Archive for the ‘higher education’ category

“…..No Fury Like the Middle and Upper Classes Scorned”

November 1, 2013

College students rage against  a lot of things (including their parents as I have learned) but tend not to channel that energy into exercising  their right of franchise.   I think that helps explain why, after decades of rapidly rising college costs and levels of student debt, the tipping point on the issue appears to have finally been reached.   As I noted in a recent study on student debt, one of the features today that distinguishes debt to pay for higher education from prior periods when the issue bubbled to the surface is how much more of the debt appears to be being borne by parents as well as students.  Parent debt is not included in the popularly reported student loan debt by college or by state, but as the chart below shows, along with unsubsidized federal loans to students, federal parent “PLUS Loans”   have been the second fastest growing category of student debt.

Growth in Student Loans by Type

PLUS Loans are still a much smaller category of debt for higher education than are loans to students but there is no data available on the non-federal loans parents have accumulated (such things as home equity loans) to pay for their children’s education so overall parent debt has likely increased even more .  It is much easier to avoid the public debate about debt for higher education when it affects just students than it is to avoid when  it affects parents as well ( whom lawmakers are much more likely to respond to as constituents).  Another troubling aspect of the chart above is the increase in federal unsubsidized student debt.  Because the interest on this debt is not paid for by the federal government while a student is in college (unlike subsidized debt), it is more costly, and thus the higher percentage of unsubsidized debt today than a decade ago suggests that student debt levels actually understate the impact on students of the increase in student debt since 2001.

Examining data from the Federal Reserve Board’s “Survey of Consumer Finances” highlights how debt for higher education has increased over 20 years as a percentage of household’s non-mortgage debt, with the biggest increases coming since the mid 2000s.

H Installment Debt by Purpose

The chart above reflects both parent and student debt.  More of household installment debt going to pay for education implies less borrowing (and spending) for other purposes and helps explain (along with generally weaker economic conditions during much of the 2000s) why consumer expenditures on other goods have been relatively weak.  The economic implications of student debt are typically seen as constraining recent graduates spending, household formation, home buying, entrepreneurship, etc., but the reality is that student debt is likely constraining more than just recent graduates.  I think some evidence of this is seen in the changes in installment debt of households by income (chart below) .

Installment debt by income

As the chart  shows, the largest increases in  the percentage of installment debt that goes for education are among middle and higher income groups.  These are income groups with the most disposable income and anything that constrains them from “disposing” of their income (like repaying debt for higher education) on goods and services that boost economic activity will have detrimental impacts on the economy.  Lower income groups have seen significant increases in installment debt for education as well, much of it attributable to increased enrollment rates over the past few decades which have been highest among students from lower-income households.  With more students from families with modest incomes attending college, the share of installment debt for education among these income groups can be expected to rise.  Higher college costs also play a role but increases in financial aid for lower-income students have helped offset some but certainly not all of those cost increases.   In no way do I want to minimize the impact of college costs and debt on lower-income students and households but that has been a fairly consistent problem that does not distinguish the current situation from the past the way changes in debt among higher income groups appears to have changed.

The large increase in the percentage of installment debt that goes to education among middle and higher income groups reflects economic conditions (savings for education dropped the most among these income groups over the past decade) but also financial aid policies and practices of governments and colleges that appear to especially squeeze middle-income households.

The rapidly rising debt for education held by parents and households, not just students, I think explains a lot about the new urgency to address college costs and student debt.  The fact that middle and higher income groups seem more affected by these pressures than in the past may say even more about why the issue has risen on the public agenda.  Beyond the political, when the households that typically have the most disposable income appear to be especially affected by higher education debt, we should not discount the role debt for higher education may be playing in constraining economic activity in the U.S.

Competition Raises Rather Than Lowers College Costs

October 9, 2013

I’ve written about student loan debt and college costs in the past.  A report I recently completed on the issue was released earlier this week, you can read it here .  It is long and detailed so even if your aren’t that interested in the topic it is an effective substitute for a strong drink or a mild sedative as a sleep aid.

