Is Trump’s DOGE Going to Be a DOGD?

President-elect Donald Trump has proposed a Department of Government Efficiency (DOGE). Trump announced he is appointing Elon Musk and Vivek Ramaswamy to lead DOGE (in reality, a commission to make recommendations), both of whom will volunteer their time. There is no question that over the last half century the federal enterprise has become larded up with excess rules, procedures, regulations, and spending. As such, the second Trump administration has a real chance to implement fundamental reform to make government work again—but it’s unclear whether that is what Trump and Republicans want.

For years, a wing of the Republican Party embraced Global Norquist’s anarchist views of wanting to shrink government down so that it could drown in a bathtub. If that’s the real endgame, radically shrinking government rather than making it efficient, calling this DOGE is a bit like the Democrats calling their massive spending package the Inflation Reduction Act. Slashing government output—as opposed to streamlining government to make it more productive, high quality, and timely—is not about efficiency. Efficiency is about producing the same with less, not less with less. If “slash and burn” is what the incoming administration is seeking, then at least it should be truthful in its advertising and call the new commission DOGD: the Department of Government Downsizing.

Given Musk’s and Ramaswamy’s statements to date, one could be excused for thinking it will be a DOGD. In other words, kabuki theatre for across-the-board government downsizing to a pre-New Deal size of government. For example, On September 5, Musk posted a picture on X (which is now removed) from the greatest movie of all time, Office Space, of “the Bobs,” two consultants brought into Inintech to fire workers, saying “The Department of Government Efficiency will be epic.” In other words, it will be epic to lay off lots of government workers. I hope DOGE is really about “GE.” There is an enormous opportunity to boost efficiency in the federal enterprise, especially through the aggressive use of technology to automate work, but I worry it will bring the government back to the age of Calvin Coolidge.

But before discussing this, let me provide some context. If you asked those in Washington, especially Republicans, about the growth of the federal enterprise, many would say it’s an out-of-control behemoth. However, we need to differentiate between government as an operation and government as a spender. Most of what the federal government does is send money to others, mainly the elderly, the military, and U.S. debt bondholders. And in that respect, yes, it has gotten pretty big.

But what about the size of the government itself: the number of people working there (including contractors) and the cost of employing those people, including buildings, pensions, etc.? As it turns out, it’s not all that big. And relative to the economy, it has been shrinking. Yes, shrinking.

Let’s start with sending money to other components. Musk has claimed he can help Trump cut $2 trillion in federal spending, one-third of the $6.8 trillion budget. Yet, about two-thirds of the federal budget goes towards mandatory programs, such as Medicare and Social Security. Another 23 percent goes to national defense, with another 14 percent to interest. So, wake me up when we are ready to have a real conversation about solving the federal deficit problem since it requires higher taxes and lower entitlements, neither of which the American public wants, or political leaders will even suggest. Everything else is rearranging the deck chairs.

Moreover, realistically, the government is not as bloated as Trump, Musk, and Ramaswamy claim. The federal government workforce is just 1.9 percent of the U.S. workforce, compared to around 3 percent in 1940. (See figure 1.)

Figure 1: Federal employees as a share of total workforce

The cost of these personnel isn’t nearly the bombshell it’s made out to be. In 2023, 7.2 percent ($674.5 billion) of the federal budget was allocated to personnel costs. The Department of Defense, which employs one civilian for every three soldiers, is responsible for most of the personnel spending in the government. The U.S. Postal Service is responsible for about 10 percent.

The growth of the federal government in terms of output should, frankly, not be a concern for policymakers. The output of the federal government has grown by 40 percent since 1997.  But that is less than half of what the overall U.S. economy has grown, 83 percent, over the same period. The U.S. government is shrinking as a share of GDP, not ballooning. Today, the U.S. government represents just 5.5 percent of the country’s gross output, compared to 7.2 percent in 1997. (See figure 2.) Again, remember that social security and other payments are not output; they are transfer payments.

Figure 2: U.S. government output as a share of GDP

Yet, Musk complains that we have way too many federal agencies (428) and should cut them to at most 99. In fairness, 428 does sound like a lot. How could there possibly be that many? Well, maybe Musk should take a look at USA.Gov, which lists every federal department and agency.

At first glance, it seems he’s right; there are a lot. But the lion’s share is pretty small. Take the Joint Fire Science Program, which helps develop firefighting technologies and shares this learning with firefighters nationwide. It has a budget of just $6 million, which equates to 3 minutes of federal debt interest payments. Or the U.S. Botanical Garden in DC, which has just 68 employees. Other agencies don’t cost the taxpayer any money at all. Bankruptcy fees fund the U.S. Trustee program, and borrower fees fund the Export–Import Bank (EXIM Bank), which actually makes money for the government.

Does Elon want to eliminate the Air Force Reserve, Census Bureau, Court of International Trade, Federal Election Commission, Food and Drug Administration, Mine Safety and Health Administration, the Mint, and Voice of America? In other words, it’s not as if a bunch of brainless members of Congress decided it would be fun to create a bunch of useless agencies and departments. Each one was established for a specific purpose.

