The Invisible Graveyard
How the precautionary principle obstructs progress
Risk & Progress explores risk, human progress, and your potential. My mission is to educate, inspire, and invest in concepts that promote a better future for all. Subscriptions are free. Paid subscribers gain access to the full archive and Pathways of Progress.
At the outset of our exploration of human progress, we discussed how cognitive biases distort our perception of reality, making it difficult for us to accurately compare the past with the present and to assess risk or danger. I called this phenomenon our “reality distortion field.” Our innate attraction to negativity, our ingrained loss aversion, may have been of evolutionary benefit to our ancestors, but today they are a liability. When these irrational fears become built into our social structures and our institutions, they work against us, holding human advancement back, paralyzing us through what is known as the “precautionary principle.”
The precautionary principle is a decision-making approach that prioritizes avoiding potential harm when the evidence about risks is uncertain. It advocates taking preventive measures, even without proof of danger, to protect public health, safety, or the environment. This burden to prove safety for novel ideas and new entrants, however, creates a Catch-22 situation because quantifying and assessing those risks is difficult. As a consequence, adherence to the precautionary principle can lead to overly cautious regulations that stifle innovation and/or protect vested interests. As explained by Cass R. Sunstein, policies flowing from the “precautionary principle” are logically inconsistent; they prohibit everything from consideration because everything new engenders unknown risk. Furthermore, they favor current over new risks, even if the latter is actually safer; stasis gets preference over change. Indeed, the precautions we take to mitigate risk and negative stimuli can be deadly by themselves.
The precautionary principle is driving the expansion of the regulatory industrial complex in the West. The Code of Federal Regulations, for example, the repository where all Federal rules are codified, grew from roughly 10,000 to 190,000 pages of regulation from 1950 to 2022. Each of these new rules imparted an accumulating cost on our ability to innovate and progress. One study in the Journal of Economic Growth estimated that since 1949, the growing regulatory thicket in the United States slowed economic growth by two percent annually through 2005. This may not sound like much, but remember that growth is cumulative; had the regulatory levels stayed frozen in their 1949 state, the US economy would have been roughly three times larger in 2005.
Safetyism
The poster child of the precautionary principle is atomic power. Initially, nuclear energy promised to make energy cheap and largely pollution-free, but events like the Three Mile Island disaster in the US (where no one died) led to risk-averse institutions regulating the industry to near-death. As I mentioned earlier, between 1954 and 1978, for example, the United States built 133 civilian nuclear reactors; it has built just 2 more since, both after 2020. As a consequence, the US is running older, less safe reactors for longer, and our skies are more polluted than they otherwise could have been.
A similar phenomenon played out in Japan in the wake of the Fukushima Daiichi nuclear accident, where again, no one died. In fear of another nuclear incident, the Japanese government rapidly curtailed the country’s use of atomic power. Researchers found that the government’s reaction to the disaster, however, led to 1,280 deaths attributable to higher energy costs in the following winter. They summarized, “This suggests that ceasing nuclear energy production has contributed to more deaths than the accident itself.” This analysis likely undersells the harm of switching off nuclear power. It did not consider, for example, the environmental harm of returning to polluting fuels like coal, which, ironically, produce more harmful radioactive waste than a nuclear power plant. When the medicine is worse than the disease, we should probably stop swallowing the pills.
Speaking of pills, in an effort to keep our food and drugs safe, the Food and Drug Administration (FDA) may, in fact, now be taking more lives than it saves. In response to food sanitation scandals, Congress passed the Food and Drug Act in 1960, creating the precursor to the modern FDA. At that time, the agency had no power to review or approve new drugs before they went to the market. It merely had police powers to enforce the law after a violation took place, and the law was primarily concerned with truth in labeling. This changed in the late 1930s when a company called S. E. Massengill began marketing one of the first antibiotics, “Elixir Sulfanilamide.” In its liquid form, the drug was prepared with a solution of diethylene glycol, a toxic chemical akin to antifreeze. Many died, not from the drug itself, but from the preparation solution.
In response, Congress passed the 1938 Food, Drug, and Cosmetic Act, which established the FDA as a regulator with the power to review and approve drugs before they could be sold. The law, however, gave the FDA just 60 days to vet new drugs, and if it could not complete its review in that time, the drug would be automatically approved. Notably, the law required that the FDA verify only the safety, not the efficacy, of new pharmaceuticals. Then, in the late 1950s, came a disaster on another continent. Thalidomide, a drug marketed as a sleep and morning sickness aid, was erroneously deemed safe by regulators in Europe despite not having been tested on pregnant women. The drug turned out to be very toxic to unborn babies, leading to tens of thousands of birth defects and deaths.
