Category Archives: Education

Higher Education Financing

Have been meaning to respond to Obama’s proposals for containing costs in higher education; the issues are incredibly complex, as there are different mechanisms driving cost pressures in the community college market, public/private non-profit university markets, and for-profit university markets. Unfortunately, I kind of missed the boat on responding, and lots of people have already said everything that needs to be said; Dylan Matthews at Wonkblog has a wonderful summary of the various drivers of cost and reasons that it’s not straightforward to address them.

Continue reading

Making College Affordable

Patricia Murphy from The Daily Beast has an article out (a few days ago) that builds on a theme I tried to emphasize in my post about the (flawed) article from the Wall Street Journal that criticized current federal student loan policies for being too generous. The Daily Beast article makes it very clear that the driving force behind the increasing cost of college in America is spending by the colleges themselves, and that this spending is rather impervious to spending constraints on the part of the students themselves:

Continue reading

Macro and Student Loans

I think the last several posts made a reasonable case that current interest rates on federal student loans are on the high side both in terms of net revenue for the government and relative to previous historical episodes with a similar interest rate climate. This post is intended to make the case that there are other macroeconomic benefits to lowering the interest rates on federal student loans that have to do with the current macroeconomic state.

I’ll take it as a given that our current economy is operating under potential due to a lack of demand, and that in the absence of government stimulus (which is not forthcoming), we need to look at alternative methods for generating private consumption again.

Indirect measures that ease private sector deleveraging are important tools to counteract the lack of fiscal stimulus over the past 4 years—the Federal Reserve has been important here, engaging in all sorts of unconventional maneuvers to bring down long-term interest rates. (Correspondingly, we are finally seeing the housing market begin to pick up after its 6 year slump). Unfortunately, even as housing debt falls, non-housing debt has been ticking up and the main factor in that has been an expansion of student loan debt (now ~ $1 Trillion in outstanding student loans across the US).

There are multiple factors driving the expansion of student debt–broadly speaking, they can be separated into structural factors and cyclical factors. Structural factors include the fact that, in general, education spending over time has increased because education is a labor-intensive industry and the cost of (highly educated) labor like professors and administrators has risen over the past decade. Additionally, colleges continue to spend more and more money on non-educational pursuits as the importance of adding amenities (quality housing, gyms, sports teams, etc) to the college experience has grown. These factors would be driving up tuition regardless of the state of the economy. However, cyclical factors resulting from the financial crisis have also contributed to an increase in student debt: stagnating (and falling) incomes during the financial crisis diminished the average family’s ability to afford college; and state cutbacks to education spending contributed to tuition increases without corresponding increases in grant aid from states or the federal government. Taken together, the amount of outstanding student loan debt has roughly doubled since 2005.

In and of itself, elevated student debt isn’t necessarily the worst thing in the world—obviously a world in which all students could receive grant funding to attend college would be better, but I would rather see more people get degrees with student debt than more people simply enter the workforce straight out of high school. The key caveats, here, are that

1) students need to earn a degree, and

2) students need to be able to find employment commensurate with their degree level after leaving college.

The absolute worst case scenario would be a student running up a significant amount of debt at a low-quality school, failing to get a degree, and then failing to find employment in the labor force (or working at the equivalent of a high-school degree job).

Unfortunately, a significant number  [page 12]  of students fail to meet the above criteria. 30% of borrowers that drop out of for-profit, less than 4-year institutions default on their loans; this is compared to the ~3.5% default rate average of all borrowers that graduate from a post secondary institution.

Furthermore, 12% of outstanding student loans are now more than 90 days delinquent, and though students that attend private, for-profit institutions make up only 10% of all post-secondary students, they account for half of all student loan defaults. There is now significant evidence that outstanding student debt is displacing other spending by young adults, primarily in the auto and housing markets. Seeing as how the housing market is one of the primary drivers of US economic growth, it seems obvious that putting measures in place to reduce the burden of student debt moving forward will allow the US economy to recover more quickly (by increasing the private-sector purchasing power of young adult consumers).

Furthermore, it seems that the federal government should use its federal financial aid bargaining power to force private, for-profit colleges to improve student performance. There are lots of options here–most for-profit, less than 4-year institutions are vocational schools focused on employing students in a specific career field. In theory, it would be relatively simple to make the schools liable for a percentage of their graduates’ loan debt based on the success rate at which graduates find work in their field of study (or implement some other similar risk-sharing arrangement between the institution and the student).

