Debt vs. Equity Finance and Interest Deductibility
In reaction to my Congressional testimony–where I said I didn’t think the federal government should tax interest payments, if they are retaining a corporate income tax–Roger Barris objected. He claimed that I was ignoring the lessons of Finance 101, and in particular the famous Modigliani-Miller theorem.
I must confess that his initial stance annoyed me; after all, I stated the “standard” position in my testimony, and then went on to disagree with it. So simply repeating back to me the standard position doesn’t seem to advance the ball much.
However, in email follow-up Barris and I had a nice conversation; apparently Twitter is not the best place to settle disputes.
Anyway, this may be a series of posts, but check out the following to see where I’m coming from. Because not just Barris, but some other people I respect have also told me that people distinguish operating costs from financing costs, I am open to the possibility that I am missing something here. However, as of right now, I still think I’m right, and that I’m adequately incorporating the perspective of the people who are arguing with me.
=======================================
Scenario A
Joe puts in $100,000 of his own money. He sells half of the company to Sally for her $100,000.
The company spends $100,000 on materials and $100,000 hiring a worker. At the end of the year they sell the resulting product for $220,000. The materials and wages are deductible expenses. The company has net income of $20,000.
Without taxes, that would be a return of 10%. However there is a corporate income tax of 35%. So the corporation pays $7,000 in tax. Joe and Sally each get $6,500 in dividend payments, which are taxed as personal dividend income.
The $200,000 is available to repeat next year.
Scenario B
Joe puts in $100,000 of his own money. He issues $100,000 worth of (perpetual) bonds to Sally, yielding 6.5%.
The company spends $100,000 on materials and $100,000 hiring a worker. At the end of the year they sell the resulting product for $220,000. Out of this they pay $6,500 to Sally as interest on her bonds. The materials, wages, and interest are deductible expenses. The company has net income of $13,500.
Without taxes, that would be a return of 13.5% to Joe. However there is a corporate income tax of 35%. So the corporation pays $4,725 in tax. Sally earns $6,500 in interest payment, which is taxed at personal interest income rate. (She has the same rate of return as in Scenario A, but is happier now because she gets that same expected rate of return but with lower risk.)
However, now Joe earns $8,775 in dividend income, compared to the $6,500 he earned in Scenario A. The extra $2,275 is the amount of tax the company avoided in Scenario B relative to Scenario A. (The company paid $7,000 in corporate income tax in Scenario A but only $4,725 in B.) Joe is earning a higher rate of return in this scenario, but his risk is higher. Depending on his preferences, he might not wanted to have taken on debt-to-equity of 100%, in which case he could’ve borrowed a smaller amount from Sally and let her buy in with equity for the rest.
The $200,000 is available to repeat next year.
=======================================
OK let’s stop to catch our breath. In light of the above two scenarios, it is standard for people to argue, “See? There are two ways for a company to obtain financing: it can pay dividends or interest. To put equity financing on an even footing with debt financing, the government should not allow corporations to write off interest payments on bonds.”
Now let’s consider two more scenarios:
=======================================
Scenario C
Joe puts in $1,000,000 of his own money. He tells Sally that if she is willing to work full-time running the business, he will make her 40% owner.
The company spends $1,000,000 on materials. Over the course of the year it earns $1,100,000 in gross revenues. Without taxes, Joe would get $60,000 in dividends and Sally would get $40,000.
However, what they report to the government is net corporate income of $100,000. So they pay $35,000 in tax, and of the remaining $65,000, Joe gets $39,000 while Sally gets $26,000, and each then pays personal income tax at the dividend rate on this. (But this is 0% for Sally, since her total income is so low.)
Scenario D
Joe puts in $1,000,000 of his own money. He hires Sally to work full-time at the business, for a salary of $30,000, but she will only be paid it at the end of the year.
The company spends $1,000,000 on materials. Over the course of the year it earns $1,100,000 in gross revenues. The company pays Sally $30,000 in salary. Sally pays 15 percent income tax on this labor income, which would leave her with $25,500 after-tax, but actually when you figure in the standard deduction etc. she ends up with more after-tax income than the $26,000 she pocketed in Scenario C. Thus Sally prefers this scenario, especially since her payment is contractually locked in, rather than being the residual of the company’s profits.
Because Sally’s wage payment is a deductible business expense, the company’s pre-tax net income is $70,000, so it pays the IRS $24,500. Thus the company saves ($35,000 – $24,500) = $10,500 on corporate income tax by going this route with Sally’s compensation. Joe is the sole owner and pays himself a dividend of $45,500. That’s $6,500 more than he earned in Scenario C. Now, it’s riskier, so maybe Joe wouldn’t want to fully exclude Sally’s equity. But it shows the possibilities.
=======================================
Now notice we can make an analogous analysis for Scenarios C and D, as we did for A and B. To wit: “There are two ways a company can obtain labor services: it can pay dividends or wages. To put equity labor on an even footing with wage labor, the government should not allow corporations to write off wage payments on labor contracts.”
Does that sound right? Or would that screw everything up even more?
Bob, I think the further objection to the uneven tax situation between debt and equity is that it makes it even easier for short sighted CEOs to run up the stock prices with borrowed money (share buybacks financed with bonds) in order to increase thier stock option value.
It seems like the combination of these to elements, financial incentives to favor debt, and management’s incentives to not dilute thier own stock or options, are a large part in creating the distortions we are now seeing in Stock valuations.
