September 24, 1987
604 September 24, 1987 PROBLEMS WITH THE NEW STUDENT LOAN PILOT PROGRAM INTRODUCTION The cost of college education in the United States has been skyrocketing.
Secretary of Education William Bennett and others blame the student loan system which encourages colleges to increase fees This system, e lains Bennett, has many One is that it spreads subsidies thinly and widely, subsidizing students from relatively affl uent families, thus leaving insufficient support available to poorer students.
Another problem is that the repayment schedule for existing loan programs can be very onerous for graduates who enter low-paid professions year proposed a program to test a new form of student loan. Passed by Congress the measure permits the U.S. Department of Education to conduct a five-year Income-Contingent Loan (ICL) pilot program. The Administration hopes that the success of this pilot program will strengthen its argument t h at the entire student loan system should be remodeled along similar lines Tver Gains. Bennett praises the ICL pilot program as "the first major advance m federal student aid in the last two decades For Bennett, a key advantage to ICL is that it gives stud ents ''potential access to large amounts of capital under manageable repayment terms geared to post-graduate earnings."
Taxpayers also gain says Bennett, because "the un-subsidized ICL promises to restore financial integrity to federal student aid programs whose costs have soared out of control in recent years 500,000 per year to ten participating colleges for revolving funds to finance loans.
The $25 million program is to begin during the 1987-1988 academic year. The stated objectives of the ICL program, says Bennett, are to failings besides the burden it places on taxpayers and the e T fect it has on fees In an effort to reform the student loan system, the Reagan Administration last In the pilot program, the Department of Education will grant an average o f -2 1) make the student's loan qmyment burden manageable by allowing him or her as much time as needed to accommodate individual post-school income patterns 2) mahtain federal support for campus-besed direct student loan programs, thereby allowing instit u tional flexibility in packaging individual student aid awards 3) enhance institutional choice by authorizing substantially higher loan 4 put m long-term burden on the federal budget by requiring borrowers to As the name "income contingent" suggests, the p r ogram will permit payments amounts than permitted under current law; and make f? ll repayment at 3 percent above Treasury borrowing costs that are related to a graduate's income. Under the pilot ICL program, the student will be able to borrow a total of $ 17,5
00. The interest rate on the loan will be adjusted annually at a rate three points above the 90-day Treasury Bill rate.
Students will pay nothing while they attend college, although interest will begin to accrue as soon as the student takes out the l oan. When the student leaves school he must begin repaying the loan plus accrued interest, after a two-year period in which a monthly payment of only $50 will be required. The result: the loans will be somewhat more expensive to students-but less so to th e taxpayer--compared with other loan programs Attractme to Students Despite the higher cost, the Administration believes that the new form of loan will stdl be attractive to students. The main attraction is that repayments will be extended in a way to ensu r e that they will take no more than 15 percent of the borrowers' after-college income. Currently, payments are a fixed sum for a fixed period (10 years) under the Guaranteed Student Loan (GSL and National Direct Student Loan (NDSL) programs In theory, the i lot program is admirable. It will reduce the interest rate subsidy to students, F or whom a college degree is an investment leading to higher average incomes. It also will limit monthly payments for those students who choose less lucrative careers. This w i ll be fairer to taxpayers without a college education who currently must pay taxes so that students can obtain higher incomes As promising as the ICL appears, there 'is only a slim chance that the ICL e nder the pilot program, as now designed, it is unlik e ly that ICL loans will prove attractive to many students Illogical Choice. In fact, the ICL program likely will lead to a much higher default rate than the Department of Education projects, despite the repayment flexibility available under ICL. Moreover, t he reduction in subsidies will lead to significantly higher payments for most students choosing ICL, when compared with payments under other loan programs-even with the improved payment flexibility of ICL. Thus it would be illogml for most students to tak e part in the pilot program.
As such, the program will not test ICL across a typical cross section of American students. Hence the results of the pilot program wll provide little or no useful information on how a comprehensive ICL program would work ilot p rogram will be a fair test of how a full-scale ICL program would work. -3 The ICL pilot experience probably will play into the hands of ICL opponents.
