Tax Notes logo

A Progressive Case for a Federal SALT Deduction

Posted on May 27, 2024
Patrick Driessen
Patrick Driessen

Patrick Driessen is a former government economist and revenue estimator.

In this article, Driessen expands the typical income distribution analysis to show in a stylized model how the availability or absence of federal deductibility of state and local taxes affects overall post-tax progressivity.

Much of the debate about the Tax Cuts and Jobs Act’s cap on federal deductibility of state and local taxes has been reduced to income distribution tables that show cap repeal as very regressive. These federal-tax-centered tables are narrowly framed: State and local taxes and most outlays themselves aren’t distributed, and states are assumed not to respond to federal tax changes.

This article relaxes these distributional assumptions in a stylized example that controls for pretax income distribution as well as state and local outlays. This is a more realistic approach: At one level, federal deductibility of SALT is mostly about redistribution between high-income people in progressive-tax states and those in regressive-tax states, without much direct effect on others. At another level, low- and middle-income people are affected as states change their tax structures in response to the availability or absence of federal deductibility of SALT.

A federal tax deduction for SALT paired with revenue-neutral top-end tax increases is shown to encourage overall progressivity. A “super-Pease” pay-for for federal deductibility is also explored.

The Narrow SALT Cap Story

SALT overall is less progressive than federal taxes and in many places SALT is regressive.1 Many individuals in low- and middle-income groups residing in both high- and low-SALT jurisdictions pay more SALT than federal (and especially federal income) tax.2 Yet the SALT cap, perhaps the most important tax policy change affecting tax federalism in some time, is shown myopically in typical distributional tables.

SALT is treated in federal-government-centered income distributions not as taxes but voluntary private personal expenditures that comprise one of the uses of post-tax income. The distributional tables imply that SALT is entirely a voluntary payment for personal benefit with no positive externalities, and generally little or no state and local expenditures are reflected in pre- or post-tax measures. Another model restriction is that for revenue estimates and income distributions in these federal-centered approaches, the states are assumed not to change their fiscal structures in reaction to changes in federal tax treatment of SALT.3

A federal-tax-centered distributional model that excludes SALT, or includes it on only a static basis, is by definition limited in what it can say about the progressivity of the tax features of fiscal federalism.4 State and local taxes essentially become static offstage characters that only indirectly enter the story through their deductibility or creditability (the latter in the context of proposals to convert federal deductions of SALT to federal credits) for federal tax purposes.5

Fruit, Layer, and Marble Cakes

Debate about the proper federal tax treatment of SALT has been ongoing for many decades. In food terms, deduction critics see SALT as no different than private expenditures for oranges and apples.6 Some of the discussion revolves around specific taxes, including income, sales, and property taxes, and whether federalism is best thought of as a separable layer cake (with SALT more of a distinct benefit lien than federal taxes) or a marble cake (with many intergovernmental interactions and less distinction between the governmental levels).7

The lead-up to the Tax Reform Act of 1986 promoted the layer cake,8 but TRA 1986 itself ended up not changing much directly other than denying federal deduction for sales taxes (later changed to a separate offering of federal deductibility of state sales taxes). Full repeal of federal SALT deductibility helped finance part of the TCJA and make the income distribution tables when the TCJA was enacted in 2017 not quite so regressive,9 but the subsequent emergence of SALT cap workarounds dinged both of those effects.

Along the way, the individual alternative minimum tax and the Pease provision eroded the value of federal deductibility of SALT (or to be more technically correct, raised tax rates on income above specific thresholds) before 2018 and are slated to come back after 2025. Some SALT cap critiques don’t say much about Pease or the AMT limitation, perhaps reflecting dislike of these provisions based on concerns about complexity and reach, but as argued below, the emergence of SALT cap workarounds should put a Pease variant in a better light.

States often respond to federal tax changes — examples from the TCJA include the rapid spread of SALT cap workarounds and a shift by some states and localities to less progressive tax systems.10 Research from before the TCJA suggested that SALT structures were changed to achieve federal deductibility,11 while state and local spending was relatively unresponsive to federal deductibility.12 These findings should be integrated into SALT cap analysis.

A Two-State Distributional Model

Table 1 shows the state tax setup for a stylized model of federalism that looks at all taxes and includes a state response to federal deductibility of SALT. State P, a progressive-tax state, and State R, a regressive-tax state, both have the same pretax incomes, including magnitude and distribution, and the same total SALT revenue collection, and it’s assumed that state and local outlays are exactly the same in total and incidence.13 The only thing that’s different is SALT incidence across income groups.

