We Found the $2 Trillion

By David Dayen
The American Prospect, Jan. 27, 2025

Sometime between the 2024 election and the 2025 inauguration, Americans discovered that they had actually voted for Elon Musk for president. Since the election, Donald Trump has faded into the background, while center stage has been taken by the South African–born billionaire with the Twitter addiction of an adolescent, whose frenetic posts are often being treated like official government statements.

Musk and Vivek Ramaswamy were named co-chairs of the Department of Government Efficiency (DOGE), which sounds like a federal agency (and was actually tucked into an existing White House department) but is actually an outside agitator, tasked with devising recommendations to reduce the size and scope of government. During the campaign, Musk said he was confident he could remove $2 trillion in unnecessary waste from the budget. If he meant annually, that’s nearly one-third of total federal expenditures.

Skepticism is heavily warranted. Presidents going back to Ronald Reagan have impaneled blue-ribbon commissions to hack away at deficits, with minimal success. Nongovernmental advisers carry no formal power, and Congress holds the purse strings tightly. “One person’s waste is another person’s vital congressional jobs program,” said Michael Linden, a former official with the White House Office of Management and Budget.

There are signals that DOGE will be even more useless than its predecessors, with early proposals consisting primarily of things that sound funny but serve valuable functions, or things that are far too minuscule to make a meaningful dent in the budget. Hilariously, Musk gave up on his $2 trillion goal before the Trump presidency even started, calling the number a “best-case outcome” in an interview on his social media site and declining to identify specific cuts. And he also gave up Ramaswamy, who was unceremoniously dumped before the project ever got started.

Many Republicans, however, are dead serious about slashing social spending and obliterating the administrative state. House leaders have been passing around a menu of $5.7 trillion in cuts over ten years, including a large chunk from health care and food assistance for the poor. Musk’s following among the rank and file could provide ballast for these long-held conservative wishes, while catering to his own pocketbook in the process.

How Democrats should deal with DOGE has become a top-level conversation. One school of thought argues for calling out Musk’s game of using spending cuts to make room for their own tax cuts. Others draw red lines on cherished, popular programs like Social Security and Medicare, or highlight serial conflicts of interest. Still others counsel constructive engagement.

But the focus on federal spending could also teach Americans how their government really works. I’ve tallied up the savings from redesigning a handful of policies to improve effectiveness, and you really could find $2 trillion in net annual federal outlays, with no direct impact on the most vulnerable. The key lies in knowing where to look: profit-hungry contractors, privatized boondoggles, systemic overpayments, and a mountain of tax avoidance.

The world’s richest man, himself a serial tax evader and one of the nation’s biggest federal contractors, isn’t likely to touch any of this.

This article should come with a warning label: We should not cancel the equivalent of 7 percent in annual GDP all at once, which would trigger a deep recession. But identifying the real sources of inefficiency in our government—the trillions funneled to elites—can preserve resources for programs to help those in need. And it can display the values of an opposition party that has strayed from its core purpose of fighting for the little guy.

Too often, Democrats have leapt to defend institutions that most Americans look upon with scorn. Now it’s the Republicans who control all branches of government; they are the establishment in every sense of the word. The Musk/DOGE plan is one of self-enrichment and outward punishment. Someone should outline a different path.

DOGE’s Game

It’s hard to critique DOGE because it’s hard to divine its actual intentions. Making government more responsive can sit in tension with cutting spending. Smaller government may be ideologically satisfying but also fantastically wasteful. And when billionaires are making the decisions, self-aggrandizement is sure to follow.

Take the preoccupation with labor costs. Ramaswamy has called for a 75 percent personnel reduction across federal agencies. This would hardly save anything. According to the Congressional Budget Office, there are about 2.3 million federal employees with total compensation in 2023 of $271 billion; that’s 4 percent of the U.S. budget. Federal employees were roughly 4.3 percent of all workers in 1960 and 1.4 percent today. As a result, we’ve seen an explosion in contractors undertaking tasks that government workers used to perform. Nearly three times as much money is spent on contractors than federal workers.

