Principles for the 2025 Tax Debate

Introduction

At the end of 2025, all the individual income and estate tax cuts enacted in the 2017 Tax Cuts and Jobs Act are due to expire. This will prompt policymakers to revisit the 2017 tax cuts, which are equivalent to about 1 percent of GDP, or about $350 billion per year over the next 10 years if the tax cuts remain in full effect, as well as the broader tax code. The debate about how to approach this challenge has already begun.

This article proposes principles for evaluating the many reform proposals that will be put forward to address the expiring tax cuts and more broadly reform the tax code. It does not offer a single proposal for doing so — there are a variety of paths consistent with the principles laid out here, and policymakers themselves will be offering their own packages. Instead, we identify core principles and sketch some of the paths for policymakers to achieve them.

We begin by providing context for the 2025 negotiation: Policymakers cannot view the 2025 “fiscal cliff” as an isolated event. It is the latest milestone in a multidecade strategy featuring large and initially expiring tax cuts that significantly reduce the U.S. revenue base and do so in a way that widens income disparities. The winners under this strategy have disproportionately been higher-income taxpayers. The cost has been to our current and future capacity to raise revenue to support the country’s needs. Ignoring this history — or worse, accepting that temporary tax cuts, once enacted, will inevitably be extended — will only serve to vindicate this approach and threaten more erosion of our tax system.

With this historical context, any tax proposal should abide by several key principles:

1. at a minimum, fully pay for tax reform relative to current law and strengthen the tax base to facilitate future increases in revenue;

2. focus the existing limited fiscal space on helping lower-income families;

3. reverse the erosion of the corporate income tax system while maintaining incentives to innovate and invest;

4. eliminate tax cuts for the highest-income Americans and increase their contributions relative to the pre-2017 tax system; and

5. improve tax administration and simplify tax filing.

Background

The dominant feature of tax policy in the two decades from 2001 to 2021 was a series of large tax cuts that undermined the U.S. revenue base and made the distribution of post-tax income less equal.

The magnitude of this effort is large: roughly 3 percent of GDP if the 2017 tax cuts are continued in full, or approaching an average of $1 trillion per year over the coming decade.¹ To put this in context, primary deficits — deficits reflecting current programmatic spending and revenue, and excluding interest payments — are projected to average just under 3.5 percent of GDP over the next decade if the tax cuts are continued.² That is not much larger than the revenue lost in these two rounds of tax cuts if they are maintained. And given that economic growth helps offset the effects of interest payments on the debt, current projections would not show a significant rise in the debt-to-GDP ratio across the coming decade if these tax cuts had not occurred (see Figure 1), and the long-term fiscal shortfall would be a small fraction of its projected size, if not eliminated.³

The federal government has significant capacity to borrow, but over time, that borrowing will need to be put on a sustainable trajectory, which will mean stabilizing the debt-to-GDP ratio and, relatedly, interest payments as a share of the economy and the federal budget. We can have a tax system that does so and finances the government’s current obligations and beyond. The tax system that existed in 2000 was roughly that. Over the past two decades, we have chosen a different path.

The tax cuts came in two rounds. The first round was enacted in several bills under President George W. Bush, the largest of which were in 2001 and 2003 and gave large individual and estate tax cuts. The 2001 and 2003 tax cuts were made entirely temporary, set to expire by 2010, as a way to lower the headline cost and avoid budget enforcement rules that prevented using the budget reconciliation process to increase the deficit over the long term.⁴ However, proponents explicitly supported continuing the tax cuts. Even as the 2001 legislation was on its way to the president’s desk, White House Press Secretary Ari Fleischer said that President Bush expected the tax cuts to be permanent. Fleischer described the president’s position as “to do anything other than that is to raise taxes on the American people, and he does not support that.”⁵ President Bush included the permanent extension in his subsequent budget proposal, and tax cut supporters intended to set the agenda for the coming decade. As Grover Norquist said then, “If the Democrats were swept into power in five or 10 years, they can do what they want. But they’d have to change the whole debate.”⁶

Consistent with that strategy, the resolution of these temporary tax cuts became the centerpiece to fiscal debates in the ensuing years. The tax cuts were continued temporarily for two years until their fate was resolved in a bipartisan negotiation in 2012 between President Obama and Republican and Democratic congressional leaders. As a result, about 20 percent of the tax cuts were allowed to expire. However, about 80 percent of them were continued — equaling over 1.5 percent of GDP, or in the rough range of $600 billion per year over the next 10 years.