Table 1. State Tax Inputs for Results in Tables 2 and 3

State P

State R

Units

Income Per Unit

State Tax Liability Per Unit

Units

Income Per Unit

State Tax Liability Per Unit

With No State Response to Federal Deductibility

With State

Response to Federal Deductibility

10

$50

$3

10

$50

$10.50

$9

17.5

$100

$8

17.5

$100

$15.70

$14.16

17.5

$300

$30

17.5

$300

$31.40

$31.12

5

$1,000

$125

5

$1,000

$78.40

$87.72

Addendum: Averages

$250

$26.40

$250

$26.40

$26.40

As shown in Table 1’s right-hand column, the response to federal deductibility of SALT is assumed to cause State R to move 20 percent of the way toward State P’s tax structure. Each state collects $1,320 in state revenue, with the state tax rate across all income groups of 10.56 percent in each state.

Federal tax liability is modeled in tables 2 and 3 using a standard deduction (or alternatively, federal tax deductibility with or without a Pease variant) with federal tax rates up to 37 percent.14 Without federal deductibility of SALT, the federal government collects $5,539 in total ($2,769.50 in each state).

A regressive-tax state like State R in Table 1 can exert a stubborn influence on an overall tax system. Federal refundable credits are one lever for remediation of SALT regressivity, but in my view, that’s kind of a fudge (because technically the refundable portion of refundable credits is no different than government outlays that are transfers) and misses some people if it targets only earned income (earned income credit) or children (child tax credit), though both of those credits are worthy for what they do.15 A broader approach, short of a full-fledged universal basic income grant, would be to encourage State R to adopt more of State P’s tax structure, and that’s where federal deductibility plays a role.

Table 2 shows Gini coefficients (lower Ginis mean less inequality) and tax rates (combined including federal and state taxes) for each state, as well as for both states together, for three scenarios.16

The first scenario is similar to what the law is in 2024 with no federal deductibility for SALT. The other two scenarios include deductibility — one that’s not revenue neutral (deductibility reduces federal revenue by 11.1 percent)17 and the other in which federal deductibility is revenue neutral when combined with the imposition of federal tax increases on the top 10 percent income group. This latter scenario is intended to be similar to the Build Back Better legislation passed by the U.S. House of Representatives in 2021.

Table 2. Combined Overall Ginis and Tax Rates for Federal Treatments of SALT for Stylized States

 

No Federal SALT Deduction

With Federal SALT Deductiona

Without Revenue Neutrality

With Revenue Neutrality, Financed by Tax Increases for Top 10 Percent

State R

State P

Both States

State R

State P

Both States

State R

State P

Both States

Ginis after federal taxes and SALT

0.4326

0.3979

0.4152

0.4279

0.4109

0.4194

0.4094

0.393

0.4012

Combined Federal and State Tax Rates by Income Group

Lowest 20 percent

20.9%

6%

13.5%

15.5%

6%

10.8%

15.5%

6%

10.8%

20 to 55 percent

20.5%

12.8%

16.6%

17.3%

12.8%

15%

17.3%

12.8%

15%

55 to 90 percent

31.1%

30.6%

30.8%

28.3%

28.1%

28.2%

28.3%

28.1%

28.2%

Top 10 percent

39.9%

44.6%

42.3%

38.5%

40.3%

39.4%

44.5%

46.5%

45.5%

All

32.7%

32.7%

32.7%

30.4%

30%

30.2%

33%

32.5%

32.7%

Note: Pretax (that is, before any federal or state and local taxes) Gini coefficient is 0.479. Tax rates include federal, state, and local taxes. There’s no accounting for SALT cap workarounds, or the increase in the federal deficit in deductibility-without-revenue-neutrality scenario.

aIncludes response to federal deductibility that bridges the no-federal-SALT-deduction progressivity gap between the states by 20 percent when there is a federal SALT deduction.

Another aspect of the federal deductibility scenarios is that State R is assumed to respond to federal deductibility by choosing a moderately more progressive rate structure that leads to lower combined tax rates in the low- and middle-income groups in State R. For example, the combined federal and state tax rate in the lowest quintile in State R drops from 20.9 percent to 15.5 percent in the federal deductibility scenarios. One of the consistent results across all three scenarios in Table 2 is that the lower two income groups always have lower combined tax rates in State P than in State R.

