Money Moves: Taxing the Wealthy at the State Level

It’s widely understood today that inequality is a major social problem that in turn contributes to other crises. By most accounts, tax systems are supposed to be our engines of equality. Yet in today’s United States, state and local tax systems mostly do the opposite: They take a greater percentage of the resources of the poor and middle class than of the rich.

Perhaps surprisingly, the traditional view among fiscal policy experts has been that this state of affairs is correct. In this standard account, only national governments should impose progressive or redistributive taxes. While acknowledging that there would be advantages to redistributive state taxation if it could be done efficiently, many experts worry that taxing the wealthy at the state level would drive taxpayers to move to a neighboring jurisdiction with lower rates, resulting in greater economic distortions and potentially little or no additional tax revenue. Similarly, politicians and advocates have opposed recent state efforts to tax the wealthy by arguing that such taxes will drive away the rich.

This Article argues that this traditional view is misguided. Recent evidence finds that relatively few wealthy households actually move in response to changes in tax policy. On the other hand, the location of taxable income—the place where wealth is legally subject to claims of the state—is quite responsive to tax rates due to a bevy of now-standard forms of tax gaming that we detail.

This distinction is highly significant because while physical relocations are hard to prevent, and, indeed, are good for a healthy federalism, the shifting of taxable income across borders has some ready legal solutions. As we detail here, a key feature of most state tax-avoidance schemes is the exploitation of the realization rule, the tax principle that imposes tax on appreciated property only when it is sold. States can greatly undercut this tax avoidance by instead imposing wealth or “mark-to-market” taxes on assets as they appreciate. Thus, critics of state wealth tax efforts have things exactly backwards: Rather than mobility making wealth taxes self-defeating, wealth taxes can counter tax-avoidance mobility.

Accordingly, we outline here how a truly progressive state tax system could operate. Building on earlier work, we show that standard critiques of wealth and mark-to-market taxes, such as that they would struggle to tax hard-to-value assets, are relatively easy to design around. We additionally explain other anti-avoidance rules that address some of the common techniques used by the wealthy to avoid state tax, such as trusts, partnerships, over-stuffed retirement accounts, and private foundations. With these new anti-avoidance tools available to states, we argue, the standard economic account shifts to favor truly progressive state tax systems.

Table of Contents Show

Introduction

By most measures, America is currently more unequal than it has been for at least a century.[1] Despite a nominally progressive federal tax system, the wealthiest tend to pay significantly less in federal taxes than the merely affluent.[2]

It might be thought that Americans just don’t mind this inequality, but that is not so. Not only do Americans dislike inequality, but they also want to do something about it. Large majorities of respondents in most surveys want to raise taxes on the rich and would support measures such as new taxes on the wealth of mega-millionaires and billionaires.[3]

Federalism is part of the problem. By most accounts, taxation is the most effective tool for distributive justice.[4] Yet nearly all state tax systems actually worsen inequality.[5] Indeed, even nominally progressive state tax systems tend to do even worse than the federal government at taxing the very wealthy. And this is not a small side note to overall national progressivity, as states bring in one-third of all national revenues.[6] Why has federalism been such a failure in the domain of progressive taxation?

A standard story advanced by opponents of progressive taxation is that wealthy residents would flee states that tried to tax them.[7] Even among academic economists, some claim that since only the federal government can effectively redistribute income from the rich to the poor, states should instead rely on regressive revenue sources such as sales or property taxes.[8]

Even ignoring the data and focusing just on normative arguments, these claims neglect some key counterpoints.[9] Sharp differences in who pays tax between the states and the federal government can distort the entire federalist structure by creating incentives to devolve policy downward. Wealthy Americans who want to escape progressive federal tax can lobby for decentralized government, even for policies where national approaches would make much more sense, because decentralized policies funded through regressive state taxes cost them less. In addition, the basic logic of federalism—its potential for citizens to shop for the policies that best suit their preferences and for subnational governments to operate as laboratories of democracy—applies as much, or more, to preferences for distribution as to other policies. Without access to progressive taxation, state efforts to reflect their citizens’ preferences may be badly limited.

Equally important, though, the argument that states just can’t tax the rich is wrong on at least one key fact or at least is greatly exaggerated. Studies of both progressive taxes generally and wealth taxes in particular find that, on net, there is relatively little physical migration in response to higher taxes, even among the wealthiest.[10] News accounts tend to highlight the occasional billionaire relocation, such as Jeff Bezos’s new home in Miami,[11] speculating, often without evidence, that such moves are for tax reasons, while typically failing to note empirical research findings that tax-induced migration has, overall, small consequences for either tax revenues or economic prosperity more broadly.[12]

This is not to say that people do not move across state lines. They do, especially in their retirement years, when relocations from more expensive (and often colder) states to lower-cost (and often warmer) states may become attractive. Relocations to states in the Southeast and Southwest have been particularly growing in recent years. Yet relative land and housing prices imply that Americans are still willing to pay large premiums to live in states like California and New York, at least during their pre-retirement years. Overall, Americans choose whether and where to relocate for a variety of interconnected reasons. The empirical literature finds that state-level tax policies are a relatively minor factor. As we will explain, there are some limitations and caveats to this empirical literature, and we ultimately urge caution in interpreting its implications. Nevertheless, despite numerous statements to the contrary by politicians and the media, there is ultimately no empirical support for the claim that state-level progressive taxation would induce sufficiently large numbers of people to move to make these policies impractical or self-defeating.

Why, then, have states mostly failed to take advantage of opportunities to implement progressive taxes? Our answer is that, although people themselves don’t seem to move much for tax reasons, money is substantially more mobile.[13] Just as multinational businesses have mastered the art of “stateless income” taxed by no jurisdiction,[14] the American rich have developed a wide range of tax-gaming moves that allow them to accumulate, and even spend, vast wealth that lies beyond the reach of their home states. Julie Roin aptly labels this phenomenon “exploitative mobility.”[15] Mobility is generally not a bad thing; it becomes exploitative when taxpayers “can extract benefits from one jurisdiction while escaping some of the costs of providing those benefits.”[16]

The tools to exploit mobility of money (as opposed to mobility of natural persons) are many.[17] There is, in fact, an entire so-called “wealth-defense” industry that self-consciously provides such techniques.[18] In many cases, no exotic planning is necessary. For example, Washington state’s richest man parks most of the income derived from his inventions inside a corporation (a little venture called Microsoft—maybe you’ve heard of it). Microsoft itself, in turn, exploits rules for multinational firms to report most of its income in Puerto Rico, Ireland, and other tax havens.[19]

Or take the case of baseball superstar Shohei Ohtani. His new contract with the Los Angeles Dodgers is structured so that the bulk of the $700 million he is owed will be deferred until after he has presumably left California; in this way, he will avoid paying California’s relatively high personal income tax rate (now 14.4 percent) on most of his earnings.[20] Some commentators point to this contract as proof that “wealthy people will find ways to avoid earning money in California.”[21] But, in fact, Ohtani has contracted to both play baseball and earn money in California. He is attempting to avoid paying tax in California through tax planning (“money moves”), not by physically moving. And, as we will explain below, if California were to adopt some of the reforms we propose in this Article,[22] then Ohtani would have to pay tax to California consistent with his California-based earnings.[23]

Behind nearly all of these tax-avoidance games is one simple concept: the so-called “realization rule.”[24] The realization rule is an income tax principle that only requires taxpayers to include a profitable investment in income at the time their property is sold, and not before.[25] It’s a handy simplifying convention for an income tax, making it easier to know what a property is worth.[26] But because wealthy people can control and access their wealth without actually selling it, the realization rule unlocks a vast array of tax-minimization strategies at the federal and state levels.