I take your point that the “fix” should not necessarily be taxing both, I think it should be neither, but having them dissimilar does create distortions.
Also, sorry if this is taking the conversation away from where you were trying to steer it.
I think there’s a problem with the analogy. Scenarios A and B can run in perpetuity. But, in scenario C the issue of when Sally’s ownership interest vests has been skipped over. To make C and D equivalent except for tax treatment Sally’s ownership can never fully vest and she must keep working for the company in order to get paid dividends. A tax court is likely to say that deal is really a labor contract and needs to be treated as one for tax purposes. If Sally’s ownership interest does vest at some point then she can take off for the beach and collect dividends while working one her tan. Now Joe has to pay dividends to his non-working partner and he still needs to hire and compensate an employee to run the business. So his dividend payments for labor services may turn out more costly to the business in the long run even though there appears to be a short run tax advantage.
Good point Capt Parker, but there are definitely cases where people are expected to contribute a perpetual flow of labor in exchange for their ownership. E.g. imagine two partners opening a pizza place together. They might each kick in some money, sure, but they both know they have to each work their tails off, at least in the beginning.
You know, It suddenly occurs to me that scenario C sounds a little bit like Sally is receiving carried interest. http://www.taxpolicycenter.org/briefing-book/what-carried-interest-and-how-should-it-be-taxed
Yes and here’s the relevant excerpt from that:
“But others believe that the general partners are more like entrepreneurs who start a new business and may, under current law, treat part of their return as capital— not as wage and salary income—for their contribution of “sweat equity.” Our tax system largely accommodates this conversion of labor income to capital because it cannot measure and time the contribution of the “sweat equity.””
I am not sure the point that Bob’s scenarios C and D are trying to make, unless it is to demonstrate that the US tax code contains more than one type of irrationality, a point which I am readily willing to concede. (Although I think that the code may be more logical than Bob thinks, as I note below.)
I think that Bob would be the first to accept that the correct analysis of scenario C is that Sally has earned a salary of $35,000 per year (in perpetuity, or at least a long time, as Capt. J notes). She has then chosen to reinvest this implicit salary into 40% of the business, for which, qua shareholder, she is entitled to dividend income. The correct economic treatment is that the implicit salary should be a deductible expense for the company and taxable income to Sally. The implicit equity investment should be treated as any other equity investment.
What both of these examples show is that the tax code does a very bad job of handling implicit costs (the implicit salary cost to Sally and the implicit cost of capital to an equity provider). However, unlike Bob’s scenario C, which does not have a material impact on the overall economy (and cannot since the opportunities to substitute sweat equity for explicit labor costs are inherently limited), the tax preference to debt does have a very material impact on the overall economy. It subsidizes higher leverage, which greatly increases macroeconomic risks. It also raises political economy problems since, in an attempt to offset some of the tax advantage coming from a single layer of taxation on debt, personal tax rates on equity returns (dividends and capital gain) are relatively low — thereby creating a plausible argument for unfairness (which even someone like Warren Buffett, who does or certainly should know better, falls into when he compares the personal tax rate he pays on his investment income to the personal tax rate on his secretary’s earnings, while ignoring that his investment income has already been taxed at the corporate level).
Bob and I are both at a disadvantage because neither of us thinks that there should be a corporate income tax (although probably for different reasons). With no corporate taxes, this debate would drop away since debt and equity would be treated the same, and capital structures would be determined entirely by risk and control preferences. I would also predict that, in this world, there would be a lot less leverage.
So, this is very much a debate about the theory of second best. Given that we have a corporate income tax, should debt and equity be treated differently? An objective of tax policy should be to avoid obvious distortionary incentives, particularly ones that have material impacts. I think that the unequal treatment of debt and equity fails this test spectacularly.
The next question then becomes, how should we treat them equally? Either the cost of both should be deductible or neither should be deductible. It is very difficult to make the full cost of equity capital deductible. Dividends can easily be made deductible (and, in fact, are in many tax systems around the world, although sometimes only partially), but the capital gain component of expected equity returns is hard to make deductible at the corporate level. Conversely, it is relatively easy to make neither deductible by eliminating the tax deduction for interest.
(Although I need to think about this further, I am pretty sure that it doesn’t matter which of these alternative routes to equality we choose. Since what really matters to investors (the suppliers of capital) is their after-tax, risk-adjusted returns on their investments and since what really matters to companies (the demanders of capital) is the after-tax cost and risk of capital, then it shouldn’t matter how we get there. If you radically changed the tax code, then prices and capital structures would have to adjust but, in equilibrium (the word that Bob hates), the after-tax returns to debt and equity would return to the status quo ante. Therefore, the choice of both or neither needs to be made on the basis of practicality and political feasibility.)
In summary, the tax preference granted debt has a very large distortionary impact on the economy. The only practical way to eliminate this distortion is by cancelling the deductibility of debt. The fact that there are other, relatively minor, distortions in the tax code does not invalidate this argument.
(Although I am no tax lawyer, I don’t think that the tax code is a crazy as Bob’s scenario C implies — as Capt. J Parker also points out. If it were persistent, the IRS would certainly look through and recharacterize the arrangement between Joe and Sally in scenario C along the lines of my second paragraph. Even if the IRS didn’t do this, there is a good chance that the granting of shares to Sally in scenario C would be considered a gift and subject to the gift tax. If the arrangement made its way through both of these hurdles, then it would certainly be caught with a capital gain liability in the event that Sally sold her shares, in which she would have no tax basis.)