Before President Donald Trump’s inauguration this January, Harvard Kennedy School scholars and practitioners examined what the then-proposed unit known as the Department of Government Efficiency could mean for Trump’s agenda. Just a few weeks into the administration, Trump has issued a flurry of executive orders and other actions that have included giving billionaire businessman Elon Musk unprecedented latitude to employ DOGE as a means to fundamentally transform the federal government. These include firing government workers across a broad swath of federal departments and agencies, seeking to eliminate agencies outright, and gaining access to sensitive government data previously held only by small groups of career civil servants, among other actions. Our faculty members discuss the widespread implications of these actions below.
- Linda J. Bilmes: Elon Musk’s missed opportunities
- John Donahue: Weak hate, rather than tough love
- Stephen Goldsmith: Distinguishing reform from retribution
- Steve Kelman: Musk is uninterested in conventional efficiency
- Elizabeth Linos: Dismantling government itself
- Paul Peterson: State lawsuits threaten Trump’s executive orders
- Lawrence Summers: Safeguarding the Department of the Treasury
Elon Musk’s missed opportunities
Linda J. Bilmes
When
Elon Musk took the reins of the Department of Government Efficiency,
many expected bold reforms from the man who built Tesla, sent rockets to
space, and promised to colonize Mars. So far, it hasn’t happened.
Rather than fostering innovation, his tenure has been defined by chaos.
Musk has been wielding a sledgehammer and tearing down existing
structures with little thought for what could replace them. A more
strategic approach could have transformed DOGE into a force for genuine
reform.
One of the biggest challenges in government is its struggle to innovate. While agencies generally function well in executing their existing responsibilities, there are few incentives to try new approaches. Civil servants are reluctant to propose innovative ideas, for fear of reprisals if they fail to deliver results. This stifles their ability to develop more efficient methods of delivering services.
Musk had a rare opportunity to break this cycle. He could have launched a government-wide challenge, pushing agencies to compete in developing new efficiencies. He could have incentivized civil servants to think like entrepreneurs, rewarding creativity. He could have met with government employees and listened to their ideas. Instead of slashing budgets and firing staff, he could have helped to build a culture of innovation and inspired a wave of problem-solving within government itself.
Musk could have empowered the Inspectors General (IGs). These watchdogs have spent years tracking waste, fraud, and inefficiency, compiling reports that if implemented, could save the taxpayers billions. But the government often ignores their findings. Musk could have convened a high-profile summit with IGs, giving them a central role in DOGE and pushing agencies to act on their recommendations. Instead, the administration has sidelined or fired the IGs. This approach squandered a major opportunity to capture ready-made plans for cost-cutting and efficiency gains.
DOGE is not the first attempt to bring private-sector-inspired ideas to the federal government. In 1982, President Reagan launched the Grace Commission to weed out government waste. The members were corporate leaders from the private sector, who Reagan instructed to “work like tireless bloodhounds” in rooting out inefficiency. While most of its recommendations were never enacted, the commission led to some important successes, including the idea for the Base Realignment and Closure (BRAC) process, which helped the military shut down outdated facilities.
“Musk has been wielding a sledgehammer and tearing down existing structures with little thought for what could replace them. A more strategic approach could have transformed DOGE into a force for genuine reform.”
President Clinton’s Reinventing Government initiative, led by Vice President Al Gore, took a different approach, focusing on making agencies more customer friendly. It introduced the “Hammer Award” (named after the infamous $400 hammer) to recognize agencies that improved service delivery. There was also the post-World War II Hoover Commission and several bipartisan congressional efforts. These initiatives show that with the right leadership, working within the institutional framework can produce consequential change. Musk has chosen to work against the system, ensuring that his tenure will be remembered more for disruption than for lasting improvements.
Musk’s presence at DOGE also had the potential to make government work exciting for STEM talent. The federal government desperately needs more tech talent to modernize aging IT systems and improve responsiveness to the public. Rather than recruiting a few untrained individuals to act as a “wrecking ball,” he could have inspired thousands of young engineers, programmers, and data scientists to bring their skills to government. A well-run recruiting campaign could have made government tech work as desirable as SpaceX or Tesla, setting up long-term efficiency gains.
