As Congress debates the reauthorization of theHigher Education Act, it - PDF document

Routing As Congress debates the reauthorization of theHigher Education Act, it should heed FriedrichHayekÕs warning that democracy is Òpeculiarlyliable, if not guided by accepted common princi-ples, to produce over-all results that nobody want-ed.Ó One result of the federal governmentÕs stu-dent financial aid programs is higher tuition costsat our nationÕs colleges and universities. Basic eco-nomic theory suggests that the increased demandor higher education generated by HEA will havethe effect of increasing tuitions. The empirical evi-dence is consistent with thatÑfederal loans, Pellgrants, and other assistance programs result inhigher tuition for students at our nationÕs collegesand universities.The diversity of objectives, resources, andtypes of governance among the thousands of col-leges and universities makes it difficult to ade-uately measure the exact amount by whichtuitions rise in response to federal student assis-ance. Therefore, estimates of the amount vary inthe literature. Congress can at best know that itspolicies increase tuitions and that some portionof the federal assistance ends up being capturedby state governments and by the colleges anduniversities.to rely on the federal government to fund theirhigher education, and the federal governmentuses this financing to affect the behavior of stateand private institutions, we should be concernedabout how the resulting loss of independence ofour colleges and universities affects the ability ofters to form opinions about public policy thatare independent of the governmentÕs position.Congress should consider a phase-out of federalassistance to higher education over a 12-yeartime frame. As the federal government removesitself from student assistance, we should expectseveral things to happen. First, sticker tuitionprices should decline. Second, the private marketshould respond to the phase-out of federal assis-ance. That response would likely take threeimportantly, the expansion of human capitalcontracts. Human capital contracts, first sug-gested 40 years ago by Nobel Laureate Miltonriedman, would allow students to pledge a por-tion of future earnings in return for assistance inpaying their tuition. Making College More ExpensiveThe Unintended Consequences of Federal Tuition Aidby Gary Wolfram Gary Wolfram is George Munson Professor of Political Science at Hillsdale College in Michigan.Executive Summary No. January 25, 2005 riedrich Hayek warned in of Libertyliable, if not guided by accepted commonprinciples, to produce over-all results thatAs Congress readies itselfAct, it should decide what the purpose of theact is and whether its component programsare accomplishing their intended goals.ax credits and deductions and other pro-grams such as the GI Bill, has over the yearsmoved the federal government further andfurther into the higher education marketwith insufficient debate over the goals of pro-grams, their effectiveness in meeting thosegoals, and the unintended effects on tuition.Direct financial aid began as a way of pro-viding benefits to World War II veterans whohad been seriously underpaid, then moved togrants to low-income students to expandaccess to higher education, and now includesax credits to help middle- and upper-incomeparents face the cost of high tuition. The netresult is a mixture of programs that may haveresults that Congress never contemplated. Inparticular, there is a good deal of evidencesuggesting that federal financial assistancehas the unintended consequence of increas-ing tuition for all students. Federal aid mayalso result in a reduction in aid by state gov-ernments to students who attend universitiesin their state and a reduction in state appro-priations to public colleges and universities.Individual colleges and universities may alsoreduce their internal financial aid when theirstudents receive federal aid. In addition, fed-and universities, both public and private.Each institution has its own organizationalstructure and operates under different objec-tives and constraints. As a consequence, thereis no single model of the effects of federalgrants, loans, and tuition tax credits on col-leges and universities. Nevertheless, both the-oretical analysis and empirical evidence indi-cate that the federal governmentÕs financialaid programs cause higher tuition costs,college and causing others to attend a collegethat is not their first choice. Basic economictheory suggests that the increase in demandor higher education brought about by thesystem of grants and loans will increase theThis year, Congress should consider