That does not mean earnings should be the sole way to gauge college quality. And earnings levels alone should never be used to rank colleges.
What earnings are is a recognition that questions of college quality should go beyond just academics. In this case, earnings are an important indicator of the quality of a student’s higher education investment and should be used in a thoughtful, careful, and limited way to capture that concept.
Earnings challengesWhen making quality judgments about colleges, earnings play a messier role than completion rates. For one, they are strongly linked to what students study, not just where they go to school. This means institutionally aggregated numbers may say as much about the programs a school offers as they do about the success of its graduates. Measuring earnings also tends to look at only completers, potentially distorting the results for lottery ticket" programs where only a few students finish, but those who do turn out well.
[pullquote]As long as graduates are not impoverished, there’s no compelling policy reason to care about different income levels by college[/pullquote]
There's also a workforce challenge with earnings. For instance, we don't know that employers actually reward academic quality with higher salaries. They may instead just rely upon what they think are proxies for quality, such as a college's reputation. And the workforce also demands a number of non-academic skills and behaviors that may not even get captured in a robust assessment of college quality.
Any use of earnings for quality purposes should be cognizant of these difficulties. This is best done in two ways. First, earnings should be used with a simple minimum standard that does not laude or penalize colleges solely for results that could be due to the programs they offer. Second, any consideration of earnings beyond a minimum level should be done only in the context of student debt.
Minimum earnings, not levelsRanking colleges by earnings level is the worst way to use these data as a quality measurement. As long as graduates are not impoverished, there’s no compelling policy reason to care about different income levels by college. That’s because regardless of whether students make $30,000, $50,000, or $100,000 we can still confidently assert that they are on a path toward the middle class or better.
[pullquote]Earnings levels should only be measured against a minimum bar[/pullquote]
Actual earnings data are also problematic at the institutional level because they reflect a great degree of noise like program mix, geographic location, gender mix of students and other characteristics that could explain as much about the earnings as the college itself. Colleges also should not be punished for producing lots of teachers, social workers, or other professions that have lower wage ceilings.
Instead, earnings levels should only be measured against a minimum bar. We should expect that any postsecondary institution is producing graduates who on average have at least some discretionary earnings and are not impoverished. Sure, a given student may suffer unfortunate circumstances and not do well, but we should be extremely concerned about any institution where the typical graduate is making less than 150 percent of the poverty line, or $17,505.
Focusing on the minimum standard at such a modest level addresses concerns about program mix, gender, and the other characteristics that can distort attempts to judge colleges solely based on earnings level. It also would not disadvantage public service programs, where the ceiling on salaries is low, but the floor is reasonable. For example, the National Education Association estimated that the average starting teacher salary in 2012-13 was $36,144--more than double the suggested minimum threshold. And the Bureau of Labor Statistics estimates the 10th percentile wage of social workers is $31,690.
The crucial relationship between debt and earningsWhile things like a love of learning are great and important, they cannot make a payment when loans come due. And at a time when a majority of students borrow for college we cannot ignore that the average graduate will have lingering college costs.
[pullquote]Any consideration of earnings beyond a minimum test should be done in the context of student debt[/pullquote]
That's why any consideration of earnings beyond a minimum test should be done in the context of student debt. The best way to do this is through a debt-to-earnings rate. This measures investment quality by judging institutions based on a comparison of how much students must pay for their college loans compared to their income after several years in repayment. Ideally, it would be done at some form of the program level.
This approach should sound familiar because it’s similar to the one the U.S. Department of Education took for its proposed rule that defines whether certain college programs are providing gainful employment. In that rule, debt-to-earnings rates are calculated for each program; non-vocational institutions should probably look at them in terms of undergraduate college (e.g. break up results by arts and sciences and engineering).
Setting a minimum debt-to-earnings rate that colleges would have to meet strengthens the connection between college prices and likely employment situations. Since colleges would be unable to give students too much debt compared to their income, the institution would be more cognizant of what the market will bear for graduates.
Colleges could employ multiple strategies to meet this measure. On the earnings side they could improve the quality of programs, create new ones with better labor market outcomes, or strengthen their job placement and career services assistance. At the same time, they could work to restrain debt through price reductions, increased aid, better counseling, faster time-to-degree, and other strategies.
A debt-to-earnings rate also provides a check on excessive use of income-driven repayment plans. New students can enroll in these plans once their loan payments exceed 10 percent of their adjusted gross incomes and receive forgiveness for amounts not paid off after 20 years. A college with a debt-to-earnings rate of 20 percent or more is thus likely going to have a lot of former students enrolled in an income-driven plan where they do not make enough payments to retire the debt and end up receiving loan forgiveness at the taxpayers’ expense.
The debt-to-earnings rate changes this calculus. A college will be judged on indebting students too much even while individuals are protected by the safety net that is income-drive repayment. This should ensure income-driven repayment options remain in place for cases where things go wrong for borrowers, not as the de facto option due to overly high prices and debt levels.
Students care about earnings and we should tooStudents clearly care about the value of the earnings return of their college educations. It's not an accident that business accounts for just over one-fifth of all bachelor’s degrees awarded, making it the most popular undergraduate major. And a majority of incoming freshmen indicate that a colleges’ reputation and ability to get jobs for graduates are among their “very important” reasons for picking their institution.
The long-term financial investment we're asking so many students to make demands that earnings become part of our discussion of college quality. They should not be the only measure, they should be used carefully, and we absolutely must still look at academic results. But we can no longer pretend the dollars don't matter."