Last week, I visited Boston College for a discussion of the Supreme Court’s recent decisions. Herewith an abbreviated version of my remarks. Comments etc. most welcome because the thoughts (some old, some new) are embryonic: I’m working on a more serious, grown-up presentation.

We are living in an age of Executive Federalism. That form of government has some deeply disturbing features, including several that should prompt a judicial response. So far, the Court has given no indication that it has a clue.

The federalism we have inherited from the New Deal is commonly called “process federalism.” It does not come from the Constitution; it comes and goes at Congress’s pleasure. Nowadays, though, nothing comes from Congress—except the occasional convoluted statute that hands an unwieldy program to the executive. This drift toward executive government is pervasive. But since so many domestic programs are federalism programs, it has huge federalism implications. And there are reasons why the tendency is particularly acute in the federalism arena.

Localism. Over time the federal government’s ambitions have become more local: land use, education curricula and services, etc. At first impression that looks like an expansion of Congress’s power—but it really isn’t. For one thing these programs have to be implemented by local agents, who have a million ways to shirk. If you want to keep them in line and try to make the programs work you need a branch that’s in business 24/7—the executive. All of these programs effectively dis-empower Congress.

Sectionalism. Political polarization is everywhere. In the federalism dimension it plays out as sectionalism—divisions between cohesive blocs of states, over salient and recurrent issues, that can’t be overcome through compromise (for example, by bribing states into a minimum winning coalition). Congress cannot really handle sectional politics, which explains why national policy on abortion or gay rights has to come from the Supreme Court. Sectionalism now has engulfed economic as well as culture-war issues: labor policy, health care, health insurance, global warming, energy policy, immigration. It’s always the same blocs of states opposing each other (in Congress, in the regulatory process, and in the courts), and the questions are existential. Federalism isn’t between “the states” and the feds; it’s between and among blocs of states—one demanding federal intervention, the other telling the feds to bug off. The executive has to manage the conflicts.

Over the decades the Executive has developed powerful tools to run its federalism. The Obama administration has been very aggressive in deploying them. But the trend is secular, and robust to partisan politics. (It began under the Reagan administration and has accelerated since.) Herewith, six features of Executive Federalism:

Centralization. Executive federalism is highly centralized. The key decisions are made by high-level political appointees, usually in close cooperation with the White House. The basic m.o. isn’t rulemaking by bureaucrats and administrative regularity; it’s political. The trend to “Presidential Administration” cuts across a wide range of issues (think net neutrality); but again it is particularly pronounced in the federalism arena. Deferred action for undocumented aliens wasn’t some bureaucrat’s idea; it was the President’s.

Vast discretion. The prototype of modern legislation isn’t “big delegation,” as under the New Deal. It’s “big waiver”: Congress writes a highly prescriptive programs but then authorizes the Executive to waive big pieces. Medicaid, education, welfare, and other programs work  that way .

Asymmetric federalism. Executive federalism is highly asymmetric: federal program requirements are worked out or waived for individual states, and the differences are huge. Medicaid is an example. There are fifty different Medicaid programs, and they have only one thing in common: not one of them has anything to do with the statute. It’s all done through Section 1115 waivers.

Federalism by Deal. Executive federalism works through dealmaking, not rulemaking. Medicaid’s operation is governed through MOUs (Memoranda of Understanding)—without notice and comment, without anyone at the table except the state bureaucrats and people from HHS and CMS, and without any opportunity for arbitrary and capricious review.

Federalism by Settlement. Over the past decade multi-billion dollar settlements with corporations that are accused of wrongdoing have proliferated. Settlements in the financial sector alone clock in at well over $120 billion. A good chunk of that is off-budget government agency finance. But over one-third is so-called restitution (for example, for mortgage debtors). These payments are privately financed grant and benefit programs, and they are federalism programs. The best example is a 2012 agreement between five big banks, the feds, and states. $5 billion went to a gaggle of federal agencies, state agencies, and “fair housing” groups; $20 billion were earmarked for “restitution” for underwater borrowers—mortgage relief, short sales, etc. Congress considered but rejected a program like that; the administration ran an embryonic (and failed) program with some money it found under a rock; the settlement is the real deal. And it’s federalism: the fund allocation is by state, and the program is run by an ad hoc committee headed by a former banking commissioner and composed of federal and state officials.

