By Richard Vedder

(Bloomberg News) U.S. employers complain that they can't find enough skilled employees. Then how do we explain why almost 54 percent of recent college graduates are underemployed or unemployed, even in scientific and technical fields, according to a study conducted for the Associated Press by Northeastern University researchers?

The cause is more fundamental than the cycles of the economy: The country is turning out far more college graduates than jobs exist in the areas traditionally reserved for them-the managerial, technical and professional occupations.

The Bureau of Labor Statistics tells us that we now have 115,000 janitors, 83,000 bartenders, 323,000 restaurant servers, and 80,000 heavy-duty truck drivers with bachelor's degrees -- a number exceeding that of uniformed personnel in the U.S. Army.

Was college worth it? A huge part of the problem relates to federal financial-aid programs. Annual student loans, Pell Grants, tax credits and other federal assistance totaled some $169 billion a year in 2010-11 -- more than 1 percent of national output. These programs are based on two erroneous premises: that almost everyone needs higher education for vocational success, and that they reduce student costs.

More than 25 years ago, Education Secretary William Bennett argued that federal aid programs benefited colleges more than students. Recent studies by Stephanie Riegg Cellini of George Washington University and Claudia Goldin of Harvard University, as well as by Andrew Gillen for the Center for College Affordability and Productivity, support that hypothesis. A new study by Nicholas Turner of the Office of Tax Analysis in the U.S. Treasury Department argues that when tax-based aid goes up, institutional scholarships go down, dollar for dollar.

Underqualified Borrowers

Consequently, we have millions of underqualified college students borrowing or getting Pell Grants to finance college.

More than 40 percent of them don't even graduate within six years, and many who do have marginal academic records. Because the average college student spends fewer than 30 hours a week on all academic activities, for about 30 weeks a year, never have so many dollars gone to teach so many students for so little vocational gain.

Besides leading to more underemployed college students of increasingly dubious academic quality, the dysfunctional federal student financial assistance programs have other pathologies:

First, universities, unlike the taxpayers, suffer no financial consequences when the underqualified students they have lured into their academic programs ultimately default on their loans.

Second, students who study six years but ultimately drop out receive more financial aid than the diligent "A" student graduating in three years: We reward mediocrity and punish excellence.

Third, there is no adjustment of student-loan interest-rate terms to meet market conditions or differing risk factors relating to individual repayment prospects. That means too much money is lent, especially to high-risk individuals with little prospect for academic success.

Fourth, the Free Application for Federal Student Aid form, associated with these programs, aside from being unbearably complex, gives colleges private information about family finances that allows them to gouge students more.

Fifth, colleges' tuition and fee policies drive the amount of loan volume, rather than the other way around, thus contributing to the college-cost explosion and the subsequent academic arms race.

Sixth, intended partly to promote greater opportunities for the poor, these federal-aid programs have been accompanied both by rising income inequality in the U.S., and a decline in the proportion of recent college graduates from poor families.

Undeveloped Credit

Proponents of federal student-loan programs argue that private student-loan markets are underdeveloped, that banks are afraid to lend to students, largely because of their lack of credit history. This argument is vastly overblown. It is amazing how students have no trouble getting credit cards and racking up debt, or little difficulty borrowing to buy a car. Why would college be any different?

Yes, the goal of providing educational opportunity for all seems commendable. Any revamping of the federal student- assistance program would have to be phased in to avoid severe hardship and enrollment disruptions. But here are some better policies:

-- The federal government should get out of the student loan business.

-- It should provide educational vouchers (similar to Pell Grants) directly to students (not schools), and make those vouchers progressive (very low-income students receive the most, fairly low-income students a little, and middle- and upper- income children nothing).

-- Add performance incentives, rewarding timely degree completion and good performance.

-- Remove the tuition tax credit that largely assists relatively affluent students and their families; perhaps use savings from all of the above to reduce the budget deficit.

-- Eliminate the Free Application for Federal Student Aid form and require that applicants give the Internal Revenue Service permission to provide family-income data.

My guess is that the total number of students attending four-year programs would fall modestly, a good thing given the disconnect between the labor market and college enrollment; that the proportion of students from lower-income families would probably increase (also good) both because the Free Application for Federal Student Aid form is a barrier for lower-income families, and the burden of aid reductions would fall mainly on the colleges and more affluent students.

Also, the total cost to the federal government would drop significantly.

More radical solutions might involve rolling many government-income security programs into compulsory tax- sheltered 401(k)-like lifetime individual security and investment accounts, allowing withdrawals for college costs. However it is done, the current system needs replacing.

(Richard Vedder is director of the Center for College Affordability and Productivity and teaches economics at Ohio University. The opinions expressed are his own.)