Notably, Thalidomide had not been approved in the US, but the disaster compelled Congress to further increase the regulatory power of the FDA. In 1962, Congress passed the 1962 Kefauver–Harris Amendment, which required the FDA to verify that drugs were safe and effective. It also lengthened the drug review period to 180 days and eliminated the automatic approval provision. This was an irrational response; Thalidomide’s issue was safety, not efficacy. The FDA had prudently blocked the drug in the US due to insufficient safety data. This is a lesson in how the regulatory industrial complex, propelled by the precautionary principle, can metastasize in our institutions over time. Since then, FDA reviews have only grown more difficult and lengthy.
The drug approval process has become so arduous, in fact, that many drug formulations and medical devices are never funded in the first place. Indeed, the FDA is one culprit behind “Eroom’s Law”, which, as I mentioned earlier, describes how the cost of developing new drugs doubles every nine years. It is also a factor contributing to the exploding cost of healthcare in the United States. Some may argue that the FDA’s review process is a necessary evil, but there is good reason to think otherwise. Foreign drug regulators, for example, routinely approve drugs far more quickly than the FDA does, with no adverse effects on public health. Furthermore, even in the US, FDA-approved drugs are often prescribed by doctors for uses for which they were not tested. This so-called “off-label” usage is 100 percent legal and common for people with severe illnesses.
Even with today’s demanding drug approval process, thousands still die every year from FDA-approved products. While this graveyard is tragic and many will consider it cause to further tighten regulatory oversight, we must remember the graveyard that we cannot see. Economists Sam Peltzman and Dale Gieringer argue that far more people have needlessly died and/or suffered because the drugs or medical devices they needed were not available due to the FDA’s regulatory hurdles. For example, by 1988, it was well known that taking aspirin reduces the risk of myocardial occlusion. For many years, however, the FDA prohibited aspirin manufacturers from advertising this. We can only speculate how many people died as a consequence.
It should be noted that other industries function quite well without a comparable regulatory authority. In electronics, for example, manufacturers can voluntarily submit their products to Underwriters’ Laboratories and obtain their “UL” mark. This is not a legal requirement, but many retailers require this certification for sale. The tort system and insurers who must insure product risks also naturally hold producers to account. When someone is harmed by a drug or medical device, the company’s reputation is tarnished, they are dragged through court, and often must pay significant damages to victims. The combination of private certification, courts, insurers, and doctors ensures that products are safe.
Risk aversion has spread upstream to scientific research itself. Institutional review boards (IRBs), tasked with reviewing research methods to ensure the safety of participants, worked well until 1998, when, during an asthma study, a patient died. Predictably, an overreaction ensued, whereby IRBs began requiring warnings against irrational safety risks to future participants. In a study that sought to test the transfer of bacteria on the skin, for example, the IRB consent form warned participants of AIDS risk, even though one cannot possibly contract AIDS through the skin. It also warned of a risk of contracting smallpox, a disease that was eradicated in the 1970s! While the death of a participant is truly tragic, overprotection is even more so. Tightening oversight may save one life, but it also makes life-saving research immensely more difficult, indirectly taking many more lives.
The precautionary principle is spreading into the workplace as well. In 1950, just 5% of American jobs required a license to perform them; today 30% do. This is due, in part, to a shift in the American labor force away from low-skilled manufacturing and toward high-skilled services. It’s also a function, however, of overzealous state regulators and lobbying by professional groups under the specious guise of “safety and health.” Studies examining the effects of occupational licensing have not found any discernible health/safety benefits. For example, Kleiner and Kudrle found no correlation between the difficulty of passing a dental license exam and the quality of dentistry. Similarly, studies have found that more stringent licensing of mortgage brokers does not result in fewer foreclosures.
Many licensing requirements are illogical, unscientific, and unverifiable. For example, 16 states require a cosmetology license to perform hair braiding, but the main component of a cosmetology license is the use of chemicals, a skillset not needed for braiding hair. In Michigan, it takes 1,460 days to become an athletic trainer, but only 26 to become an emergency medical technician. In some jurisdictions, fortune tellers even require a license. How could one even judge the quality of a fortune-teller for licensure? Additionally, licensing has spread beyond those occupations that might pose a genuine risk to health and safety. For example, there is no reason that interior designers, auctioneers, travel guides, or scrap metal recyclers need a license to work, yet many states require them.
