Senator Warren’s Interest Rate Follies

First-term Senator Elizabeth Warren (D-MA) is a darling of the progressive Left, and she has been mentioned as a possible Presidential candidate in 2016 (although she has stated she’s not running). One of the ways she has gained support with the Democratic base is through her many public statements about the federal government’s purported profit on student loans, which she cites to be $51 billion in Fiscal Year 2013. Given the huge profit, she has introduced legislation to drop interest rates to the overnight borrowing rate at the Federal Reserve: 0.75%.

Her argument suffers from one main problem: student loans carry risk for the federal government. (She made my 2013 not-top-ten list for this reason.) The Congressional Budget Office, where the $51 billion estimate came from, uses federal borrowing costs as a discount rate. This discount rate is very low, in part because the federal government is viewed as very unlikely to default (even with the possibility of debt ceiling shenanigans). As a result, numerous groups have suggested the use of fair-value accounting, in which the risk of default is considered. Indeed, the Washington Post’s fact-checking blog gave Senator Warren’s statement of a $51 billion profit “two Pinocchios” because it did not consider fair-value accounting.

[On Twitter, the wonderful Libby Nelson notes that my explanation of fair-value accounting vs. federal regulations is unclear. Here is a nice CBO summary of the different methods.]

With the debate over student loan profits and accounting methods as a backdrop, the release of Friday’s Government Accountability Office report on federal student loans was eagerly anticipated in the higher education community. The title of the report succinctly summarizes the rest of the document: “Borrower Interest Rates Cannot Be Set in Advance to Precisely and Consistently Balance Federal Revenues and Costs.” This resulted in a few howlers from policy analysts, including this gem from Matt Chingos at Brookings:

Karen Weise at Bloomberg was a little more diplomatic with her summary of the report:

The report itself is fairly dry, but it does emphasize something that should be kept in mind when considering the costs of student loan programs. Due to the growing prevalence of extended payment plans, increased rates of income-based repayment plan usage, and the continued risk of defaults, the actual amount of the subsidy or cost on student loans will not be known for 40 years after disbursement.  Each of these individual variables could also have a large effect on the long-run subsidy or cost; for example, a higher-than-expected rate of income-based repayment participation could increase program costs.

The following paragraph on pages 18 and 19 sums up a key point of the report:

“As of the end of fiscal year 2013, it is estimated that the government will generate about $66 billion in subsidy income from the 2007 to 2012 loan cohorts as a group. However, current estimates for this group of loan cohorts are based predominantly on forecasted cash flow data derived from assumptions about future loan performance. As more information on actual cash flows for these loans becomes available, subsidy cost estimates will change. As a result, it is unclear whether these loan cohorts will ultimately generate subsidy income, as currently estimated, or whether they will result in subsidy costs to the government. This will not be known with certainty until all cash flows have been recorded after loans have been repaid or discharged—which may be as many as 40 years from when the loans were originally disbursed.”

I read this paragraph as providing possible evidence that interest rates may have been set relatively high compared to the federal cost of borrowing. (Recall that the interest rates for subsidized Stafford loans declined from 6.8% in 2007 to 3.4% in 2011, while Treasury rates were at historic lows.) The spread between the 10-year Treasury yield in May versus the interest rate on subsidized Stafford loans has been the following for the past seven years:

Year Stafford (pct) 10-yr T-note (pct) Spread (pct)
2007 6.8 4.75 2.05
2008 6.0 3.88 2.12
2009 5.6 3.29 2.31
2010 4.5 3.42 1.08
2011 3.4 3.17 0.23
2012 3.4 1.80 1.60
2013 3.86 1.93 1.93

This interest rate spread is statutorily set at 2.05% for subsidized Stafford loans in the future, roughly the long-run average. So while future GAO reports a few years after disbursement may find similar results, what we’ll all be waiting for is longer-term data to see if the estimates hold true. The federal government doesn’t necessarily have a great history of long-run cost projections, so I’m expecting this spread to disappear over time. (And keep in mind this report doesn’t fully account for risk.)