At the federal level, the realization rule enables a series of tax-planning strategies, often labeled as “buy, borrow, die.”[27] Wealthy taxpayers avoid selling their appreciated assets and thus defer paying income tax during their lives. Then, they completely escape their deferred tax responsibilities upon death, thanks to another provision of the income tax, known as stepped-up basis.[28] When it comes to state-level taxes, wealthy taxpayers need not even wait until death: They can buy, borrow . . . and then eventually move to a state like Florida or Texas (where there is no income tax) to escape deferred state-level tax responsibilities.[29]

Take Elon Musk as an example.[30] Musk built most of his wealth as a resident of California, benefiting from the services and protections offered by California while doing so. But because Musk did not take much salary or sell much stock while living in California, he paid minimal California income tax.[31] He instead borrowed to finance his investments and to pay for his lavish consumption. Because borrowed money is not subject to income tax, he thus paid very little tax to California.[32] Now, assuming that Musk has actually moved to Texas, as he claims, he can escape California’s income tax on most of the wealth he built while he was a California resident.[33]

The Musk example is, in some ways, unusual because Musk, apparently, actually moved (and did so pre-retirement), but, in many ways, it is typical because he was able to accumulate great wealth for years in California without paying commensurate taxes. This is because his great wealth mostly consists of appreciated stock. Thus, because of the realization rule, it was easy for Musk to avoid taxes in California. Only after having benefited from the realization rule did Musk apparently actually leave.

In this manner, realization-based income taxes create perverse incentives for taxpayers, who build up great financial wealth within states with substantial income taxes (like California), to wait until after they eventually move out of state before they realize those financial gains. These perverse incentives could be eliminated, or at least greatly alleviated, by reforming the design of state-level taxes.

We will explain how it is relatively straightforward, using modern tax administration principles, including information reporting, to collect a progressive tax without relying on the realization rule. Indeed, with our economist colleague, Emmanuel Saez, we have helped to author bills in California[34] and Washington[35] to impose an annual wealth tax on individuals with extremely high net worth (over $50 million in the California proposal and over $250 million in the Washington state proposal), and bills in New York,[36] Illinois,[37] and Vermont[38] to collect income tax from similarly very wealthy households when their assets increase in value, regardless of whether they are sold. We also have helped to pen Senator Elizabeth Warren’s federal wealth tax bill[39] and the bill implementing President Biden’s “billionaire minimum income tax,” which, like the New York, Illinois, and Vermont bills, would impose a tax on the very rich as their assets appreciate.[40] Tax mavens call this latter approach a “mark-to-market” tax.[41]

Wealth and mark-to-market taxes of these sorts would go a long way toward solving the states’ taxable-income mobility problem. Under a residence-based wealth or mark-to-market income tax, a wealthy individual in New York or California would pay tax as their wealth grew, regardless of whether that wealth was accumulating inside a multinational corporation, a South Dakota trust, a complex multi-tiered partnership, or any of the dozen other supposedly out-of-state places the rich now move their money. If a wealthy taxpayer has been paying taxes on their income accrued within a state while a resident, then their moving does not pose the same threat to the integrity of state tax systems.

Thus, recent critics of these new state tax proposals have gotten their arguments exactly backwards. They claim that states cannot impose wealth or mark-to-market taxes on the rich because the wealthy are mobile.[42] But, in fact, a primary purpose and function of these taxes is to reduce the mobility of merely paper capital, allowing it to be taxed at the state level. Once capital is being taxed appropriately, physical taxpayer relocations would generally serve the positive goals described in the federalism literature.

To be sure, critics also have other arguments, such as the common refrain that it is too hard to implement taxation without the realization rule because it would be too difficult to measure a taxpayer’s net worth.[43] We have set out elsewhere a detailed explanation of how successful systems in other countries, such as the Swiss wealth tax (itself imposed not at the national but at the “canton” or Swiss state level), have approached this issue.[44] There, we also explain how using a few modest administrative improvements, such as allowing governments to collect payment in the form of notional equity claims against a portion of the taxpayer’s assets, can solve valuation and liquidity concerns.[45]

In short, if Swiss cantons can do a reasonable job, then so can U.S. states. Indeed, over forty states already impose corporate taxes on multijurisdictional businesses[46] and, among those that do not, such as Texas, most impose a different, but still complicated, business tax.[47] Thus, although taking our advice would likely require some building up of state capacity, this would be in proportion to the administration already in place for business taxes. Further, we should understand the baseline, in the cases we are concerned with, is that the states are reaching little of the income accrued in the jurisdiction by the ultra-rich. The goal is not perfection. If a state managed to raise even a substantial portion of the revenue that should be raised from a god’s eye perspective, as to the unrealized gains of the super wealthy, then that would be an important achievement.

Accordingly, in this Article, we extend our prior work to show how our proposed solutions to the valuation challenge also help states overcome some of their own administrative obstacles. For instance, we’ll explain how the notional equity claim helps a state make sure it can still take a taxpayer to court to collect an outstanding tax debt, even after the taxpayer (or their money) has moved somewhere that would otherwise be outside the personal jurisdiction of the state’s enforcement efforts.

In addition, we outline other design considerations for architects of a state wealth or mark-to-market tax. For example, we propose that, instead of a strict binary definition of “residency” that would subject a resident’s full wealth to state tax, states should instead tax only a portion of the wealth of new and departing residents to reflect the fact that the state’s claim on these assets is less than total, but also greater than zero. In the case of a mark-to-market tax, we offer recommendations for how to manage the ways in which the revenues from such a tax would swing up and down with the business cycle, such as by spreading out the taxation of gains (and refunds for losses) over multiple years.

We also consider state-specific rules for trusts, pensions, family foundations, and gifts, all of which currently feature prominently in common state-level tax avoidance strategies. For example, we propose reforms for states to tax large pension and retirement accounts of the very wealthy, such as Peter Thiel’s infamous billion-dollar individual retirement arrangement (IRA). There is little policy reason why states should add further benefits on top of federal exemption for these savings, and we explain how state wealth taxes, in particular, would allow states to sidestep federal statutory limits on the states’ ability to tax pension accounts.

In short, we argue that the central objection to progressive state taxation is not a theoretical or normative argument, but instead is a practical question about whether and how states can effectively tax mobile capital. We then dig into this practical problem and show how it can be overcome using a combination of tools and approaches tested worldwide, plus some new innovations we propose. With this practical barrier removed, the case in favor of significant state taxes on the rich becomes convincing, at least with respect to the states where supermajorities of voters say that they want more progressive taxation of the wealthy. State-level progressive tax reforms can help to reflect local preferences for distributional fairness, prevent serious distortions to the political economy of federalism, and, by offering smaller-scale experimental proofs of concept, help to advance the national project of progressive taxation.

I. Fiscal Federalism and State-Level Taxation in the United States

In this Part, we offer some important background theory on fiscal affairs in a federation and an overview of U.S. arrangements. The main takeaway we emphasize is that under conventional analysis, progressive state taxation is self-defeating because wealthy taxpayers will move away in response. If this problem could be overcome, however, theory suggests that there are important benefits to subnational progressivity.