Slashing the federal workforce, almost two-thirds of which is at the Departments of Defense, Veterans Affairs, and Homeland Security, would likely lead to more expensive contractors, and also increase the $247 billion in improper payments the government makes every year. “When you talk about cutting people in the Pentagon, these are people overseeing military contracts,” said economist Dean Baker. “There’s already fraud and there would be a lot more.”

Another chunk of the bureaucracy is devoted to processing federal benefits. Social Security’s administrative costs are legendarily low, down to 0.5 percent. Cutting 75 percent of its staff would delay benefit claims, make checks more difficult to process, and invite fraudulent scams. That’s the opposite of the alleged efficiency goal.

Or take what Public Citizen co-president Rob Weissman calls “policymaking by anecdote.” Musk’s brain-poisoned antipathy to anything that sounds woke led him recently to attack the “fake job” of Ashley Thomas, the U.S. International Development Finance Corporation’s director of climate diversification. But Thomas has nothing to do with diversity, equity, and inclusion; she works with farmers to diversify crops, to deal with changing weather patterns. So is efficiency the goal, or merely reverse political correctness and slipshod word-policing?

At times, DOGE’s goals collide with ignorance over governmental procedures. Ramaswamy claimed that $516 billion could be excised by ending programs that are “unauthorized” by Congress. But in every case, lawmakers have allocated funding and therefore inherently authorized those programs, which include things like health care for nine million veterans. Congress can and should reauthorize and improve programs, but any savings would be far lower than throwing out the entire Veterans Health Administration, the Justice Department, or NASA (where Musk has had $11.8 billion in contracts over the past decade).

The closest thing to an official vision for DOGE was laid out in a Wall Street Journal op-ed with Musk and Ramaswamy’s byline. The only named programs slated for cuts are the Corporation for Public Broadcasting and Planned Parenthood, which total about .012 percent of federal expenditures. And there’s a legally suspect claim that the president can nullify congressionally appropriated spending, which would be an enormous grab at Congress’s power of the purse that would trigger a constitutional crisis.

But the main target of the op-ed is regulations written by “unelected bureaucrats.” (I’m straining to understand who elected Musk or Ramaswamy.) Supreme Court rulings, the DOGE duo claim, prohibit agency discretion on rulemaking, and therefore allow the president to simply pause any regulation that he imagines exceeds congressional authority, and subsequently fire the workers overseeing them.

This isn’t what the Supreme Court said; Loper Bright Enterprises v. Raimondo specifically protected past regulations devised under the older framework of deference to agency interpretation. And a president cannot pause regulations without going through administrative procedure, nor fire employees with civil service protections. But these half-truths do serve the goal of rolling back or curtailing enforcement of regulations, which happens to benefit companies that are habitually under regulatory scrutiny, like, oh, I don’t know, Tesla and SpaceX.

Guided by this vision, Musk will pick the lists of regulations the president can insta-vanish. You can see how the incentives would run. “Every regulation has net positive savings for the country,” said Weissman. “They may cost corporations, they may cost billionaires, but they do not cost the country.”

In this sense, DOGE intends to rewire government for personal use. There has been an unusual interest in a $6.6 billion Department of Energy loan to electric truck manufacturer Rivian. Which electric-vehicle maker do you suppose would like to cancel government support for a rival? Musk has also targeted the bipartisan infrastructure law’s program to build out rural broadband; might he be mad about how his own broadband service, Starlink, was excluded from the grant process, and want to reverse that? Is high-speed rail development truly a “wasteful” program, or an old vendetta from a car manufacturer who opposes mass transit? Do bank regulators have to go because they impede the free flow of capital, or because the Securities and Exchange Commission has been fighting with Musk for seven years?

Maybe it seems too parochial to suggest that the “government efficiency” effort is simply cover for defunding Elon Musk’s regulatory police, steering more contracts his way, depriving rivals of the same treatment he gets, and building oligarchy in America. Maybe Musk is a real efficiency expert who wants to bring his style of business cuts to government as a public service. Maybe he’s truly concerned about burdens being left to his children and grandchildren. Maybe he will nobly share in the sacrifice.