The reason for this is that as long as the ICL pilot propam is voluntary and other highly subsidized loa n programs exist alongside it, there is no reason to suppose that there will be significant demand for it Bennett believes that decreasing student subsidies will give students an incentive to re-evaluate the financial return to higher education, and that t his in turn will put pressure on colleges to reduce costs to attract students. But this cannot happen through a small volunta pilot program, leaving other subsidies intact. It is only likely to come about if the government withdraws completely from the lo a n business and eliminates subsidies to all but a small minority of needy students. The Department should be focusing on this political task, not on an experiment that cannot work WILL THE ICL PROGRAM MEET m GOALS A major difficulty with investments in suc h "human capital" as skills and education is that they cannot be capitalized to serve as a source of collateral, unlike investments in buildings and machinery. This creates considerable risk for any institution lending money to individuals based on their h uman capital and expected The result, understandably, is that banks are much less inclined to lend to edtY stu ents than, say, to businessmen.
The ICL program tries to reduce the default risk by linking the repayment schedule to after-college income. on th e level of a student's future income, but with monthly payments no more than a maximum of 15 percent of his or her adjusted gross income. Presumably, this means that if post-schooling income is zero, then paymenis also are to be zero.
Interest on the loan, however, will continue to accrue and be converted to principal.
Subject to the 15 percent maximum, ICL payments are, on the other hand, to be adjusted for changes in interest rates to make students bear the risk of future financial market conditions. Un der rograms with fixed interest rates, it is taxpayers Loan payments under ICL would be contingent who bear the risk of future market g uctuati0ns.l shifting the Risks. Most individuals are risk averse, as demonstrated by the willingness of most home buye r s to pay a premium (higher interest rate) to avoid the uncertainty in payments under variable rate mortgages. Thus, since the ICL program not only decreases the interest subsidies to students, but also shifts the risk of changes in future interest rates f r om taxpayers to students, students will be less inclined to borrow.2 1. The Higher Education Amendments of 1986 phase in variable interest rates for existing loan programs 2. Besides the risk allocation as .ct of a variable interest rate, which is distrib u tional, there is also an should be allocated to investments with the highest rate of return. For exam le, if students had been allocative or efficiency argument !r or variable rates. From an efficiency standpoint, loanable funds char ed the T-biU rate plu s 3 percent in the early 1980s (see Table l it is e, ely that some students wo uf d have reevaluated the returns from college 4 Because of the limited market for unsecured investment loans in human capital, it is difficult to determine what would be an uns u bsidized market interest rate for student loans. A cautious estimate might be the interest charge on unsecured credit cards or personal notes An additional complication is that students are not required to start payments until they finish school, which ca n be five years or more after they take the loan. The potential repayment period moreover can stretch indefinitely into the future. This p0tent.d for prolonged repayment periods represents a radical departure from current loan programs, which limit the rep a yment period to ten years. Thus the distinction between short-term and long-term interest rates is much more important under the ICL program than under the current loan system The loan payments will be based on the 90-day T-bill rate lus 3 percent. To rev i ew the implication of choosing this rate, in contrast to rates P or longer term borrowing, Table 1 compares the 90-day T-bill rate with the 10-year Treasury security rate during the past 10 years TABLE! 1 Year 1976 1977 1978 1979 1980 1981 1982 1983 1984 1 985 Ave Consumer 90-Day Increase interest Price Index T-bill rate 05.80 percent 06.50 percent 07.70 percent 11.30 percent 13.50 percent 10.40 percent 06.10 percent 03.20 percent 04.00 percent Q3.10 percent 04.99 percent 05.27 percent 07.22 percent 10.04 p e rcent 11.5 1 percent 14.03 percent 10.69 percent 08.63 percent 09.53 percent 07.24 percent 07.16 percent 08.91 percent 10-Year Treasury interest rate 07.61 percent 07.42 percent 08.41 percent 09.44 percent 11.46 percent 13.91 percent 13.00 percent 11.10 p e rcent 10.75 percent 10.64 percent 10.37 percent Except 'for years of high inflation, the T-bill rate was significantly below the 10-year rate. The average difference over the 10-year period was 1.5 percentage points Thus, if the ICL program is intended to eliminate the interest subsidy, then a longer term rate should be used as the basis for setting the interest to be paid by the student This longer term rate more closely corresponds to the period. for which the federal government is obligated than does th e T-bill rate. Moreover there is an advantage in using the longer term rate, since it tells students what the market anticipates the future interest rates will be. Hence they can make a more informed choice about borrowing funds. -5 Hiding the Default Risk . The most serious problem with calling the ICL program an unsubsidized program, however, is the hidden risk of default. About 10 percent of all the dollars loaned out under the Guaranteed Student Loan (GSL program go into default. Even when loans to defau lting students are referred to debt collection agencies, there is a hal loss rate of approximately 4 percent.