Yet Senator Warren and eight other Democrats released a press release on Friday afternoon with the headline of “Democratic Senators Highlight Obscene Government Profits Off Student Loan Program.” They focused entirely on the initial projection of a $66 billion profit over five years and entirely ignored the long-run uncertainty highlighted by the GAO. This press release is a great example of selecting only the most favored parts of a report, while ignoring other important details along the way. Again, the Twittersphere (myself included) expressed its thoughts:

On a more fundamental note, I think that Senator Warren and colleagues are misguided in their efforts to continue lowering student loan interest rates. Given the reality that higher education funding is a zero-sum game, I would much rather see funds used to support the Pell Grant, work-study, and other upfront sources of aid for students than slightly lower loan payments after students have already left college. (The same argument holds against tax credits.) Senator Warren may not be running for President (yet), but she’s in the running for my 2014 not-top-ten list.

Will Holding Colleges Accountable for Default Rates be Effective?

As student loan debt continues to climb and Congress enters a midterm election year, three Democrats in the United States Senate (Reed, Durbin, and Warren) recently introduced a piece of legislation designed to hold certain types of colleges and universities accountable for their students’ loan default rates. If enacted, the bill would require colleges to pay a fine of a percentage of its students’ total defaulted loans to the Department of Education, part of which would be used to help borrowers avoid future defaults and the other part would go to a fund to help support the Pell Grant in case of any future funding shortfalls.

The proposed fines are the following:

  • 5% fine if the most recent cohort default rate (CDR) over three years is 15-20%
  • 10% if CDR is 20-25%
  • 15% if CDR is 25-30%
  • 20% is CDR is 30%+

As an example of what these fines could mean, consider their potential implications for the University of Phoenix’s online division. Data from the Department of Education’s Integrated Postsecondary Education Data System (IPEDS) show that Phoenix collected roughly $1.4 billion in student loan revenue during the 2011-12 academic year, while 34.4% of students who took out loans defaulted in a three-year period. This default rate would place them in the 20% fine category, resulting in a fine of roughly $100 million per year based on an estimated $500 million per year in defaulted loans. This would represent roughly four percent of their total tuition revenue ($2.7 billion) in the 2011-12 academic year—which is far from a trivial sum.

Daniel Luzer on Washington Monthly’s College Guide blog (where many of my pieces are cross-posted) notes some of the potential positives of this legislation, including encouraging colleges to spend more time and energy counseling students and providing more information about financial aid.

But, in order for this legislation to actually benefit students, three things must happen:

(1) Some colleges must actually be affected by the legislation. The sanctions in the bill would not apply to community colleges, historically black colleges and universities (HBCUs), and likely other colleges designated as minority-serving institutions. This excludes a substantial number of nonprofit institutions, many of which have higher default rates. A provision in the bill excludes colleges at which fewer than 25% of students take out federal loans, which further diminishes the number of nonprofit institutions on the list.

But even if a college is not exempt from the legislation, it is still possible to avoid fines if default rates are over 15%. The legislation grants the Secretary of Education the authority to grant waivers, which would be the first time the Secretary has ever been granted that authority. (Kidding!) Colleges can submit remediation plans in order to avoid or reduce fines. It will be interesting to see the reaction to the first waiver request, as colleges’ lobbying efforts tend to be well-organized.

A more interesting case will involve the for-profit sector. Given the three Senators’ general distrust of for-profit institutions, it would not surprise me if nearly all of the colleges facing fines are proprietary in nature. But the way the bill is targeted seems to be similar to previous attempts at gainful employment legislation, which have been the subject of massive amounts of litigation. Expect this proposal to face litigation if it ever became law.

(2) Colleges must be able to improve their financial aid offices without restricting students’ access to financial aid. One of the underlying premises of this legislation is that financial aid offices are not helping students make sound financial decisions that help them complete college. Aid administrators would likely disagree with that statement, although additional resources targeted toward financial counseling may be beneficial.

