Calvin H. Johnson is the John T. Kipp Chair Emeritus at the University of Texas School of Law. He thanks Eugene Steuerle, Stephen Shay, and Daniel Halperin for helpful questions and comments.
In this article, Johnson supplements his argument that interest deductions should be limited to a ceiling of the interest rate times adjusted basis, and he clarifies when the remedy is inappropriate, that capitalization doesn’t remedy the negative tax harm, and that all exempt income should be subtracted from adjusted basis for the ceiling calculation.
Copyright 2022 Calvin H. Johnson.
All rights reserved.
A publication on an important subject, it is said, is not the final word but only the most recent draft, ripped by the printer from the vise-like clutches of the author.1 Sometimes an annex is then necessary to meet new objections or concede hasty error, which is what this article is about.
In my recent report on adjusting the interest ceiling, I argued that a combination of expensed investment costs and interest deduction yielded an indefensible negative tax, a subsidy better than no tax.2 Under free-market economics, God resides in the demand curve, and the negative tax subsidy costs money and violates the norm that capital must go to the investment with the highest real demand, reflected in pretax revenue. The negative tax warps investment into wasteful targets.
I proposed a remedy, short of taxing all economic income or a full move to a cash-flow consumption tax, but as immediately necessary as putting out a fire in the kitchen. The interest deduction must be limited to a ceiling of adjusted basis times the interest rate. Just as section 265, around since 1917, denies deduction of interest matched with tax-exempt municipal bond income, so the remedy needs to be generalized to take away an interest deduction stacked against income that is effectively tax exempt because of the value of upfront deduction of investments.
The core thesis is sound and simple, but around the edges I sometimes need to call off its application with patches; for instance, for the farmer who borrows for seed and planting in April and repays the loan at harvest. I need to explain why market capitalization of tax benefits won’t fix the problem, and I need to take exempt income out of adjusted basis when adjusted basis might include exempt amounts.
I. Borrowing for Expenses
My interest ceiling report defined an expense — consistent with sound and orthodox accounting — as an investment that happens to generate all its return by the end of the year and then expires in full. The report looked at “expenses” as just another intra-accounting-year detail — just one of those problems that arose because tax and nontax accounting and even cash flow diagrams treat events as year-end events. The report erroneously concluded that expenses were just like any other investments, and they had zero adjusted basis by year-end. With zero basis, there was no room under the ceiling to allow an interest deduction.
That conclusion is an error that doesn’t account for the outside world. Take a farmer who borrows $100 in April for seed and planting of a crop, and then has a harvest in October large enough to repay the loan, $10 interest on the loan, and $2 left to feed himself and his family. Assume a tax that takes away one-sixth of income, a middle-class tax rate. The farmer has zero basis from the $100 planting expense by year-end because seed and planting are deducted when incurred. So if there is no interest deduction, the farmer reports $12 taxable income and owes one-sixth of that, $2, as tax, leaving the farmer with nothing. The $2 profit should be taxable in a modest tax bracket, perhaps even a zero tax bracket, but interest must be deducted to get to that.
Upfront expensing of an investment is ordinarily so valuable that the tax does not reduce the pretax return. That isn’t what is happening to the farmer. The expense and the harvest come in the same tax year, so there is no early tax reimbursement. Within the year-end convention of accounting, the tax saved by expense and tax paid on harvest are simultaneous.
Moreover, the negative tax comes from a mismatch of debt added to upfront investing, and by the time of the tax-savings reimbursement, the debt has been paid off. Estimated taxes might sometimes be modeled as continuous tax accounting that isn’t settled at year-end, but estimated tax depends on estimates of net income, not unnetted expenses, so estimated taxes can plausibly be ignored. Now, for those crops planted late in one tax year and harvested in the next tax year, there is early reimbursement of upfront tax savings in year 1, which would be expected to yield no reduction of pretax returns.3 Not all farming can be exempted from the proposed interest ceiling.
When outlay and return happen within the same tax year, however, the tax reimbursement isn’t upfront, and the deduction of interest isn’t a mismatch. Interest must be allowed in cases like the farmer’s crop, on top of the interest-times-adjusted basis ceiling, as a kind of annex to the basic proposed ceiling.