Under the National Direct Student Loan (NDSL) program, the default rate is 15 to 16 percent of all money borrowed. The loss rate after the collect ion process decreases b less than 1 percentage point because of poor collection efforts by institutions d Si 'ficantly, the authority for the re lations to be issued for the ICL program based on default rates that are based on total money borrowed. But th e experience of the NDSL program suggests strongly that unless the ICL program results 111 a significant reduction in the loan default rate, the 3 percent interest rate add-on will not be sufficient to cover both the default rate and administrative costs I f this is the case, then contrary to claims by the Department of Education, the program will involve a subsidy to students comes Em the NDSL program. It is di fr 'cult to determine an annual risk premium WILL THE ICL PROGRAM DECREASE OR INCREASE THE DEFAUL T RATE default rate. On the plus side, the ICL program, compared with other federal loan programs, is advantageous to taxpayers in that it decreases the annual interest subsidy to students while in school, eliminates the risk of future market fluctuations a nd reduces the total taxpayer subsidy to students On the negative side, taxpayers face costs and risks because the repayment period under the ICL program is substantially longer. The interest subsidy each year is lower than for other loan programs, but th i s is offset by the longer period of subsidy, the larger loan amount available per student, and a possible higher default rate of ICLs relative to other loan programs subsidized loan programs, there are additional costs to students. But the student does en j oy some benefits. The student can borrow larger amounts than under other loan programs, has a wider range of choice among lending institutions, and has a more flexible repayment schedule conditioned on the level of future income The Advantage of Repaymat F lexibili Compared with other loan programs the larger ICL amount is insignificant, especially when the reduced subsidy is factored into the calculation? Thus, if there is a student advantage, it lies with the flexible repayment terms. Indeed, Bennett has e mphasized the ICL repayment The crux of the subsidy issue in the ICL program is the likely effect on the b To the degree that taxpayers benefit in the ICL program, relative to other 3. These figures are approximate and based on conversations with the U.S. Department of Education in the Higher Education Amendments of 1986 also increased per student loan amounts 4. under o er rograms. For example, a student can borrow $17,250 under a GSL-only $250 less than a ICL (S17,SOOf. -6 flexibility as the principal st u dent benefit. He claim too that repayment flexibility will also benefit t increased or extended over time ayers by reducing the default rate.5 But the default rate is likely to increase K w en the subsidies are reduced and individual loan payments are How will a student fare under the program? Table I in Appendix A indicates the repayment schedule under the terms of the ICL program. The calculations show that at an interest rate of 15 percent on the loan (12 percent T-Bill plus 3 percent the student would n eed an after-school income of $40,083 just to make the mterest ayments on the maximum loan and yet keep these payments below 15 percent of Kis income. An income below this figure means that the maximum permissible repayments under the ICL program would fa l l below the interest charged on the loan and so thestudents debt would grow. Moreover, the figures indicate that the longer repayment period does little to lower monthly payments. The annual income needed to pay off the loan over 30 years would still be $ 40,560, compared with $42,240 if the debt were to be cleared in 20 years college, was $18,118 in 19
81. Adjusting for inflation, the figure for 1986 was $21,9
07. Assuming a 5 percent rate of inflation, the average after-school income for graduates by 1993 will be $30,8
25. Thus under the ICL program, unless interest rates fall to unusually low levels a student would need to earn significantly above the average graduate salary to be able to make payments, subject to the 15 percent of income maximum rule, without falling deeper and deeper into debt.
