In a 1999 Hoover Institution interview, economist Milton Friedman was asked which federal agencies he would abolish. As host Peter Robinson rattled off the Cabinet list, Friedman gave a blunt verdict on most: "Abolish." Departments of Agriculture and Commerce? "Abolish." Education and Energy? "Abolish." Housing and Urban Development? Gone. Labor? Gone. Transportation? Gone. Even Veterans Affairs, he argued, could eventually be eliminated (with veterans compensated in other ways).
By the end of this exercise, Friedman had effectively reduced 14 Cabinet departments down to about 4.5. The only agencies he'd clearly keep were those handling essential duties like defense, justice, foreign affairs, and treasury functions—the minimal state required to protect the nation and uphold the law.
Fast forward to 2025. The Trump administration too wants to slash federal agencies. The headline-grabbing effort is led by the new Department of Government Efficiency (DOGE), helmed by billionaire Elon Musk and directed by a large number of executive orders. In a few weeks, DOGE has been caught in a whirlwind of activity, with sweeping claims of slashing and cutting, though with little clarity on the forces driving these abrupt shifts.
To be sure, the DOGE team is raising the profile of the need to cut spending and claims it has "cut" billions of dollars in grants and interrupted payments for many contracts. Musk, like Friedman, has talked about the need to cut entire agencies. They have already started taking knives to a few of them. In fact, several of their targets have long been on libertarians' abolish lists.
Yet I can't shake the feeling that the administration may not be playing the long game. It isn't because the cuts will be minuscule compared to what truly needs to be cut. Neither is it because the DOGE team is causing chaos. Any serious effort to cut down the size of government is bound to be chaotic because it throws a wrench into the usual functioning of government and pushes against the desires of many special interests.
My trepidation boils down to two things. First, for all the talk about cutting government waste and fraud, the DOGE-Trump team seems mostly animated by rooting out leftist culture politics and its practitioners in Washington. It feels that it is less about smaller government than it is about political transformation. While the two intersect, this strategy could fall short.
That's in part—and this is my second point—because for those of us who care about permanently downsizing government and keeping it bound by constitutional rules to prevent the exercise of arbitrary power, DOGE is mixed. While there is a small probability the approach will succeed in reining in spending or the administrative state, it will be at the heavy cost of reinforcing the power of the executive branch and opening the door to the same abuse when the left is in power.
The probability may be higher, however, that they will fail to make a significant difference at all. If that is the case, we will be left with both a presidency on steroids and no meaningful reduction in government.
Friedman vs. DOGE
Throughout his career, Milton Friedman championed a government that does only what is strictly necessary to protect individual rights, leaving individual adults otherwise free to make their own choices. In his vision, the state should be limited, decentralized, predictable, and constrained by the rule of law—not an instrument for shaping culture, redistributing wealth, or "managing" the economy.
This philosophy of limited government, along with his training as an economist who understood the power of incentives and tradeoffs, shaped all of Friedman's policy recommendations. First, he was clear about the fundamental role of the federal government, which he believed should be limited to "preserv[ing] the peace, defend[ing] the country, provid[ing] a mechanism whereby individuals can adjudicate their disputes, and protect[ing] individuals from being coerced by other individuals." That explains why only 4.5 agencies survived Friedman's scrutiny.
When asked questions about specific cuts, Friedman grounded his argument in economic reasoning, revealing the unseen costs of government actions that many people overlook. He would explain that nobody takes care of somebody else's property as well as that person takes care of his own, suggesting why the private sector would deliver better services in areas where people have taken the role of government for granted. With the same ease, he would explain that public parks would be cleaner and safer under private ownership or that the Department of Housing and Urban Development exacerbates rather than solves housing problems with subsidies that fuel housing demand while other government policies restrict housing supply.
The underlying principle was always the same: Shrink the government to expand individual freedom and economic efficiency. That means ending subsidies to the private sector, privatizing what needs to be privatized, ending payments for functions that are state and local functions, and watching for inevitable inefficiencies in functions that are legitimate government functions.