theeffect of federal tuition aid on college costs asshould phase out the current federal assis-ance program over a period of 12 years andallow the charitable sector to provide assis-ocus on developing the legal structure thatould allow for a system of human capitalobel Laureate Milton Friedman more than40 years ago and whose time has now come.The System of Financial AidThe role of the federal government insince the Founding. George Washingtonthought that we should have a national uni-ersity.The federal government helped estab-lish land grant universities in the 19th centu-. Although we have no national universitytoday, and the land grant universities are nowprimarily associated with the state of theirlocation, the federal government is heavilyinvolved in higher education, subsidizingattendance and using incentives to affect thebehavior of colleges and universities. It doesso with grants, loans, work-study programs,and tax credits.The Bush administrationÕs 2005 budgetould provide $73.1 billion in overall federalnancial aid to students through the Depart-ment of Education under HEA, an increase of$4.4 billion over the 2004 level. The number ofstudents receiving assistance through grants, theory suggeststhat the increasein demand forbrought about by the system of grantsof higher clear is what the different effects are on vari-ous income groups by type of aid. As Harvardprofessor Susan Dynarski has recently point-ed out: ÒThe effect of a given subsidy may varyacross groups due to relative differences inaccess to information, the form taken by theIn this and a second paper, Dynarskiexamined the effect of legislative changes inaid programs and found that, consistent withthe other literature, financial aid increasesThus, while it is notknown to what extent federal financial aid ismeeting the goals of the different programs,what is clear is that federal financial aid hasincreased demand for college. Basic economictheory shows that this increase in demandwill cause the unintended consequence ofempirical question is simply: how much doesThe Effect of Federal Aid onpressure to expand the HEA and provide taxcredits and deductions for higher educationcosts, which leads to further expansion ofHEA and use of the tax code to affect taxpay-er behavior. This is the type of cycle thatHayek and Ludwig von Mises suggest hap-pens when government acts outside of itsfundamental role and fails to take intoaccount how the market works.Before HEAbillions of dollars on financial aid, it shouldbe fairly certain of the effect of this spending.The amount of federal (as well as state) aidhas grown substantially over the past decade. ns of billions of dollars oncertain of theeffect of this 0.02.06.08.0 1996199719981999200020012002200320042005 Source: U.S. Department of Education, FY2005 Budget Summary.Ž Note that, for technical reasons, aid available tostudents will differ somewhat from appropriations. Figure 1Pell Grants and Numbers of Applicants Year 0.02.06.08.0 1996199719981999200020012002200320042005 iations (billions of dollars) Valid Applicants (millions) Source: U.S. Department of Education, FY2005 Budget Summary.Ž Note that, for technical reasons, aid available tostudents will differ somewhat from appropriations. Figure 1Pell Grants and Numbers of Applicants Y tuitions, then the aid will not have much effecton tuition.The observant reader will notice thatgiven a downward sloping demand curve, theonly way that tuition would not increase is ifthe supply curve were perfectly horizontal,which would indicate that as a whole collegesand universities expand to accept more andmore students at the given market price forhigher education. That does not seem to be areasonable assumption, as anyone who hasbeen rejected by a college will attest to.tunately, there is little literatureregarding the elasticity of supply, or theresponsiveness of institutions of higher edu-due to the lack of an accepted theoreticalmodel of university behavior. As Judith Linoted recently, Òvery little is currently knownabout the objective functions of higher edu-trations and boards of universities are.However, there has been some recent work indeveloping a model of university behaviorsupply curve for higher education services isply is likely to vary by institution.Journal of Economic Perspectivessymposium on higher education, GordonWinston built upon the earlier work of econo-mists Michael Rothschild and Larry White toprovide an interesting discussion of the type ofmarket within which universities operate.paper emphasizes the combined role of univer-sities as charitable organizations that rely ondonations for support and as