Different Story. The average T-bill rate in the last 10 years was 8.91 percent Table 1 If this average is assumed for the next several years then the average interest rate on ICLs would be about 12 percent (T-bill plus 3 percent Those students who will earn at least $30,825 in 1993 (the average inflation-adjusted income based on the 1981 average income) would be able to pay off the principal in 20 years (see Table I, Appendix A Thus, for an average or better student income, and assuming interest rates in the future no higher than the last decade the ICL program would be manageable. But for students earning below the a v erage income, the story is different A students income does not have to fall much below the average at a 12 percent interest rate before payments are insufficient to cover even interest, let alone any principal. And if the prevailing interest rate were to be just 1 or 2 percentage points above the average rate in recent years, then the payments of many students would be insufficient to cover interest only there would be a greater probability that the student would default or declare The average gross incom e of males in the U.S age 25 to 29 with 4 years of If the principal owed increased, or the repayment period had to be extended 5. Address by William J. Bennett, United States Secretary of Education, The Future of Federal Student Financial Aid, sponsored by the Institute for Educational Mairs, Catholic University, Washington, D.C., November 19, 1986, pp. 7-11. -7 bankruptcy.6 Moreover, if payments, prove to be insufficient to cover the interest costs of the ICL program, it will not sustain revolving loan fun ds.
Interest rates of course, could remain at low levels, and the future incomes of college graduates could increase at a much higher rate than inflation. Yet even under these hishly favorable conditions the ICL program would not be attractive to students An rmplicit assumption of the ICL program is that its terms are more favorable to students than other loan programs. Two implications follow if this assumption is not correct. First, there may not be any student demand for an ICL if other loan programs ar e available. And second, if ICL repayment terms are more onerous relative to other loan programs, then the ICL default rate will be higher than existing loan prograqu. So the comparative terms of ICL and other student loan programs will have a significant i mpact on student attitudes to the new program HOW DOES ICL COMPARE WITH GSL The Higher Education Amendments of 1986 permit a student to borrow up to 17,250 under the Guaranteed Student Loan program--only $250 less than the ICL program Table II in Appendix B compares the payments faced by students under the two programs. The calculations show that only low-income graduates could expect lower monthly payments under ICL, with its higher interest rates but extended repayment period, than under the 10-year payo f f required under the GSL program, The claimed benefits of flexibility and extended payment schedule, in fact do nothing to make an ICL loan more attractive to the average student. Under all reasonable assumptions the graduate will end up paying more each m onth under the ICL program than under GSL, and for as many as 20 years longer Thus as long as other loan programs are available, it is unlikely there will be any significant demand for ICL. Those students with high expected future incomes certainly would n ot choose an ICL because their payments could be higher and the re ayment period shorter compared with a GSL. Indeed, if there is any demand at al P for ICL, it is likely to come from students with very low expected future incomes. This will result in adv e rse selection, such that payments to the program might not even cover interest costs, making it impossible to sustain a revolving fund higher than that of other loan programs since default rates can be assumed to be positively related to the amount and du r ation of monthly payments Table II in Appendix B also suggests that the ICL default rate is likely to be 6. There are costs dated with bankruptcy such as the loss of any credit for an extended period of time. It is possible that a payment of 15 percent of a student's post-schooling income is less costly than declaring bankruptcy even though there is no expectation of ever paying off the principal of an ICL 9 -8 The ob'ectives of the ICL program are laudable. But the analysis points to the futility o 1 offe r ing students one financial aid program when other more heavily Under almost any set of assumptions, the ICL de H ault rate would likely be higher subsidized programs are available. If students can be persuaded to take part in the ICL program then there mi ght be a tiny reduction in the federal deficit. But any taxpayer savings from the ICL program are contin ent upon student default rates than other federal loan programs.
