Three Months on the Way to FAIR

By Kelvin Droegemeier, Barbara Snyder, Toby Smith, Nancy Andrews, Willie May, Kurt Marek, Farin Kamangar

Members of Congress have long questioned the government practice of covering the facilities and administrative (F&A) costs of federally funded research. These so-called indirect costs include the expenses of facilities operation and maintenance, library services, building depreciation, and research administration. Rates of reimbursement for indirect costs vary by institution; they are determined by negotiations between individual research organizations and federal research agencies, and governed by a set of byzantine regulations and accounting practices. Questions about indirect costs arose as early as 1946, only a year after the concept was implemented. Since 1980, the Government Accountability Office alone has released more than 30 reports on the subject of indirect costs, and more than 100 people have testified before Congress about them.

In 2017, before he became presidential science advisor in the first Trump administration, Kelvin Droegemeier testified before Congress at a hearing on a proposed cap on reimbursements that was included in the president’s FY2018 budget request. Droegemeier’s testimony explained the history of indirect costs within the social compact between universities and the government that began during World War Two. Congress eventually blocked the implementation of the cap in the FY2018 spending legislation, and policy around indirect costs remained unchanged.

In early April the Joint Associations Group (JAG) was formed, and in mid-July it announced an alternative model for accounting and funding indirect costs called the Financial Accountability in Research (FAIR) model.

Even before President Trump took office for a second time, there were signs that the administration intended to try again to reduce indirect cost payments to research institutions. In February 2025, Droegemeier, now professor of atmospheric science and special advisor to the chancellor for science and policy at the University of Illinois Urbana-Champaign, began convening an informal group of individuals from research organizations to think about an alternative to a cap. In early April the Joint Associations Group (JAG) was formed, and in mid-July it announced an alternative model for accounting and funding indirect costs called the Financial Accountability in Research (FAIR) model. The FAIR model presents a more understandable, transparent structure that’s designed to be usable by all kinds of research institutions, applicable to all research proposals and awards from all federal funding agencies.

Issues asked Droegemeier and others involved in the effort to explain how they managed to work toward agreement on a notoriously difficult subject in such a short time.

February 4, 2025: Déjà vu all over again

Kelvin Droegemeier: There were a lot of rumors that there was going to be action around indirect costs, particularly because of the Project 2025 report from the Heritage Foundation, which proposed capping reimbursements. But then on February 4, we heard that the National Institutes of Health would set a cap of 15%. That was enough to lead me to call the chairman of the House Appropriations Committee, Representative Tom Cole (R-OK), to say, “Here we go again, what’s different this time?”

During the first Trump administration, before I became presidential science advisor and director of the Office of Science and Technology Policy, NIH proposed a 10% cap. At that time, Chairman Cole came out strongly against the cap, and pulled together a subcommittee hearing where I testified that some of what makes indirect costs controversial is that they are misunderstood—but they are still a vital part of the nation’s research enterprise.

But here we were eight years after that testimony, and I realized we weren’t going to be able to just explain our way through, like we’ve done in the past. Change was on the horizon. It’s better to be a part of that change than just wait until something drops. Because once that happens, it’s very hard to make a shift: The train’s already running down the track.

So I started calling around to see what groups might be interested in strategizing.

February 13: The Thursday Group

Toby Smith, Association of American Universities (AAU): Kelvin convened some of us on the phone. He let us know that he’d spoken to some key Republican members of Congress and that it was not going to work for the community to fight to maintain the existing system for reimbursing indirect costs—that just wasn’t going to work this time. The community was going to have to come up
with a new model. That was the genesis for starting the bigger group. And then we needed to get the associations’ principals involved, and the only time those people could all meet was six o’clock on Thursday nights. So that became the Thursday Group.

Barbara Snyder, AAU (JAG principal): I helped Kelvin reach out to some other higher education association leaders that he did not know yet. We really wanted everybody to be a part of this—public, private, large, small, all kinds of institutions and not just universities. We knew they all had a stake in the outcome of this, and we needed to bring everybody together.

The only time those people could all meet was six o’clock on Thursday nights. So that became the Thursday Group.

The group quickly grew to include representatives from AAU, the Association of Public and Land-grant Universities, Association of American Medical Colleges, American Council on Education, Association of Independent Research Institutes, Council on Governmental Relations, National Association of Independent Colleges and Universities, American Association of State Colleges and Universities, Science Philanthropy Alliance, and National Association of College and University Business Officers.

