Years ago, when I still owned restaurants and a brewery in California, we tried to make a beer using olive leaves.
We weren’t allowed to.
Olive leaves weren’t some novel chemical compound we’d invented in a laboratory. People drink olive leaf tea. They take olive leaf tinctures and extracts. Olive trees have been part of the human food and medicinal landscape for thousands of years. But as I remember the regulatory process, olive leaf had not been approved for use specifically in beer.
We could potentially prove that our olive leaf beer was safe, but doing so would require studies and an expensive regulatory process that made absolutely no economic sense for a small brewery experimenting with one beer. So we didn’t make it.
I remember thinking at the time how absurd it seemed. But recently, as I’ve been learning more about a very different corner of our food regulatory system, that experience came rushing back to me.
Because while a small California brewery could be effectively prevented from putting olive leaves into beer, food manufacturers have historically been able to introduce certain new ingredients into the American food supply after determining for themselves that those ingredients are “generally recognized as safe,” or GRAS, without even being required to notify the Food and Drug Administration.
How can both of those things be true?
I am generally not an advocate for more government regulation of food. Quite the opposite. I think Americans should have considerably more freedom to buy food directly from farmers, drink raw milk, slaughter their own animals, sell homemade foods, grow vegetables in their front yards, and decide for themselves what they are willing to eat.
I don’t think the government needs to stand between every farmer and every customer.
But I also believe you have to know what you are eating in order to make that choice. And this is where I find myself surprisingly conflicted about the Trump administration’s effort to reform GRAS.
The GRAS designation grew out of the 1958 Food Additives Amendment. The basic idea makes perfect sense. Congress wasn’t going to require manufacturers to prove from scratch that ordinary substances with long histories of human consumption were safe. Salt is safe. Vinegar is safe. We don’t need a multimillion-dollar study every time someone figures out a new way to use cinnamon.
But over the decades, GRAS evolved into something considerably stranger.
Manufacturers could make their own GRAS determinations based on scientific evidence and expert evaluation. Although companies could voluntarily notify the Food and Drug Administration (FDA) of those determinations, they weren’t necessarily required to do so. That means the agency responsible for overseeing the American food supply couldn’t necessarily tell you every substance that companies had decided was safe enough to put into it.
The Government Accountability Office raised concerns about this exact problem more than 15 years ago, concluding that the FDA lacked information on GRAS determinations that companies had never submitted to it.
Think about that. We have built one of the most elaborate food regulatory systems in human history, yet the government charged with regulating food hasn’t even necessarily known every substance introduced under this pathway.
At the same time, Americans are remarkably sick.
A widely cited analysis published in JAMA estimated that suboptimal diet was associated with 318,656 American deaths from heart disease, stroke, and type 2 diabetes in a single year. That number deserves some perspective.
Before the measles vaccine was introduced in 1963, approximately 400 to 500 Americans died from measles in an average year. We have spent an extraordinary amount of political, cultural, and emotional energy fighting about vaccines.
During COVID-19, people lost jobs over vaccination decisions. Families fractured. People were excluded from restaurants, schools, and workplaces. We argued endlessly about mandates, risks, efficacy, and whether government or employers should be allowed to compel a medical intervention.
And I don’t think the public has always been given transparent information in that debate either. Pharmaceutical companies have enormous financial interests. Regulators depend in part on data generated by manufacturers. Government agencies, pharmaceutical companies, medical institutions, and the media have relationships and incentives that deserve scrutiny rather than blind trust.
That doesn’t mean vaccines never work or that every regulator is corrupt. It means financial incentives and institutional capture don’t magically disappear because the product being discussed is pharmaceutical rather than food.
Yet consider the scale of what barely produces the same urgency. Suboptimal diet was associated with an estimated 318,656 American cardiometabolic deaths in a single year.
That doesn’t mean food additives killed 318,656 people. They didn’t. That number encompasses the much broader consequences of poor diet, and it would be irresponsible to pretend otherwise. But perhaps that’s exactly the point.
We can spend years tearing ourselves apart over one category of health risk while something we put into our bodies three times every single day contributes to an enormous burden of disease almost in plain sight.