While a substantial share of the tax cuts expired for the top 1 percent, the top more broadly, and especially the 95th to 99th percentiles, ended up the largest winners in the deal, even as the immediate benefits extended down the income distribution. In other words, tax cuts disproportionately benefiting the top were enacted, and although the effort to claw back the tax cuts managed to reduce them significantly for the top 1 percent, the top more broadly defined still ended up as disproportionate winners.

Table 1 shows the cost and distribution of the 2000s tax cuts as made permanent in 2013 — and the share of the tax cuts retained — based on Urban-Brookings Tax Policy Center’s analysis at the time. Note that this includes expansions in refundable tax credits benefiting the lowest-income Americans, which were enacted in the 2009 stimulus legislation under President Obama, thus expanding the benefits to those at the bottom of the income spectrum.

The second round of tax cuts came in 2017 under President Trump. This legislation combined individual, estate, and corporate tax cuts. As with the 2001 and 2003 tax cuts, there was an explicit expiration strategy — setting tax cuts to expire to lower the headline cost and abide by reconciliation rules. As Budget Director Mick Mulvaney said at the time in justifying the expirations, “One of the ways to game the system is to make things expire. The Bush tax cuts back in early 2000 did the same thing.”⁷

All the individual income and estate tax cuts expire after 2025, setting up the focus on that year for tax reform. By contrast, major elements of the corporate tax cuts, such as the 14 percentage point cut in the corporate rate, are permanent. Given that the reconciliation rules require permanent measures to be paid for, taxwriters offset the permanent tax cuts, paying for them in the 2017 law by changing to a slower inflation adjustment for the individual income tax system and reducing health coverage by eliminating the individual mandate. To lower the headline cost and to reduce the scored cost of the permanent corporate tax cuts, the legislation also contained expirations of parts of the business tax cuts and future-year business tax increases, some of which are already taking effect.

The 2017 tax cuts, if fully continued, amount to around 1 percent of GDP, or around $350 billion per year over the coming decade.⁸ They, too, were heavily skewed toward the top of the income distribution, with the largest winners again in the 95th to 99th percentiles and the top 1 percent also disproportionately benefiting. This is even as those toward the bottom of the income distribution get very little benefit, and as explained later, simply continuing these tax cuts can turn into an eventual tax increase for some low- to middle-income families.

Table 2 shows the cost and distribution of the 2017 tax cuts when fully in effect in 2018. Note that some elements of the business tax cuts — especially changes in expensing — involve timing shifts, so the 2018 cost is somewhat higher than the long-term average for these tax cuts, which is closer to 1 percent of GDP than the 1.4 percent of GDP reflected there.

Across this period, Democrats also enacted temporary tax cuts — though in much smaller magnitude and with very different distributional effects. The 2009 stimulus plan included refundable tax credits that the Obama administration sought to make permanent. In the end, modest expansions of the earned income tax credit and child tax credit for lower-income families were extended in the 2012 tax deal and eventually made permanent. The American Rescue Plan Act of 2021 included expansions in the child tax credit and Affordable Care Act premium tax credits, which the Biden administration has sought to extend. The child tax credit expansion — on the order of $100 billion per year — expired after 2021 despite efforts to continue it, and the ACA premium credit expansions — on the order of $30 billion per year — are scheduled to expire after 2025, having been extended in the Inflation Reduction Act of 2022. As a result, these expanded tax credits for low- to middle-income families will be on the agenda as the broader 2025 expirations approach.

The tax system is defined not just by the laws on the books but also by how those laws are administered and enforced. And the 2010s saw a sharp reduction in funding for the IRS and a resulting deterioration in both service and enforcement, further undercutting the revenue base. From 2010 to 2021, the IRS’s budget adjusted for inflation fell by about one-fifth.⁹ These cuts plus the effects of the pandemic then produced a stunning collapse in service, including a sharp rise in unanswered phone calls and backlogs of unprocessed returns. At the same time, audit rates fell across the board, especially for those with the highest incomes. From 2010 to 2019, audit rates for millionaires fell by 70 percent,¹⁰ and evidence suggests considerable revenue loss as a result. Using administrative records, researchers have recently calculated that every $1 spent on audits above the 90th percentile generates $12 in revenue.¹¹

Since 2021, the Biden administration and Congress have pursued tax changes to address elements of this two-decade erosion. The IRA makes a historic investment in IRS enforcement and service. When it comes to tax avoidance by the largest corporations, the Biden administration successfully concluded negotiations with the OECD on a global minimum tax system that would help to reverse the erosion of the U.S. corporate tax base if the United States moves forward with implementation. The IRA also included a new 15 percent minimum tax on corporations based on their book income, which sought to reclaim some of the corporate revenue lost to the 2017 tax cut. However, all these measures are politically contested and will be in dispute in the debate around the 2025 tax cut extensions. As has been the case since 2001, American tax policy for the next two years will once again be framed by an expiration fight around large, regressive tax cuts that were made temporary precisely under the theory that they would then be hard to alter or eliminate.

Principles for 2025

This background underscores that policymakers should not view the 2025 “cliff” in isolation. It is the next stage in a multidecade strategy of large, expiring tax cuts that have undercut the revenue base and disproportionately benefited those with greater resources. Reorienting the trajectory of tax policy toward more sustainable long-term finances and a fairer distribution of resources will not happen all at once. However, progress along several key dimensions is important and achievable. This article lays out five specific principles to guide tax policymaking looking toward 2025.

1. Pay for reform and strengthen the tax base.

The tax system today is insufficient for the government we have. As we have described, this is largely a result of policy choices made over the past two decades; it was not inevitable. A minimum goal for reform in 2025 should be a bill that is fully paid for and strengthens the tax base — making it harder to engage in wasteful activity to avoid paying taxes, which can then facilitate raising additional revenue in the future, as will likely be needed. This is a “pay as you go” principle, or what we might call “pay-go plus,” combining a bill that should be fully paid for and should also leave our tax code with a stronger tax base that will serve us in the years ahead.

This pay-go-plus principle is important not only for setting our future fiscal path but also for setting precedent. It would signal a commitment to change course from the past two decades’ strategy of enacting large, temporary, and regressive tax cuts with the expectation that most will be continued without being paid for. A pay-go-plus principle also helps establish policy discipline — the more revenue raisers that policymakers are willing to accept, the broader and deeper the tax cuts they can offer. The fewer revenue raisers they are prepared to enact, the more constrained any continued tax relief should be. Any policymaker or candidate offering a 2025 tax proposal should have to answer this question first: Is the proposal at least fully paid for, in a way that leaves our tax code with a stronger tax base — a tax base that is fairer and involves less wasteful tax avoidance — for future generations?

Adhering to pay-go-plus in the context of the 2025 tax debate would also be an important step forward in positioning our country to answer the ultimate question of what the tax system should look like over the long term. To finance our existing government commitments and stabilize debt and interest on that debt as a share of the economy, that system would likely need to raise more revenue — on the order of 2 to 3 percent of GDP over the long term — than would be the case even if reform in 2025 is fully paid for.¹² This goes to the “plus” part of pay-go-plus. Just as it is harder to fill a leaky bucket, it is harder to raise revenue efficiently and fairly within a leaky tax system that encourages wasteful tax avoidance behavior — some legal and some not. Reforms that plug the avoidance holes in our tax system and improve tax administration will help set the stage for future policymakers to more easily address the long-term fiscal shortfall and also help restore public trust in the tax system itself.

How to address.

Several leaders have been willing to consider the types of measures that could fully pay for tax reform in 2025 and strengthen the tax base. In 2021 and 2022 Congress developed a list of revenue-raising measures as various iterations of what eventually became the IRA were considered. As the investment side of that reconciliation bill got cut down in negotiations, the revenue envelope was reduced as well. In 2025 that revenue envelope should be reopened. President Biden’s budget goes in that direction, incorporating ideas considered in the previous legislative efforts, and others too. Table 3 gives a basic breakdown of the revenue sources in the president’s budget, adding up to about 1.5 percent of GDP.¹³

Of course, the president’s budget, and Democrats in the context of the recent reconciliation bill, used that revenue for some combination of financing new investments and deficit reduction — not to offset any extension of the 2017 tax cuts, which were assumed to be financed on a revenue-neutral basis and with offsets yet to be specified. This again goes to the “plus” element of the pay-go-plus principle: More revenue will be needed in the future, even if tax reform in 2025 is fully financed.

The sections that follow will discuss both revenue-raising corporate tax reform and increasing taxes on the high-income individuals in greater detail.

2. Focus limited fiscal space on helping lower-income families.

The 2017 tax law provided little to no benefits to the lowest-income families. In fact, it was designed in a way that will increase taxes on some lower-income taxpayers who started with a tax cut if the law is simply extended in its current form.