Musk’s approach to USAID is another missed opportunity. Foreign aid isn’t just a budget line-item —it’s a tool of soft power. China recognizes this and aggressively expands its global influence through aid programs. Rather than pursuing a wholesale shutdown of USAID, Musk could have shifted its focus toward projects that enhance U.S. influence and align with Trump’s geopolitical and diplomatic goals.
Ultimately, knocking things down is easy—anyone who has renovated a house knows that the demolition process is quick, but rebuilding takes much longer. By focusing on destruction rather than construction, Musk is making it much harder for DOGE to produce sustainable efficiency gains. Even if legal and institutional safeguards eventually rein in some of the collateral damage, it will take years for government to recover from the shocks of his tenure.
Linda Bilmes is the Daniel Patrick Moynihan Senior Lecturer in Public Policy.
Weak hate, rather than tough love
John Donahue
Now
we know what Elon Musk is up to with Department of Government
Efficiency, and it’s a big letdown for those of us who care about, well,
government efficiency.
He could have done a magnificent job at diagnosing the ills of the clunky and waste-ridden federal contracting system and proposing efficiency-boosting fixes to Congress.
He also could have brought some “tough love” to the federal workforce, championing reforms that Democrats often balk at because of worker resistance. No doubt he and his proudly inexperienced team would have wreaked some havoc along the way, but they could have done some good.
Instead, he is performing brain surgery with a sledgehammer, treating the workforce not as a neglected asset to be fixed but as a pathogen to be purged.
Rather than tough love, he has chosen weak hate. “Hate,” because what else can we call the bilious sentiments that have overthrown a first-rate mind? “Weak” because his work is anchored in the sand of executive orders and ephemeral politics and will not last.
The Civil Service Act that DOGE is defying has governed federal employment for 142 years, or about three-fifths of America’s history. Does anyone believe that Musk’s model will make it to 2167? Even 10 years from now it will likely be remembered only as a strange episode behind a blip in the budgetary and payroll statistics.
This is because most Americans value most of what Washington does. The abstract slogan of “less federal spending” gets a lot of heads nodding, but deficit hawks have long searched in vain for a robust constituency behind eradicating any major federal function.
“No matter how creatively DOGE delegates tasks to private contractors, states and localities, non-profits, and even AI bots, federal work will still require federal workers.”
At some level Musk clearly knows this. The “tell” is that the featured target for his performative stomp-down is USAID, a small, obscure agency that serves foreigners. And sustaining his assault on it still requires extravagant lying.
There may be enough niche spending that only benefits non-MAGA people and purposes to keep the game going for a while. But the big-buck items are broadly popular. They will survive DOGE, or spring back quickly once it fades away. Abraham Lincoln wrote that government’s purpose “is to do for a community of people what they need to have done but cannot do … in their separate and individual capacities.” The list of such tasks differs community by community and era by era. But there is always a list.
Likewise for the resilience of the administrative state. No matter how creatively DOGE delegates tasks to private contractors, states and localities, nonprofits, and even AI bots, federal work will still require federal workers. And agencies can be rebuilt. Not instantly, not costlessly, and not without some serious damage in the meantime. But the federal workforce—unlike, say, America’s honor among nations—will take years, not generations, to restore.
In “The Rise and Decline of Nations,” Mancur Olson argued that the cataclysmic consequences of starting, waging, and losing World War II seared away cultural and economic rigidities in Japan and Germany and thus enabled their postwar prosperity. As Olson himself recognized, this was a hideously wasteful route to social change. But something similar is likely for beleaguered Washington. Since we are fated to endure the DOGE firestorm, we can find solace in the prospect that its targets will come back stronger, smarter, and fairer. In the long run, Musk might improve government efficiency despite himself.
John Donahue is the Raymond Vernon Senior Lecturer in Public Policy.
Distinguishing reform from retribution
Stephen Goldsmith
Elon
Musk’s Department of Government Efficiency is off and running,
proclaiming on its website that “the people voted for major reform,”
which, indeed, many voters did.
As we see news stories about nuclear security workers laid off and rehired, of worries about safety risks due to air-traffic control understaffing, and concerns that National Park Service and U.S. Forest Service layoffs could weaken capacity to fight wildfires, we are reminded that fewer workers do not ipso facto equal reform or efficiency. Changing large bureaucratic organizations in a way that improves effectiveness is a whole lot more challenging to achieve.
DOGE can accomplish its stated goal of reform, not just smaller government, and do so with fewer unintended consequences. However, achieving deep, permanent change will require a broader array of tools than we’ve seen so far—tools that require nuance and changes in laws to prove legal and sustainable.
Even a straightforward initiative like hiring and firing requires nuance. Weeding out underperformers takes work. Firing dedicated civil servants en masse, not so much. I am reminded of when I served as deputy mayor for Mike Bloomberg during a tough budget year in New York City. Mayor Bloomberg imposed a flexible hiring freeze that allowed data-driven productivity improvements to compensate for reductions in force. At the same time, he approved limited hiring where demands for a certain skill could not be easily absorbed elsewhere.
President Trump’s view that bureaucratic reform takes a large dose of top-down leadership, even disruption, is unmistakably true. Anyone who has wrestled with a large bureaucratic organization in search of innovation can provide painful stories that support the need for reform. Comprehensive change requires a customer-focused and bottom-up culture of innovation as well from an empowered, not discouraged, workforce.
Government’s workloads will not shrink. Technology allows breathtaking advances in effectiveness but to fully harness the benefit requires broader data literacy among workers and large doses of business process reengineering. Genuine change must include reforming a civil service that forces too many public employees into mindless routines, preventing them from using problem-solving discretion.
“Genuine change must include reforming a civil service that forces too many public employees into mindless routines, preventing them from using problem-solving discretion.”
Internal regulations present infuriating obstacles. When I served as chairman of the federal AmeriCorps program, we sought to replace a paper application process with a digital one; federal lawyers insisted we needed six months to get permission from the enforcers of the Paperwork Reduction Act! Once, while serving as mayor, I offered HUD officials and a Congressional committee to reduce our annual funds by 20% in return for relief from administrative restrictions that consumed almost a quarter of the spend. Both the chair of the applicable subcommittee and HUD immediately rejected the idea.
Many changes require Congressional action, including consolidating departments, removing layers of oversight, sunsetting obsolete and duplicate agencies, changing civil-service rules, and focusing oversight on outcomes not process. DOGE can assist by promoting government-wide simplification. In so many cases, the problem rests not with the goal of a well-meaning regulation but rather with the tangle of red tape associated with compliance.
DOGE can accomplish its reform goal of slimming down an overbearing government but only through a legal and comprehensive manner that that, by distinguishing reform from retribution, empowers those federal employees committed to service and innovation.
Stephen Goldsmith is the Derek Bok Professor of the Practice of Urban Policy.
Musk is uninterested in conventional efficiency
Steve Kelman
Elon
Musk was unleashed on the government in an executive order by President
Trump on his very first day in office, establishing the so-called
Department of Government Efficiency, which was inserted into (one might
say foisted on) the U.S. Digital Service, an organization in the
Executive Office of the President originally established by President
Obama as a way to encourage young public-service minded techies to do a
one-to-two year stint in government to help agencies with IT
modernization. Until very recently, he had no formal government job at
all, which conveniently exempted him from conflict-of-interest rules, or
from having to divest his stock holdings or put them in a blind trust.
As soon as Musk started, it became clear that he had little interest in conventional ideas of efficiency, which involve getting better organizational performance without spending more money. The only actual efficiency issue Musk discussed was getting rid of the penny (which costs 4 cents each to produce), which may be a good idea. Instead, Musk quickly focused on slashing or eliminating entire government programs he didn’t like. An early list by his erstwhile sidekick Vivek Ramaswamy included eliminating the IRS, FBI, and the Department of Education. Now, after going after USAID, Musk has set his sights on the Consumer Financial Protection Bureau and benefit fraud in social security. What these two objects of his ire have in common are that both primarily help people of modest means. This focus is particularly distasteful coming from the richest man in the world.
“As soon as Musk started, it became clear that he had little interest in conventional ideas of efficiency, which involve getting better organizational performance without spending more money.”