Overall, the no-deductibility scenario shows big inequality and tax rate differences between the two states. The scenario with federal deductibility of SALT in a non-revenue-neutral manner increases overall inequality and especially the inequality in State P, though even in that scenario, the Gini coefficient of State P (0.4109) remains lower than it is in State R (0.4279).18

Federal deductibility for SALT with revenue neutrality achieved with additional taxes imposed on the top 10 percent residing in both states is the scenario with the best inequality results. In this case, federal deductibility comes with higher tax rates for the top 10 percent, particularly in State R where the combined rate rises from 39.9 percent with no federal deduction permitted to 44.5 percent with revenue-neutral federal deductibility. But even with federal SALT deductibility (without and with revenue neutrality), the combined tax rate for the top 10 percent income group in State R is still lower than the combined tax rate faced by the top income group residing in State P.

One general conclusion from this exercise is that on a static basis, the federal tax changes described here result directly in redistribution among the top income groups only. Part of that is by design of the example, but that also comports with the idea that high-income groups pay a lot of federal tax, so rearrangements of federal taxes and deductions will redound largely in that group whether members of that group reside in State P or R. The big nonstatic impact in Table 2 for lower-income groups happens when State R itself is assumed to become less regressive in its tax structure in response to federal deductibility, a choice that likely would be motivated both by economics and by political economy.

Super-Pease to the Rescue

The TCJA $10,000 cap is unindexed, the same for both single and joint filers, and seems low given the variety of circumstances and SALT that people face. Even if taxpayers who lost SALT deductions still got net TCJA tax cuts, there were process and substance concerns that can’t be dispelled by elective pairing of certain TCJA provisions, and if there is analytical interest in vertical equity, state and local taxes should be treated as taxes in income tax rates and distributions.

When radical provisions are enacted without enough buy-in, other extreme things often happen in response, but the IRS guidance that permitted passthrough entity workarounds of the SALT cap is its own semi-horror story.19 Workarounds either were a result of sloppiness in statutory law drafting and the TCJA’s technical explanation or a masterful long game (that is, TCJA supporters could claim to end federal deductibility for progressive reasons but encourage or wink at a passthrough exception). I’m not aware that any legislators who voted for the TCJA have repudiated the workarounds that have corroded both the revenue and distributional rationales for the SALT cap. Workarounds now seem to be perceived as a distinct tax increase test and may outlast the SALT cap and affect other tax provisions in the future, including other tax expenditure limitations.

Looking back, removing state and local sales tax from federal deductibility in TRA 1986 probably was too much in one direction, and the TCJA’s stern SALT cap in 2017 too much in the other direction. One might believe that in all this, the Pease provision — a sort of middle ground position on SALT deductibility enacted in 1990 — coupled with federal deductibility would be a good choice. But the Pease provision is disfavored because it is considered too complex or mild by some.

Thus, the test here is of a Pease-inspired alternative that delivers a big marginal incentive, at a relatively low loss of federal revenue, for states and localities to use progressive taxes. The variant of Pease used in Table 3 for both federal deductibility scenarios (without and with revenue neutrality) permits federal deduction for SALT above a 4 percent rate (a 5 percent threshold could work too), assumed to be defined as necessary to prevent gaming.

Table 3 shows that federal deductibility with a Pease variation at a 4 percent threshold is a bit more equitable overall than no deductibility (Gini of 0.4151 versus 0.4152) even without revenue neutrality. With federal revenue neutrality achieved by taxing the top 10 percent, the reduction in inequality is a full 0.01 improvement (0.4057 versus 0.4152, not to be sneezed at in the Gini world).

Note also that with the Pease variant, the effective tax rates in both states needed for revenue neutrality in the top income group are lower by a couple of percentage points than what was required for the revenue-neutral scenario shown in Table 2. That result is attributable to the amount of lifting the Pease variant does.20

Table 3. Combined Overall Ginis and Tax Rates, Including Pease Treatment for Deductible SALT Options

 

