[Economic Report of the President (1996)]
[Administration of William J. Clinton]
[Online through the Government Publishing Office, www.gpo.gov]

[DOCID: f:erp_c7._]
Economic Report of the President - - - - - - - - - - - - H. Doc. 104-161
[From the online service of the U.S. Government Printing Office]
[wais.access.gpo.gov]

CHAPTER 7

Investing in Education and Training

THE FEDERAL GOVERNMENT HAS BEEN a vital partner in education for
more than 200 years. Even before the Constitution was adopted, the
Ordinance of 1785 set aside a section in every township in the new
territories west of Pennsylvania to support a school. In 1862 the first
Morrill Act authorized Federal land grants to States for the
establishment of colleges. As World War II came to an end, a grateful
Nation offered the G.I. bill, which eventually served nearly 8 million
returning veterans--and fundamentally changed the educational landscape
of the country. Today, Federal educational loans and grants open the
doors to college for millions of students who could not otherwise
attend, and Federal grants to low-income schools help more than 6
million children learn to read and to do math.
Learning is a lifelong process, not limited to those between the
ages of 5 and 25. From early childhood education to college to training
for the unemployed, this Administration has sought to complement the
efforts of State and local governments in responding to the new demands
of the labor market. The Nation is in the midst of an educational
renewal, and families, teachers, local school districts, colleges,
States, employers, and the Federal Government all have a role to play in
the transformation.
The renewed Federal interest in education and training is in part a
response to the two challenges outlined in Chapter 1: the slowdown in
the growth of productivity and the increase in earnings inequality.
Education and training policy is one of the few policy levers available
to address both problems simultaneously.
One of the most dramatic changes in our economy during the past 15
years has been the increased economic payoff to skills, as reflected in
the increased inequality in earnings between high school and college
graduates. In 1979 full-time male workers aged 25 and over with at least
a bachelor's degree earned on average 49 percent more per year than did
comparable workers with only a high school degree. By 1993 the
difference in wages had nearly doubled, to 89 percent. To the extent
that this rise in the payoff to education reflects an increase in the
value of skill, improving our schools and expanding access to
postsecondary training stimulate economic growth. Based on estimates
from the Bureau of Labor Statistics, the rise in the average educational
attainment of the workforce accounted for one-fifth of the annual growth
in productivity between 1963 and 1992. International evidence reveals
that, all else equal, those nations with the highest school enrollment
rates in the early 1960s tended to enjoy the most robust growth in
subsequent decades.
Education and training policies can also help address the problem of
growing inequality. A primary goal of Federal policy must be to ensure
that educational opportunities are not restricted to those whose parents
can finance an education out of their own pockets. Federal programs such
as Head Start, which helps low-income children prepare for school; Title
I of the Elementary and Secondary Education Act, which provides
supplemental Federal assistance to low-income schools and school
districts; and Federal financial aid for college students are all
designed to support those who would otherwise not have an equal
opportunity to invest in learning.
The sharp rise in family income inequality should not be allowed to
cause greater inequity in access to educational opportunities. The
widening disparity in earnings prospects between the more and the less
educated makes such efforts to equalize educational opportunities even
more imperative. Since the 1980s the Nation's track record in equalizing
educational opportunity has been mixed. In elementary and secondary
schools, racial gaps in test scores in mathematics, reading, and science
have closed somewhat, even as mean scores have risen for whites as well
as blacks and Hispanics. The black-white gap in high school graduation
rates has also narrowed since the mid-1970s, as high school graduation
rates rose for blacks.
However, gaps in college enrollment rates between low- and high-
income youth and between minority and white, non-Hispanic youth have
widened since the late 1970s (Chart 7-1). Although all groups have
responded to changes in the labor market by attending college at higher
rates, the increases have been larger for middle- and higher income
youth than for low-income youth. Because blacks and Hispanics are
overrepresented at the bottom of the income distribution, the racial and
ethnic enrollment gaps have widened as well.
The widening gaps in college enrollment are troubling for at least
two reasons. First, they may imply an increasing perpetuation of
inequity from one generation to the next--with access to higher
education increasingly allocated on the basis of ability to pay, not
ability to learn. In this country, which values the principle that
children's success in life should not be held hostage to their parents'
lack of resources, this is unacceptable. A second reason is that low
enrollments deprive the economy of the skills of those unable to finance
those investments. The labor market is demanding high-



er levels of skill, and the economy will grow more quickly if we succeed
in producing more skilled workers.
Education and training policy can contribute to reversing the growth
of inequality in the country in two ways. First, by targeting
educational resources more effectively, education and training policy
may enable more of our citizens to benefit from the rising payoff to
skill. Second, a robust supply response that creates an abundance of
skilled labor and causes less-skilled labor to become relatively more
scarce may slow the rise in the price of skill in the labor market,
reducing the growth of wage inequality and possibly even reversing it
somewhat.
In short, the Administration's education and training policies are
predicated on the three principles outlined in Chapter 1. They encourage
students and schools to embrace change by developing the skills demanded
by the new labor market. They create opportunity by targeting resources
to the disadvantaged, providing greater opportunity to participate
fruitfully in that market. And they promote personal responsibility, by
stressing to young people and workers that they are responsible for
making their own educational choices, and by requiring them to share
some portion of the cost: through their efforts in school, through the
earnings they forgo to remain in school, through their participation in
the Federal Work Study program, and through their obligation to repay
educational loans.
This chapter first reviews the good news on the extent to which the
Nation has responded to the rise in the value of education since the
early 1980s, as well as the sobering news on how far we still have to
go. The chapter then examines the evidence from the economics literature
on the payoff to investments in schooling and training. Finally, we
describe the Federal role in education and training policy in
complementing State and local efforts.

AMERICANS ARE RESPONDING TO THE DEMAND FOR SKILLS

Americans have always placed a high value on education, seeing it as
a ladder of opportunity. Therefore, the country was ready to respond
when A Nation at Risk, the 1983 report of a commission appointed by the
Secretary of Education, sounded the alarm over declining nationwide test
scores. Since then a number of States and local school districts have
launched ambitious reform projects. After a decade of effort, progress
clearly has been made:
 Students are spending more time on homework than they did
at the end of the 1970s. The proportion of 13-year-olds
reporting that they had no homework or that they had not done
their homework declined from 38 percent in 1980 to 25 percent in
1992.
 The proportion of 11th- and 12th-grade students taking
advanced placement courses grew by 138 percent between 1984 and
1992.
 In 1992 the average public high school graduate had
completed 49 percent more courses in algebra or higher
mathematics, 33 percent more coursework in science, and 8
percent more coursework in English than his or her counterpart
in 1982.
 Between 1980 and 1993, the proportion of students in grades
10 through 12 remaining in school rose for whites, blacks, and
Hispanics. The decline in the dropout rate was particularly
steep for blacks.
The hard work of students, parents, teachers, and school
administrators has borne fruit in the form of higher test scores and
higher college enrollment rates. Some year-to-year fluctuations
notwithstanding, most of the trends suggest that progress is being made:
 As measured by scores on the National Assessment of
Educational Progress, average mathematics proficiency rose for
nearly every age, gender, and racial or ethnic group between
1978 and 1992.
 Average mathematics scores on the Scholastic Aptitude Test
(SAT) rose by 13 points overall and by 28 points for blacks
between 1980 and 1994. These gains are particularly impressive
given the large increase in the proportion of high school
students taking the SAT, which would have tended to reduce
average scores.
 The proportion of college-age youth (those 18 to 24 years
old) enrolled in college grew by more than one-third between
1980 and 1994, from 26 percent to 35 percent.
 The numbers of associate, bachelor's, and doctoral degrees
awarded grew by 28 percent, 25 percent, and 29 percent,
respectively, between 1980 and 1993, even though the population
of college-age youth declined by 15 percent.
However, much remains to be done. Although average scores have been
rising in mathematics and science, much of the gain has occurred in
lower level computational skills rather than in higher level problem
solving. Reading and writing test scores declined slightly for the
weakest students during the late 1980s. Perhaps most disturbing,
students in the United States continue to lag behind their counterparts
in many Asian and European countries in math and science (Chart 7-2).