Constitutional Hardball. When Congress goes AWOL, the Executive will tend to push its authority to the outer limits, and beyond. Examples spring readily to  mind: immigration; executive waivers for states that have decriminalized marijuana; a clean power plan for which EPA plainly lacks authority; and so on. Some states welcome these escapades, sometimes; others fight back.

Should we worry? A “National Federalism” literature (emanating from the New Haven metropolitan area) says, no. States can shirk, and they can participate in the regulatory process. So all is in good federalism order. I’m not sure. For one thing, Executive Federalism is debt-financed federalism (which helps to explain why we are now extracting the needed funds for the programs and even just running the federal agencies from private entities). We’re now bailing out Puerto Rico. Admittedly that’s not a state but as goes Puerto Rico, so goes Illinois, and in its trail any other state that relies on federal transfers to support a failed business model.

My larger worries are about lawful government. Executive federalism happens in the shadow of the law, and sometimes the bright sunshine of patent illegality. And the federal courts have nothing consequential to say about it.

Historically the Supreme Court has played a formative role in federalism; and even the modern Court prides itself of safeguarding federalism’s “usual balance.” Most of the time, however, courts don’t get to see Executive Federalism at all. Big waivers are usually unreviewable because no one has standing. A few big settlements are reviewed; the vast majority aren’t (not even when the deal is cut, let alone the implementation). MOUs are effectively unreviewable. Even some agency rules have become practically unreviewable. EPA routinely proposes rules that are strictly in terrorem: draconian, arbitrary and capricious, in excess of statutory authority. They’d be struck down in any court. But that will take four years, and utilities and investors and state commissions cannot wait that long. They figure something will happen, and so they move in EPA’s direction and then EPA counts those upfront investments as “benefits.” If a court eventually invalidates the rule, the agency has gotten most of what it wanted. The D.C. Circuit is a paper tiger.

Another example: during the financial crisis Secretary Paulson did not do anything he wouldn’t have done at Goldman Sachs: Buy. Sell. Low-ball them. Be long and be loud. Above all be unpredictable. Even when he had a legal basis for his actions he declined to invoke it because that would suggest he’d do the same thing in the next case, and there goes the leverage. That is now the ordinary operation of our federalism: it’s how the ACA is being implemented, and how the clean power plan is being implemented. There ought to be some law or doctrine to control the basic operation of our government. For the life of me I can’t think of one. If an agency cuts deals with 30 states and then says “no” to the 31^st, is that arbitrary or capricious? Abuse of discretion? Who knows?

Even when courts do get to look at Executive Federalism in action, constitutional constraints are exceedingly weak; and that includes federalism constraints. There is the anti-commandeering rule; but for the most part, the Executive has found ways to commandeer states without running afoul of the rule. There is the made-for-one-case Medicaid holding of NFIB fame; but no one has ever identified another statute to which it might apply. There are various federalism canons and presumptions; but no one has figured out how those should operate in administrative law cases. And that’s the crux. The federalism stuff that the courts do see is Administrative Law stuff; and unfortunately, the Court appears at sea both about AdLaw itself and about its interplay with federalism.

By “the Court” I mean the nominally conservative majority. The four liberals justices have concluded that administrative acts need no legal basis: whatever is not directly prohibited is permitted. (Case cites available upon request; the justices’ dissent in UARG v. EPA is a good start.) On that assumption, administrative law becomes easy, and so does federalism. The conservatives, in contrast, still struggle—and they struggle to resolve even the most basic questions of Executive Federalism.

