The newly inaugurated second Trump administration has arrived, and among the changes that the president and his allies have proposed is large-scale simplification and elimination of government regulations. President Trump stated in a press conference in December that he wants 10 old regulations eliminated for every new regulation added.
The arguments for deregulation are not new. Regulations function as an indirect tax that slows economic growth: Compliance costs money, and the expense is often passed on to the consumer. Regulations also cost the government (and, therefore, the taxpayer) money to promulgate, revise, and enforce. Deregulation, thus, is a win for everyone because it frees money for economic growth—which increases tax revenue while reducing government spending.
Regulations function as an indirect tax that slows growth: Compliance costs money.Excessive regulations slow down goals for governments and citizens. Sixteen years after being approved by California voters, the state’s high-speed railway system has yet to carry a single passenger. One major reason the project has been repeatedly delayed is because of the state’s onerous environmental and other regulations, which put up red tape requiring years to cut through.
Scientists are also burdened by regulations that add costs and increase delays. The most important regulation for science is 45 CFR 46, which governs ethical approval of human-subjects research. It is known as the “Common Rule” because 20 government agencies have agreed to be governed by the regulation. This Rule is a worthy target for the new administration’s deregulation efforts because the Rule adds to the cost of scientific progress, delays research projects, and is probably unconstitutional, to boot.
The Rule mandates that all federally funded research receive ethical approval from an Institutional Review Board (IRB) before researchers can start any human-subjects research. The reach of IRBs is breathtaking, including on university campuses. Because the regulations define research as “a systematic investigation … designed to develop or contribute to generalizable knowledge,” many forms of scholarly pursuit require IRB approval. Likewise, the regulation’s definition of “human subjects” includes research that most Americans would not place in that category, for example analyzing existing archival data that contains identifying information, or conducting laboratory tests on frozen tissue or blood samples from a hospital’s patients.
The IRB’s purview extends to nearly all research at universities, hospitals, and other organizations that receive federal funds, because these entities must provide “assurance” that they will comply with the Rule; the only feasible way to do this is to monitor all research performed by researchers, students, and staff. As a result, the regulations do not explicitly state that non-funded research must receive approval from an IRB, but there is a de facto mandate for it.
The result of this regulatory breadth is a massive inefficiency. Researchers must wait weeks or months for IRB approval before they can begin projects, and projects must stop and receive a new approval if the researcher wants to change any aspect of the study that involves interaction with the participants. Complying with this red tape is expensive. A 2003 study found that the average cost per IRB action was $494 to $1,426 in 2024 dollars. These costs mount quickly because an “IRB action” includes an initial application, any revisions, periodic updates (usually annual), and reports filed when a study ends. No one knows how many studies per year are submitted for IRB approval, but a survey of IRB directors at large research universities and hospitals found that the average IRB handled over 1,300 new study submissions annually. Conservatively, the cost of IRB compliance at these institutions is over $100 million/year. With over 2,300 IRBs in the United States, the full cost of the Common Rule is much higher. Researchers do not pay these costs out of their own pocket. Instead, universities, hospitals, and other organizations pay for the IRB’s time, staff, and other expenses through the costs tacked on to grants as “indirect costs” or through taxpayer funds that support universities. Like most regulation costs, the expenses of an IRB are passed on to the public.
The IRB system is inherently unconstitutional when regulating many research activities.If the time and savings of reducing regulations of scientific research were not sufficient reason for the incoming administration to act, there is a compelling legal argument. Legal scholars have argued that the IRB system is inherently unconstitutional when regulating many research activities. Columbia law professor Philip Hamburger says the IRB system functions as a licensing system, wherein researchers must get permission from a government-mandated board before conducting research activities. The legal problem arises when the research consists of activities that would otherwise be constitutionally protected speech. This describes most social-science research, which often consists of merely asking people questions in interviews, surveys, focus groups, or laboratory settings. Most behavioral observations and psychological tests are also constitutionally protected speech.
Hamburger’s argument is obvious in a thought experiment on how the IRB system would apply to other constitutionally protected speech. He imagines the “constitutional danger” if federal regulations required journalists to submit their proposed articles to a government-mandated board that could demand changes to interview questions, halt a journalist’s work in progress, monitor ongoing work, or even veto a project. If it would be unconstitutional to create a review board for journalists, then it is equally unconstitutional to create one to monitor researchers engaging in speech.
No one relinquishes their constitutional rights when they work for a university; indeed, the Supreme Court has ruled and reaffirmed that “teachers and students must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding.” It is hard to square that judicial decree with the federal regulations mandating IRB permission to conduct research.
In response to concerns about the constitutional problems of IRB, Zachary Greenberg of the Foundation for Individual Rights and Expression stated
Institutional Research Boards, when properly limited to ensuring safe and ethical research on human subjects, are a useful tool for safeguarding academic standards. However, IRBs pose free speech concerns when they exceed this narrow purpose by restricting research into controversial topics, especially when there are no apparent safety or ethical concerns. Faculty and student researchers have the academic freedom to delve into a wide array of issues. Universities, acting through IRBs, improperly restrict this right when they stifle scholarly inquiry because of political or ideological concerns divorced from safety or ethical considerations.