By contrast, there are well-known examples of government inefficiency that have escaped his eagle eyes. The government subsidizes insurance premiums for hurricane insurance in Florida (given Gov. Ron DeSantis, he doesn’t want to go after that) and for coastal tourist properties on the southeast coast. FEMA actually paid for new insurance policies for people in Florida who submitted an earlier claim for insurance from a hurricane, as well as giving these people $3,200 in cash. Most economists would say this gets incentives exactly backwards, since they encourage building in flood-prone areas. To take an example closer to Musk’s interests, the government spends billions on missions to far-off planets such as Jupiter—there have been five over the years—where SpaceX has launched the rockets, programs that many would regard as unnecessary.
Musk has no idea of how our government works. He seems unaware that our system establishes multiple institutions sharing power in addition to the president, most obviously Congress. Given his ignorance, we should not be surprised that Musk seems to believe he has the authority to freeze government spending and to “delete” a major government agency through a memo, on his own say-so.
I spent four years in the federal government working on increasing efficiency in our procurement system, a big spending item for the government. I worked with Republicans and Democrats. It’s something I believe in. But I don’t believe in Musk’s cruel rampage.
Steve Kelman is the Albert J. Weatherhead III and Richard W. Weatherhead Professor of Public Management, Emeritus.
Dismantling government itself
Elizabeth Linos
Before
DOGE officially launched, many of us thought that it would act as a
hollow advisory body—one that might not accomplish much operationally
but would still inflict harm by spreading misleading narratives about
government, exacerbating stereotypes and falsehoods about government
workers. The past few weeks have revealed a reality that’s far worse:
DOGE is not just misrepresenting how government works; it is actively
dismantling government’s ability to function.
The administration’s “deferred resignation” offer was the first major test of its assumptions about public servants. The idea was simple: pay employees to leave and, in theory, shrink a so-called bloated workforce. But the results tell a different story. Only about 75,000 employees took the deal—a figure that aligns with natural retirement rates and accounts for just half of turnover in any typical year. If federal employees were merely looking for an easy payout, participation should have been much higher. Instead, the lackluster response should force a reckoning: the premise that government is full of unmotivated, unproductive workers simply doesn’t hold up.
What followed was even more damaging. The administration has now purged probationary employees—workers who have been hired in the past two years to provide health care for our veterans, protect our nuclear arsenal, and coordinate wildfire responses, among other critical functions. If DOGE’s true goal were efficiency, this would be the worst way to achieve it. These employees represent the future of government: they bring in specialized expertise, technological skills, and fresh ideas at a time when the civil service is already facing a “silver tsunami” with twice as many employees over 60 than under 30. Rather than assessing performance or focusing on retaining top talent, DOGE has implemented indiscriminate cuts that erode government’s capacity to improve itself.
“At this point, it is impossible to argue that these cuts stem from a mere misunderstanding of government operations. The past few weeks have made it clear: DOGE is not about fixing inefficiencies—it is about dismantling government itself.”
The administration also seems to be weakening, or fully eliminating, the teams that were doing exactly the kind of work DOGE claims to value: teams that focus on data, evaluation, and customer experience have spent years reducing bureaucratic red tape, modernizing service delivery, and bringing in critical tech talent. These teams were already working—under both Democratic and Republican administrations—to make government more efficient and responsive to the public. Gutting them is not a strategy for reform; it is a strategy for dysfunction.
At this point, it is impossible to argue that these cuts stem from a mere misunderstanding of government operations. The past few weeks have made it clear: DOGE is not about fixing inefficiencies—it is about dismantling government itself. If we truly care about making public services work better, the answer isn’t mass layoffs and blind cuts. It’s investing in the people and processes that can drive real improvement. Anything less isn’t reform—it’s sabotage.
Elizabeth Linos is the Emma Bloomberg Associate Professor of Public Policy and Management.
State lawsuits threaten Trump’s executive orders
Paul Peterson
Is the president tied down by Lilliputians?
President Donald Trump’s executive orders and agency directives are terrorizing Washington’s insiders, but the most consequential are provoking legal actions that could delay their implementation or eviscerate them altogether. Trump speaks as if he is the biggest bully in town, but he may soon find himself a Gulliver tied down by Lilliputians.
The constitutional design of the federal system presents a fundamental challenge to the administration’s agenda. According to long-standing legal doctrine, the Constitution gives sovereign power to both the federal government and the several states, giving states the authority to challenge federal actions. When different parties are in power at the state and national level, conflict between tiers of government escalates—and the winner is for the courts to decide, not the man residing inside the Oval Office.
