Once upon a time, there was a little biotech startup. It had novel technology and a bright idea: what if we could genetically modify the Atlantic salmon, a staple of modern diets, to grow faster and require less food while doing so? The science worked out superbly well, the concept was proved. All that remained was to acquire regulatory approval.
Almost thirty years later, American consumers will, in a few months, finally be able to buy AquAdvantage Salmon (AAS) courtesy of AquaBounty Technologies. I am very bullish that transgenic salmon will succeed on the market, and AquaBounty will finally be profitable. In due course I suspect that all kinds of genetically modified fish will find their way to our plates, with enormously beneficial consequences for wild fish stocks.
But even assuming that all ends happily ever after, there is no justification for AquaBounty’s thirty years in regulatory purgatory. The company’s journey is a showcase of a more general malaise that cuts the growth rates of Western economies, lowers productivity, and dramatically slows the rate of innovation. The story has many villains beyond the FDA, and this is not a simple tale of an overzealous regulatory agency. Both Democrat and Republican politicians at the highest levels have deliberately sought to destroy the company, aided by a whole host of parasitical civil society organizations. My conclusion is that the core problem of the West is not over-regulation per se, but the growth of a political culture that offers far too many informal vetoes to too many actors.
The original Nature paper describing the transgenic salmon now about to come to market was published in 1992. Three years later, the company filed an Investigational New Animal Drug file with the FDA’s Center for Veterinary Medicine. That a company explicitly aiming to develop a food for human consumption wound up under the regulatory umbrella of the Center for Veterinary Medicine should tell you something about how ill-adapted the regulatory environment of the time was to a novel technology such as transgenic fish. The FDA did not issue “Guidance for Industry 187: Regulation of Genetically Engineered Animals Containing Heritable Recombinant DNA Constructs” until 2009.
Eventually, in 2003, AquaBounty filed its first study to the FDA. It took another six years for the company to file its final study. What caused the delays between 1995 and 2009 is not entirely clear, but part of the problem seems to have been the FDA’s desire for the company to demonstrate the stability of the genetic line over multiple generations. Whether seven generations worth of data was necessary is debatable.
Some part of the holdup may also have been related to the exhaustive allergenicity data the FDA demanded. Why the FDA were so concerned about this remains unclear. Certainly it seems unlikely that one would expect a faster-growing salmon to possess higher allergenic properties than a wild salmon. Furthermore, as a scathing editorial in Nature Biotechnology pointed out, all kinds of non-GM crops and foodstuffs exhibit extremely high natural variability in allergenicity, so there existed no obviously appropriate comparison class to evaluate the AquAdvantage salmon against.
The heaviest regulatory burdens placed on AquaBounty, however, were undoubtedly environmental. In order to avoid even the remotest possibility of escape by their transgenic salmon, the company has been confined to landlocked inland farms that use recirculating aquaculture systems (RAS), whereas traditional salmon farming uses oceanic pens just off the coast. The AquAdvantage salmon grown for sale are all-female and infertile.
No doubt some of these measures were due to justified risks of escape, although even then it has never been clear that AquAdvantage salmon actually enjoy a fitness advantage. But many of the actions taken by both the FDA and the company are best explained as a form of preemptive defense. This became apparent in 2010 when, the company having finally satisfied the FDA that its salmon was ready for approval, congressional politics kicked into high gear. Alaskan Republicans sponsored a budget amendment banning the FDA from approving AquaBounty’s product. Alaska, naturally, has a large fishing industry.
The Senate did not take up the measure and come 2012, the FDA was on the verge of formally approving AquAdvantage salmon for human consumption, despite enormous pressure from concern trolling NGOs such as the “Center for Food Safety”, the “Union of Concerned Scientists”, and “Food and Water Watch”. At this point the White House suddenly sought to block approval. Evidently the Obama administration did not want the public embarrassment of alienating the “crank left” part of its voter base just before the President sought re-election - a base already upset over Obama’s refusal to block the Keystone pipeline. All communication between the FDA and the company suddenly ceased.
At this point, John Entine of the Genetic Literacy Project stepped in to expose the Presidential foot-dragging in Slate Magazine, which somehow found space to publish the article amidst the five different reviews of each episode of Girls that they were running at the time. Faced with the choice between the votes of aging hippies and “I fucking love science” bros, Obama’s team opted for the latter. The White House logjam was removed.
