DOGE, Done Right
Efficiency as a path to fiscal consolidation
This article is part of The Boyd Institute’s quarterly policy print on debt and deficits. To learn more about them and the work they do, or submit your own article, click here. Viewer’s discretion is advised.
In the third episode of the last season of Euphoria, Nate and Cassie Jacobs’ wedding night ends horrifically when Naz, a loan shark, severs Nate’s pink toe for not paying his debts. Plot holes and bad writing aside, the lesson here is simple: pay your debts and tackle problems head-on rather than kicking the can down the road.
Sadly, this lesson is clearly lost on Congress, which has kept its head in the sand regarding the dire warnings about our fiscal health. As of Q1 2026, the federal government owed roughly $31.5 trillion, or about 100% of GDP, in public debt (i.e., debt owed to nongovernmental creditors). That amounts to $91,265 for every man, woman, and child in the country. More recently, budget deficits (i.e., the gap between spending and revenue) have stabilized at a jaw-dropping $1.8 trillion, or roughly 6% of GDP, an unprecedented level during peacetime economic expansions.1
And it’s only getting worse: higher interest rates and ballooning costs for Social Security and Medicare risk taking the debt to unprecedented levels. The Committee for a Responsible Federal Budget projects that if no changes are made, public debt will reach 120% of GDP by 2036 and an eye-watering 175% by 2056. And, as the Boyd Institute notes in its helpful fact sheet on the national debt, these debt projections are likely to worsen over time.
With our utter fiscal annihilation imminent, our federal government has snapped into (in)action. Years of brinksmanship and partisan polarization have kept our lawmakers mired in gridlock as the problem worsens. Congress, as usual, is asleep at the wheel, filled with politicians who prattle on about hard choices but ultimately cave when the rubber hits the metaphorical road. Indeed, our ruling class has been completely spoiled by fiscal depravity since at least the George W. Bush administration, if not earlier (I blame Ronald Reagan, but to each their own);2 they have neither the inclination nor the will to cut the problem down to size.
But what if we could trim the deficit without the hassle of raising taxes or cutting spending? What if we could do so merely by improving the enforcement of existing laws and programs? What if there were literal billion-dollar bills just ripe for the taking? Economists never fail to remind us that there ain’t no such thing as a free lunch, but when it comes to improving government efficiency, the economists are dead wrong.
I do not wish to suggest that efficiency alone can solve the nation’s budget problems. This may have been true a couple of decades ago, but years of ruinous tax cuts, foreign policy boondoggles, and multiple once-in-a-lifetime crises (the Great Recession and COVID-19) have already screwed the metaphorical pooch. Eventually, Congress will have to put on its big-boy pants and do some actual lawmaking. Nonetheless, I believe efficiency plays a much larger role in improving the nation’s finances, and I also wish to rescue the cause of government efficiency from certain DOGE-minded lunatics who shall not be named.
This article is divided as follows. Part I will clear up some widely-held misconceptions about public finance, misconceptions that bedevil even the sharpest fiscal watchers. Part II explores potential efficiency savings on the spending half of the equation, primarily by implementing reforms already recommended by internal government auditors. Part III will address the revenue side, which primarily entails narrowing the massive gap between taxes owed and taxes actually paid. Part IV will briefly estimate the scale of savings that we can reasonably expect from these reforms. Part V will round out the discussion by addressing any potential criticism that may come our way.
Part I: Clearing Up Some Misconceptions
Before I get into the weeds, however, I want to make a few things known. First, I personally find much of the kvetching around the deficit and deficit-adjacent topics to be counterproductive at best and downright misleading at worst. Even the sharpest budget wonks tend to get hysterical once deficits become the talk of the town, which helps explain why nobody in government actually talks about these problems in the first place. Nothing pisses me off more than when half-educated fiscal hawks throw large numbers around in a bid to scare up a gullible audience (okay, I did it too, but so what? At least it got your attention).
In any case, most of these very same fiscal hawks are nowhere to be seen when their favorite politicians enact ruinous policies that drag the nation further into debt, like the Trump tax cuts (or the Bush tax cuts, or even the Reagan tax cuts). Such hypocrisy only goes to show that, in the immortal words of Dick Cheney, “deficits don’t matter.”
Anyway, if you want a comprehensive and well-balanced summary of the nation’s fiscal problems, I recommend the Boyd Institute’s recent primer on the very topic. While I have a few quibbles with the wording and topical choices, which I will elaborate on later, I think it demonstrates a clear grasp of our nation’s fiscal challenges. That said, I intend to use this part of my essay to clear up a couple of misconceptions that even the Boyd Institute overlooked.
Misconception #1: We Need to Pay Off the Debt
Debunking the first misconception is simple: we actually don’t have to pay down the national debt. Hell, we don’t even have to stop its growth, at least not in nominal terms. A national debt that grew at a fraction of the current pace would more than adequately put the nation’s finances on a sounder footing.
The reasoning is straightforward: government debt =/= personal debt. Put another way: the federal budget is not like a household budget. Households, which are (usually) composed of human beings with finite lifespans, are obligated to pay off their debts, ideally over the course of their lives (though personal debts can and do often haunt families from beyond the grave). Even allowing for bankruptcy, debt forgiveness, and other technical exceptions, the general rule remains that ordinary households cannot roll over their debts indefinitely: that’s just not how this works. You rack up debt over the course of your life, and either you or your descendants will be forced to pay it back, plus interest. If you don’t, your credit rating will tank, and you and/or your descendants will be hard-pressed to find anyone willing to lend you money.3
But for the government, things don’t quite work that way. The government is a giant bureaucratic entity that is effectively immortal: elected officials (and their constituents) come and go, but the government itself stays in place (at least until we get to Full-Automated Gay Space Communism, but that’s another story). You know the old saying about how the only certain things in life are death and taxes? Well, that’s where it comes from: you may live, grow old, and die, but the tax man will always be there waiting for y’all to pay up.
Since the government’s lifespan is theoretically infinite, it borrows money under conditions that most people find unimaginable. Because a government’s ability to honor its debts hinges entirely on its ability to survive and generate tax revenue, governments can borrow money more cheaply than people (or even most businesses). And assuming they remain on good terms with the bond markets, governments can rollover debts at will for practically eternity. Hell, the British government didn’t fully redeem its Napoleonic-era debts until 2015.