Although it is tempting to extrapolate from current trends and to
assume that the rise in skill-related earnings inequality will continue
unabated, economic historians tell us that the payoff to education has
fluctuated over the past 50 years, rising and falling with changes in
supply and demand. For example, the ratio of the average earnings of a
college graduate to the average for high school graduates is today
roughly what it was in 1940. Economic theory predicts that positive
shifts in demand will be met by increases in the quantity supplied.
Although Americans have responded by enrolling in college in record
proportions, so far the demand for skill has outpaced the Nation's
ability to produce more skilled workers. But the demographic tide is
gradually turning, as the number of 18- to 24-year-olds is expected to
rise by 20 percent over the next 15 years. Eventually the rise in the
labor market value of skill, and the wage inequality it has brought
about, may be dampened if these new workers are better equipped to meet
the demands of the labor market. The remainder of this chapter discusses
the role of government policy in aiding that response.

DO EDUCATION AND TRAINING LEAD TO HIGHER EARNINGS?

Throughout the 1980s the gap in real annual earnings widened between
American workers with different levels of education (Chart 7-3). Labor
economists have argued for decades over whether education actually
causes differences in earnings, or whether those with better earnings
prospects--because of more favorable family backgrounds or greater
native ability--simply consume more education. After literally hundreds
of studies of the economic importance of education, most economists now
agree that education does, indeed, lead to higher earnings (although
they may disagree about the size of the effect). Each additional year of
formal schooling is associated with a 5 to 15 percent increase in annual
earnings later in life. Even without counting the other benefits offered
by education--a more active citizenry, breakthroughs in science and the
arts, less reliance on social welfare programs--such benefits are often
large enough to justify the public and private investments involved (Box
7-1).
Questions of causation are difficult to resolve, however, because
unlike natural scientists working in the controlled setting of the
laboratory, researchers cannot simply assign people randomly to
different educational careers. Even if one tried to perform such an
experiment, those assigned to lower levels of educational attainment or
training could always decide to pursue their options elsewhere. This
implies that random assignment experiments can only evaluate the
incremental impact of specific programs over that of opportunities
available elsewhere--not the full value of the training. The more
options available for education and training, the smaller will be the
incremental impact of any specific program--even if the training itself
is quite worthwhile. Therefore, in addition to using experimental
evidence, economists have exploited sev-



eral other sources of variation in educational attainment in studying
the effect of additional education and training on earnings.

COMPARING THE EARNINGS OF SIMILAR WORKERS WITH VARYING EDUCATIONAL
ATTAINMENT

For decades survey researchers have collected information not just
on education and earnings but on other characteristics, such as
standardized test scores, parental education, and family income, which
might be related to both educational attainment and future earnings. In
analyzing these data, economists have attempted to control for prior
differences in earnings prospects between the more and less educated, by
studying the relationship between education and earnings only among
those who might be expected to have similar earnings given their other
characteristics.
In such studies, more than 75 percent of the estimated impact of
education typically remains even after controlling for test scores prior
to entering college. One recent study compared the earnings 14 years
after high school of a sample of graduates of the high school class of
1972 who had attended different types of postsecondary institutions.
Although those who had attended 4-year institutions had higher earnings
than either community college students or those with no postsecondary
training, they also had higher grades, higher standardized test scores,
and more favorable family

Box 7-1.--Is a College Education a Worthwhile Investment?

Calculating the return on any investment involves assessing both
costs and benefits. Here we do some back-of-the-envelope calculations of
the economic return to a college education.
Although a college education certainly yields other benefits,
earnings differentials after college--the additional wages that a
college graduate earns compared with a high school graduate--are perhaps
the easiest to measure. It remains to be seen how today's college
graduates will fare over the next 45 years of their careers; absent that
information, the most straightforward approach is to assume that the
difference in earnings observed among people of various ages and
educational attainments today will persist into the future.
A college education clearly has high costs as well. In addition to
the $10,000 in average educational costs per year of college, students
forgo potential earnings while in school. Since a full-time college
student would typically miss 9 months of work experience in a year,
three-quarters of the average annual earnings of an 18- to 24-year-old
male high school graduate, or $12,200, is a reasonable estimate of
earnings forgone for each year of full-time college study. Therefore the
total cost of a year in college is the combination of educational costs
and forgone earnings, approximately $22,200.
If these measures of costs and benefits are accurate, the internal
rate of return on 4 years of college for a male, 13 percent, is higher
than that for most financial instruments. Even if one attributes only 75
percent of the earnings difference between high school and college
graduates to schooling, the internal rate of return is still 11 percent.
Despite the high costs, then, a college education continues to be a
worthwhile investment.
backgrounds upon graduating from high school--all characteristics that
would have predicted higher earnings for them even if they had not
attended college. Comparing those who had similar family backgrounds and
academic characteristics in high school, the researchers found that a
year of community college was associated with an increase in earnings of
4 to 7 percent, roughly the same as that associated with a year in a 4-
year college.

STUDIES USING TWINS

Admittedly, however, many of the characteristics that affect
earnings are difficult to measure. Such easily quantifiable variables as
family income or years of education received by one's parents may not
fully capture the myriad differences in family background. Rather than
attempt to collect information on a seemingly infinite list of
characteristics, some survey researchers have gone to great lengths to
follow the experience of pairs of identical twins. Because identical
twins growing up in the same household share a variety of environmental
and genetic factors, analyzing differences in their earnings and
educational attainment eliminates the need to measure the subtle ways in
which backgrounds may differ between families.
The conclusion of this research is that, even among identical twins,
those with more education tend to earn more. In some studies, the
difference in earnings associated with a year of education has been as
great as the 5 to 15 percent earnings difference per year of education
observed in the broader population. For example, a recent study of this
type found that each year of education was related to a difference in
earnings of between 12 and 16 percent.

NATURAL EXPERIMENTS

Just as individuals from different families may differ in ways that
are not easily measured, identical twins may have different experiences
growing up that would lead one twin to attend school longer and to earn
more in the labor market than his or her sibling. A third approach,
therefore, is to identify laws or institutional differences that may
have an effect on educational attainment but are expected to have no
independent effect on earnings.
Compulsory schooling laws provide one such opportunity. Many States
once had regulations that allowed only those turning 6 during the
current calendar year to enter first grade in the fall. In other words,
5-year-olds with their 6th birthdays falling on or before December 31
could begin classes in the fall, while those born on January 1 or later
had to wait an additional year. Because compulsory schooling laws
specify a minimum age of mandatory attendance (usually age 16 or 17) and
not a minimum grade level, those born during the first calendar quarter
reached the age at which they could drop out after having completed a
year less of school than those born in the last calendar quarter. As
long as the earnings of those born at different times of the year do not
vary systematically for reasons unrelated to educational attainment, the
interaction between compulsory schooling laws and calendar quarter of
birth provides a ``natural experiment'' for measuring the impact of
education on earnings. Researchers have found that those with birthdays
in the first calendar quarter were indeed slightly more likely to drop
out at lower grade levels than those born later in the year. Moreover,
each year of additional education was associated with a 5 to 10 percent
increase in hourly wages later in life.
The study of compulsory schooling laws is particularly important
because it identifies the payoff to a year of schooling only for those
who are constrained by such laws to remain in school, rather than
describing the average return to education for all who remain.
Therefore, the results suggest that even those who would have dropped
out earlier than compulsory schooling laws allowed seemed to benefit
from additional schooling. This is a strong argument for measures to
deter high school students from dropping out (Box 7-2).