On paper, DOGE's action might sound similar to Friedman's goal of smaller government. DOGE is using smarter tech to identify improper payments and straight-up fraud. If it succeeds in reducing both—which cost taxpayers hundreds of billions of dollars a year—the accomplishment will be long overdue. DOGE is also attempting to cut federal employees so far at a lesser pace than the Clinton and Reagan administrations. Federal employment was reduced by over 331,000 under Bill Clinton and by 90,000 under Ronald Reagan.
However, in other areas, DOGE's other cuts look to be far less grounded in a consistent philosophy, despite the exercise being framed as slashing spending and bureaucracy. That doesn't mean the early targets of DOGE's "war on waste"—such as the U.S. Agency for International Development (USAID) and the Consumer Financial Protection Bureau (CFPB)—shouldn't be gutted. But the decisions appear to be primarily driven by culture war judgments about whether or not "diversity, equity, and inclusion" (DEI) is in the contract, or if money is going to organizations that conservatives don't like, as opposed to a consistent limited government ethos.
For instance, talking about what DOGE will do when it will audit the Department of Defense, Defense Secretary Pete Hegseth said, "They care just like we do, to find the redundancies and identify the last vestiges of Biden [administration] priorities—the DEI, the woke, the climate change B.S., that's not core to our mission, and we're going to get rid of it all." As former Congressional Budget Office Director Douglas Holtz-Eakin notes, the cuts seem to avoid agencies "doing things [the administration] like[s]" and instead zero in on agencies "they disagree with." Musk is purging agencies like Education and USAID because "if you walk down the halls…you're not going to find any conservatives," economist Stephen Moore bluntly put it. "It's really important they defund the left."
This is where DOGE and Friedman diverge. Under the Friedman selection process, these grants would rightly be canceled too—but not simply because they fund left-wing culture war activities. Instead, they would be canceled because a sound determination was made that the federal government shouldn't be in the business of subsidizing projects in other countries or that should be funded only by the private sector or the state and local governments. The contracts would be canceled because it's not the role of the federal government.
In his essay "Why I Am Not a Conservative," Friedrich A. Hayek observed that "the conservative opposition to too much government control is not a matter of principle but is concerned with the particular aims of government." This could be in action here.
If this proves to be the case, the DOGE cuts will fall short of what libertarians and classical liberals would like to see done. Parts of the government may be gutted, while other parts—equally costly or duplicative and outside of the role of the federal government—are spared. While it is true that it would at least get us partially where we want to get, there are other limitations to this approach.
Centralization of Power
Cutting wasteful programs, eliminating unnecessary agencies, and reducing bureaucratic overreach are causes classical liberals and libertarians should champion. We also care, however, about the rule of law, the process under which the cutting and eliminating takes place, and who does the cutting. If shrinking the scope of the administrative state is done via an expansion of executive power and/or in violation of legal and constitutional rules, that's a problem. The ends do not justify the means.
Perhaps more importantly, this strategy carries risks.
Friedman was concerned about that too. "Freedom is a rare and delicate plant," he once wrote. "Our minds tell us, and history confirms, that the great threat to freedom is the concentration of power. Government is necessary to preserve our freedom, it is an instrument through which we can exercise our freedom; yet by concentrating power in political hands, it is also a threat to freedom. Even though the men who wield this power initially be of good will and even though they are not corrupted by the power they exercise, the power will both attract and form men of a different stamp."
This is where DOGE may make libertarians and classical liberals uneasy, even if they support the bulk of its outcomes.
DOGE is a creation of the executive branch and is clearly extending its reach. This is not new in Washington. It's true that many presidential powers today go way beyond what the Founders intended. Over time, executives from both parties have accumulated authority both at the expense and with the blessing of Congress. From President Franklin Delano Roosevelt's New Deal bureaucracy to President Barack Obama's executive orders on immigration to President Joe Biden's student loan forgiveness efforts, every modern president has contributed to the erosion of constitutional checks and balances.