Consider the “executive” part of the equation. Nowadays, agency rules at the outer boundaries of an ambiguous statute will come from the White House. (Agency “expertise” is the art of making up some numbers.) Does that progeny make a rule more suspect, or less? Should we favor “accountability” (and it doesn’t get more accountable than the President), or should we insist on administrative regularity and protection against politics? The case law provides no clear answer. It does not even present an open acknowledgment of the problem.

Consider the “federalism” piece. The most foundational AdLaw canon is Chevron; the most foundational federalism is the clear statement rule. How do they go together? They come and go, usually without explanation. The foundational clear statement case is Gregory v. Ashcroft (1992). It’s a Chevron case as well, except there’s no mention of it in the opinion. City of Arlington (2013) is a federalism case, and the question of when—as opposed to whether—jurisdictional provisions command deference stares you in the face. The Court brushes that aside: it’s all Chevron metaphysics. If the justices cannot make the foundational canons hang together, they cannot hope to organize this universe.

I conclude with two premises that need examination if one wants to get this right. The first is about AdLaw; the other is about the states.

Regulation, the Courts, and Congress. In the 1970s, Administrative Law was about a “partnership” between Congress, agencies, and the courts. That is no longer so: the Court now thinks it has to ensure regularity and accountability. It’s a check. Is it?

Maybe—except when it matters. The Court thinks it can monitor the government’s mechanics, even as its objects or its legal basis remain beyond judicial purview. Regulation must go on. The leading examples are global warming, and the ACA. (Mind you: both are federalism examples.)

Mass v. EPA (2007) says, EPA must regulate GHGs under the Clean Air Act (not quite, but close). The CAA, everyone knows and the Court knew at the time, isn’t built for greenhouse gases. Sure enough, EPA re-writes the statute, including its numerical thresholds. The Court says, “we will not let EPA embark on this voyage.” (UARG v. EPA). Who is we? The Supreme Court? The D.C Circuit? The courts are supposed to pilot an agency that no longer gives rip about judicial review? Shamefacedly the majority opinion then says, we are giving EPA 83 percent of what it wants. We just have a point about statutory interpretation. Thank you.

This year’s decision in King v. Burwell is yet more astounding.  The IRS—an agency we rarely associate with federalism—makes a decision that shapes the federal structure of the entire ACA. It does this with a rule that looks like something from the 1950s: no legal analysis, no response to comments. It does this in the teeth of a statute that quite plainly seems to say the opposite. The administration’s legal defense is so lame that the Court doesn’t even bother to mention it, let alone rely on it. But the justices waive this thing through.

Had King involved any other statute, the case would have gone the other way. But the show must go on.  The Will of Congress Must Prevail: That’s the rock-bottom premise of modern AdLaw, and of modern “process” federalism. It now conflicts with the reality that there really is no Congress with a discernible will. In the ordinary case, the Court fakes its way through and cites Chevron. And in big cases, when even that acknowledged fiction won’t do, the Court simply makes it up, on terms that lets the system grind on. And it’s not just the ACA or the CAA. The Court’s decisions on the “waters of the United States,” and securities fraud, and civil rights have the same structure: unleash a firestorm, and then pretend that the Court is standing watch. It’s a partnership after all.

Federalism is about “the States,” versus the feds. That’s the Court’s paradigm: you see it in every case. But the Court hasn’t had a case that pits “the states” against “the federal government” in a very long time. Usually the states oppose each other. So who, in those cases (about energy, or health insurance, or antitrust) is the authentic representative of federalism—and who is being opportunistic? Could it be the case that private individuals are the only ones who can be trusted, federalism-wise? You need a theory, which the justices don’t have.

Some legal fictions are good or at least useful. These aren’t: they’re counterfactuals to manage a world that’s dead and gone, and they are no match for an Executive that does as it pleases. To borrow a line from a great thinker: the Supreme Court’s law is Neronian in nature. The Court fiddles, while Rome burns. It is excused by two facts: it does not know that it fiddles, and it does not know that Rome burns.