 2,744

________________________________________________________________________________

Table F

Sixty-Percent Tax Rate Investor;

Kahn Depreciation

________________________________________________________________________________

(5) (6F) (7F) (8F)

Tax on

Cash Out Interest Present

at year end [60% of After-Tax Value of

[=Col(3) col. (3) of Cash [(5) (7F) at

Year of Table D] Table D) less (6F)] 6%

1 $ 1,200 $ 93 $ 1,107 $ 1,044

2 1,200 177 1,023 910

3 1,200 246 954 801

_____

Sum $ 2,755

________________________________________________________________________________

With Kahn depreciation, columns (8E) and (8F) show the Dido is worth more than $2,740 after tax for both the prince and pauper. But, the value is higher at $2,755 for the higher-bracket taxpayer than it is for the lower-bracket taxpayer. The difference between Kahn and economic depreciation is the difference between present value of the first and the last cash flow (assuming all are equal); for long-lived investments, the differences between the bids of different bracket investors would be material. In a world of razor-sharp competition, Kahn depreciation would ensure that high-bracket taxpayers outbid low-bracket taxpayers for the ownership of all Didos. Bright, but relatively low-bracket investors would be priced out of the ownership of depreciable assets, solely because of tax. Kahn, however, thinks that is okay, well within his definition of neutral tax.

When the Dido is debt financed, moreover, Kahn depreciation generates artificial tax losses that shelter unrelated income from tax. If we assume that debt payments are $1,200 per year, the investor has neither real gain nor loss from the Dido, ever. But, the income given Kahn depreciation, combined with normal interest deductions, give excess losses in early years:

________________________________________________________________________________

Table G

Debt and Kahn Combined

________________________________________________________________________________

(1) (2) (3)

Kahn

Income Interest Noncash

[Col. (3) Expense loss

Year of D] (at 15%) [(1)-(2)]

1 $ 155 $ 411 $ 256

2 295 293 (2)

3 410 157 (253)

________________________________________________________________________________

The gains and the losses balance each other out in Table G, ignoring time value, but the big tax losses come early, as in tax shelters, and they are recaptured only at the end. The early tax loss can be used to exempt unrelated income from tax. Because the deferred tax can be financed with less than the tax on $256, the early deduction allows consumption that is tax free from a transaction that is only a wash as a matter of true accounting. Kahn depreciation does not describe the debt-financed investment; it creates tax shelters and tax-free consumption.

C. Doctrinal Argument. Kahn argues that his depreciation method follows, by a process of legal analogy, from our general nontaxation of unrealized appreciation. As noted, Kahn's method would allow depreciation of perpetuities like farmland. Kahn argues that in buying land, the farmer is buying the harvests. He would allocate the total costs of the land to each harvest. As each harvest goes by, part of the cost of the land would be deducted. Under current law, an infinite-life asset, like the land, is not depreciable. After a year goes by, the investor has not lost anything; she still has the land. The infinite stream of harvests yet to come on the land comes closer, moreover, and replaces the harvest that has been lost. Kahn argues, however, that the harvests gone by are really lost and that the infinite stream of harvests that replace them are immune from tax as unrealized appreciation.

Kahn depreciation is inconsistent with current law at its first step. An investor has no separate basis in the harvests. The harvests are 'carve-out interests,' that is, chronological subdivisions of the investment. As a matter of positive law, basis may not be allocated to a chronological subdivision when the taxpayer retains a reversion in the underlying property. After the harvest, the investor still has her land and her capital is intact. The harvests are therefore income, taxable in full, and not a recovery of capital costs, even in part. If the harvest has in fact caused a diminution of the investor's land, she can get tax recognition for the loss, but only when she recognizes the loss by selling the land. /7/

The rationale for current law is clear and persuasive. When the investor purchases an investment she has no loss. 'In theory,' Kahn tells us, 'the taxpayer has not actually spent the dollars he paid for [an investment] but rather has converted them into a different type of property.' /8/ That theory applies as much after a harvest or two as it applies at the initial investment. Deductions are, in general, meant to measure declines in net worth and in ability to pay tax. An investor still holding the land and a depositor still holding the bank account have not lost anything and should not deduct anything.

