What Musk and Ramaswamy Don’t Get

A couple of weeks ago, my kitchen started to smell like sewage. The garbage disposal began to drip. The dishwasher wouldn’t drain. The toilet bubbled when my kids emptied the kitchen sink.

At first, I thought the root problem was the broken disposal. But we fixed that and the kitchen still stank. Maybe it was the dishwasher? No, the dishwasher was fine. Only then did I call a plumber, who immediately knew what the problem was: “Your drainpipe is clogged. We see it all the time.” A guy came over to bust the clog. Problem solved.

I had a mental model of what was wrong in my kitchen, but my mental model turned out to be bad. That’s only natural: I’m a lawyer, and not a handy one. The plumber had a good mental model. He’d seen the same constellation of symptoms before. He made a sharp guess about the root cause, and he knew how to fix it.

I thought about mental models when Elon Musk and Vivek Ramaswamy released an op-ed in The Wall Street Journal making their first major statement about the soon-to-be-created Department of Government Efficiency, or DOGE. In some ways, I liked what I read. I share their conviction that government has become too bureaucratized, rigid, and slow. I agree that radical change is needed.

Musk and Ramaswamy smell the same crap that I do. But I fear they’ve got a bad mental model about what’s wrong. They think we’ve got too many civil servants—their equivalent of the garbage disposal. They aren’t paying nearly enough attention to the clogs in the drainpipe, including the finicky legal and procedural rules that will predictably frustrate their reform efforts. Unless they change what they’re up to, I doubt they’ll make much progress.

In Musk and Ramaswamy’s telling, the chief problem with the administrative state is that it’s full of unelected mandarins who force their diktats down the throats of a reluctant public. Worse, those bureaucrats often act without legal authority from Congress. They adopt regulation upon regulation without regard to their costs, blithely unconcerned about the drag they’re placing on the economy.

If that’s your model, going to war against the bureaucrats makes sense. And that’s what Musk and Ramaswamy aim to do—with, they seem to expect, the help of the Supreme Court. They promise to quickly rescind the regulations adopted by hyperactive officeholders. With fewer rules on the books, they say, there will be less need for bureaucrats to enforce them, providing a justification for “mass head-count reductions across the federal bureaucracy.”

For support, they point to two recent Supreme Court cases, Loper Bright v. Raimondo and West Virginia v. EPA. In the first, the Court overturned the doctrine of Chevron deference. As a result, the federal courts will no longer defer to agencies when those agencies interpret ambiguities in the laws that they administer. In the second, the Supreme Court admonished the EPA after it adopted a creative approach to reducing greenhouse-gas emissions. Only Congress, the Court said, can resolve “major” economic or political questions.

Musk and Ramaswamy believe that “these cases suggest that a plethora of current federal regulations exceed the authority Congress has granted under the law.” They want to target “illicit rules” adopted in the prior, more permissive regime.

Sure, agencies sometimes push the legal envelope. And those cases tend to attract headlines, which is why they shape our thinking about the administrative state. They are the exception, however; agencies have clear legal authority to adopt most of the rules they’ve adopted.

A few days after Loper Bright came down, for example, the National Park Service banned bear baiting under a law telling it “to conserve … the wild life” of the parks. It’s very hard to see anything “illicit” about such a routine exercise of delegated power. The U.S. Code is full of delegations like this because Congress doesn’t have the bandwidth or the expertise to establish every detail of government programs. Instead, it writes general laws and instructs agencies to fill in the specifics. And agencies still get deference, even under Loper Bright, when they act within “the boundaries of th[eir] delegated authority.”

That describes most of what agencies do. Last month, for example, the IRS adopted new rules about tax credits for new semiconductor facilities under the CHIPS Act. Why? Because Congress told the IRS to “issue such regulations” as needed to carry out the law. Last Wednesday, the National Marine Fisheries Service adopted a rule limiting the sardine catch in fisheries off the Pacific Coast. The agency had similarly unimpeachable legal authority to do so.

Don’t get me wrong, there are lots of dumb rules. It’s just that most of those rules are squarely within their agency’s remit. Although that doesn’t make them any less dumb, it does mean that pointing to Loper Bright and West Virginia v. EPA won’t help get rid of them.

What’s more, Musk and Ramaswamy get it backwards when they say that the cases give them extra powers to undo existing regulations. In fact, the cases constrain their authority. Imagine that an agency, for example, has an old rule on the books that is based on an interpretation of the law that DOGE dislikes. Before Loper Bright, the agency could have changed that interpretation so long as the new interpretation was also “reasonable.” The agency was free, in other words, to toggle between different ways of reading an ambiguous law.

After Loper Bright, toggling is verboten. An agency that has already adopted the soundest interpretation of a law can’t change its mind. It’s stuck. If the agency were to try to adopt a new reading of the law—perhaps one that DOGE prefers—and to use that to justify rescinding the rule, the courts would stop the agency. Saying that Loper Bright gives DOGE flexibility is about as sensible as saying that handcuffs help when throwing a baseball.

Much as DOGE might wish it were otherwise, rescinding a rule requires agencies to go through a cumbersome, multiyear rule-making process. Working to streamline that process would be a terrific mission for DOGE, and I hope Musk and Ramaswamy pursue it. Instead, they said that they’ll take a shortcut: The president will simply “pause the enforcement of those regulations” while they’re being reviewed.

A unilateral pause won’t be as helpful as Musk and Ramaswamy seem to think. Many businesses, especially big businesses, have to certify their legal compliance to government agencies—most notably via financial reports to the Securities and Exchange Commission, where false certifications can trigger criminal penalties under Sarbanes-Oxley. Few will feel comfortable ignoring rules that are still on the books just because DOGE tells them they might someday be rescinded.

