State Policymaking Doesn't Require a Congress Limited by Enumerated Powers

Richard Primus

            Last Friday, I appeared on an episode of Eric Segall’s podcast Supreme Myths.  Most of the conversation was about my view of congressional power.  In a series of articles published over the last several years, and in a forthcoming book, I advance the view that orthodox ideas about Congress’s being limited by its enumerated powers are systematically misconceived.  Among other things, I think that on the best interpretation of the Constitution, Congress has non-enumerated as well as enumerated powers.  It doesn’t follow that Congress has a police power, but I do think that the scope of Congress’s legislative jurisdiction is very broad.  I won’t try to lay out the full view and its supporting reasons in this post.  Readers who want to hear the podcast episode can find it here; readers who want a more complete exposition of my view can read things like this, and this, and this.  Or they can wait for the book.  In this post, I’m writing to respond to a useful question that Orin Kerr posed after hearing the podcast episode. 

The question is whether, on my view, Congress could enact a general preemption statute that would nullify all state legislation, or at least all state legislation not falling within express constitutional language directing that certain legislation be done by states.  In a short Twitter thread, Orin presented this question as follows:

Let’s say Congress passes a new law, the State Law Is Dumb Act (SLID). It provides: “All state law of any kind on any topic is expressly preempted.”  I take it Congress can’t do this now because the federal government has limited power under the “enumeration” understanding of Article I; Congress can only preempt where it has power to regulate, which is limited. 

Here’s the general shape of my answer.  First, I’m going to affirm the suspicion that animates the question, which is that on my view Congress would have the authority to eliminate or at least radically reduce the ability of state and local governments to act as consequential policymakers—and that that would be a bad thing.  Second, I’m going to explain why this possibility isn’t a problem for my theory.  Third, I’m going to point out that the concern animating this question echoes a common concern about the scope of congressional power in the eras before the New Deal, and I’m going to suggest that just as we today can easily see the solution to the pre-New Deal concern, we should be able to see why we shouldn’t be worried about something like SLID.

            So first: Yes, on my view of congressional power, Congress could preempt an awful lot of state law.  Not all of state law, but an awful lot of it.  It is not my view that the Constitution properly interpreted vests Congress with a police power, and even if Congress had a police power, there would be affirmative constitutional prohibitions against Congress’s displacing several core pockets of state law.  (Two examples: Under Article I, Section iv, Congress cannot override a state law designating the places at which states choose their U.S. Senators, and under Coyle v. Oklahoma Congress cannot designate the location of a state capital.)  But I do think that Congress has legislative power at least as broad as indicated in decisions like Wickard v. Filburn and Katzenbach v. McClung, that the individual mandate of the ACA was easily within Congress’s power to enact under the Commerce and Necessary and Proper Clauses, that Congress should be understood to possess some amount of legislative power beyond that which the text of the Constitution expressly specifies, and that it is not necessarily the case that any particular policymaking domain is beyond the reach of congressional legislation, excepting those domains that are the subjects of affirmative constitutional prohibitions (textual or otherwise) on federal lawmaking.  Again, I don’t intend to explain or defend these conclusions here.  I mean just to acknowledge that those are my views and that if those views prevailed, Congress could enact extremely broad preemptive legislation.  Something as totalizing as SLID would still be overreach, but maybe not by much.  A slightly moderated version—call it AASLID, for Almost All State Law Is Dumb—would be within Congress’s power.  And AASLID’s preemption would be bad, because there’s a lot of value in having a lot of consequential policymaking done at the state and local levels.

            But second: The fact that Congress has the power to enact a destructive law doesn’t always point to a problem, let alone an incorrect interpretation of the Constitution.  Congress has the power to tax all American incomes at 99%, and it would bad for Congress to do so.  Ditto for the power to prohibit all commercial activity on Thursdays and the power to completely disband the U.S. military.  But none of these facts about congressional authority is a problem, because Congress isn’t going to do any of those things.  No plausibly foreseeable congressional majority would want to do them, and any momentarily insane Congress that tried would be quickly turned out by the voters.  (I’m not saying that congressional behavior reliably consists of reasonably people pursuing reasonable ends reasonably.  But there are limits to how pathologically Congress will behave, even under the distinctly unfortunate conditions of 2023.)  The constitutional system has several mechanisms for avoiding bad legislation, and the most important mechanisms are process limits, like the fact that large numbers of actual officeholders have to vote for something before it becomes law.  That limit, rather than anything about the scope of congressional authority, is the thing that prevents most of the bad things that Congress could theoretically do.  And I’m confident that it would prevent Congress from enacting something like AASLID, even if my view of congressional power were accepted.