Still the saga continued. Issuing a preliminary Environment Assessment of “no adverse effect”, the FDA invited a 60-day public commentary period, later extended to 120 days for no good reason. It took until 2015 for AquaBounty to finally get their New Animal Drug application approved, only for Lisa Murkowski to squeeze in a rider to the 2016 Appropriations Act that directed the FDA to issue an import alert for genetically engineered salmon, preventing AquaBounty from selling AAS grown in their Canadian facility (at the time, their only facility) in the US. The FDA was directed to maintain the import alert until it developed new labelling guidance.
It took until 2019 for the FDA to remove the import alert. Murkowski tried one more time to finish Aquabounty off, this time by slipping in a rider directing the FDA to undertake a “label comprehension study”, a process that aims at understanding whether or not consumers understand a drug’s intended use cases and potential side effects. This time, however, AquaBounty managed to mobilize some political support. It had completed construction of a new farm in rural Indiana, and the state GOP swung into action. Murkowski, it seems, partially backed down, and barring some new labelling requirements, the company has finally been allowed to proceed unhindered to commercialization. Friends of the Earth and other troll NGOs are still abusing the legal process to get the FDA’s approval revoked on the grounds that the agency did not give sufficient weight to environmental consequences in the approvals process, but at this point their chances of success seem slight.
I have a few takeaways from this story:
For the last few decades, we have had the technology available to dramatically increase productivity in the farming of a key foodstuff. We just haven’t used it, and even now there is little prospect that anywhere in Europe will allow transgenic fish to come to market any time soon, even if AquaBounty becomes a big success in the US. This makes me wonder how many other magic growth-enhancing money trees we are leaving to wither.
The FDA - and AquaBounty - were working with a false model of how to minimize opposition. Both seem to have laboured under the presumption that a long, painstaking, and highly burdensome approvals process would be the best way to assuage irrational safety concerns around a controversial product. In reality, all the lengthy process did was give opposition more chance to mobilize. At one point, the Center for Food Safety even argued that AquaBounty’s lengthy battle with the FDA was a sign that AAS must be in some way unsafe!
Some people are just temperamentally opposed to progress, and others have commercial incentives to stop new products coming to market. A regulatory regime aimed at appeasing the first group will never work, and all it does is allow them more time to form an effective coalition with the second group. Furthermore, the length of the process arguably serves as much of a negative signal to consumers as it gives assurance of safety.
As we’ve seen with covid vaccines, a very good way to assuage consumer concerns over something new is just to let people try it. Vaccine skepticism has collapsed in US and UK polling over the last few weeks, as people can simply see that the millions of vaccine recipients have been fine. If US consumers had been eating AAS since 2003, how different might attitudes be today towards genetically modified meat and fish?Much of “civil society” has been hollowed out, and enormous amounts of NGO activity is destructive. Perhaps this is a long-term consequence of the growth of the state leaving less room for genuinely useful charitable activity, but the AquaBounty story is another instance of numerous NGOs functioning purely as well-funded lobby groups, drawing from the same pools of foundation money. In the UK, eminent political commentators have recently drawn attention to how much public policy discourse is shaped by sockpuppet organizations funded by the Joseph Rowntree Foundation, but the influence of similar left-wing foundations across Europe and the US remains insufficiently examined by the press.
Altruistic persons interested in the pursuit of progress should spend far more time trying to pre-emptively fix regulatory regimes, in advance of - or at least in parallel with - the development of new technologies. AAS should never have been regulated as a New Animal Drug. Arguably it should never have come under the FDA’s remit at all, and would have received far more reasonable treatment from the Department of Agriculture. I can understand a certain degree of regulatory caution when it comes to new pharmaceutical drugs, especially those with a novel mechanism of action. This is just a fish that grows faster, and the mindset suited to one was clearly ill-adapted to regulate the other.
Far too many people working on “AI ethics” and so on are sympathetic to, and funded by, anti-growth NGOs. Although it is entirely natural that most people interested in innovation will want to go work for innovative companies, there is evidently a crying need for techno-futurists to go into government. Pro-growth firms, foundations, and individuals should consider how best to create and train networks of like-minded people that can shape the regulation of the future, and act as focal points for coalitions that fight to overcome the informal vetoes that block progress today.