The Fda vs. Recycling:
Has Food Packaging Law Gone Too Far?

My name is Alexander Volokh of the Reason Foundation, and I'm going to be talking to you about food packaging regulation.

Food packaging is not a sexy topic. I know of only one instance where it has affected popular culture in a lasting way. This is in The Music Man, the musical by Meredith Willson and Jerry Herman, where in the opening piece, traveling salesman from the turn of the century discuss how the trappings of modern society have changed their profession. In that piece, one salesman describes the effect of food packaging: "Why it's the Uneeda Biscuit made the trouble. Uneeda, Uneeda, put the crackers in the package, in the package, the Uneeda Biscuit in an airtight, sanitary package, made the cracker barrel obsolete, obsolete." At this, the other salesmen chime in in agreement: "Obsolete, obsolete, obsolete."

Most people, though, don't find food packaging very interesting, and this is a shame. Food packaging is living proof that some of the most wonderful, life-enhancing, freedom-increasing innovations happen in some of the most mundane-seeming areas. And people who don't appreciate the crucial role that food packaging plays in our lives are unlikely to spend much time critically evaluating the regulations the food packaging industry lives under, or feel outrage when such regulations go overboard. Many people have strong feeling in general about the FDA; some folks excoriate the FDA for allowing people to take drugs that can kill them, while others bash the FDA for not allowing people to take drugs that might save them. Most of these people are silent when it comes to FDA regulation of food packaging. Who ever got excited over a Coke bottle?

But food packaging is your life, and the silent miracles of packaging innovation touch the lives of your customers, and their customers, all the way down to me, the supermarket shopper. Of all the areas regulated by the FDA, I think food packaging is the field where the FDA is most reasonable and gets it right most consistently. Still, there are some major problems, which I will try to point out to you today in the context of recycled plastics. As a wise man once said, "I think sometimes the FDA lives in an ivory tower."

Most of you probably already know the surprising, and counter-intuitive, nature of food packaging law. Packaging isn't a food, it's not a drug, and it's not a cosmetic, but the FDA still regulates it under the Food, Drug, and Cosmetic Act. Officially, food packaging is regulated as a food additive. Unlike NutraSweet and Yellow #5, food packaging isn't intended to do anything to the food. But by the Second Law of Thermodynamics, substances diffuse over time, so given enough years, the components of the Coke bottle will become part of your Coke, and vice versa. So the components of food packaging law are regulated as "indirect food additives."

The most surprising aspect of food packaging law as it relates to recycled materials is that there isn't really any. The Food, Drug, and Cosmetic Act doesn't mandate any special regulatory review or additional FDA clearance for recycled materials. This is because the FDA regulates food-contact materials based on their composition, and based on the identity and amount of the material that can migrate to food -- not based on whether the materials are virgin or recycled plastic. Any recycled material can be used if it complies with current regulatory requirements.

The trouble, though, is that the people who wrote the original regulations weren't really thinking about recycled materials. The regulations assume that packaging is being made out of virgin materials, or out of materials that come from a controlled source (rather than recycled materials, which could come from thousands of unknown sources). Materials have to be "suitably pure," but the regulations doesn't say how "suitable" is "suitable." For instance, the regulations officially allow recycled paper and paperboard as long as neither "bears or contains any poisonous or deleterious substance which is retained in the recovered pulp and that migrates to the food." This is a very general statement, and could be interpreted as requiring a zero tolerance for contaminants. In fact, FDA regulation of recycled materials has been a mixture of discretion, rulemaking, and informal opinion letters. These informal opinion letters are called "non- objection letters."

At first, food packaging developers didn't have much guidance as to what to expect from the FDA. They submitted their best guess as to what the FDA wanted, and hoped that the FDA would issue a non-objection letter. Sometimes, the FDA would respond by asking for more information, and the whole process took several months. The FDA issued guidance documents, which informally established that chemicals with dietary concentrations below 0.5 parts per billion (ppb) -- that is, chemicals which could be expected to be less than one two-billionth of your entire diet -- were O.K. In reality, though, because of the conservatism of the FDA testing procedures, the actual dietary concentration is much, much lower than whatever the testing might reveal. As the FDA put it, testing assumed various worst-case scenarios; all material entering the waste stream was assumed to be contaminated.

In October 1993, the FDA announced that it was considering adopting a "threshold of regulation" for food additives; the rule became final in July 1995. The FDA has already been following this rule informally for a few years. Under the threshold of regulation policy, certain substances are formally exempted from regulation as food additives. To qualify for the exemption:

• The substance shouldn't be carcinogenic, because the use of carcinogens as food additives is prohibited by the Delaney Clause. There should also be no reason to suspect, based on the substance's chemical composition, that it might be a carcinogen.
• The expected concentration of the substance in the diet should be less than 0.5 ppb. If the substance is already also being regulated as a direct food additive, then the amount of the substance that gets into food through the packaging should be no more than 1 percent of that substance's acceptable daily intake.
• The substance should have no technical effect in or on the food.
• The use of the substance should have no significant adverse effect on the environment.

