Tax Analysts®Tax Analysts®

My Subscriptions:

Featured News

April 27, 2012
On Tax Increase Limitations: Part II -- Evasion and Transcendence

Full Text Published by Tax Analysts®

by David Gamage and Darien Shanske

David Gamage and Darien ShanskeIn our previous column in this series we argued that tax increase limitations (TILs) are analytically incoherent. We further suggested that this incoherence contributes to the observed ineffectiveness of TIL regimes.1 In this, our second column, we analyze the implications of the incoherent nature of TILs. In particular, we argue that because of their incoherent nature, TILs can be effectively evaded by a legislative majority wanting to do so.

There are several reasons why it is worth discussing strategies for circumventing TILs. First, and perhaps most obviously, that TILs can be effectively evaded illustrates our point in the prior column about TILs being incoherent. Second, we believe that these evasion strategies have in fact been used, at least to some degree, and that reflecting on the evasion strategies can thus help in understanding the failure of TILs. Third, the evasion strategies that we will discuss in this column may have positive normative features; by using the term "evasion" we do not mean to imply that there is necessarily anything inappropriate about using these strategies. Fourth, and finally, reflecting on the potential for evading TILs is instructive for developing more effective mechanisms for controlling the growth of government -- which we take to be the primary goal of the advocates of TILs. Paradoxically, it turns out that a promising strategy to evade TILs also potentially functions as a mechanism for controlling the size of government. Indeed, that strategy -- the increased use of benefit taxation combined with refundable tax credits -- could work still better were it not for poorly designed TILs.

Two Strategies for Evasion

Without further ado, we will explore two related strategies for circumventing TILs regimes: the "tax expenditure" strategy and the "benefit charges with refundable tax credits" strategy.

The Tax Expenditure Strategy

Suppose that the language of a given TIL provision was the following:

    Any changes in state taxes enacted for the purpose of increasing revenue collected pursuant thereto whether by increased rates or changes in methods of computation must be imposed by an act passed by not less than two-thirds of all members elected to each of the houses of the Legislature.2

Note that what is essential to this provision is that an overall change in state taxes does not increase state revenue; this still leaves a lot of room for maneuvering. Thus, instead of increasing taxes in order to fund new spending, a majority party can just pass a new tax credit to fund a desired program while increasing other taxes or reducing other tax expenditures. The overall tax package would be revenue neutral in that it would not increase overall taxes, but it would in effect accomplish the majority party's spending and taxing goals because the tax expenditure would substitute for the increased spending while the tax shift would shift tax liability to where the majority wanted it to be.3

In theory, there are no limits to the types of programs that could be funded in this manner. Businesses and high-income individual taxpayers could be given dollar-for-dollar tax credits against their higher taxes in exchange for donating to state spending programs (for example, universities or healthcare programs), thus systematically exchanging public funding for nominally private funding. The more tax instruments a state has, the easier that will be, but even a state without an income tax can favor all manner of programs through sales tax expenditures (while simultaneously raising the sales tax on other goods or services).4

In short, the tax expenditure strategy relies on TILs not applying to revenue-neutral packages that reduce taxes on some taxpayers (through tax expenditures) while increasing taxes on others. For most existing state TILs, this strategy should suffice for the majority party to evade TILs to the extent the majority party so desires. But if the tax expenditure strategy is unavailable, a majority party might still use our second strategy -- the benefit charges with refundable tax credits strategy.

The Benefit Charges With Refundable Tax Credits Strategy

The essence of this second strategy is to transform the funding mechanism for government programs to benefit charges instead of general fund expenditures. To defray the distributional impact of these benefit charges, the state would then provide refundable tax credits against the state income tax. Those refundable credits should phase out with income, so that low-income taxpayers could be completely reimbursed for benefit charges, whereas higher-income taxpayers would be reimbursed only partially.5 For example, consider tuition at public colleges and universities. Even today, tuition at those schools is a relative bargain compared with many private colleges and yet recent dramatic increases in tuition still undermine the public purpose of state schools to provide affordable public higher education. The solution to the riddle here could be to allow tuition at public colleges and universities to remain high -- or even to become higher---- but to then use tax credits to keep those schools affordable for lower-income students. State higher education credits could be administered, for example, through the state income tax and could be modeled on federal higher education tax credits. The credits could be made refundable so that taxpayers without tax liability would still be helped.