The report didn’t get a lot of media coverage and in any case it is difficult for media coverage to capture more than a few highlights of a long study that deals with complicated issues (although Kevin Landrigan at the Nashua Telegraph did his usual good job, as did Kathleen Ronayne at the Concord Monitor).  One small portion of the study that I think is interesting and important examines the issue of competition in higher education as a contributor to higher tuition levels.  Colleges in New England and the Northeast have higher tuition prices on average than colleges with similar characteristics (public, private, selectivity, state support, etc.) located elsewhere in the country.  No doubt the region has some of the finest higher education in the country but not every institution in the region is world-class with an ability to charge a premium for educating students.  I thought state-level differences in cost-of-living would account for some of New England’s and the Northeast’s higher tuition prices but multivariate (regression) models using data from about 700 four-year colleges and universities across the country failed to demonstrate that cost-of-living differences were significantly related to differences in tuition prices among universities with similar characteristics but in different regions of the country.

Tuition FactorsOne characteristic of higher education in New England and the Northeast that distinguishes them from other regions of the country is how many colleges and universities are located here.   It is possible that increased competition for students and for faculty could increase college costs in New England and elsewhere.   To date, few colleges and universities have been willing to compete on price.    Colleges generally compete by offering more and better faculty, facilities, student services, and amenities. Colleges want to be the best they can be and they compete for students, for faculty, and with businesses for talent (PhD’s in fields in demand). This is especially true in a region like New England where there is a high concentration of higher education institutions and where there is a large base of technology, business, and professional employment that is more likely to employ individuals with advanced degrees and to compete with colleges for available “talent”.

To test the degree to which competition among colleges might influence regional college costs and prices, I had to develop a meaningful measure of competition among colleges. The study operationally defined the level of competition among colleges in each region as the percentage of regional employment that is employed in higher education.  Specifically, I calculated a “location quotient” for higher education employment in each state and averaged the location quotients for each state in a region to arrive at regional location quotients for nine census divisions (regions).  This measure roughly approximates the degree of choice students in each region have regarding college enrollment.  The figure below highlights regional differences in the concentration of higher education employment.  I won’t get into a discussion of location quotients (LQ) but a LQ of greater than 1 for a service industry indicates it is serving more than local markets and is also “exporting” its services.  We know colleges in New England attract student  nationally and internationally.  The darker regions of the country are where there is a greater concentration of higher education employment (and thus the potential for competition among them).

Higher Ed Emp Concentrations

Testing the impact of the LQ measure of regional competition on tuition and fee levels shows the variable to be significantly related to tuition levels at both public and private colleges, with regional location quotients showing a larger impact on tuition and fees at private colleges than at public colleges (implying that competition for students at private colleges has a greater impact on prices than it does at public colleges) .  The elasticity of tuition and fees with respect to our measure of competition was small (.10) at private colleges, and even smaller (.076) at public colleges. However, as the chart below shows, the magnitude of the difference between higher education employment in New England and other regions is great (indicating much higher levels of competition among colleges in the New England region), so this small elasticity still implies that competition has a relatively large impact on tuition and fees at colleges in New England.

Higher Ed Location QuotientsAs an example, New England’s location quotient of 2.35 is about 140 percent larger than is the location quotient (or concentration of higher education employment) in the South Atlantic region. The elasticity of tuition and fees at private colleges with respect to this measure of competition suggests that for every 10 percent increase in higher education competition in a region, tuition and fees will be one percent (1%) higher. Thus the 140 percent difference in higher education competition in New England compared to the South Atlantic region implies that all else equal, we can expect tuition and fees at private colleges in New England to be 14 percent higher than in the South Atlantic region.  For public colleges, with a smaller elasticity of tuition and fees with respect to competition, these results imply that tuition and fees would be about 10 percent higher in New England as a result of higher levels of competition among colleges in the region.