Thresholds of regulation are necessary because with advancing technology, instruments can detect smaller and smaller amounts of any substance. But just because they're there doesn't mean they're dangerous. The threshold of regulation standards are the ones now being used to judge non-objection letter requests. Substances that fail these conditions can still be approved, but they need to go through the regular food additive process, which takes a pretty long time.

As I've said before, food packaging regulation is one of the most reasonably regulated areas of all the fields regulated by the FDA. The reason it's reasonable is that it recognizes that the dose makes the poison, and that a chemical that is dangerous in high concentrations might be harmless at low ones. This philosophy dates back at least to the 1979 case of Monsanto v. Kennedy. In the late 1970s, Monsanto made a Coke bottle out of an acrylonitrile styrene copolymer. The FDA had at first approved the bottle, but after finding out some bad toxicity data on acrylonitrile -- one of the components of the bottle -- it withdrew its approval. Monsanto sued the FDA, because there was so little acrylonitrile in the bottle that it could never pose any real health risks. Ultimately, the court told the FDA that food and drug law doesn't prohibit teensy-weensy amounts of chemicals. This case was important because it established that there's such a thing as a "reasonable expectation of no migration." In the decade following the Monsanto case, as analytical models became more sensitive to minute quantities of chemicals, the FDA became more and more willing to say "this is so small that we don't care." In the end, both consumers and plastics companies came out ahead.

But even though FDA food packaging regulations are less troublesome than they might be, they still exaggerate risk, and they do so consistently. As a general rule, it makes sense that recycled material should be more suspect than virgin material. Recycled material usually comes from many sources, and as the saying goes, "You don't know where it's been." Or what's been in it. Recycled plastics are less "pure" than virgin plastics, because they may contain trace amounts of the original contents of the package, or whatever the previous consumer happened to store in the container -- pesticides, gasoline, rat poison, and so on.

Even so, it's important to make sure that our safety standards aren't too strict. We should remember that whenever we tighten a safety standard, we may improve safety, but at the cost of shutting out potentially promising technologies. As we get safer and safer, the increased safety from tightening the standards gets smaller and smaller, while the costs get higher and higher. Since recycled materials are potentially more troublesome than virgin materials, an overconservative standard would hit recycled materials harder, and shut recycled materials out of the market altogether without improving safety.

One problem is that under the new, finalized threshold of regulation policy, an environmental assessment has to be done before a non-objection letter is issued. If the review goes well, the FDA issues a "Finding Of No Significant Impact" (FONSI). Otherwise, the FDA conducts a full-blown "Environmental Impact Statement" (EIS). EISs take a very long time; for instance, the FDA has spent seven years not completing an EIS on polyvinyl chloride (PVC) bottles. The FDA has only had to conduct an EIS twice in the last 30 years. But even the less rigorous environmental assessments take time -- about three to six months. Now that they're required, they could double the time required to get a non-objection letter.

But the FDA's threshold of regulation continues to be conservative in other ways as well. The Delaney Clause prevents the FDA from approving a substance that causes cancer in any laboratory animal except under certain circumstances, even though it's a very large leap from cancer at high doses in a laboratory animal and cancer at low doses in a person. If something causes cancer in any laboratory animal, the FDA assumes that it causes cancer in people, even though tests on mice and rats give contradictory results 25 percent of the time, and people are more different from mice and rats than mice and rats are from each other. If many studies have been done on the same chemical, the FDA chooses the most sensitive species, sex, and organ combination for its carcinogenic potency. The FDA assumes that there is no threshold below which a chemical is harmless. And if testing methods don't find a chemical, the FDA doesn't assume that it's not there; instead, it assumes that the concentration of the chemical is just the limit of the detection method. In other words, if a testing method can detect anything larger than 50 ppb, then a negative result is interpreted to mean a concentration of exactly 50 ppb.

Some say the Delaney Clause may be repealed in whatever watered- down FDA reform may come out of the current Congress, but that's not certain, and as Samuel Goldwyn is said to have said, "I never make predictions, especially about the future." As for the rest of the conservative assumptions, they're pretty much entrenched in government rulemaking procedures, and will be tough to change. Is that much conservatism necessary? Given that we don't know everything about toxicology, any one of the conservative assumptions may be justifiable, but this many layers of conservatism is ridiculous.

A bit paradoxically, the FDA has noted that the threshold of regulation, which was supposed to lighten the agency's workload by loosening regulations, may in fact end up tightening the regulations instead, and increasing the agency's workload. It used to be that any chemical with a concentration under 50 ppb only required a rodent feeding study and a literature search to find published studies on the substances' effects. Now, a threshold level of 0.5 ppb would focus regulatory attention on exposures far below 50 ppb.