What’s more, you need smart bureaucrats to make sure that rescissions hold up in court. Under settled law, established way back in the Reagan administration, “an agency changing its course by rescinding a rule is obligated to supply a reasoned analysis for the change.” Compiling that analysis requires technical skills that agency bureaucrats will have and that DOGE will lack. Slashing the federal workforce will thus work at cross-purposes to deregulation.

Blanket nonenforcement is also, well, not so legal. As the federal courts have said, “An agency’s pronouncement of a broad policy against enforcement poses special risks that it has consciously and expressly adopted a general policy that is so extreme as to amount to an abdication of its statutory responsibilities.” When that happens, the courts—with review up to the Supreme Court—are likely to intervene.

Will they? Musk and Ramaswamy seem to think that the Supreme Court will be a foot soldier in the war they hope to wage on the bureaucrats. Again, I think their mental model is off. Six of the justices are conservative, that’s true. It’s also true that those justices have reservations about the size and scope of the modern administrative state. In general, though, the Supreme Court’s preferred approach to keeping agencies in check has been to insist on procedural fastidiousness.

That’s not a passing fancy or a political spasm. It’s a cornerstone of the conservative legal movement, which is committed to the view that the courts must stand as a bulwark between the excesses of federal agencies and the public. The justices may like Donald Trump’s policies, and they may go somewhat gentler on those policies than they did on Joe Biden’s. They are unlikely, however, to abandon their commitment to the scrupulous enforcement of procedural rules to cater to his whims.

Wishful thinking also characterizes Musk and Ramaswamy’s approach to cost cutting. In their op-ed, they seem to appreciate that more than half of all government spending comes from entitlement programs, including Medicare and Medicaid. They likewise admit that only Congress—not DOGE—can shrink those programs. Nevertheless, they think they can make substantial headway because of “the sheer magnitude of waste, fraud and abuse that nearly all taxpayers wish to end.”

They’re right about the waste, fraud, and abuse. I hope they tackle it. The trouble is that, in their apparent mental model, making those cuts will be easy, because no one likes waste, fraud, and abuse. On that, they’re mistaken.

For a characteristic example, look at how Medicare pays for drugs. When physicians dispense (say) a chemotherapy drug in an outpatient facility, they’re allowed by law to bill the government for 106 percent of the drug’s average sales price. Because 6 percent of a big number is more than 6 percent of a small number, physicians have a huge financial incentive to prescribe the most expensive drugs, even when a cheaper and equally effective alternative is available. That helps explain why Medicare drug spending has exploded, growing an average of 9.2 percent a year from 2008 through 2021.

I’d call that wasteful. Would oncologists? Back in the Obama administration, Medicare proposed changing the statutory formula to make it mildly more sensible. Unusually, Medicare has broad legal authority, conferred in the Affordable Care Act, to do so by legal fiat. Nonetheless, the attempt fell victim to an intense lobbying campaign from hospitals and doctors.

The point generalizes. The health economist Uwe Reinhardt called it the Cosmic Law of Health Care: “Every dollar of health spending is someone else’s health-care income, including fraud, waste and abuse.” If you really wanted to cut federal spending, you wouldn’t declare war on bureaucrats. You’d declare war on hospitals and physicians. Does DOGE have the stomach for that?

If not, Musk and Ramaswamy’s claim that they will reduce government spending by “impoundment” won’t come to much. Their idea is that Trump could simply refuse to spend some of the billions of dollars that Congress has appropriated. A Nixon-era law called the Impoundment Control Act prohibits the president from doing so; Musk and Ramaswamy insist that the law is unconstitutional—and that the Supreme Court would agree.

I wouldn’t be so sure. Trump’s allies have tried to build the legal case to support this constitutional argument, but it’s both untested and unpersuasive. Congress’s powers are at their zenith when it comes to federal spending on domestic programs. As early as 1838, the Supreme Court rejected the claim that the president can refuse to spend money as Congress has directed: “To contend that the obligation imposed on the President to see the laws faithfully executed, implies a power to forbid their execution, is a novel construction of the constitution, and entirely inadmissible.”

In any event, if all of this spending is so unpopular, why not pass a law to cut it? Republicans will hold the White House and both chambers of Congress come January 20, and the Senate filibuster is no impediment to spending cuts. Yet going to Congress is not Musk and Ramaswamy’s style. They want to “driv[e] change through executive action based on existing legislation rather than by passing new laws.”

That dismissive attitude toward Congress betrays the limited scope for DOGE’s reforms. It also replicates the problem they say they want to fix—that unelected people (like, ahem, Musk and Ramaswamy) are making law instead of elected officials. If they really believed what they’re saying—that “our nation was founded on the basic idea that the people we elect run the government”—Congress would be at the center of their plans, not an afterthought.

DOGE may make progress on selected problems, of course. Musk and Ramaswamy are dead right, for example, that the civil service is in desperate need of fixing, and that Trump has an unusual degree of freedom to rethink it. The procedures that apply to federal hiring are Kafkaesque, and firing civil servants is next to impossible.

Even there, however, Musk and Ramaswamy seem to care only about the firing part, and not about the hiring. That’s a problem. As the economist Tyler Cowen has written about the administrative state, “dismantling it, or paring it back significantly, would require a lot of state capacity—that is, state competence.” If Musk and Ramaswamy have ideas about how to bring the best and the brightest into government, they’re not sharing them.

Maybe Musk and Ramaswamy can pivot. Maybe they will be more creative, daring, and capable than I expect. For now, however, it looks to me like they are coming at the problem with the wrong mental model and a half-baked belief that they can achieve change through sheer force of will. I admire the ambition, and I share their concern about government dysfunction. But I fear they have no clue how to fix it.