Individual taxpayers might still face liquidity issues even with refundable credits because lower-income taxpayers could have trouble paying benefit charges, like tuition, upfront and then waiting for a state income tax return. To address that problem, the tax credits could be made advanceable. The Affordable Care Act's premium tax credits in new IRC section 36B is an example of how tax credits can be made advanceable. Under section 36B, state Exchanges can make advance payments of the premium tax credits to pay for health insurance for low-income taxpayers, with the taxpayers then reconciling the advance payments with the amount of the tax credits that they are allowed when they file their tax returns.6 Similarly, for example, state universities could receive advanceable state tax credits to cover low-income taxpayers' tuition.7

By using the benefit charges plus refundable tax credits strategy, which we will henceforth call BCPP (benefit charges plus progressivity), a majority party can effectively evade TILs because funding can be increased for state spending programs without actually needing to raise explicit taxes or spending. The limit on this strategy is the preexisting state income tax (or other preexisting state taxes).8 Refundable tax credits can be used to completely alleviate the expense of benefit charges for low-income taxpayers, but positive tax liabilities cannot be assessed on high-income taxpayers in excess of their preexisting tax liability.

Hence, taken to the limit, preexisting state taxes become the mechanism for achieving progressivity in state spending under the BCPP strategy. Further, the preexisting income tax and other general taxes (or other sources of revenue, like federal grants) fund any government programs not entirely fundable through benefit charges.

What About California's Proposition 26?

In November 2010, California's Proposition 26 modified its TIL regime partially in response to the success of a version of the tax expenditure strategy.9 The new rule is as follows:

    Any change in state statute which results in any taxpayer paying a higher tax must be imposed by an act passed by not less than two-thirds of all members elected to each of the two houses of the Legislature.10

This new rule prevents the simple tax expenditure strategy for evading TILs. The kind of shifts in tax burden required to achieve a revenue-neutral package will increase the taxes on at least "any" taxpayer (one will do!), meaning that any such proposal requires a supermajority.

As for this new rule, we should immediately note that, as we suggested in our previous column,11 its apparent success comes at a great cost. Proposition 26-like TILs interfere with traditional "base broadening plus rate lowering" tax reform -- the model for traditional, and efficient, bipartisan tax reform. That is because closing any tax loophole increases the taxes on "any taxpayer" even if the overall package reduces rates on most taxpayers. The Tax Reform Act of 1986 and all the various bipartisan proposals now floating about Washington and the states would require a supermajority under this new rule.

But Proposition 26 does not prevent the BCPP strategy. Proposition 26's supermajority rule applies to "taxes," but taxes are defined to exclude benefit-type charges. For instance, the following is not a tax and is thus not subject to the supermajority rule:

    A charge imposed for a specific government service or product provided directly to the payor that is not provided to those not charged, and which does not exceed the reasonable costs to the State of providing the service or product to the payor.12

Thus, the California State Legislature (or local governments)13 could increase benefit charges by majority vote just as it could add tax expenditures by majority vote, and so even Proposition 26-style TILs can be evaded.

Perhaps the next step is for TILs to include all possible government charges, as the TIL-provision in Missouri seems to do:

    Counties and other political subdivisions are hereby prohibited from levying any tax, license or fees, not authorized by law, charter or self-enforcing provisions of the constitution when this section is adopted or from increasing the current levy of an existing tax, license or fees, above that current levy authorized by law or charter when this section is adopted without the approval of the required majority of the qualified voters of that county or other political subdivision voting thereon.14

Despite the broad language restricting "any tax, license or fee," the Missouri Supreme Court has interpreted this language not to require elections in connection with "fee increases which are 'general and special revenues' but not a 'tax,'" holding "that increases in the specific charges for services actually provided by an ambulance district are not subject to [the Missouri TIL]."15

And even without a favorable judicial interpretation of this sort, a variation on the BCPP strategy would still be viable. All that would be needed would be to partially privatize state spending programs (like higher education), while keeping them highly regulated so that they continue to operate in a fashion similar to how they were run as state spending programs, and then providing tax credits for payments made to those new quasi-private entities.16 In other words, no fee that could be argued to be a tax would be required.

