ACT Scores Fell Last Year. Relax!

As a shareholder of the Green Bay Packers, I keep an eye on what Butte Community College’s most famous student-athlete has to say. Packers quarterback Aaron Rodgers famously told fans in “Packer-land” in 2014 to “R-E-L-A-X” after the team got off to an uncharacteristically slow 1-2 start. Fans relaxed after the team went 11-2 the rest of the way in the regular season as Rodgers played like his regular self.

In the education policy niche of the world, few things get people more upset than declining standardized test scores. Last year, I wrote about the fuss about SAT scores declining—and how at least part of that decline is due to more students taking the test instead of the American education system failing young adults. Now it’s ACT’s turn to release their newest scores—and my message again is R-E-L-A-X.

Between 2015 and 2016, average ACT scores declined from 21.0 to 20.8 nationwide, the lowest score in at least five years. But as the now-dominant test in the United States (much to the surprise of many folks who grew up on a coast where the SAT is still common), the percentage of students taking the ACT rose from 52% in 2012 to 59% in 2015 and 64% this year. This sharp increase in ACT takers is in large part due to more states requiring all students to take the ACT as a graduation requirement. In 2016, all graduating high school seniors took the ACT in 18 states, up from 13 states in 2015.

The five states that required all students to take the ACT for the first time in 2016 all saw large decreases in their average scores, as shown below. Wisconsin, Missouri, and Minnesota all had about 75% of their students taking the ACT in 2015 and had drops of about 1.5-1.7 points when all students took the test, with South Carolina having a drop of 1.9 points as the last 38% of students took the test. Nevada had a decline of 3.3 points in 2016, but the percentage of students taking the ACT more than doubled.

State Pct tested (2016) Avg score (2016) Pct tested (2015) Avg score (2015)
Nevada 100 17.7 40 21.0
South Carolina 100 18.5 62 20.4
Wisconsin 100 20.5 73 22.2
Missouri 100 20.2 77 21.7
Minnesota 100 21.1 78 22.7


Among the other 45 states that had very small changes in ACT participation rates, the average change in scores at the state level (not weighted for size) was effectively zero. So R-E-L-A-X about test score declines when they are due to more students taking the test (some of whom won’t be going to college, anyway) instead of collegegoing students suddenly performing worse.

Income-Driven Repayment Plans Continue to Grow

The traditional way to repay federal student loans was for students to pay back their loans over a ten-year period of time, generally by making the same payment each month. But as student loan debt has generally risen over time (although falling ever so slightly in the most recent quarter), paying off larger loans in a short period of time has become more difficult for many borrowers. This has made income-driven repayment plans, expanded during both the Bush and Obama Administrations, an appealing option for more students (although the future price tag of the programs is something to watch closely in the future).

The U.S. Department of Education recently released new data (updated every three months) on the federal student loan portfolio showing the growth in income-driven plans. The chart below shows the percentage of dollars in the Direct Loan program that are in one of four broad categories: 10-year payment plans not tied to income, longer payment plans not tied to income, income-driven plans, and miscellaneous plans that don’t fit well in any of the above three categories.1


Since 2013 (when repayment plan data first became available), the federal government’s holdings in the Direct Loan program have risen from $361 billion to $673 billion. The amount of loans in the standard ten-year repayment plan rose from $168 billion to $267 billion during this time, but the amount in income-driven plans rose from $72 billion to $269 billion in just three years. Income-driven plans now make up 40.0% of all Direct Loan dollars, while 39.7% of dollars are now in ten-year plans.

The Department of Education also released data for the first time on the number of students seeking employment certification in the Public Service Loan Forgiveness (PSLF) program, which will allow students working in approved fields to make ten years of payments instead of 20-25 years under other income-driven plans. While students aren’t officially in PSLF until they complete ten years of payments (the first students will do so in October 2017), this is an interesting measure of potential interest in PSLF. The below figure (from Federal Student Aid) shows the number of students who have submitted employment certification forms in possible preparation for receiving PSLF.


Notably, about one-third of all requests have been denied to this date, suggesting that quite a few students will get an unpleasant surprise when they go apply for PSLF in the next few years. But at least 430,000 students look to be on track for PSLF at this point—a number that is likely a significant understatement of the number of applications that the federal government will receive.


1 The Direct Loan program represents about 90% of all loans held by the federal government. The other 10% are in the older Federal Family Education Loan (FFEL) program, which has not disbursed new loans in years but has about one-third of its loan dollars in income-based plans. I excluded FFEL here because repayment plan data are only available for 2016.

On Student Loan Debt and Negative Wealth

A recent analysis by economists at the Federal Reserve Bank of New York looked at the approximately 14% of American households that had negative wealth in 2015 and pointed out student loan debt as a key driver of negative wealth. As a key figure from the report (which is reprinted below) shows, student loan debt is responsible for between 40% and 50% of total negative wealth among households with a net worth of below -$12,500.


This isn’t a tremendously surprising finding, although it’s always helpful to document something intuitive with actual data (although self-reported data always come with caveats). Student loans are one of the few types of debt aside from medical or legal bills that can be taken on in large amounts without having outstanding credit or collateral. Credit card debt is another way to take on debt, but most people with negative wealth won’t be able to access large lines of credit this way. It’s also possible to be underwater on a house (by owing more than its current value), but this affects a relatively small percentage of households.

The authors of the analysis then wrote the following about the implications of student loan debt:

“It is likely that the steady growth in student debt and borrowing, combined with the very slow rate of student loan repayment we have documented elsewhere, has materially contributed and will continue to contribute to negative household wealth and wealth inequality.”

As I told Inside Higher Ed in their summary of the analysis, I only partially agree with that assessment. The challenge with this analysis is that it combines students who completed and did not complete a degree (likely due to sample size issues, as the dataset includes questions about educational attainment). As the authors note, households with a bachelor’s degree or higher and negative net worth tend to have a young head of household. For example, my household is just now leaving negative net worth territory, five years after our head of household completed law school.1 Paying off student loan debt is difficult, but the rapid growth in takeup of income-driven repayment plans among high-debt individuals (as shown in this recent White House report and in the chart below) has the potential to reduce this burden.


I’m far more concerned about the implications for wealth inequality among students who did not complete a degree and are unaware of income-driven repayment options. Although there are positive economic returns on average for students who attend college but do not graduate, they are far smaller than students who finish. A better measure of wealth inequality would look at how wealth progresses over a ten-year window after a student leaves college. If he or she is able to repay loans and build assets, the picture is far less bleak than if a student still has negative wealth due to student loans and other types of debt.


1 In 2016, the term “head of household” is outdated. In households where both adults work, it’s far from clear who should be the head and who isn’t. It would be better to know the highest educational attainment of either of the adults.