In the world of finance, a write off is a polite way to describe a massive investment that incinerated capital with nothing left to show for it but a tax deduction and a lesson learned. In Washington DC, however, it appears that they call it the Department of Government Efficiency. Elon Musk finally offered a candid post mortem on his brief but chaotic tenure as the spiritual leader of the
department. You may recall that he wasn't the legal head, as former leadership would have required a Senate confirmed appointment. He spilled the beans earlier this week on a podcast hosted by Katie Miller, a name that might ring a bell if you followed the soap opera of Washington staffing. Katie Miller is the wife of Stephen Miller, Donald Trump's deputy chief of staff. She's also a former Trump administration official who served as the spokesperson for DOGE between January and May of
this year. She left that role to work directly for Musk in the private sector before launching her podcast. The interview was a friendly affair, studiously avoiding the awkward elephant in the room. A viral incident last June where a fake screenshot circulated online showing Elon Musk bragging that he had stolen Katie Miller from her husband. Musk's truth seeking AI assistant Grok analysed the screenshot and declared it was likely genuine. Because the metrics and content
aligned with Musk's behaviour. Musk had to publicly correct his own invention, which, as it turns out, was a perfect metaphor for the entire Doge project, an automated confirmation of a reality that didn't exist. For a project that ended with a whimper, the bang at the beginning was truly deafening. During a televised rally at Madison Square Garden last year, Howard Lutnick invited Musk on stage to ask him a simple question. All right. But we set up Doge, yes?
How much do we you think we can RIP out of this wasted $6.5 trillion Harris Biden budget? Well, I I think we can do at least 2 trillion. Yeah, yes, 2 trillion. It was a staggering figure, nearly a third of all federal spending. It was also, according to Lutnick, himself, completely off script. Lutnick later admitted to the Financial Times that the deal was that he was going to say he would cut $1 trillion, adding somewhat helplessly. What was I supposed to say?
Apparently, in the heat of the moment, Musk decided to simply double the already ambitious promise. This tendency to amplify reality was an ongoing feature of the operation. Donald Trump famously described the initiative as the Manhattan Project of our time, a comparison that implied the department would be building an atomic bomb to drop on the federal bureaucracy. Now, that comparison was apartment, though perhaps not in
the manner intended. Like the Manhattan Project, DOGE was shrouded in secrecy and driven by a sense of existential urgency. But unlike the physicists at Los Alamos who successfully split the atom, the workers at DOGE discovered that the laws of statutory spending could not be rewritten on a whiteboard. Within months, the target of $2 trillion was quietly walked back to 1 trillion. By April, Musk was intimating that the savings might be closer
to 150 billion. The Manhattan Project was rapidly turning into a bottle rocket. Now, before we dig into the specific reasons for the failure of Doge, we should acknowledge that the premise itself was not unsound, as rooting out fraud, waste and abuse is 1 political 'cause that unites almost everyone. Whether you're a libertarian who wants to starve the beast or a socialist who wants a welfare beast to do more heavy lifting, you want your tax dollars to be spent efficiently.
If there is waste, we should all be delighted to see it end. Reducing it with theoretically either either lower our tax bills or improve the quality of government services. We would, in short, get more for our money.
Even if we put aside deliberate malfeasance, It's not crazy to think that a businessman with a background in technology who has built out huge manufacturing facilities around the world might be able to upgrade antiquated government systems, automate manual processes, and introduce rigorous metrics to measure performance. Efficiency is a noble goal. However, a worrying sign that this particular efficiency Dr. might not live up to The theory appeared early on, and it came
from Musk himself. Shortly after the department was announced, he posted a meme on his Everything app depicting himself and his Co head Vivek as the Bobs, the two consultants from the cult classic film Office Space. For those who haven't seen the movie, The Bobs are brought put into a poorly managed software company called Initech to streamline operations. They are the embodiment of
sterile corporate absurdity. They speak in buzzwords, they've no understanding of what the employees actually do, and they ultimately fire the wrong people like the competent software engineers while promoting the incompetent ones who simply know how to play the corporate game. Comparing your new government department to a pair of fictional consultants famous for being useless antagonists was certainly a choice, and as it turns out, it was a prophetic one.