RANDOM ASSIGNMENT EXPERIMENTS

Even though, as noted above, random assignment experiments can
identify only the incremental impact of specific programs and not the
value of training itself, some programs do indeed seem to raise the
earnings of those who are assigned to them. The primary advantage of
being able to randomly assign some subjects to training and others to a
comparison group is that one can expect that any resulting difference in
average earnings for the two groups is due to the incremental training
provided and not to some other difference between the two groups.
Although the studies are usually conducted on a small scale, random
assignment evaluations have often found that education and training
raise the earnings of participants. For instance, in recent years the
Center for Employment Training (CET) in San Jose, California, has
achieved impressive results in two different random assignment
evaluations. Out-of-school youth receiving an average of 4.1 months of
training at CET earned 40 percent more per year (approximately $3,000
per year in 1993 dollars) than the control group during the third and
fourth year after being assigned. The total cost of the program per
enrollee was $4,200. In a separate random assignment evaluation of a
program for minority single female parents, participants earned $1,500
(again in 1993 dollars) more than the control group in the second year
after training. Earnings increases remained large in the fifth year of
the study, by which time those who had received training and job
placement services were still earning 16 percent more than the control
group.
Education and training for experienced workers yield economic
benefits as well. A recent random assignment evaluation of the Job
Training Partnership Act (JTPA), a Federal program providing training
for economically disadvantaged clients, found that participation
increased the earnings of adult male participants by 7 percent and those
of adult female participants by 10 percent. These earnings gains were
one and one-half times greater than the costs of producing them.

LEARNING OR SORTING?

Although labor economists would generally agree that education and
training do lead to higher earnings, it is more difficult to deter-

Box 7-2.--New Opportunities for Potential Dropouts

One of the eight goals set out in the Goals 2000 Act is to raise
high school graduation rates to 90 percent by the year 2000. Indeed,
dropping out of high school is not a good financial decision. A male
youth who finishes the last 2 years of high school will reap a net
lifetime earnings increase of $99,000 (stated in present value terms at
a 3 percent discount rate). Even when one considers the cost to
taxpayers of 2 additional years of public secondary education ($5,600
per year), the internal rate of return for a male completing high school
is 9.5 percent. Persuading young people to remain in high school seems a
particularly worthwhile investment.
Between 1987 and 1989 the Department of Labor conducted a random
assignment evaluation of JTPA programs for out-of-school youth. The
average youth assigned to JTPA did not receive higher earnings during
the 30-month evaluation than did those assigned to the control group,
many of whom participated in other non-JTPA education and training
programs. In other words, the availability of JTPA programs did not seem
to add much to the existing array of services for out-of-school youth.
In response, the Department of Labor is exploring alternative
strategies. For instance, rather than providing training to students
once they drop out of school, the department is funding a replication of
a promising high school dropout prevention program. The Quantum
Opportunities Program (described in more detail in the 1995 Economic
Report of the President) will be replicated with over 1,000 participants
at seven sites around the country.
The Labor Department is also conducting a major evaluation of the
Job Corps program, a comprehensive, residential job training program for
high school dropouts. Treatment and control subjects will be followed
for 5 to 6 years to determine the impact of the program on employment
and other social outcomes.
The Labor Department has also experimented with ``geographic
targeting,'' saturating high poverty communities in inner cities and
rural areas with job training, work opportunities, school-to-work
programs, and sports and recreation activities. The aim is to reach
enough young people in a neighborhood to reverse the effect of peer
pressure. Although the saturation approach made random assignment
difficult, a nonexperimental evaluation is yielding promising results.
mine why they matter. Do employers pay their highly educated workers
more because of the skills they have learned, or do the more educated
earn more because educational attainment provides other signals to an
employer about them, such as their perseverance or level of motivation?
The question is very difficult to resolve empirically, since it is
difficult to measure acquired skill as distinct from educational
attainment. For instance, we infer the extent of a physician's training
not by directly measuring his or her medical knowledge but by observing
his or her educational credentials.
It is likely that some portion of the observed payoff to schooling
is due to both the ``skills'' and the ``sorting'' explanations. However,
it appears that technological change has increased the value of some
skills more than others. Even if sorting accounts for some portion of
the value of education, higher level problem-solving skills have almost
certainly increased in value with the availability of computers.
Furthermore, it would be difficult to attribute the large increase in
the payoff to schooling, even among those who have been in the labor
market for decades, to an increase in the value of education as a
signal. Greater success in producing these skills not only would raise
the earnings of those benefiting, but also would contribute to economic
growth. Moreover, when it comes to improving the earnings prospects of
the disadvantaged, whether it is the skill learned or the credential
acquired that opens the door, such investments improve the prospects of
those who may lack the resources to invest in themselves and reduce the
perpetuation of poverty.

THE PAYOFF TO PUBLIC INVESTMENT IN EDUCATION

Since the publication of Equality of Educational Opportunity
(commonly known as the Coleman Report) in 1966, researchers have
struggled with the question of whether increased expenditure on schools
improves student performance. The debate is often quite contentious
because of the large differences in expenditure per pupil between rich
and poor school districts. For example, during the 1992-93 school year,
New Jersey spent more than $9,400 per pupil in public elementary and
secondary schools, while Alabama and Mississippi spent less than $3,900.
Regional differences in the cost of living can explain only a small part
of such variation. Furthermore, given the importance of local financing
of public education, expenditure per pupil can differ by a factor of two
or three even between districts in the same State.
Typically, analysts compare average test scores in high-spending and
low-spending districts to learn about the effect of additional resources
on scores. Not surprisingly, the high-spending districts have higher
average scores. However, since high-spending districts also tend to have
higher average family income and parental education, the differences in
student performance may be caused not by differences in the level of
spending but by differences in family resources. When analysts compare
test scores in high- and low-spending districts with similar family
incomes and parental education, the results are often considered
provocative: districts that spend more are often found not to have
higher test scores.
However, additional resources could have other beneficial impacts.
The standardized tests used in much of the research may not reliably
measure the kinds of improvements that parents or policymakers would
expect schools to produce with additional resources. The benefits of new
courses in American history, geometry, or calculus or improved learning
opportunities for the disabled--valuable as they may be--would not be
captured by such measures.
Consistent with this hypothesis, studies of the long-term impacts of
school expenditure on earnings and educational attainment--in contrast
to those that focus on test scores--yield more optimistic evidence that
public investment in elementary and secondary schooling does generate
benefits later in students' lives. For instance, better paid and better
educated teachers and smaller classroom size have been associated with
greater educational attainment and higher payoffs to education later in
life, even if they have not had large effects on the particular test
scores used. One recent study concluded that the payoff was not only
positive but financially lucrative: a 10 percent increase in
expenditures from kindergarten through 12th grade would produce
additional lifetime earnings valued at 1.2 times the additional cost (in
present value terms). Admittedly, studies of this kind remain few, and
some authors have reported less positive results, but some evidence
suggests that past increases in spending on education did bear fruit,
even if the results did not register on the particular tests used.
But the debate over such findings often misses a more relevant
question: rather than continue to debate how much of a difference
additional resources have made in the past, we should be asking how
programs and incentives could be structured today to ensure even greater
benefits from resources invested now and in the future. It is difficult
to believe that a knowledgeable school principal could not find a way to
use additional resources to improve student learning, as long as the
incentives in the environment rewarded such gains. The task of
policymakers should be to create an environment in which incentives
dictate that resources be invested profitably.
On this question, Federal, State, and local governments are already
a step ahead of the academic debate. Many of the educational reforms
being pursued today seek to produce more decentralization and greater
accountability, both of which are designed to create an environment in
which resources are used more efficiently. The charter school movement
is a good example. Minnesota was the first State to pass a law allowing
for charter schools in 1991. Since then 19 other States have enacted
laws permitting the development of charter schools. A charter school is
usually the brainchild of a committed group of teachers or set of
parents who want the flexibility to try a different approach. Typically,
they apply to the local school board or the State department of
education for a charter allowing them to open a new school with public
funding. Since charter schools are public schools, they do not charge
tuition. Such charters typically waive many of the regulatory
requirements imposed on other public schools for 3 to 5 years, at which
time they are subject to review.
Charter schools enhance accountability in two ways. First, charter
contracts often specify benchmarks for performance, such as scores on
specific State assessments. In exchange for the freedom to innovate,
charter school organizers are expected to produce results. Some
contracts are more specific in spelling out such performance
expectations than others. As States develop better assessment tools
under the Goals 2000: Educate America Act (described below), these
performance expectations can be more explicitly stated. Second, the
presence of charter schools is intended to encourage innovation by
nearby public and private schools, through the demonstration of
successful educational strategies and through the threat of lost
enrollment.
The Department of Education has helped to nurture the charter school
movement by providing seed money for the establishment of charter
schools. In the 1995 fiscal year, the Federal Government provided nearly
$6 million in grants to help cover startup costs for charter schools.
The Administration hopes to increase this commitment significantly over
the next few years.
But the establishment of charter schools represents only one way in
which States and local school districts are seeking to provide better
incentives for schools and teachers. School report cards, performance
bonuses for schools, magnet schools, and other forms of public school
choice are also being tested.
Publicly funded vouchers for use at private schools are another,
more radical approach. But vouchers have several problems. Their
advocates fail to recognize the many ways in which education for
children differs from conventional goods. The primary risk of vouchers
is that they may produce a dramatic increase in social stratification.
The cost in terms of the resulting damage to social mobility and social
cohesion could exceed any benefit in terms of better school performance.
Because they are public schools dependent upon public support, charter
schools can be more carefully planned to serve all children's interests
by locating them in urban areas, by insisting on open admissions
policies, by holding them directly accountable for results, and--when
oversubscribed--by requiring them to establish lotteries for admission.
Charter schools provide a framework for an improved educational system,
with parents and teachers working together to develop new and creative
solutions to the challenges they face, and demanding accountability of
all participants in the educational process.
Some approaches to accountability are better suited to some
environments than others. For instance, school report cards are better
indicators of school performance when mobility between schools is low
and when one can control for differences in student characteristics.
Charter schools and magnet schools provide better incentives when the
quality of local transportation is good and parents are engaged and well
informed. Still another approach, which several European countries
employ, raises the stakes for students, through more widespread use of
achievement tests as a criterion for high school graduation and college
admission, or even by employers in their hiring decisions (Box 7-3).
Given the diversity of circumstances around the country, it is
appropriate that each State and school district pursue its own strategy
for encouraging more decentralization and accountability. The next
section discusses the various ways in which the Federal Government has
chosen to complement these efforts.

