Admittedly, dear readers, I had a lot to say about one of this week’s articles, but when headlines tell an incomplete story that could impact enrollment, I think it’s important to understand the details. In this week’s issue, I look at the results of a recent study about the economic returns of graduate education before moving on to an article that questions whether the value proposition presented by most postsecondary institutions is still valid. I close with our new feature, Sparks, a series of short articles designed to spark additional conversations on your campus. 

After reading today’s issue, share your thoughts about graduate degrees and value in the comments! 

Graduate School Investments 

From Graduate School Pays Off for Pharmacists, but Not Psychologists | Inside Higher Ed 

New research from the Postsecondary Education & Economics Research Center at American University finds that not all graduate degrees provide a positive economic return.  

Our Thoughts  

This is an interesting piece worth spending some time with, if only so you better understand the conclusions and how the researchers got there. The underlying research is a NBER working paper by Joseph Altonji at Yale and Zhengren Zhu at Vassar, and it is already drawing attention beyond higher education news sites. The headline numbers are the kind that travel fast: some graduate degrees yield a negative return on investment, psychology is the worst at negative 8 percent, and even popular fields like curriculum and instruction land in negative territory. If you work at an institution that relies on graduate enrollment to balance budgets, you should be thinking about how this narrative could reach your prospective students and the policymakers who fund your programs. It is the kind of study that tends to get simplified in the retelling, and the simplified version is not going to be kind to some graduate programs.  

That said, the study, while methodologically rigorous, applies a broad economic framework that does not always reflect how graduate education actually works in practice. While I’m not an economist, I spent some time with the full 73-page paper to try to better understand the methodology, and I think there are a few things worth exploring before accepting the headline numbers at face value. 

First, the study treats tuition uniformly as a cost borne by the student, using average public institution tuition rates from NCES data. It does not adjust for employer-paid tuition. This matters a great deal for fields like curriculum and instruction, where the decision to pursue a master's degree is frequently subsidized or fully covered by a school district's tuition reimbursement program. If you zero out the tuition cost because someone other than the student is paying, the math changes significantly. Even a modest earnings effect starts generating a positive return when the investment denominator shrinks to near zero. The paper acknowledges employer-paid tuition for MBA programs in a footnote but does not extend that consideration to education fields or social work, where tuition reimbursement from school districts, state agencies, and nonprofit employers is common practice. Public service loan forgiveness programs further complicate the picture, particularly for social work graduates who may never bear the full cost of their degree. 

Second, the headline comparisons across degree types are skewed by something the paper reports but does not emphasize. When calculating the return for each field, the researchers combine part-time and full-time students into a single number. That sounds reasonable until you look at how different the enrollment mix is across fields. JD programs are 98 percent full-time. MD programs are 98 percent full-time. Meanwhile, education administration is 93 percent part-time and nursing is 88 percent part-time. The paper consistently finds that full-time students see higher economic returns than part-time students, so when you combine everyone together, fields dominated by full-time students naturally look stronger than fields dominated by part-time students. The result is that when you line up 18 degrees and compare them, you are not just comparing the economic value of different fields. You are comparing degrees that are overwhelmingly pursued full-time by people who stop working against degrees that are overwhelmingly pursued part-time by people who keep working. Those are fundamentally different investment profiles, and the combined numbers hide that distinction.  

Third, and this is one that matters for anyone who has worked in K-12, social services, or clinical settings, the paper's methodology cannot distinguish between market-driven wage increases and administratively determined ones. Teacher pay is largely governed by district salary schedules where a master's degree triggers a defined pay bump. Licensed clinical social workers and clinical psychologists operate in a similar environment, where graduate credentials unlock licensure tiers that come with structured pay increases at agencies and healthcare systems. These are not market returns in the way an MBA's return is. They are contractual or credentialing-based entitlements. The paper treats all post-degree earnings increases the same way, but the mechanisms generating those increases are entirely different. For a curriculum and instruction or social work degree, the "return" is really a question of whether the guaranteed compensation bump exceeds the cost of the degree. If someone else paid the tuition or the loan is forgiven, the answer is almost always yes. 

There is a broader question the paper's framework cannot account for. The fields at the bottom of the return on investment rankings are not there because they are low-value work; they are there because we have collectively chosen to underpay the people who do them. Teachers, social workers, and clinical psychologists serve functions that no community can do without. No school district is looking at this study and concluding that it should stop hiring teachers with master's degrees. No state child welfare agency is deciding it no longer needs licensed clinical social workers. Society depends on these professions and then compensates them at levels that make the graduate credentials required to enter them look like a bad investment. The return is low because the public investment in these roles is low. That is a policy outcome, not a market signal. 

It is also difficult to ignore that the fields clustered at the bottom of the ROI rankings are disproportionately held by women. Education, social work, clinical psychology, and curriculum and instruction are all majority-female professions, while the fields at the top of the rankings, medicine, law, and engineering, have historically skewed male. The paper does find that women see higher returns to graduate education than men on average, which is a useful data point, but the structural reality underneath that finding is that the professions women enter in the greatest numbers are the ones society compensates the least. We then produce research that frames those professions as poor investments. The ROI lens is not wrong, but it is incomplete, and it risks reinforcing an undervaluation that is already baked into the system. 

None of this means the paper's findings are wrong. The methodology is sound within its framework, and the variation across fields is real. The finding that students from lower-paying undergraduate majors benefit more from an MBA or JD is genuinely useful for advisors and career development staff. The observation that program rankings correlate with returns for law and business but not for nursing or social work is interesting and worth further study. Zhu's closing comment in the article is also one of the most important sentences in the piece: Harvard Business School graduates would have had high earnings even if they had not gone to Harvard Business School. That kind of selection bias is exactly what makes these analyses so tricky, and to their credit, the researchers try hard to control for it. 

For higher education leaders, the takeaway is not "graduate school is a bad deal in some fields." It is that broad economic frameworks, applied uniformly across 121 degree types, will systematically undervalue programs designed around working professionals who enroll part-time, whose employers pay their tuition or whose loans are forgiven through public service, and whose pay increases are governed by salary schedules, licensure requirements, or agency pay bands rather than market forces. Those are not edge cases. Those describe the majority of students in education, social work, and clinical psychology graduate programs. The fields that land at the bottom of these rankings are not there because the work lacks value; they are there because the compensation has never reflected the value. If this study gets picked up more broadly (and it will), institutions would be wise to have a more nuanced version of the story ready to tell. 

Why Go to College 

From Colleges Need a New Sales Pitch | The Chronicle of Higher Education  

Karin Fischer looks at the current value proposition of colleges and universities and asks whether it works in today’s economy.  

Our Thoughts  

I will keep this one brief since my first article this week ran long, but this piece deserves attention because it sits right alongside the graduate degree ROI conversation above. Both articles are asking the same underlying question: How should we measure the value of higher education?  

The core argument here is that higher education's value proposition has been reduced to individual job outcomes, and that narrative is losing its footing. The article traces how that happened, and it is worth noting that colleges did not arrive here entirely by choice. As state funding declined and students absorbed a larger share of the cost, the relationship became transactional almost by necessity. When a student is paying the majority of the bill, asking "what will this get me?" is not unreasonable. Legislatures reinforced the dynamic with performance-based funding, earnings scorecards, and now proposals to eliminate majors that don't clear a salary threshold. Colleges responded to the incentives in front of them, and for decades, the ROI story held up. Now, as AI disrupts entry-level knowledge work, as the wage premium shrinks, and as graduates increasingly find themselves underemployed, the narrative is cracking. The problem is not that colleges told the wrong story. The problem is that everyone—institutions, legislators, and the public—converged on a single story, and there is no shared fallback for when that story stops working. 

What struck me most was Brian Rosenberg's observation that old-fashioned liberal arts might actually be what this moment requires. The ability to adapt, think critically, pick up new skills, and navigate uncertainty may matter more in an AI-disrupted economy than any specific technical training. That is not a new argument, but it lands differently now that the fields once considered golden tickets, like computer science and programming, are themselves feeling the pressure of automation. If institutions are going to rebuild a compelling case for the value of a degree, it probably starts with being more honest about what a degree actually does. It does not guarantee a job. It offers a foundation for a career that will change multiple times over a lifetime. That is a harder story to tell on a billboard, but it is a story that won’t break the next time the economy shifts. 

Sparks 

• Web accessibility: the 2026 deadline you can’t ignore (University Business) - Just a gentle reminder that if you are a public institution, you have until April 24, 2026 to ensure your online presence complies with Title II of the Americans with Disabilities Act. Not sure what this means or what compliance looks like? This article is a great, short introduction to get you started.
• Dropouts on the rise globally as university participation surges (Time Higher Education) - A new UNESCO report finds that while the enrollment rate in postsecondary education has more than doubled globally over the past 25 years, the graduation rate has not kept pace. This short read provides a good perspective that no one seems to have figured out how to improve retention rates at scale.
• Nearly Everyone’s Using AI at Cal State. And Nearly Everyone’s Worried About It. (The Chronicle of Higher Education) - A survey of 94,000 students, faculty, and staff at 22 California State University campuses finds AI deeply embedded in campus life, even as concerns about accuracy, ethics, and job security persist. Two-thirds say it improves their learning and two-thirds are skeptical about its role in education...hmmm