Another concern is that in order to reduce default rates, aid offices will not offer students loans if they perceive the student as having a higher risk of default. While there is a prohibition written into the legislation against denying loans based on the perceived risk of default, this would be extremely difficult to prove and enforce. Colleges are not required to offer students the full amount of loans available in the initial aid package, and indeed some community colleges decline to offer any federal loans to their students. Some colleges would like more authority to limit loan offers to students, and this legislation could reduce access to credit for needy students.

(3) The legislation must adequately address students who transfer. If a student takes out loans while attending multiple institutions, would each college be held responsible for a student’s default—even if most of the debt was at one institution? Consider a student who attends a regional public university for one year and takes out the maximum in subsidized Stafford loans ($3,500). She then transfers to an expensive private college and accrues an additional $30,000 in debt before graduating. If she defaults on her principal of $33,500, should both colleges be held responsible? That is unclear at this point.

So would holding colleges accountable for default rates (in the method of this legislation) help students? I’m skeptical because I don’t see many colleges actually facing sanctions, nor do I see the fines being particularly effective. This is one of those ideas that is great in theory, but may not work as well in practice.

I don’t think this legislation is likely to become law in its current form, but it’s worth keeping an eye on as the Department of Education works to develop the Postsecondary Institution Rating System (PIRS). Many of the potential discussions this legislation raises will certainly come up again once the draft ratings are released.

Does This Explain Opposition to Market-Based Interest Rates?

As of this writing, it appears that the U.S. Senate has finally reached an agreement on student loan interest rates after subsidized Stafford rates doubled from 3.4% to 6.8% on July 1. The general terms of the agreement are similar to what President Obama proposed in his FY 2014 budget and what the House of Representatives agreed to back in June, with some compromises on each side.

The big difference between current law and the Senate agreement is that interest rates for nearly all student loans will be tied to the 10-year Treasury rate, which currently sits at about 2.5%.  (Undergraduates would pay a 2.05% premium above the Treasury rate on Stafford loans to account for program costs and the risk of offering the loans.) However, the Treasury rate is expected to increase to 5.6% by 2016, pushing the interest rate for undergraduates to 7.65% from less than 5%. The plan includes a cap at 8.25%, which may be reached according to the CBO report.

The Senate agreement is not without its critics, particularly on the political Left. Senator Bernie Sanders, a Vermont independent and a self-described “socialist,” criticized the plan as “dangerous” in an article in The Hill. His criticism lies in the fact that interest rates can rise well above the current 6.8% over time, a very real concern given the interest rate projections. While the plan is expected to pass the Senate (and the House), some other Senate Democrats will likely vote no as well.

At this point in the great interest rate debate, I have to wonder if there is another reason some politicians oppose market-based interest rates. Tying student loan interest rates to the 10-year Treasury note directly connects students’ future payments to the cost of federal borrowing. And that cost of federal borrowing is influenced by the federal government’s fiscal policy.

This connection between federal borrowing and student loan rates could potentially have the following repercussions. If loans are tied to Treasury notes—and there is no way to fix the rate as has been done for the past seven years—students should have an incentive to push for federal policies which lower the federal government’s cost of borrowing. (With the decline in home ownership rates among younger adults, fewer 20- and 30-somethings have mortgages, which are affected by federal borrowing costs.)

The policy that best reduces the cost of borrowing is a balanced budget, which reduces the need for additional borrowing. The passage of the Omnibus Budget Reconciliation Act of 1993, which reduced the budget deficit through a combination of tax hikes and spending cuts, had the effect of driving down long-term interest rates. (For more on this, I highly recommend reading Bob Woodward’s Maestro about Alan Greenspan’s role in the policy discussions.)

