Archive for the ‘Student loan’ category

Mismeasuring the Burden of Student Loan Debt

October 14, 2015

Rising higher education costs along with the volume of outstanding student loans, now in excess of $1.3 trillion nationally and greater than the volume of credit card or motor vehicle loan debt, are prompting  concerns about the impact that student loan debt is having on economic growth. Student loan debt grew at the fastest rates on record during the 2000s, doubling from $600 billion to $1.3 trillion over the past decade. Popular reports annually rank the debt loads of students graduating from colleges and universities in each of the 50 states. New Hampshire, is notable for being at the top of the list as having students graduate with the highest levels of debt in the nation.

debt of grad 2013

But the average debt levels of recent college graduates in any state says little about the impact that student loan debt has on a state’s economy. First, the schools from which students graduate aren’t necessarily the states in which students choose to reside (and repay their debts) after graduation, and second,  reports of the average debt levels of recent graduates provide no information about the outstanding balance of student loan debt (and thus overall student loan debt burden) held by residents of each state. The latter is necessary to understand the impact that student loan debt is having on a state’s economy.  I had not seen data on the balance of student loan debt on a state-by-state basis until a journalist (Ryan Lessard of the Hippo Press here in New Hampshire) passed along data from the U.S. Department of Education that was recently released by the White House. The data includes information on federal student loan debt only, and does not include private student loans or other loans used to pay for college – such as home equity loans taken out by parents, but is still extremely useful in understanding the differential impact of student loan debt in each state. The data present a different view of the student loan debt issue than do the data released annually on the debt of recent college graduates. In this post I add some economic and demographic data to the student loan debt data from the Dept. of Education to examine different measures of the relative burden that student loan debt places on individuals, and thus the economy of each state.

As of January of 2015 there were 212,000 individuals residing in New Hampshire with outstanding federal student loan debts totaling $5.1 billion dollars according to the U.S. Department of Education. The $5.1 billion compares to my estimate of $4.5 to $5.6 billion in credit card debt and $37.8 billion in home mortgage debt in the state.  In contrast to reports showing that the most recent graduates of colleges in NH have the highest student debt levels, the average outstanding loan balance among all of NH’s borrowers (regardless of where or when they graduated), at $24,048, was near the bottom of all states.

outstanding balnace by state

As I documented in a recent study of student debt, New England and the Northeast have the highest college costs in the nation, with graduating student’s debt levels similarly high. So why would NH’s average outstanding student debt balances be among the lowest in the nation? If NH residents with student debt had been paying off those debts for a longer period of time (that is borrowers were longer removed from college i.e. older on average) then their debt levels would be relatively lower even if their original debt levels were higher on average. In addition, if recent grads in NH, and their higher debt levels, leave the state, while somewhat older individuals move into the state, the state would be trading individuals with higher debt levels for those with more modest student debt levels. This seems like a plausible explanation based on some of the analysis of NH’s demographic trends I’ve written about in this blog and elsewhere. In addition, some of the discrepancy results from the new data on total student loan balances by state that includes all debt from students at two and four year colleges, as well as graduates and those with debt but who did not graduate. Thus the data released by the White House is a much more comprehensive measure of student loan debt at the state level. In addition, because it aggregates student loan debt of individuals who reside in each state, it is a more appropriate measure of the burden of student debt on any state’s economy.

Student loan debt is a problem, it has retarded household formation in NH and the U.S. and contributed to a slower than anticipated recovery in the housing market.  It has other negative impacts on younger individuals and families as well, but how large of a burden is student debt on any state’s economy and what is the best metric to assess it? It is not an easy question to answer.  The White House (Dept. of Education) data helps tremendously but analyzing it raises almost as many questions as it answers. The $5.1 billion in federal student loan debt held by borrowers living in NH represents about 7.1 percent of the state’s 2014 gross state product. Using this measure , NH ranks in the middle of all states on student loan debt burden, higher than indicated by the average student loan debt in the state. Because NH has a high percentage of students who have attended (and graduated) from college, even with below average student debt levels among all borrowers, the aggregate debt as a percentage of the state’s economy is higher than in states with lower average levels of debt among borrowers.  States with a high percentage of college attendees and graduates in their populations are likely to have a higher student loan debt to GSP ratio regardless of the average outstanding loan balance of borrowers. But is 7.1 percent a problem for the state’s economy?