producers of aproduct in which students are both consumersand inputs. In other words, the quality of thestudent body affects the quality of the educa-tion provided. The implications of his modelare that the elasticity of supply will vary acrosstypes of institutions, from highly inelastic atelite private universities such as Harvard to rel-atively elastic at for-profit non-elite institu-tions such as the University of Phoenix.In their 2002 paper, Dennis Coates andBrad Humphreys use a model of bureaucracyto provide an empirical estimate of the supplyof university enrollment in 11 Maryland uni-Their estimate for price elasticity is.94 for one specification of the model and.97 for another. What this means is that forevery one percent increase in tuition, thenumber of places in those 11 universitiesincreases by between 1 and 2 percent. Thesupply curve for these universities, at least, isdefinitely upward sloping, so any increase indefinitive measure of the elasticity of supplyis not well established, there is no evidence,either theoretical or empirical, that the sup-ply would be perfectly elastic.that at least some portion of the increase inderal financial aid that has occurred overthe years has ended up in the form of higherDetermining the Effects of deral Student Assistance on TuitionThere are several situations that compli-cate the issue of determining the amount bywhich federal aid increases tuition. First, notall students are eligible for the various grantsand loan programs that exist. Thus, some stu-dents will receive no subsidy but will be affect-ed by the higher tuition. Second, the demandand supply for colleges is probably quite het-erogeneous. Students may have inelasticdemand for some colleges and elastic demandr others. For example, the demand for slotsat Harvard may be inelastic, while thedemand for slots at Adrian College may beelastic. The supply of slots at the University ofennsylvania may be inelastic, while the sup-ply of slots at Lake Superior State Universitymay be very elastic. Thus, federal assistancemay be passed on to some colleges and uni-ersities and not to others.Another factor that makes it difficult todetermine the final effect of the subsidy isthat colleges and universities may reducetheir own internal financial aid in response to At least some portion of thefederal financialoccurred over theyears has endedup in the form of funds, according to some New York offi-cials. Theodore Hollander, deputy com-missioner of education for New York atthe time, explained that if tuition weremade necessary by the fiscal crisis, it wasmade feasible by New YorkÕs large statestudent-aid program and the rapidlyexpanding federal Basic EducationalOpportunity Grant Program.Stampen also points out: Assistance Act) of 1978 is there anystatement providing guidance or estab-lishing objectives regarding institution-al tuition level vis-ˆ-vis federal studentfaces the choice of maintaining tuitionsat the lowest possible level or of raisingtuitions to ÒharvestÓ the federal studentaid as an indirect institutional sub-sidy.Congress has recently taken note of this,as a number of House Republicans proposedlegislation that would have made collegesand universities that raised tuition toosteeply ineligible to receive work-study dol-lars and other federal grants.bill was later withdrawn, it drew furtherattention to the possible link between finan-Empirical Evidence of theEffect of Federal FinancialAid on TuitionMost empirical studies of the effect ofnancial aid have focused on its effect on stu-As discussed above, this lit-that federal financial aid increases the demandthat one result of these federal programs is anincrease in tuition. However, as Judith Lipointed out in her 1999 study of the effect ofell Grants, Òthere have been surprisingly fewstudies on the impact of federal financial aidon college tuitions.ÓA comprehensive search of the literaturecolleges and universities to capture for them-selves federal student aid in the form of highertuition or reduced in-house assistance. Ofthese ten, eight found evidence that federal aidshowed up in higher tuitions or smaller insti-tutional scholarships. One of the two thatund no evidence of tuition increases admit-ted to the poor explanatory power of its modeland the other looked only at large publicresearch universities. Some of the authors inthe eight supporting studies found evidence ofpublic institutions increasing tuition, whilethers found that private institutions in-creased tuition. The complications of identify-ing what portion of change in tuition is due toederal aid, the variety of models, differences inthe data being used, and the varying