Americans increasingly eat food that would have been unrecognizable to our great-grandparents. Ultra-processed foods now make up a huge portion of the American diet. They can contain combinations of refined starches, sugars, industrial fats, emulsifiers, stabilizers, colors, preservatives, and flavor systems designed to make food inexpensive, consistent, shelf-stable, transportable, and extraordinarily appealing.
Not every processed food is unhealthy, and not every food additive is dangerous. Processing itself isn’t inherently bad. Cooking is processing. Fermentation is processing. Cheese is processed milk.
The question is: What have we been processing food for?
For the better part of a century, we have relentlessly optimized the American food system for efficiency: make food cheaper, make it last longer, make it travel farther, make every package identical, make it convenient, make it scalable, and make it microbiologically safe.
And on that last measure, we have accomplished extraordinary things. Commercial kitchens obsess over refrigeration temperatures. Restaurants have handwashing requirements. Slaughterhouses have inspectors. Milk is regulated. Farmers face rules about processing and sales. Labels contain pages of required information.
I have spent my adult life producing food, farming, and running restaurants. I know firsthand how regulated food production can be. Yet somehow we have managed to create a population suffering staggering rates of obesity, diabetes, cardiovascular disease, and other chronic illnesses intimately connected with diet.
Maybe we need to ask an uncomfortable question: What exactly do we mean by calling food “safe”?
Our food safety system is exceptionally concerned with whether something will make you sick tonight. Did the chicken remain below the proper temperature? Was the milk pasteurized? Did the cook wash her hands? Could bacteria grow in this jar? Those questions matter. Acute foodborne illness can kill people.
But there is another question that may matter just as much: What happens when you eat this food every day for 40 years?
A food can apparently pass every regulatory hurdle, sit legally on a supermarket shelf, and never give you food poisoning while still being part of a dietary pattern associated with metabolic disease and premature death. Perhaps we have confused “won’t poison you today” with “safe to eat for a lifetime.”
That doesn’t lead me to the conclusion that Washington should control our dinner plates. It leads me somewhere almost opposite.
Maybe we’re regulating the wrong things.
Why should a small brewery face an expensive regulatory hurdle for experimenting with an olive leaf, a plant humans have consumed in other forms for generations, while a genuinely novel industrial ingredient can find a pathway into processed food without the company even being required to tell the FDA about its safety determination?
Why do traditional foods and small producers so often encounter extraordinary regulatory friction while industrial food manufacturers have access to regulatory pathways most ordinary Americans don’t even know exist?
There is a legitimate reason for GRAS to exist. We should not force every farmer, chef, or food producer to prove again that humans can safely eat salt, vinegar, or herbs with centuries of traditional use. But a substance with no meaningful history in the human diet is different.
If a corporation invents a genuinely novel ingredient and intends to put it into food consumed by millions of people, I don’t think it is unreasonable to ask what it is, who determined that it was safe, what evidence they relied upon, and whether anyone without a financial interest looked at that evidence.
That isn’t necessarily an argument for government controlling what we eat. It is an argument for transparency because freedom of choice requires information.
I cannot make an informed decision about what I feed my children if I don’t know what is in their food. A free market doesn’t work particularly well when the seller possesses information about a product that the buyer could never reasonably discover.
This is the tension I can’t quite resolve. I don’t want a government powerful enough to decide what every American is allowed to eat. I’ve spent too many years watching small farmers and food producers struggle under regulations written for an industrial food system.
But I also don’t want a food industry powerful enough to conduct a population-scale experiment without meaningful transparency simply because an ingredient fits through a regulatory loophole.
Perhaps the answer isn’t more regulation. Perhaps it is a different regulation.
Get government out of the way of traditional foods, small farmers, and direct relationships between people who produce food and people who eat it. Stop pretending an olive leaf suddenly becomes an unknown substance because someone puts it into beer.
But when we invent something genuinely new and plan to feed it to hundreds of millions of people, the burden should run in the opposite direction. Show us what it is. Show us the evidence. Tell us who paid for the research. Tell us who decided it was safe.
And then let us decide whether we want to eat it.
After decades of making food safer, cheaper, more convenient, and more efficient than at almost any point in human history, Americans are remarkably sick from diseases closely connected to what we eat.
Maybe the problem isn’t that we haven’t regulated food enough.
Maybe we’ve spent all this time regulating the wrong things.
