The Mythical College Savings Penalty

Richard Vedder has a column over at Bloomberg View today exposing what he labels the “stealth tax” on family savings. Citing uncited “considerable anecdotal evidence,” Vedder claims that savers pay almost three-quarters of their earnings for college, a much higher rate than a family of comparable wealth with no savings. It’s an alarming argument, but it’s also a largely unsourced account, lacking lacking any information about how savings and their treatment in financial aid calculations actually work.

Most Families Save Through Mortgages or Tax-Advantaged Accounts…

Financial aid formulas do not treat all savings as the same. Most middle-income families today save money either though building up home equity or making use of tax-advantaged retirement savings vehicles like 401(k)s, Individual Retirement Accounts (IRAs), Roth IRAs, etc. Someone under 50 can contribute $17,500 to the 401(k) and $5,500 to the IRA, netting them $23,000 in annual savings plus whatever they are building up through a mortgage. And since accounts are for individuals, a married couple with two incomes could potentially double those amounts. It’s after these limits that folks would likely turn to other investments that do not have special tax treatment.

 …And the Federal Government Doesn’t Touch Them for Student Aid Calculations

The Free Application for Federal Student Aid (FAFSA) does indeed ask students (and parents if the student is a dependent) for information on their assets. But the instructions for the form (see page 2) explicitly exclude: the value of the home the parents/student live in, retirement plans, such as 401(k)s, non-education IRAs, pension plans, annuities, etc., and life insurance In other words, before the family has entered a single cent, the form already excludes the predominant savings vehicles used by most families. 

Remaining Assets are Heavily Discounted

You Get Credit for Assets, But Not for Most Debts

Another suggestion in Vedder’s argument is that free-spending is actually rewarded. But apart from non-business real estate, where the amount counted is the value minus remaining debts, balances on credit cards, auto loans, etc. are not deducted treated as “negative” assets. And the contribution from assets cannot be negative, so you can’t use being underwater on a second home to try and offset contributions from income.  Heavy debts may reduce the amount of cash in checking and savings accounts treated as assets, but the large protection allowance probably makes that meaningless. 

So how much is that savings penalty really?

Vedder’s example for the savings penalty involves two otherwise identical families that both make $125,000 a year, with one having $100,000 in savings and the other having nothing. If the $100,000 is held entirely in retirement accounts, then there’s no difference in the expected contribution for the thrifty versus free-spending families. If those additional savings are held in assets that are counted, then the contribution for the saver is about $4,000 higher. Yes, that’s a bigger contribution, but the family also has $100,000 more in assets to make paying for college easier and avoid debt. If anyone thinks that’s an unfair trade and would like to trade me $100,000 for $4,000 feel free to contact me. 

A Glimmer of Truth in Institutional Aid

While Vedder’s argument falls apart with respect to federal aid, it’s nearly impossible to evaluate his claim with respect to institutional aid.  That’s because many of selective, private nonprofit institutions rely on additional documentation through the CSS/Financial Aid Profile to calculate aid based upon a completely opaque and customized “Institutional Methodology.” There is no public base formula and it varies by school so there’s no way to compare it to the federal formula. (The Financial Aid Journal has a summary here and a less clear brochure from College Board is here.) The institutional formula does include houses, so it’s possible that those who save though their homes could be hit more than non-savers, but it still relies on a very low share of assets–between 3 and 5 percent depending on the level.  

Wrong Argument, Mostly Right Conclusion

The irony is that while Vedder’s arguments on the savers’ tax are dubious, his conclusion that aid determinations should rely largely on income alone makes a lot of policy sense. At this point, getting rid of assets is the next logical step in FAFSA simplification, as that section  presents a lot of questions that take much more time to answer than those related to income and make little difference in the end result. It would also lay the groundwork for experimenting with awarding aid based upon older income information straight from the IRS or other changes that could make it even easier for students to get aid. Sure it might have some small effect on contributions for richer people, but probably not enough to offset the gains at the other end of the income spectrum. 

Simplifying the federal formula will go a long way, but it won’t do enough as long as a select group of colleges keep relying on additional information and opaque formulas to generate aid estimates that are wildly different than what the federal government suggests. Such machinations in the name of rooting out every last little false positive do nothing to help lower-income students understand what college is actually going to cost and likely increases confusion for everyone about how their aid packages really work. And that’s before colleges start deviating from the expected contribution through so-called merit aid, tuition discounts, gapping, and the like. When it comes to stealth penalties, the issue isn’t taxing savers, but what schools are doing for low-income students and how they get there.