ability ofstate governments to capture some of the aidby lowering appropriations to public universi-ties, are no doubt factors that result in differentestimates of the magnitude of the effect of fed-eral aid. However, there is enough evidence toconclude that federal aid has been a factor inrising tuition in our higher education system.Judith Li found, using data on individualstudents and institutions, that private four-more than two dollars for each dollarincrease in Pell Grants, and public four-yearcolleges increased their listed tuition by 97cents for every dollar increase.She foundthat public four-year institutions were able toincrease net tuition by 68 cents for every dol-lar of Pell Grant increase, while private four-ear institutions raised their net tuition by 60cents. That means that both public and pri-vate colleges and universities actually raisedtuition by more than the amount of the PellGrant. Li did estimate that public two-yearinstitutions decreased tuition, 17 cents fornet tuition and 18 cents for list tuition, forevery dollar increase in Pell revenue. She alsonoted that since tax credits might be more Private four-yearlisted tuitionprices by morethan two dollarsincrease in PellGrants, and public four-year their listedtuition by 97cents for every NCES foundthat when federal aid was subtracted fromtuition costs adjusted for inflation, there wasstill a real increase in net tuition costs over thegraduate borrowing increased substantially.This is consistent with the tendency forincreased financial assistance to result in high-er tuition. The study also found that nettuition did not increase once all grants, includ-ing state and institutional aid, was accountedtice that this is still consistent with ris-ing tuition as a response to increased federalaid. Suppose federal aid increases by $8 andtuition rises by $10. If the university increasedits in-house assistance by $2 to offset some ofthe tuition increase, there would be no nettuition increase, even though the federal aidincreased tuition. Some students may be pay-ing much higher tuition costs as others receivemore state and institutional aid. The study alsoound that costs such as room and board rosesufficiently that the total net cost of atten-dance increased after all aid was considered.Singell and Stone, in their 2003 study ÒForWhom the Pell Tolls: Market Power, TuitionDiscrimination, and the Bennett Hypothesis,Óused data on 71 public and private universitiesto look at the effect of Pell Grants on tuition.They found that the top-ranked private uni-ersities increased net tuition (list tuitionminus institutional aid) by $3.96 for each dol-lar in Pell Grant. They did not find evidencethat public universities or lower-ranked pri-vate universities capture Pell Grants in theorm of net higher tuition.The National Center for Education Statis-tics devoted a chapter in their two-volumestudy on college costs to the effect on tuitionof federal financial aid.Examining institu-tional-level data, the studyÕs authors usedregression analysis in an attempt to measurethe effect of various aid variables on tuition.While the models they employed did not finda significant statistical relationship betweenederal aid and tuition levels, the models didnot do a very good job of explaining changesin tuition. In fact, the authors themselvesnoted that the lack of explanatory power ofplained than explained by the models.ÓIn a 2003 NBER working paper, MichaelRizzo and Ronald Ehrenberg examined theresponses of 91 public research universitiesto changes in state appropriations and feder-al financial assistance over the period1979Ð1988. The authors were primarily con-cerned with the enrollment of out-of-statestudents as a revenue source.of their study, they did not find that increas-es in Pell Grants led to increases in tuition atthese universities. There are at least twocaveats to this result. First, these flagshipuniversities may behave differently fromther colleges and universities, such as two-ear colleges or private four-year colleges.And second, it may be that the states are cap-turing the federal subsidy in the form of low-ered appropriations by states for their majorpublic universities.Ehrenberg and two colleagues found in anearlier paper, ÒHow Would Universitiesespond to Increased Federal Support foring universities respond to changes in thestudents supported on federal funds by reduc-ing the number of graduate students support-ed by internal funds.Such behavior is con-sistent with a scenario in which the subsidy isused to displace spending that would other-wise have occurred by either the university orthe state government. This may