In practice, the problem is contained.4 A farmer has adjusted basis in barn, tractor, and land, and that adjusted basis will raise the ceiling for the interest deduction, year after year. Land is capital, and crops are only annual income on capital, say 10 percent of the land value, so that the ceiling of interest multiplied by the cost of land will ordinarily be much larger than interest multiplied by cost of crops.
Many corporations now strip tangible assets off their balance sheets by having some other entity own the assets and paying that entity rent, but they can easily bring back tax ownership of the tangible assets if the proposed interest ceiling begins to bite. Bringing back ownership of assets would raise the ceiling measured by the total adjusted basis of all the corporation’s assets. Still, if it is possible to solve or mitigate the problem of borrowing for crops and like expenditures without too much clutter or opening up to abuse, we should do so.
A. Don’t Relitigate Post-Year Value
In theory, if an expense generated no revenue by year-end, it isn’t an investment with an upfront reimbursement of tax and shouldn’t be reached by the proposed interest ceiling. The question, however, whether expenses generate future income beyond year-end requires a prediction about a future beyond the tax year that hasn’t yet happened. For some expenses, like those for seed and planting, we can be confident that the expense expires by the harvest, but other expenditures conventionally categorized as expenses generate tax-exempt-equivalent revenues beyond the tax year.
Thus, start-up businesses usually expect that operating expenses will exceed revenue until the business attracts more customers: The new business may have built a better mousetrap, but it may take some time for the world to beat a path to its door. The existing tax and nontax accounting conventions get expensing decisions very wrong. Thus, the behemoths — Amazon, Apple, Google, and Microsoft — have made world-dominating investments worth more than $1 trillion per company, as validated by the economic judgment of the smart public market, but accounting doesn’t recognize the investment value of what the behemoths have built and treats their expenditures on intangibles all as just incurring worthless expired expenses, throwing money down a rat hole.5
The logic of the interest ceiling is that it is a secondary line of defense, preventing the harm of negative tax, but the behemoths and other producers of intangibles can have their tax-exempt income without the negative tax subsidy arising from an interest deduction as well. The behemoths have a modest adjusted basis in cash and cash equivalents, and their negative tax from interest mismatched with expensed intangible investments should stop. The ceiling on interest isn’t the place to relitigate the distinction between investments theoretically but not in fact capitalized and expenses expired by year-end. The point of the ceiling is to let the exemption equivalence go for investments expensed by ordinary tax accounting, but prevent the revenue loss from negative tax, and the departure from justifying investments by the real demand curve.
B. Loan-Based Annex?
Instead of separating expired expenditures from continuing-value investments, one strategy would be to create a supplemental rule allowing interest on the seed and planting loan to be deducted, without adjusted basis, on the ground that the loan will be repaid before the end of the tax year. A remedy based on the short-term nature of the loan looks like a patch, somewhat complicated and never a perfect fit to the problem, but plausibly that still is sufficient in ameliorating the problem.
The crop loan is repaid upon harvest in the same year. A farmer borrowing for seed and planting in April with repayment at harvest is not the intended target of the ceiling restriction: The targeted negative tax depends on a combination of upfront tax reimbursement of the input that allows lesser borrowing (or equivalently with gross-up, a greater investment for fixed borrowing). But for planting and harvest in the same year, the loan for seed and planting must be repaid before the time the tax reimbursement comes in. There is no negative tax from a mismatch of tax reimbursement and loan.
Now estimated tax can be modeled as a source of tax reimbursement, sort of, but estimated tax depends on the year’s profit ($2 in the simple hypothetical) not the gross seed-and-planting outlay ($100 in the hypothetical). It is the final tax payment that plausibly best marks the time when deductions save tax — so let’s ignore the estimated taxes here. If a loan is paid by the end of the year of the borrowing, there is no combination of reduced borrowing from reimbursed tax from investment that yields a result better than exemption. Thus, the interest on the paid-before-year-end loan would be deductible even for a taxpayer without an adjusted basis in any asset.