I think these findings have important implications for tuition prices, student debt, and higher education in New Hampshire and New England and their economies.  The biggest reason student debt levels are so high in New Hampshire is the fact that the primary strategy used by families for reducing the cost of college and for limiting student debt is attending a lower cost, in-state, public institution.  Over the past decade more New Hampshire families have attempted that and with little success because of the high cost of attendance at New Hampshire’s public colleges and the low levels of grant aid awarded by them.  State aid plays an important role in tuition prices at public colleges (just as endowments do at private colleges) and has a big impact on NH’s prices but even accounting for differences in state aid, competition for students and faculty appears to be contributing to higher tuition prices among  public (and private) colleges in the region.

Like all of us colleges want to be the best they can be and they compete with other colleges on national and regional rankings to prove it.  A sad byproduct of that is that most rankings count per student spending heavily in their assessments so that any college that can offer as good or better educational services while restraining expenses will suffer in national rankings.  It is important for public institutions to remain a key strategy for families in limiting the cost of college and student debt.  It is probably more important now than national rankings and competing for a shrinking population of college age students.

Competition is a wonderful thing and it is addicting (I know –  it is my drug of choice).  I am a believer in markets and I believe that the market for higher education is changing and will result in more price competition among private and hopefully public institutions. The competition won’t be on price exactly,  it will be based on “value” not just price (it won’t do anyone any good to say I got the “cheapest” education unless the result of that education is greater opportunity for success in the labor market and in life).  Competition among public colleges and public colleges with private colleges could accelerate market-based changes  in higher education by focusing on value, not on facilities, amenities, high-profile faculty or programs, or whatever basis colleges typically compete on.  Given the fiscal realities of states, colleges will not be able to compete with private institutions by spending more to be the best.  When you can’t effectively compete on one field you are wise to move the competition to one more favorable to your chances of success.  Public colleges can do that and parents and families in New Hampshire and New England looking to limit college costs and debt would be winners if they did.

Is Your State Overrated?

May 24, 2013

My penance (and your burden) for being an absent blogger over the past week or so is a longer post with extra graphics today.

A lot of people, including me, are accustomed to assessing the overall skill level of a state’s or a region’s workforce (and thus its potential to capture growing industries that rely on more highly educated workers) based largely on the percentage of the workforce with a college degree.  It is simple, intuitive,  and more than a  little lazy.  It is also becoming a  less useful indicator of the supply of labor that is in demand by businesses.   Populations with higher levels of educational attainment confer a lot of benefits on a state or region but today, having a high percentage of a state’s or region’s population with at least a BA degree probably says as much about the state’s political and cultural sensibilities (as well is its “demand” for services rather than its “need” for services but that is another post)  than it does about its economic performance and potential.

The sense of self-satisfaction we in New England and in New Hampshire enjoy about  having a population with among the highest levels of educational attainment in the country is palpable, but the reality is that more states are increasing their levels of educational attainment and New England and the Northeast stand-out far less than in the past.  Moreover, in an economy that is increasingly rewarding particular skills and degrees more  than  just high educational attainment, it is not as clear that much of the region still has an edge  on the one resource that  it has that is always in demand – talent.

Much of New England and the Northeast has a high percentage of its adult population with a four-year college degree or higher (see chart below).

% With BA or Higher

Just looking at levels of educational attainment tells only part of the story.  I can’t blog for too long without talking about the “skills-gap”  so here goes.  Much of the demand for college-trained labor is in fields that require scientific, technical, engineering,  or mathematical (STEM) skills and degrees.  The percentage of a state’s population with a BA or higher degree tells a lot about the availability of STEM skills but for a number of states it tells a lot less.  I calculated the percentage of a state’s population with a STEM degree (based on first college degree earned) and included it above as the dark blue portion of the bar graph.  The official listing of STEM fields is maintained, surprisingly, by the Dept. of Homeland Security ( I categorized 171 college degrees into STEM and non-STEM degrees and I think my listing is close but not a perfect match).  If you compare  the percentage of the population with a four-year or higher STEM degree (chart below) with the percentage of the population with a BA degree or higher (chart above) it shows a large change in the relative rankings of a number of states, and a some in New England in particular.