explain whysome of the studies mentioned earlier werenot able to find a statistically significant rela-tionship between federal student aid andtuition levels. Some universities may be cap-turing the aid in the form of reduced institu-tional aid rather than higher tuition.In summary, studies that have focused onthe impact of federal student aid on tuitiongenerally find some effect, whether on publicuniversities, private universities, or both. Thedegree of effect is no doubt due to a numberof the complicating factors discussed above.Some universities and colleges have fairlyinelastic supply, some have inelastic demand,some may keep listed tuition the same and of trustees of types of publicand private different sets of mentioned. In recent testimony before theSubcommittee on 21st Century Competitive-and the Workforce, F. King Alexander, presi-dent of Murray State University, commented,ÒIronically, federal programs in totality giveincentive for institutions to increase tuitionand to set high sticker prices.ÓAlexandershowed that he is aware that other institutionshave been able to increase tuition when theederal government increased student aid: or have we (Murray State) opted todramatically shift the educationalcosts away from the state and to theederal government indirectly throughthe student by inflating tuition likemany higher cost states and institu-tions have done over the last twoIn an interview with USA Today, Alexanderwas more direct: ÒIt is a shell game, pure andsimple. A lot of schools set tuition prices tomaximize grant money and then use institu-tional (financial) aidÑwhich isnÕt real moneyÑto set the real tuition.Óition Rising: Why College CostsSo Muchat Cornell University, discusses the tradeoffbetween university internal aid and federalthe slowed growth in federal aid resulted inincreased financial aid costs for Cornell, theimplication is that increased federal aid willresult in lower university-provided aid byCornell, similar to the Bowdoin College expe-Other Reasons for Rising There are, of course, a number of other rea-sons for rising tuition. In addition to othering higher education has been increasing.William Baumol, in his famous 1967 paper,productivity in certain services, in particularteaching, that will inevitably cause the cost ofproviding these services to increase relative tother goods and services.Howard Bowenand hypothesized that universities will, ingeneral, spend all available revenue as theyseek to gain prestige and influence.recently, Ehrenberg used his experience as asenior administrator at Cornell University todiscuss the various reasons why the cost ofhigher education is rising and what the effectThese costs include competi-tion for students that results in higher tuitionulty salaries, new advanced equipment forresearch, and so forth. My experience as atrustee for a state university has shown howreduced state appropriations, higher healthcare costs, and increased utility costs can drivethe decision to raise tuition. However, therecan be little doubt that federal aid has, at aminimum, allowed colleges and universitiesto increase tuition beyond what would other-A Threat to Higher EducationIn addition to raising tuition costs, govern-ment tuition and institutional aid threatensFormany Americans, the point at which theybegin to develop their opinion about the roleof government and possible solutions to pub-lic policy problems is in college. For democra-cy to work, it is important that the institutionsthat educate those who will participate in thedemocratic system be truly independent of thegovernment. As Friedrich Hayek points out inConstitution of LibertyThe conception that governmentshould be guided by majority opinion Search as onemight throughone will not findthe power to granted to the to higher education. Certainly HEA results inhigher list price tuitions for all students andhigher net tuitions for an unknown number.university, student, and time, as demand andsupply conditions differ. The lack of certainty in the results of thisprogram, other than increasing tuition, sug-of federal assistance to higher education overa 12-year time frame. This would allow peo-ple who have made decisions under the cur-rent system to continue with the same basicstructure while the current program is beingphased out. It would allow colleges and uni-ersities time to respond to the removal ofederal government interference in the high-As the federal government removes itselffrom student assistance, we should expect sev-eral things to happen. First, we would expectsticker tuition prices to decline. Second, theprivate market would respond to the phase-out of federal