A.      The Standard Story

Fiscal federalism is an economic framework for determining which level of government—assuming a federation with distinct national and subnational governments—should carry out any given governmental task.[48] Usually this prescription involves balancing competing considerations.[49] For instance, when there is significant popular disagreement about a policy, fiscal federalism would generally favor assigning that policy task to state and local governments, so that governments can pursue diverse solutions and citizens can opt for the government that matches their preferences.[50] On the other hand, when policies implicate significant externalities or spillovers across subnational borders, fiscal federalism would favor nationalization to help ensure that policymakers take all the impacts of their policies into account.[51]

Traditional models of fiscal federalism tend to support allocating all progressive taxation to the federal government.[52] The idea is that because wealthy residents can move from one state to another more easily than they can leave the country entirely, the federal government has a comparative advantage (over state governments) at taxing the wealthy.[53] This is not to say states can’t collect taxes or redistribute wealth at all; many taxpayers may be tied or drawn to a given state by business needs, family, or a fondness for snowshoeing, permitting states to impose taxes or regulations up until the value of those ties becomes strained.[54] When state-level taxation causes wealthy residents to leave for other states, the literature calls this “horizontal distortions” or “horizontal externalities”: A state that sets a low tax rate, in effect, imposes fiscal pressure on its neighbors to lower their own rates.[55]

As we have explained in prior scholarship, “[d]ue to these horizontal distortions, the ‘classic theoretical result’ of the fiscal federalism literature is that both distribution policy and the taxation of mobile capital are best left to the federal government rather than the state governments.”[56] We further explained that “[t]he rationale for this result is that state governments carrying out these activities generate all of the same problems as when the federal government does so, in addition to inducing horizontal distortions related to economic activity being relocated to other states.”[57]

In contrast, state and local governments arguably can make efficient use of some other revenue sources, but those revenue sources are comparatively regressive. Localities can more efficiently collect sales taxes and economically similar taxes, such as gross receipt taxes and even the state corporate income tax to some extent because it is typically apportioned by sales.[58] That is because these taxes are a kind of user fee, the price businesses must pay in order to sell to local consumers.[59] But these sales and sales-like taxes are generally regressive,[60] since poorer households tend to spend a larger share of their income on consumption—and are thus subject to proportionally more sales tax—than richer households.[61]

These observations lead some commentators to argue that the federal government should take on most or all progressive taxation and then fund state governments with grants or other forms of revenue sharing.[62] This is not just an abstract thought experiment: Revenue sharing is an important feature of the fiscal systems in Canada and other countries.[63]

At the same time, the literature recognizes that there would be serious costs to depriving states of the ability to deploy progressive tax systems. Voters in different states might have different tastes for distributive fairness and progressivity, so disallowing state-level progressive tax systems could limit voters’ ability to choose their preferred government policies.[64] Beyond differing state preferences, there are also likely divergent state capacities to engage in redistribution.[65] States that are home to valuable economic and amenity aggregations (think California and New York) should be able to impose more progressive taxes than states without such aggregations—and, in fact, they do so.[66] That is, the more attractive the state is to potential taxpayers, the greater its likely ability to redistribute without changing where the taxpayers choose to live.

It would, therefore, improve efficiency and democracy (in the sense of improving the fit between voters’ preferences and policy outcomes) to permit different states to pursue varying tax policy goals.[67] In addition, commentators often argue that state-level experiments with different forms of taxation can generate helpful lessons, both for other states and for the federal government, and can thereby help overcome political and administrative obstacles to enacting welfare-enhancing policies.[68] State-level experiments with early income taxes paved the way for the federal government to successfully enact the modern federal income tax in 1913.[69] Redistribution may also be a local or club good best provided at the local level.[70]

Some prior authors (including some of us) have therefore proposed hybrid systems in which the federal government uses grants, or designs its own tax system, to entice state governments to impose fewer externalities on one another, thereby freeing up states that prefer more progressive tax systems to do so without being undercut by their neighbors.[71] For example, matching grants or tax deductions can reward states for raising revenue.[72] Some of these tools may not even be intentional; the federal government might simply be indifferent to “vertical” externalities in which states are able to increase their own revenue by reducing the federal tax base.[73]

But these hybrid options have important limitations. The Supreme Court has said that federal grants cannot “coerce” states into compliance.[74] A grant system large enough to push states to adopt a new progressive income tax system could well raise constitutional questions under that doctrine.[75] Although shifting some of a state’s tax pain to the federal government can effectively encourage the state to impose taxes, it also means that the state has sharply diminished incentives to choose an efficient tax.[76] Some states might even enact taxes that are on the wrong side of the Laffer Curve (i.e., are losing money on net) once the costs to both states and the federal government are taken into account.[77] State fiscal needs often vary with the business cycle, but federal support rarely matches this pattern or, at best, does so only with a damaging lag.[78]

Ultimately, the current literature generally concludes that states should mostly not impose progressive taxes; however, this conclusion depends on an empirical question about the extent of taxpayer mobility.[79] If the voters in some states believe that the federal government is not doing enough to combat inequality through tax policy (and polling suggests that a supermajority of voters in many states do in fact believe this),[80] then theories of fiscal federalism support state-level progressive tax reforms. On the other hand, the literature claims that taxpayer mobility makes these taxes so inefficient that the benefits are not likely to be worth the cost. Assessing this is an empirical question, as it matters whether horizontal distortions are likely to be small or large.

Our argument hinges on the way that the case against state progressive tax depends on whether tax revenues do, in fact, move easily across states. In Part III, we explain that the empirical literature generally finds relatively small horizontal distortions caused by wealthy people moving to escape taxation. By contrast, horizontal distortions caused by wealthy people moving their money to escape taxation can be much larger. Crucially, we will argue that with wise design and some innovations we propose, state governments have the tools to reduce these larger distortions. As a result, our conclusions strengthen the case for progressive state tax systems.[81]

B.      The Modern Fiscal Picture

To further set the stage for our arguments, it’s worth noting that modern American fiscal arrangements don’t look much like the ideals of the fiscal federalism literature.[82] On the tax side, things are somewhat in line with theory. The federal government levies a relatively progressive income tax,[83] and modern state-level tax systems are increasingly regressive, mostly exacerbating inequality rather than combatting it.[84] States have largely abandoned the general taxes on wealth that they imposed throughout the nineteenth century.[85] Local governments do impose taxes on real property, but these turn out to be relatively regressive.[86] Moreover, modern wealth and inequality are mostly tied to financial assets, not bricks and mortar.[87]

As to the spending side, the current system does not closely approximate fiscal federalism theory. We note two deviations in particular. First, state and local governments have extensive fiscal responsibility to provide frontline government services, particularly public education.[88] In 2020, the United States spent about 3 percent of its GDP on K-12 education, slightly less than the Organisation for Economic Co-operation and Development (OECD) average.[89] Across the OECD, central governments provide 55 percent of funding,[90] whereas the federal government in the United States provides only 13.6 percent, with states picking up the rest.[91] As a result, education funding makes up about one-third of state expenditures.[92]

Second, states have huge responsibilities for safety-net programs and other efforts to fight recessions.[93] Over 40 percent of states’ expenditures go to public welfare programs such as Medicaid, Temporary Assistance for Needy Families, and Supplemental Security Income, which are funded through a mix of state and federal dollars.[94] By design, these safety-net responsibilities increase during a downturn.[95] Unemployment insurance benefits are paid for by state taxes, although federal funds backstop them during recessions.[96]

Fiscal federalism generally holds that subnational units should not be responsible for these kinds of large counter-cyclical spending obligations and that, if they are, the central government should then race in to help during downturns.[97] Among other reasons, this is because states do not internalize any economy-boosting or -sapping effects of their spending decisions for other states.[98] Public education is a partial example of this issue: It does not make a lot of sense to lay off lots of teachers every time there is a recession, yet that is often what states do. States also cannot easily save for recessions or borrow during them.[99]

In short, modern arrangements in the United States are a bit of a puzzle. Why do we engage in so much safety-net and redistributive spending at the state level and below? Perhaps policy makers believe that progressive state spending can make up for regressive financing. Or perhaps regressive financing is exactly the point. We consider those two possibilities in the next Part.