Again, it’s worth reiterating that government debt is a different beast than personal debt. Government debt, which usually takes the form of bonds and other debt securities issued by the Treasury, is generally seen as a safe and secure asset by creditors, ranging from large investors and foreign central banks to ordinary households and businesses. We see government debt as an asset on our books; as a result, U.S. Treasury debt (thanks to our reserve currency status) is one of the building blocks of global capitalism. The Federal Reserve has a good explainer on how the dollar (and dollar-denominated assets) play a major stabilizing role in global capital markets.
All in all, the takeaway is simple: government debt doesn’t really need to be paid off. Governments can take out new loans to pay off old debts, and then take out even more loans to pay off these new debts, so long as they maintain the confidence of their creditors. Actually paying off the debt would undermine a key pillar of international capital markets, thereby unraveling the global economy. The last time we paid off the national debt was during the presidency of another controversial populist named Andrew Jackson. Jackson’s chaotic presidency was followed by a series of financial panics and painful recessions, partly caused by his austere fiscal policies and a paranoid distrust of bankers.
Misconception #2: We Need to Balance the Federal Budget
Now that I’ve shown that we don’t really need to pay off the debt (entirely or in part), it’s time to deal with the second misconception that plagues everyday fiscal discussions: “balancing” the budget. In most discussions about fiscal policy, including those entertained by otherwise astute fellows at the Boyd Institute, everybody assumes that we need to balance the budget to get the debt under control. In other words, we need to treat the government like a household and ensure that we don’t spend more than we take in.
The logical flaws are similar: governments are like households, households need to balance their budgets, so governments must do so as well. But as we’ve shown above, public finance is an entirely different beast, with constraints and opportunities that differ vastly from those available to ordinary people. Fiscal hawks who insist on treating the public coffers as they would their own purses (or as Trump treats trade deficits) are running a fool’s errand, or deliberately misleading the public for their own nefarious purposes.
In truth, governments don’t really have to balance their budgets. In fact, they probably shouldn’t: reducing deficits to zero, or generating a surplus, generally takes money out of the economy, either in the form of excessive taxation or miserly spending. This needlessly slows economic growth (especially if the spending cuts target public investment) and raises unemployment beyond what is necessary, all else being equal.
Instead, governments should focus on stabilizing debt relative to its potential income: in most high-level fiscal discussions, the debt-to-GDP ratio is the most important metric. Ideally, governments should aim to stabilize their debt-to-GDP ratios over the long run, using above-average deficits to combat recessions and below-average deficits to keep debt growth slower than economic growth. This means that, to achieve a baseline level of stability, the government should target the deficit at a level that stabilizes the debt-to-GDP ratio (ideally, you’d want the ratio to shrink over time, but let’s not let the perfect be the enemy of the good here). Most experts believe this level is around 3%: at that rate, debt-to-GDP stabilizes over time. And since debts are largely set in nominal terms, their real value will gradually decline.
Unfortunately, as stated above, the U.S. government seems incapable of satisfying even this minimal standard. Federal deficits are roughly double what they should be, which means that we need to cut spending and raise taxes by roughly 3% of GDP annually to avoid bigger debt problems down the line. 3% of GDP is not nothing, though it’s nowhere near the fiscal apocalypse that many self-appointed fiscal hawks make it out to be. It also provides a perfect segue to the first part of my deficit-reduction proposal.
Part II: The GAO’s New Groove
Between 2018 and 2022, taxpayers lost $233-$521 billion annually due to waste, fraud, and abuse, plus an additional $186 billion from improper payments. Extrapolated over a ten-year horizon, the savings accrued from curbing such gross abuses would really help improve the nation’s fiscal outlook.
Thankfully, the underpaid and underappreciated staff of the Government Accountability Office (GAO) are committed to investigating these very instances of fraud and inefficiency in federal programs. The GAO is a congressional watchdog charged with providing “timely, fact-based, non-partisan information” to improve government efficiency. It employs a dedicated team of auditors, investigators, and economists to comb through the depths of the federal bureaucracy, identifying inefficiencies and providing policymakers with the information and the tools needed to fix them.
And these reforms have paid off, bigly: since 2003, the GAO saved taxpayers nearly $1.5 trillion, arguably more if one takes foregone interest payments into account. Roughly 75% of GAO recommendations proposed within any given fiscal year are eventually implemented, a testament to the agency’s integrity and judiciousness among the congress critters in D.C. And it does all this on a paltry $860 million budget, boasting a marginal return on investment of roughly $145 for every $1 spent over the past five years.
Unfortunately, that’s where the good part ends. You would think that since most GAO reforms are eventually enacted anyway, there’s relatively little juice to be squeezed from the other 25% of outstanding recommendations. But you’d be wrong: in FY 2023, the federal government lost at least $236 billion due to improper payments, most of which was concentrated amongst six high-profile government programs that attract fraud like a bug zapper attracts mosquitoes. Not coincidentally, most of the GAO’s outstanding recommendations involve those same six programs. They could achieve considerable savings and efficiency gains, but agency heads are notoriously slow to respond, let alone implement, these reforms. At the same time, most elected officials are too preoccupied with the news cycle to focus on actually making government work, and when they do, they often make a mess of things.
And it gets worse: over the past three decades, the GAO’s budget has stagnated in real terms while absolutely plummeting as a share of total federal spending.4 While the country rightly (and perpetually) debates the public sector’s role in everyday life, hardly anyone could disagree with the notion that funding for government watchdogs should scale with the size of government more generally. And yet the GAO is compelled to carry on its mission while fighting for crumbs in the appropriations process. We thus find ourselves in the worst of all possible worlds: shrinking budgets for fiscal watchdogs amid a broader splurge of government spending, much of it falling through the leaky buckets of the administrative state.
This is a shame, especially considering how achievable these savings are. Between 2011 and 2024, the GAO helped achieve roughly $775 billion in savings by reducing duplication and fragmentation in government programs, while also rationalizing Medicare payment formulas and clawing back ineligible claims for COVID tax relief. To put this into perspective, these savings (adjusted over a ten-year horizon) are comparable to axing major tax expenditures like the charitable contribution deduction.5
But if these savings are so achievable, a clever critic might ask, why haven’t they already been achieved? Bureaucratic intransigence, mostly. GAO recommendations are not self-executing: instead, they are largely informative, providing Congress and the executive branch with information to press absent-minded agency heads. To be sure, this culture of ambivalence is less about negligence and more about priorities: bureaucrats have a lot on their plate these days and can be forgiven for looking askance at the prospect of upsetting the applecart for mere pennies on the dollar. Nonetheless, it is remarkable how much time and money are wasted due to bureaucratic inattention.
This question of inattention is doubly true for elected officials, who spend an increasing amount of time begging for campaign contributions and reacting to the latest soundbite. In the era of social media and the 24-hour news cycle, political issues need to be sexy and/or salacious to cut through all the slop and innuendo on CNN and TikTok. All you need is a clear villain, an endearing victim, and a cheap, silver-bullet solution.
Unfortunately, improving public sector efficiency has absolutely none of these characteristics. The taxpayer may be an appealing victim to save, but hardly anyone (normal) identifies as a “taxpayer” tout court. The villain, meanwhile, is nonexistent: while there are a few dastardly interests who profit from government inefficiency, including some of the largest campaign donors in the country,6 most public-sector inefficiency emerges slowly and organically, sort of like body fat. And, like unwanted body fat, the solution to inefficiency is more pedestrian and unappealing than inefficiency itself: better data sharing, improved inter-agency collaboration, and the gradual pruning of redundant rules and programs. An attention-grabber this most surely is not.
Like the problem itself, most existing proposals for reforming the GAO (and government more generally) are unsexy and largely tinker at the edges: incremental budget increases, regularized communication, and formal notice-and-comment periods for released recommendations. Given the agency’s stagnant budget, some advocates have proposed raising its funding to roughly $1 billion a year, allowing it to hire more investigators and spend more time poring over the government’s books. They also recommend requiring agency leaders to make public comments on GAO findings and to provide official reports on their progress in implementing agency recommendations within generous timeframes.
But in the spirit of the Boyd Institute’s pithy motto, I insist we go bigger. First, the GAO should have a much bigger budget to reflect the size and scope of government today. Major publicly traded corporations (i.e., those that gross more than $5 billion annually) spend approximately 0.057% of gross revenue on external audits of their accounts; I’m not usually a fan of the whole “run-the-government-like-a-business” style of governing, but at least here it makes some sense. Using FY 2024 as a benchmark and making the generous assumption that external audits can be analogized to the GAO’s mission, imitating Wall Street’s example would require boosting GAO outlays to about $2.8 billion per year, roughly four times the current budget. Since government bureaucracies are notoriously sluggish, this additional funding should be phased in over time, with the budget then permanently fixed at 0.057% of the previous fiscal year’s gross receipts.
Furthermore, we need to change the process for how GAO proposals are implemented. Rather than frantically waving our hands to capture the attention of agency officials or pushing through boring notice-and-comment requirements, I propose shifting the balance of power in GAO’s favor. Under my plan, we should transition towards a more “opt-out” approach to government reform: all outstanding GAO recommendations with demonstrable financial benefits (a category already tracked in GAO reports) must be completely implemented within two years of their initial promulgation, unless GAO recommends that implementation can or must occur within a longer timespan.
Moreover, the GAO will provide “GAO ratings” for every federal department and its senior leaders to assess which programs (and which officials) are most effective at implementing its directives. Congress will receive regular reports on these ratings, which will also be thoroughly integrated into the disciplinary criteria for senior departmental officials. Officers who fail to implement GAO recommendations promptly (as evidenced by poor GAO ratings) will have their tenure placed under review; in other words, stonewalling will constitute “cause” for termination. All unchallenged recommendations that remain outstanding will be subject to the two-year implementation deadline. In other words, agency heads will be compelled (by law and self-preservation) either to challenge GAO proposals substantively or to rigorously implement them, thus resolving the problem of GAO reform proposals being lost in the daily avalanche of paperwork produced by the federal bureaucracy.
To be sure, this is not a silver bullet for our nation’s fiscal problems. Eventually, the nation will have to reckon with the structural drivers of project deficits: lackluster tax revenue, runaway healthcare costs, and a deteriorating worker-dependency ratio for retirement benefits. But I forthrightly believe that a more proactive and effective GAO could lessen the need for painful fiscal austerity, which will be further elaborated upon in Part IV.
Part III: Unleash the IRS
Before I start ranting about the wonderfully entertaining topic of tax enforcement, I want to get something out of the way. Nobody likes paying taxes: anybody who does ought to get a brain scan. I say all this despite being a diehard tax-and-spend leftist who nonetheless finds the “taxes are the price we pay for civilization” argument as trite and off-putting.
That’s not wrong. We don’t live in a libertarian utopia where tax cuts reduce the deficit. However much you hate to admit it, higher taxes are inevitable if we’re actually serious about stabilizing the nation’s finances, especially after decades of fiscally irresponsible tax cuts implemented by Donald Trump and George W. Bush (don’t even get me started on Reagan). Indeed, one way to distinguish the rare genuine fiscal hawk from the more opportunistic ones is whether they seriously consider raising taxes, particularly on middle and working-class Americans, as an option. The Boyd Institute has made it clear: tax cuts don’t starve the beast.
If it were up to me, we’d go back in time and restore the tax rates that existed right before the Reagan administration. I’d even settle for the Clinton tax rates, adjusted for inflation. But as Mont Pelerin Review so eloquently argues, tax hikes are generally painful means of achieving fiscal consolidation, since they discourage economic activity and thus reduce economic growth over the long run. I’m somewhat skeptical (not of the substance of the argument, but rather its severity), but I do not wish to adjudicate these criticisms here. More relevant for our purposes is that actually passing a tax hike (or letting existing tax cuts expire) is a dead letter in Congress for obvious reasons. We need to look at other ways to boost federal revenues, primarily by improving the enforcement of existing tax laws. Welcome to the wonderful world of the “tax gap.”
The tax gap is what it sounds like: it’s the difference between gross taxes owed and the total actually collected by the IRS. In an analysis by the nonpartisan Congressional Research Service (CRS), roughly 80% of the gross tax gap results from underreported income (i.e., underreported income or invalidly claimed deductions and credits). Self-reporting taxpayers are far more likely to underreport than third parties (i.e., employers, banks, etc.); this is especially true for underregulated income sources such as sole proprietorships and rent/royalty rights.
More generally, tax noncompliance overwhelmingly arises among high-income households, which is not surprising, given that they pay more in taxes. Leading studies have estimated that, on average, the top 1% underreport their incomes by roughly 20-25%, whereas the bottom half of taxpayers underreport by less than 7%. Far from being a problem of general deviancy and disorder, tax evasion is largely a function of elite behavior.