My question to readers is whether you think that some politicians may oppose market-based interest rates because more young adults may place pressure on Congress to find some legislative solution to balance the budget—although the solutions certainly vary by political persuasion. Say it’s 2016 and undergraduate Stafford rates are 7.5%, with hitting the 8.25% cap becoming more likely. Could we see student advocacy organizations pushing for a balanced budget to bring down interest rates? I don’t know how many people will think this way, but it’s something to consider.

The Great Student Loan Interest Rate Debate

As I write this post, the House of Representatives is currently debating the future of student loan interest rates. Under current law, the rates on subsidized Stafford loans for undergraduates (the rates which get the most attention) will double on July 1 from 3.4% to 6.8% without Congressional action. The same debate was held last year under the same parameters, but Congress and the President agreed to extend interest rates for an additional year.

There have been a wide range of proposals put forth regarding plans to address the interest rate cliff, an outstanding summary of which was written by Libby Nelson in Inside Higher Ed. (In addition to the plans listed in that article, some Senate Democrats have supported a two-year extension to current law in order to allow for the Higher Education Act to be reauthorized.) Most proposals move to tie interest rates to the market—represented here by borrowing costs for the federal government—but the plans vary widely in their ideas of what the relevant market should be.

Proposals put forth by the Obama Administration and House and Senate Republicans all tie interest rates to long-term Treasury bills, but vary in their other features. (I’ve previously written on the Obama Administration’s proposal.) While the President has threatened to veto the common House GOP proposal over certain aspects, there is enough common ground here to reach an agreement.

However, proposals put forth by certain Democratic senators, particular Sen. Elizabeth Warren from Massachusetts, confuse long-term lending risks with short-term credit markets. She has proposed tying student loan interest rates (which are repaid for at least ten years once a student leaves college) to the interest rate the Federal Reserve charges banks for very short-term borrowing. Jason Delisle of the New America Foundation, hardly a bastion of conservatism, crushes her argument in a great piece of writing. He notes the confusion between short-term and long-term rates, as well as accounting for the probability of default. I would also note that if Congress wishes to help make college more affordable, it’s a better idea to give the funds upfront to students than to lower interest rates later on–long after the enrollment decisions have been made.

The federal government should move toward some sort of a market-based strategy for interest rates with certain student protections. This would allow for the costs of student loans to be more adequately reflected in the federal budget. (And if interest rates get too high, maybe it’s a reminder for Congress and the President to produce a balanced budget!) With that being said, I would still expect to see a short-term extension of the current interest rates as Congress may end up deadlocked on this issue until the Higher Education Act is reauthorized.

Why Expanding Student Loan Interest Deductions is Unwise

The United States Congress has rarely met a change to the tax code that it doesn’t like. Since 2011, Congress has made nearly 5,000 changes to the tax code, making the tax code even more Byzantine and causing many Americans headaches each spring. While the exact levels and types of taxes preferred by the public tend to differ by political and economic ideologies, a general consensus exists among economists that a stable, predictable, relatively simple tax code achieves revenue goals at a reasonable cost.

Yet Congress is constantly faced with pressure to change the tax code to allow more credits and deductions to incentivize certain activities. We have certainly seen incentives in the tax code to increase educational attainment through the use of credits and deductions. These incentives have a difficult time being effective in inducing students to obtain more education because students and their families do not receive a financial benefit until well after the enrollment decision has been made, and awareness of these programs is uneven at best.

The deduction for student loan interest payments is designed to help student be able to repay their educational debt; currently, students can receive a $2,500 deduction on their taxes (if they itemize) in a given year and their income is less than $75,000 per year if single or $150,000 per year if married filing jointly. It is hard to imagine that such a program, which does not provide a financial benefit to students until after they have left college, will help increase educational attainment.

Yet Congressman Charles Rangel (D-NY), who is well-known for his 2012 censure over failing to pay taxes, has introduced the Student Loan Interest Deduction Act, which would double the tax deduction for student loan interest and get rid of the income cap for receiving the deduction. This bill has picked up the support of many higher education associations, including the American Council on Education and the National Association of Student Financial Aid Administrators, but it is an example of inefficient and misguided public policy.