II. Why Should States Tax the Rich?

Before exploring the empirics of tax mobility, we want to introduce some additional arguments for progressive state taxation. Specifically, we argue that state tax progressivity is essential to what we call “federalism neutrality”; that it can’t be replaced by other forms of progressivity; and that it is, perhaps, encouraged by unique federal constitutional constraints. We think it is worthwhile to put these arguments on the table here explicitly because fiscal federalism is a balancing of competing considerations. As we noted in the last Part, other commentators generally agree that if mobility were not an overwhelming obstacle, state redistribution would be beneficial because it would better match voters with their preferences, better utilize local fiscal capacity, and, perhaps, better allow states to bear the fiscal burden of handling counter-cyclical spending. Here, we make some additional points because stronger arguments for state progressivity should lead to higher tolerance for increased inefficiency or horizontal distortions in order to achieve it. 

A.      Federalism Neutrality

Just as there can be sound reasons to prefer federalist solutions, such as matching citizens with their policy preferences, there can be perverse ones. Consider the incentives of politically powerful people and organizations who would bear higher tax burdens under a progressive tax system. If the central government funds its programs with progressive taxes but the states do not, then that potentially creates a perverse incentive for those likely to pay the progressive taxes to lobby for expensive redistributive programs to be pushed onto the states.

A system that allows this to happen fails what we call “federalism neutrality.” We define a fiscal system as “federalism neutral” when voters and other relevant political actors are indifferent to whether policies are funded at the national, state, or local level. That is, political players would have no reason to think that they will pay more or pay less based on which government is imposing the tax.

While, of course, government choices are complex and have many causes, we argue that the perverse incentives resulting from failure to achieve federalism neutrality can help explain the serious misalignment of spending and revenue we described in Part I.B. Failures of federalism neutrality can undermine efficient government by shifting policy responsibilities to levels of government that are poorly equipped to handle them. While powerful interests that lobby for devolution may also be harmed by the lower efficiency of assigning a policy to the state level, that cost is likely widely shared by society. Thus, the powerful interests’ losses from government inefficiency would likely be much smaller than their gains from shifting the tax burden.[100] All else equal, then, federations should aim for fiscal arrangements under which there are not likely to be large and systematic differences in the distribution of tax burdens between federal and subnational tax systems.[101]

Admittedly, all else will rarely be equal. As we described in Part I, there are often strong economic arguments for collecting revenue differently at the national, state, and local levels. Our point here is that there can be significant political-economy costs when the tax burdens that powerful political actors face vary substantially depending on which tier of government collects the tax. Thus, we suggest that the economic benefits of any particular tax system should be weighed against the potential harms of failing to minimize perverse incentives for devolution. Balancing these considerations might lead governments to make state and local tax systems more progressive than they would otherwise have been, without necessarily making them equally as progressive as the federal system.

B.      Spending is No Substitute

Next, we argue that states should be able to redistribute through their tax systems, rather than having to rely on alternatives, such as progressive spending policies. Some argue that even if state taxes are flat or regressive, states can target their spending to poorer households so that state policy is progressive overall.[102] In reality, states with more progressive taxes have historically also had more progressive spending,[103] but we want to at least consider the possibility that this could change.

One big problem with this approach is that any means-tested spending program is, in a sense, just another tax system under a different name. If each additional dollar you earn causes you to lose twenty cents, it doesn’t matter much whether that loss comes from a higher tax or a lower amount of benefits received.[104] For the person who is being taxed or receiving the benefit, their decision about whether to move in order to keep their twenty cents is the same.[105] Indeed, there is a large literature on the extent to which mobility constrains states from providing more generous safety net programs.[106]

Further, it is likely very difficult for spending programs to make important distinctions among relatively wealthy households.[107] For the wealthiest families, government benefits represent such a tiny share of household income that the benefits are almost meaningless to them.[108] Taxes, of course, do not have that constraint, so tax systems are likely much more effective at addressing inequality between the mega-rich and everyone else.

Lastly, shifting redistribution over to spending programs does little to help the federalism-neutrality problem. Reducing or shifting the benefits from a program is mostly an externality for the interests that want to relocate policies to the state level to reduce their tax bills. For example, moving school lunches off the federal budget and onto state ledgers would predictably save high earners money, even if states respond by phasing out meal benefits for wealthier families.

C.      Constitutional and Veto-Gate Considerations

One final observation we should make is that states may be the only legally or politically viable place to pursue wealth or mark-to-market taxation, at least initially. Polling and other evidence suggests that the majority of voters in the nation actually favor both more progressive taxation (more generally) and wealth tax and mark-to-market reforms (more specifically).[109] Nevertheless, the prospects for enacting such reforms federally are challenging due to both the numerous veto points that proponents must overcome[110] and the constitutional limitations that apply specifically to federal-level tax reforms. As to the latter, many commentators predict that the current U.S. Supreme Court would strike down an unapportioned federal wealth tax or mark-to-market reform.[111]

The recent case of Moore v. United States heightened these constitutional fears.[112] As one of us has argued previously, even if the Supreme Court were to strike down an unapportioned federal wealth tax or mark-to-market reform, modern fiscal tools allow the federal government to levy an apportioned version of such reforms equitably and effectively.[113] Nevertheless, the politics of apportionment could be challenging. Explaining further, and also what is meant by apportioned versus unapportioned reforms, is beyond the scope of this Article. The key point for this Article’s purposes is that state-level reforms would not be subject to these constitutional constraints and related political challenges.[114]

Thus, if veto points or constitutional limitations that apply just to federal-level reforms would hinder the passage of such reforms despite the preferences of strong majorities of voters, then state governments could potentially operate as a kind of “repair shop” to jump-start these policies.[115] As we have already noted, it has long been understood that an advantage of federalism is better tailoring policies to the preferences of local voters. Thus, if the voters in some states have stronger preferences for progressivity, then this supports pursuing more progressive tax reforms in those states.

This is not to say that state-level adoption of these policies faces no obstacles. Unlike the federal government, many state constitutions have special legislative supermajority requirements for tax increases.[116] However, the diversity of states—with differing rules, institutions, and politics—may make it much easier to pass reforms in at least some states than at the federal level. Notably, in many states with strict legislative supermajority requirements to raise taxes, there is a robust tradition of direct democracy.[117] Thus, state tax increases that cannot pass the legislature might well pass as ballot measures.

*          *          *

Overall, we think the current condition of U.S. fiscal federalism supports a potentially powerful case for reform. Regressive state financing contributes to overall inequality and may be a driver of the decision to allocate so much safety-net spending responsibility to states. And this spending consequently does little to counteract many important dimensions of inequality.

III. On the Mobility of Wealthy People and of Their Money

We have argued so far that there are good reasons for states to pursue progressive tax reforms, but is that a realistic goal? A key argument against progressive taxation at the subnational level, and especially against taxing the very wealthy, is that taxes could drive the rich away. For example, when California was debating a tax hike on its higher-bracket households in 2011, the California Senate minority leader quipped: “There’s nothing more portable than a millionaire and his money.”[118]

He was half right. In this Part, we argue that while there are some significant practical obstacles to progressive state-level taxation, the out-migration of wealthy residents themselves should not be a primary consideration—but with some caveats and cautionary notes that we will explain further below. While some millionaires do sometimes move after states raise taxes, those relocations have only minimal impact on revenues. The empirical literature simply does not support the claim that wealthy people’s movement decisions are highly responsive to taxation.

