To address several points that were just raised:
I can't speak for everyone, but I'm certainly not trying to say that state-level regulation would automatically be "better" than federal-level regulation. In fact, my expectation is that some states would do a worse job, and some would do better. That's somewhat the point of the exercise, actually. The only way to honestly ascertain what kind of regulation works best (whether in relation to food and drugs, health care, or anything else) is to actually try it a bunch of ways and see what happens. But federal-level regulation is, by its nature, homogeneous. Consequently, we get a whole lot of talking heads bickering about what will work best, but very little actual data to make that determination. With a diversity of regulation, it should be possible to directly observe over time which states' systems actually work the best. Eventually, I would expect the various states' regulations to settle towards a fairly common set of policies that have demonstrably worked better than others.
I think that homogeneity in government is generally a bad thing. Especially homogeneity across very large populations. People are simply not homogeneous, and the larger the population is, the less desirable it is to treat it is if it were. Uniformity's only upside, really, is that it
may be more efficient than diversity. The efficiency argument is a fair point, but I think there are several counter-points worth considering:
One, efficiency isn't the end-all-be-all of regulation. That's not to say that it shouldn't be given any consideration -- clearly it should -- but it's not the only important factor, nor necessarily the most important one. It's important only inasmuch as less efficiency usually translates into greater cost. However, it might be the case that giving up a little efficiency is worthwhile if doing so purchases something of greater value. See my argument in the first paragraph. Additionally, I feel like I need to reiterate my argument that The One Best System likely doesn't exist. If, for instance, a majority of people in a state are dissatisfied with the way food and drugs are regulated, should they be forced to comply with that system simply because it is more efficient? If they would be happier with a different, but less efficient system, why should anyone stop them? A state government is in a much better position to shape itself to the actual desires of its constituents than the federal government. That's not a slam on the federal government; it's just the nature of the beast. Federal things are federal: your state's constituents are
not its only concern.
Secondly, it isn't necessarily the case that diversity would be less efficient than uniformity. Having 50 different systems might be less efficient in some respects, but it is likely to be more efficient in others. As I've said, people are not homogeneous. Neither are markets nor really anything else about the underlying subject of regulation in this case. The price system is far more efficient at assessing value and lubricating economic transactions than any centrally planned system precisely because it is not uniform and therefore can take advantage of local knowledge and conditions better. Similarly, a more granular approach to food and drug regulation may produce more efficient results than a federal system because it better accounts for the local conditions. This may well be enough to offset loses in efficiency due to increased overhead for multi-state businesses and so on.
I also think that the potential diversity (and therefore presumed inefficiency) is being overstated -- at least in the long run. I've already outlined one reason that I believe a state-regulation system would ultimately settle into substantially similar systems. Additionally, I should point out that there's no reason to believe that the individual states, in pursuing their own regulation, would be incapable of making pragmatic considerations, and would chose to adopt regulations so radically different from one another as to be unworkably inefficient. That certainly hasn't been the way of things in other areas. Case in point: no one regulates the underlying technologies that make up the Internet and its many services and protocols (ex. HTTP, SMTP, HTML, etc., etc.). And yet, standards have not only coalesced, but have by and large been adopted (to a reasonable degree) world-wide. Not because someone forced them to from on high (no one is going to come arrest you because you sent a malformed HTTP request), but because people, left to their own devices, are actually quite capable of seeing that cooperation has value which may mitigate the value of individualism.
Finally (I think...), Corolinth's point needs to be repeated, because I feel it was dismissed out of hand. Given the at least comparable population and geographic size of individual U.S. states to European nations, the idea that state-level regulation is too granular to be efficient is suspect at best. Lydiaa offered an explanation as to why she thinks it's more workable in Europe, but I see no reason why the same sort of thing wouldn't happen in the U.S. if it were deemed beneficial to do so. States adopt other state's policies, or mutually adopt a set of policies developed by some independent body all the time. For example, many (most?) state highway departments have adopted the
AASHTO design standards more or less wholesale. This doesn't constitute an interstate treaty, and isn't unconstitutional. I don't know why people think food and drugs would be automatically different somehow.