Here are a few reasons why the proposed bill is poor public policy:

(1)    Students get the tax deduction years after beginning college. Yes, students with perfect information about the deduction (and who believe the policy will remain in place) will adjust their total cost of college down. But that requires a large amount of information and a low discount rate, neither of which can be assumed.

(2)    Increasing the deduction cap allows students to benefit more if they took out more debt. This can provide an additional incentive for colleges to raise their price and capture additional revenue if they think students will pay.

(3)    Getting rid of the income cap shifts the benefits more toward higher-income families, who are more likely to itemize deductions and be able to use the deduction. Advocates on the left should be particularly steamed by this point.

(4)    This acts as an additional tax on saving for college relative to borrowing, and can place families who chose not to borrow at a disadvantage.

I would prefer to see all interest deductions removed from the tax code and replaced with more efficient upfront payments. Instead of forgoing billions of dollars in revenue through tax credits and deductions, I would rather see the money used in grant programs, lowering the interest rates for federal loans, or used toward other educational improvements. Rangel’s bill just complicates the status quo and does little to actually help students afford college.

Something Old, Something New: The FY 2014 Obama Budget

Even though I know that it has no chance of being passed in anything resembling its current form, I am excited to get my hands on President Obama’s long-delayed budget for Fiscal Year 2014 (short version, long version, six-page summary of the education portion). The funding request for the Department of Education is for $71.2 billion in discretionary spending, 4.6% higher than this year’s (pre-sequester) budget; ED is unlikely to see an increase of greater than inflation this year given the current political climate.

I tweeted my way (follow me!) through some of the key points relating to higher education yesterday, and am now back with a more detailed summary of the budget. (I also recommend Libby Nelson’s excellent summary in today’s Inside Higher Ed.)This year’s theme is “something old, something new,” as many of the proposals are recycled from last year—but with one key difference that will affect millions of students.

First of all, not much changes with respect to the Pell Grant. The President proposes a $140 increase in the maximum Pell Grant to $5,785, while the program is on more solid financial footing for the next few years. He is again trying to get a higher education version of Race to the Top passed this year, which will look similar to the plan from last year. Again, there is a strong focus in the STEM fields and for program evaluation (the latter of which is welcome from my perspective). The biggest program boost I could find was to FIPSE (the Fund for the Improvement of Postsecondary Education), going from under $2.4 million to $260 million. Although it is unlikely to be adopted, it does show a commitment to demonstration projects in K-12 and higher education.

The most controversial part of the President’s budget is the proposed shift to market-based interest rates. A day after Republican Senators Coburn, Burr, and Alexander introduced a bill to tie all interest rates to the ten-year Treasury rate (currently 1.8%) plus three percentage points, the President’s budget also proposed tying interest rates to the same measure. His plan is more nuanced, with different loans having different premiums over the Treasury rate (see p. 344-350):

Subsidized Stafford: Treasury plus 0.93% (about 2.75% currently)

Unsubsidized Stafford and Perkins: Treasury plus 2.93% (about 4.75%)

PLUS: Treasury plus 3.93% (about 5.75%)

GOP plan: All loans are Treasury plus 3% (about 4.8%)

These rates are far lower than the current rates (3.4% for subsidized Stafford, 6.8% for unsubsidized Stafford, and over 8% for graduate unsubsidized loans), but do shift risk onto students as the rate for new loans would change each year. There would also be no interest rate cap, which is lamented by many advocates. (Income-based repayment provides another alternative, however.)

If either of these plans is adopted, the interest rate cliff would be eliminated as students would no longer have to wait on Congress to know their rates. However, students are likely to see rates rise as Treasury yields return toward their historical norm. The Congressional Budget Office predicts that 10-year Treasury notes will yield 5.2% by 2018, which would put unsubsidized loans just over 8%. (This is still lower than the recent rate for unsubsidized graduate loans, with which I am quite familiar.) If rates go higher than that, I expect Congress to enact an interest rate cap in several years.