Money, on the other hand, is much more portable. We sketch a few of the many ways in which wealthy individuals can readily shift their taxable income across state borders to minimize their tax burdens. We then show how these opportunities help to explain some of the observed empirical findings on the impact of progressive wealth or income taxes. That is, the dominant proportion of the “large” mobility results found in certain studies seemingly consists of paper mobility. Accordingly, we subsequently explain in Parts IV and V how state governments can use tax reform solutions to combat the ways in which money can currently move to escape state-level taxation. In other words, we argue that the movement of money is not an inherent limitation on state-level taxation but rather should be largely understood as a design problem.

A.      On the (Non)Mobility of Wealthy People

By nearly all accounts, the recent empirical literature does not support claims that the migration of wealthy people poses a substantial constraint on states’ taxing capacity. As John R. Brooks explained in 2014, the evidence that migration substantially limited subnational redistribution was then weak, and data in the last decade have strongly confirmed that view.[119] For example, a thorough review of the evidence commissioned by the U.K. government concluded in 2022 that “actual migration responses are small.”[120] In other words, although people do move, their decisions as to whether and where to move do not appear to be heavily influenced by tax considerations.

We will not repeat the work of these thorough literature reviews here[121] but rather will explain a few important points and some especially relevant recent studies. Because most of the literature has focused on existing realization-based income taxes (along with business taxes and the like), the research on Spain’s recent experience with subnational wealth taxes is especially interesting for the purposes of this Article. Prior to 2008, when it was suspended for two years, the Spanish wealth tax was mostly levied and administered in a uniform manner across different regions of Spain, meaning that moving regions would not have had substantial tax consequences for migrating residents.[122] This changed with the reintroduction of the wealth tax in 2011, after which different regions began enacting substantially different tax rates and exemptions. Notably, Madrid effectively abolished its regional wealth tax (establishing an effective 0 percent tax rate), whereas other regions in Spain levied progressive wealth tax rates ranging from 0.16 percent to 3.75 percent.[123]

As the leading academic study on the effects of the Spanish wealth tax explains, the tax savings of moving from a region with a higher wealth tax rate to Madrid “are sizeable.”[124] This is because a number of regions in Spain levy very high top wealth tax rates—much higher than those that have been proposed for state-level wealth tax reforms in the United States. For instance, compare the top 1.5 percent rate of California’s proposed wealth tax or the top 1 percent rate of Washington state’s proposed wealth tax with the top rates in higher-tax Spanish regions—2.75 percent in Catalonia or 3.75 percent in Extremadura.[125] Depending on certain assumptions, these Spanish rates can impose an economic burden comparable to a capital-gains tax of upwards of 60 percent, triple the U.S. federal rate.[126]

Yet, despite these relatively high top rates levied in these regions and the large tax savings that can thus be achieved from moving from one of these high-tax regions to Madrid, the study finds that mobility responses are relatively small as compared to the revenues raised by Spain’s wealth tax.[127] Specifically, the study finds that regions other than Madrid only lose about 5 percent of the potential revenue that would otherwise have been raised from their wealth taxes due to mobility responses, plus an additional loss of 2.5 percent of personal income tax revenues.[128] Thus, although these mobility responses are measurable and significant, they do not greatly undermine the revenue-raising potential of Spain’s subnational progressive taxes.

Moreover, the study finds that “nearly all tax-induced mobility is driven by the salient zero-tax region of Madrid.”[129] Despite the quite significant tax rate differentials among regions other than Madrid, the study found negligible tax-induced mobility between these regions.[130]

Other studies of tax-induced migration responses find similar results. Consider a study of California’s 2012 tax hike on millionaires, authored by two economists at the conservative Hoover Institute, Joshua Rauh and Ryan Shyu,[131] which has been trumpeted by many opponents of progressive state-level tax reforms.[132] This study finds a dramatically larger responsiveness of wealthy taxpayers to California’s 2012 tax hikes than most other studies in the literature.[133] Yet only a small portion of this responsiveness is due to taxpayers leaving California, with over 90 percent of the reported responsiveness instead arising from reductions in taxable income by taxpayers who remain in California.[134] The authors note that there are a number of mechanisms that could explain how taxpayers remaining in California reduced their reported taxable incomes, including “labor supply effects, offshoring to other countries, shifting of sales of pass-through businesses to other states under California’s single sales apportionment rule, and shifts to forms of compensation for which taxation is deferred.”[135]

In other words, most of this reported responsiveness may be a result of taxpayers moving their money to escape tax, as we discuss further below. The study found that California only lost about 4.2 percent of the revenue that would otherwise have been raised from its tax hikes due to taxpayers leaving the state.[136]

Thus, although both the study of Spain’s wealth taxes and the study of California’s 2012 income tax hikes find mobility responses that are measurable and significant, the migration responses reported by both studies are relatively small as compared to the revenue-raising potential of the taxes being studied. Moreover, neither Spain’s wealth tax nor California’s income tax has implemented the reform measures we propose in Part IV to mitigate migration responses. As we will explain, implementing these reforms would likely have further alleviated the migration responsiveness (or at least the revenue loss from the migration responsiveness).

In our view, the most convincing study is one from 2016, based on confidential IRS data on the universe of million-dollar earners in the United States from 1999 to 2011.[137] The authors of that study concluded:

The most striking finding of this research is how little elites seem willing to move to exploit tax advantages across state lines in the United States. Millionaire tax flight is occurring, but only at the margins of statistical and socioeconomic significance . . . . [M]illionaires are not very mobile and actually have lower migration rates than the general population.[138]

There are some subcategories of the rich who do seem to be more apt to move in response to tax. “Super-star” inventors who hold multiple patents tend to relocate to low-tax places more than members of the general population, as do some European soccer players and entertainers.[139] These are logical exceptions because these groups include earners who move far more than other individuals and whose ability to bring in money is relatively unattached to any one place.[140] Patent-holders are the most mobile, for instance, when they already work for a multinational business and have an alternative workplace to which they can readily relocate.[141] In contrast, most revenue is deeply embedded in a community: Workers, customers, and suppliers cannot easily be uprooted, and it is difficult to manage a successful business from afar.[142] Studies that have examined the question indeed find that business owners are considerably less mobile than other millionaires, particularly during their working years.[143]

One Swiss study did observe somewhat higher migration responses than the other studies we have surveyed, finding that claimed relocations cost about nineteen cents of every potential dollar that would have been collected without any moves.[144] But that was the result of tax differences between two adjacent, relatively tiny (total population: 1.5 million) German-speaking cantons, connected by Switzerland’s famously extensive and efficient public transit.[145]

In the U.S. context, researchers find some tax responsiveness in border regions and cross-state metro areas, but only incrementally so.[146] It is thus worth cautioning that our conclusions about the limited impact of interstate mobility do not necessarily apply when considering metro areas sitting at the boundaries of a higher-tax state and a much lower-tax state. In these cases, taxpayers could just move neighborhoods to escape some tax while remaining in the same general region.[147] But we view these as exceptions that are consistent with our general conclusions: For the most part, the empirical literature finds relatively small migration responsiveness of wealthy taxpayers to taxation.[148]

Moreover, there is evidence that a good portion of this supposed “migration” is actually fake.[149] Infamously, tax authorities struggle to know where taxpayers really live, as it can be challenging to verify whether a beach house or pied-à-terre is really a primary residence or vice-versa.[150] In Spain, for instance, researchers concluded it was plausible that all claimed relocations by second residence owners in Madrid were tax evasion.[151] That is, much or even most of the measured mobility responses to the Spanish wealth tax were likely a result of taxpayers with second homes in Madrid fraudulently claiming to have moved their principal residence to Madrid without actually having done so.[152] These sorts of fraudulent “paper” migrations could potentially be prevented with stronger auditing and enforcement policies.[153]

Similarly, in the U.S. studies, many of the “migrations” observed by researchers involve changes from full to part-year residency, a change that can bring major tax benefits (as we detail more below).[154] Of course, tax fraud reduces revenues, just as real relocations do, but the policy implications of fake moves are quite different from their real counterparts. For example, tax authorities have had considerable success in some jurisdictions in cracking down on fraudulent relocations,[155] whereas we likely would not want to (and perhaps constitutionally could not) constrain real moves.