assistance. This would likelythree forms: additional private-sectorloans, additional private scholarship funds,and perhaps most importantly, the expansionof human capital contracts, which are similarto owning stock in the future earnings of a col-greater role in providing student loans.onfederal borrowing has increased from 7percent to 16 percent of education loan vol-ume over the past five years. Nonfederal bor-rowing reached $11.3 billion in 200304, up39 percent in real terms over the previousear and almost 150 percent in three years.It is quite likely that this market will becomemore and more developed over time and willbe a normal place for students to look forassistance.It is likely that private assistance for high-of HEA. Because federal programs affect theprivate market, one cannot take the currentamount of private and institutional financialederal government has stopped providingdirect assistance to students. If the federalsystem of grants and loans were phased out,some persons who do not now contribute toprivate grants and loans would be likely tocontribute, and others who are already con-tributing would likely increase their contri-butions. Although this effect is unlikely tooffset federal aid completely, at least some ofthe reduction in government assistanceThe most interesting substitute for feder-al government intervention may be humancapital contracts. A human capital contractallows a student to go to the venture capitalmarket and obtain investors in his educa-In return for that financing, the stu-dent pledges a specific percentage of laterincome over a specified period of time to bepaid to the investor.Milton Friedman proposed such a contractin his 1955 paper ÒThe Role of Government inCapitalism andMiguel Palacios in his 2002 Catolicy Analysis ÒHuman Capital Contracts:ÔEquity-likeÕ Instruments for Financing HigherEducation,Ó provides a detailed analysis of suchcontracts, their benefits and limitations, andhow they would work in practice.When fully accepted, human capital con-tracts would come to be combined by invest-ment funds. Those funds would purchaselarge numbers of such contracts, allowing therisk of default to be spread over a sufficientnumber of students that the law of largenumbers would generate sufficient profit formarket for these contracts developed further,shares in the funds would be traded in thesame way that individuals purchase shares insuch things as real estate investment trusts.This would create an economically efficientway to finance higher education that wouldallow students to graduate without having toear that their future earnings would not besufficient to pay their student loans.There are other issues that may limit theextent of this market. For example, the interesting substitute for intervention maybe human capital sidy that will occur will depend upon the elastici-ty of demand and supply. The interested readermay find a full discussion in Hyman.19. Technically, the elasticity of demand is thepercentage change in the quantity demandeddivided by the percentage change in the price, andthe elasticity of supply is the percentage change inthe quantity supplied divided by the percentagechange in the price. For an introductory discus-sion see Roger Miller, Economics Today(Boston: Pearson, 2004), chap. 20.20. Judith Li, ÒEstimating the Effect of FederalFinancial Aid on College Tuitions: A Study of PellGrants,Ó Harvard University, April 1999, p. 11.21. Gordon Winston, ÒSubsidies, Hierarchy andeers: The Awkward Economics of HigherJournal of Economic Perspectives12, no. 1(Winter 1999): 13Ð36. In 1993 Michael Rothschildand Lawrence White made a number of interestingobservations about university behavior and raisedsome important questions for further research intheir article ÒThe University in the Marketplace:Some Insights and Some PuzzlesÓ that appeared inan important National Bureau of Economicesearch volume, udies of Supply and Demand in, ed. Charles Clotfelter and MichaelRothschild (Chicago: University of Chicago Press,1993). They followed this with their 1995 paper,he Analytics of the Pricing of Higher Educationand Other Services in Which Customers AreInputs,Ó Journal of Political Economy103, No. 3 (June586, in which they use universities as anexample of firms with the characteristic that thecustomers are also inputs to show that a competi-tive market in such a situation can result in someof the pricing and output behavior associated withuniversities. In the case of higher education, thecustomer is the student, but the type of students bywhich he or she is surrounded affects the quality ofeducation that any student gets. They note thatsuch