% with STEM degrees

The final chart makes just that comparison, it shows the change in ranking  between a state’s position on the percentage of its adult working population with at least a BA degree and the percentage of its population with a STEM degree.  The chart highlights states that may be over and underrated on the skill level (at least skills in demand) of their workforce.   Vermont stands out as having the biggest drop in rankings between the percentage of its population with at least a BA degree and its ranking on the percentage of the population with a STEM degree.  Maine also fares poorly.  But New Hampshire, Rhode Island, and Connecticut also drop in rankings when measuring “talent in-demand” among the workforce.  Only Massachusetts does not  drop in ranking  ( it is ranked number one on both measures so there is no way it could show anything but a drop in relative rankings).   On the other hand, states that are often derided by Northeastern “elites”, such as Texas, Arizona, Florida and Alabama  have a smaller percentage of college-trained labor but more of them (on a percentage basis) are trained in the STEM fields most in demand.  Still, they  don’t have as high a percentage of their adult populations with a STEM degree as do some New England and some other states, but with population and migration trends, and as more individuals with those skills and more companies that want access to them agglomerate in those states, how long before some take the lead in “talent”?   I don’t think Massachusetts has as much to worry about as do other states in  New England because of their unique higher-education assets.  The question for the rest of us is, can we continue to “beggar our neighbor” and benefit from the Bay State’s ability to churn-out and attract individuals with the degrees and skills in demand?

over and under rated states

As New Jersey, New York, and Connecticut show, having a lot of “talent” in your workforce doesn’t guarantee strong economic growth.  The business, political, environmental, and even cultural and social climates also play an important role in promoting prosperity.  I look at states with a relatively higher  percentage of their college trained workforces  in STEM fields as “up-and-comers.”   Most don’t have the history of high educational attainment in their populations that New England does, so their overall ranking on educational attainment tends to be lower.  Some, like Texas and Arizona also have had a large influx of individuals with traditionally lower levels of educational attainment.  Nevertheless, they are accumulating and growing a larger portion of the nation’s “talent” in STEM fields and over the long-haul, that is the biggest threat to New England’s most valuable and most in-demand resource, and thus the biggest threat to its prosperity.

Productivity and Student Debt in NH

April 15, 2013

Student debt is in the news again today (as it is pretty much everyday) and because I am doing some work on higher education costs and debt, today I will highlight what some students (and others) at UNH were drawing attention to with their pencil sculpture using more than 30,000 pencils to highlight the average debt of students graduating from NH colleges and universities.

I’ve written before about how the cost of a higher education is affected by many things.  The level of tuition and fees, the volume and type of student financial aid, the demographics and characteristics of students at each institution as well as the interactions among all these factors affect student costs and debt.     It’s a complicated issue that seems to generate a wealth of simple and intuitive explanations.   NH policymakers debating the appropriate level of support for higher eduction are concerned with tuition prices, the efficiency of  higher education institutions, the impact of college costs on access to college, graduate’s debt, and the larger impacts of all of these on the economy.

The chart below highlights what NH students (and students everywhere) are concerned about.  The chart shows the average debt of graduates with debt (not all graduate s do so with debt) from UNH along with the average debt of students  from “flagship” public universities across the country.  Each state has one flagship university and I chose to use them here to control for the fact different types of institutions will have different characteristics that can affect average debt levels.  Just as importantly, state lawmakers in NH and elsewhere decide funding levels for public colleges and are often most interested in how public  higher education is affected by important issues.

UNH grad debt

The chart shows that the average debt of  UNH graduates (who graduate with debt) is higher and has risen faster than the national average for flagship universities.  The chart also shows that the per capita debt of graduates has been rising faster than the average debt of graduates with debt, both at UNH and nationally.    The more than 70 percent rise in per capita debt of graduates is especially troubling because it is a sign that more is being borrowed by each graduate but also that more students are graduating with debt.  Debt at UNH is higher but the problem is occurring everywhere.  The economic and social implications of the larger amounts of aggregated student debt on each successive cohort of young people are significant.