We acknowledge, however, that mobility responses to tax depend on the total tax enforcement environment. In a world where fake moves are easy or, as we argue below, where individuals can easily move money without physical relocation, taxpayers may have little incentive to relocate. Tighter enforcement on these other margins might result in more actual relocations.[156] But available evidence is reassuring on this front. Enforcement crackdowns in Norway, for instance, did not result in any measured increase in real relocations.[157] And, as we will argue, the tax rewards of moving are much smaller if common abusive techniques are curtailed, so that tax enforcement may reduce real mobility rather than increase it.

We also acknowledge that there is limited available evidence on whether subnational taxes affect initial decisions about where to live or start a business, as opposed to moves by entrepreneurs who have already succeeded.[158] Yet theory predicts that these initial location choices should be less sensitive to tax rates because the local tax is more of a benefit tax. That is, as we already mentioned, people and businesses decide where to start out based on the combined bundle of costs and benefits a jurisdiction offers. In contrast, the retired entrepreneur has already reaped many of the benefits of, say, a dense network of other nearby entrepreneurs, and may be more focused on the tax rate.

In any event, to the extent that taxes affect initial location decisions, we view that more as a factor to be considered in setting the tax rate rather than the tax structure. For states, mark-to-market and wealth taxes are likely far more efficient than a realization-based system. Even if it makes sense to set taxes low in order to attract entrepreneurs, we argue that it would be foolish to do so by creating a maze of complex tax dodges that lower effective rates by exploiting the realization rule. Under such a structure, low rates are available only to entrepreneurs who pay a lot for tax lawyers and engage in economically wasteful forms of tax planning.    

B.      E Pur Si Muove[159]: On the Mobility of Money

While wealthy taxpayers themselves do not appear to be especially mobile, their reported taxable income and wealth often are.[160] As the evidence of fake relocations suggests,[161] taxpayers often find it easier to escape tax through tax-gaming and tax-planning strategies than through real changes in their behavior.[162] This accords with the widely accepted view that “capital,” or the combination of business and investment assets, is more mobile than humans.[163] Profits and losses are intellectual concepts, not natural things. They do not exist in any one place but instead arise through a series of transactions between parties who may be far apart from one another. Thus, the law must impose somewhat arbitrary rules to tie any given bit of profit to a particular taxable location.[164] Often it is relatively easy for taxpayers to tweak minor features of their transactions to change the law’s arbitrary assignments.[165]

When it comes to taxing peoples’ income, the most important of these arbitrary rules is the concept of realization. As we noted in the Introduction, the realization rule is the feature of most global income taxes providing that taxpayers only have to include profitable investments in their income at the time those assets are sold.[166] Although the rule makes some sense as an easy way for tax systems to determine the value of a taxable asset,[167] it also arbitrarily assigns taxing rights to the jurisdiction where a taxpayer happens to live at the time they sell their property.[168]

Realization can easily be gamed to avoid state income taxes.[169] For instance, as Julie Roin recently explained,[170] imagine that Musk’s stock increases in value from $0 to $100 billion while he lives in California, but he never sells it. He then moves to Texas, where the tax rate is zero. If Musk sells his shares while in Texas, he will never pay income tax to California (or any other state) on the $100 billion in appreciation he enjoyed. The reverse is also true: If someone plans to permanently relocate to a high-tax state, they have the option to sell appreciated property before moving to take advantage of their pre-move lower rates.[171] Thus, even if there is no net loss of population from higher state taxes, it is still possible that states could lose a great deal of revenue from mobility.[172] Departing taxpayers will tend to take their taxable gains with them, while new arrivals will not bring any in.

Taxpayers do not even have to move permanently to exploit this strategy. They can relocate (or claim to) for one year, sell, and then move back if they want, assuming they can convince state tax authorities their move was genuine.[173] This greatly lowers the subjective cost of tax avoidance, as it minimizes the time taxpayers must spend in their less-desired location. This may be why some researchers find that the location of taxable estates is somewhat responsive to effective state tax rates[174]: The wealthy person can move in late retirement years or even just before their expected end, making it easier to shop for a preferable tax rate.[175]

Realization and its close cousins also form the foundation of numerous other state-tax reduction strategies. For instance, another relatively simple tactic is to gift appreciated or income-producing assets to relatives in low-tax states.[176] This is far superior to earning the income in high-tax states, then transferring to the relative later. Gifts are not “realization events” in the American tax system,[177] so the transfer entirely eliminates any tax that might have been imposed by the transferor’s home jurisdiction.[178] Some commentators note that this strategy could trigger federal transfer (i.e., gift and estate) taxes,[179] but that outcome has long been fairly easy to avoid for aggressive planners[180] and was further defanged by 2017 changes to increase the gift-tax exemption to more than $10 million.[181] Moreover, since transfer taxes are not imposed on transfers for fair market value, the aggressive move is to “sell” the transferred assets to relatives, often at very low prices that the parties will nonetheless maintain are fair market value so that no transfer tax applies.[182] Many planners use trusts for this purpose,[183] both because of favorable Tax Court and IRS rulings,[184] and perhaps on the expectation that the presence of the trustee adds a veneer of independence to the deal.

Trusts further amplify this basic strategy by allowing wealthy individuals to move assets to low-tax states, even if they don’t (yet) have any relatives there they want to give money to.[185] Many states tax trusts based on where the trust is managed, not where its money originated.[186] If the transferor (called the grantor or settlor in trust lingo) gives up certain elements of control, they can easily shift taxation of trust assets elsewhere.[187] There is currently a split between state supreme courts over whether it would even be constitutional for a state to tax a trust based only on where the grantor lived.[188] And once the grantor passes, trust funds can often grow free of any tax, even if the eventual beneficiaries live in a high-tax state.[189] Here again, this loophole has a constitutional source, as the U.S. Supreme Court recently held that states cannot tax trusts based only on the residency of a beneficiary who isn’t currently entitled to receive funds from the trust.[190] More complex trust strategies, such as charitable remainder trusts, offer even more possibilities.[191]

Partnerships and limited liability companies (LLCs) also provide a wealth of planning options. Probably the simplest example is that partners and LLC members can store a great deal of unrealized value inside the entity and then claim that value upon the sale of their equity interest.[192] Partnership sales (and sales of LLCs, which are generally taxed as partnerships) are typically “sourced” to the state of residence at the time of sale,[193] so if the sale happens after the partner has retired to a low-tax state, the high-tax state loses out on all the stored value.[194]

How do partnerships and LLCs store value? They could buy and hold appreciating property, of course. But another common route is for the retiring partner to sell off the valuable reputation, customer lists, and other intangibles associated with their business (think of the sale of a used-car business or a dental practice).[195] A good share of these assets are, in actuality, a product of the partner’s “sweat equity” or labor, not a classic investment.[196] In theory, labor income is supposed to be taxed in the place where the labor happened,[197] but by converting their labor into an increase in the value of their partnership and then moving, partners escape that rule. Partnership tax rules also routinely grant partners large deductions in early years at the price of income in later years or at sale; partners who move to a low-tax state can escape this deferred-tax obligation.[198] While state rules for capturing some of this value are possible,[199] and some do exist,[200] even federal auditing of partnerships is essentially toothless; it would be naïve to believe states are at all effective on this front.[201]