A federal system induces moderation and compromise. “The different governments will control each other” in the 21st century, just as they did when James Madison first penned these words.
Within days of his inauguration, Trump announced the elimination of units engaged in “diversity, equity and inclusion” activities, fired senior staff at the FBI, cut federal grants, reduced university research funding, began the deportation of those who lack proof they entered the United States legally, and much more.
But the design of the federal system may preclude the president from achieving many of these and other declared objectives. In 23 states, including California, New York, Pennsylvania, Illinois, and North Carolina, the attorney general was either elected as a Democrat or appointed by a Democratic governor. The office of the attorney general is responsible for defending the interests of the state against any perceived abuse of federal power.
Democratic attorneys general are just as prepared as the Trump team for the current political struggle. Coalitions of attorneys general formed prior to election day so coordinated responses to executive orders could be mounted within days, sometimes hours, of presidential announcements. Forty lawsuits and counting have been filed so far, a hefty share emerging from the offices of attorneys general. “We saw this coming. … We started preparing… almost two years ago,” explained former Oregon Attorney General Ellen Rosenblum. “I believe that there’s no group better prepared to push back,” confirms Letitia James, New York’s attorney general and gubernatorial hopeful. She takes pride her that her “office has been preparing… I am ready to do everything in my power to ensure our state and nation do not go backwards.”
Coalitions of attorneys general have numerous advantages in the upcoming legal battles. For one thing, they do not need to worry much about standing to sue. An attorney general has the duty to defend the state whenever federal action appears to impinge on its sovereignty. As California’s attorney general, Rob Bonta, put it: “It’s hard work, but ... that is what we signed up to do.” Oregon’s legal head views Trump’s actions as an “existential threat.”
“The office of the attorney general is responsible for defending the interests of the state against any perceived abuse of federal power. Democratic attorneys general are just as prepared as the Trump team for the current political struggle.”
Just as important, attorneys general, as plaintiffs, are generally able to pick the judge who will hear the case. Democratic presidents nominated 60% of all the judges sitting on the federal district court bench. Attorney general coalitions need not hunt far before finding one likely to bend a friendly ear.
A temporary injunction by a district judge will halt the impact of executive orders and agency actions. That ruling stands until overturned by a higher court, a process that can take months, even years, especially if the matter is to be determined by the Supreme Court. Meanwhile, the district-court decision applies nationwide. A district judge in Rhode Island can hand down a ruling that applies in California, Florida, and everywhere in between.
With all these advantages, plaintiffs against the Trump administration have won nine out of 10 decisions in district courts thus far, a completion rate most quarterbacks will take. If anything close to this rate continues, it will hollow out Trump’s boast to make America great again.
The president can hope to win back early defeats as decisions are appealed up to higher federal courts. Eventually, a conservative Supreme Court may be convinced by the arguments offered up by the President’s solicitor general. But months, even years, pass before the highest court in the land finally decides. Even then, district judges can, if so disposed, interpret the Supreme Court’s ruling in creative ways that undermine the administration purposes. It took decades before the Brown v. Board of Education decision took effect in southern states determined to resist its impact.
A term-limited president cannot wait that long. Theodore Roosevelt became a lame duck the day after he announced he would not run for re-election. Trump cannot avoid the same fate, as the Constitution’s 22nd Amendment precludes another term. That Trump regularly cracks jokes about serving beyond his second term reveals how much the passage of time is on his mind.
Time will run out quickly if Democrats win control of the House of Representatives in the mid-term elections. In 21 of the 24 most recent mid-terms, the party of the president has lost ground in the lower chamber. If history repeats itself, Trump will find himself knotted down firmly by potent Democratic opposition.
Tiny majorities in both House and Senate help explain Trump’s reliance upon constitutionally problematic executive actions. In all but financial measures, Republicans have to attract at least seven Democratic votes to end filibusters necessary to pass any bill not enacted via the reconciliation process, which is exempt from the 60% cloture rule. Reconciliation bills are (for the most part) limited to fiscal matters, and they may not increase government expenditure, making them a complicated vehicle for introducing policy change. Any small group of Republicans can kill a reconciliation bill that includes material not to its liking. After all the deals are made, much of the Trump agenda could find itself left on the cutting room floor.