Color me unconvinced. And let me submit as evidence a litany of items that DOGE is likely to leave mostly untouched in its drive for austerity.

Health Care

The federal government is often described as a health insurance company with an army. About 75 percent of all spending is concentrated in four buckets: Social Security payments, health care programs (e.g., Medicare and Medicaid), veterans’ benefits, and the Department of Defense. If you’re not touching them, to cut $2 trillion you’d have to eliminate everything else government does, in total.

But health care in particular is lousy with private-sector profiteering, providing several options for savings.

Nearly 33 million seniors are enrolled in Medicare Advantage, a private insurance substitute for traditional Medicare. It’s heavily advertised to seniors as offering better benefits (including gym memberships and wellness programs) at a lower cost. And it’s true that MA plans typically include dental, hearing, and vision coverage, which is not part of traditional Medicare, as well as reduced premiums.

But Medicare Advantage has been criticized heavily for overbilling the government, not just by liberal activists but also by gold-standard independent auditors. The Medicare Payment Advisory Commission (MedPAC), a congressionally established expert panel, calculated that Medicare Advantage overpayments came to $83 billion in 2024 alone.

There are two reasons for these overpayments. First, Medicare reimbursement is weighted depending on the health of the patient. Insurers are compensated at higher levels for enrolling sicker patients with more diagnostic codes that correspond to ailments. Insurers have exploited this in Medicare Advantage by “upcoding” patients to make them appear sicker, regardless of the actual care they receive. The Wall Street Journal recently found that UnitedHealth, the leading MA plan sponsor and also the largest employer of doctors in the U.S., routinely encouraged its doctors to add codes to their patients.

Physicians for a National Health Program (PNHP), which advocates for a single-payer system, noticed even greater savings potential in the MedPAC report. Traditional Medicare sets a “benchmark” for spending on the average beneficiary. Several studies have shown that MA plans spend between 11 and 14 percent less, because they cherry-pick healthier patients, even after accounting for upcoding to make them look sicker. Increasing denials of care allows MA plans to rake in even more profit.

In all, PNHP found that MA plans charge the government at rates $140 billion per year higher than traditional Medicare. Dr. Ed Weisbart, national board secretary with PNHP, estimated that Congress could use savings from MA overpayments to add an out-of-pocket spending cap, a public drug benefit, and dental, hearing, and vision benefits to traditional Medicare, and have tens of billions left over.

“If you’re serious about DOGE, here’s something you can do,” said Weisbart. “At least let’s agree that Medicare Advantage is being outrageously subsidized.” But with the incoming administrator of Medicare and Medicaid, Dr. Oz, a longtime booster of Medicare Advantage, that’s an unlikely avenue for DOGE.

Reforming physician pay schedules for Medicare could yield more savings. As my colleague Robert Kuttner has written, pay rates are determined mostly by a secretive advisory committee mostly made up of specialists, who give themselves higher reimbursements at the expense of primary care. The government rubber-stamps the recommendations, and private insurers typically use them as a benchmark. Lowering these rates, which incoming Health and Human Services Secretary Robert F. Kennedy Jr. has expressed interest in, would not only slash specialist pay but properly value primary care, reducing health expenditures over time by finding medical problems before they fester.

Dean Baker estimates that bringing U.S. doctor pay (now at $350,000 per year on average) to the level of physicians in Germany or Canada would reduce national health expenditures by around $200 billion per year. Some of that could be done through reductions in the federal pay schedule, but allowing qualified doctors in other countries to practice in the U.S. could also constrain costs through competition. “We have free trade in manufactured goods but we don’t do anything in services,” Baker said. “It would still be a well-paid profession, just not as much as it is now.”

Federal health programs like Medicare (which serves 68 million enrollees) and Medicaid (72 million) account for roughly half of all national health expenditures. So a good estimate for federal savings would be $100 billion per year.

The government also spends massive amounts of money on prescription drugs. In 2022, U.S. drug prices were 178 percent higher than in 33 other industrialized nations, according to a report funded by the Department of Health and Human Services. Some of these drugs are sold at 20 to 30 times the cost of production and distribution; pharmaceutical profit margins are significantly higher than private-sector counterparts.

