By: Sam Brunson
Friday, the New York Times‘s DealBook section had an article about law school debt. (H/t Paul Caron.) It focused on John Acosta, a recent Valparaiso graduate who is starting a defense and family law practice.
Although he’s done well for himself so far—top third of his class, passed the Bar Exam on his first attempt, and successfully convinced a former prosecutor to join him—he has a significant problem: debt. From the article:
Yet in financial terms, there is almost no way for Mr. Acosta to climb out of the crater he dug for himself in law school, when he borrowed over $200,000. The government will eventually forgive the loan — in 25 years — if he’s unable to repay it, as is likely on his small-town lawyer’s salary. But the Internal Revenue Service will treat the forgiven amount as income, leaving him what could easily be a $70,000 tax bill on the eve of retirement, and possibly much higher. [Emphasis added]
Up to $70,000 in taxes, or maybe more? Could that be right? And, if so, what’s up with that?
First things first: if the government forgives his loan in 25 years, he must have federal student loans and be taking advantage of the Revised Pay as You Earn (“REPAYE”) repayment plan. Basically, under that plan, borrowers cap their monthly payments at 10 percent of their monthly “discretionary income.” And, after 25 years, the government forgives whatever hasn’t been repaid.
But debt forgiveness comes with a cost. Specifically, in most cases, discharge of indebtedness is treated as income to the debtor, and the debtor has to pay taxes on the amount discharged. And under certain circumstances, the discharge can be significant.
Of course, it is impossible to know Mr. Acosta’s economic future, but using some assumptions, we can at least get an idea of it.
For these purposes, I’m going to assume that Mr. Acosta’s income remains steady over the next 25 years, that the tax rates and brackets remain steady over the next 25 years,[fn1] and that his familial situation also remains steady. [Update #2: It turns out that the DoE’s Repayment Estimator assumes a 5% growth in a borrower’s discretionary income plus an increase in the poverty level linked to inflation.]
To figure out the amount that will be forgiven, I looked at the income of Indiana attorneys. In May 2015, their mean annual income was $110,240, the 25th percentile was $60,750, and the 10th percentile was $44,590.[fn2]
To figure out how much debt would be forgiven, I used this Repayment Estimator from the DOE. I entered a $200,000 loan at 5.84% interest (the current rate for direct unsubsidized loans for grad students).
I ran the calculation at $44,590, at $60,750, and at $110,240, both for a single Mr. Acosta and a married Mr. Acosta with two children. For the married version, I assumed his spouse had no income and no loans. (I mean yes, the article says he’s engaged to another lawyer who has her own debt and is starting her own practice, but this post was already taking too long.)
So what happens to Mr. Acosta under these assumptions? If he’s single, after 25 years, the government will forgive:
|Income||Amount Forgiven, single||Amount Forgiven, married|
(I’m a little confused here, frankly. Clearly, the Repayment Estimator believes that any forgiven interest will be capitalized into principal. Everything I can find says that under REPAYE, interest is only capitalized if you leave the REPAYE program, and I can’t find a definition of “leave,” so I don’t know if that includes being forgiven after 25 years. But, because this is outside my area, and I don’t have the additional hours I’d need to run down the answer, I’m going to use the numbers the Estimator spit out for the rest of this post.)
Mr. Acosta will have to pay taxes on the amount forgiven in the year that it’s forgiven. Of course, the actual amount of taxes he has to pay will depend on the tax rates and brackets in 25 years, but again, I’m going to assume they’re the same as today’s. I’m also going to assume that in the year of forgiveness he earns the same amount of money as I’m assuming he earns today. The question is, by how much will the REPAYE forgiveness increase his tax bill?
To calculate that, I figured out his tax liability would be (treating his income as “taxable income” and ignoring any potential deductions and personal exemptions he might have) without including the forgiven debt, and then calculated it including the forgiven debt. I subtracted the first number from the second, and that gave me the increase in taxes he faces as a result of the debt forgiveness.
And what is that increase?
|Income||Additional Tax, single||Additional Tax, married|
That’s kind of shocking: assuming that the Estimator works right, if Mr. Acosta is in the 10th or 25th percentile, when he reaches year 25, his tax bill will be higher than his annual income as a result of the debt forgiveness.
What should we do about this? I don’t really know. We could say that student loan debt forgiveness should be treated differently than other debt forgiveness (we already do in certain circumstances). But I’m not sure that’s a good answer—why should student loan debt forgiveness be treated differently than mortgage loan or auto loan debt forgiveness, or any other kind of debt forgiveness?
Maybe we should make student loan debt dischargeable in bankruptcy. Debt discharged in bankruptcy doesn’t have to be included in gross income.
Maybe we address the problem at the other side, somehow addressing the skyrocketing college tuitions so that nobody incurs $200,000 of debt.
I really don’t have an answer.[fn3] But even though REPAYE and other income-based repayment options strike me as better than not having such options, they still come with a significant cost at the end, a cost I’m not sure we want students to have to bear.
Update: as Michael Simkovic points out elsewhere, there may be other problems with the article. The Valparaiso grad was apparently in the PAYE, not the REPAYE, program, which offers 20-year debt forgiveness, and functions differently. (I assumed REPAYE because of the 25-year forgiveness cited in the article.) What are the tax consequences of PAYE forgiveness? I don’t know—I’d have to rerun all of the numbers, which maybe I’ll do later.
Also, there is an exemption to discharge of indebtedness if the debtor is insolvent. It’s not an entire exception, though, and it affects tax attributes in various ways that I really don’t want to get into. I was laying out the worst-case scenario, which is kind of a bad scenario.
[fn1] Or, at least, that both match inflation precisely, so that his income and the marginal brackets grow in proportion to one another. And I realize that there’s at least a chance his income growth will outstrip inflation, or vice versa, but I don’t have any way to quantify that, even if it were a sure thing.
[fn2] The data’s available on this Excel spreadsheet from the BLS.