THE FEDERAL ROLE IN EDUCATION AND TRAINING

The environment facing providers of education and training is
changing. Today parents and taxpayers increasingly expect results from
their investments. In partnership with State and local policymakers,
Federal policy is helping to create this new environment in several
ways: by providing seed money to States developing content standards in
core subject areas, by supporting States in the development of
assessment tools for measuring progress, by helping States to invest in
their teachers, and by supporting the establishment of charter schools.
But in addition to these efforts the Federal Government serves many
other roles in our education and training system, such as guaranteeing
student loans, channeling resources to low-income schools and school
districts, helping disadvantaged children prepare to enter kindergarten,
and helping States develop new pathways from school to the world of
work. As mentioned at the outset of this chapter, the Federal Government
has played a vital role in education since before the Constitution was
signed. There are at least five reasons why.

Box 7-3.--Raising the Stakes for Students

Despite recent gains, American youth continue to perform poorly in
science and mathematics relative to their counterparts in many other
industrialized countries. American students also seem to spend less time
on their studies than students in other countries. The Organization for
Economic Cooperation and Development has suggested that one of the
causes of the poorer U.S. performance is the lack of connection between
high school achievement and employment or schooling opportunities.
Unless they are planning to attend a selective college, high
school students in America often have little incentive to do well
academically. Surveys suggest that employers have difficulty collecting
and interpreting transcripts from many different schools. And except for
the most competitive colleges, a student's performance in high school
has little impact on his or her chances of admission to college. The
skills developed in school may well matter later in students' careers,
but many students may fail to see a connection between performance in
school and immediate prospects for a job or college admission.
In contrast, many European countries require students graduating
from high school to take tests in various subject areas. Universities
use these scores in making admission decisions, as do employers in their
hiring decisions. Some precedent for such high-stakes testing exists in
the United States--the Regents Examination in New York is an example. By
raising the stakes for high school performance--or, possibly more
important, making the actual consequences more visible--these tests may
induce students to work harder.
An achievement test may also strengthen the incentives of students
and teachers to work together. Absent an external standard, schools
judge individual students relative to their classmates. But the relative
scale gives students an incentive to discourage their peers from
``wrecking the curve.'' In contrast, an external standard unites
teachers, students, and their classmates in a common objective: to
perform well.
To focus attention on the value of high school achievement, the
Administration has proposed providing $1,000 scholarships to the top 5
percent of every high school class, public and private, for use at
college. Although the reward is still based on a relative standard, the
goal of the awards will be to make the new realities of the labor market
more salient, giving students in school a more immediate reason to
strive harder.
First, Americans are a mobile people. Between 1993 and 1994 alone,
6.7 million Americans moved from one State to another. The consequences
of a good--or a bad--educational system therefore extend well beyond the
borders of a single State. For this reason, education is a national
concern as well as a local one.
One consequence of that mobility is that the Federal Government has
a distinct advantage in administering educational loan programs. The
average cost of a year at a public 4-year college is approximately
$10,000, not counting room and board, earnings forgone while attending
school, college expenditures on sponsored research, or scholarships and
fellowships. Even though States often pay a large share of these costs
through subsidies to public institutions, relatively few families have
the resources to finance such large investments out of pocket. Moreover,
because an education cannot be repossessed like a car or a house,
private lenders have not been willing, absent government guarantees, to
lend at reasonable rates, even to the most promising student. Given the
mobility of the population, the Federal Government is in the best
position to guarantee these loans and to pool the risk associated with
them.
Second, the Federal Government must share the responsibility of
guaranteeing equality of opportunity for all children. The commitment to
equal opportunity is founded upon both moral imperatives and economic
interests. The commitment to opportunity for all children has long been
a fundamental American value. The economic interest is also clear.
Without intervention by higher levels of government, many communities
would not be able to invest to the full extent worthwhile in their
children's educations. Although many State governments do target
resources on the most disadvantaged schools and school districts, as
argued in Chapter 4, Federal involvement may be necessary to avert a
``race to the bottom'' in the provision of State services to the
disadvantaged. And even if there were no race to the bottom, differences
in resources would mean children in disadvantaged communities or poor
States might receive an inadequate education. The Federal Government can
help to equalize access to educational opportunities across States and
school systems.
Indeed, some progress has been made over the past decades. As
already mentioned, black youth have closed part of the gap in test
scores with their white classmates in elementary and secondary school.
Nevertheless, students continue to come out of our school system with
enormous disparities in basic skills. One recent study has suggested
that differences in basic skills among youth emerging from our school
system may account for a significant share of the difference in average
earnings between black and white males in their late 20s.
Third, the Federal Government must play a role in research and
evaluation and in informing local decisionmakers about the payoffs to
alternative strategies. This is true of research and innovation in
education no less than in other areas. How much does classroom size
matter? Which teaching techniques produce better student performance?
Which training programs best meet workers' and employers' needs? To
deploy a school's resources wisely, teachers and administrators must
know which strategies work best for which youth. The answers to these
questions are public goods, of value to educators everywhere. Although
some school districts have conducted evaluations of their own, no
individual school or school district has a sufficient incentive to
invest, to the full extent worthwhile, in the kind of careful, expensive
random assignment evaluation necessary to resolve critical issues. The
Federal Government--through the Departments of Education and Labor, in
particular--has an important role in promoting, analyzing, and
disseminating this knowledge.
Fourth, the Federal Government has a critical role to play in
encouraging States to set content standards in education and to develop
testing methods that are consistent with those standards. Just as
industries have found it essential to set national standards to support
a national market for their goods, so it is with education: the national
labor market is more effective and efficient when employers in
California know that a job applicant graduating from school in New York
was held to a reasonably stringent set of standards. The recently
enacted Goals 2000: Educate America Act provides seed money to States to
develop standards and assessments.
Fifth, the Federal Government has a particularly important role to
play as a catalyst in developing a national response whenever change
occurs as suddenly as it has in the labor market over the last 15 years.
It performed this role admirably in the post-Sputnik era, leading
reforms in the math and science curricula of our Nation's schools. It is
playing that role today in a number of areas. For instance, the School-
to-Work Opportunities Act allows the Departments of Education and Labor
to jointly offer relatively small, short-term grants to States to begin
developing pathways to careers for high school students. Although the
Federal funding is short-term, scheduled to be phased out by 2001, the
presumption is that thereafter States and local governments will
continue to finance the experiments that worked and drop those that did
not. Similarly, in response to an evolving labor market in which some
workers find themselves in need of retooling, the Administration has
been working to transform the unemployment system into a reemployment
system. A third example is the Federal Government's encouragement of
charter schools. In these and other areas the Federal Government acts as
a catalyst, providing startup funds to encourage States to think in new
ways about the problems presented by a changing world.
Federal efforts--in particular, research and evaluation and the
encouragement of standards and assessments--complement States' systemic
reform efforts. With the knowledge gained from rigorous experimental
evaluations of alternative educational interventions, school principals
will make better decisions. With well-defined standards and assessments,
parents and local school administrators will have better information to
back their demands for accountability from the schools. Teachers, too,
will have a clearer idea about where to invest in their own training and
classroom preparation, so that they can effectively teach the material
defined in content standards at the State and local level.