No Federal SALT Deduction

With Federal SALT Deduction and Pease Set at 4 Percenta

Without Revenue Neutrality

With Revenue Neutrality, Financed by Tax Increases for Top 10 Percent

State R

State P

Both States

State R

State P

Both States

State R

State P

Both States

Ginis

0.4326

0.3979

0.4152

0.424

0.4063

0.4151

0.4143

0.3971

0.4057

Combined Federal and State Tax Rates by Income Group

Lowest 20 percent

20.9%

6%

13.5%

15.5%

6%

10.8%

15.5%

6%

10.8%

20 to 55 percent

20.5%

12.8%

16.6%

17.6%

12.8%

15.2%

17.6%

12.8%

15.2%

55 to 90 percent

31.1%

30.6%

30.8%

29.8%

29.6%

29.7%

29.8%

29.6%

29.7%

Top 10 percent

39.9%

44.6%

42.3%

39.9%

41.8%

40.9%

42.8%

44.5%

43.9%

All

32.7%

32.7%

32.7%

31.7%

31.2%

31.4%

33%

32.5%

32.7%

Note: Pretax (that is, before any federal or state and local taxes) Gini coefficient is 0.479. Tax rates include federal, state, and local taxes. There’s no accounting for SALT cap workarounds, or the increase in the federal deficit in deductibility-without-revenue-neutrality scenario.

aIncludes response to federal deductibility that bridges the no-federal-SALT-deduction progressivity gap between the states by 20 percent when there is a federal SALT deduction.

Reclaiming Progressivity

Today’s SALT cap is mostly about redistribution between high-income individuals in progressive and regressive tax states. A common pro-SALT-cap argument is that with federal deductibility, high-tax states overdo government provision at the expense of low-tax states. Yet there’s evidence that federal deductibility affects state tax progressivity more than state government size.21

Is there “paternalism” in this advocacy of a federal SALT deduction? Sure, but federal influence can go either way, it’s not neutral. Just like those who enacted the TCJA couldn’t directly tell states what kind of tax system they should have but could — by enacting the SALT cap — encourage states to adopt less progressive tax structures, so too can federal deductibility (perhaps with a Pease variant) encourage states to adopt or retain progressive structures.

There’s a clear revealed interest by the U.S. population as a whole in overall tax progressivity, even if the view on the magnitude of government spending is more equivocal. Some of the states on behalf of which there have been made objections to the effects of a federal deduction for SALT have the most regressive tax structures in the nation. A practical way to further overall progressivity is for these states to adopt more progressive SALT structures, not for the federal tax code to encourage the unwinding of the remaining progressive SALT structures.

States with progressive tax systems understandably are concerned about losing residents because of tax-related migration, but those of us who desire overall progressivity in our own and other states should also worry that overall and specific-location progressivity is undermined by states with regressive tax structures. Federal deductibility with a Pease variant that uses a properly crafted threshold is a good recipe choice in the SALT deduction cake and political bake-offs.

FOOTNOTES

1 See, e.g., Institute on Taxation and Economic Policy, “Who Pays? A Distributional Analysis of the Tax Systems in All 50 States” (Jan. 2024).

2 That is even more likely if the refundable portion of federal tax credits is treated as a negative federal income tax.

3 Assuming no change in government behavior (including foreign, U.S. federal, and state and local governments) aids tractability and consistency (though it’s not required by another estimating convention that assumes U.S. gross national product doesn’t change in the budget period). Unfortunately, the static approach effectively undermines analysis of proposals that by design and intent are meant to encourage big changes in behavior by other nations and U.S. states and localities. Proposals for which this static government assumption is a problem include a lot of SALT issues (such as federal attempts to prohibit or regulate formulary tax methods used by states) as well as the OECD’s pillar 1 and 2 proposals.

4 Some of the references for federal deductibility of SALT available before the TCJA were terse — heavy on revenue or tax expenditure enumeration, and light on context. Notably absent was a description of SALT structures, state and local behavior, and fuller context on horizontal equity, including federal tax treatment of foreign tax payments by U.S. individuals and of SALT deductions afforded to subchapter C corporations. Even the more elaborate explanation offered by the Congressional Budget Office in its budget options right before the TCJA was enacted was cryptic or noninclusive on these issues. CBO, “Options for Reducing the Deficit: 2017 to 2026,” at 140 (Dec. 2016).

5 Some thoughtful articles from before and after the TCJA have gotten lost in the din, including Frank Sammartino, Gordon B. Mermin, and Noah Zwiefel, “Alternatives to the TCJA Limit on the State and Local Tax Deduction,” Urban-Brookings Tax Policy Center (May 6, 2020); Leonard E. Burman, Tracy Gordon, and Nikhita Airi, “Congress Can Help State and Local Governments Prepare for a Rainy Day Without Repealing the SALT Cap,” TaxVox Blog, Jan. 14, 2021; and testimony of Sammartino before the Senate Finance Committee, “Federal Support for State and Local Governments Through the Tax Code” (Apr. 25, 2012).