Taxpayers can avoid state tax even more effectively with corporations, although sometimes there is an offsetting federal disadvantage. While partnerships and some corporations “pass through” their annual income to their owners, a so-called “C Corporation” (named for its subchapter of the tax code) is treated as a separate taxpayer. Equity owners only pay tax on the corporation’s profits when they sell (or otherwise dispose of) their stock, or when they get dividends. Both actions are discretionary, particularly in the case of a C Corp that is controlled by just a few shareholders who can thus choose when to issue any dividend. Therefore, it is simple to store value inside a corporation: The owners simply leave the profits in the corporation’s bank account and don’t sell their shares or take dividends.[202] They can easily borrow against the value of their stock if they need spending money.[203]

U.S. states generally tax corporations based on where their sales are, not where their owners are, so often the entity’s overall rate will be lower than the owners’, and frequently close to zero for multi-national firms.[204] If the owners’ home state would impose any tax on the corporation’s income, there are an assortment of common planning techniques to “strip” the income out and send it to another corporation owned by the same people, but taxable only in a tax haven such as Delaware, Liechtenstein, or Ireland.[205]

While there is a federal cost to the C Corp strategy, that cost decreased significantly in 2017 when Congress slashed the corporate tax rate from 35 percent to 21 percent.[206] At a minimum, the strategy is appealing to owners who were likely to form a C Corp anyway, such as those who hope to go public or are funded by private equity.[207] In theory, very old federal rules are supposed to impose penalty taxes on corporations that are used to accumulate excess profits or serve as “personal holdings companies.”[208] But those rules are in desuetude—we could find only three cases applying them after 1989, and one since 2000.

*          *          *

It is worth reemphasizing that we do not consider mobility in and of itself to be a problem. Quite the contrary, and consistent with most of the literature on fiscal federalism, we favor a system in which people can freely move to whichever jurisdictions offer them their most preferred packages of tax-funded benefits and other amenities, so long as they pay the tax costs of funding those benefits.[209]

Rather the problem, in our view, is what Julie Roin has labeled “exploitative mobility”—that is, when taxpayers “can extract benefits from one jurisdiction while escaping some of the costs of providing those benefits.”[210] If a taxpayer decides not to move to a jurisdiction with tax policies that are too progressive for their tastes, we would not consider this to be exploitative mobility, but rather just the normal operations of fiscal federalism. Similarly, a taxpayer who has earned substantial income and wealth in State A, where they reside and have also fully paid taxes on this income and wealth, is not engaging in exploitative mobility if they relocate because this income and wealth were taxed as it was accrued. Instead, we think reform efforts should be focused on taxpayers who benefited from the amenities provided by State A but engaged in tax planning (such as the schemes we discussed above)[211] to avoid paying much tax to State A.

Ultimately, just as we argue that there are good reasons for states to pursue progressive tax reforms, there may also be good reasons for states to limit the rates or aggressiveness of such reforms. We caution here that our assessment that the empirical literature does not find large migration responsiveness from existing forms of progressive taxation is not dispositive as to all possible futures.[212] If a state were to shut down all of the ways in which money moves to escape tax—implementing the reforms we propose in Parts IV and V—and especially if the state were to then hike its tax rates well above existing or historical levels, then this might well produce larger migration responsiveness. (Though it also might produce less, because although it will be harder to escape tax without moving, the tax rewards to moving will also be smaller.) With those cautions in mind, we think it prudent that both California’s and Washington state’s proposed wealth tax reforms would levy top rates well below the top rates currently being levied in many regions of Spain.[213]

A comprehensive evaluation of all the factors that states should consider in deciding the top rates for progressive tax reforms is beyond the scope of this Article. In this Part, we have aimed more modestly to explain why the existing empirical literature does not support claims that migration of people themselves is a primary obstacle to progressive tax reforms at even relatively low rates. Instead, we have argued that the primary obstacle is the movement of money. We thus next explain, in Parts IV and V, how states can solve the exploitative mobility problems associated with the movement of money.

IV. Addressing Exploitative Migration with Wealth Tax or Mark-to-Market Reforms

As we have explained, we do not consider mobility to be a problem in and of itself. Instead, this Article is concerned only with “exploitative mobility.”[214] It is worth further subdividing exploitative mobility into two subcategories that we will label as “exploitative migration” and “exploitative money moves.” Exploitative migration is when taxpayers physically relocate in a manner that extracts benefits from one jurisdiction while escaping that jurisdiction’s tax costs. Exploitative money moves are when taxpayers do not physically relocate but rather just move their money to escape their resident jurisdiction’s tax costs. This Part explains how states can use wealth tax or mark-to-market reforms to address exploitative migration. The next Part will explain reforms that states can implement to address exploitative money moves.

Exploitative migration typically begins with taxpayers accruing income and wealth in states with relatively progressive but realization-based income tax systems (e.g., California or New York). Those taxpayers benefit from the services and benefits provided by those states without paying much state tax because their incomes come mostly in the form of unrealized gains. Then, the taxpayers relocate either to a state without an income or wealth tax (e.g., Texas or Florida), or to a state with much lower-rate or less-progressive taxes, prior to selling their appreciated assets or otherwise triggering tax on their unrealized gains. Through these steps, the taxpayers can escape income tax in their state of initial residence, the state in which they accrued their income and wealth, while benefiting from that state’s services and benefits.

In this Part, we discuss two sets of options for solving this exploitative migration problem. First, a straightforward state-level wealth tax could partially solve this problem even without any further refinements. This is because the base of a wealth tax is a taxpayer’s net worth, with no need to wait for assets to be sold. However, this approach would be limited by the fact that wealth taxes are inherently backward-looking. Wealth is built up over time and then taxed periodically—economists often say that wealth taxes are assessed on a stock measure, whereas income taxes are assessed on a flow measure.[215]

To illustrate, imagine if Lucky Taxpayer won $100 million in the state lottery in year one while living in New Delaware (ND), a state. Under an income tax, Lucky would have $100 million of income in that year and no further income in subsequent years unless there were subsequent gains. All of Lucky’s income would be subject to income tax in ND, even if the Lucky were to relocate to another state in a subsequent year. By contrast, under a wealth tax, Lucky would have $100 million of wealth in the first year and also in each subsequent year, assuming for simplicity that there were no subsequent changes in their wealth. Thus, if ND levied an annual wealth tax, and if Lucky remained in ND, then Lucky would owe wealth tax to ND in both the first year and in each subsequent year. But if Lucky instead left ND at the end of that first year, then they would only owe a single year of wealth tax to ND, even though the wealth was accrued entirely while Lucky was a ND resident.

Because wealth taxes are inherently backward-looking in this fashion, we argue that residency tests for purposes of wealth tax assessments should similarly be backward-looking. Specifically, we propose that residency for purposes of wealth tax assessments should be phased both in and out, symmetrically, over a period of multiple years. We explain this proposal further in Part IV.A below.

The second set of options for solving the exploitative migration problem is mark-to-market reforms. The primary cause of exploitative migration is realization-based state income taxes, so mark-to-market reforms could solve this problem straightforwardly by taxing gains as they accrue without waiting for the sale of assets or other similar “realization events.”[216] However, there are a number of design issues and potential obstacles with implementing mark-to-market reforms at the state level. We discuss some options for resolving these issues in Part IV.B below. We also discuss in that Section how state-level wealth tax reforms might be integrated with mark-to-market reforms as an alternative approach for solving the exploitative migration problem with respect to state-level wealth taxes.