Droegemeier: It became clear that we were maybe in a position to do something here. But then it became a question of how do we organize this? How do we think about developing a new model? We started thinking about how to organize formally but not as a legal entity, by bringing the groups together and then working with subject matter experts on models. For a long time, we didn’t even have a name—we just called it the Thursday Group, which included government relations experts from the JAG associations and some senior research officers from a few universities.

February 24: Meeting with Chairman Cole

Droegemeier: I visited Chairman Cole’s office to get a better handle on what we were trying to accomplish. I was joined by Leslie Belcher, Chairman Cole’s first chief of staff and now a lobbyist with Steptoe LLP. I had worked with Leslie and Chairman Cole for many years when I was at the University of Oklahoma because the university is in Cole’s district. The purpose of our meeting was to understand what concerns were really vexing Congress. Is it transparency, accountability? Is it cost? We knew from the last go-round that one of the big issues was this idea of reimbursement, of how the reimbursed funds are used.

We shared with him how we thought we should go about developing a new model. You know, he didn’t want to get into the weeds, and appropriately so, in terms of how this would be done, or what the model would look like, or anything like that. It was more him saying, “This is the world you live in, now you know what the challenges are—come up with something. If you can do it, we will help you every possible way we can.”

As time went on and we started talking to members of Congress and others, it became clear that what we’re actually trying to achieve is transparency, accountability, and explainability.

April 8–9: Introducing JAG and gathering the subject matter experts

On April 8, the newly formed JAG announced that 10 national organizations representing America’s academic, medical, and independent research institutions were joining forces to create a more transparent model for funding indirect costs of research in federally funded grants. The next week, Droegemeier assembled a group of subject matter experts with the intention of drawing together representatives of many political constituencies from a variety of institutions across the country. Every one of the 26 people he invited to serve as subject matter experts agreed to join the project.

Snyder: Our hope was that if we put together a group of subject matter experts from across this wide range of institutions, we would ultimately be able to agree to something that—while everybody probably doesn’t love it equally—they could still get behind. We needed to come together around a model that would work for all kinds of institutions.

Nancy Andrews, Boston Children’s Hospital (subject matter expert): Like all of us, I guess, I’ve been involved in other efforts over the years, where people set out with good intentions but don’t accomplish much. But it was clear that Kelvin had a plan for taking this across the finish line, and that he had the connections and the political know-how. I joined, feeling like I was representing biomedical research and American research broadly—not any particular organization and not my own interests. I felt a profound responsibility to do what I could to keep American science strong, and particularly to make sure that we continue to nurture and develop the scientists of the future.

I’ve been involved in other efforts over the years, where people set out with good intentions but don’t accomplish much. But it was clear that Kelvin had a plan for taking this across the finish line, and that he had the connections and the political know-how.

Willie May, Morgan State University: I got an email from Professor Droegemeier; the fact that he asked us was an honor. I accepted on behalf of Morgan State University, not only to represent the university, but also to represent organizations like Morgan State—the R2s and historically Black colleges and universities. And then I volunteered our associate vice president for research, Farin Kamangar, as a subject matter expert. You know, if you’re not at the table, you might be on the menu.

April and May: Wrestling with the models

The subject matter experts group started with 26 members. They were tasked with coming up with a workable model. They began by dividing into two groups to work independently on different approaches. One group took on reenvisioning the current F&A model. The other team set out with a “blank sheet of paper.”

One of the challenges the groups faced was considering how different institutions relied on the current indirect cost structure, which reflected differences in how institutions manage their F&A services, the types of research they perform, and the ways geography affects costs, including for utilities.

Kurt Marek, Sanford Burnham Prebys (subject matter expert): One of Kelvin’s driving efforts was to ensure that all different types of institutions were represented. I was there as a member of the Association of Independent Research Institutes (AIRI), representing the interests of the independent institutions, of which there are about 82. Unlike universities, which have other sources of income like tuition and support from the states to help subsidize federal research dollars, AIRI members are independent nonprofits without other revenue streams to make up for a federal shortfall. About 95% of our revenue is from federal grants. With a 15% indirect cost rate, we would have to shut down. This is an existential threat to our ability to do science.

Andrews: Not all American hospitals do research, but for the hospitals that do, it would be impossible without government indirect cost reimbursement. For us, the rate is currently about 78%. This doesn’t fully cover the hospital’s costs associated with federally sponsored research, but if the indirect rate were cut to 15%, we would pretty quickly have to stop doing research—at least the kind we mostly do. It would be devastating for all our institutions.

You know, if you’re not at the table, you might be on the menu.

Farin Kamangar, Morgan State University (subject matter expert): There is no model that can be very simple and still consider all the intricacies of real-world facilities and administrative costs, right? Many models were suggested, and the question was, which one do you pick? What level of simplicity or complexity do you use so that it optimizes real-world outcomes, versus something that is merely feasible? Then, when you change the model, a lot of other stuff needs to change. Universities have software systems that have been working for decades. Can you retrain the principal investigators? Can you retrain their staff?

You also have to make changes to statute, to the Code of Federal Regulations. You need to make changes to program rules, and sometimes even cost accounting standards, to some extent. So, negotiations were enormously complex and many models were presented.

Marek: From the beginning, everybody recognized this was a national emergency, essentially—that implementing indirect cost caps across agencies would affect national security and the economic leadership of the United States in innovation and technology. And so people really came together with open minds and a willingness to listen. Because of that, I think we were able to make impressive progress in a very short amount of time.

Droegemeier: The whole timeline was determined by Congress. We wouldn’t have chosen to do it this fast. But it turns out that sometimes that’s the best way. You just really dig into it, and you get it done.

May 17 and 18: Fly-in retreat at Chicago O’Hare

After holding a virtual retreat in early May, about 20 subject matter experts met at the Hilton Hotel at O’Hare Airport in Chicago for an in-person retreat. There, the two groups presented four models for consideration.

Droegemeier: The takeaway from my meetings with the Office of Management and Budget was that we were not oceans apart. That the model we were building would make further conversations possible, would keep us at the table for future conversations. And members of Congress made very clear that if the community comes across as being divided, they would not support us.

Kamangar: One inflection point came at the first meeting in Chicago, when everybody agreed to bring down the number of all proposed models to two. One model that was pretty simple, and one that was somewhat more complex, but more comprehensive. And I think everybody understood that if they didn’t come to some reasonable agreement, the consequences could
be dire.

Andrews: Kelvin is a great coach and cheerleader. Every time my motivation started to flag, he could sense that we were getting tired and somehow magically got us motivated again.

May: It is so much easier to come to consensus and work through very difficult and maybe intractable issues if you meet face-to-face. That’s very difficult to do via Zoom or electronic media. But being in the same room, being able to read the body language, have side meetings—all those things add to the ability to come to consensus on very, very difficult issues.

Is this $26 carrot cake a metaphor or what?

By the end of the fly-in retreat, we had two models. We knew at that time that neither one of those was the solution, but they indicated the extremes. We called them the bookend models, and we figured out that probably we’d end up somewhere in the middle. We wanted the community to look at these and give us feedback, which they did.

How in the world can carrot cake cost $26? Could this really be worth the price? So we ordered it.

Droegemeier: We all met for dinner at a restaurant the first evening of the retreat, and on the menu was a slice of carrot cake for $26. We said, How in the world can carrot cake cost $26? Could this really be worth the price? So we ordered it. We had 8 or 10 people at the table. This thing was, like, a foot tall, and it came with a gallon of ice cream in this big blob that had chocolate sauce and fudge all over it. I was like, okay: Question asked and answered. That was a moment I’ll never forget.

June 12–17: Presenting to the broader research community

The subject matter expert team presented the bookend models in concept to the JAG association principals, who approved them for rolling out to the broader community. Then the team presented the same models to the national research community in a webinar, which attracted over 2,000 attendees who submitted upward of 370 questions and comments.

A few days later, the group hosted an open, virtual town hall to entertain community questions and to present a guide for institutions to test the model. The event attracted a similar crowd of over 2,000 attendees, but this time the panel spent more than an hour entertaining questions and comments, which reached 465.

Andrews: A huge amount of time went into what I think of as the politicking—preparing and cultivating our relationships with people in government, the public, and our scientific community. Kelvin did almost all the cultivating with people in government, and then we, in various combinations, did it with the larger scientific community. I think overall that’s probably taken a lot more time than coming up with the model.

Droegemeier: We received a lot of formal feedback, and the Texas A&M University System was fabulous in attending these webinars, organizing the Q&A, and then using AI to synthesize it. People say, “Well, I’m one of, you know, 400 commenters. Are you really listening to what I have to say?” And the answer is yes, we are. We’re listening to it, and we’re using your feedback to improve the models.

People say, “Well, I’m one of, you know, 400 commenters. Are you really listening to what I have to say?” And the answer is yes, we are. We’re listening to it, and we’re using your feedback to improve the models.

A lot of questions at the town hall asked, “Why?”

Marek: People in the community had a hard time understanding that simply defending the status quo—as we’ve done for the last several decades—wasn’t an option anymore, that the current indirect model was going to change. I think a lot of the government relations folks understood it, a lot of the more senior executives at institutions understood it, but a lot of the rank-and-file people, a lot of the faculty, they just didn’t … see it. They’re far enough removed from Washington that they didn’t really understand that we will not be able to keep the current system as it is.

The town hall concluded with a call for institutions to test the models with their own data, and to report their results back to the JAG through an online form over the following two weeks. One hundred seventy-six institutions completed the testing.

June 29–30: Second fly-in retreat at O’Hare

The subject matter experts met again in person to review the community input and continue testing the models.

Marek: During our second retreat in Chicago, we reached an impasse over how we were going to address the issue of facilities costs. In the first iterations of our models we had tried to present one that was incredibly accurate and very granular. That model would have required an onerous amount of work on the part of institutions to be able to deal with it—and we got that feedback. But as we met in Chicago after getting that feedback, we had a hard time trying to work through it, and there were some tense moments—people with different ideas about how to do it.

One of the biggest surprises was the incredible benefit that the group saw from the presence of the industry and foundation representatives. As we worked toward the final model, the industry folks were really well informed about how the government expects the costing to be done. They pointed out flaws and gaps in our thinking by recognizing the similarities between what we came up with and the ways foundations fund research.

In the end, we came up with a useful compromise that both met the objectives of what we knew Congress wanted—more granularity on costs—and kept it simple enough for institutions to implement without bogging them down in details. That was the moment I understood that everybody was willing to work together to come up with an approach that benefited the country.

July 15: Announcing the FAIR Model

Droegemeier: When the model came out, that’s when we put the accelerator to the floor and said, okay, now we’ve got something to talk about. Here’s the model, here’s why we think this is the right approach.

In the end, we came up with a useful compromise that both met the objectives of what we knew Congress wanted—more granularity on costs—and kept it simple enough for institutions to implement without bogging them down in details.

Our model makes it clear that you have the research performance costs (formerly referred to as direct costs), and then you have the essential research performance support costs (restructured from some of the components of former indirect cost items). This reframes both as being essential. You can’t do one without the other. You can’t go to a restaurant and get a steak without paying for electricity. Steaks are a direct cost, the electricity and other things needed to provide them are support costs. When I mention that kind of thing to members of Congress, they get it.

Going forward

Smith: We have language in four of the House appropriations bills and three of the Senate bills. We also have language in the defense authorization bill in the Senate. And the Senate’s Departments of Labor, Health and Human Services, and Education, and Related Agencies Appropriations Bill mentions by name the efforts of the JAG, which I didn’t expect they would ever do. That shows that it’s getting traction. The fact that this effort spanned working with the research community and Congress—that’s a big accomplishment.

May: I think it’s very important for our community to speak with one voice. Sometimes even the different groups that represent the universities are not on the same page. As the administration unifies its position on higher education and the academic community, this is an issue that is forcing us to begin to speak with each other and speak with the same voice.

Snyder: I think we’ve shown that we can pull together as a research community and act effectively and quickly when we need to. We’re proving it to ourselves, really. What’s important is that that ability is powerful for the research community—maybe as much as anything else that’s come out of this.

Andrews: This work drove home how important diverse perspectives are for solving tough problems. There’s literature on that, of course, but this has been a compelling, real-life example. Our group is exceptionally diverse in where we live and work, in the types of institutions that we represent, in the roles we have at those institutions. And almost certainly in our political leanings as well, although we never went there.

Droegemeier: I had an epiphany at one point, where I just realized how blessed we are to have such a group here. When you’re working on it, your head’s down—and then you step back and show people what you’ve done: put the model out. And then you look at those people involved and say, we did this as a family. It was just unbelievable. Still to this day, I’ve never had an experience like this in my whole career.