What does the declining birthrate
mean for colleges and universities and the students who hope to get a
college degree a decade from now? The answer depends on where you live
in the United States and how selective the college is. For most colleges
and universities, the outlook is grim. But that could be a good thing
for their future students.
Nathan
Grawe, an economist at Carleton College in Minnesota, predicts that the
college-going population will drop by 15 percent between 2025 and 2029
and continue to decline by another percentage point or two thereafter.
"When
the financial crisis hit in 2008, young people viewed that economic
uncertainty as a cause for reducing fertility," says Grawe. "The number
of kids born from 2008 to 2011 fell precipitously. Fast forward 18 years
to 2026 and we see that there are fewer kids reaching college-going
age."
Birthrates failed to rebound with the economic recovery. The latest 2017 birthrate data from the Centers for Disease Control and Prevention posts new lows, marking almost a decade of reduced fertility.
But
not all colleges will feel the pain equally. Demand for elite
institutions – the top 50 colleges and 50 universities, as ranked by
U.S. News & World Report – is projected to drop by much less during
the 2025 to 2029 period (18 years following the birth dearth). And
student demand for elite institutions may be 14 percent higher in 2029
than it was in 2012. Meanwhile, regional four-year institutions which
serve local students are expected to lose more than 11 percent of their
students, from 1.43 million in 2012 to 1.27 million in 2029.
The
Northeast, where a disproportionate share of the nation's colleges and
universities are located, is expected to be the hardest hit. By
contrast, mountain states where there are fewer students and fewer
colleges, such as Utah and Montana, may see slight increases in student
demand.
Grawe's
forecasts for the number of students at two-year community colleges and
four-year institutions are published in his book, "Demographics and the Demand for Higher Education," with updates on his website.
He breaks the numbers down not only by type of school, and how
selective it is, but also by geographic region and race/ethnicity.
Why
do the forecasts sometimes move in opposite directions? Grawe explains
that elite colleges are less affected by the birth dearth because
they're a small niche market of fewer than 200,000 students that has
benefited from the explosion in college education since the 1980s.
"The
people who went to college 20-30 years ago and got a degree, they're
now the parents of kids who are looking at going to college in the next
10 years or so," says Grawe. "If your parents went to college, your
probability of going to college is much higher and your probability of
going to a highly selective four-year college is a lot higher."
Meanwhile, less selective institutions are not as likely to buck the powerful demographic trend of declining birthrates.
Northeastern
schools, especially those who cater to students who live nearby, are
feeling more pain because of demographic shifts that began well before
the Great Recession hit. Americans are continuing to move away from the
Northeast to the South, to places like Texas. In addition, birthrates
are lower in the Northeast where there is a smaller Latino population.
Latinos have historically had the highest fertility rates among U.S. racial and ethnic groups.
This
may be good news for students who are currently in fifth grade and
younger. Grawe predicts they'll have an easier time getting admitted to
schools as colleges fight more fiercely for the available students.
"Students
are going to be a hot commodity, a scarce resource," says Grawe. "It's
going to be harder during this period for institutions to aggressively
increase tuition. It may be a time period when it's a little easier on
parents and students who are negotiating over the financial aid
package."
For
the colleges themselves, declining student enrollments will likely
translate into leaner budgets. Regional colleges will be under pressure
to cut liberal arts courses and expand professional programs, such as
law enforcement, that students feel will translate into a good-paying
job. "As a liberal arts professor, it's heartbreaking," says Grawe. "But
you can understand. The institution's existence is dependent on meeting
the expectations of the student."
Some colleges won't make it. Moody's Investors Service is predicting
an uptick in closures of private colleges. And public colleges may have
trouble convincing state legislatures to fund them amid declining
enrollments.
Grawe
argues that colleges might be able to avoid closures and budget
shortfalls if they can reduce their dropout rates and focus on keeping
students – and their tuition dollars – on campus. Grawe cites the
example of the University of Southern Maine, which is coping with fewer
students but operating with a larger budget because of its efforts to
keep students through to graduation. Expect more colleges to launch
student retention and student success initiatives.
Of
course, Grawe's predictions may turn out to be wrong. Economists
predicted a similar drop in college enrollments in the 1980s following
the baby boom generation. Instead, the college-going rate skyrocketed.
Women started going to college in larger numbers. More young Americans
wanted a college degree as it became more difficult to get a good job
with only a high school diploma. Even older Americans went back to
school.
Could
something like that happen again? It's possible that the Latino
college-going rate could surge. It has already increased to more than 70
percent from 60 percent since Grawe first calculated his forecasts
using data from 2011 and earlier. But Grawe says it would be a "very
risky" strategy for college administrators to cross their fingers and
hope this demographic slump goes away.
This
story was written by Jill Barshay and produced by The Hechinger Report,
a nonprofit, independent news organization focused on inequality and
innovation in education.
No comments :
Post a Comment