ONGOING EFFORTS IN EDUCATION AND TRAINING

State and local governments have traditionally borne most of the
burden of financing elementary and secondary education. As recently as
1920, the Federal Government provided only 0.3 percent of nationwide
funding for public education from kindergarten through 12th grade.
(Currently, 9 out of 10 youth attend public elementary and secondary
schools.) With the advent of the Great Society programs of the 1960s and
the growth in Federal aid to low-income school districts, the Federal
share rose, reaching a peak of 10 percent in 1980. That share has
generally declined over the past decade and a half, however. In 1992-93
the Federal Government provided only 7 percent of total funding for
public elementary and secondary education, with State and local
governments roughly splitting the remaining 93 percent.
The Federal Government has traditionally played a larger role in
higher education than in elementary and secondary education. In 1993
Federal spending accounted for approximately 25 percent of the revenues
of all American institutions of higher education. (Of that 25 percent, 9
percent went to provide student grants and loans, 12 percent was for
sponsored research, and the remaining 4 percent for direct
appropriations and unrestricted grants.) In part, the greater Federal
role in higher education may reflect the fact that highly educated
people are more likely to move across State lines. In 1990, 49 percent
of 25- to 34-year-olds with a bachelor's degree, but only 33 percent of
those with less education, lived outside their State of birth.

EARLY CHILDHOOD EDUCATION

The Head Start program, begun in 1965, provides educational,
nutritional, and health services to children up to the age of 5; 90
percent of program beneficiaries must be from families with incomes
below the poverty level. The program has enjoyed bipartisan support, as
reflected in the fact that funding for Head Start more than doubled
between 1989 and 1995. In the 1995 fiscal year, the Head Start program
cost $3.5 billion and provided funds to approximately 2,000 programs and
750,000 children. In addition to increased funding, the Administration
has sought to improve program quality by increasing the number of
expanded day slots for children from families with working parents and
by seeking to improve the quality of program staff.
Evaluations of Head Start have reported short-term gains in IQ among
children enrolled in the program; enrollees are also less likely in
their later school careers to repeat grades or be assigned to special
education classes. The long-term impacts of Head Start are more
difficult to assess, given the long lag between investments and results.
One recent evaluation reported sustained improvements in cognitive test
scores for white participants, whereas initial favorable impacts seemed
to diminish for black youth. Early benefits may wither if they are not
nurtured in elementary school. Evaluations of Head Start have also
pointed to its significant improvement in the delivery of preventive
health services to children from low-income families, as reflected in
measures such as immunization rates.
Despite recent additional investments in Head Start, children from
high-income families remain much more likely to start school having had
the benefit of early childhood education. In 1993 only 33 percent of
children from the poorest 20 percent of families were enrolled in
preschool or kindergarten, compared with 59 percent of children with
family incomes in the top quintile. Because Head Start still serves
fewer than 40 percent of eligible families, the Administration has
proposed its continued expansion. If we are to reach the goal of equal
access to high-quality early childhood education, the Head Start program
deserves continued and expanded bipartisan support.

ELEMENTARY AND SECONDARY EDUCATION

To sustain the gains achieved in early childhood programs,
elementary and secondary schools must provide challenging and engaging
curricula that set high expectations for all their students. Three major
initiatives over the past 2 years--the Goals 2000: Educate America Act,
the reauthorization of the Elementary and Secondary Education Act, and
the School-to-Work Opportunities Act--were designed to complement and
support the reform efforts of State and local school officials.

The Goals 2000: Educate America Act

The Goals 2000: Educate America Act, passed by the Congress in 1994,
is the centerpiece of the Administration's effort to support State and
local school reform to raise standards of achievement. Its purpose is
twofold: to provide grants to States to set rigorous standards for
academic achievement, and to support local grassroots efforts to ensure
that all students meet those standards. In the first round of grants
every State but two applied for funding to support statewide systemic
reform efforts as well as promising local initiatives. In the first year
of the program, total funding for State grants was $90 million. States
were required to distribute 60 percent of these grants directly to
school districts, to support innovative programs to improve student
achievement in core subjects. The remaining 40 percent could be used for
statewide planning, such as the development of academic standards and
better statewide assessment tools. In the second year of the program, 33
States have so far received grants totaling $274 million, of which
States are obligated to pass 90 percent along to school districts.
As argued above, educational investments are most likely to pay off
when the objectives are clear and when some measure exists for tracking
the progress of students and schools. Accordingly, States applying for
funding under the second year of the program must develop or adopt
challenging content and performance standards and a means of assessing
whether the standards were met. States must also outline their plans for
helping teachers develop their abilities to teach to the challenging
standards. States, school districts, and schools are given a great deal
of flexibility in their planning to achieve these goals. Indeed, the act
expressly proscribes Federal mandates, direction or control of a
school's curriculum or program of instruction or the allocation of State
or local resources.
According to a survey by the Council of Chief State School Officers
in May 1995, 47 States were working on more rigorous content standards
and means of assessment. In Vermont, for example, the assessments
encompass a broader range of student achievement than do standardized
tests. The mathematics standards are typically the furthest along,
drawing on the efforts of the National Council of Teachers of
Mathematics during the mid-1980s. Perhaps it is no coincidence that
mathematics test scores have shown the greatest gains since 1980.
In addition to providing grants for systemic reform, the Goals 2000:
Educate America Act codified into law eight national goals, for
improving high school graduation rates, student achievement and
citizenship, math and science performance, adult literacy, teacher
education, school safety, school readiness, and parental participation.
The act also provided funding for the National Education Goals Panel, to
monitor the Nation's progress toward meeting those goals. The panel, an
autonomous body established in 1990, is charged with publishing regular
progress reports and with making suggestions to Federal, State, and
local governments that will further the achievement of those goals.