Of course, in practice, the Dept. of Education doesn’t even have the ability to track graduate outcomes at that level of detail right now. On top of that, any change to institutional eligibility for Dept. of Education loans would have to be worked out through Congress, so I think we can assume that this isn’t going to happen anytime soon…

PS: shout out to Lil Laurie for the topic suggestion.

Are student loans profitable…for the government?

Like most complex financial issues, the answer to this question depends, a bit, on how ‘profit’ is defined and the accounting procedures used to calculate profit. If you go directly to the student loan projections by the Congressional Budget Office (CBO), it appears that the government is indeed profiting quite handsomely from direct loans. In 2013, the government will loan out 105 billion in student loans and expects a negative subsidy (profit) of ~36.5%, or over $37 billion dollars (closer to $35 billion once administrative costs are taken into account). These figures are presented as the net present value of future expected cash flow from the loans, accounting for expected future defaults and delinquencies.

Some groups dispute the CBO analysis; for example, the New America Foundation [page 7] explains that lenders (in the case of federal student loans, taxpayers) take on additional risks beyond default in any case where credit is extended:

“Like bond holders who earn interest for the risks that they bear, stock holders also seek a return on the capital that they have put at risk by investing in [a] loan company. When the federal government issues student loans, taxpayers are effectively stock holders in those transactions because they will have to bear default losses while Treasury bond holders who finance the government’s loans are theoretically never at risk of loss. [Additionally], in times of economic stress, loans may default at a higher rate than expected and these defaults are likely to be severe and costly. Lenders consider this type of risk an additional cost that is over and above expected losses from default and include an additional premium in the interest rate they charge borrowers to compensate for it.”

It is not uncommon for some people to argue that due to these additional risks beyond the cost of default, government accounting principles should include these additional risk factors in their discount rate (rather than using the risk-free discount rate of US Treasury Bills when calculating net present value of future cash flows). The term for this sort of calculation is ‘fair-value’ accounting, and the calculations substitute market interest rates for risk-free Treasury rates. The CBO also releases a ‘fair value’ accounting of student loans, and the difference in discount rates is enough to drop the subsidy on federal loans from $36 billion dollars to $5.5 billion dollars.

Regardless of how you discount for the risk in lending students money, however, the federal government is profiting from its current student loan practices (anywhere from $5.5 billion to $36 billion this fiscal year). Personally, I don’t think that the fair-value method for evaluating loan risk is an improvement because the federal government is not a corporation—the primary difference is that the US government can ‘lose’ money year after year (ie run a budget deficit) but actually improve its long-run fiscal position (if corresponding growth is higher than the budget deficit). Banks and other private lenders cannot do the same; it seems silly to me to evaluate risk factors equivalently when the market forces governing each type of lender are very different. Furthermore, and perhaps more importantly, private lenders’ sole revenue source from a loan is the interest paid by the debtor. However, in the case of the US government and student loans, the government makes money from interest payments, but the government also makes money from the increased earning potential of the college graduate.

For example, over the course of a lifetime, laborers with a Bachelors degree can expect to earn more than $500,000 more than an equivalently talented person with a high school diploma. For the person that wouldn’t otherwise be able to afford college, the government not only benefits from the interest payments on the student loan, but benefits from tax revenue on an additional half million dollars over the course of that student’s lifetime. When viewed in that light, it seems that there is really no way to argue that the US government doesn’t profit enormously from its student loan portfolio. In fact, given the positive spillover effects from higher education (increased tax revenue, more educated workforce, etc), the $36 billion projection may, in fact, understate the federal government’s profit from direct student loans. Given the abnormally high interest rates on today’s student loans, the positive externalities to a more educated workforce and populace, and the fact that the US government can (and should) be engaging in more fiscal stimulus to encourage growth in the aftermath of the financial crisis, I think there is a very strong case to be made that interest rates on student loans are much too high today.

Federal student loans are a part of the Department of Education budget, so running federal student loans at a significant loss would directly take spending away from other federal education programs (if there were no corresponding increase in the federal education budget). However, net subsidies from the program are deposited as receipts in the General Fund [page 14, footnote 1], which means that the Department of Education in no way benefits from making money off of the student loans. (In other words, it wouldn’t impact the federal education budget in any way if the net profit on student loans was 0 as opposed to $36 billion; that would just impact the overall deficit of that particular fiscal year.) To me, this is sufficient evidence to remove any doubt about whether the federal government should renew the 3.4% interest rate on subsidized Stafford loans to undergrads; in fact, I think there is a compelling case to be made that all student loans issued for the next fiscal year should be paying interest rates in the ~3% range.