Kahn depreciation, in sum, cannot be deduced from current law. It is not a privileged property right. No taxpayer has any expectancy right to have the method. A taxpayer could, in fact, go to jail for reporting tax in reliance on the argument.

The Kahn argument, moreover, is not really about realization. The crops from the land have been harvested and sold. The real interest income from the Dido was realized, even by cash method taxpayers, in the receipt of the $1,200. Once the income is received, the realization convention provides no protection. Kahn is also not really talking about unrealized appreciation in the value of the capital. One can, for instance, concede that increases in the value of unsold land are not ordinarily taxed, without, however, it following that the costs of land may be deducted upon each harvest.

Kahn's argument, then, deals with costs, not appreciation, and the syllogism is that costs may be deducted because gain is not taxed. That is a non sequitur. It is as if he were arguing that section 1211, which limits the deduction of capital losses, is illegal because capital gain is not taxed until sale. But, in fact, the rationale for section 1211 is exactly that losses must be limited because gains are not taxed. Depreciation, moreover, is itself a deduction of unrealized amounts, /9/ so that Kahn is arguing that some unrealized losses may be deducted because [sic] unrealized gains are not taxed. Kahn would also concede that the investor in land has not lost any capital overall. The loss from a bountiful harvest can be seen only with a certain kind of trained eye that ignores the offsetting gains from the future crops. Out of a transaction, the holding of land or bank accounts, that gives no realized gain nor loss on net, Kahn finds some bad news and concludes that costs invested in the land or the bank account have truly been lost. Looking at the land or bank account without blinders, of course, the taxpayer has lost no costs and she still has the land and account. Kahn argues in his letter that 'there is no need to require income inclusion to balance a deduction taken by anybody.' /10/ But of course, it is the investor herself who is taking the excess deductions under Kahn depreciation, and to describe her true nonloss, once each crop or cash flow is viewed separately, the good news must be balanced with the bad. Kahn is in sum arguing for the deduction of fully hedged, unrealized amounts that are not losses. Nothing about a realization convention requires that. /11/ If it did, it would be time to ignore such a silly convention.

A final and necessary element of the Kahn's argument is the key premise that every taxpayer is entitled to a tax treatment equal to the least-taxed taxpayer. In his original article, Kahn argued that a taxpayer could get basis in a harvest or other investment cash flow, by eschewing a reversion. If each $1,200 cash flow were purchased from a separate seller of harvests (or other future cash flows), an investor could achieve both a separate basis for the future harvests and unrealized appreciation treatment as the future harvests grew nearer. Kahn was making an analogy, as we lawyers sometimes do, between a thought experiment and the client at hand and arguing that the client should be able to keep up with the thought experiment, without regard to policy.

I doubt sales of separate future harvests are a very big commercial deal or a very serious threat to the tax base. If they were, we would probably have to do something about them. What is important is not how we treat his thought experiment transaction, but how it affects normal investment income. It is important to maintain the tax base. Just because one investor slithers out of the tax base cannot mean that all investors must be let go. Our tax system is far from perfect. It has many mistakes, many more serious than the failure to tax separately sold future harvests. If all taxpayers are entitled to the weakest link in the tax base, as a matter of right, then as a matter of right, there can be no tax. If we grant any validity to the argument that tax must keep up with the least-taxed investor, then we cannot maintain any revenue.

There is, in sum, not much to Kahn's argument. It is bad accounting, bad distributional policy, bad in its allocation of resources, bad law, bad legal principle, and bad logic. But then again, Kahn does not himself seem to think that Kahn depreciation is mandatory, but only that the possibility of such a depreciation rebuts the concept of government costs. To finish the debate, accordingly, we need to discuss government costs.