The Improving America's Schools Act

Whereas the Goals 2000: Educate America Act intends to provide
momentum and direction to State education reform efforts, the Improving
America's Schools Act (IASA) seeks to better coordinate Federal aid with
those State reform efforts. The most important part of this act was its
reauthorization of the Elementary and Secondary Education Act (ESEA) of
1965. The most significant budgetary change was the overhaul of Title I
(formerly Chapter 1) of the ESEA, which provides grants to States and
local school districts for the education of disadvantaged students. The
program, for which $6.7 billion was appropriated in 1995, was improved
in five important ways.
First, the act allows more schools with high proportions of students
from poor families to use their Title I grants for schoolwide reform
programs. Until the IASA was enacted, only schools in which more than 75
percent of children came from poor families had been allowed to use the
money for schoolwide programs. The IASA lowered the threshold further:
eventually it will allow schools with more than 50 percent poor children
to use Title I grants for schoolwide reforms. This corrects a
longstanding problem that prevented some students and teachers even in
high-poverty schools from using equipment purchased with Chapter 1
funds.
Second, States and local educational authorities are required to
monitor the progress of students in Title I programs using the same
standards and assessments used for other students. State and local
educational authorities are given greater authority to intervene in
schools that fail to show progress. Both measures should allow local
administrators to better coordinate Title I programs with State and
local reform efforts.
Third, the IASA eliminated the perverse penalty imposed on low-
income schools that succeeded in raising test scores. Prior to the IASA,
while poverty rates determined school eligibility, resources were
distributed among individual schools according to the performance of
their students. Low-income schools that raised their performance could
actually lose funds. Thirteen percent of principals in a survey of
elementary schools reported that their Chapter 1 (now Title I) program
had lost some funding as a result of improved performance. Under the
reauthorization, disbursement within local educational authorities
depends only upon the number and percentage of poor children, not on
their academic performance.
Fourth, school districts are required to involve parents and
communities in the education of their children, and to use 1 percent of
their Title I money for such programs. Research consistently finds that
close parent and teacher collaboration is needed to help students learn.
Fifth, Title I establishes two new, better targeted formulas for
disbursing money to poor districts and schools. As part of its 1996
budget, the Administration proposed distributing an additional $1
billion through the more targeted of the two new formulas, combining
$700 million that was to have been distributed under the old formula
with $300 million in new money.
The IASA includes other legislation intended to improve teaching and
learning. For instance, the Eisenhower grants (Title II of the ESEA) are
designed to support the efforts of schools and communities to develop
high-quality teacher training in all core subject areas, with particular
emphasis on math and science. The Safe and Drug-Free Schools Act (Title
IV of the ESEA) provides funds to States and communities to support
prevention of drug abuse and violence in their schools. In combination
with the Goals 2000: Educate America Act, the IASA for the first time
also grants the Secretary of Education waiver authority to give States
and local schools more flexibility in implementing their reforms.

Promoting Uses of Technology in Education

The Administration has supported the creative use of technology in
schools. The Technology Learning Challenge, funded under Title III of
the ESEA, provides challenge grants to partnerships of schools,
colleges, and the private sector for the development and demonstration
of educational technology. In 1995 the initial challenge grant
competition for elementary and secondary education attracted over 500
proposals and resulted in 19 grants totaling $10 million. The challenge
grants have been matched by $70 million in private sector contributions
in the first year. For example, the Capital School District in Dover,
Delaware, received a challenge grant to bring educational curricula and
communication links into students' and teachers' homes. Using a device
connected to their telephone or cable lines, students use their family
television sets to communicate with their teachers and classmates, and
so replace passive television watching with learning time. The project,
intended eventually to reach all 16 of Delaware's school districts, also
receives considerable support from the State government and private
sources.
During 1995 the President and the Vice President appealed to a group
of firms to bring Internet access to schools in California. The goal of
the privately funded effort is to establish Internet access to all
elementary and secondary schools and set up local area networks within
20 percent of them by the end of this school year. Before this effort,
California ranked near the bottom in the ratio of students to computers
available in schools, even though it is home to much of the computer
industry.
The Star Schools program provided $25 million in matching grants in
fiscal 1995 for projects using telecommunications technology in distance
learning. For instance, a Star Schools grant supported the development
of software to allow teachers from around the country to contribute and
draw from a data bank of lesson plans in various topic areas such as
math and science.
The IASA also provided $10 million in funding in fiscal 1995 for six
regional technology consortia. For instance, the South Central
consortium is made up of the Kansas State Board of Education and
colleges of education at Texas A&M University, University of Oklahoma,
University of Missouri-Columbia, and University of Nebraska-Lincoln. The
consortia are intended to provide consulting services to States and
school districts interested in finding new uses for technology in their
schools.
To give teachers, school administrators, and researchers around the
country better access to the inventory of educational research
maintained by the Educational Resources Information Center (ERIC), the
Administration created the AskERIC service. Educators and researchers
are able to send questions to the service by electronic mail and receive
a response within 48 hours.
Although the Federal investment in each of these programs is
relatively small, the lessons learned from experimenting with the uses
of technology in education may eventually have much broader applications
in elementary and secondary schools around the country.

The School-to-Work Initiative

Young people leaving high school often lack the skills and the
social networks to make the transition to work. A successful transition
means that a young person soon finds a job that puts him or her on a
career ladder at the hiring firm or imparts skills that make him or her
more widely employable. The experience of other countries and some of
the experiments in the United States have shown that programs that help
young people learn skills in the context of an actual workplace make
successful transitions from school to work more likely. For instance,
Germany's apprenticeship system is often given credit for the low
unemployment rates for youth in that country.
The School-to-Work Opportunities Act, passed in 1994, provides
States and communities with funds to assist young people in making the
transition to work after secondary schooling. Through the combined
efforts of the Departments of Education and Labor, the Federal
Government is to act as a catalyst, providing venture capital to States
for the development and implementation of school-to-work systems. In
1994 the Federal Government gave 52 development awards--one to each
State, the District of Columbia, and Puerto Rico--to assist in the
initiation of these systems. Also included were eight implementation
awards: funds competitively awarded to States with operating school-to-
work systems. The States receiving the implementation awards in 1994
were Kentucky, Maine, Massachusetts, Michigan, New Jersey, New York,
Oregon, and Wisconsin. By the end of 1995, 27 States had received
school-to-work implementation grants, as had almost 90 urban and rural
communities. Since the inception of the program, the Departments of
Labor and Education have provided $345 million to advance the school-to-
work initiative.
For example, the Socorro High School for the Health Professions in
El Paso, Texas, combines a traditional college preparatory course of
study with applied health occupations classes. In the first 2 years of
the 4-year program, students take an introductory course in the health
professions, a health occupations laboratory, enhanced mathematics, and
a foreign language, in addition to standard subject matter. In the 11th
grade, students spend half of each school day in clinical rotations;
they undertake 12 unpaid 3-week rotations, formally observing health
care providers and administrators at work. Students also visit local
colleges to learn about postsecondary education in health fields. In the
last year of the program students work between 15 and 20 hours per week
in competitively allocated, year-long internships. Students receive
performance evaluations from supervisors in these internships; those
receiving positive evaluations are typically hired as part-time regular
employees. The program receives guidance from the El Paso Hospital
Council, a coalition of senior executives from all the major health care
facilities in the city. More than three-quarters of the students in the
Socorro program are from low-income bilingual families; the school
receives funds from Title I of ESEA and the Job Training Partnership
Act.
An apprenticeship program in rural Pickens County, South Carolina,
accepts exemplary students for youth apprenticeships. The program offers
high school courses at the district career center, where students learn
skills from agricultural mechanics to graphic communications to welding.
Even in traditional subject areas, students apply their knowledge in
situations that simulate the workplace. During their senior year
advanced vocational students work as apprentices for 20 hours a week,
earning an average of $6 per hour at local businesses while taking
classes both at their high school and at the district career center.
After graduating from high school, the apprentices continue to work
part-time while studying for an associate degree at a technical college
in the area. Local businesses and large corporations with local
establishments have taken apprentices in the program. The Partnership
for Academic and Career Education (PACE), a consortium of businesses and
educators, assists with curriculum development, provides staff
development opportunities, and contributes materials to area high
schools. The Department of Education recognized PACE with the first
Award for Technical Preparation Program Excellence in 1991.
Both these programs have some degree of employer involvement, a
critical component of success. Employers can be counted upon to maintain
their investments in apprenticeships and worker training only to the
extent that they learn that it is in their economic interest to do so.
If employers are expected to share the costs, they must be rewarded with
some of the benefits. Some evidence suggests that there are indeed
benefits to be shared. A recent study of small manufacturing firms in
Michigan that received training grants from the State government
significantly raised productivity by reducing wastage. Another survey of
manufacturing firms that introduced formal training programs in 1983
suggested that these firms enjoyed faster productivity growth than other
firms. How these benefits are shared will depend upon turnover rates
among trained workers. The experience of those firms that have been
willing to participate in the school-to-work initiative, or have
invested in incumbent workers, will have an important impact on future
investment in education and training by the private sector.