Greenberg’s concern about censorship is justified. In one study, researchers sent three versions of an application for a proposed study to university IRBs. When the topic of the proposed study was discrimination against short or obese applicants in hiring, the proposal sailed through IRB approval. However, an identical study on discrimination against women or minorities was denied about half the time, and about two-thirds of proposed studies on discrimination against white males were not approved. Some IRB members in the study were brazen about why they were rejecting the proposed study of discrimination against white males: “The findings could set Affirmative Action back 20 years,” one stated.
There is a fundamental difference between doing something to a person’s body and asking them questions.Proponents argue that IRBs are necessary to protect research participants from harm. Indeed, the IRB regulations get their legal authority from the National Research Act, passed in 1974 in reaction to the Tuskegee Syphilis Study. In that study, medical researchers withheld treatment for decades from African American men who were already infected with syphilis. Other IRB proponents point to the atrocities committed by physicians in Nazi concentration camps.
The problem with using Tuskegee or Auschwitz to justify IRBs for social-science research is that neither were social-science research studies. Additionally, there is a fundamental difference between doing something to a person’s body and asking them questions (not the least of which is that the latter is constitutionally protected, while the former is not). This is why no one questions the need to regulate drug and medical-device research, which is handled under different regulations. Conversely, there is little evidence that social-science research is harmful.
Another problem with the argument that IRBs prevent harm is that the federal regulations try to prevent much more than the bodily harm that can come from medical research. In the Rule, “harm” includes “discomfort” and damage to a person’s reputation. This gives IRBs permission to interfere with research even when the actual harm is trivial or non-existent. Consent forms mandated by IRBs often warn participants that participating in social-science research could produce boredom or restlessness. Quelle horreur!
Broadening a definition to increase interference is not limited to concerns related to “harm.” Research on vulnerable populations requires higher ethical scrutiny, which is an uncontroversial perspective among scientists. The problem is that the definition of who is considered “vulnerable” has ballooned since the National Research Act was passed. Whereas the original law defined vulnerable people as children, prisoners, and “the institutionalized mentally infirm,” today the regulations include “economically or educationally disadvantaged persons,” and the government-established National Bioethics Advisory Commission’s definition includes people disadvantaged by “gender, race, or class inequalities,” “seriously ill individuals,” and people who “belong to undervalued social groups.” The expanded definition of who is “vulnerable” gives IRBs permission to meddle in a wide array of social-science research. Unfortunately, this inhibits research on the very groups that government busybodies see as vulnerable, thereby making it harder for science to improve their lives.
Even if one agrees with the regulations and definitions in their current form, there is little evidence that IRBs actually protect anyone. A 2023 report from the Government Accountability Office found that there was no procedure for measuring how well IRBs protect research participants. And pointing to the dearth of research scandals in the IRB era is poor evidence that IRBs protect anyone. The abhorrent Stanford Prison Experiment (conducted before the National Research Act was passed) received approval from the university’s ethics committee, for example.
One reason IRBs are not effective at protecting research subjects is that research ethics are often specific to disciplines, methodologies, and contexts. It is impossible for IRB members to know the ethical standards of conduct for every situation that scholars encounter. I saw this firsthand when I was a psychology professor and applied to use a popular intelligence test in a study. My university’s IRB chair demanded a copy of the test before approving the study. But complying with his request would violate ethical standards in my discipline, which include maintaining confidentiality of psychological tests. Ironically, the IRB chair was asking me to act unethically to pass ethical review.
New regulations should assume that research does not need to be regulated unless it falls into explicitly enumerated categories.Given the expansive nature of the Common Rule and its application, reform is warranted. I have some recommendations that I hope the second Trump administration will implement. First, the underlying assumption of the Rule should be reversed. Current regulations assume that research needs to be regulated unless it falls into narrow exceptions listed in the Rule. New regulations should instead assume that research does not need to be regulated unless it falls into explicitly enumerated categories that warrant oversight.
Second, regulations should state that research activities that consist of constitutionally protected speech cannot be regulated. This would immediately free up most social-science research and some medical research. Such a provision would also put to rest the canard that “research is a privilege, not a right.” After all, “if flag burning, porn movies, and nude dancing are constitutionally protected, so too is academic research and its publication,” according to Hamburger.
Third, regulations should focus less on general guidelines and instead empower the ethical codes of professional societies and disciplines to provide authoritative guidance. This will prevent IRBs from trying to impose improper or nonsensical ethical requirements onto disciplines and methodologies.
There is recent precedent for reforming the Common Rule. In 2018, the Rule was revised so that legal research, oral histories, and some other types of research were exempt from IRB oversight. The 2018 revisions also permitted research projects conducted at multiple locations to get approval from a single IRB instead of the IRBs at every site. This experience shows that simplification is possible. If the new administration’s appetite for deregulation extends to revising the Rule again, then the benefits for science and society will be immense.
Russell Warne is a former associate professor of psychology in the Department of Behavioral Science at Utah Valley University.