a model still fails to explain some commonuniversity pricing and output decisions.22. Dennis Coates and Brad Humphreys, ÒTheSupply of University Enrollments: University110, no. 3Ð4 (March 2002):23. John Quigley and Dan Rubinfeld, ÒPublicChoices in Public Education,Ó udies of Supply andDemand in Higher Education. Quigley and Rubinfelddeveloped an empirical model of supply anddemand for public higher education based on atheoretical model that focused on the legislaturebeing the entity that determines the supply ofpublic education. Using 1984 state data on enroll-ment, average tuition, and other variables, theyound a negative relationship between tuition andpublic supply in a complicated regression analysisthat combined two- and four-year institutions.This would mean that tuition increases result inless supply of public higher education, or a down-ward-sloping supply curve. On page 268, theauthors describe this result as Òdifficult to inter-pret; given the difficulties of identification, itcould merely reflect the negative relationshipbetween student demand and tuition.Ó In separateregressions on two-year and four-year public insti-tutions, they get a positive relation betweentuition and supply for two-year colleges, and apositive relation between supply of four-yearenrollment and four-year tuition. This makessense and means the supply curve is upward slop-ing. Their separate regression on two-year publiccolleges resulted in a negative relation betweentuition and supply. It would thus appear that theircounterintuitive results of the combined regres-sion are affected by the legislaturesÕ decisionsregarding the provision of two-year colleges.24. See Li, p. 3.25. In this paragraph, revenue is Òcurrent fundrevenue.Ó This is money that can be used to payobligations currently due and surpluses reappro-priated for the current fiscal year.26. U.S. Department of Education, NCES, of Education Statistics, 2002, June 2003, Table 330. 27. Quoted in Larry Singell and Joe Stone, ÒForWhom the Pell Tolls: Market Power, TuitionDiscrimination, and the Bennett Hypothesis,Óiversity of Oregon working paper, April 2003.28. Chester Finn Jr., Scholars, Dollars, andBureaucrats(Washington: Brookings Institution,1978), p. 62.29. Jacob Stampen, The Financing of Public Higher(Washington: American Association forHigher Education, 1980), p. 41.30. Ibid., pp. 6Ð7.31. See Greg Winter, ÒHouse G.O.P. to Drop Ideaof Penalty for Steep Rises in Tuition,Ó New , March 3, 2004.32. For example, see Susan Dynarski; HidehikoIchimura and Christopher Taber, ÒSemiparametricduced-Form Estimation of Tuition Subsidies,ÓAmerican Economic Review 92, no. 2 (May 2002):286Ð92; and Marcus Stanley, ÒCollege Educationand the Midcentury GI Bills,Ó Quarterly Review of118, no.2 (May 2003): 67133. Li, p. 11. 17 Tower(Washington: Regnery, 1994), chap. 5.60. The percentage enrollment in public institu-tions of higher education was 76.8 in 2000. SeeDigest of Education Statistics 2002, Table 172.61. For a discussion of the constitutionality of theHigher Education Act, see Arnn, especially pp. 13Ð17.62. The FY 2005 Education Department BudgetSummary requests $22.256 billion for studentnancial assistance under HEA. This does notcount other federal programs, such as theMontgomery GI Bill, nor various tax credits.63. College Board, rends in Student Aid, 2004, p. 5.64. In the interest of brevity, please read ÒhisÓ as ÒherÓif one chooses a different gender for the example.65. Milton Friedman, ÒThe Role of Governmentin Education,Ó in ed. Robert Solo, (Rutgers: Rutgers UniversityPress, 1955) and Capitalism and Freedom iversity of Chicago Press, 1962).66. Miguel Palacios, ÒHuman Capital Contracts:ÔEquity-likeÕ Instruments for Financing HigherEducation,Ó Cato Institute Policy Analysis no. 462,2002, http://www.cato.org/pubs/pas/pa-462es.html. Also see Miguel Palacios, Investing in HumanCapital: A Capital MarketÕs Approach to Student Fund-(Cambridge, U.K.: Cambridge University Press,67. Those familiar with this literature may recognizethe similarity between human capital contracts andleÕs ill-fated Tuition Postponement Program.However, a major difference is that YaleÕs programrequired all participants to continue to pay a portionof their income until the loans of all members of theircohort had repaid their loans. Thus, those who par-ticipated in the program were forced to cover the loss-es of those who failed to repay their loans. For a dis-cussion of the Yale program see Bret Ladine, Ò70sDebt Program Finally Ending,Ó Yale Daily NewsMarch 27, 2001, www.yaledailynews.com/article.asp?68. Palacios. The website for MyRichUncle ishttp://www.myrichuncle.com/.