UNH’s high relative tuition among public, flagship universities across the country is not completely responsible for the debt levels among its graduates (although it is clearly a driving factor).  Demographic characteristics of student, financial aid policies, as well many other factors also play key roles.  UNH takes a lot of heat for its high tuition price and some see the high tuition as a sign of UNH’s inefficiency.  There are a lot of factors that help explain the rise in college costs but UNH’s high tuition isn’t readily explained by inefficiencies or “waste” compared to other public institutions.  Below is one  measure of efficiency I developed to compare colleges and their costs.  The chart below shows the direct educational and general expenditures per degree awarded at “flagship” public  institutions.  It is not a measure of quality.  A university could reduce its costs per degree awarded by teaching every class with 150 students or if it offered no student support services.  It will also reduce its costs per degree if it offered less grant and scholarship money.  The chart shows that UNH has the lowest cost per degree of any flagship institution in the country on this metric. In part it is a measure of efficiency as well as the characteristics of students because  a greater number and percentage of students who graduate will also lower expenditures per degree awarded.

exp per degree awarded

A couple of words about this metric.  First, I use a weighted degree measure that uses a bachelor’s degree as the baseline and assigns higher values to PhD and professional degrees and lower values to masters degrees to reflect the differing time (and thus costs)  related to obtaining each degree and to account for difference between institutions  on the mix of degrees awarded.  A university that awards medical degrees, for example, can be expected to have higher costs to educate each degree recipient (I need to examine how strongly the results are influenced by the weighting scheme).   I also adjusted the dollar figures to reflect differences in the cost of living in each state.  Wages, salaries and labor costs differ greatly across the country to compensate for differences in the cost of living and that should be reflected in the expenditures of public universities across the country.  It isn’t done often but when comparing expenditures of similar types of college I think it is a good idea.

Controlling College Costs One Student at a Time

April 8, 2013

Higher education costs, productivity, and student debt are issues in the cross hairs of policymakers and the public. Public colleges blame government support for trends in college costs, lawmakers blame colleges for poor productivity, parents blame both colleges and government, and the academy blames almost everyone for misunderstanding the issues or not being given the unconditional love that they believe they deserve.   Aside from my professional interest I have a strong personal interest in the issue as the father of college students.  One thing that I have learned from the blurred lines between my professional and personal interests in higher education is that both colleges and students (and their parents), have far more control over the costs of higher education than either seems to want to acknowledge and until each more fully exercises that control, some disturbing trends aren’t going to change much.

Trends in higher education costs and productivity are more complex than their popular treatment suggests but as I noted in a February post, the surest way to limit the cost of higher education is for students to graduate on time.  I’ve been looking at college completion and grad rate data and since it is impossible to highlight the rates of thousands of colleges I have chosen to highlight the rate at which students at  the “flagship” public university in each state graduate “on-time” (within 4 years for a Bachelor’s degree).  A quick look at the chart below shows what tremendous variation there is in “on-time” grad rates at public universities.  It is nice to see the University of New Hampshire with a relatively high “on-time” graduation rate.  I am sure there are other factors as well but “eyeballing” the chart also suggests how important student qualities and characteristics are to these data, as the “top” public universities in the nation that are more selective (UVA, UNC, UMICH, U Cal Berkely etc.) also have very high on-time grad rates.

On time grad rates at public universities

Speaking of the University of Virginia, here is a picture of one of the authors of this blog at the foot of greatness (a free, lifetime subscription to this blog for anyone who suggests that, in fact, that is where he is spends every day).

Gully and Jefferson

Yes it costs a lot to send a kid to college and I wish it were cheaper but the chances of that happening during the matriculation of my (or any) individual’s child are slim.  But we can price shop and perhaps more importantly make sure we have to pay for as few years of college as possible by graduating early or on-time and by choosing a college where we have the best prospects for that to happen.   I’ve been looking at college cost, finance, and completion and graduation data a lot lately, along with demographic, student ability, and other data on the characteristics of college students and their families.  I think it would be pretty easy to develop a discriminant function that would fairly accurately predict whether a student of differing characteristics and abilities would graduate and graduate on time from a particular institution and that might give parents a better metric of potential costs than does ‘sticker price” or “net price”. Choosing a college where can you pursue your calling is most important but price shopping and choosing to graduate on time are keys to minimizing college costs and they are both factors that are largely within the control of students and their parents.


%d bloggers like this: