Toggle Menu
October 11, 2022

How to Actually Reduce the Administrative Burden on Research

Tl;dr version: So far, it has been pretty much impossible to reduce the burden of bureaucracy on researchers via normal politics. We need a national commission — like the military base closure commission in the 1990s — that is empowered to make recommendations that go into effect automatically unless vetoed by Congress. That way we’ll actually make progress here.

* * *

In official surveys, researchers say that they spend well over 40% of their research time on complying with bureaucratic requirements, such as reports, budgets, and more.

This is unacceptable. We would never make a top athlete like Steph Curry or Lionel Messi spend 40% of his time filling out expense forms for his shoes and travel, writing up extensive reports on his games (as if the funding agency had no access to the games themselves), and the like.

As a top MIT scientist recently said to me, filling out NIH reports makes him spend time extensively rewriting his published articles so as to fit NIH’s format, even though they could just look at his articles for themselves. As he put it, progress reports are basically like this:

Q: “What did you accomplish with this grant?”

A: “I made the discoveries that are in my published articles, which I will now rewrite in your format.”

Q: “How did you disseminate those accomplishments?”

A: “Perhaps you could look at my published articles to see how they were disseminated.”

And so forth. Even though progress reports are mandated with the best of intentions (i.e., making sure that federal money is spent wisely), they can make some of our nation’s top scientists feel like this guy:

There have been many reports and efforts to fix this problem, but they inevitably bump up against the following dilemma:

1) Any bureaucratic requirement probably has a good motivation that came from some scandal or problem in prior years;

2) Any bureaucratic requirement probably has a minimal impact in and of itself;

3) Anyone who proposes to get rid of a particular bureaucratic requirement will be met by a small set of opponents who can point to the good reason for that requirement, but won’t have much if any public support for reform (because that one lonely requirement doesn’t impose enough of a burden for anyone to mobilize against it).

4) Thus, public choice theory predicts what will happen: the bureaucratic requirement will remain in place.

Here’s how that dynamic played out in just one of many examples.

The Stanford yacht scandal in the 1990s had broad repercussions for government funding of research. Throughout the 1980s, Stanford had allegedly been charging the federal government for indirect costs that weren’t really indirect costs of research, but instead included items such as “depreciation charged for a yacht that had been donated for the use of the Stanford student sailing program.” As of 1994, Stanford paid $1.2 million to the US government to settle the claims.

But this wasn’t just an accounting dispute between Stanford and the US government. Instead, in the words of some scholars, it “led to a vast increase in research administration rules and regulations, at least some of which seem (from convergent indicators) to be red tape.”

Which isn’t surprising at all! No one wants to be a national headline for allowing yacht expenses to be charged to federal research grants.

But the result was that many other universities had to spend time and resources implementing rules that were arguably onerous. Indeed, a full accounting of the situation would almost certainly show that it would have been better to let Stanford charge for the yacht than to make all other universities scramble to implement more rules and regulations over time.

Eliminating Bureaucracy is Like Closing Military Bases

We need to flip the usual public choice scenario on its head, just as we did with military base closures 30 years ago.

Back in the 1980s, there were around 890 military bases in the United States and various territories. Everyone knew that we didn’t need 890 military bases, and that many of them were duplicative, outdated, inefficient, etc. Yet up till 1988, “not one major base was closed.”

Loring Air Force Base, from

The reason why is fairly obvious: Any given military base was a large employer in some congressional district. It meant jobs and votes.

Thus, even though everyone wanted (in theory) to close unneeded bases, the public choice calculation for any given military base was this: At least one congressperson and possibly the state’s senators would vociferously object because they were worried about local votes. Meanwhile, there was no national coalition that would mobilize to close just that one base.

As the New York Times editorialized in 1988:

No waste in government is more obvious than keeping military bases open only to benefit a particular congressman’s or senator’s constituents. . . . Consider Loring Air Force Base in far northeastern Maine. Building it there, as close as possible to Moscow, made sense in 1946. Modern bombers and intercontinental missiles have long since wiped out that justification. . . .