We should worry about loans that are payable before year-end on the face of the loan document, but that are always rolled over in fact. In retrospect, the rolled-over loans aren’t paid off by the end of the year, which we need to have proven under the alternative way of identifying that this isn’t the negative tax targeted by the ceiling. Still, if we insist that a loan repayment be identified by a reduction of total liabilities on the year-end balance sheet, that should cover the rollover problem. Loans are like water added to a swimming pool, supplying the capital (water) drawn out of the pool to pay any expenditure. Loans are never properly identified to any specific use drawing out of the (water or capital) pool. Thus, it is the overall reduction of liabilities by year-end that marks repayment of the loan. That test would cover the rollover loans. Sometimes liabilities are reduced for a corporation because of contributions to equity by shareholders; that equity would be adjusted basis for the proposed test, but reductions in liabilities by reason of equity contributions shouldn’t also justify an annex of interest deduction without regard to adjusted basis.
If a farmer has a bad harvest, the seed and planting loan won’t be paid at harvest. There must be a further exception to allow the interest to be deducted on loans extended beyond the year-end obligation to pay because of the borrowing taxpayer’s distress. Distress loans involuntarily extended by the creditor might be hard to identify, but if they are found they are worth exempting from the proposed interest-times-adjusted-basis ceiling.
In my judgment, finding an exemption for the seed and planting loan and the like is worth it. The complication won’t weigh down the proposal so much as to break the limbs of the tree.6 Also, clearly provably expired expenses like the seed for annual crops could be combined with an exemption for loans paid by year-end. But the complications in favor of the taxpayer shouldn’t justify defeating the necessary limitation on interest that yields a negative tax.
II. Capitalizing Drops in Return
The interest ceiling report argued that interest rates have never risen enough to take away the advantage of negative tax for the investing borrowing taxpayer. In theory, if the creditor in the same tax bracket paid tax and raised interest enough to cover that tax, the borrower would be left with no advantage on net from negative tax. That has never happened. Each creditor has its own tax rate on interest income and cannot shift the personal tax to borrowers.
Among other reasons, creditors cannot price-discriminate between investors with adjusted basis and those who have none; only the fair market value of the collateral counts to the creditor. Creditors and borrowers also arrange themselves into tax constituencies. Creditors are low-rate investors; high-rate taxpayers gravitate toward tax-advantaged zero-basis investments and away from fully ordinary-tax interest income. A creditor trying to raise interest charged to cover its tax would be undercut by a creditor who pays less or no tax.7
The interest ceiling report did not, however, discuss the possibility of capitalizing away the investor’s advantage on the investment side. Full capitalization would mean that the pretax return from zero-basis or low-basis investment would drop to fully offset the investor’s advantage from negative tax.
That drop in return rate on tax favored investments would be the fruition of harm from the negative tax. The decrease in pretax return is equivalent to the increase in the capital price of low-basis investment. The added price means that capital has been wasted in investments that cannot produce enough from the real pretax demand curve to progress into the Valhalla of no tax distortion based on real demand. With capitalization, bad investments drive out those that better meet real demand.
Full capitalization is unlikely. There are too many competing tax-exempt investments, and they swamp the possibility of capitalization. Under today’s law, tax-exempt returns are generated by movies, oil drilling, ore exploration, intangible investments, municipal bonds, owner-occupied housing, qualified plans, and step-up in basis at death. If return rates drop on everything else, the capital appreciation on property to be held until death will increase as refugees from other investments flock into capital gains, and that will increase the returns from capital gains held until death, whereas capitalization shifting away a tax advantage would need to decrease returns.8 Far from being capitalized away, the drop in returns elsewhere will enhance the value of property that can be held until death. The creativity intensity of tax planners — who are continually inventing new ways to avoid tax — confirms the assumption that there is an unlimited supply of tax avoidance strategies and that we certainly cannot limit the supply in a way that would drive up the price enough to remove the borrowing advantage enjoyed by investors that have no adjusted basis.
The most likely result is that tax-advantaged investments will rise in price some because of the tax advantage — but not enough to offset all of the tax advantage that the investor can get from the negative tax. The result will be a mix of both inequity and unfairness: not as much inefficiency as full capitalization would cause, nor as a much inequity as no capitalization would generate, but a combination of both inequity and inefficiency.