Stalled on Capitol Hill and with time running out, Trump must use the power of the pen, as Barack Obama put it, to effectuate his policies. That Trump cannot afford delay explains the frequency with which his signature is appearing.
Executive orders can be written on day one, but they last only as long as the executive branch is in friendly hands, as Trump discovered when his successor canceled Trump’s first round of orders. Unless Republicans consolidate their national majority, they will undoubtedly happen again.
This is not to say that Trump’s actions will have no impact. Genetically male athletes may not participate in female sports during Trump’s time in office. Diversity, equity, and inclusion initiatives may wane under federal pressure, though institutions may practice what they don’t preach. Undocumented immigration may slow, but actions taken by cities, states, and lower courts will depress deportation rates. Government spending will retract from its COVID high, even if it remains above pre-pandemic levels. Pennies will disappear from cash registers, but nickels will become more expensive to produce.
In sum, the pace of change in American government will be more measured than current rhetoric suggests. Democrats shall become increasingly grateful for federalism’s inherent, impassive, massive strength.
Paul Peterson
is a professor of public policy at HKS and the Henry Lee Shattuck
Professor of Government at FAS. A longer version of this piece was originally published here.
Safeguarding the Department of the Treasury
Lawrence Summers
To
protect citizens from arbitrary taxation and spending, our founding
fathers designed the Constitution such that only Congress would have
“the power of the purse” or the prerogative to set the federal budget.
Since then, with few exceptions, the Department of the Treasury has
faithfully executed the will of Congress by ensuring that federal
payments were made on time and according to law.
I was concerned by recent reports suggesting that the newly-created Department of Government Efficiency sought to use the Treasury’s payment system to unilaterally freeze the funding of USAID. Thankfully, Treasury Secretary Bessent has committed to maintaining the integrity of federal payments, but even the idea that such a system could be managed by political actors is worrying. While I agree with the importance of sensible debt management policy and government waste reduction, enacting fiscal policy via the Treasury payment system would be inappropriate for three reasons.
First, allowing unvetted political appointees to handle taxpayers’ sensitive financial information would amount to a serious invasion of privacy. Individuals with access to the federal payment system can easily view taxpayers’ social security numbers and bank account information. Historically, this system has been operated by a small group of nonpartisan civil servants to ensure the security of taxpayer data. Therefore, it is imperative that the Trump administration adopt a similar degree of care and sensitivity towards the privacy of its constituents.
Second, freezing funds subverts democracy by overriding the authority of Congress. Since the Nixon presidency, the Supreme Court has reaffirmed in key cases that Congress has primary authority to decide how government funds should be allocated. Both Chief Justice Roberts and Justice Kavanaugh, whom President Trump appointed in his first term, have stated that the president has no right to decide whether financial promises made by Congress will be kept.
“Millions of Americans take for granted the fact that the government keeps its financial promises.”
Finally, interfering with government payments to pick winners and losers opens the door to political favoritism and capture by special interests. Suppose the Treasury began to decide on an ad hoc basis which payments should be honored. There would be few guardrails preventing political appointees from privileging the contracts and payments of friendly industries while blocking the payments of political enemies.
Millions of Americans take for granted the fact that the government keeps its financial promises. Retirees expect their Social Security checks every month. Medicare providers expect to be reimbursed. Members of the military expect to be paid on time. Disrupting these payments or risking their cutoff threatens the livelihoods of these people and those who depend on them. Furthermore, any perception that the government is willing to renege on its financial promises could be seen as a threat to the creditworthiness of government debt. If U.S. Treasuries cease to be seen as a safe asset, it would have severe implications for the health of our financial system.
For now, this battle is playing out in the courts, where a federal judge has blocked DOGE from further access to the Treasury payment system. The ultimate question is whether the Trump administration will respect the courts and the authority of Congress. For the sake of the rule of law and the integrity of the American financial system, I hope that they do.
Lawrence Summers is the Frank and Denie Weil Director of the Mossavar-Rahmani Center for Business and Government and the Charles W. Eliot University Professor.
—
Banner photograph by Samuel Corum/Getty Images; portraits by Martha Stewart