Policymakers in 2025 should flip this script. A second principle for reform should be that any new tax benefits relative to the 2017 law should be focused on lower-income families. Those with modest incomes should not be left with paltry tax cuts — or, for some, tax increases — as a result of the changes enacted in 2017 and the changes to come in 2025.

The modest and, in some cases, shrinking benefits from the 2017 law result from several structural choices.

Almost no expansion in refundability for the lowest-income families.

The 2017 law did almost nothing to give benefits to the lowest-income Americans. Given that these families do not have positive income tax liabilities, even as they pay other taxes both at the federal and state levels, the way to help them is to expand refundable tax credits. Refundability refers to the amount of the credit available for those without offsetting income tax liability, and the 2017 law did very little when it comes to refundability for those with the lowest incomes.

For the lowest-income Americans, the 2017 law reduced the threshold at which the refundable child tax credit begins to phase in from $3,000 to $2,500. Given the phase-in rate of 15 percent, this translated into a benefit of only $75 for those with earnings of up to around $16,000 for a family with two children (and no benefit whatsoever for families with earnings below $2,500).¹⁴

Disappearing benefits for many families of modest means.

For many families with higher but still relatively modest incomes, the 2017 law features a set of tax cuts that disappear over time and would in fact turn into tax increases over the long term as a result of the lack of indexing for the child tax credit and the effects of the slower inflation adjustment enacted in the TCJA.

The 2017 law expanded the child tax credit from $1,000 to $2,000 but did not index that amount to grow with inflation. However, at the same time, the law eliminated personal exemptions, including the personal exemption for dependents. The personal exemption allows a write-off of a specific amount — scheduled to be $4,150 per person in 2018 and growing with inflation. The value of that exemption then depends on a family’s tax bracket. For a family in the 10 percent tax bracket, the exemption would have been worth $415 as of 2018.

In general, there are good reasons to replace an exemption with a tax credit. Among other things, the value of the credit does not vary with a family’s tax rate. But the fact that the child tax credit amount was fixed in nominal terms and replaced an exemption that was indexed to inflation translates into a shrinking benefit over time.

For instance, for a family in the 10 percent tax bracket with one child, the $1,000 increase in the child tax credit initially was worth $585 more than the personal exemption it replaced in 2018. Now, five years later, that has fallen to about $510, and by 2035 it will have fallen to $335 under the Congressional Budget Office’s inflation assumptions.

At the same time, the 2017 law changed the way that the tax system adjusts for inflation by shifting to a chained consumer price index. The chained CPI, on average, measures a lower rate of inflation annually than the traditional CPI because it more accurately considers the degree to which families substitute toward lower-priced goods. There are good reasons why this is a more meaningful measure of price changes, but that doesn’t change the fact that families, including lower- to middle-income families, see tax increases as a result of the adoption of chained CPI. The chained CPI frequently goes unmentioned in discussions of the 2025 expirations because this was a permanent change to the tax system used to help finance the permanent corporate tax reductions. But its consequences are significant over time and could have been balanced with more generous tax savings to lower-income families elsewhere.

For instance, for a family with two children on the EITC, the introduction of chained CPI has reduced the maximum EITC by over $120 in 2023. And based on the CBO’s inflation projections, that would be expected to grow to over $530 by 2035. For low- and middle-income families, the chained CPI also affects thresholds like the start of the phaseout for the EITC and where the tax brackets start (pushing more income into higher brackets).

To give a sense for how these factors can combine, consider a family of four (a married couple with two kids) making $40,000 in 2018, and then compare their tax cut in that year with the result for a family earning the same amount (adjusted for inflation) in future years. As a result of the 2017 law, the family got a tax cut of about $1,420 in 2018. A family in the same position — earning the same amount adjusted for inflation — would get a tax cut of about $1,040 in 2023, a tax cut of just over $750 in 2026, and a tax increase of over $100 by 2035 under the CBO’s inflation assumptions and assuming the 2017 law is continued without changes.¹⁵ Figure 2 shows how various elements of the tax law combine over time to produce this effect.

How to address.

Any deal on taxes in 2025 should be judged by whether it prioritizes using limited fiscal space to help lower-income families. That would require changes relative to continuing current law, even as there are a variety of paths to accomplishing this.