debt as a pct of gspI think the student loan debt burden is probably better understood from its impact on individuals.   Only about 20 Percent of the adult population (age 18+) in New Hampshire have student loan debt and the debt has its greatest impact on a subset of the adult population. The typical repayment period of student loan debt is 10 years so, in theory, the population between graduation (or leaving school) and the age of about 35 should be most affected by student loan debt and assessing the impact of student loan debt should focus on impacts among this demographic group. For this analysis I use the characteristics of each state’s population ages 24-34 to assess the relative impacts of student debt on each state. The chart below uses the average outstanding student loan debt in each state and the average annual earnings of residents age 24-34 in each state to calculate how much of the annual earnings of 24-34 year working individuals with at least an associate’s degree go to student loan repayment in each state. Using the average outstanding loan balance in each state and assuming a combined federal subsidized and unsubsidized loan  interest rate of 4.5 percent, on a monthly basis, almost all states have average student loan burdens that require monthly payments of less than $300. The one exception is DC, not presented on the graph, where the $40,000+ average loan balance and $413 monthly payment is attributable to the high percentage of law school and other professional and advanced degree student who reside in the city.

monthly paymentA monthly student loan payment of $300 is not an inconsequential amount but less than most new car loan payments. Still, as a percentage of annual earnings, student debt payments clearly could influence the ability of younger people to purchase a home or make other significant financial commitments.  Combining monthly payments (annualized)  with the average annual earnings of college graduates ages 24-34 living in each state provides a measure of student loan debt service as a percentage of the earnings of graduates in each state.  Again, the chart shows that New Hampshire, along with several other states with both high college costs and high debt, rank relatively lower on repayment as a percentage of annual income.

burdens as a pct of earnings

The examples of several states highlight the importance of different variables in assessing the impact of student debt on any state’s economy.  The average debt of recent graduates from colleges in Vermont is in the middle among all states, yet the average loan balance of all borrowers in the state is higher than the debt levels of recent grads.  As a percentage of the earnings of working college grads ages 24-34, however, student loan debt in Vermont is the highest among all states. This suggests that recent grads (with their moderate level of debt) may be leaving Vermont while the state attracts or retains individuals with higher levels of student debt. It also suggests that the high percentage of the earnings of 24-34 year olds in the state that is absorbed by student loan debt service is, in part, a function of relatively modest average earnings  in the state.

avg debt and pct of earnings scatterplot

Another illustrative example is Georgia, a state with a relatively low average debt among recent graduates from its colleges, but with the highest level of debt among all borrowers of any state. From my limited experience in Atlanta, it is seems the city hasn’t been as overrun with northerners since Sherman’s march to the sea. This time the northerners have come armed with college degrees and promissory notes.  A state with below average student debt among recent graduates from its colleges but with above average student debt among all residents can’t address it’s high student loan debt burden by increasing state support for colleges or by providing more student aid.  Georgia appears to be gaining individuals with higher levels of educational attainment (“talent”) at a cost of higher student debt levels and greater debt burden among its residents. That is not a bad tradeoff as the state gets a more skilled workforce at a low cost to state government. Georgia reinforces a point that I repeatedly make, the importance of being attractive to skilled individuals with higher levels of educational attainment. NH makes this point as well, it has the highest average debt levels of recent graduates but relatively low average student debt for all borrowers in the state. We know NH losses a lot of its recent graduates to other states as I have documented in this blog and elsewhere, but attracts a lot of college graduates from other states, especially in the 25-40 age range.  These individuals, if they have student loan debt, have likely paid-off a good portion of it.  NH too has upgraded the skill of its labor force at a relatively low public cost by importing or attracting talent from other states.

New Hampshire, Vermont and Georgia are just three of many examples of how the debt levels of recent college graduates in a state must be interpreted with caution and in particular, when debating state-level policies directed at rising student debt levels. This brief analysis suggests different ways to assess the burden that student loan debt places on the residents of any state as well as on a state’s economy and shows that those burdens cannot be simply assessed by the most common assessment, looking at the average debt of recent college graduates. Reports on the average student loan debt of recent graduates by state can be an especially misleading indicator of the burden student loan debt places on any state’s economy.  I am not arguing here that student loan debt is not a problem, but like most public policy issues it is subject to errors of popular sentiment and conventional wisdom that can distort decision-making by policymakers. My purpose in this post is to explore some alternative measures (other than the average debt of recent graduates) of the impact that student loan debt has on each state’s economy. I welcome suggestions for better measures or criticisms of the ones examined or the methodology in this post.