Bennett believes that decreasing or eliminating loan subsidies will provide students w ith the incentive to reevaluate the return to higher education, which in turn will put pressure on higher education institutions to decrease costs. But this effect is likely to come about only if the government gets out of the .loan business or truly elir m nates subsidies in all loan programs As long as Guaranteed Student Loans and other highly subsidized loan programs are available, the ICL pilot program will provide little or no useful information regarding the impact on default rates or the deficit of ch a nging the entire federal loan program to a version of ICL In fact, the five-year pilot program will reveal very little, except perhaps that the students who enroll in it know little about financial principles L he ared for The Heritage Foundation by Ro er t J. Staaf Professor of Economics Clemson University 9 Repayments Under the ICL Program Table I indicates the financial situation a student could face under the ICL The table shows the princi al balance owed at the end of 7 years and the required possible i nterest rates charged on the loans. Column 2 shows the princi al balance owed at the end of 7 years for each interest rate, assuming the student g orrows 2,500 for the first and second year 3,500 the third year, and $4,500 in the fourth and fifth years fo r a total amount borrowed of $17,5
00. These are the permissible maximum amounts under the ICL program percent plus 3 percentage points) the student will have a princi al balance 24,030 principal balance and the $17,500 amount that the student actually bor rowed is the accrued intere~t.~ That is, the 6,530 difference is a measure of the additional taxpayer subsidy that is paid under other loan programs such as NDSL or GSL, where no interest is charged until repayments begin. Column 2 is the principal balanc e remaining at the end of 7 years with an assumption that the student pays the maximum $50 per month payment for two years upon completion of his five years of study, as specified under the ICL program retiring any of the principal, and column 4 represents the adjusted gross.income required by the student simply to pay the interestP Thus if the interest rate were 15 percent, the principal balance owed at the end of 7 years would be $40,083 the accrued interest over the 7-year period of $22,583 40,083-$17,50 0 and the post-schooling income nee B ed to make payments. Column 1 provides a range of For example, at an interest rate of 9 percent (that is, a T-bill rate of 6 obligation of $24,030 at the end of five years of study. The d' li erence between the Column 3 is the monthly payment necessary to pay the interest only, without 7. Interest was calculated b taking the compound interest on the $22,500 borrowed in the fist year for subsequent amounts borrowed in the third, fourth, and fifth years to arrive at the t o tal compounded interest for the five-year period 8. Secretary Bennett has stated that "Payments would begin nine months after aduation. For the the end of this two-year period, the ayments would become contingent on the borrower's Adjusted Gross Income fo r the prior year. dere would be no minimum payment, and year ayments would Address by Secret column 2 are an un 7 erestimate if the student is given a grace period of nine months after graduation in and not 5 years and 9 months. Note that the $50 per month pa Table 2. for five years, the compoun B interest on $2Joo borrowed in the second year for four years, and so on first two years they would be a fixed rate between $20 and $50, depending on t I e size of the loan. At never exceed 15 percent of the benefi c iary's income. Special deferments would still ge available nine-mont no g. grace period and assumes Y? t e student makes $50 per month payments at the end of 5 years on the principal balance owed at the end of 5 years regardless o r what interest rate is a ssumed in Bennett, November 19, 1986 p. 11 Thus, the principal balances calculated in yments are required, altho interest would continue to accrue. Column 2 ignores the ent is insufficient to cover interest 9. Adjusted Gross Income under the 1986 Tax Refo r m Act is likely to be similar except for adjustments such a alimon paid or IRA or Keogh deductions. The required income figure in column months to obtain an annual ayment, then dividing the annual payment by 15 percent (maximum 4 IS calculated by taking t E e interest-only monthly payment in column 3 and multiplying it by 12 percent of income required P or payment 10 TABLE I 7-YEAR mmAL BALANCE UNDER 1 INCOMES AND PAYOFF PERIODS Assumiog maximum loan of $17,!500 1 2 3 4 5 6 No 20-Yr. 30-Yr.
Principal Month ly Payoff Payoff Payoff Percentage Balance Interest Interest End of Payment Required Required Required Rate 7-Years Only Income Income Income 08.0 09.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 27.722 $29121 1 30,785 $32;446 $34,201 $36,055 $38,014 $40.083 4 2;271 $44,582 185 $219 $257 297 $342 391 443 $501 564 $632 14,785 $17,527 $20,523 $23;794 $27,361 $31,248 $35,480 $40.083 45;089 $50,527 18,560 $23,760 $26,800 $30,160 $33,760 $37,840 $42,240 47,040 $52,320 21,040 16,240 $21,600 $24,720 $28,160 $31,920 36 ,000 $40,560 45,440 $50,880 18,800 interest-only monthly payments would be $5
01. The required post-schooling income to make the interest payments would be $40,0
83. Note the sensitivity of principal owed at the end of 7 years (column 2) to the interest r ate (column 1 A one percentage oint change in the interest charged can change the principal owed by more than P 2,000 Column 5 calculates the income necessary for the student to pay off the loan in 20 years. For example, if the interest rate were 9 percen t , then a monthly papent of $263 would be required to cover interest payments and to pay off the nncipal balance of $29,211 in 20 years. The required income would have to be 21,040 .lo At an interest rate of 15 percent, the required income for a payoff in 20 years is $42,2
40. Column 6 is calculated in the same manner as columns 4 and 5 except that a 30-year payoff is assumed differ considerably when the payoff period is extended from 20 years to 30 years, or to an indefinite period for that matter. This is especially true at high interest rates.
For example, at an interest rate of 15 percent the required income for interest payments only is $40,083, while a 20-year payoff requires an income of $42,240 and a 30-year payoff an income of $40,5
60. Thus the f lexibility of extended payments It should be noted that the required incomes in columns 4, 5 and 6 do not 10 263 X 12) divided by 15 percent 11 has erhaps been overestimated by the Department. As columns 5 and 6 in Table difference in payments and the req u ired incomes to support these payments I1 ilustrate P a 50 percent increase in the allowable payoff period makes very little Thus Table I indicates that the ICL program would require students to earn fairly high post-schooling incomes in order to pay the i nterest, without even reducing the ICL pmcipal. If the required income is not earned, then the unpaid interest will be added to the principal, making the principal owed by the student grow even larger. The average gross income of males in the U.S age 25 t o 29 with 4 years of college, was 18,118 in 1981.11 If this figure is adjustedfor inflation, then the average gross income for 1986 would be $21,907.12 If the inflation rate is assumed to be 5 percent per year for the 1987-1993 period, then the average inf lation adjusted income for such students could be assumed to be $30,825 in 19
93. Table I1 (column 4) indicates that the required post-schooling income merely to pay interest at the end of 7 years and a 13 percent interest rate is $31,2
48. Thus if the ICL interest rate were to be 13 percent or more, or a students actual income in 1993 turned out to be less than the 1993 inflation-adjusted average income of 30,825, the ICL principal would increase indefinitely as long as the students pos t schooling income continued to be insufficient to support interest-only payments 11. Money Income of Househol Families, and Pensons in the United States, 1981, Bureau of Labor Statistics, 1985, Table 48, p. 164 12. The 1981 mean gross income for the expan ded age groups of 25-34 males was $20,4
68. If the same inflation adjustments are made, then the 1993 mean gross income would be $34,8
25. As Table 11 illustrates, at high interest rates, the required income may exceed $34,825 12 APPENDIX B Payments Under ICL and GSL Table II compares the Income-Contingent Loan (ICL) program with the Guaranteed Student Loan (GSL) program, assuming that the amount borrowed is 17,500 under each program and that the GSL qualifies for federal interest subsidies. The first col u mn in Table 11 is the assumed interest rate. Column 2. is the monthly payment on a GSL of $17,500 over a 10-year pay-off period (the maximum repayment period for a GSL is 10 years Because interest on a GSL is subsidized while the student is in school, the pMcipal balance owed at the end of the fifth year is the same as the amount borrowed. Column 3 is the monthly payment on a ICL over the same 10-year period if the student were to start payments at the end of the fifth year. The higher monthly ICL payment in column 3, compared with the GSL payment in column 2, is due solely to the absence of interest subsidies while the student is in school. All other terms are identical.