President Donald Trump and DOGE are not simply using the excessive powers granted to the executive branch. They are also testing the boundaries of what the executive branch can do. Instead of restoring constitutional limits, Trump is accelerating a model of government in which policy is determined by a single leader rather than by a constitutional process. Even if this discretionary power is exercised only to cut wasteful spending, and even if it's all perfectly consistent with Congress' intended delegation of its legislative powers (which remains to be seen), the current spate of unilateral executive actions should worry those who dislike rule by decree.
Take, for example, the Trump administration's unilateral efforts to dismantle USAID. Many libertarians, myself included, believe that USAID is a wasteful bureaucracy that should never have existed. In practice, it has often propped up dictators and funded other mischief. But the agency was effectively established by legislation in 1998, meaning that the proper route to its elimination is through Congress. Instead of following that process, Trump simply moved to close the agency unilaterally, only later to decide to maintain it for legal reasons, in a much-shrunken form, under the State Department.
It is a good thing that the administration abided by the law, which makes the reform less likely to be legally challenged. However, it remains true that this kind of hasty, arbitrary, heavy-handed decision-making should bother us. It's a symptom of an increasingly imperial presidency that agencies and policies are shaped not by clear legal principles and congressional intent but by the whims of whoever happens to be in the Oval Office. The same logic applies to efforts to weaken or eliminate the Department of Education—which I believe should be terminated by Congress, with education left to the private sector and to the states. I have no such reservation for the fate of the CFPB, which was designed intentionally to be immune to any congressional attempt to defund it, and as such was designed with a fatal flaw.
But in the end, this is a striking distinction between conservatives and libertarians. "Like the socialist, [the conservative] is less concerned with the problem of how the powers of government should be limited than with that of who wields them," Hayek noted in that same essay, "and like the socialist, he regards himself as entitled to force the value he holds on other people."
That means that the risk we face then is that if we cheer when a president mostly governs by decree or circumvents legal constraints to impose libertarian-leaning policies, we have no principled basis on which to stand to object when the next administration uses, as it will inevitably do, the same unchecked powers to expand government, raise taxes, and impose new restrictions on liberty.
Then, there is also the danger that the changes, even the legal and applause-worthy ones, might not last—something that should bother anyone who supports DOGE's general mission.
If You Want Lasting Reform, Congress Must Be Involved
Trump's entire executive agency, including the creation of DOGE, was born out of conservatives' frustration with Congress' unwillingness to do anything and pass any of their priorities. That doesn't change the reality that reforms imposed unilaterally by the executive branch are fragile because they lack legislative permanence. Unlike statutes passed by Congress, executive actions can be easily reversed by future administrations. If reforms are to stick, they need to be enacted through the legislative process, ensuring durability beyond the tenure of any single president.
As noted above, one major reason executive-driven reforms don't last is that they can be undone as easily as they were implemented. For instance, Trump rolled back many of Obama's executive orders, only for Biden to reinstate them, and now Trump is again rolling them back. This constant, dizzying policy seesawing undermines long-term stability and makes it nearly impossible to achieve durable reforms through executive action alone.
Legal challenges also pose a major threat to executive actions. Perhaps most importantly, legislative reforms tend to be more durable because they usually reflect broader political consensus. Most legislation passed by Congress requires majority approval and often involves bipartisan negotiation, making them harder to repeal even when power shifts. The 1986 Tax Reform Act, championed by President Ronald Reagan and passed by Congress, remained intact through multiple administrations. Similarly, the 1996 welfare reform signed by President Bill Clinton survived despite Republican control of the government in later years. When Congress passes a law, it becomes embedded in the legal and political framework, whereas executive orders remain vulnerable to reversal.
While executive action can set the tone and signal priorities, it is a weak substitute for genuine reform. Without congressional backing, most executive-led initiatives are easily reversed.
The bottom line: The DOGE approach to cutting spending should make libertarians uneasy if they want to see lasting reductions in government size and scope, care about the rule of law, and fear executive overreach. Although it could bring some positive changes, it may come with a big price tag, setting a true limited-government agenda back for years. Establishing a smaller and more effective government in the long term is possible—but only if it's achieved the right way.
The post How Would Milton Friedman Do DOGE? appeared first on Reason.com.