6 “Most state and local taxes can be thought of as being fairly closely related to benefits received; certainly the ‘layer-cake’ model of fiscal federalism tells us that state and local governments should use benefit-type taxes. There is not much more reason to allow deductions or credits for such taxes than for private purchases of apples and oranges.” Comments by Charles E. McLure Jr. in Comprehensive Income Taxation, The Brookings Institution, at 71 (1977).

7 Regarding differentiating between types of SALT, John F. Due, “Personal Deductions” in id. at 51: “The strongest case can be made for the federal deductibility of state income taxes because the gains from fiscal coordination are greatest in this field, and the potential effect in lessening opposition to increases in state taxes is greater with this levy than with others because of the progressive rate structure.”

8 Treasury, “Tax Reform for Fairness, Simplicity, and Economic Growth” (Nov. 1984); and The White House, “The President’s Tax Proposals to the Congress for Fairness, Growth, and Simplicity” (May 1985).

9 The TCJA distribution is even more regressive when the full effects of the loss of federal healthcare benefits caused by the zeroing out of the Affordable Care Act shared responsibility payment are included. CBO, “Reconciliation Recommendations of the Senate Committee on Finance” (Nov. 26, 2017).

10 It’s perhaps not a coincidence that in the absence of a federal deduction for SALT, there’s been a shift to less progressive SALT. Jared Walczak and Katherine Loughead, “The State Flat Tax Revolution: Where Things Stand Today,” Tax Foundation (Feb. 15, 2024).

11 CBO, “The Deductibility of State and Local Taxes,” Pub. No. 2906, at 15 (Feb. 2008): “The interaction between the presence of high-income taxpayers and state and local tax burdens also influences how the benefits from the taxes-paid deduction are distributed among the states. Although taxpayers in states that have a large percentage deduction from AGI tend to claim larger deductions at all income levels than do taxpayers in states that have a small deduction share, the ratio increases as income rises. That is, the difference between the claimed deductions of taxpayers in large-share states and small-share states is greatest for the highest-income taxpayers. That finding implies that the benefits from deductibility are more closely related to the progressivity of state and local taxes than to their average level.”

12 Id. at 6. The absence of much of a link between federal deductibility and state and local spending can work both ways — it suggests that federal deductibility isn’t that effective at getting states and localities to spend more to generate more positive externalities, but it also suggests that federal deductibility may affect SALT structures without necessarily encouraging states to do more on the outlay side.

13 The focus in this exercise on SALT progressivity rather than SALT levels aids in highlighting the difference in state tax structures. That focus also accords with the idea that “the benefits from deductibility are more closely related to the progressivity of state and local taxes than to their average level.” Id. at 15.

14 The standard deduction is $10. There is a federal 12 percent marginal tax rate between $50 and $100, 22.5 percent between $100 and $200, and 37 percent above $200.

15 Further, to the degree SALT regressivity is a reason to use federal refundable tax credits or even something like a universal basic income grant, the uneven after-SALT inequality starting point for low-income people in regressive-tax and progressive-tax states creates its own targeting problem. Federal refundable credits or a universal basic income grant are relatively blunt instruments to the extent that they can’t directly discriminate between states that have regressive and progressive SALT structures. Advocates of national sales taxes or even carbon taxes must also deal with the existing state and local sales tax infrastructure.

16 Gini coefficients are imperfect but useful here as a summary statistic. A Rawlsian approach that places more weight on low-income inequality would reinforce the general outcome of this exercise that emphasizes the embedded regressivity in some SALT structures.

17 Federal revenue with no deductibility is $5,539. With federal deductibility of SALT but no other changes, federal revenue falls to $4,920.

18 Ideally federal budget deficits should be distributed too, but that’s omitted here for simplicity.

19 Daniel J. Hemel, “The Passthrough Entity Tax Scandal,” New York University School of Law Working Paper No. 23-01 (Feb. 2023).

20 This Pease variant could be added on to, say, an indexed $15,000/$30,000 single/joint cap that could be phased out at higher incomes.

21 CBO, supra note 11.

END FOOTNOTES

Copy RID