Whenever policymakers argue over ways to lower the budget deficit, one of the most popular ideas on both sides of the aisle is "means testing" programs like Medicare or Social Security. Instead of cutting everybody's benefits, the idea is to reduce them for the rich and middle classes while leaving them intact for the poor.
In theory, means-tested programs should be more efficient and progressive because they don't spend money on those who can pay their own way. But one concern that dogs these proposals is that the programs will lose support and funding as soon as budgets get tight. As the saying often goes: "Programs for the poor are poor programs."
Over the last several years, we've witnessed a high-profile example of that principle in action. Inadvertently, America's higher education system has become a massive lab experiment, the results of which suggest that means testing social programs can ultimately hurt the very people it is meant to protect.
In the past, almost all public universities were heavily subsidized by state taxpayers, which kept tuition low for everyone. In recent years, however, legislatures have cut support even as the pressure to spend -- often on new buildings and other "amenities" -- has grown and become baked into the system. As a result, universities have ended up strapped for cash.
In response, many schools have increased their "sticker prices" -- the tuition they advertise to the public -- while offering many students discounts through financial aid packages. Affluent undergraduates pay the full cost of their educations, while lower-income students receive aid and pay a much lower final net price. This is called the "high-tuition, high-aid model," and it is essentially a means-tested program. Some higher ed experts have praised this shift, arguing that it would help low-income students by decreasing their tuition.
Unfortunately, this argument has not held up in practice.
Reality: High Tuition + High Aid = Higher Prices for All
As more schools have adopted the high-tuition, high-aid model, prices have gone up for low-income and high-income students alike. The reason is simple: Tuition is going up faster than aid. In the end, low-income students -- as well as middle- and upper-income students -- face much better prices at schools that keep total price low across the board. As Edward St. John and Douglas Priest wrote a few years ago in Inside Higher Ed, "The reality of high-tution/high-aid did not match the vision advocated by progressives."
At Penn State, which uses the high-tuition, high-aid model, the sticker price for tuition at the school's flagship campus was $15,250 for the 2010-2011 school year, the most recent year for which all data is available. Including room and board and other expenses and subtracting financial aid, public college students from Pennsylvania families making less than $30,000 a year faced a bill of almost $17,000 per year. Families had to pay more than half of their yearly income to send one child to school. That's after factoring in financial aid from any source, including the maximum federal Pell Grant awarded to the student.
A few hundred miles south in North Carolina, students at the main Chapel Hill campus of low-sticker-price UNC faced tuition of around $6,700 for the same year, less than half of what Penn State charged. With all additional costs and aid factored in, UNC students from families making less than $30,000 paid only $4,000 in final net price.
Penn State and UNC have the same share of students with Pell Grants, meaning that they have a similar percentage of low-income students. But despite having a much higher sticker price, PSU actually offers less aid to students than UNC. Only 29 percent of Penn State students receive scholarships or grant aid from the school, while nearly half of UNC students do - and the average amount of aid is smaller at Penn State. Thus, the much larger gap for students at Penn State is not nearly covered.
The model is not high-tuition, high-aid. It's just high-tuition.
To test whether these examples are accurate more broadly, I analyzed data from the largest public four-year colleges and universities from the National Center for Education Statistics. The aggregated data indeed show that, while there are exceptions, schools with higher sticker prices tend to have higher net prices for the lowest-income students.
One reason that the high-tuition, high-aid model isn't working is that schools are not using the increased revenue to pay for more aid. On average, according to Andrew Gillen of Education Sector, only 60 percent of the increased revenue from a higher sticker price is recycled back to aid. A mere 3% of schools surveyed increased aid by as much as they increased tuition. The schools that do increase aid often target the aid at particularly desirable - and often wealthy - students to try to lure them onto campus. Studies show that the money for this "merit aid" mostly benefits the affluent and has come at the expense of money for the need-based aid that is essential to making the high-tuition, high-aid model work as intended.
In short, financial aid for the poor has become a poor program because the money gets used elsewhere.
The Dark Side of Means Testing
The state of public higher education today serves as a powerful cautionary tale about means testing more broadly. Universal programs - think Medicare and Social Security in their current incarnation, or state university systems that have kept tuition low for all - tend to enjoy stronger support and have greater protections for low-income recipients. While under a means tested program, funding shortfalls are often made up by raising the costs for the poor (after all, they are still getting a discount), universal programs share the brunt of price increases across the board. This creates more political leverage for people who are otherwise viewed as recipients of charity that can be taken away. "Universal programs generally have a greater and more stable effect upon life-course security than targeted programs because they have a larger political base of support," write health policy scholars Barbara Starfield and Anne-Emanuelle Birn.
Not every means tested program is problematic, of course. Social Security's Supplemental Security Income program, for example, provides additional money to disabled adults and children beyond normal old-age and survivor's insurance benefits. In the education world, there is evidence that some means tested supplemental programs for the least-advantaged students, like Pell Grants, help increase equal access. The crucial issue, however, is how central the means test is to the whole system. SSI and Pell Grants are supplements rather than core features of the system. Means testing and targeted programs might work as additional benefits to boost up the least-advantaged individuals in an otherwise universal system, but they cannot be the crux of the system themselves.
Like the products being hawked on late-night infomercials, policies that claim to be more progressive and less costly often come with a catch -- they don't work as well. In the case of means tested programs, such as the high-tuition, high-aid model, data suggest that the benefits they have on paper are not maintained in practice. Universal programs might cost more money up front, but they're the best bet to ensure fairer access to quality higher education and a supportive social contract.