            One way to explain why I’m confident on that matter is to point out that Congress has never legislated anywhere near the limits of its preemptive authority.  Even on the orthodox view by which Congress is limited to and by its textually enumerated powers, Congress has the authority to preempt enormous swaths of state and local law.  All of contract law could be preempted under Congress’s power to regulate commerce (or, if you follow Justice Scalia’s analysis in Gonzales v. Raich, Congress’s power to regulate commerce combined with its power under the Necessary and Proper Clause).  The same is true for all of corporate law, employment law, products liability law, professional licensing, and all other law connected to the production and consumption of goods and services, not to mention traffic safety law and all other law connected to transportation (including, say, the licensing of drivers).  A creative legislative drafter using the techniques that Congress used to reenact most of the coverage of the Gun-Free School Zones Act after United States v. Lopez could easily reach many more things, too.  Given Congress’s ability to attach conditions to the receipt of federal funds, Congress could probably also preempt a much broader subset of state regulation than it currently does in domains like education, policing, firefighting, environmental protection, and on and on.  But it doesn’t.  To be sure, there’s a bunch of federal preemption, and some of it is unfortunate, and different people will have different views about where it goes too far (or not far enough).  But in the aggregate, Congress leaves a great deal of state law standing in areas where it has the constitutional authority to preempt state law.  It does so for a combination of reasons: because the federal government lacks the resources to regulate all those domains effectively, because Congress thinks it would be unwise to federalize all those domains, because Congress correctly perceives that there would be strong political resistance to federalization, and so forth. 

            The point is that whatever limiting force the Constitution’s enumeration of congressional powers has is not what prevents Congress from preempting most state law.  Most state law exists in a zone of concurrent jurisdiction, where the state and national legislatures are both competent to regulate and state law operates to the extent that it is not inconsistent with national law.  For good reasons, Congress has let most of that law stand.  And there is no reason to think that the reasons why Congress does not displace most state law would disappear if we acknowledged that the Constitution’s enumeration of congressional powers should not be thought to constrain Congress in the way that prevailing doctrine holds.  Congress would have somewhat broader authority, but the incentives and other dynamics that prevent unlimited preemption over the domain of Congress’s already wide authority would still be in force.  Yes, in theory a future Congress not constrained by enumerated powers could go crazy and enact AASLID, just as Congress today could go crazy and disband the U.S. military.  But if we think it’s possible that a Congress not constrained by its enumerated powers could go crazy and preempt everything within its legislative reach, we should also think it’s possible that a Congress operating within current enumerated-powers limits (whatever they might be) might similarly go crazy and preempt everything within its legislative reach—and that preemption too would be plenty broad enough to be a terrible idea.  We don’t seem to be worried that that will happen, though, and rightly not.  It’s not going to happen, even though Congress has the authority to do it.  And if we see that we don’t need a judicially enforceable rule preventing Congress from legislating to the limits of its preemptive authority in our current world, even though it’d be terrible for Congress to do that, we should be able to see that for the same reasons we wouldn’t need a judicially enforceable rule preventing Congress from legislating to the limits of its preemptive authority in a world where Congress’s legislative authority was somewhat broader than the extremely broad authority it already enjoys.  We don’t need judicially enforceable rules preventing that legislation any more than we need a judicially enforceable rule preventing Congress from enacting a 99% income tax.

            Third: The worry that Congress might destroy American federalism by enacting preemptive legislation to the full extent of its legislative mandate is a modern version of a worry that preoccupied some jurists during the Constitution’s first hundred and fifty years.  During that earlier period, many people thought of federal legislative jurisdiction—either in its entirety or in some of its most important parts—as inherently exclusive of state jurisdiction, rather than as existing concurrently with it.  Some members of the Founding generation assumed that giving any particular policymaking domain to the federal government meant withdrawing that domain from the state governments and therefore opposed extending federal power not because they feared the federal government would govern abusively but because given the federal government’s limited practical capacity it was implausible that the federal government alone could regulate adequately, so disabling state regulation meant that domains of social life needing regulation for the public welfare would be dangerously neglected.  In the nineteenth and early twentieth centuries, many judges thought that anything coming within Congress’s commerce power could not be regulated by states, and that made it imperative to construe the commerce power narrowly, lest dormant commerce preemption nullify massive amounts of local regulation that Congress could not plausibly replace.  Often, that's what courts between 1890 and 1937 had in mind when they wrote that on a more expansive view of the commerce power, the American constitutional system would be destroyed: not that letting Congress regulate would be tantamount to that destruction, but that eliminating state policymaking would be.

The transition to modern commerce doctrine was made possible partly by the legal profession’s acceptance of the model of concurrent jurisdiction, on which the fact that Congress can regulate something under its commerce power doesn’t deprive the states of the power to regulate in the same domain.  Once we accept that the existence of federal legislative authority does not by itself eliminate state policymaking, the imperative to preserve meaningful state policymaking only requires constraining the scope of federal legislative authority if we expect that Congress will in fact occupy too much of the policymaking space to which it has access.  For more than eighty years, constitutional law has operated on the correct understanding that Congress will in fact not occupy anything close to all of that space, even though it has the authority to do so.  And the fact that Congress will not occupy anything close to all of that space is a robust feature of the system, rather than a matter of happenstance or easily revoked grace.  Put differently, part of what has kept the states meaningful policymakers in the last eighty years is the reality that the system has features other than enumeration-based constraints that prevent Congress from overriding most of state legislation.  That being the case, there is little reason to worry about SLID or AASLID.

So: Yes, wall-to-wall preemption would be bad.  So would the massive amount of preemption that Congress could legislate tomorrow, with no changes to current ideas about enumerated powers.  The fact that we don’t find that last possibility terribly worrisome—and we shouldn’t—indicates that it isn’t the constraining force of enumerated powers that gives us reassurance.