What does this mean in actual numbers? In FY 2022, the IRS estimated that total tax liability was roughly $4.6 trillion, of which only $3.9 trillion (85%) was voluntarily paid on time. Including late and enforced payments, which add an extra $90 billion or so to federal coffers, the resulting net tax gap amounts to roughly 13.1% of total taxes owed, or roughly $600 billion a year. That same year, the federal deficit stood at $1.4 trillion, down nearly 50% from the year before (largely due to the expiration of COVID relief). Had the government fully closed the tax gap, the deficit would shrink to roughly 3.1% of GDP, bringing it within earshot of our 3% target and reducing overall deficits by at least $6 trillion over the next decade. And since revenues generally grow with the economy over time, the deficit would shrink even more, stabilizing overall debt levels and turning our looming fiscal crisis into a manageable embarrassment.
Now I know what you’re thinking: completely closing the tax gap is infeasible. And I agree. However, improving enforcement and reporting requirements should claw back a substantial part of the shortfall. The CRS identified several policy options for narrowing the tax gap: simplifying the tax code, expanding enforcement, and greater oversight over tax preparers. Since simplifying the tax code requires our do-nothing representatives to actually do something, this analysis will focus on the other options: enforcement and regulation.
Being the big government leftist that I am, my first gut reaction would be to drown the IRS with more funding. Ensuring that our nation’s revenue collectors are well-funded shouldn’t be an issue in normal times, but then again, these are not normal times. As part of its general distaste for taxes and fiscal rectitude more generally, the Republican Party has long worked to freeze and even cut funding from the IRS; you might consider it the conservative way to “Defund the Police.”
In fairness, the IRS (like American law enforcement more generally) has long been haunted by allegations of political misconduct. Recent accusations center around Donald Trump’s tax returns and the infamous Obama-era Tea Party scandal, where newly created political nonprofits affiliated with the Tea Party faced negligent bureaucratic foot-dragging. IRS funding peaked in real terms around 2010, right after the scandal broke, and the budget hovered around that level afterward. The Biden administration tried to revitalize the agency with an $80 billion infusion over 10 years, but the Trump administration later rescinded it, and Congress followed with deeper cuts. For FY 2026, the IRS budget is expected to be about a third smaller (in real terms) than its 2010 peak.
But even this overstates the damage that years of fiscal depravity have inflicted on the IRS and the country as a whole. As a share of total government spending, the IRS has declined considerably since its last recorded peak in the mid-1990s.7 Headcount has also shrunk dramatically, with the IRS employing roughly 95,000 employees, down significantly from its 1988 peak of 123,000. Audits and investigations have plummeted over the same period. In 1995, the IRS audited roughly 1.67% of individual tax returns; in 2010, this fell to 0.90%, and in FY 2021, it bottomed out at a pitiful 0.38%. Given the long-term decline in both funding and manpower, it’s no wonder that IRS service quality has deteriorated in recent decades.
More Money, More (Tax) Problems
My case for revitalizing the IRS relies on a single premise: penal certainty. In other words, the best way to deter lawbreaking activity (in this case, not paying your taxes) is to increase the frequency at which tax cheats get caught, thereby incentivizing all the other would-be cheaters to follow the rules. The relative efficacy of penal certainty over severity (i.e., more severe punishments) is well-supported by existing social science research and is a lodestone of modern criminology: all I’m doing here is transferring this knowledge to the realm of public finance.
My plan for reforming the IRS entails three main approaches: the first focuses on tax enforcement, aiming to improve tax collection by increasing budget and staffing levels, the second requires strengthening regulations on tax filers and third-party preparers, and the third approach entails providing a free, public option for online tax preparation.
Any substantive plan to take on the tax gap must start with giving the tax police all the tools and manpower they need to enforce the law. We should restore IRS funding to its relative peak in the early 1990s at roughly 0.60% of gross federal revenue: in FY 2024, that would boost the IRS budget to roughly $29.5 billion.8 Obviously, a more realistic approach would be to adopt the IRA approach and allocate this additional funding over a longer time span.
Moreover, these funds should come with significant strings attached. The priority should be bolstering overall IRS headcount: in 1988, total agency employment peaked at roughly 123,000 employees processing 110 million individual tax returns, or roughly one employee for every 895 returns filed. In 2025, over 162.8 million individual tax returns were filed; restoring manpower to peak levels would require increasing the existing headcount from roughly 95,000 to about 182,000 employees.
The second priority would be to increase the frequency of audit rates. As mentioned previously, the number of audits performed by IRS agents has shrunk by roughly 80% since the mid-1990s: a revitalized IRS should be able to restore audit levels to their previous highs. Audits should be specifically targeted at both business and high-income individual returns: the bulk of unpaid tax revenue arises from these areas, thereby increasing the potential revenue boost from such targeted enforcement. Moreover, business and high-income tax returns are more likely to be processed through third-party tax-preparation entities, which decreases enforcement costs, as third-party preparers can provide more precise and accurate information about taxable income and the resulting liability.
Along with more agents and greater funding, the IRS must tighten regulations on income reporting and verification for tax filers and third-party tax preparers. To improve income reporting, the IRS should expand data collection from third-party filers (i.e., employer payroll records and third-party payment services like PayPal), which would better enable the agency to verify taxpayers’ income reporting and claims for various credits and deductions. The IRS should extend this to various domestic financial institutions and require tax filers to provide detailed information to justify their claims for deductions and credits. This must be done to cut down on erroneous claims of the various deductions and credits that plague our byzantine tax code: indeed, underreporting is the largest source of unpaid tax liability. The IRS should also require third-party preparers and commercial tax software to satisfy minimum competency requirements: nearly 53% of the 84 million tax returns filed via hired paid preparers lack adequate registration and/or testing by the IRS.
The final plank of my plan to revitalize the IRS entails undoing a rather specific policy error of the Trump administration: ending Direct-File. For those who don’t know, the United States is an outlier among developed countries in not offering a free public option for filing one’s taxes. Indeed, because of our complex tax code, taxpayers spend billions of dollars (and hours) every year filing their taxes, often hiring third-party preparers to handle the work electronically. Under the Biden administration, the IRS planned to roll out an online public filing option, which was subsequently revoked once Trump retook office. This error reinforces the existing oligopoly on tax preparation, dominated by major tax preparers like TurboTax and H&R Block. By restoring Direct File as a public option, I seek to reintroduce greater competition in the tax-preparation industry and minimize the time and resources taxpayers spend filing their returns. While this individual policy would only have a negligible effect on revenue collection, it would hopefully work to improve the service quality of the IRS and ease the overall process of filing taxes, which is a major point of stress for American taxpayers as well as a major source of dissatisfaction with the IRS (and the federal government more generally).
Part IV: Estimating the Savings
As crazy as it may sound, creating the proposals was the easy part. Now comes the hard part: figuring out if improving public-sector efficiency can seriously help resolve our looming fiscal challenges.
GO GO GAO Savings
Estimating the potential savings from tackling inefficiency in the federal government is no small task. But first, a few clarifications are in order. The Boyd Institute estimates that eliminating fraudulent and improper payments would still leave the country approximately $1.1 trillion in the red. Yet as I illustrated in Part I, eliminating the deficit is not (and should not be) our goal; rather, the goal should be to shrink the national debt relative to the broader economy. This entails getting the budget deficit below 3% of GDP, and according to the Boyd Institute’s own figures, eliminating waste, fraud, and improper payments more than amply achieves that goal.
However, no serious person believes the federal government can actually achieve this. Waste and improper payments are undoubtedly bad, but they’re also inevitable: inefficiency is inherent to large entities, even in the highly competitive private sector. Eliminating waste and improper payments is unrealistic. In this sense, the Boyd Institute is correct: waste and fraud are not and cannot be our only source of deficit reduction, even if we dispense with the dubious assumption that balancing the budget is a vital imperative. That said, there’s no reason to believe that the fiscal impact of tackling inefficiency is negligible. Since 2002, the GAO has helped secure roughly $1.5 trillion in savings due to its recommendations. And these are only for recommendations that are eventually implemented: they do not include recommendations that lawmakers have let fall by the wayside.
Since 2003, the federal government has incurred nearly $3 trillion in improper payments, averaging $136 billion annually. In FY 2025, the federal government estimated that improper payments constitute roughly 2.7% of current outlays, up from 2.1% in 2004.9 Returning to 2004 levels of improper payments would save the government roughly $42 billion using FY 2025 figures: over ten years (and making some highly unrealistic simplifying assumptions), that amounts to savings of roughly $420 billion (or 0.1% of GDP). Assuming we could go further and limit improper payments to an aggressive (but by no means unachievable) 1.5% of total outlays, we could reduce the ten-year deficit by roughly $840 billion, or 0.2% of GDP. According to the center-right Economic Policy Innovation Center, implementing all GAO proposals (including those involving improper payments) could generate a median annual savings of $151 billion, or $1.51 trillion over ten years. The Committee for a Responsible Federal Budget has estimated that the U.S. would need to reduce deficits by roughly $10 trillion over 10 years to achieve the 3% deficit target needed to stabilize overall debt-to-GDP levels. Full implementation of these GAO reforms would get us roughly one-seventh of the way there.
Shrinking the Tax Gap
As with spending-side reforms, it’s tricky to precisely calculate the reasonably expected savings from boosting revenue collection. In her ambitious deficit-reduction proposal, the Manhattan Institute’s Jessica Riedl estimated that refunding the IRS as per the IRA would yield savings of roughly 0.07% of GDP, or a paltry $200 billion over ten years; more optimistic predictions estimate savings as high as $700 billion over a similar timeframe.10 Of course, my proposal is far more expansive than merely reimplementing the IRA, so my estimations are likely to be even more speculative. Ultimately, the overall fiscal impact of better revenue collection hinges on whether these enforcement policies will shift the tax compliance rate, either by improving voluntary compliance, boosting the revenues generated by enforcement, or some combination thereof.
For simplicity’s sake, I will extrapolate from the tax figures gleaned from FY 2022, while also making some simplified assumptions about how my policies would move the needle on overall revenue collection. Assuming we can boost net tax compliance to an unprecedented 93% of total tax liability (compared with the recorded 86.9%), the federal deficit in FY 2022 would have shrunk by roughly 1.10% of GDP, or about $281 billion (which translates to $2.81 trillion over ten years). By contrast, a more pessimistic goal of 88% compliance would generate a more modest 0.20% of GDP, or roughly $51 billion ($510 billion over 10 years). The middling-scenario, which envisions net tax compliance reaching roughly 90% of total tax liability, would have reduced the deficit by 0.56% of GDP, or roughly $140 billion ($1.4 trillion over ten years). Moreover, these likely understate the actual savings achieved, since they don’t include foregone interest payments on the debt that would have been issued absent these reforms.
Of course, these are highly imprecise guesstimates achieved through back-of-the-envelope arithmetic. There are a couple of unknown factors that would push the total level of savings in either direction. Optimistically, one must take into account the issue of growth: these numbers assume that GDP will be frozen (in real terms) from FY 2022 onward: yet even the most conservative estimates forecast growth over the next ten years, and given that tax revenues rise in some proportion to overall growth (assuming no more budget-busting tax cuts), the actual savings achieved under either of these scenarios is likely to be higher. More pessimistically, one could argue that tighter enforcement and greater revenue collection will ultimately have negative second-order effects on economic growth, or that taxpayers will find new ways to evade liability, thereby reducing future expected revenues. I plan to address these issues in the final section of this paper, but I recognize that they are important questions worth dwelling on further.
Taken together, my efficiency-centered approach to fiscal consolidation has the potential to reduce the ten-year deficit by between $2 and $4 trillion (let’s just split the difference and say $3 trillion), getting us roughly one-third of the way towards our target of 3%. While by no means sufficient, this actionable program significantly improves the nation’s fiscal outlook and lays the groundwork for more ambitious reforms to stabilize the nation’s long-term finances.
Part V: Addressing the Critics
Of course, I’d be doing my beloved readers a disservice if I didn’t engage in a round of rhetorical struggle sessions. Here, I’m going to borrow from the work of the intrepid economist Albert O. Hirschman and his best-selling book, The Rhetoric of Reaction. Despite being a man of the Left, Hirschman’s experience in the great ideological conflicts of the twentieth century and his professional commitment to scholarly rigor compelled him to engage his many interlocutors in good faith. He observed a consistent pattern in traditional arguments about new social reforms that are generalizable to novel proposals, regardless of their ideological valence. Indeed, such arguments are only truly “reactionary” in the narrowly technical sense of the term.11 Hirschman organized these argumentative approaches into three broad categories that I will shamelessly present out of order: futility, perversity, and jeopardy. I will thus structure potential criticisms of my efficiency-based plan along these lines.
Futility
The futility thesis is the most intuitive of the three categories and the easiest to fall into. Futility-based arguments assert that any novel reform will fail to achieve its desired goal, whatever that may be. In his work, Hirschman illustrates this concept by reviewing historical arguments against various progressive reforms, such as the welfare state, and observing how reactionary critics argued that such programs have little, if any, impact on poverty or economic deprivation.
Here, this argumentative style can assume various forms, not all of which are necessarily ideological. J.K. Lundblad makes several futility-based arguments in his article on deficit reduction, where he ridicules the idea that greater tax revenue can help close the budget gap. He bases this position on two arguments: the Laffer Curve and his tendentious belief that, even if more revenue were raised, the government “will always expand to use the revenue available to it, and then some.” Related to this argument (though left unmentioned by Lundblad) is the hypothesis of “Hauser’s Law,” which holds that federal revenues to GDP have averaged around 18% since the end of World War II. Many commentators use this observation alongside the Laffer curve to claim that raising taxes to fix the deficit is worse than useless: the idea being that, with federal revenues relatively fixed at 18% of GDP, the existing tax regime is either at or dangerously close to the revenue-maximizing peak.
Normally, I hate to single out a single person for ridicule, especially given how easy it is to criticize and how hard it is to create. But neither Lundblad’s assertions nor the hypothetical invocations of Hauser’s Law pass the sniff test. Lundblad spends some time articulating the technical parts of the Laffer Curve, which is the rather banal observation that, at some level, higher taxes reduce overall revenue by stifling economic growth. Left unsaid is where precisely the current tax code is on that curve: Lundblad makes no clear assertion either way, but his subsequent arguments (which will be dealt with in turn) suggest that he believes we’re either at or on the wrong side of the optimal point on the curve. One way to test this is to see whether previous rounds of tax reduction have resulted in higher tax revenues: after all, the Laffer Curve presumes that, in some rare cases, tax cuts can actually pay for themselves. The empirical evidence doesn’t really bear this out, at least at the federal level: various studies have shown that tax cuts generally depress revenue relative to what it would otherwise have been, thereby raising the deficit (or diluting the surplus).
The other arguments are even less impressive. Lundblad’s bald assertion that governments will simply spend any additional revenue they may raise sounds like common sense, but it is actually false. Governments are quite capable of exercising fiscal restraint under certain circumstances, as evidenced by a deficit reduction program implemented across the Reagan, George H.W. Bush, and Clinton administrations in the late 1980s and early 1990s. Meanwhile, Hauser’s Law is completely unsupportable as a concrete economic proposition: federal revenue has varied considerably over the past seventy years, in no small part due to substantial changes in the structure of the tax code. Evidence from other developed nations suggests that there is no structural reason to believe that federal receipts are forever stuck at 18% of GDP: the more likely explanation, as argued by arch-conservative economist Daniel J. Mitchell, is the absence of a general consumption tax and the broadly federalist division of sovereignty.12 Finally, Hauser’s Law fails to account for the downward trend in effective tax rates since the 1980s, especially on middle and low-income households.13 The federal income tax is highly progressive, which ultimately limits how much revenue it can raise, given that the rich only account for a small share of overall income. While it is undeniable that higher taxes entail distinct tradeoffs in the form of slower growth, it is moronic to argue that they have no practical effect on closing budget gaps.
Peter Banks advances a more sophisticated version of the futility thesis, arguing that efficiency-based reforms would be insufficient to balance the federal budget. This is true even if we also closed the tax gap on top of these spending-side reforms and used all the accumulated savings to balance the books: it’s even more true when you adopt a more realistic picture of how much can be achieved through efficiency reforms and improved tax collection. I concede as much in my argument above: efficiency-oriented reforms, by themselves, are no substitute for a broader plan for fiscal consolidation.
That said, there is no reason to discount the value in using efficiency as a first step towards broader deficit reduction. As mentioned above, we don’t actually have to balance the budget: all we really need to do is stabilize (and later reduce) debt levels relative to GDP, which only requires us to keep deficits below the rate of nominal economic growth, with inflation and growth doing the rest of the work for us. Deficit reduction is not an all-or-nothing proposition: to get the nation’s fiscal house in order, we must rely on several different approaches.
The reforms I have outlined here have the potential to get us roughly one-third of the way towards our target deficit of 3% of GDP, and it’s worth reiterating that these savings are relatively painless. They require no growth-stifling tax increases or drastic cuts to existing benefits; if anything, they will likely increase the quality and value of federal services as experienced by ordinary people. They also have the potential to build public trust and foster a collaborative political culture, enabling tougher choices down the line. This is especially important given how difficult fiscal consolidation is in practice: The Mont Pelerin Review argues as much in their piece on workable austerity: too often, governments rely excessively on tax hikes and cuts to public investment because those are politically easier to stomach than cuts to public benefits. This only reinforces my point: deficit reduction is a tricky issue to manage even in good times (and these are not good times); this places a high premium on positive-sum changes (like the reforms outlined above) that can move the needle in the right direction.
Perversity
In contrast to the futility thesis, perversity-based arguments are a bit less intuitive. Perversity-based arguments go beyond futility and claim that any reform under debate will ultimately backfire, exacerbating the very problem it was meant to amend. In other words, the perversity thesis invokes the specter of unintended consequences, with which the road to hell is duly paved. Hirschman discussed how similar arguments were used in debates over universal suffrage and its relationship with political liberty. The more primitive-minded critics insisted that reformist pieties like “one man, one vote” constituted an attack on the freedom of existing voters, mainly propertied elites, who were outnumbered by the disenfranchised masses. This crude invocation of majoritarian tyranny was often elaborated by more sophisticated reactionaries, who argued that mass democracy would subject the government to the fickleness of the mob, which could easily elect a tyrannical demagogue and thus undermine the very political liberties that pro-suffrage advocates rightly cherished.
Here, sophisticated adherents to the perversity thesis would argue that relying too heavily on efficiency-based savings without dealing with structural policy defects merely kicks the can down the road. Lundblad indirectly makes this point in his critique of relying on additional tax revenue to plug the deficit: drawing on his earlier government-as-a-drunken-sailor metaphor, he cites a 1996 paper by Richard Rahn to assert that economic growth declines once government spending exceeds 20-25% of GDP. He also alludes to an infamous 2010 paper by Carmen Reinhart and Kenneth Rogoff that claims that public debt above 90% of GDP directly reduces overall economic growth. Concluding that governments “try[ing] to tax their way to fiscal solvency never can,” Lundblad’s overarching point is that tax increases fuel the structural growth of government spending, thus negating any potential effect higher revenues might have on the deficit. Moreover, additional government spending is presumed to have a net negative effect on the economy due to the implications of the Rahn curve, so we face the dual issues of an unresolved deficit problem and a steadily eroding tax base. Mon dieu!
As you can probably tell, I’m not convinced. Putting aside the (un)timeliness of the Rahn study, there’s little agreement among economists as to whether the underlying point (that spending more than 25% of GDP directly undermines growth) is actually true. While the general concept of diminishing returns to government spending is virtually unassailable, most economists doubt that there is a precise moment when such spending degrades growth, let alone that 25% of GDP is the so-called “sweet spot.” Moreover, this conflates the connection between government spending and GDP: it could just as easily be the case that slowing economic growth swells government spending in relative terms, either due to higher unemployment levels or a shrinking tax base. Things get even worse when you get to the Reinhart & Rogoff study. The 2010 paper was widely panned shortly after publication for methodological issues, data-coding errors, and a broader conflation of correlation with causation. The data errors, in particular, are instructive: once those errors are corrected, the actual effects of debt on economic growth at the 90% level shrink dramatically (though they remain negative). Economists generally agree that high debt levels have a deleterious impact on economic growth, but the flawed Reinhart & Rogoff study overstates the level of consensus around how and when those impacts actually occur: unfortunately, various European policymakers relied on the paper’s findings to justify their austere policies after the Great Recession, to their ultimate detriment.14
Having done my best to pigeonhole Lundblad’s arguments, I think it’s time to make a concession: there is an underlying logic to Lundblad’s arguments, however poorly made. Quick fixes (like boosting tax enforcement) might help reduce budget gaps in the near term. Still, they leave unaddressed broader issues facing the economy, like runaway healthcare spending and a worsening worker-dependency ratio. This requires a tough look at the American economy, its tax structure, and its welfare state. More broadly, the second-order effects of improving government service and structure warrant brief reflection. While there are various historical reasons why the United States has a smaller welfare state than other peer nations, one understated reason is the piss-poor state capacity at all levels of government. We’re talking about a nation that overspends on virtually every public good, from healthcare to transportation, and designs its public benefit programs in such an opaque manner as to make the whole application process an ordeal. It’s no wonder American voters don’t trust the government with more of their hard-earned money. It’s also perfectly conceivable that if the federal government put more effort into improving the efficiency and quality of public services, the broader American public would be more amenable to big government programs (more than they already are, at any rate).
For a big government leftist such as myself, this is no issue: tax and spend is my political mantra, so to speak. But for small government-minded folks on the Right, this is a real issue, to which I have two responses. The first is simple: questions about government size and fiscal consolidation are two separate debates and should be handled as such. Tying the two together only accelerates partisan polarization and gridlock, as evidenced by the past quarter-century of hemming and hawing about the national deficit. Here, conservatives (and critics of big government more generally) should compromise on their opposition to big government to avert a much greater catastrophe: insolvency. Second, I’d remind them of the immortal words of Henry David Thoreau, the lauded antebellum writer/activist, who argued that the best response to bad government is not no government, but better government. Getting conservative politicians to make the government worse or forestall improvement doesn’t improve overall fiscal outcomes, and it sure as hell doesn’t improve people’s opinions of conservative politics (at least, it didn’t for me). By improving the quality and efficiency of government services, conservatives may regain some of the political legitimacy they’ve lost over the past few decades, giving them a new lease on life in the post-Trump era.
Jeopardy
In my opinion, the jeopardy thesis is the most interesting of the three categories. Jeopardy-based arguments presume, for the sake of argument, that a policy will perform as advertised by its proponents. However, the jeopardy-minded interlocutor contends that the successful implementation of the policy will ultimately endanger another goal or normative value. In other words, the jeopardy thesis reminds us that there are tradeoffs to every decision, a reality that all policy advocates must consider. Any economist with half a brain knows this. Here, the jeopardy thesis has its greatest purchase in challenging my outlined reforms on technical grounds, which should be a relief to anyone who has grown tired of abstract arguments about tax elasticity curves and the normative merits of government intervention in society.
The sophisticated jeopardy-minded critic would contend that my plan envisions giving more money and political authority to two administrative agencies: the GAO and the IRS. And the clout I plan on giving them is not negligible: indeed, it may implicate our very constitutional framework, which I will discuss below. In other words, I’m proposing to expand the power and reach of the administrative state, the very “swamp” that President Trump once promised to drain (how times change!). Cheeky jowls aside, there are serious, legitimate concerns about such a plan and its effects on the delicate division of power within the federal government.
Let’s take my beefed-up IRS plan, for starters. Beyond merely showering it with more money, my plan entails granting the agency greater capacity and legal authority to conduct audits, regulate tax preparers, and process sensitive financial information for hundreds of millions of taxpayers nationwide. Granting this power to any agency would be hard to digest, but the IRS is not just another run-of-the-mill regulatory agency. As mentioned before, the IRS has a long and infamous history of corruption, misconduct, and selective enforcement at the behest of powerful interests. The Tea Party scandal in the early 2010s was unsavory enough that Congress froze the IRS’s budget for over a decade, potentially costing the country tens (if not hundreds) of billions in uncollected revenue. And the Tea Party scandal was mild compared to the more notorious cases of IRS misconduct, which largely date back to the early twentieth century. Notable instances of IRS-enabled skullduggery include: FDR’s surreptitious investigation of potential rival Huey Long, the Kennedy administration’s targeting of tax-exempt right-wing foundations, the infamous COINTELPRO program deployed against leftists and civil rights groups, not to mention the long list of quasi-authoritarian overreaches that proliferated under the Nixon administration. Even more recently, there were serious concerns that the Clinton administration used the IRS to hound Paula Jones and Juanita Broaddrick, who both accused Clinton of sexual harassment. Suffice it to say that the IRS has long been used as a vehicle for partisan hackery, and liberty-minded citizens are justified in questioning its expansion.
