U.S. post-secondary education enrollment has been declining in recent years including, most recently, a 5% decrease in Fall 2024¹. This trend has opened the discussion of what affects enrollment for universities. Common variables considered in that conversation include job and unemployment statistics, tuition costs, and the wage premium for obtaining a degree. One question not commonly explored is how does college football affect enrollment?

Millions of people watch college football every year, for example, almost 12.5 million people watched Michigan State University (MSU) play Ohio State University in November 2024.² This begs the question, could the performance of MSU’s football team affect how many people want to study there? As another noteworthy example, Boise State University began as a small institution in the 1960s, with a modest football program in its early years. However, the program gained momentum in the early 2000s, achieving multiple successful seasons and winning conference championships in consecutive years. This success peaked in 2006, culminating in a memorable victory in the Fiesta Bowl.³ Today, Boise State is now the largest University in Idaho, illustrating how a university's enrollment numbers can often correlate with the success of its athletic programs.

Figure 1 shows a visual association between how many games MSU wins and enrollment at the university. On a strictly visual basis, when MSU wins more games, there is sometimes an increase in enrollment the following year. In 2010, MSU won 11 football games and 2011 enrollment increased by 2.2%. Likewise, sometimes when MSU loses more games during the season, enrollment decreases the following year. MSU only won 3 games in 2016 and 2017 enrollment decreased 0.6%.

Figure 1: Comparison between Fall Enrollment and Football Wins at Michigan State University 2004-2023

 Source: Points Consulting using MSU Spartan Athletics database; Michigan State Football Statistics & Year-By-Year Results, Michigan State University Athletics; MI Capital Region Data Hub

To investigate further how much football affects enrollment, we can use a statistical technique called Ordinary Least Squares (OLS) regression analysis to further explore the relationship between the two variables. An OLS Regression is a statistical method to find the relationship between one dependent variable and one or more independent variables. An independent variable explains, or influences, changes in another variable while a dependent variable is the one being affected, or explained, by the independent variable.

Table 1 shows how many individuals enrolled at MSU is the “dependent variable” and “Wins”, our independent variable, represents how many games the MSU football team won.

Table 1: Correlation between MSU Fall Enrollment, Games Won, and Other Variables

	OLS Regression Results
Variable	Coefficient	P -Value
Tuition Per Credit	31.14	0.00
Wins	-67.43	0.24
MGRP	< -0.01	0.01
Unemployment: State of Michigan	-0.01	< 0.01

Source: Points Consulting using MSU Spartan Athletics database; Michigan State Football Statistics & Year-By-Year Results, Michigan State University Athletics; Rate History, Michigan State University; MI Capital Region Data Hub

Table 1 displays results of an OLS Regression with more variables considered. Additional factors considered here are those that, one may theorize, would have some impact on enrollment for one reason or another.

• Tuition Per Credit: How much the university charges students per credit they take
• Wins: How many games MSU’s football team won
• MGRP: The State of Michigan Gross Regional Product (GRP)
• Unemployment: State of Michigan: How many people are unemployed in the State of Michigan

Though there are many numbers generated by such an analysis, the Table has been reduced to just the most important factors. Firstly, this model’s R2 (pronounced R-squared) has a value of 0.98. R2 indicates how much of the movement of enrollment is “explained” or accounted for by the movement of the variables listed. In this case, the 0.98 indicates that it’s a very strong model (in that almost all of the movement of the data is accounted for).

The P-Value indicates how likely it is that the given variable is correlated with enrollment. Counterintuitively, the lower the value the higher the likelihood. For example, “tuition per credit” is almost certainly related (P-Value of less than 0.01) and “wins” is less so (24% chance that it’s actually just randomness in the data). Lastly, the coefficient explains mathematically how the given variable has an effect on enrollment (which will be explained in the subsequent example).

By adding more variables to the OLS Regression, the impact has changed on how variable, “wins” affects enrollment. The association is not very strong when accounting for other possible factors. And, surprisingly, the direction of the effect has reversed compared to Table 1. Wins has a coefficient of -67.43, implying that for every additional game MSU wins the model predicts a decrease of approximately 67 students enrolled. This seems to indicate that football performance has less of an effect than it initially seemed while other factors have more power in explaining changes in enrollment.

The State of Michigan’s GRP has a negative effect on enrollment. With a coefficient less than 0, our model indicates that as GRP increases, the number of people enrolled at the MSU will decrease.  A higher GRP indicates a thriving local economy and high levels of employment. Hence, when individuals are employed and the economy is doing well, they are less likely to attend a university for higher education.

Additionally, unemployment in the State of Michigan has a negative effect on enrollment. Having a negative coefficient suggests that as unemployment in the state increases, the number of people enrolled decreases. It might be expected that as unemployment rises, more individuals would return to school to pursue further education to improve their job prospects. However, it is possible that higher unemployment often leads to financial instability for individuals and families, making it harder to afford tuition and other costs associated with higher education

According to the model, Tuition Per Credit increases enrollment. It has a coefficient of 31.14, implying that for every dollar tuition increases, the model predicts an increase of approximately 31 students. It would seem with the rising costs of tuition that it would be a decrease instead. However, it could be a reflection of the increased cost actually being due to increased quality of service and therefore people are willing to pay for it.  Figure 2 shows an association between Tuition Per Credit and how many individuals are enrolled at the university. In recent years, as tuition rates have increased, enrollment numbers have increased. Enrollment dropped two percent between 2019 and 2021. However, many universities experienced this due to the pandemic.

Figure 2: Comparison between Fall Enrollment and Tuition Per Credit at Michigan State University 2004-2023

 Source: Points Consulting using MSU Office of the Controller; Rate History, Michigan State University; MI Capital Region Data Hub

Despite what may have originally appeared in Figure 1, after accounting for more variables it appears that how many games the MSU football team wins do not have a statistically significant effect on tuition. The high P-Value indicates that football performance is not statistically significant in predicting enrollment changes. In contrast, Michigan’s GRP and how many people are unemployed in the state have a greater impact on enrollment. These findings suggest that broader economic conditions play a more substantial role in influencing enrollment at MSU than athletic success. Other external factors, such as tuition costs at nearby Universities or other economic variables could influence enrollment trends and could be explored in future research.

---

¹ Janae Bowens, "College Freshmen Enrollment Drops," The Baltimore Sun, December 14, 2024, https://www.baltimoresun.com/2024/12/14/college-freshmen-enrollment-drops/.

² "College Football TV Ratings," Sports Media Watch, accessed December 2024, https://www.sportsmediawatch.com/college-football-tv-ratings/.

³ Aspen Shumpert, “A look back at BSU football history ahead of Friday's big game,” KVTB 7, December 5, 2024, BSU Bronco football history ahead of Friday's big game | ktvb.com

---

Wage comparisons between public and private employers are always a topic of interest since both sectors seek to offer competitive packages to attract talented workers. Generally speaking, talent has come to the conclusion that private sector jobs pay better while public sector jobs offer more benefits. Interestingly, this pattern does not prove to be true in Michigan where public sector jobs often pay more than the private sector. In a surprising twist, Michigan’s public sector salaries outpace those in the private sector, further challenging the conventional wisdom on job compensation.

Figure 1 shows that average annual salary for both State and Federal government jobs are significantly higher than private sector jobs in Michigan. Over time, the average annual salary for private sector jobs in Michigan has increased but nowhere near as much as State and Federal jobs. Between 2021 and 2022 average annual salary for private sector jobs in grew 5.1%, the highest of the three categories, while Federal earnings increased only 1.9%. As of 2022, the average annual salary for State employees has almost reached the same amount as Federal jobs. Part of the reason for this trend is a decline in Federal pay as well.

In 2015 and 2016 there was a significant difference between average annual salary for State and Federal jobs in Michigan but over time the gap has narrowed. This could partially be due to some “outlier” State employees who have an incredibly high salaries, thereby pushing up the averages for all State employees. To cite one example, the head football coach of Michigan State University, Mel Tucker, has an annual salary of $5.56 million.

Figure 1: Average Earnings between Public and Private Sector 2015-2022

Source: Points Consulting using U.S. Bureau of Labor Statistics and Federal Reserve Bank of St. Louis, Average Hourly Earnings; FederalPay.org, OPM; OpenPayrolls.com, State Employee Salaries by Year

Figure 2: Annual Payroll Growth Rate between Tri-County Region and Other Sectors, 2016-2022

Source: Points Consulting using U.S. Bureau of Labor Statistics and Federal Reserve Bank of St. Louis, Average Hourly Earnings; FederalPay.org, OPM; OpenPayrolls.com, State Employee Salaries by Year, Tri-County Regional Planning Commission  

Payroll growth in the Tri-County Region has generally mirrored trends in Michigan's private and public sectors but contrasted with Federal government patterns. In 2020, Federal Government payrolls grew by 9.2%, while the Tri-County Region saw a 0.5% decline. By 2022, however, the Tri-County Region outpaced national growth, with payrolls increasing 9.8%, compared to a 1.9% rise in Federal Government jobs.

While salary across the three sectors varies, how many people are employed in each sector shows a completely different picture. Figure 3 shows that the private sector has significantly higher employment growth over time compared to the public sector in Michigan and the Federal Government.

Figure 3: Percent Change in Employment Growth between Sectors 2015-2022

 

Source: Points Consulting using U.S. Bureau of Labor Statistics and Federal Reserve Bank of St. Louis, Average Hourly Earnings; FederalPay.org, OPM; OpenPayrolls.com, State Employee Salaries by Year

Excluding 2018 and 2020, employment in Michigan’s private sector has consistently outpaced State and Federal jobs in annual growth. The private sector experienced the sharpest decline during the pandemic but has shown a stronger recovery since 2021 compared to Federal and State employment. While Michigan Government and Federal jobs often offer higher pay, Michigan’s private sector provides a significantly larger number of employment opportunities, making it a more accessible option for job seekers.

Figure 4: Percent Change in Employment between Different Sectors, 2020 to 2022

Source: Points Consulting using U.S. Bureau of Labor Statistics and Federal Reserve Bank of St. Louis, Average Hourly Earnings; FederalPay.org, OPM; OpenPayrolls.com, State Employee Salaries by Year

Between 2020 and 2022, employment in both the public and private sectors in the State of Michigan increased, while Federal employment declined. However, private sector jobs in Michigan experienced significantly higher growth, increasing 9.5%, compared to the more modest growth in public sector jobs. Although government jobs tend to offer higher pay, the data suggests that the private sector may provide greater opportunities for employment due to its faster growth rate and potentially larger job market.