But to be fair to Musk, he did manage one genuine efficiency gain right out of the gate. He must have realized that a department dedicated to eliminating redundancy shouldn't start with two CE OS, and so he managed to shake off Vivek almost immediately. The first crack in the efficiency narrative appeared early on on on February 12th, 2025, to be precise. That was the day The Economist published the piece with the blunt headline Elon Musk is failing to cut American spending.
While Doge was posting viral videos of staff counting beans, literally because posting memes is what government officials do nowadays and promising a revolution, the magazine's data team was doing something far less exciting but far more revealing. They were reading the Daily Treasury Statements, or DTS. The DTS is effectively the federal government's daily checking account statement.
Every working day, the Treasury publishes a detailed list of cash withdrawals from its primary deposit account. It doesn't care about press releases, ceiling values, or projected savings. It only cares about cash out the door. The Economist found that in the first three weeks of the new administration, despite the announce instruments of massive savings, federal outlays averaged $30 billion a day. When compared to the same period the prior year. Spending hadn't fallen.
It had actually risen. Naturally, the excuse factory went into OverDrive. Defenders of the department argued that government spending is like a super tanker. It takes a lot of time to turn. They claimed that the savings would appear with a lag once the zombie contracts were fully terminated and the fired employees were off the payroll. And I don't really disagree with that argument.
It was quite plausible. But by May, the Financial Times ran the numbers again and found the exact same thing. U.S. Treasury data, they wrote, has so far shown no drop in government spending. Now that we're a year into this experiment, I've updated the analysis myself. I looked at the daily withdrawals for the full fiscal year, and the results are well, let's just say the Super tanker didn't just fail to turn, it
seems to have accelerated. If the Treasury's bank account was showing no savings, what had happened to the billions Elon Musk claimed he had cut? The answer could be backed out from the wall of receipts that Musk had posted on the department's website. This was supposed to be a testament to radical transparency, a live accounting for every dollar cut from the budget.
But when auditors began to go through the receipts, they discovered that the wall was load bearing only in the sense that it was holding up a fantasy. Investigations by Politico and Futurism revealed that Doge was using a Met district, known in government procurement as ceiling value, to calculate its wins. When they cancelled a government contract, they didn't count the amount of money actually being spent.
They counted the maximum theoretical amount that could have been spent over the lifetime of the deal. One of the more egregious examples was a shelter for migrant children. DOGE claimed a savings of $2.9 billion for cancelling this contract. In reality, the shelter had been empty for a year year. The agency was only paying a small retainer to keep the facility on standby. The actual cash savings from cancelling the contract was only about 4% of the headline figure.
This wasn't an isolated error, it was the entire DOGE methodology. Further investigation revealed that roughly $4 billion of the claim savings came from indefinite delivery vehicles, which are flexible framework contracts that allow agencies to order services if and when they
need them. As procurement law expert Jessica Tilotman noted, claiming these as savings is the equivalent of basically taking out a credit card with a $20,000 credit limit, cancelling it, and then saying I've just saved $20,000. You obviously haven't saved a dime. You've simply destroyed your line of credit. In the end, the department itself vanished just as silently as the savings it claimed to
have found. In November 2025, with eight months still left on its official mandate, Reuters contacted the Office of Personnel Management to check on the status of the DOGE initiative. The OPM director gave the kind of brutally efficient answer that Musk himself might have appreciated. That doesn't exist. There was no press conference, no final report, and no mission accomplished.