________________________________________________________________________________

II. Identifying Government Costs

________________________________________________________________________________

There is a fallacy in Washington that tax give-aways involve no cost. Sophisticated people, who ought to know better, sometimes argue that programs giving up billions in tax revenue are cost-free to the government because the give-ups are not part of the budget. The argument seems to assume that if the government can avoid cash method receipt, the government has no jurisdiction or responsibility for the cost. Off-budget means cost-free. Beneficiaries of various tax advantages encourage the fallacy. The ongoing game in Washington is to find a way to get the juice for one's favorite programs off- budget. Budgets are nasty things, involving competition among programs for real resources and annual reviews by the administration and appropriations experts. Budgeted costs are hard to hide and their nature as costs are understood by the political process. The watchdogs of Treasury oppose budgeted costs, but not off-budget costs; budgeted costs can be sequestered, but not off-budget costs. If cost is considered, the benefits from one's favorite programs might be considered too expensive. If the benefit is considered cost- free, however, then the program will survive.

The idea behind the tax expenditure budget is that a tax advantage is sometimes a government cost as much as a budgeted government check is. Tax advantages will increase the deficit and give away government revenue, just like a government check does. The W-2 people,who bear the residual burden in tax, must pay for both government checks and tax advantages. Tax advantages channel capital and distort private choices in exactly the same way that government checks do; that is, by increasing the after-tax wealth of the beneficiaries of the program at the expense of the government.

There is a pressing need for the polity to take government costs more seriously. There are many programs that look wonderful, ignoring the costs, but which flunk a cost-benefit analysis once costs are taken into account. Look at section 103, for instance, which allows state and local governments to borrow money paying interest that is not taxed to the lender. The benefit of section 103 is that it reduces the interest expenses state and local governments must pay. But section 103 is wasteful because its costs exceed its benefit. The revenue lost by the federal government in foregoing tax on the interest is larger than the reduction in interest which the issuing authorities enjoy. So long as one investor has an effective tax rate that is higher than the 'implicit tax' (i.e., the lowered interest rate on the bonds), the federal cost exceeds the benefit to state and local governments. The middlemen in the program -- high effective tax rate investors -- take out too much of the cost. The program would be made much less wasteful by repealing it and replacing it with government checks. Once we look at the costs of section 103 as real costs, moreover, then for the first time we might even ask why we should subsidize cities that push off their operating costs into the future or why we should subsidize a shift in capital projects from private sector to municipal hands? /12/ Far better to subsidize reduction in borrowing, rather than an increase.

Section 103 survives in an environment in which revenue sharing died, because section 103 is not a budgeted cost. Budgeting is the primary process by which government strives for rationality and compares programs in competition with other costs. Section 103 was adopted by a hidden, less rational, less legitimating process than budgeting. Section 103 was adopted on the assumption that it was cost-free.

Whatever the outcome on section 103, in any event, analysis of cost is crucial. If the cost of program is greater than its benefit, then one should properly 'tip the political scales' kicking them really hard toward repeal.

The tax expenditure budget is an attempt to make the government more responsible for the effects of tax. A tax expenditure is considered to be a government cost because it is a departure from some base line of revenue that government could collect. A tax expenditure is an opportunity cost and it is sometimes difficult to measure the true opportunity. But then discount rate, too, is an opportunity cost concept, measured from alternative returns that an investor could make, and we can rarely specify what alternative returns will really be in the future with any accuracy. For both discount rate and tax expenditure, we so desperately need the concept of cost that we must tolerate some fuzziness in the definition of what the opportunity really is.

The ideal baseline from which to measure a tax expenditure would separate government costs from government noncosts. General tax cuts, for instance, are not reasonably treated as government costs because government does not expect any special activity to be subsidized. /13/ It would be silly to do a cost-benefit analysis of what the government got for its revenue loss in a general tax cut. Taxpayer's use of the money is privileged waste; the money is for the taxpayer to spend and the government has no responsibility for it. Similarly an increase in the standard deduction is not a cost, although it losses revenue, because it is not intended to affect behavior or resources in a way that would be measured by a cost benefit analysis.