The federal budget process does not move quickly, especially with a divided Congress. While I do not expect large increases in the Department of Education’s budget, I am optimistic that a market-based solution to interest rates will be adopted in order to provide more certainty in the short run and to bring graduate loan rates closer to what the private market would otherwise offer.

Should Campuses be Able to Limit Student Loans?

The National Association of Student Financial Aid Administrators jumped into the financial aid reform debate this week with the release of their policy paper as a part of the Gates Foundation’s Reimagining Aid Delivery and Design (RADD) project. Many of the recommendations are similar to other papers in the panel (including proposals to increase the maximum Pell Grant for certain students and providing more information for students and their families to make better college decisions)—and an additional recommendation of exploring an early commitment program for Pell recipients is informed by some of my research, which is pretty nifty.

The NASFAA report does make one recommendation which will likely prove to be highly controversial—limiting eligibility for student loans for certain groups of students in a clear effort to reduce student loan default rates. First, NASFAA suggests that students who do not meet a baseline level of academic preparation (perhaps a combination of ACT/SAT scores and high school GPA) would not be initially eligible to take out federal student loans. This proposal would be similar to the academic eligibility index used by the NCAA to determine student-athletes’ ability to play college sports. This proposal could have the effect of ending the open-access institution as we know it, depending on exactly where the cutoff is set. While it is true that students with lower standardized test scores are less likely to complete college, I’m very hesitant to place a substantial barrier to college entry—especially for students who did not enroll in college directly after completing high school.

The report also contains a recommendation allowing colleges to restrict groups of students’ ability to borrow if the financial aid officer feels that the loan funds are not needed or risky. For example, education majors’ loans may be limited compared to business majors because of their lower annual earnings (and reduced repayment abilities). Restricting access to loans by program characteristics (instead of individual characteristics) reduces the burden on financial aid officers, but also fails to take individual characteristics into account unless a student appeals for professional judgment.

The proposal to limit student loans will penalize students who cannot pay for college by any other means—especially for dependent students who cannot get parental support to pay for their expected family contribution. Additionally, many students cannot borrow the maximum amount of loans under current rules, which base eligibility in part on the estimated cost of attendance. Research suggests that this posted cost of attendance may be much lower than the actual cost of attending college, as institutions have an incentive to make the college look as affordable as possible.

While I am concerned about these particular portions of NASFAA’s proposal, they raise concerns that are of genuine merit and concern in the financial aid and policy communities. I would be surprised if they become a part of federal rules in any meaningful way, but this does show the diversity of opinions within the RADD group and the importance of listening to as many stakeholders as possible before redesigning the financial aid system.

Predicting Student Loan Default Rates

Regular readers of this blog know that there are several concerns to using outcome measures in a higher education accountability system. One of my primary concerns is that outcomes must be adjusted to reflect a college’s inputs—in non-economist language, this means that colleges need to be assessed based on how well they do given their available resources.  I have done quite a bit of work in this area with respect to graduation rates, but this same principle can be applied to many other areas in higher education.

The Education Sector also shares this concern, as evidenced by their recent blog post on the importance of input-adjusted graduation measures. In this post (at the Quick and the Ed), Andrew Gillen examines four-year colleges’ performance in student loan default rates. He adjusts for the percentage of Pell Grant recipients, the percentage of part-time students, and the average student loan size to get a measure of student default rate performance.

I repeat this estimate using the most recent loan default data (through 2009-10) and IPEDS data for the above characteristics for the 2009-10 academic year. This simple model does a fair job predicting loan default rates, with a R-squared value of 0.422. Figure 1 below shows actual vs. predicted loan default rates for 1876 four-year institutions with complete data:

figure1_jan17

The Education Sector analysis did not break down student default rate performance by important institutional characteristics, such as type of control (public, private not-for-profit, or for-profit) or the cost of attendance. Figures 2 and 3 below the performance between public universities and their private non-profit and for-profit peers:

figure2_jan17figure3_jan17

Note: A positive differential means that default rates are higher than predicted. Negative numbers are good.