To help the lowest-income families, there must be a meaningful expansion in refundability, and this should be the first priority. In 2021, under ARPA, there was an expansion of refundability with enhancements to the EITC for childless workers and full refundability of the child tax credit. Notably, there has been some support within the Republican Party for enhanced refundability. Former House Speaker Paul Ryan used to champion increasing the EITC for childless workers.¹⁶ And Sen. Mitt Romney, R-Utah; then-Sen. Richard Burr, R-N.C.; and Sen. Steven Daines, R-Mont., proposed significantly enhancing the refundability of the child tax credit.¹⁷ However, their proposal has critical differences from what the administration has proposed, in that there is still an earnings test, and they would partially pay for the expansion through a cutback in the EITC. The best path forward for low-income families is to permanently extend the refundability expansions that were enacted for 2021. Alone, full refundability of the child tax credit if the credit were extended at its current level and the expansion of the earned income tax credit cost about $30 billion per year.

In combination with expanded refundability, there should also be an expansion in support beyond that so that lower- to middle-income families do not see disappearing benefits. This could be done through an increase in the size of the child tax credit or another credit, such as one that goes to all filers based on the number of people in the family. A credit of this kind should be increased and then indexed, preferably to reflect the broader growth in incomes in the economy and not just inflation. The indexing of the credit reflects that this is meant to raise living standards and invest in lower- to middle-income families on a permanent basis and to reflect the broader prosperity of the country.

Importantly, though, benefits of a further expansion in credits should truly be focused on lower-income taxpayers, especially if other elements of the 2017 tax law are extended. This should not be used as reason to further increase the size of tax cuts received by upper-middle- and upper-income families, who already receive significant tax cuts on average under the law. For instance, a fully refundable child tax credit that is expanded for families making up to $100,000 to $125,000 in income for couples and not beyond that would squarely focus the tax credit expansion on those at about the 50th percentile and below.¹⁸

3. Reverse the decline of the corporate income tax system.

Over the decades, federal corporate income tax receipts have fallen considerably as a share of the economy — even as that has not been true of corporate profits. (See Table 4.) In fact, in the most recent two decades, corporate profits rose as a share of income, even as tax receipts fell relative to the economy. In the 2010s corporate profits were up by almost 20 percent as a share of national income, but federal corporate tax receipts were down by almost 20 percent as a share of the economy.

There are many explanations for the drop in corporate tax receipts. One is the rise of S corporations, which get taxed at the individual level rather than the corporate level. Another is increased shifting to low-taxed foreign countries of profits made in the United States.¹⁹ And then came the 2017 law, which cut the corporate rate from 35 percent to 21 percent and only partially offset the revenue loss through base broadening while not adequately addressing profit shifting to tax havens.

Those supporting corporate tax reductions have had two main and related justifications: First, they argue that this helps enhance incentives for investment and innovation with attendant benefits for workers as well as business owners. Second, they say that corporate rate cuts and other tax reductions are necessary because of dropping corporate rates around the world and competition for the reported location of the largest profits.

However, it is possible to raise significant, additional revenue from the corporate tax system and address both concerns. Given our fiscal challenges, this must be a third principle of any serious 2025 resolution.

Relative to the revenue lost, corporate rate reductions are a particularly ineffective incentive for innovation and investment. The corporate rate reduction provides windfall gains for past investment and also rewards those earning rents — whether because of market power or other factors — when we should be doing the opposite. This contrasts with direct incentives for investment. These kinds of direct incentives include the tax credits in the IRA. They were designed so that smaller and newer companies and technologies could access them, and they are focused on activity that also has broader benefits by helping to address climate change. As a result, these credits are particularly important, and their positive effects are already being seen in business activity. There are also less targeted incentives, like expensing, that still focus on new investment and largely avoid cutting tax rates on economic rents. Ironically, given the stated concern with innovation, the 2017 law eliminated the expensing of research and experimentation expenses and made expensing for other forms of investment only temporary.

When it comes to international tax competition, the world has changed substantially, even since 2017. One credible objection to raising the corporate tax rate, despite its weak connection to innovation and investment, is that corporations might react by shifting where they report profits on paper and, also, by locating some “lumpy” investment associated with those outsized profits in very-low-tax jurisdictions. A quintessential example is pharmaceutical companies’ shifting of profits and some investment to Ireland and other tax havens despite much of those profits being associated with economic activity elsewhere, including the United States. However, in 2021 nearly 140 countries around the world came together to agree on a global minimum tax of 15 percent. This will help address this kind of competition and eliminate the ability to shelter profits in extremely-low-tax countries. While it does not eliminate all forms of tax competition, it gives the United States even greater ability to tax the largest profits of its corporations at higher rates and without the same concern about those profits disappearing to tax havens.