Moreover, there is a sound reason why proponents of TILs, and courts in interpreting their intentions, have not applied TILs to benefit charges. As a practical matter, what would it be like for tuition at state colleges, public parking rates, building permit fees, and so on all to be subject to a supermajority requirement?17 As a matter of theory, why should the voters impose extraordinary constraints on legislators in connection with charges that they, the voters, will usually have to pay only by opting to engage in a voluntary activity?

How Should These Strategies Be Evaluated?

Let us assume that these strategies have been explicitly used.18 Should we consider these strategies to be unscrupulous dodges? We think not. There is ample evidence that voters desire both lower taxes and increased spending on all of the major programs on which governments spend significant resources.19 TILs are one outgrowth of that bias in voters' fiscal preferences. That bias is particularly problematic because voters appear to have little understanding of what state governments actually do. And thus these evasions are perhaps just a way of responding to voters' inconsistent demands regarding taxes and spending.

But there is a deeper point to be considered about the shift to benefit-type taxes. The primary mechanism through which state governments have historically functioned in the face of voters' fiscal biases and irrationality is through representative government. Elected representatives made the hard choices that voters were unwilling to make. But TILs undermine the elected representative model; they stem from voters' loss of trust in that model.

The benefit taxes model can function as a partial replacement for the elected representative model. In effect, the benefit taxes model relies on the market to determine fiscal priorities, because taxpayers must pay for more of the costs of governance through direct benefit charges. Instead of wholly relying on elected representatives to determine fiscal priorities, the benefit taxes model relies much more on the choices made by individual state citizens acting as consumers. By using the BCPP strategy -- combining benefit charges with progressive refundable tax credits -- a state can use the market-based benefit taxes model for making allocative fiscal decisions while continuing to use the elected representative model for making distributive fiscal decisions.20 Markets are generally superior to elected representatives at making allocative decisions, but markets on their own are not capable of enacting most forms of distributive policies that voters might desire.

It is generally (and correctly) maintained that, by mimicking the market to the extent possible, providing a service with a benefit charge should usually be more efficient than paying for a service with a general tax. By directly connecting payments to the services received, benefit charges mitigate the incentives to change behavior that results in traditional forms of taxation creating excess burden (also known as deadweight loss).21 The use of the BCPP strategy can thus limit the size of government in at least two ways. First, to the extent that benefit charges better reflect the level of government services that people want, benefit charges are more politically efficacious in shaping the government in accordance with the voter's wishes. Second, to the extent that benefit charges raise funds while creating less excess burden or deadweight loss, benefit charges reduce the distortionary effect that government activity imposes on the larger economy.

Toward a Better Approach for Fiscal Accountability

Our second -- and deeper -- strategy for evading TILs, BCPP, is in part justified because this kind of tax system better controls the size of government. Of course, controlling the size of government is exactly what TILs are supposed to do. At this point a supporter of TILs might breathe easier. Surely two ways to limit the size of government are better than one?22 That is not true if the two ways undermine each other. We have already seen how BCPP undermines TILs. How do TILs undermine BCPP?

The main answer is that TILs undermine jurisdictional competition. The BCPP solution to the size of government conundrum in part results from better matching of individual citizens to individual services, but, to reach its full potential, this solution must also match individuals to entire jurisdictions (in a Tiebout fashion) -- The heart of the Tiebout Model is that individuals will "vote with their feet" by moving to jurisdictions with their preferred level of taxes and services). There are only so many government services, such as higher education, that can be provided individually. Critical government services, particularly at the local level, tend to come in bundles -- for example, K-12 education, police protection, and parks.23 There is a local government levy that, to some extent,24 acts as a blended price for those local amenities, and that is the property tax. Yet TILs at the local level obstruct the proper functioning of jurisdictional competition because with them localities cannot modulate their rates in competition with one another.25

TILs at the state level have a similar effect on interstate jurisdictional competition. Suppose, for instance, that California did want to become more like Texas. California could lower or abolish its non-Texas taxes (for example, the corporate income tax) by majority vote, but adding a new Texas tax (that is, the margin tax)26 or increasing an existing superior tax (that is, the property tax) or improving another existing tax (for example, taxing sales of services) would require a supermajority (or, in the case of the property tax, a constitutional amendment). Another debilitating effect of local TILs is that they put pressure on state budgets to fund services that could have been more efficiently funded locally. Not only is that a less efficient use of revenue, but also states typically rely on more volatile revenue sources, and thus TILs at the local level increase volatility at the state level -- yet state-level level TILs then make it more difficult for states to adjust their tax rates to cope with that combination of greater responsibility and volatility.27