Banner. The Manhattan Project of our time had simply become a non centralized entity, its functions quietly absorbed back into the very HR departments it was meant to destroy. The department that was supposed to revolutionize the American state had been quietly disbanded, and the Super high IQ small government revolutionaries that Musk had recruited on his Everything app were either let go or moved into the national design studio to work on
beautifying government websites. It turned out that the only thing Doge successfully deleted was itself. If the accounting was creative, the personnel management was catastrophic. Musk's stated strategy was to inject a start up culture into the federal workforce. He recruited a cadre of engineers and venture capitalists to disrupt the status quo.
Leading this charge was a 19 year old technologist named Edward Karestein who commanded respect in the halls of Washington under the Nom de gar Big Balls. Mr. Balls, as we must assume he was addressed in meetings, was not your typical government employee. His previous qualifications included being fired from a cybersecurity internship for leaking secrets and reportedly providing tech support to a cyber crime gang that bragged
about harassing FBI agents. Naturally, Doge decided that he was the perfect candidate to parachute into the Cybersecurity and Infrastructure Security Agency and the Social Security Administration. The epicenter of Doge's disruption, however, was the IRS. Doge's approach here perfectly illustrates the difference between cutting costs and destroying value.
Early in the administration, Musk installed his own acting commissioners, leading to a revolving door of leadership where the agency averaged a new boss every month. The chaos culminated in a shouting match in The West Wing between Musk and Treasury Secretary Scott Percent over the firing of a mega aligned whistle blower who had been installed at the agency. But the real damage was operational. Doge pushed for the firing of tax cops to save on salaries. The result?
A projected $64 billion loss in tax revenue over the next decade because there were fewer auditors to catch tax cheats. In economics, as in life, you often have to spend money to make money, and indiscriminately firing the people who collect the bills is rarely a path to profitability. The operational reality on the ground was grim.
The Economist reported that the IRS cafeteria in Austin had to be used to store mountains of unprocessed paper tax returns because the staff required to digitize them had been let go or demoralized. The agency's widely praised Direct File tool, which had allowed Americans to file taxes for free, was put on the chopping block. While the chaos at the IRS hit the government's wallet, the dismantling of USA targeted its global reach.
This was the one area where Doge truly succeeded in slashing a department, feeding it, in Musk's own words, into the wood chipper. Because US aid was created by Congress, Doge lacked the legal authority to abolish it. Instead, they used executive power to fire its staff and cancel its contracts, effectively closing it by Fiat. Interestingly, while this move horrified foreign policy experts, it was a political winner at home.
The Financial Times reported that the gutting of foreign aid enjoyed broad support among voters, many of whom were reacting to viral claims from mosque that the agency was sending $50 million in condoms to Hamas. As it turns out, the efficiency experts at DOGE had confused the Gaza Strip with the Gaza province in Mozambique, where the US funds HIV prevention programs. USAID had sent 0 condoms to the Gaza Strip, but the correction came too late and the agency was already in the wood chip.
The geopolitical cost associated with these savings may be steep. Bill Gates told the FT that the picture of the world's richest man killing the world's poorest children is not a pretty one. Even Marco Rubio, the Secretary of State, testified to Congress that the death toll from these abrupt cuts was surely not 0. The cancellations were so haphazard that Dodges own website listed the details of over 3000 cancelled US aid contracts as simply unavailable for legal reasons.
DOGE treated the agency less like a government department in need of reform and more like a distressed asset to be liquidated overnight. And while it may have saved roughly $45 billion on paper, or just 0.6% of federal spending, it could be argued that it cost the US significantly in soft power and global influence. By May, the administrative collapse of Dodge had bled into a bitter personal feud.
The relationship between Trump and Musk disintegrated into a public brawl that would have been shocking if it weren't so perfectly on brand for both men. The end came with a visual metaphor so heavy-handed a scriptwriter would have rejected it. At his farewell event in the Oval Office on May 30th, Musk appeared sporting a visible black eye. When pressed by reporters, he claimed that his five year old son serial number had punched him in the face while they were horsing around.