And numbers like 0.5 ppb and 1 percent of the diet, while they may be nice, round numbers, are in fact drawn out of thin air. They were derived, in a circuitous way, from the standard assumption that risks below one-in-a-million are acceptable while risks above one-in-a-million aren't, but even that number is drawn out of thin air. It's a convenient number to use, but it plays into the myth that we need to have one standard for everything and everyone, even though the food inside the package may be far more dangerous than the package itself, even though unregulated natural chemicals may pose far greater risks than the regulated manmade ones, and even though people willingly, without giving it a second tought, expose themselves to risks greater than one-in-a-million every day.

These, in short, are the main problems behind the FDA's risk assessment. According to Jerry Heckman, the general counsel of the Society of the Plastics Industry, the system is "archaic," "unnecessarily demanding," and "inefficient in the use of manpower and scarce resources." According to former FDA general counsel Peter Barton Hutt, indirect additives need "a completely separate and different process, with far less government involvement." Reforming these areas should make a significant difference to the development of new technologies, including recycled-content food packaging applications.

It may seem, in all this, as though I'm arguing against safety.

In a sense, I'm not. As I've pointed out, the FDA follows risk- assessment methods that consistently exaggerate risk. Given all the levels of conservatism built into the process, the true risk is almost guaranteed to be lower, perhaps significantly lower, than the FDA's calculated risk.

Now, conservatism is not a bad thing, particularly with food. As Charles Lamb put it, "I hate a man who swallows his food, affecting not to know what he is eating. I suspect his taste in other matters." Many of us are risk-averse in our daily lives, and would feel uncomfortable if we took risks that were above our risk threshold, whatever that was. Many of us even have different risk thresholds for different activities.

But that's risk management. Those are decisions about how we ought to respond to risks. Before we make these decisions, we should have accurate information about what those risks are. That's risk assessment. Conservatism can be useful and beneficial at the risk management stage; at the risk assessment stage, it's generally counterproductive. So in a sense, I'm not arguing against safety. Rather, I'm arguing for a return to accurate and honest calculations of risk. These honest calculations will reveal that many food packaging applications, especially applications with recycled content, are in fact a lot less dangerous than they now appear.

But in another sense, yes, I am arguing against safety. Not against safety as such, but against a particular safety ideology that many people in the public health establishment are in the grip of. This ideology holds (1) that chemicals in food, given current technology, pose substantial health risks, (2) that public health risks can and should be reduced to zero, or as close to zero as we can possibly get, (3) that public safety is the most important thing of all, and (4) that man-made substances are more troublesome than natural substances.

This philosophy is easy and sounds plausible, but suffers from a subtle but crucial defect, which is that it is wrong.

First, as I've pointed out in this talk, public health agencies, including the FDA, the EPA, and state and local health and safety agencies, consistently overestimate risk. Is the risk really there? Let's consider. Before the Food Additive Amendments in 1958, there was never any dramatic health-threatening incident like thalidomide in food packaging. No government or responsible authority in the United States had ever raised serious concerns about food packaging safety. Since 1958, there has not been a single FDA case aimed at condemning or seizing food because of hazardous packaging. What's more, market forces already do a good job of regulating packaging. In most cases, problems are signaled by a food taste or odor problem. No manufacturer would use -- because few consumers would buy -- such a package.

Second, risk can never be reduced to zero. Everything has some chance, however slight, of killing or injuring us, and there's literally no way of avoiding that. We can make risks arbitrarily small -- one in a million, one in a million, one in a trillion -- but we can't make them zero.

Third, we shouldn't even want to reduce risks to zero. Safety, like everything else, comes at a cost, and any dollar spent on reducing the risks of packaging is a dollar that can't be spent on reducing other risks or otherwise improving the quality of life. Any policy that puts safety above all other goals will ultimately lead to unnecessary high costs and harm safety in the long run.

And finally, it makes no sense to single out man-made chemicals as evil while exonerating natural chemicals. There is more risk of cancer in apple juice, peanut butter, and coffee than in the containers they come in. This risk is all-natural, and people accept it without reservations. A public policy that insists on reducing the risks of packaging while still recognizing that people have the right to choose coffee must have crooked priorities.

So when we call for loosening FDA restrictions on packaging in order to allow new and better technologies to flourish, including recycled-content packaging, we're not arguing against real safety. Instead, we're arguing against a fake, ideological concept of safety, exaggerating risk, biased against man-made chemicals, and driven by an unrealistic desire to reduce risks to zero regardless of costs. We're arguing against a food packaging law that has gone too far.

Alexander Volokh is the author of "The FDA vs. Recycling: Has Food Packaging Law Gone Too Far?", a policy study available from the Reason Foundation, 310-391-2245.