III. Other Exemption Income
The interest ceiling report argued that the section 1014 step-up in basis should be taken out of adjusted basis to compute the interest ceiling. The origin of section 1014 was the idea that capital, originally the castle and manor, could pass to the heirs without income tax, but that rationale didn’t justify the added harm of negative tax. In calculating the adjusted basis for the proposed interest ceiling, only the predecessor’s adjusted basis would carry over. If the original cost or adjusted basis of the inherited property is unknown, that would imply the basis should be zero because the heirs get the property as a windfall without cost to themselves. It is difficult to see how the heirs could be said to have a detrimental cost for inherited property if they didn’t even know what the cost was.
The original report also should have noted that every other kind of exempt income should be subtracted from adjusted basis for the ceiling calculation. The original model for the prevent-negative-tax remedy is section 265, which since 1917 has prevented the deduction on debt to buy tax-exempt municipal bonds. Cash from tax-exempt bonds is itself basis and gives basis to property purchased with the cash, but it cannot be part of adjusted basis for the interest ceiling calculation. So similarly cash from all tax-exempt sources must be taken out of adjusted basis for the interest ceiling calculation.9
Borrowed cash or cash contributed to partnership or corporate equity is tax exempt, but it does increase basis. By presumption, the cash was from taxable sources of the creditor or equity contributor, and that basis appropriately carries over to the investing taxpayer. Seller-provided debt, however, does not have a basis in the creditor’s hands, as long as section 453 installment sale treatment is available, so that seller-provided debt shouldn’t be included in the adjusted basis for the proposed ceiling.
IV. Summary
The interest ceiling report argued that we need to end the negative tax subsidy arising from a mismatch in our treatment of debt and expensing investments by limiting the deduction of interest to a taxpayer’s adjusted basis in all assets times the interest rate. This article supplements that argument by saying:
Loans like those for planting and seeds that are repaid at harvest by year-end are not the target of the ceiling. The problem is limited because basis from any source will justify the interest deduction, but interest on loans paid by year-end would properly be deducted without regard to the basic limitation of interest times adjusted basis ceiling.
The negative tax might in theory yield a reduced yield, and a higher price for tax-advantaged investments. That is a waste of capital, however, because it distorts investment away from real demand, identified by pretax revenue. Capitalization is likely to be partial, yielding some perhaps small reduction in the taxpayer-investor’s inequity from the negative tax and also some inefficiency.
Exempt income of all kinds must reduce the adjusted basis in the calculation of the ceiling.
All three of those points were best addressed in the original interest ceiling report. Clarification of the points here, however, strengthens rather than undercuts the original report.
FOOTNOTES
1 I didn’t originate this idea, but I cannot now remember nor find who said it.
2 Calvin H. Johnson, “Interest Ceiling Must Be Adjusted Basis Times Interest Rate,” Tax Notes Federal, Sept. 26, 2022, p. 1987.
3 In the early 1980’s there was a syndicated shelter called the Mexican herbal rollover, in which the crop was planted at the end of the calendar year (when the New Jersey dentists knew how much money they needed to shelter from tax) and harvested in spring of the next year. The rollover could be repeated year after year. The “herbal” in the name carried a joke that the herb was marijuana, not something that Congress at the time was wanting to subsidize.
4 I am grateful to Stephen Shay for this argument.
5 Johnson, “A Fair Income Tax on the Trillion-Dollar Behemoths,” Tax Notes Federal, May 24, 2021, p. 1199, argues for capitalizing investments in intangibles as justified by the public market for stock.
6 The image of tax complications breaking the tax system like snow that breaks a tree limb is Michael Graetz’s.
7 Johnson, supra note 2, at 1989-1990.
8 Johnson, “Inefficiency Does Not Drive Out Inequity: Market Equilibrium & Tax Shelters,” Tax Notes, Apr. 15, 1996, p. 377.
9 Tax-exempt income increases the basis of the partner’s partnership interest, so that the cash from exempt sources can be distributed from entity to owner without tax to the partner. Section 705(a)(1)(B). No similar adjustment would be made for the interest ceiling.
END FOOTNOTES