Fintech and Millennials — Student Loans and refinancing

Published in

Build the future

8 min readSep 28, 2017

In this post, I’ll give an overview of increasing student debt levels and their performance over time — slicing and dicing the data by degree types, education level, loan amount etc. As we take a deeper dive in the problem, we’ll see who is most likely to default on loans and zero in on a few insights about those borrowers.

After that, I’ll take a look at various student loan refinancing providers and their target borrowers. We’ll see how various refinance lenders are focused on various types of borrowers and their profiles.

Student loans is the second largest category of debt for millennials in the U.S. There is a “student debt crisis” as income levels have stayed stagnant and the cost of education has outpaced everything else. Cumulative student debt levels crossed $1.3 Trillion in outstanding balance to more than 43 million students (11.9% of the U.S. population). This is the largest segment of debt after mortgages.

Below is a graph of student loan volume by year since ‘95:

Federal Student Loan Volume by year — dollar volume and distribution by loan type

In recent years, students borrow ~$100B in student loans every year.
Total loans per year have doubled over the past 20 years with $106 billion disbursed in 2015–16 calendar year.
Total loans peaked in 2010–11 at $124 billion (because of people going back to school after the crisis).
Graduate students are borrowing a lot more than undergraduate students on average.

The interesting thing here is that there are no credit requirements for Unsubsidized and Subsidized Direct Loans or Perkins Loans.

Federal Subsidized and Unsubsidized loans are the highest proportion of total loans.
There is a significant decrease in subsidized loans in the past 3 years. They are now only 22% — down from a peak of 55%.
Unsubsidized loans increased from 28% to 48%.

Interest Rates:

The average interest rate varies from 3.7% t0 6.9%. Federal Direct Subsidized loans are offered the lowest rates (3.7%) and Direct Unsubsidized Graduate loans are offered the highest (6.8%).

Below is a high-level overview of interest rates by loan type:

Borrowers and loan balance:

As we look at students taking loans for graduate and undergraduate degrees, we notice that a large portion of students going to graduate schools have higher debts. This is because graduate school programs such as MBA tend to be expensive.

Total Amount Borrowed for Undergraduate and Graduate Studies for 2011–12 cohort

Amongst the total debt outstanding by undergraduate and graduate borrowers, graduate borrowers owe 71% of it.
It’s even more prominent in the 120k+ category where 80% of the total debt is by graduate school students (we’ll see later that higher debt doesn’t mean higher default rates).

Average Loan Balance by Degree Type for 2011–12 cohort

43% of all graduate students have a debt > $40,000 while only 11% of undergrads do.
92% of graduate students have a debt of $5,000+.

Let’s take a deeper look at a few graduate school degree types and borrowing behavior of students going to school for those courses. Here, we will discuss doctoral degree and master’s degree students.

Percentage of graduate school borrowers by degree type for 2011–12 cohort

The percentage of graduate school borrowers in debt for doctoral degrees is really high
Only 45% of PhDs take student loans, most likely because their education is financed through research stipend
85%-90% of Medical and Law students take student loans for school and 75% of them take >$60,000. This forms a significant portion of debt by dollar loan volume.
We should keep in mind that these borrowers with high debt also earn significantly higher salaries after graduation. Doctors and Lawyers are one of the highest earning professions in the U.S.

Now, we will look at the outstanding debt of these borrowers for the same year. The average loan balance is only for students who had taken a loan.

Average loan balance of graduate school borrowers by degree type for 2011–12 cohort

Medical students have the highest average loan balance at $163k. The employability for medical doctors is really high — the average salary of a doctor in the U.S. is $200,000. As we will see later, this is one of the primary reasons that default rates of graduate students is low compared to other degree types.
Law schools are also expensive but they also make an average of $140,000 in income after graduation.
There is a correlation in income and cost of education. The degrees that are expensive to obtain also pay well in the professional world.

Default Rates:

The average loan amount per borrower has been increasing and incomes have been stagnant. As a result, default rates have slowly increased over the recent years. The increase was significant for borrowers entering repayments right after the 2008–2009 crash.

Five-year default rate by school type

Five-Year Federal Student Loan Default Rates by Institution Type

There is a clear trend that borrowers who only borrow for graduate school had defaults at a lot lower rate than others.
Even the most selective four-year schools had a default rate twice of graduate school borrowers.
For-profit institutions perform the worst. 47% of borrowers were in default within 5 years of starting repayments. This is mind-numbingly high.

Two-year default rate by school type

The relative ranking of defaults by school type is the same as that of 5 years. So, if borrowers default in the earlier years, they are less likely to get back on track and make payments.
Private non-profit universities perform the best in 2-year default performance.
Public two-year program default rates are the highest and have increased significantly in recent years — students are finding it difficult to find jobs with non-professional qualifications.
Students going to school for an associates degree or vocational qualifications at public universities may don’t complete schools and drop out at higher rates than students going to undergraduate/graduate schools.
Major reasons for not attending or finishing school includes increasing cost, family responsibilities, losing the drive to study and wanting to work.

A natural inclination is to determine why these students default. Can parents, government, or companies do something that can reduce the default rate?

The first thing to look at is whether borrowers are graduating from school.

Two-Year Federal Student Loan Default Rate by Completion Status

There is a significant difference in default rates by just degree completion status. Borrowers who graduate default at a much lower rate than students who don’t.

If we can get more students to graduation, we could significantly increase their chances of employment and a higher salary. Just these two factors can cut the default rate by 2/3rd in most cases.
We also see that borrowers who didn’t graduate during the recession defaulted at 3x than borrowers who did.
Due to limited jobs and cutbacks on blue collar jobs, most student borrowers who didn’t finish school couldn’t find work to pay for loans.

The chart below looks at default rate only for borrowers who took a loan in 2010–11. Let’s try to understand the possible reasons for a high variation in default rates.

Next, we look at default rates of borrowers by type of university/education sector based on degree completion status. This is a more detailed look at the borrowers in the previous chart (for loans in the year 2011–12).

Two-year default rate by sector for 2011–12 cohort

We see that public 2-year and for-profit sector have a 3x default rate compared to public and private non-profit four-year degree programs.
Getting an undergrad both from non-profit private and public schools have similar default rates for both student populations — students who completed degree vs who didn’t.
If we want to lower default rates, we need to get students to complete their degrees so they can be competitive in the job market. Students who don’t finish school significantly reduce their chance of getting a job.

Now, we look at default rates by the loan balance and the results are counterintuitive.

Three-year default rate for 2010–11 by Loan Balance

It is interesting to note that borrowers with less than $5,000 in loans have the highest default rate.
This is because students were enrolled in vocational or associates programs rather than bachelors or a masters degree. These courses tend to be cheaper.
These borrowers may have left the school midway and were not able to find jobs.
Borrowers with a high loan balance were perhaps enrolled in undergrad or grad programs that have a higher employment rate. These degrees tend to expensive. The engineers, lawyers, and doctors tend to have high debt, but also high income after graduation.

The high-level analysis of student debt underscore a few key points:

Students who go to school for higher education and complete their degrees have a significantly lower rate of defaults than students who go to school for professional/associate degrees.
There is a high correlation between the cost of school and average income after graduation.
If we want to lower default rates across the board — government, companies, and parents need to implement programs to ensure that students finish school irrespective of the level of education.
We need more students going to higher education non-profit schools as the quality of education is better and students learn skills that make them employable.

Data Sources: