QTRS March 6, 2025

Graphs and interesting content for the curious

As I see it…

The Trump executive order to reduce the percentage of indirect costs for NIH grants has received plenty of attention in higher ed circles. Most people aren’t aware that when, for example, a faculty member gets an NIH grant for, say, $1 million, 60% or more, $600,000+, goes to the university to cover indirect costs. The indirect percentage is set by the university and not by NIH. This is supposed to cover, for example, administrative overhead and maintenance. To my knowledge, the university does not list or is audited on how indirect money is spent. The other 40% or less are costs listed in the grant to pay researchers, for example. The indirect costs aren’t listed as line items in a budget. Basically, the indirect costs, which also exist for NSF grants, are a way of subsidizing higher ed or really mostly R1 institutions. The Trump proposal is to reduce the indirect percentage to 15%. Now, certainly there are indirect costs to the host university, but 60% or more seems really high, while 15% might be too little.

As I see it, higher ed has already lost here. Even if there is no change in indirect percentages, they have lost the PR battle as society in general is not going to view the 60% overhead as a good investment of their tax dollars. Worse, here is a graph from the AEI article How Many Administrators Do Colleges Have? (5/3/2025). The purple bars are the number of instructional staff. The other bar is a breakdown of non-instructional staff. The graph is grouped by the highest 20% of non-instructional staff to student ratio and on down by 20% groupings. It is hard not to look at this graph and think that there is some real administrative bloat at a lot of our colleges.

It turns out that the places with the biggest gap between instructional staff and faculty are more likely the R1 schools. From the article:

If administrative bloat is so concentrated at a few schools, couldn’t students just avoid them? Unfortunately, the top quintile of colleges by administration includes quite a few selective, in-demand institutions. The group includes many flagship public schools like the University of Texas at Austin (which has 281 FTE noninstructional staff per 1,000 FTE students), the University of Wisconsin-Madison (323 per 1,000) and Ohio State University (482 per 1,000).

The top quintile also includes most elite private universities. Washington University in St. Louis reports 17,019 FTE students and 17,128 FTE noninstructional staff—meaning there are more nonteaching staff than students!

This is probably no accident. Students want to attend elite universities for their prestige, and are willing to pay high tuition for that privilege. These schools’ clout also leads to lavish funding from state governments and private donors. Flush with revenue, top universities are free to hire legions of nonteaching staff. This is consistent with economist Howard Bowen’s “revenue theory of costs,” which holds that the revenue schools can raise largely determines what their costs are, as there is little incentive for cost minimization.

While it's true that research universities require additional administrative staff to oversee all grants, it's difficult to ignore the possibility that they would be just fine with a decrease in indirect costs.

Take our local university, Cornell, as an example. This is from our local online paper (2/27/2025):

According to the lawsuit, Cornell spent $452 million on research for over 1,600 NIH awards last year, receiving $137 million in reimbursements for indirect costs. The lawsuit states that Cornell has 1,207 NIH awards for the rest of 2025, and estimates that the school would lose over $42 million the rest of the year if the cuts are enacted.

The $452 million seems like a lot, but presumably much or most of that is from grants. Now the $137 million is a big number, but not relative to their overall budget of $5.8 billion. It is $137,000,000 / $5,800,000,000 = 0.024. In other words, NIH indirect money accounts for 2.4% of their budget, and they are using the possible loss of this money as an excuse to freeze hiring, while they have an endowment (2022) of over $9 billion or almost $400,000 per student. Maybe they could lose a few administrators first, and maybe society doesn’t think they should spend 60% on indirect costs given the money they have. It is almost as if universities forget that this money comes from taxpayers. Meanwhile, it isn’t unreasonable to question if some of that $137 million in indirect money could be spent on more researchers doing actual research. Anyway, The Free Press has a good debate (3/5/2025) on this issue if you want to learn more.

Right now higher ed is its own worst enemy. They have gone through a time of growth with more than enough students and funding, and that has made the industry complacent and bloated. They would be wise to get their houses in order themselves as a way to rebuild public trust. They are supposedly smart; this shouldn’t be so hard, right?

Stress, income & demographics

This is a really interesting graph from the paper Higher income is associated with greater life satisfaction, and more stress (2/19/2205), most of which makes sense, but there are two surprising results to me.

First, the lines show reported stress levels by income levels. The dots are turning points where stress levels go back up. Too little income, and your stress levels are higher, but at some point your stress goes back up with income.

For panel (a), females have generally higher stress levels than males, and the turning points are close. Interestingly, female stress doesn’t go up after the turning point.

For panel (c), there is little difference by political affiliation beyond Republicans being a little less stressed, and the gap widens as income goes up.

But what’s up with panel (b)? First, the narrative is that if you are a minority in the U.S., you are being bombarded with microaggressions and racism. Yet, Hispanic and Black people have lower stress levels than White people at all income levels. If you asked me to sketch this graph ahead of time, I would have the Hispanic and Black curves slightly above White. Not only are they below White people, but it’s a decent gap, bigger than between females and males. Keep in mind my data rule: Doubt something everyone knows is true.

The second surprising result is that the turning point for Hispanics is at an income of about $40,000 less ($30,000 compared to $70,000) than for both Black and White people and has a sharper increase after the turning point. I don’t even have a good guess hypothesis as to why.

Female-male ratio by county

This is one of a number of interactive graphs from the AIBM article Patterns and Trends in County-Level Sex Ratios (2/12/2025). Overall, the country is 98 men to 100 women, largely due to women living longer as there are more men than women in the 0-39 age group. Still, there are places that are highly skewed one way or another. Some of this is explained by the location of colleges, predominantly female, and urban areas, which are also often skewed female. The article notes that “more research is needed to fully understand these implications,” but they do make this one interesting point.

If there are many more men than women in a particular community, it is likely that those women have more bargaining power in the relationship market and can be more selective when picking a partner.

A country divided

The increasingly large viewpoint or value divide by political affiliation is one thing that concerns me. It seems to me that when the divide becomes big enough, violence against the other side becomes more easily justified. Anyway, here is one example from Gallup (2/24/2025). The article has a list of views of other counties, but the Israel divide is the largest, with Mexico (36-point gap) and Ukraine (30-point gap) coming in second and third.

Data Centers, climate change & smartphones

From DCD (3/3/2025)

US utility First Energy has reported that its pipeline of potential data centers has surged to 2.6GW by 2029, according to its Q4 results.

This is almost double its reported pipeline in November of last year, with First Energy utilities set to have potentially 5.5GW of data center load by the end of the decade, including existing and contracted facilities.

A regular reader will not be surprised by this. It is the next paragraph that got my attention:

To meet the new demand, First Energy is considering delaying the retirement of its coal-fired generation or replacing it with gas-fired generation. Currently, the utility has 3GW of coal-fired generation in West Virginia set for retirement between 2029 and 2035.

Solar and wind can’t supply this demand. We’ll keep burning coal and gas, and I don’t hear anyone suggesting we cut back on our data storage. Here is one example from 18+ Mobile Photography Statistics for 2025 (2/14/2025)

Smartphones capture 92.5% of all pictures, leaving just 7.5% to conventional cameras.

The typical smartphone user stores ~2,795 photos in their camera roll.

About 92M selfies are taken daily in the world.

An average American takes out their smartphone to take a photo six times a day.

Most, if not all, of this is backed up on a data center (the cloud) somewhere. Climate change or fewer selfies, which is more important?

The spinning CD

I missed this one from the Struts 4 months ago. Fun, catchy song.

Please share and like

Sharing and liking posts attracts new readers and boosts algorithm performance. I appreciate everything you do to support Briefed by Data.

Comments

Please let me know if you believe I expressed something incorrectly or misinterpreted the data. I'd rather know the truth and understand the world than be correct. I welcome comments and disagreement. We should all be forced to express our opinions and change our minds, but we should also know how to respectfully disagree and move on. Send me article ideas, feedback, or other thoughts at briefedbydata@substack.com.

Bio

I am a tenured mathematics professor at Ithaca College (PhD Math: Stochastic Processes, MS Applied Statistics, MS Math, BS Math, BS Exercise Science), and I consider myself an accidental academic (opinions are my own). I'm a gardener, drummer, rower, runner, inline skater, 46er, and R user. I’ve written the textbooks “R for College Mathematics and Statistics” and “Applied Calculus with R.” I welcome any collaborations. I welcome any collaborations.