As we have argued over these two columns, TILs should not be assumed to shrink state governments. Instead, TILs primarily serve to undermine the effectiveness of government programs without necessarily reducing the size of government. That is reason enough to eschew TILs. But there is another reason to avoid TILs and that is that they actively impede a superior means of controlling the size of government, namely the increased use of benefit taxes and jurisdictional competition. Those alternative approaches to managing the growth of government are far from panaceas, but, unlike TILs, they are analytically coherent.28

Academic Perspectives on SALT is a column by David Gamage, an assistant professor at the University of California Berkeley School of Law (Boalt Hall), and Darien Shanske, an associate professor at University of California Hastings College of the Law. During the 2011-2012 academic year, Gamage is on academic leave from Berkeley while serving as special counsel and senior Stanley S. Surrey Fellow at the U.S. Department of the Treasury, Office of Tax Policy. Nothing written in these columns should be attributed in any way to the Treasury Department, the Obama administration, or anyone other than the authors.

* * * * *


1 David Gamage and Darien Shanske, "On Tax Increase Limitations: Part I -- A Costly Incoherence," State Tax Notes, Dec. 19, 2011, p. 813, Doc 2011-25440, or 2011 STT 243-3. See also, Michael Leachman, Nicholas Johnson, and Dylan Grundman, "Six Reasons Why Supermajorities are a Bad Idea," State Tax Notes, Feb. 27, 2012, p. 703.

2 This language is from Calif. Const. Art. 13A, section 3, which was California's TIL before the passage of Proposition 26 in November 2010. See also, e.g., Arizona Const. Art. 9, section 2.

3 For some examples, see Gamage and Shanske, supra note 1, at 814-815.

4 And enacting that strategy through business tax expenditures would be even easier.

5 For more on this approach, see generally Darien Shanske, "Going Forward by Going Backward to Benefit Taxes," California Journal of Politics and Policy, vol. 3, issue 2, art. 14, available at

6 For further discussion of how this works, see, e.g., Edward A. Zelinsky, "The Health-Related Tax Provisions of PPACA and HCERA: Contingent, Complex, Incremental and Lacking Cost Controls," Cardozo Legal Studies Research Paper No. 301 (2010), available at (summarizing the many tax provisions of the Affordable Care Act).

7 There are trade-offs involved in making tax credits advanceable. Doing so requires a reconciliation process, wherein taxpayers whose income ends up being higher than predicted might be required to pay back excess advanceable credits received, which could create complicated enforcement issues. We cannot fully analyze those trade-offs here. Instead, we merely mean to point out the possibility of making credits advanceable in order to deal with liquidity issues.

8 Refundable tax credits can be implemented through other state taxes in addition to the income tax, although doing so is somewhat more complicated.

9 See, e.g., Lenny Goldberg, "California Governor Approves Gas Tax Swap," State Tax Notes, Mar. 29, 2010, p. 903, Doc 2010-6510, or 2010 STT 58-2 (describing complicated revenue-neutral package); Proposition 26, Findings, sec. 1(d) (new stricter TIL seems to target this "swap"); Calif. Const. Art. 13a, section 3(c) (Proposition 26 is retroactive to Jan. 1, 2010, and thus appears to invalidate the swap); AB 105, 2011-12 (61.2011) (California Legislature modifies swap in part because of Proposition 26).

10 Current Calif. Const. Article 13A, section 3(a).

11 Gamage and Shanske, supra note 1, at 817.

12 Calif. Const. Article 13A, section 3(b)(2).

13 Proposition 26 made parallel changes to the ability of local governments to raise taxes. See Calif. Const. Article 13C, section 1(e).

14 Mo. Const. Art. X, section 22.

15 Keller v. Marion County Ambulance District, 820 S.W.2d 301, 303-305 (Mo. banc 1991); see also Arbor Investment Company, LLC v. City of Hermann, 341 S.W. 3d 673 (Mo. banc 2011) (reviewing history of tax-fee jurisprudence in Missouri, affirming use of five factor test as useful, and then affirming lower court finding that utility charges were fees and not taxes).

