Jodie Beggs of Economists Do It With Models posed the following via twitter:
Question- how is it that ppl are advised against variable rate mortgages but pushed into variable rate student loans?
I never turn down the chance to lecture condescendingly to a Harvard trained economist, so let me see if I can clear this up for Jodie by arguing why it might make sense for an individual to have a fixed rate mortgage but variable rate student loans.
First note the similarities that lead one to ask this question: a college education and a home are sort of assets. Well, a home is literally an asset, and an education is kind of like an asset, but they both share the features of an asset that they are a long-lived investment that pays investment returns. In addition -and central to this issue- purchases of both are typically financed by debt. A third important similarity is that these are typically the large components of household debt, and two of the most important investments households make.
Given these similarities, it is natural to assume that it would be optimal for households to choose similar interest rates. But it is the nature of the returns these investments produce that explains why an individual might choose different interest rates. For a house, the investment return produed is a flow of housing services. Think of it as a factory that produces a months worth of housing each month, which the individual owning it consumes. Education, in contrast produces higher human capital which translates (usually) into higher value of output that their labor creates. The difference is that inviduals sell their labor output, whereas they consume the flow of housing services.
This matters because if inflation -and thus nominal interest rates- go up then the cost of debt service will go up for both housing and education debt if the interest is variable. In addition, inflation should also drive up the price of the output produced by the assets and the nominal value of the assets themselves: house prices and the value of the flow of housing services -equal, in theory, to rent- will increase, and so will wages. Because households are selling their labor, their income rises with their costs of debt. But households are consuming their housing services, their debt service goes up without an offsetting increase in income. For this reason, it is more risky to use variable rate interest on housing debt than on education debt.
13 comments
Comments feed for this article
Sunday ~ April 29th, 2012 at 10:53 am
George
Im afraid i dont see the distinction here. If inflation goes up then the real value of housing services stays the same but the nominal value of said services increases at the same rate as the nominal value of the person’s labor. In either case, fixed rate loans are better for the borrower if inflation is higher than expected and worse if inflation is lower than expected.
Sunday ~ April 29th, 2012 at 12:36 pm
Kyle
@George I think the issue comes down to liquidity and/or transaction costs. If, like most people, you live in your house, then it can be somewhat painful to take the gains from the change in nominal value of your house (moving, selling the house,…). Since most people can adjust their wages yearly (ie negotiate with their boss or whatever) it is far easier to adjust to new interest rates.
I think that in the case of an investor who was renting a house this point would be somewhat less important…or perhaps I have also missed the point
Sunday ~ April 29th, 2012 at 12:39 pm
Marginally Thoughtful
I believe the idea is that a higher nominal value of housing service doesn’t matter for someone who plans to stay in her home.
Then someone who buys a house with plans to sell it down the road shouldn’t mind the variable rate loan? Or someone (straying further into hypothetical territory) who plans to lease the house?
Sunday ~ April 29th, 2012 at 2:26 pm
Matt Rognlie
Nice post.
My wife is going to law school next year, and I’ve been thinking a fair amount about loan terms. The core issue to me is that education debt is much, much more expensive than housing debt; you can get a 30-year mortgage at 3.95%, while most fixed-rate education loans at in the neighborhood of 7-9%. (The subsidized Stafford is temporarily cheaper than that, but it’s up in the air politically and it doesn’t cover a huge fraction of the cost anyway.)
In part this is probably due to federal support for 30-year mortgages. But I think it’s mainly because in practice, human capital is much less effective as collateral than physical capital. This is true even after the housing crash — or perhaps because of it, since banks know that house prices can’t possibly fall much more. Sure, you can’t discharge student loan debt in bankruptcy, but it’s also a pain in the ass for banks to collect on it; on the other hand, a bank can be pretty sure that homeowners with a substantial amount of equity in their homes simply won’t default.