How to address.

Given the fiscal challenges we face, corporations should be paying more under any 2025 resolution. A goal of increasing the projected federal corporate tax base to approximately 2 percent of GDP on a continuing basis, rather than the current course of only 1 to 1.5 percent of GDP, is not only achievable but can be done in a way that maintains or even enhances the incentives for innovation and investment. Proposals that represent a net tax cut to corporations should be a non-starter.

The corporate rate should be increased substantially. President Biden has proposed increasing it to 28 percent and raising in the rough range of $1 trillion in revenue over the next decade, or about 0.4 percent of GDP.²⁰ That can be done at the same time incentives for innovation and investment are maintained and even potentially enhanced. The incentives enacted in the IRA should be maintained. The TCJA provision ending the expensing of R&E should be reversed, and expensing of other forms of investment, which the 2017 law only made temporary, could be made permanent — though it would be sensible to pair this with further limitations on the deductibility of interest expense, to avoid a combination producing negative tax rates on new investment. These provisions initially cost in the range of 0.3 percent of GDP, but because they are largely a timing shift, their cost falls to less than 0.1 percent of GDP on a continuing basis.²¹ Notably, however, the challenges created by the TCJA when it comes to investment and expensing of R&E should be addressed only in the context of reform that bolsters corporate tax revenue and tackles the long-term decline of the corporate system.

Reform should also take full advantage of the agreement reached on the global minimum tax and bring the United States into compliance — helping to address the tax competition that has motivated corporate tax cuts of the past and undercut the corporate tax base. The 2017 law contained the kernel of a good idea by imposing a minimum tax on the foreign profits of U.S. corporations under a provision focusing on global intangible low-taxed income. But that minimum tax was set at too low a rate and allowed gamesmanship by giving corporations flexibility to pool taxes and profits across all countries rather than imposing it country by country. The 2021 global minimum tax agreement was in part inspired by the effort in the TCJA, even as it addressed several of these flaws. The United States should bring its system into full compliance with the deal, and it would be sensible to set our minimum tax rate on foreign profits higher than 15 percent and closer to the domestic corporate rate, given the many advantages of operating in the large and dynamic U.S. market.

International tax reform would help restore the U.S. corporate tax base. But it should not be viewed narrowly as a significant tax increase on corporations. In fact, if we fail to fully adopt the global minimum tax, U.S. corporations will still be paying at the global minimum tax rate — they will just be paying those taxes to other countries. Therefore, a key reason to bring the U.S. system into compliance with the global minimum tax is that this revenue should be paid to the United States — to help finance our needs and stabilize our corporate revenue base — and not to European and other capitals.

This kind of sensible international tax reform could raise hundreds of billions of dollars over the coming decade.²²

4. Eliminate tax cuts for the highest-income Americans and increase their contributions relative to pre-2017 law.

As we have described, almost 25 years of tax policymaking have been dominated by two rounds of tax cuts that have disproportionately benefited those toward the top of the income spectrum, both through corporate tax cuts as described above and also through individual income and estate tax cuts. The tax cuts will eventually have to be financed, and unless the tax cuts for those toward the top of the income spectrum are reversed, it’s likely that lower- and middle-income families will be the ones who pay for them — whether in current or future generations.

It is important to differentiate among those at the top of the income spectrum and how these tax cuts — and the preexisting tax system — affect them. The two rounds of tax cuts over the past two decades delivered real benefits at the very top, including the top 0.1 percent. But the largest benefits relative to incomes were repeatedly delivered to the 90th through 99.9th percentiles, and especially the 95th to 99th percentiles. For context, the income cut off for the 90th percentile is expected to be around $230,000 in adjusted gross income as of 2026, and the 95th percentile is expected to be around $335,000.²³

Several provisions in the TCJA delivered these higher-end income tax cuts. The most important were (1) rate cuts and reforms to the alternative minimum tax; (2) the new special deduction for passthrough income that’s highly concentrated in its benefits at the top of the income distribution and available only to business owners and not employees; and (3) a further increase in the estate and gift tax exemptions, which reduced the share of taxable estates from 0.2 percent to less than 0.1 percent and cut taxes for that remaining sliver.²⁴

Yet for Americans who have the greatest resources, rely heavily on investment income, and represent the top of the top, their relatively low tax rates derive largely from preexisting elements of the tax code. As reported by Danny Yagan based on earlier joint work with Greg Leiserson, the top 400 wealthiest families have paid an average tax rate of 9.6 percent in nominal terms — or 12 percent, adjusting for the effects of inflation — from 1992 to 2020.²⁵ This includes accrued but unrealized income, which is a dominant source of income for these families. Under the existing tax code, this income can entirely escape income taxation if it is passed down to heirs through the step-up in basis at death, which eliminates any taxation of the gain at that point. This is in addition to substantial evidence that the IRS has struggled to enforce the existing tax laws on the very wealthiest families — a situation that worsened in the 2010s as Congress significantly slashed IRS funding and audit rates dropped dramatically, especially at the top.