Tax expenditure analysis should identify how tax shifts resources. If industry A pays tax at 34 percent, the statutory tax rate, but industry B, pay tax at an effective rate of 1/2 percent, then resources will shift from industry B to industry A. What commonly happens is that industry B established its advantage by combination of legal doctrine and historical accident. Some kind of effective rate or tax expenditure analysis is needed to see how tax is distorting investments. /14/ Tax expenditure analysis is just a kind of revenue estimate and without a revenue estimate, policymakers have no possibility of understanding what they are doing on a national level.

Kahn argues that Kahn depreciation is the proper baseline from which to measure the government opportunity. Only depreciation faster than Kahn depreciation is a tax expenditure. He could not claim that Kahn depreciation best describes the world as a matter of good accounting and he could not claim that a taxpayer has a property right in the method under current law or tax doctrine. Instead, his argument is based on an analogy to a tax avoidance scheme that he believes an investor might be able to achieve. An investor who bought cash flows separately and sold them without retaining a residual interest could get an allocation of costs among cash flows and that is Kahn's baseline.

There are, however, many other opportunities for a taxpayer to avoid tax in our far from perfect tax world. A salaried employee can avoid taxable income by becoming a beachcomber. A wealthy investor can avoid U.S. tax by renouncing citizenship and moving overseas. Generalizing beyond Kahn's one small tax scheme, Kahn's baseline would measure the government cost from the least tax that might be achieved under some possible (but not winning) tax scheme. With enough imagination about the scheme, one could find a baseline for estimating revenue loss would be 'mintax,' that is, government revenue that is about as close to absolute zero as liquid hydrogen. The tax expenditure budget, Kahn believes, is defeated by a scheme that fails both under current law and sound policy. The government is not responsible for the effect of any tax, he is arguing, if the tax is greater than the mintax. Kahn has found a magical way to reduce government costs with a kind of voodoo accounting. He defines all government costs away.

Kahn argues that Congress might adopt a depreciation method less generous than Kahn depreciation if it finds that there are 'overriding' policy considerations. But if Kahn were right about his definition of cost, then there would be nothing that could do any overriding. If Kahn depreciation entailed no government cost, then it would be Pareto optimal: Beneficiaries of the method would like the advantage of Kahn depreciation and no one else would have any cost. If advantages are free lunch, how can anyone rationally oppose them? Kahn in fact seems to have an intuitive concept of cost that is part of what he calls 'overriding policy considerations.' It is just that his measurement of cost prevents any possibility of quantifying the cost or using it in any political decision played under hardball rules.

I also wonder what it is that Kahn thinks policy considerations have to 'override.' There is no property right, policy consideration, or accounting sense to his method. Kahn's best argument is the taxpayer could achieve a tax scheme by buying cash flows separately and selling them without retaining a reversion. I cannot see that the tax scheme carries much weight. I hope that decisionmakers are not up nights worrying about it, when they decide how to tax depreciable property.

I would not, in the end, want to use Kahn depreciation for a system of government accountability. What system of government accountability, I wonder, does Kahn think that his mintax baseline fits into? What is it that he is trying to measure? The tax expenditure budget is not the only conceivable quantification of the affects of tax. But I would hope that whatever system is used prevents hidden costs and misjudgments based on ignoring costs. The government should be responsible for government costs, and I cannot see how Kahn depreciation helps that responsibility along.

________________________________________________________________________________

Sincerely yours,

Calvin H. Johnson

Arnold, White & Durkee

Centennial

Professor of Law,

University of Texas

December 12, 1991

________________________________________________________________________________

FOOTNOTES

________________________________________________________________________________

/1/ Kahn, 'Accelerated Depreciation--Tax Expenditure or Proper Allowance for Measuring Net Income?' 78 Mich. L. Rev. 1 (1979). My short attack on Kahn depreciation is found at 53 Tax Notes 858 (Nov. 18, 1991), which is quick review of Johnson, 'Soft Money Investing Under the Income Tax,' 1989 Ill. L. Rev. 1019 (1990).

/2/ Dido is also the name of the queen of Carthage who fell in love with Aeneas (The Aeneid of Virgil, Bk. I, lines 712-728 & Bk. IV) -- just as Raphael, Michelangelo, Donatello, and Leonardo are also the names of Italian painters.