I know that we have an extremely low probability of default, so that the credit risk premium we’ll be paying with student loan debt will never be compensated by any benefits from credit delinquency. It’s natural, therefore, for me to want to prepay the law school debt as soon as possible to avoid wasting money on debt service; within just a couple years, even, after we graduate. This means that if we choose fixed rate debt, we’ll be hammered unfairly by the yield curve, paying interest at a fixed rate that reflects long-term yields even when we’ll only hold the debt during a short period when the average fed funds rate will be something like 1-2%. Variable rate debt gets us closer to a more reasonable cost of capital. Also, in the short term, interest rates are highly positively correlated with strong economic performance (the Fed will only raise rates if the economy is doing decently), giving variable interest rates an extra risk-hedging benefit. (This is related to the point in your post, though I think the effect is even more powerful in our idiosyncratic current situation. Better be careful to have the loan indexed to prime rather than LIBOR, though, since pending another financial crisis I wouldn’t say that LIBOR is positively correlated with economic performance.)
Of course, this is a lot less general than your (much broader) economic story: it relies on (1) near-metaphysical certainty that we can prepay the debt almost immediately and (2) a strongly upward-sloping yield curve.
One interesting point here is that interest rates serve two functions: (A) governing intertemporal prices and (B) determining the amount of cash you must pay to service the debt at any given time. I think most people avoid variable-rate debt* because they want to avoid cash-flow shocks from (B), but in principle there’s no reason why functions (A) and (B) should even be connected with each other. You can imagine being required to pay 7% of your principal every year, and then having “interest” capitalized into the principal amount next year, so that a shock to the interest rate doesn’t imply an immediate cash-flow shock. This makes a lot more theoretical sense.
*Well, actually, most people avoid variable-rate debt for housing because it’s an absurdly bad deal compared with fixed-rate debt; in fact, it has been for many years, except for people who are either fooled by low rates in the short-term or smart enough to take advantage of the “teaser” rate combined with a prepayment option. Even for student loan debt, variable-rate offerings are pretty bad unless you’re planning to hold the debt for a very short time. But I can’t see why this should be true in general, rather than being an idiosyncratic feature of U.S. household credit.
Monday ~ April 30th, 2012 at 2:44 pm
Craig
Also consider that you can’t really insure lifetime earning potential if a newly trained lawyer decides to go collect seashells on Kauai and live in a tent; houses make it easier for a lender to avoid losses. A lender can’t repossess and sell your law degree at auction, and the degree doesn’t enter your estate if you die. So for lots of reasons, we’d expect education money to be more expensive than housing money.
Sunday ~ April 29th, 2012 at 4:37 pm
Benny Lava
Adam, I think you missed the point. If students could get a fixed rate for their loans they would, and it would involve less risk than a variable rate. But alas, they cannot.
You may not know this (I relish the opportunity to lecture economists as well), but before the 1930s all mortgages were variable. It was the FHA that created the 20% down 30 year fixed rate plain vanilla mortgage during the New Deal. Before that mortgages required a higher down payment (usually at least 40%), a shorter amortization term (usually less than 20 years) and a variable interest rate. America has a housing policy which pushes people into owning single family residences, through tax breaks and through market manipulation via Fannie, Freddie, and the FHA, and it has been that way for the past 80 years.
This is exactly why Fannie and Freddie are in limbo, and why neither the Democrats nor the Republicans could even float legislation to do something about them. Because they both know that if they killed the GSEs then things would go back to how they were before. And that would be too politically unsavory.
Jodie is absolutely correct here that student loans have bad terms. Higher interest, variable rates, and the inability to discharge them stack the deck against student loans. The promise of greater lifetime earnings has, for many young graduates, failed to produce. This is all because of our public policy. Economist should think more deeply about this, and take this issue more seriously.
Young people should be advised against taking student loan debt. If this means going to a different college or taking a few years off between high school and college to work, save money, and really think about long term careers goals they would probably be better served by it.
Monday ~ April 30th, 2012 at 2:51 pm
Craig
I do advise against taking a lot of student debt, but I disagree that student loans have “bad terms.” They have hard terms. But I don’t regret a penny of the debt I took on (for a STEM degree at a state school), or my wife did (for a top-tier MBA). These were good choices; they put us in careers we are well-suited for, and boosted our lifetime earnings enormously.