Democrats did take some action in 2022 by allowing Medicare to negotiate drug prices with manufacturers for the first time; new prices on ten drugs will begin to come online in 2026. But more can be done. “It’s not all drugs right away negotiated to the best price possible. Anyone would look at that deal and say you should get the best deal right away,” said pharma activist Alex Lawson of Social Security Works.

Using federal statutes to seize certain drug patents and distribute them to generic manufacturers that charge less would also save billions. But more structurally, we could overhaul the monopoly patent system that gives drug companies exclusive rights to charge whatever they want for a set period.

Baker has proposed having the government pay for clinical trial research up front, rather than distributing patents to private companies so they can recoup research and development costs. Paying for clinical trials wouldn’t be cheap—maybe $100 billion annually—but the savings realized by free-market prices for drugs without monopoly protection would be considerable: $500 billion per year by Baker’s estimates. Again, some of that would accrue to patients and private health plans; let’s call it $200 billion per year in government savings. The fact that Trump and RFK Jr. literally dined with pharma executives during the transition, however, makes this unlikely to get onto the DOGE list.

There are smaller opportunities. Group purchasing organizations, which help hospitals buy bulk supplies, have been shown to inflate health care prices and cost Medicare and Medicaid $17.3 billion a year, according to a 2010 report; with conservative inflation assumptions, let’s raise that to $20 billion. Medicare and Medicaid made $101 billion in improper payments in 2023, according to the Government Accountability Office, and have been criticized for weak enforcement even when they find health care scofflaws. Moderately better enforcement could yield $10 billion a year. Boosting funding for community health centers, which efficiently fund direct primary care, saves Medicaid $2,371 per enrollee according to one study. Spending $3 billion to get ten million more people care through these clinics would therefore save about $20 billion a year on net.

Of course, moving to a single-payer system wholesale could yield over half a trillion dollars in savings from administrative expenses alone, per the People’s Policy Project. But even if the nation isn’t ready for single-payer, limiting private-sector profit-taking and boosting public provision comes to roughly $490 billion per year.

Military Spending

In 2023, Congress appropriated $841 billion in military spending, nearly equivalent to all nondefense discretionary spending combined. And the Department of Defense (DOD) is not a model of efficiency. In 2024, it failed its seventh consecutive audit, which means it could not account for all its spending in the last fiscal year, or the six preceding ones either. The best the Pentagon could muster in response was a press release claiming it “has turned a corner in its understanding of the depth and breadth of its challenges.” Even Musk had to admit that “DoD gets terrible value for money.”

He ought to know! His companies have $3.6 billion in contracts with the Defense Department, and SpaceX is in talks with a consortium of tech firms seeking to win a greater share of military spending. But this merely doubles down on a prime source of waste in how we finance our military.

“Our budget wouldn’t be justified if the DOD did pass this year’s audit,” said Julia Gledhill, research associate for the National Security Reform Program at the Stimson Center. “Contractors continue to be rewarded for not doing their jobs terribly well.”

One of the best examples of this underperformance is Lockheed Martin’s F-35 Joint Strike Fighter, now slated to cost $2 trillion over its lifespan. As with most new weaponry, the Pentagon invested heavily in producing F-35s before the design was complete, and before thorough testing had been conducted. The F-35 is not optimal for traveling long distances or close-range combat; in 2021, over 800 continuing defects were found on the plane. Yet after 20 years of spending, it was greenlit for full-rate production in 2024, because to do otherwise would be too colossal a waste of prior funds.

Weapons systems that are excessively over budget breach the Nunn-McCurdy Act, triggering reviews of whether to continue the program. But when Northrop Grumman’s Sentinel, the land-based missile part of the nuclear triad, exceeded the Nunn-McCurdy threshold last year, DOD determined that it could go forward, arguing that it was critical to national security. “There haven’t been any significant consequences for that critical breach, and we’re just chugging along,” Gledhill said. “We do have guardrails in place, but corporate interests are so entrenched.”