Now Musk was famous for bringing his son to the White House. The child was a fixture at high level meetings, often seen hiding behind the Resolute Desk while his father discussed gutting the civil service. But in hindsight, bringing a toddler to work every day may have been unwise given what Musk apparently believed about his bombs. Just days after leaving the White House, the bromance evaporated. Musk took to the Everything app to call Trump's new spending bill a disgusting abomination.
When Trump retaliated by threatening to cut Musk's government subsidies, Musk went nuclear. On June the 5th, he tweeted. Time to drop the really big bomb. Real Donald Trump is in the Epstein files. That's the real reason they haven't been made public. Have a nice day, DJT. He deleted the tweet shortly after, but the damage was done. The man who had spent months as the president's right hand had effectively accused him of shielding a sex trafficking
ring. It was a chaotic end to a chaotic tenure, leaving Musk with a black eye that was both literal and reputational. Before we critique the execution, we should clarify exactly what Doge was, or rather, what it wasn't. Despite the official sounding name and the presidential seal on the podium, the Department of Government Efficiency was not a real government department. Only Congress can create new departments, and Congress did no such thing legally. Doge was an advisory committee.
This concept is nothing new. In 19 O 5, Theodore Roosevelt established the Keep Commission to reform the executive branch. In 1984, Ronald Reagan appointed the industrialist J Peter Grace to lead the Grace Commission, which famously concluded that 1/3 of all income taxes were consumed by waste. Historically, these commissions follow a polite, predictable script. A group of business leaders spends a year writing a thick report.
The president thanks them for their service, and the report is promptly filed in a basement, never to the be seen again. Musk, however, had little interest in writing reports. While the legal setup suggested he was merely A consultant offering advice, the reality on the ground was very different. Consultants usually produce PowerPoint decks. Musk produce termination notices. Through a combination of executive orders and sheer administrative bullying, Musk's team bypassed the usual
separation of powers. They didn't just suggest that the IRS should be leaner, they fired the staff. They didn't just recommend that USA be reformed, they cancelled its contracts and locked the doors. In the past, the business approach to government meant bringing in ACEO to just look at the books. This time it meant bringing in a corporate Raider to strip the assets. Before we mourn the premature end of this experiment, we should probably ask if the
mission was ever possible. To start with, the claim that DOGE could cut $2 trillion from the federal budget was not just ambitious, it was, to use a technical economic term, ridiculous. As we discussed in a prior video on this topic, the US federal budget is not a discretionary
slush fund. Roughly 65% of all spending is mandatory, meaning that it is. Locked in by law for Social Security, Medicare and Medicaid. Another 10% is interest on the national debt, which has to be paid unless the US wants to default and trigger a global financial crisis.
That leaves just 25% for discretionary spending, and half of that goes to the Pentagon. Since the Trump administration had vowed to increase defense spending, Musk was effectively promising to cut $2 trillion out of a non defense discretionary budget that only contained about $900 billion to begin with.
He could have fired every park Ranger, air traffic controller, and FBI agent, shut down the Department of Justice, and turned off the lights in the White House, and he still would have missed his target by a trillion dollars.
While Dodge does not appear to have made any significant savings for taxpayers, they may have still made perfect sense for Elon Musk. Critics warned from the start about the obvious conflict of interest inherent in appointing a major government contractor to oversee government spending. Senator Richard Blumenthal noted that Musk's companies were facing at least $2.37 billion in potential fines and investigations from the very federal agencies DOGE was
targeting. DOGE pushed for deep cuts at the Federal Aviation Administration and the National Highway Traffic Safety Administration. These are the specific regulators responsible for overseeing SpaceX rocket launches and Tesla's self driving software. By cutting the budgets of these government watchdogs, Musk effectively removed the oversight that had been slowing down his own product testing. Then there's the matter of his
other ventures. While Musk was slashing the budgets of rival government contractors, questions arose about how his own satellite Internet company, Starlink, seemed to navigate the chaos unscathed. Critics noted that while legacy providers faced the wood chipper, Musk's companies were positioned to pick up the pieces, with his team pushing to integrate private AI solutions, presumably modeled on his own technology, into the heart of
the federal stack. It appears that in the rush to delete government waste, the department also managed to delete a significant amount of regulatory risk for its spiritual leader. Underlying the entire DOGE project was the idea that the US government is a uniquely swollen leviathan. But the data suggests otherwise. According to the International Labor Organization, the US employs roughly 64 public sector workers per thousand people.