The problem is that this debt is too-readily available to naive consumers: kids who have been told for eighteen years that life is all about possibilities and following your dreams, practicalities be damned. College becomes another four, or six, or ten, years of prolonged adolescence, when it ought to be the start of adulthood.
(Dreams? My passion is Shakespeare; my wife’s is animal rescue. Both of these fill our evenings and weekends with considerable joy. But first we work, and then we play.)
Monday ~ April 30th, 2012 at 7:20 pm
Benny Lava
No, the problem is that college is more expensive now than ever before and kids are told that going to college is the best way for them to succeed. They are getting bad advice from everyone on this. Even college professors like Stanley Fish make mendacious arguments in the NY Times that college is more affordable than ever.
Based on your reply I’ll bet you are old and have no idea just how expensive college got.
Tuesday ~ May 1st, 2012 at 9:13 am
Craig
Conceivably, thirty-seven is old; I’m probably not the right person to judge. College _is_ much more expensive now than it was in the mid-nineties (my undergraduate) or the late nineties (my wife’s master’s). I know that; I pay a lot of attention to it, because I’m trying to work out what college is going to cost those near and dear to me in another ten years. You should see the planning summits we have.
But, again, the question is not really how much college costs in dollar terms. The question is whether it is worth the expense. And the answer depends entirely on where you go and what you study. College is still the deal of a lifetime if you go to a public school for a marketable degree, working your way through so as to minimize your debt. It’s a mug’s game if you go to Harvard to take a BA in anthropology. That’s for rich people, not those of us who have to make our way in this world.
I will agree without reservation that college is not for everyone, and much harm has been done by flogging marginal students into for-profit institutions from which they will plainly never graduate. That’s vile. And if you’ve learned nothing else, you’ve at least learned not to trust Stanley Fish farther than you can throw him. He did write a pretty good book on Paradise Lost, once upon a time, before he became an old fool and shill for the establishment.
Monday ~ April 30th, 2012 at 3:07 am
FT Alphaville » Further reading
[...] – The differences between housing and education debt. [...]
Monday ~ April 30th, 2012 at 11:46 am
Craig
I’m not following you here. The household income of mortgage holders goes up becasue they’re earning more imputed rent. Yes, they’re living in exactly the same house. But they’re working the same forty hours (or what have you) in exchange for a paycheck that buys roughly the same basket of stuff. You assume the borrowers are working for a living, and analyze the student debt, but not the home debt, through this lens. I don’t see how that works.
Just add a zero to every dollar amount: nothing really changes. One third of your paycheck pays for one month of your mortgage, and a tenth of it services your student debt (or whatever). Nothing changes the fact that fixed-rate debt is a hedge against inflation, and variable-rate debt is a bet that rates won’t rise.
Tuesday ~ May 1st, 2012 at 8:09 am
Campaign Notes – Tyler’s AM Reads – May 1, 2012 « Blog of Rivals
[...] Adam Ozimek comments on variable rate student loans versus ARMs. Share this:EmailTwitterFacebookLike this:LikeBe the first to like this post. ▶ No Responses /* 0) { jQuery('#comments').show('', change_location()); jQuery('#showcomments a .closed').css('display', 'none'); jQuery('#showcomments a .open').css('display', 'inline'); return true; } else { jQuery('#comments').hide(''); jQuery('#showcomments a .closed').css('display', 'inline'); jQuery('#showcomments a .open').css('display', 'none'); return false; } } jQuery('#showcomments a').click(function(){ if(jQuery('#comments').css('display') == 'none') { self.location.href = '#comments'; check_location(); } else { check_location('hide'); } }); function change_location() { self.location.href = '#comments'; } }); /* ]]> */ [...]
Wednesday ~ May 2nd, 2012 at 9:10 pm
richardhserlin
Still, often better to have fixed rate student debt too. When interest rates go up, nominal incomes for the college educated tend to go up too, but there are lots of exceptions. By having fixed rate student loan debt you insure against the scenario that you end up as an exception, that interest rates soar, but your particular field or job doesn’t, or your spouse gets laid off when the interest rates soar, etc.
It’s a risk, and if another party can pool and absorb that risk a lot better than you can, like the US government, then it can make a lot of sense for them to take on that risk and not you the individual.