/3/ The internal rate of return (IRR) is the constant discount rate that makes the positive cash flows from the investment have the same present value as the negative cash flow. See, e.g., Brearly & Myers, Principles of Corporate Finance 71 (3d ed. 1988). IRR has anomalies that mean that it is not always like positive interest on a bank account, but none of them are relevant here.

/4/ In Eisner v. Macomber, 252 U.S. 189 (1920), the Supreme Court held that the government could not tax a stock dividend because a stock dividend is not separated from capital. For more recent willingness of the Court to tax without separation, see e.g., Hillsboro Nat. bank. v. Commissioner, 460 U.S. 370 (1983) (no recovery needed to reverse prior deductions into income); Helvering v. Horst, 311 U.S. 112 (1940) (tax on son's affection); Helvering v. Bruun, 309 U.S. 461 (1940) (tax on return of taxpayers own land with unseparated building attached). See also, Cottage Savings Ass'n v. Commissioner, 111 S. Ct. 1503, 1510 (1991) (standard required to satisfy the administrative purposes underlying the realizations requirement is not demanding).

/5/ The seminal article is Brown, 'Business-Income Taxation and Investment Incentives,' in Income, Employment and Public Policy: Essays in Honor of Alvin H. Hanson 300 (1948), but the thesis is now standard public finance economics.

/6/ Cf. Portland Golf Club v. Commissioner, 110 U.S. 2780, Slip Opinion No. 89-530 at 15 (June 21, 1990) (the statute reflects an attempt to measure economic income, not serve some ancillary purpose).

/7/ Commissioner v. Gillette Motor Transport, 364 U.S. 130 (1960) (payment for lease was rent, not sale of underlying property); Hort v. Commissioner, 313 U.S. 28 (1941) (tenant's payment for cancellation of lease was rent to landlord and not recovery of capital); Reggio v. United States, 151 F. Supp. 740 (Ct. Cl. 1957) (payment in lieu of interest was not recovery of capital). See J. Dodge, The Logic of Tax 252-257 (1989).

/8/ Kahn, supra note 1, 78 Mich. L. Rev. at 13.

/9/ Michael McIntyre, in a Letter to the Editor, 53 Tax Notes 1319 (December 16, 1991) attacks Kahn by arguing that depreciation is an unrealized loss. Kahn's response is that his depreciation is not measuring loss at all, but merely allocating cost of income to the income. Id. at 1320. The issue comes down (again) to whether the costs reside in the land or the harvest. The harvest has been sold (realized) but the land has not and its cost is unrealized. As the text argues, current law insists on looking at the costs as residing in the land.

/10/ 53 Tax Notes at 1081.

/11/ See Cottage Savings Ass'n v. Commissioner, 111 S. Ct. 1503, 1510 (1991) (standard required to meet the administrative purposes underlying the realization requirement is 'not demanding').

/12/ See, e.g., M. Mussa and R. Kormendi, The Taxation of Municipal Bonds 190 (1979) (section 103 causes municipalities to undertake capital projects that private capital would reject).

/13/ An editorial in The Wall Street Journal attacked the tax expenditure budget on the ground that the budget assumed that the government was entitled to all money. The Journal asked, rhetorically, who was doing the cost benefit analysis for the trillions of dollars of personal income that the government let taxpayers keep. Wall St. J. (March 20, 1975) quoted in Graetz, Federal Income Taxation; Principles and Policies 55 (2d ed. 1988). But the assumption that the government should have all the money is in fact an unworkable baseline for a tax expenditure budget because it does not measure the inefficiency (nor the inequity) caused by differential tax impacts, that is, it does not identify the government cost.

/14/ Thuronyi, 'The Concept of Income,' 46 Tax L. Rev. 45 (1990) argues that a tax expenditure is properly a measure of unfairness. If a tax advantage is fully capitalized by the party on the other side of the transaction, however, an investor will get a return from an investment benefiting from a tax expenditure that is no higher than normal after tax returns. Tax expenditures are thus not necessarily unfair. Tax expenditures are about distortion in use of resources and hence they are more about efficiency than inequity.

________________________________________________________________________________