The narrative that licensing is necessary to protect public health and safety is, for the most part, false. It’s a comfortable lie we tell ourselves to create the illusion of safety and to satisfy the precautionary principle. If you still don’t believe me, consider this: licensing rules usually grandfather existing practitioners into the new regime. This means practitioners established before the licensing rules do not have to meet their own industry standards. In other words, the rules only apply to newcomers, not to themselves. Read between the lines, more often than not, licensing is a rent-seeking mechanism used to restrict competition. It’s not for you, it’s for them.
One study found no difference between the quality of floral arrangements made by florists in Louisiana (who must be licensed) and those in Texas (where they do not). Yet, consumers paid more for floral arrangements in Louisiana. Additionally, licensing restricts economic mobility, the brunt of which lands on the poor who have greater difficulty jumping through the legal hoops required to obtain a licensed occupation. The poor frequently lack the luxury of time and money to pay licensing fees, take courses, and pass exams. Licensure, in effect, traps many in a continuing cycle of poverty. Indeed, estimates suggest that occupational licensing results in 2.8 million fewer jobs in America and costs consumers $203 billion annually
Environmentalism
The environmental movement is similarly plagued. Signed into law in 1970, for example, the National Environmental Policy Act, or NEPA, was intended to give a voice to environmentalists. The law required federal agencies to produce a “detailed statement” of the environmental effects of any “major action” before securing permits. Initially, those assessments were less than ten pages. However, because NEPA created an avenue for lawsuits, each new lawsuit set a precedent for the next, expanding its scope far beyond its original intent. By 2020, Environmental Impact Statements were typically hundreds of pages in length, and the mean preparation time was 4.8 years. What began as an effort to curtail environmental risks evolved into a tool that blocks and delays projects that reduce humanity’s environmental footprint, including clean energy projects like wind and solar power stations. Paradoxically, NEPA protects the existing fossil fuel-driven infrastructure by making it difficult to build anything else.
Our efforts to protect the environment through regulation are hypocritical and unscientific. Maxwell Tabarrok, writes that the US Fish and Wildlife Service, for example, plans to kill 470,000 invasive barred owls to protect the Northern and California spotted owls, who are unable to compete. If the goal is to minimize human environmental impact, however, human conservation actions run precisely contrary! Every species we try to protect was once an “invasive” species itself, such is the nature of the natural world. In the abstract, our conservation efforts are deepening human environmental impact as we humans are picking and choosing which species to protect. Now, many argue that the purpose of conservation is to preserve biodiversity, but even viewed in this light, our efforts are futile, at best. Human activities, including the clearing of land and hunting, are certainly exacerbating the natural extinction rate. What isn’t clear yet, however, is to what extent human activities increase the speciation rate. Some research suggests relative balance; for every species that goes extinct due to human activity, one new species, better fit for the human world, emerges, with no net loss of biodiversity.
None of this is to suggest that we ought to abolish all regulators and regulations or act with wanton disregard for human safety or our environment. Some regulation is certainly beneficial, and where the risks are systemic and broadly existential, adherence to the precautionary principle may even make sense. Too often, however, our efforts to procure safety and security, initially beginning with good intentions, grow into unnecessary and irrational barriers to human advancement and fulfillment. Reality, distorted by our cognitive biases, renders our gaze so affixed on illusory remediation that we neglect the invisible graveyard of lost hopes, dreams, and lives accidentally constructed to protect us.
Next in this series:
Previous:
Another important distinction between regulations are those that are performance based versus those that are process based.
The Clean Air Act and Clean Water Act are performance based. There is an evidence based process at EPA to set acceptable pollution levels based on best available science. This involves direct open debates about health impacts and cost benefits. Then from those standards permitting systems are developed to reduce pollution below the limits.
Process based regulations like NEPA are completely different. They have no objective, agency defined standards. They are supposed to be public disclosure documents but decades of lawsuits make them ever larger and larger.
For the FDA how much of the growth in expenses for approval is tighter standards and how much is more elaborate processes? I see the IRB craziness as a process step gone awry.
Do you think as a country becomes more "developed", the bureaucratic bloat around regulations deepen which inherently ushers in risk aversion which slows down productivity and progress? Is it a developed vs. developing world phenomenon and the stage in their economic cycle that they are in?