By contrast, my plans for GAO are relatively tame. As a congressional watchdog, GAO has no actual authority to enforce its recommendations: the best it can do is communicate these in various reports to Congress and the White House, which it does ad nauseam. By shifting towards an “opt-out” approach to GAO reform, one could argue that I am subverting the separation of powers by placing the GAO above the duly appointed departmental heads and the elected members of Congress. Indeed, part of the proposal entails fast-tracking GAO recommendations for a straight up-and-down vote before both Houses and creating a compliance-ratings scheme for departmental executives, bypassing the minutia of congressional procedure and giving the congressional watchdog an aura of authority possessed by few other government agencies. Overnight, I would turn the GAO into one of the biggest power centers in Washington, D.C., which raises important questions about the balance of power among the various branches of government.
In response to these valid concerns, I have two rebuttals. The first is rather technical: if people have a genuine problem with extending so much power to GAO and the IRS, my proposed changes can be modified to alleviate these concerns. For the GAO, we’ve already given departmental officials the power to rebut or challenge GAO recommendations within a generous timeframe, while giving both Congress and the White House the authority to override GAO recommendations, thereby ensuring that the congressional watchdog remains subordinated to the broader system of checks and balances. IRS reform is a bit trickier: misconduct here is largely the result of pressure from elsewhere in the executive branch, with Congress often belatedly responding by withholding funding and thus degrading the agency’s long-term enforcement capacity. Having said that, I think it’s fairly obvious that the first step on the path to fiscal consolidation is not undermining the nation’s chief revenue collection agency.
Thankfully, this leads me to my second (and final) rebuttal, which I hope will cap off this behemoth of an article: time is running out. I’ve said it before, and I’ll say it again: I find the word “crisis” to be heavily overused, especially in relation to the national debt. But the fiscal problems we face are not insignificant, and they are becoming increasingly troublesome the longer we let them fester. Decades of fiscal extravagance and cowardice have forced us into a position where we must make tough choices, including those about our structure of government. The time to resolve our fiscal problems within the current balance of governmental power has long passed: to borrow from our tech overlords in Silicon Valley, we need to move fast and break things to get this problem under control. That means everybody has to give something up, if only for the greater good of the republic: liberals must swallow cuts to popular programs, conservatives need to quit defunding the IRS, and hard-bitten institutionalists must come to terms with the fact that fiscal irresponsibility is a feature, not a bug, of our gridlocked kludgeocracy. The time for wailing and gnashing of teeth is long past: now is the time for action.
As you can see from this graph, the United States didn’t really start running chronic and unmanageable deficits until the Reagan administration, which tried mightily to “starve the beast” yet chickened out when the time came to cut spending. Après moi, le déluge.
According to Demand Progress, the GAO budget peaked in real terms in 1992 at a little over $800 million; in FY 2021 it was $661 million, a budget cut of roughly 13%. In relative terms, GAO outlays also peaked in 1992 at roughly 0.082% of total discretionary spending, nearly twice its 2021 level at 0.045%. DANIEL SCHUMAN, A NEW FUNDING MODEL FOR THE GAO AND FINANCIAL RELIEF FOR CONGRESS (2021), https://s3.amazonaws.com/demandprogress/reports/A_new_funding_model_for_ GAO_--_October_2021.pdf.
A year’s worth of GAO savings from removing redundant programs amounts to roughly $55.4 billion. In FY 2024, the JCT estimated that eliminating the charitable interest deduction would raise revenues by roughly $55.4 billion. What are the largest tax expenditures?, TAX POLICY CENTER, (2024) https://taxpolicycenter.org/briefingbook/ what-are-largest-tax-expenditures.
Here I’m mostly talking about tax prep companies like TurboTax and H&R Block. Anna Massoglia, Tax prep companies that spent $90 million lobbying against free tax-filing system face new scrutiny from lawmakers, OPENSECRETS (Sept. 1, 2023), https://www.open secrets.org/news/2023/09/tax-prep-companies-lobbying-against-free-file-face-scrutiny-from-lawmakers/.
Experts often compare the IRS budget as a share of gross revenue to measure the “cost of collecting $100.” According to the Tax Foundation, the cost of collecting peaked at 0.60% in 1993. Scott Hedge, IRS is Doing More with Less, But New Funding Misses the Mark, TAX FOUNDATION (Aug. 29, 2022), https://taxfoundation.org/blog/irs-budget-increase-technology/.
In FY 2024, total federal revenue amounted to about $4.92 trillion; 0.60% of that is roughly $29.52 billion. FEDERAL BUDGET IN PICTURES, https://www.federal budgetinpictures.com/federal-revenue-vs-spending/.
There were at least $186 billion in improper payments made in 2025 out of a total expenditure of roughly $7.01 trillion, which amounts to 2.65% of total outlays. In 2000, improper payments accounted for roughly $46 billion out of a total budget of $2.23 trillion, which amounted to 2.06%.
Ultimately, the size of the estimated savings depends on how effectively the additional funds are used and how taxpayers react to improved enforcement. Taxpayers may respond by either voluntarily increasing compliance, spending more resources to hide existing income, shifting their behavior to reduce their taxable liability, or some combination thereof.
Encyclopedia Britannica defines “reactionary” as “a person strongly opposed to new political or social ideas.”
More specifically, the subnational governments (i.e., states and municipalities) of the United States have far greater power to raise revenue through taxation and fees than in other nations, even those with similar federal constitutions. Additionally, the United States has fewer mechanisms for revenue sharing between state and federal governments than other countries have formally integrated into their constitutional frameworks.
As you can see here, effective tax rates have generally declined for virtually all income levels over the past half-century. Effective tax rates for the bottom 20% have dropped from 9.3% in 1979 to 0.6% in 2019, while for the top 20% they have declined gradually from 27.1% to 24.3% over the same period. The top 1% saw its effective tax rate decline from 35.1% in 1979 to 29.9% in 2019.
Over the past few months, there’s been a prodigious discussion about Europe’s lackluster economic performance following the financial crisis, especially when compared to the United States. There are many factors behind Europe's sluggishness in recent years, including political instability and high fuel prices, but one underrated factor is the brutal fiscal consolidation that occurred during the sovereign debt crisis, which even right-leaning commentators, such as The Mont Pelerin Review, admit contributed to Europe’s relative stagnation.