POSTSECONDARY EDUCATION AND TRAINING

As described above, many young people seem to have responded to the
rising payoff to college. The proportion of 18- to 24-year-olds enrolled
in college increased by one-third between 1980 and 1994. Moreover,
college students are increasingly likely to earn degrees in the fields
where earnings are rising the most, such as in engineering, the
sciences, and the health occupations. But not all young people have
reacted similarly. Although college enrollment rates have increased for
most groups, differences in college enrollment rates by race and by
family income have widened since 1980.
One possible cause of the widening gaps in college enrollment rates
is the dramatic increase in the cost of a college education, at public
as well as at private institutions. Between 1980 and 1994 the real
average tuition at public 2-year and 4-year colleges rose by 70 percent
and 86 percent, respectively. Over the same period, however, the value
of the maximum Pell grant, the primary Federal grant program for low-
income students, fell by more than 25 percent in real value. Not
counting parental borrowing, the maximum amount a dependent
undergraduate student could borrow over 4 years of college also declined
by 5 percent in real value (Chart 7-4). Even if one takes State and
institutional need-based aid into account, the net cost at a public 4-
year college for the average youth with family income in the bottom
quartile rose between 1987 and 1993.



The college entry decisions of young adults, particularly those from
low-income families, seem to be quite sensitive to increases in tuition.
A number of studies have attempted to measure this price sensitivity by
comparing enrollment rates in high- and low-tuition States. These
studies suggest that a $100-per-year difference in college tuition
levels is associated with a 1.2 to 1.6 percent difference in college
enrollment rates among 18- to 24-year-olds. Some recent evidence also
suggests that those States that have raised tuition see slower rates of
growth in enrollment, and that the gaps in enrollment rates between
high- and low-income youth have grown most in those States that have
raised tuition.
Rising costs were not the primary cause of rising tuition at public
institutions. Educational expenses per full-time student (including
costs of instruction, administration, student services, libraries, and
operation and maintenance of physical plant, but excluding sponsored
research and scholarships and fellowships) rose by only 15 percent in
real terms between 1980 and 1992 at public 4-year colleges and by only
12 percent at public 2-year colleges. Rather, public tuition rose
primarily because State and local taxpayers were paying a smaller
percentage of the cost than they had in the past. As enrollments have
risen and as other demands on State budgets have grown, States have
responded by raising tuition rather than increasing their appropriations
proportionately.

Reforming Student Aid Policy

Given the forces at work, the Nation faces a number of difficult
choices in the financing of higher education. In addition to a
continuing increase in the demand for a college education, demographic
trends indicate a 20 percent increase over the next 15 years in the
population of traditional college-age youth. In some States, such as
California, the demographic shift will be even more pronounced. Unless
State budgets for higher education grow, public tuitions are likely to
continue rising, not because costs are rising, but because State
appropriations will be spread over larger enrollments. This will make a
college education even less accessible for many Americans. Therefore
Federal student loan and grant programs are likely to be more critically
important than ever before.
To meet these new challenges, the Administration's direct lending
program has sought to provide educational financing in a less costly,
less cumbersome manner, with more flexible terms of repayment. The
Federal Government issues loans to students through the financial aid
offices of colleges, bypassing the more than 7,500 private lenders, 41
guaranty agencies, and 90 secondary market participants that make up the
Federal Family Education Loan (FFEL) program.
Under the FFEL program, the Federal Government guarantees a return
to banks that provide financing for student loans. Under the direct
lending program, on the other hand, the Federal Government provides the
capital. Whether or not direct lending saves taxpayers money depends on
whether the Department of Education can service the loans for less than
the subsidies it pays the private banks to carry the loans. Based on the
prices it has already negotiated with private contractors to service the
loans, the Administration believes that the program can deliver
substantial budgetary savings. At the time the Student Loan Reform Act
was passed in 1993, gradual conversion to direct lending was projected
to save more than $4 billion over 5 years.
However, the debate over the cost savings generated by direct
lending has overshadowed discussion of the quality of service received
by students and colleges participating in the program. On this question
there seems to be little disagreement, at least among the colleges and
students themselves. Direct lending clearly provides more timely, more
accessible service to students and universities. After the first year of
direct lending, in which 104 schools participated, a survey funded by
the Department of Education revealed that 61 percent of participating
schools reported themselves very satisfied and an additional 28 percent
were satisfied. The General Accounting Office (GAO) also evaluated the
program. Officials interviewed at 11 of the 17 schools examined by the
GAO described themselves as greatly satisfied with direct lending, and
the remaining 6 reported being generally satisfied. None of the schools
reported serious misgivings. The GAO report also cited a number of ways
in which direct lending helped students and universities: parents and
students do not have to file separate loan applications to banks;
students receive their loans more quickly; students know whom to contact
for deferments or other questions, because their loans are not resold;
and each college works with a single lender, the Federal Government,
rather than hundreds of financial intermediaries.

More Flexible Options for Repayment

The average student borrower completing 4 years of undergraduate
education today leaves school approximately $11,000 in debt. As loan
burdens grow with ever-rising tuitions, flexibility in the terms of
repayment can lighten the burden significantly. The direct lending
program offers four different repayment options to provide such
flexibility: the standard plan, the extended plan, the graduated plan,
and income-contingent repayment. Private banks also can offer some
choice in the form of repayment.
Under the standard repayment plan, borrowers pay fixed nominal
monthly payments over a 10-year term. At an annual interest rate of 8.25
percent, a borrower with the average debt for someone finishing a
bachelor's degree pays $135 per month. Under the extended repayment
option the same borrower would pay $107 per month, with payments spread
over 15 years.
Under both the standard and the extended plan, the nominal payment
is fixed over the term of the loan, so that the real value of the
payment actually declines over time. However, a declining real payment
schedule may impose unnecessary hardship since young college graduates
often earn significantly more after a few years on the job than they did
immediately out of college. The graduated plan therefore attempts to
ease their debt burden by matching payments more closely to this
expected rise in earnings. For instance, a borrower with $11,000 in debt
would make payments of $77 per month during the first 2 years and end
with a $175 monthly payment during the 15th year.
The income-contingent option is even more flexible: monthly payments
are calculated on the basis of the borrower's adjusted gross income, as
reported by the borrower and verified by the Internal Revenue Service.
The above graduate starting his or her career making $18,000 and
enjoying annual earnings increases of 5 percent would begin by paying
$90 per month and end, after 15 years, paying $121 per month. Borrowers
whose earnings are so low that they still have loan balances after 25
years of repayment will have those balances forgiven. Income-contingent
student loans may thus be viewed as an innovative form of ``forward-
looking'' means testing (Box 7-4). Although it is too early to tell,
more flexible terms of repayment may also lower default rates by helping
to deter borrowers from getting behind in their payments early in their
careers.