This isn’t to say that some could not be eliminated, privatized, or merged. Sure, but not 329. Here’s one to put on the chopping block: the Small Business Administration, which supports less productive small companies. I can hear the screaming now from the Republican small business lobby. How about Fannie Mae, Freddie Mac, and Ginnie Mae, letting all housing financing be done by the private sector? And while we’re at it, let’s get rid of the wasteful and unfair mortgage interest deduction. Yeah, I didn’t think so. Or the Farm Credit Administration and all agricultural subsidy payments to farmers, including to sugar producers. I bet red state farmers would love that.

Musk also tweeted that we will “cut wasteful expenditures.” What exactly is wasteful? Spending that does not yield a net present value return above the cost of capital? Spending that provides a safety net for at-risk Americans? Or spending that Republicans just don’t like? Some subsidies clearly are not about a safety net and have likely no rate of return. But don’t call most or all government spending “wasteful” just because you have free market idolatry. The Defense Advanced Research Projects Agency, National Science Foundation, National Institute of Standards and Technology, EXIM Bank, and a host of other agencies don’t waste money; they make critically needed public investments.

Musk also proposed term limits for bureaucrats. Besides the condescending name “bureaucrats” for federal personnel, would Musk or Ramaswamy fire their companies’ mid-level and senior executives after 10 years of service, regardless of performance? Of course not, unless they want to drive their companies into the ground. If you want to force even more good people out of government or dissuade talented young people from joining, term limits for “bureaucrats” would be just the ticket.

Mr. Ramaswamy’s proposals are even more outlandish, at least the ones he made when he ran for president. He proposed eliminating 50 percent of all federal jobs in year one and 75 percent by year two. That would mean no federal agency workers outside the DOD and 24 percent fewer in the DOD by the end of year two.

So, if this is not to be a dodge for DOGD (e.g., radical downsizing of federal output), what should this new enterprise do?  First, it should focus on radical application of IT to automate as much as possible. It should start with the IRS, Medicare, Social Security, State Department passport services, and all the other customer-facing activities of the federal government. To say that these agencies are anywhere close to the private sector’s best-in-class use of IT for efficiency is a vast overstatement. The large majority of government agency websites are terrible. Full stop. The goal should be to eliminate half the customer-facing jobs within a select timeframe through IT while dramatically slashing wait and processing times for citizens within the same period.

Second, create some moon shots. Start with the USPS. Absent a massive federal bailout, the USPS is a deadman walking. And the current leadership of Postmaster DeJoy has only made things worse, especially as he has worked to expand USPS full-time, unionized employment while cutting more efficient work-share programs with the private sector. Much of that work should be contracted out to the private sector when they can show the Postal Regulatory Commission that they can do it more cheaply. Musk and Ramaswamy should reverse DeJoy’s union-supporting job creation and call for the delivery of half the U.S. mail by robots within a decade.

Third, don’t be afraid to try some agency consolidation. Why not start with federal statistical agencies? We have multiple statistical agencies, unlike many nations with just one. Not only is that inefficient, but it makes data collection and use much more difficult.

Fourth, make it easier to fire federal workers. I have a peer who works in a government agency who regularly complains about several of her colleagues who are deadwood but can't be let go because of civil service rules. I used to attend a regular lunch of Federal CIOs, during which many regularly complained about OPM and other rules. They often paid deadwood to leave and go work at another agency just so they didn’t have to waste their time managing them. But, of course, the challenge is to find a happy medium that enables poor performers to be dismissed while not using that flexibility to impose a partisan political agenda.

Finally, eliminate and reform federal regulation. The new commission should start by working with Congress to repeal the National Environmental Policy Act. They should ensure that regulators at the Federal Railway Administration and the Federal Aviation Administration stop delaying more automated freight trains and drone delivery. But they also need to support more permanent reform by creating a unit within the Office of Management and Budget’s (OMB) Office of Information and Regulatory Affairs (OIRA) to systematically consider how proposed agency regulations impact innovation and competitiveness. While OIRA is tasked with reviewing major regulations from a cost-benefit perspective, it does not explicitly review regulations for impacts on longer-term dynamic effects on innovation or competitiveness. The next administration should appoint people to head OMB and OIRA who are committed to this approach and establish a branch within ORIA to serve as an “innovation and competitiveness champion” in the regulatory process. The office should have the authority to push agencies to either affirmatively promote innovation and competitiveness or achieve a particular regulatory objective in a manner least damaging to innovation and competitiveness.

Trump and Ramaswamy must remember that while we need to cut some spending and boost efficiency, the next administration will have to significantly increase public investment if we are not to lose to China. This means more funding for the DOD and more funding for “competitiveness weapons systems,” i.e., funding to support advanced technology firms in the techno-economic war with China.

The Trump administration could signal a turning point for the country and the federal government in many areas. For too long, Washington accepted things that were flawed and broken. The real question is whether this proposed department is about reform or reduction. Will Trump, Musk, and Ramaswamy take advantage of this opening to make the federal government work again, or will they work to drown it in a bathtub? Only time will tell.