The default rate performances of public and private not-for profit colleges do not differ in a meaningful way, but a significant number of for-profit colleges have substantially higher than predicted default rates. This difference is obscured when all colleges’ performances are combined.

Finally, Figure 4 compares default rate performance by the net price of attendance (the sticker cost of attendance less grant aid) and finds no relationship between the net price and loan default rates:

figure4_jan17

Certainly, more work needs to be done before adopting input-adjusted student loan default rates as an accountability tool. But it does appear that a certain group of colleges tend to have a higher percentage of former students default, which is worth additional investigation.

Am I On the Wrong Job Market?

In light of being on the academic job market this year, I was amused to get the following mailing from the local branch of Globe University. (Even though the mailing was addressed to me, it was also addressed to “Or Current Resident.”)

Globe Ad

The message is quite simple: graduates of this university get jobs. The mailing advertises that 100% of graduates with a bachelor’s degree in business administration get employment, although the fine print does mention that “employment is not guaranteed.” However, not much can be said about the graduation rates of students attending any of the Globe campuses, both because very few students attending Globe are first-time, full-time students (which are the only students counted in federal graduation rate calculations) and because many campuses (including Madison) have not been open long enough to have a graduation rate cohort.

To get an idea of graduation rates, I looked at the oldest Globe campus, in Brooklyn Center, MN. The reported graduation rate is 23%, with an overall career placement rate of 72%. Meanwhile, tuition is over $5,000 per quarter before mandatory course fees. I’m not saying that Globe University is a bad bet for students, but some students are likely to benefit more by attending the local technical college.

Moral of the story: Don’t believe colleges which imply that everyone gets a job. This isn’t even true at the most prestigious colleges, let alone for relatively unknown for-profit institutions. (Now, I do hope that my UW-Madison PhD helps me get a great job!)

Paying It Forward: A Different Take on Income-Based Repayment

In prior blog posts, I have been less than charitable toward the federal government’s changes to the income-based repayment policies for student loans. (As a reminder, these changes provide large subsidies to students who attend expensive colleges and particularly those who earn good salaries after having attended law or medical school.) My criticism of the federal government’s way of implementing the program does not mean that I am not open to a better way of income-based repayment. With this in mind, I look at a proposal from the Economic Opportunity Institute, a liberal think tank from Washington State, which suggests an income-based repayment program for students attending that state’s public colleges and universities.

The EOI’s proposal, called “Pay It Forward,” would charge students no tuition or fees upfront and would require students to sign a contract stating that they would pay a certain percentage of their adjusted gross income per year (possibly one percent per year in college) for 25 years after leaving college. It appears that the state would rely on the IRS to enforce payment in order to capture part of the earnings of those who leave the state of Washington. This would be tricky to enforce in theory, given the IRS’s general reticence to step into state-level policies.

I am by no means convinced by the group’s crude simulations regarding the feasibility of the program. This is currently short on details and would also require a large one-time investment to get off the ground and enroll an initial cohort of students. Additionally, it is not clear whether the authors of the report accounted for part-time enrollment patterns in their cost estimates. I also urge caution with this program, as this sort of an income-based repayment program decouples the cost of attendance from what students actually pay. Colleges suddenly have a strong incentive to raise their posted tuition substantially in order to capture this additional revenue.

With all of these caveats, the Pay It Forward program does have the potential to serve as a simple income-based repayment program once analysts do more cost-effectiveness analyses. But this will only work if policymakers keep a close eye on the cost of college in order to result in a revenue-neutral program. My gut feeling is that the group’s estimates understate the cost of college under current rules and don’t consider the possibility of the incentives that will increase cost.