Other egregious examples abound: Fort Douglas in Utah was built to protect a stagecoach route. Fort Sheridan in Illinois provides Army brass with a golf course and two beaches. Virginia’s Fort Monroe, moated against the British in 1812, now serves no essential military purpose.

Indeed, in 1976, President Ford had announced a plan to close or realign some 160 bases, and Congress was so upset that it not only blocked that plan, but passed a law that “required waiting periods for congressional consultation and for the preparation of environmental impact statements.”

[Side note: This isn’t the only occasion on which “environmental impact statements” are demanded for reasons having nothing to do with protecting the environment!]

In the Defense Authorization Amendments and Base Closure and Realignment Act of 1988 (codified in a note here), Congress came up with a genius way to move forward: It created the Base Realignment and Closure Commission.

This independent commission was required to recommend base closures by early 1989, and those recommendations would automatically go into effect unless Congress passed a joint resolution within 45 days disapproving of the recommendations.

That is, instead of saying, “This inefficient military base will stay open forever unless a majority coalition emerges to close that specific base,” we said, “Here’s a commission that can objectively decide which bases need to be closed everywhere, and that will happen unless a majority coalition wants to keep bases open.”

That flipped the public choice calculation on its head. We created a way for all congresspeople to vote for a more effective military—while giving them plausible deniability if it happened to close a base in their state.

As Jerry Brito has written, this was the most important feature of the legislation: it gave political cover for Congress to “vote for the popular budget-cutting measure, and then deflect blame to the Commission if a base in their district is ultimately selected for closure.”

Congress was well aware of this political dynamic, as Phil Gramm hilariously pointed out in a committee hearing (quoted by Brito):

The beauty of this proposal is that, if you have a military base in your district . . . under this proposal, I have 60 days. So, I come up here and say, “God have mercy. Don’t close this base in Texas. We can get attacked from the South. The Russians are going to go after our leadership and you know they are going to attack Texas. We need this base.” Then I can go out and lie down in the street and the bulldozers are coming and I have a trusty aid there just as it gets there to drag me out of the way. All the people . . . will say “You know, Phil Gramm got whipped, but it was like the Alamo. He was with us until the last second.’”

[Another side note: in today’s environment of social media and video, Gramm probably wouldn’t have been nearly as candid about his political calculations in trying to do the right thing for the US . . . .]

The results? The closure commission almost immediately issued a report recommending that 11 major bases and 80 smaller installations be closed or realigned. Congress did try to pass a joint resolution of disapproval, but it overwhelmingly failed.

At last, we had found a way to close unnecessary and inefficient military bases.  

We need the same political dynamic as to research bureaucracy.

Either the National Science and Technology Council or a “Research Policy Board” (as already envisioned by the 21st Century Cures Act and in Cures 2.0) could independently assess all of the bureaucratic obstacles – from proposal requirements, to reporting formats, to conflict of interest reporting – and decide on a list of 10, 20, or 50 procedures or rules that need to be streamlined.

If that list automatically went into effect unless Congress rejected a specific line item, now the burden would be on people who want to preserve a bureaucratic requirement to figure out how to build a coalition on that front. Otherwise, the reduction would go into effect along with the rest of the list.

By flipping the public choice calculation, we’d finally be able to make progress after decades of stagnation. We’d be able to get billions of dollars of research — for free.


As a second best, we could have an Executive Order or legislative mandate that, unlike any previous effort, consisted simply of the following mandate: The NIH Director (or the NSF Director) shall work with the OMB Director and university officials to take whatever actions are needed within the next 5 years to reduce the administrative burden to 20% of researcher time rather than 44%.

A goal, a timeline, and a mandate seem like a good way of getting agencies to actually take action here. As well, such a mandate capitalizes on agencies’ expertise: The NIH Director and OMB Director will be better situated to have staff figure out which processes create the most burden and how to streamline them, and university officials would have a substantial incentive to comply with whatever NIH and OMB say.