This is hardly gluttonous compared to Australia's 143 per 1000 or the public sectors in the UK and Scandinavia. Interestingly, the US federal workforce actually shrank by 8% between 2008 and 2023 relative to the population. As the World Bank notes, government efficiency often correlates with better trained bureaucracies, not necessarily smaller ones. Singapore runs a very lean state not by firing experts, but by by paying them well to ensure
competence. Doji's strategy of mass firings didn't move the US closer to the Singaporean model, it moved it closer to the Haitian 1 Possibly the most foolish thing that DOGE did was to send mass emails to staff offering paid redundancy. Such an offer will be accepted by the best workers who could easily find work elsewhere, leaving the government staffed by the people who have no other outside opportunities.
Ultimately, the size of a government is a reflection of what its citizens expect from it. Different cultures have different views on this. In 1691, the English government raised less than 2% of GDP in taxes. That sounds appealing until you remember that the government also didn't manage a nuclear arsenal, a National Highway system, or air traffic control. As societies modernize, they tend to demand more from the
state. It's entirely reasonable for a country to decide it wants a smaller government with a lighter tax burden, but that debate should be conducted honestly. If the goal is to reduce government spending significantly, politicians need to admit that this requires cutting services, benefits and protections. Pretending that the difference can be made-up solely by eliminating waste, fraud and abuse without touching the services people actually use is
a fantasy. The real question for taxpayers is not simply how big the government is, but whether they feel they're getting value for their money or not. As I mentioned early on in the video, I updated the research done by The Economist in February, and when the fiscal year closed on September 30th, the scoreboard was unambiguous.
Total federal outlays for 2025 reached $7.01 trillion, a 4% increase in government spending from the 6.75 trillion spent in 2024. Despite all the chainsaw waving, the wall of receipts, and the Manhattan Project rhetoric, the US government had managed to spend more money this year than last. To the untrained eye, the start of fiscal year 2026 might have looked like a real breakthrough as federal spending collapsed in October. This, however, had nothing to do with DOGE.
The decline was caused by a government shutdown that froze operations once the lights came back on. The spending didn't just return to normal, it surged as agencies paid back pay to furloughed workers and process delayed contracts. The November 2025 deficit spiked to $366 billion. And not the only person who did
this sort of research. The Hamilton Project at the Brookings Institution built a daily tracker to monitor these flows, and their data confirms that there was no durable reduction in aggregate outlays attributable to Doge's actions. The Committee for a Responsible Federal Budget went further, questioning whether the entire project was a distraction rather than an effective focus as spending continued to climb materially. Perhaps the most damning is the
calculation of net cost. CBS Money Watch reported an estimate from the Partnership for Public Service suggesting that Doge's disruptions, including paid leave for suspended workers, the cost of rehiring during wrongfully terminated staff, and lost productivity, may have actually cost taxpayers roughly $135 billion. In the end, the Efficiency Dr. didn't just fail to save $2 trillion, it may have finished deeply in the red.
In his interview with Katie Miller, Musk attempted to frame the venture as having been somewhat successful, claiming that Dodge had identified $200 billion in zombie payments. But the Treasury's Ledger paints a very different picture. If DOGE truly stopped $200 billion of waste, those savings were immediately consumed and exceeded by new spending.
To call a project that presided over a net increase in federal outlays somewhat successful is a bit like the captain of the Titanic describing his voyage as a partial success because the band played beautifully until the end. Perhaps the most revealing moment of the podcast came when Musk let his guard down. When asked if he would do it all over again, he replied. I don't think so, he added, almost wistfully. Instead of doing Doge, I would have basically worked on my companies.