Box 7-4.--Income-Contingent Student Loans as Forward-Looking Means
Testing

Means testing in student aid programs ``taxes'' the income and
assets of parents and students at a high rate by providing less aid for
those with higher incomes or more assets. Because the implicit taxes
apply for every year that one has a child in college, the marginal tax
rates on savings can approach 50 percent for families with two children
attending college for 8 years. In other words, for every dollar in
savings above a threshold, parents may lose 50 cents in financial aid,
lowering parents' incentive to save. In the past these very high tax
rates did not apply to very many families, because many families'
incomes were too high to qualify for any aid. However, as tuition levels
rise, the marginal tax rates apply to an increasing number.
High marginal tax rates are an inevitable result of ``backward-
looking'' means testing, in which financial aid is distributed according
to the recent past income and assets of applicants and their parents
(usually a single year of income and assets). In contrast, the income-
contingent loan program may be thought of as a form of ``forward-
looking'' means testing. It has three advantages: it targets resources
on those with low earnings after they leave college (rather than just
low family incomes in the year before they enter college); it provides
some ``insurance'' to students from middle- and higher income families
who may be anxious about their future labor market prospects given a
large debt; and it broadens the base of income used for means testing
from a single year to the student's whole career. Because parents'
savings are not taxed when means testing is forward-looking, parents may
even save more to contribute to their children's education. Moreover,
this forward-looking means testing is more suited to the needs of older
workers seeking to return to school, since the traditional backward-
looking financial aid formulas were often designed with traditional
college-age dependent students in mind.
In a time of rising tuition and strained public budgets, publicly
guaranteed loans make the most of public resources while ensuring that
young people use the Nation's educational resources prudently. The
availability of the income-contingent repayment scheme protects those
with very low or highly variable earnings later in their careers. If
tuition levels continue to increase, limits on student borrowing under
both the direct lending and the FFEL programs may need to be raised in
coming years. At present, dependent undergraduate students (those who
are unmarried, not veterans, with no dependents, and less than 24 years
of age) can borrow only $2,625 during their first year in college,
$3,500 during the second year, and $5,500 per year during the junior and
senior years. Parents are allowed to borrow more under the Parental
Loans for Undergraduate Students (PLUS) program. However, since payments
on PLUS loans begin immediately, many parents may be reluctant or unable
to take on the additional burden. Tuition expenses alone exceed the
$2,625 limit at a group of public 4-year institutions that together
enroll 42 percent of all undergraduate students. As a result, unless
borrowing limits are raised, an increasing number of dependent students
will not even be able to borrow enough under the Federal programs to pay
their college tuition and living expenses.

Default Rates

Ever since the inception of the Federal student loan programs,
defaults have been a significant concern. This concern was heightened,
however, when default claims paid to lenders exceeded $2 billion for the
first time in 1989. Under this Administration, the Department of
Education has made lowering student loan default rates a high priority.
Default rates differ markedly according to the institution the borrower
attended. Therefore the Department of Education has imposed standards to
preclude schools whose attendees have high default rates from receiving
federally guaranteed loans: postsecondary institutions can lose
eligibility to participate if they have a default rate in excess of 25
percent for 3 consecutive years. (The default rate is calculated as the
percentage of loans going into repayment in a given year that default by
the end of the following year. This threshold has been lowered from 35
percent in 1991 and 1992.) Approximately 250 schools have been declared
ineligible to participate in the loan programs based upon their 1992
default rates. An additional 190 schools have appealed the calculation
of their default rates, and it is anticipated, based on past appeals,
that many of these institutions will also lose eligibility. Although it
is difficult to distinguish the impact of regulatory efforts from the
effects of an improving economy, the default rate has been cut nearly in
half over the past few years: from 22 to 12 percent for debts going into
repayment in the years 1990 and 1993, respectively.

Future Challenges

A college education is becoming both more expensive and more
important for a successful career. The combination of these two trends
is making parents and students increasingly anxious. The Federal
Government provides a number of separate grant, loan, and work-study
programs for college students, but this variety of programs may itself
add to the lack of transparency in the financial aid process, increasing
families' anxiety. Students and their parents could make better
decisions regarding college if they knew more about how much they could
borrow or receive in grants and how much they were likely to have to
finance out of their own income and savings. Complicated means tests
necessarily make it difficult for students to anticipate the exact
mixture of grants and loans they will receive. Even so, there could be
much better information about the size of the total package available.
Moreover, parents and students who are worried about rising debt burdens
may find that the more flexible options for repayment now available help
relieve their concern.

BETTER OPTIONS FOR THOSE ALREADY IN THE LABOR FORCE

As different skills appreciate or depreciate in value, workers must
have the opportunity to react to these changes in the labor market. As
proposed in the G.I. Bill for America's Workers, the Administration has
also been working to reinvent how the Nation delivers education and
training services to those already in the workforce. Both the Congress
and the Administration have proposed consolidating many of the separate
education and training programs now administered by the Departments of
Labor and Education and providing block grants to the States. These
reforms are intended to convert our unemployment system into a re-
employment system. Although the proposals differ in some details--
particularly in the level of funding--they are similar in at least two
important dimensions.
First, States would coordinate the delivery of employment and
training services through one-stop career development centers. The goal
of the one-stop centers would be to allow workers to find out about
employment opportunities, apply for jobless benefits, learn about
available training programs, and receive assistance in financing that
training all in one place. Sixteen States have already received
multiyear implementation grants from the Department of Labor to begin
integrating an array of education, training, and employment programs
into the one-stop centers. The remaining States, which are at an earlier
stage in the process, have all received grants to plan the transition to
the one-stop concept.
Second, the Congress and the Administration have both proposed
consolidating more than 70 existing training programs and giving
training recipients the ability to choose the program that best meets
their needs. Under the Administration's proposal, dislocated and low-
income workers would be eligible for so-called skill grants of up to
$2,620 per year to complete an associate degree, enough to cover
tuition, supplies, and fees at a typical community college. Other
proposals would provide the funding to States in the form of block
grants but would also encourage States to allow recipients more
discretion in choosing the training program that is right for them.
Unlike the current system, in which government agencies often choose
what training workers will receive and who will provide it, grants could
be used by workers themselves to find the best match among eligible
training providers. But any worker, regardless of his or her income or
employment status, could use the centers to learn about training and
education options and would receive guidance in applying for educational
loans.
Both reforms are intended to enhance accountability among providers:
training providers that do not attract workers' interest would be
allowed to founder and the more successful programs to flourish.
Accountability will be enhanced if the quality of information available
to workers for assessing different programs, such as graduation rates or
placement rates (using, for instance, unemployment insurance wage
records to track the employment histories of graduates of each program),
can be improved. By voting with their feet, workers themselves will be
empowered to shut down ineffective training programs and expand those
that meet the changing needs of the labor market--decisions that may be
more difficult for program administrators to make.
The $10,000 tax deduction for tuition expenses in the Middle Class
Bill of Rights (described in Chapter 3, Box 3-4) will also lower the
cost of further training for those workers going back to school, as well
as for families with dependent children struggling with large tuition
increases.

CONCLUSION

Ever since the Nation's founding, the Federal Government has been a
partner in education and training. It has served as a clearinghouse for
research and evaluation results, contributed to equality of educational
opportunity by targeting resources to low-income schools and college
students, and guaranteed educational loans for college students. No
other layer of government could assume these responsibilities as
effectively and efficiently.
In addition to these traditional responsibilities, the Federal
Government must also help coordinate a national response to the dramatic
changes in the labor market. The Federal Government has responded by
providing funds to States interested in developing new pathways from
school to work. To add focus and momentum to school reform efforts, the
Department of Education has offered seed money to States for the
development of voluntary content standards in core subject areas and has
encouraged States to develop testing tools for measuring their progress.
Federal grants have supported the startup of charter schools and
investments in educational technology. In these new endeavors, the
Federal role is properly understood as that of a catalyst--vital but
temporary.
Progress has been made. Despite some year-to-year fluctuations, test
scores in math and science have risen for all age groups since 1980.
High school graduation and college enrollment rates have also risen. But
this is no time to drastically scale back those efforts. The shift in
demand has continued to outpace the increased output of more skilled
workers: earnings differences between the more and the less educated
continue to widen. Someday the increase in supply may begin to overtake
the increasing demand of the labor market and dampen future increases in
wage inequality, but at least until that day arrives, the Federal
Government must continue to support State and local efforts to transform
their educational systems.
In the midst of efforts to balance the Federal budget, it is
important to keep in mind that the objective of deficit reduction is to
spur long-term economic growth by freeing up more of the Nation's
savings for productive investment. To cut investment in education and
training simply for the sake of balancing the Federal budget in the
short term runs counter to that goal. Education and training have always
been a major source of U.S. growth; as the economic returns have
increased, these undertakings should represent a larger share of the
Nation's investment portfolio, not a smaller one. As families and
communities respond to the rise in the payoff to skill by investing in
themselves, the Federal Government should not shrink from the task of
encouraging and complementing their efforts.