The ways in which the very-best-off escape taxes cost revenue, lead to unproductive tax planning, and undercut faith in the tax system — faith that will be needed to address the long-term fiscal shortfall. The federal tax system is progressive, with average rates that rise as resources grow, until one reaches the very top, where that reverses.

How to address.

The highest-income Americans need to pay more in taxes as part of any 2025 resolution. The first step is to allow the tax cuts they enjoyed from the 2017 law to expire and, to the degree possible, limit tax cuts for others who are still well-to-do. This means expiration of the rate cuts at the top; expiration of the deduction for passthrough income, or at most extending it below a reasonable income threshold; and expiration of the increase in the estate and gift tax exemption. The AMT could be allowed to revert to something like its previous form, with much broader effect, or the revenue could be replaced with increases in the tax rates at the top and more direct limits on write-offs.

However, it is not just a question of allowing tax cuts to expire but having the highest-income taxpayers pay more relative to pre-2017 law and help finance the rest of the legislation. There are any number of ways to do this. One is to continue the cap on the deductibility of state and local taxes. That cap raises roughly $100 billion per year,²⁶ and it disproportionately falls on those at the very top of the income spectrum. Over 50 percent of the additional liability comes from the top 1 percent, and an additional 25 percent comes from the next 4 percent.²⁷

Between continuing the cap and simply raising tax rates, we would prefer raising rates since, at least theoretically, there is some potential for the deduction to help encourage state and local investment with broader positive spillovers (though the evidence is mixed). Investment in roads, bridges, education, and other such spending help not just the state or locality involved but others too. However, given that this measure has been in effect for several years, it may be the most likely path to significantly increase taxes on the top, and continuing the cap in something like its current form would be better than having no significant tax increase at the top at all. Extension of any version of the cap should also shut down the now-expanding workarounds that allow some of the very highest-income Americans — specifically, those working as partners in firms rather than as employees — to avoid the cap.

This is also an opportunity to address at least some of the approaches used by the very richest Americans to significantly reduce their effective tax rates. We believe that a more fundamental reform of the realization rule for the extremely well-to-do — such as Biden has proposed in his billionaire minimum tax — is the best course for the tax system. But there are intermediate steps to be taken in the right direction as well. Step-up in basis could be addressed above a high-dollar threshold by either taxing those built-in gains at death or switching to carryover basis, meaning that the gains are eventually taxed when the property gets sold, even if that’s by heirs. Also, abuses of specific provisions could be addressed — such as the qualified small business stock exclusion, which has exploded into a giveaway, especially to tech investors; carried interest; and the use of private foundations to generate charitable deductions without public benefit. And, as discussed in the next section, the 2025 debate can be used to resolve the issue of long-term IRS funding and provide for stronger, sustained enforcement at the top.

Even if all of this were accomplished — and that would require a major political lift and sacrifice from both parties — steps like these are unlikely to give us a tax system that can fully finance the government’s long-term commitments. The political system will need to grapple with the fact that the top decile and quintile of earners have benefited disproportionately from two decades of tax cutting, which is not sustainable over the long term. However, credible progress in increasing tax rates and reducing tax avoidance among the highest-income earners in the United States would represent an important step toward strengthening our revenue base and restoring public confidence in the fairness of our tax system — and, hopefully, lay the groundwork for further reform in the years ahead.

5. Improve tax administration and simplify filing.

As we’ve noted, the tax system is defined not only by the laws that are on the books but how that system is administered, and the 2010s featured a substantial reduction in funding for the IRS, which helped lead to a collapse in both taxpayer service and enforcement of the laws. At the same time, there are long-standing issues in the cost and complexity of tax return filing. The IRS had negotiated with private software companies to provide free filing, for which 100 million low- to moderate-income Americans should have been eligible in 2020 but was used by only 4 percent, according to the Government Accountability Office.²⁸ This was partly because the companies themselves redirected households toward paid products when they could have filed for free.