While Musk was playing shadow president and getting the president to pretend to like his cars on the White House lawn, everything's computer. Tesla stock was getting pummelled as the company ceded its crown as the world's largest EV maker to Chinese rival BYD. The Reuters report on Doge's disbandment noted that while most federal agencies faced strict headcount caps, exceptions were carved out for positions necessary to enforce immigration laws and protect
public safety. This new spending, which does nothing to increase growth, is particularly painful when viewed against the backdrop of the national debt. As we've discussed in prior videos, the US national debt has expanded massively and is growing as a percentage of GDP. When the government runs an unbalanced budget, spending $7 trillion while collecting far less in taxes, it has to borrow. The difference that borrowing comes out of cost.
Interest payments on the national debt have surged to nearly $1 trillion per year. This interest expense is dead money. It doesn't buy aircraft carriers, pave roads, or provide healthcare. It's simply services. The past. By failing to cut the deficit and instead ramping up spending, the efficiency Dr. has ironically accelerated the one expense that no amount of efficiency can reduce the compounding interest on our own legacy.
In the absence of spending cuts, the administration has pivoted to its favorite revenue raising tool, tariffs. The government collected $195 billion in custom duties in full year 2025, a massive increase from the previous year. Supporters claim that this windfall will pay down the national debt. The problem is that even with this historic surge, tariff revenue covers less than 3% of total federal spending. And worse still, the revenue
stream may be temporary. The Supreme Court is currently hearing a challenge to the constitutionality of using emergency powers to impose these sweeping tariffs. During oral arguments in November, several justices appeared skeptical of the president's authority to bypass Congress on trade taxes. If the court rules against the administration, the Treasury could be forced to refund up to 168 billion dollars to businesses.
This would blow an even larger hole in the budget, leaving the government with higher spending, lower revenue, and a Manhattan Project that produced nothing but a puff of smoke. Perhaps the most frustrating aspect of the Doge failure is that a mechanism to cut waste and fraud already exists. And unlike Elon's chainsaw, it actually works. the United States has a long standing statute called the False Claims Act, which dates back to the Civil War.
It allows private citizens, known as whistleblowers, to sue companies or individuals defrauding the government and crucially, to keep a portion of the recovered funds. This is the eat what you kill approach to regulation, and it appeals to free market principles far more than a centralized politburo of efficiencies. Ours. It aligns incentive perfectly. The more fraud you find, the more you get paid. Last year alone, this system quietly recovered huge sums.
The SEC and IRS paid out nearly $379,000,000 to whistleblowers who flagged financial misconduct. This decentralized approach doesn't require big balls to insult civil servants on Twitter. It just requires a lawyer and a spreadsheet. But unlike DOGE, it's unglamorous legal and slow qualities that don't play well at a campaign rally. There is possibly a broader cultural lesson to be learned here. For the better part of a decade, markets and politics have been dominated by laws.
The idea that irony, memes, and sheer audacity could override fundamentals. We saw it in Dogecoin, the cryptocurrency. We saw it in the meme stock mania, and we saw it in the creation of a government department named after a Shiba Inu meme. I think the finest name is is DOG the the Doge Department of Government efficiency. As the Financial Times columnist Jemima Kelly noted recently, the joke just isn't funny anymore. When the laws meet the bond market, the bond market usually
wins. The 10 year treasury yield does not care about your meme game, it cares about the debt to GDP ratio and that ratio is not improving. In his final days, as the shadow efficiencies are, Musk was asked if the project could survive without him. He replied with a cone. Is Buddha needed for Buddhism? It is a lofty comparison, but perhaps a more fitting question for the bond vigilantes watching the US debt pile grow. Is this is a calculator needed for accounting?
The answer, it turns out, is yes. And for now, the calculator says that despite all the memes, the US government is still spending $30 billion a day. Thanks for tuning into this week's podcast, with special thanks to our supporters on Patreon whose funding makes this all happen. Have a great week and talk to you again soon. Bye.