16 The eligibility for tax credits for payments to those quasi-private entities could be made conditional on the quasi-private entities complying with state regulations. In that fashion, the state can ensure that tax credits are issued only to the extent that those entities fulfill a public purpose in a similar fashion to how the entities would have been run had they remained state spending programs rather than quasi-private entities.

17 That is not to say that provisions like Proposition 26 and its predecessor, Proposition 218, do not complicate using benefit-type financing. As we observed in our last column, the ambiguities in both measures have resulted in an enormous amount of litigation and uncertainty. Gamage and Shanske, supra note 1, at 816; see also, supra note 15 for some of the resulting litigation in Missouri. On California's Proposition 218, which targeted special assessments, a particularly venerable and potentially useful type of benefit charge, see, for example, Darien Shanske, "Clearing Away Roadblocks to Funding California's Infrastructure," State Tax Notes, Nov. 23, 2009, p. 567, Doc 2009-22866, or 2009 STT 223-7.

18 In this column we can only offer educated intuitions about the extent to which these evasion strategies are actually in use, but we do think that these strategies are being used, at least to some extent (even if implicitly). In the last decades, and particularly since the imposition of limitations on the local property tax, there has been an explosion in the use of benefit-type charges. There has also been an increase in the use of state-level tax expenditures, though those two phenomena have generally not been explicitly linked. And, in at least some instances, state legislatures have explicitly made use of our first evasion strategy, passing revenue-neutral packages in order to evade TILs.

19 See, e.g., David Gamage and Darien Shanske, "Three Essays on Tax Salience: Market Salience and Political Salience," 65 Tax L. Rev. 19, 96 (2011); David Gamage, "Managing California's Fiscal Roller Coaster," State Tax Notes, Sept. 8, 2008, p. 659, Doc 2008-17246, or 2008 STT 175-12.

20 For the original distinction between allocative and distributive fiscal policy, see Richard A. Musgrave, The Theory of Public Finance (1959).

21 For discussion of these concepts, see David Gamage and Darien Shanske, "Three Essays on Tax Salience: Market Salience and Political Salience," 65 Tax Law Review 19, 61-65 (2011).

22 See, e.g., Geoffrey Brennan and James M. Buchanan, The Power to Tax 197-198 (1980) (arguing how limits on the property tax can be a complement to jurisdictional competition as a means of controlling the size of government).

23 Darien Shanske, "Above All Else Stop Digging: Local Government Law as a (partial) Cause (and Solution) to the Current Financial Crisis," 43 Michigan Journal of Law Reform 663, 703-704 (2010) (discussing local amenity "bundling rules").

24 See, e.g., Darien Shanske, "How Less Can Be More: Using The Federal Income Tax To Stabilize State And Local Finance," 31 Virginia Tax Review 413, 455-458 (2011) (reviewing the evidence and concluding that there is an argument that property taxes function as benefit taxes at least partially).

25 William Fischel, The Homevoter Hypothesis 98-128 (2001).

26 California's proposed Business Net Receipts Tax resembled Texas's Margin Tax. California Commission On The 21st Century Economy, Final Report (2009), available at

27 For further discussion, see David Gamage, "Preventing State Budget Crises: Managing the Fiscal Volatility Problem," 98 California Law Review 749 (2010).

28 And, not surprisingly, there is some empirical evidence that they are effective. See, e.g., Fischel supra; Wallace E. Oates, "The Many Faces of the Tiebout Model," in The Tiebout Model at Fifty 21, 34-37 (William A. Fischel ed., 2006). Even critics of the Tiebout model on normative grounds acknowledge its relative explanatory power. See, e.g., Richard Briffault, "Our Localism: Part I -- Localism and Legal Theory," 90 Columbia Law Review 346, 405-406, 416-417 (1990).


About Tax Analysts

Tax Analysts is an influential provider of tax news and analysis for the global community. Over 150,000 tax professionals in law and accounting firms, corporations, and government agencies rely on Tax Analysts' federal, state, and international content daily. Key products include Tax Notes, Tax Notes Today, State Tax Notes, State Tax Today, Tax Notes International, and Worldwide Tax Daily. Founded in 1970 as a nonprofit organization, Tax Analysts has the industry's largest tax-dedicated correspondent staff, with more than 250 domestic and international correspondents. For more information, visit our home page.

For reprint permission or other information, contact