But there have also been important steps in the right direction in recent years, some of which are beginning to address these challenges.

First, as we have discussed, the TCJA swapped an increase in the standard deduction and child tax credit for elimination of personal exemptions. One benefit of that swap is that fewer Americans now choose to itemize their deductions — because the standard deduction is larger than the specific deductions (such as the home mortgage interest deduction or the state and local tax deduction) they might take. The swap meant that we went from about 30 percent of filers itemizing to only about 10 percent of them doing so.²⁹ This is a true simplification for those who used to itemize.

Second, the large cuts in funding to the IRS during the 2010s have now been reversed, albeit temporarily. By the end of the Trump administration, funding for the IRS had roughly stabilized in real terms, though at a reduced level, and the Trump administration had begun to seek small increases. Bipartisan support for increasing IRS funding continued initially during the Biden administration. In fact, a $40 billion multiyear investment in the IRS was at first included as part of the publicly announced bipartisan infrastructure deal and as a way to help finance the legislation, since the investment would generate considerably more revenue than its cost. However, Republican negotiators were ultimately unwilling to support this provision, and Democrats eventually enacted an $80 billion increase in IRS funding in on a party-line basis in the IRA, which was estimated to raise far more than $80 billion in revenue. Unfortunately, in the wake of this, parts of the Republican Party have focused on IRS funding and enforcement as a partisan issue, and House Republicans demanded an over $20 billion cut to the funding as part of the debt limit deal, with that cut apparently focused on enforcement. What was once a 10-year investment plan through 2031 has effectively been transformed into a shorter one.

Notably, the benefits from the new IRS funding are already being seen. While effectively and fairly increasing enforcement will take time, the IRS has moved to immediately address the dramatic drop-off in service for taxpayers and the backlog of unprocessed tax returns. At the end of filing season this year, the backlog in returns had been cut by 80 percent, and the IRS had gone from answering only about 10 percent of calls to about 35 percent — a number that is still too low but represents a real improvement.³⁰

Finally, the Biden administration in May announced that it will be piloting a “direct file” option with the IRS that could eventually give tens of millions of Americans easy access to free return filing with the government. As recently issued reports from the IRS and an independent third party have suggested, several critical questions remain — such as how the system would interact with state tax return filing — but this has the potential to make filing less expensive for many Americans who have not been well-served by the current software providers.

How to address.

Even though recent years have seen historic challenges for tax administration, the coming tax debate presents an important opportunity for reform. A final key principle for 2025 is that tax return filing should be simple and free for the vast majority of American households, and the highest-income Americans should be accountable to pay the taxes they owe.

This is an area in which we need to reduce partisanship over commonsense policies. Fewer tax itemizers are better than more, all else being equal. The IRS should be able to provide good service to taxpayers needing help. Return filing should be simple and free for middle- and lower-income families. Finally, the laws on the books should be fairly and effectively enforced — and these enforcement efforts can raise considerable revenue without actually increasing taxes owed under the law.

The 2025 tax debate is the time to address the future of tax administration. The additional IRS funding from the IRA had been designed to last through 2031, but with the cuts demanded by the Republicans in the debt limit deal, the runway has become significantly shorter.³¹ Resolving the future of the IRS should not wait until the last minute. The IRS will be unable to effectively continue to invest in service and enforcement as the funding cliff approaches. The uncertainty would surely affect how IRS leadership manages the agency’s resources and affect its ability to recruit and retain talent, which is critical to tax administration objectives.

Both of us have lived through recent tax debates and understand how partisan many of these issues have become. But we have also seen opportunities for bipartisanship, including in the initial agreement on the bipartisan infrastructure law. That came in a context when the tax system wasn’t even up for broader debate. In 2025 it will be, and improving tax administration and simplifying filing should be goals for legislation that year — and ideas championed by both parties should be included in an agreement.

Conclusion

As the tax cut expirations in 2025 approach, policymakers should avoid the tendency to view the coming “tax cliff” in narrow terms. In the face of a multidecade strategy to starve our country of revenue and tilt the tax system toward benefiting upper-income taxpayers, 2025 is a moment when we need to think big and do better.

If policymakers can keep their focus on the principles outlined here, 2025 will represent a historic opportunity. While 2025 will not be the last word on tax reform in America, it could be the most consequential moment in a generation to lay the groundwork for the tax system that this country deserves — a tax system that is fairer and can finance the services and investments that Americans support.