Interes Rates 8.0 9.0 10.0 11.0 12.0 12.5 13.0 14.0 15.0 16.0 17.0 TABLE3 II A COMPARI SON OF GSL AND ICL MONTHLY PAYMENTS 2 3 4 5 6 Commencing payments Commencing payments after 5 Yrs. after 7 Yrs 10 Yrs 10 Yrs 10 Yrs 20 Yrs 30 Yrs payoff GSL ICL ICL payoff ICL payoff ICL Monthly Monthly Monthly Monthly Monthly Payment Payment Payment Paym e nt Payment 212 $222 231 241 $251 $256 $261 $272 $282 $293 304 284 304 $325 348 $371 $383 $396 $422 $450 $479 510 336 370 407 $447 491 $514 538 $590 647 $708 $775 232 $263 $297 335 $377 $399 $422 $473 $528 $588 654 203 235 270 $309 352 $375 $399 450 $507 5 6 8 $636 The Department of Education argues that the advantage of the ICL to students is that payments are contingent on future incomes, thereby making payments less onerous than the arbitrary 10-year payoff required under GSL. But this advantage exists onl y if incomes are low. For example, consider a 9.0 percent interest rate. The rnonthl payments under GSL are $222 over 10 years, whereas the monthly payments un B er an ICL are $304 over the same 10-year period. A 13 student with a post-schooling income of $ 17,760 upon graduation thus could make payments of 15 percent of his income or $222 per month under GSL while payments of $304 per month would be required under an ICL. The savings to the student of a GSL over an ICL.for the same payoff period are $9,840. n Now consider the claim that the student benefits from the extended repayment eriod. Under the ICL propam, column 4 assumes the student pays the maximum O per month after graduauon for two years and then makes monthly payments based on the principal owed a t the end of 7 years for a period of 10 years (that is 50 er month for 2 years and then the monthly payment in column 4 for 10 years Starting at an interest rate of 12.5 percent, the ICL payment proves to be more than double that of a GSL payment. Thus th e benefit of aying $50 per payment for the remaining ten years. Even those students expected to earn a low mcome upon the completion of school may be at a disadvantage with a ICL Suppose the interest rate was 9.0 percent, for instance, and the student made IC1 payments of 15 percent of his income, or $370 (column 4) per month at the end of 7 years for a 10-year period. The required income would be $29,600.l But a student with GSL payments of 15 percent of his income, or $222 er month income Thus, the ICL be n efit of only having to pay $50 per month for two years comes at the expense of the ICL student hawng to earn considerably more than a GSL student for the same 10-year payoff period. The claimed flexibility of an ICL does not seem to benefit the student un d er these circumstances month for the sixth and seventh year contributes to a doubling o F the monthly column 2) for 10 years would only require an income of $17,760 or $11,840 less Now consider the more extended repayment periods under the last two column s , which list the ICL monthly payments on the principal balance owed at the end of 7 years with 20- and 30-year payoff respectively. For exam le, at a 9 percent interest rate, the monthly payment for a 20-year payoff is P 263 and $235 for a 30-year period. Even with an extended payoff of 30 years, the students monthly ICL payments exceeds that of GSL payment 222 with a 10-year payoff. So the ICL student would have to earn more than the GSL student with payments based on 15 percent of income, make higher pay m ents, and make three times as many getween the ICL and GSL increases with an increase in the interest rate Thus under almost any set of assumptions, a student who chooses an ICL over a GSL is going to be worse off ayments. Moreover, as Table II indicates, the required income and disparity 13. All the calculations in Table II in Ap ndix B ignore the 9-month grace riod after graduation owed under a ICL actually is underestimated because interest would accrue into principal during the grace period 14 370 X 12 ) divided by 15 percent 29,600 15 222 X 12) divided by 15 percent 17,760 when payments are not required under eit g. er the GSL or ICL. This means t g. at the prinupal balance