Federal aid to college students is on the fiscal chopping block again. This is an old story; the Reagan administration has repeatedly sought cuts in student aid, even though Congress, with the exception of changes enacted in 1981, has refused to go along.

But the 1986 budget proposals are the harshest yet. They set a maximum of $4,000 on the total amount of federal aid a student could receive, establish a $32,500 income ceiling for students and their families seeking Guaranteed Student Loans, and set a $25,000 income ceiling for students seeking grants. Coupled with other proposals, these changes would reduce the number of student aid awards by more than 1 million in 1986 alone. That's a decrease of over 10 percent.

Education Secretary William Bennett's strong defense of the proposals -- and the furious response -- have generated much heat but little light. The reason is not hard to find: Both parties are right. Federal student aid has served countless deserving young people and the nation well. But the system is in disarray and desperately needs reform.

Part of the problem is simply design. The major student aid programs are incredibly complicated. Few campus officials understand how the programs work, and students and parents find them a nearly incomprehensible maze.

The design problem exists because there is no coherent philosophy about student aid. Originally, the centerpiece of federal efforts was grants for low-income students. In recent years emphasis has shifted to Guaranteed Student Loans -- which frequently benefit the middle class. Loans for everyone have become more important than grants for the poor.

Cost is also a critical problem. Last year Congress set aside nearly $8 billion for student assistance, an increase of 20 percent from the previous year. It wasn't enough. The best estimates are that a supplemental appropriation of nearly $1 billion will be needed.

Finally, there is some waste and abuse involved. In 1985 the government will spend $800 million to pay off defaulted loans. And, some student aid is misused. Nobody knows how many student loans buy motorcycles and stereos, but nobody doubts that some of them do.

But the Reagan proposals are not the answer. The proposed ceilings would have a devastating effect on the students who attend the nation's private colleges and universities. Historically, federal aid sought to treat public and private institutions equally, and President Reagan's ceilings would break faith with the nation's private schools.

As well, reducing aid by 1 million awards is not a trivial matter. According to a recent study by the American Council of Education, half the nation's college students get federal aid; 75 percent of private- school students do. The government cannot walk away from them as cavalierly as some might wish.

Finally, student aid remains a major public concern. Most Americans regard higher education as a vehicle of social progress and economic growth and want the federal government to help students finance it. Politicians seeking deep cuts in these programs will find little public support.

But cost pressures, complexity and an enrollment decline mean that major changes will be made, sooner or later.

Instead of replowing old ground, what might be done to prepare for the future? First, in light of enrollment declines, we must recognize that the structure that worked well in the past will not work in the future. The current system is not designed to contract.

Second, we must reexamine fundamental purposes. Students do not have a "right" to go to Harvard, and the federal government has no obligation to pay for a Harvard education. But the federal government has established a system that enables them to go, and there is strong support for maintaining this principle. This country -- founded on the idea of a natural aristocracy of talent -- has attempted to put the world's best higher education programs within the reach of all.

The solution to the problem raised by the president's budget is not conceptually difficult. Fairness and simplicity are its hallmarks. Two basic elements are essential. One is an enlarged and more generous grant program for the very poor. There is simply no better social investment than the education of poor, ambitious young people. We owe it to ourselves if no one else. The second is that all students' eligibility for Guaranteed Student Loans should be based on income. As it pres the program is both a loan of convenience and an entitlement. It should be neither.

A more ambitious and more difficult solution is to replace the whole loan structure with a National Student Loan Bank, which would issue loans based on financial need, with repayment determined by the borrower's income after graduation. Students would pay off the loans out of future income, collected by the Internal Revenue Service as it collects income taxes.

In an income-contingent system, low earners would pay less, high earners more. A loan bank is an old idea, supported -- in theory -- by many. A major barrier to its creation is the high cost of creating it.

Where might billions of dollars be discovered to start a National Student Loan Bank? One solution is a special federal bond issue. However, given the deficit, further federal borrowing is both unattractive and unlikely.

But the federal government could loan money to the bank from a specially appropriate source, the surpluses building in the Social Security Trust Fund. Thanks to the 1983 rescue package, according to The Wall Street Journal, by 1990 Social Security will have an accumulated surplus of nearly $180 billion dollars. By the year 2000 the surplus will exceed a trillion dollars. Currently the surplus is used to float the deficit; it is certainly more sensible for the Social Security Trust Fund to support the education of tomorrow's taxpayers than to underwrite today's government spending binge. The intellectual and political symmetry of using Social Security surpluses to help establish a Student Loan Bank is most appealing.

Regardless of the revisions eventually enacted, changes in student aid are essential. With limited resources, the federal governmen can assist a small number of students in a big way or a large number of students in a small way. The complex web of programs now in place attempts to do both, and does it very poorly. In this lies the fallacy of trying to fine-tune the present system. It is time to go back to the drawing board.