For both wealth tax and mark-to-market reforms, the primary administrative and implementation challenges concern valuation and liquidity. Many opponents of such reforms have argued that determining what assets are worth and ensuring that taxpayers have enough cash to satisfy their liabilities are so difficult as to make either wealth taxes or mark-to-market reforms infeasible, especially at the state level.[217] In Part IV.C below, we explain how either wealth taxes or mark-to-market reforms can be implemented to mostly resolve valuation and liquidity issues.

Beyond addressing exploitative migration, either wealth tax or mark-to-market reforms can also go a long way towards mitigating exploitative money moves. For instance, in the case of business entities, such as partnerships and corporations, the value of any entity owned in part by an in-state taxpayer would be subject to wealth or mark-to-market taxation. Any wealth stashed inside the entity would thus be subject to tax, no matter where the entity is legally located.[218] However, additional reforms and anti-abuse rules are needed to more fully address the problem of exploitative money moves, as we detail in Part V.

A.      Proposing Phased Residency Rules for State-Level Wealth Taxes

1.       The Basic Legal and Policy Argument

Jurisdictions in which wealth or income is accrued have a claim to taxing that wealth or income.[219] In a world of perfect information and ignoring administrative and compliance costs, this principle might imply that state-level wealth taxes should employ tracing rules so that each jurisdiction would only tax wealth that was accrued within that jurisdiction. Thus, wealth earned before entering and after leaving the state would be exempt.

To ground the intuition for why, consider Mark Zuckerberg, who built his fortune (from Facebook, now Meta) as a resident of California and, in doing so, took advantage of California’s services and benefits. If California were to levy a wealth tax and then Zuckerberg were to move out of state, we would consider it to be an inappropriate result if California were to consequently have no claim on taxing Zuckerberg’s wealth that was accrued within California. Conversely, imagine if Jeff Bezos were to move to California either just before or after California levied a wealth tax. Bezos built his fortune (from Amazon) as a resident of Washington state. We would consider it inappropriate if California were to tax all of Bezos’s wealth upon his becoming a resident.

Although this tracing approach has normative philosophical appeal, it would be excessively cumbersome to implement in practice.[220] And even if we could simplify recordkeeping, because money is fungible, such tracking would likely prove incoherent. Consider if a taxpayer won the lottery in State A, then moved to State B and invested those lottery winnings in growth stocks, then moved to State C and sold some of those stocks to buy real estate. What portion of that real estate value should be attributed to each state?

But something like the binary approach of current law, in which all realized investment income is taxed to the current residence state, regardless of when that investment wealth actually accrued, is not satisfying either. While administratively simpler than tracing, the current law’s approach departs very far from the normative principle of assigning taxing rights to the state that hosted the taxpayer while their wealth accumulated.[221] Moreover, because this rule is easily manipulable and causes large swings in tax outcomes, it is highly inefficient.[222] Thus, several commentators have recently proposed sliding-scale or partial-residency systems for income taxes, in which taxpayers split their income among multiple jurisdictions, even for items that have not traditionally been divided that way.[223] A full discussion of all options and their comparative advantages and disadvantages is beyond the scope of this Article.

Instead, we will just explain an approach that we call phased-residency rules, an approach that we incorporated in the proposed wealth tax reform for California (that we designed and drafted along with economist Emanuel Saez and other collaborators)[224] and in the proposed mark-to-market bill in Vermont.[225] We argue that this phased-residency approach is greatly superior to the binary approach in terms of accurately and fairly attributing wealth among jurisdictions while still being practically implementable. Most relevant here, the phased-residency rules approach also substantially mitigates the exploitative migration problem.

The essence of the phased-residency rules approach is to gradually phase in wealth taxes for new residents over a multi-year period and to symmetrically phase out those taxes over the same number of years for former residents who have migrated out of the state. For example, the California wealth tax reform bill uses a four-year phase-in and phase-out period: New residents migrating to the state would only have a fourth of their net worth attributed to California for wealth tax assessment purposes in the first full year of residence, rising to half in the second year, three-quarters in the third year, and the taxpayers’ entire net worth only attributed to California in the fourth year. [226] Symmetrically, a former resident migrating out of California would still have three-quarters of their net worth attributed to the state in the first year after their move, lowering to half in the second year, a quarter in the third year, and then none of their net worth attributed to California in the fourth and subsequent years following their migration.

2.       Legal Defense of Phased Residency Rules

Despite some claims to the contrary,[227] partial-residency rules for wealth taxes are constitutional under the current principles that govern interstate taxation. To be sure, a future Supreme Court could change these principles, or a lower court could misapply them. Thus, we cannot offer certainty. The principles we rely on are not just old but are grounded on notions of federalism and separation of powers that are important from many different ideological perspectives.

The general rule of state taxation is that intangible wealth and income of persons is sourced to (that is, taxed in) each person’s state of residence. However, as the Supreme Court has made clear, granting the resident state taxing rights to all of the income from intangible investments is just a tradition grounded in administrative convenience;[228] there is no reason such income cannot be apportioned or split between multiple states if it can be reasonably sourced to particular jurisdictions.[229] Thus, the Court has appropriately allowed the intangible income of a corporation to be apportioned.[230] The Court has ruled similarly with respect to corporate franchise taxes, that is, taxes that tax a stock of wealth.[231] It is not just income that can be apportioned. Indeed, these apportionment rules and the related “unitary business” principle originated in the context of property tax disputes, and property taxes are a form of partial wealth tax.[232] Further, the Court has long recognized that a state need not adopt a binary approach to sourcing intangible wealth, because the Court recognized that such a system leads to inappropriate and illogical discontinuities.[233]

Just as the Court has permitted states to tax reasonably apportioned shares of intangible wealth, so too the Court has (correctly) permitted a state to tax the intangible income of nonresidents that was “fairly attributable to” the state.[234] And states have continued to do so.[235] In fact, states already use temporal apportionment to source income even after the taxpayer has migrated to another state. Suppose an employee earns options as part of their employment but then does not exercise them until many years later (and typically in a jurisdiction without an income tax). The fact that the compensation was delayed does not defeat the source state’s claim for a share of the income.[236] Indeed, it is because there was a general consensus that the law permits the source state to tax deferred compensation that Congress specifically preempted some forms of such taxation.[237] A further striking point about the deferred compensation example is that the state in which the compensation was earned needs some method to apportion the value of the compensation if it is in the form of equity, as some of the accumulation occurred when the taxpayer was out of state. Accordingly, the rule in California is to measure relative time between the two states.[238] Our apportionment rule similarly uses a time formula to divide up the accumulation of wealth between states. Thus, not only is our proposal well-grounded in state tax principles, but it is also well-grounded in state tax practice.

We view the California reform’s use of a four-year phase-in and phase-out period to be conservative, a deliberate choice to err on the side of having a shorter period for a novel reform proposal. We think a strong case could be made that the period should be substantially longer. That said, at some extreme lengths, we think there would be due process concerns caused by the compliance burdens in later years. Recognizing that the Due Process Clause is often conceived of as containing a kind of fairness gut check, we thus think that (say) a hundred-year apportionment rule, though perhaps defensible for taxpayers whose wealth accumulated long ago, would likely offend this rough sense of fairness. We, therefore, chose four years for a measurement period, not because we have identified any literature on the analytically most appropriate timeframe to consider the relative contributions of a state to accumulations of great wealth, but rather because four years is the statute of limitations for a deficiency assessment in California.[239] That is, Californians already seem to accept that their tax affairs with the state might not be completely finished for four years after they physically leave. We draw upon that intuition as to fairness to guide our suggested phase-in and phase-out periods.

Also, although this would add to the complexity somewhat, it may be desirable to just use the phase-in and phase-out rules as a rebuttable default presumption to be accompanied by equitable apportionment rules. This way, either taxpayers or a state could argue with respect to specific fact patterns that less or more of a taxpayer’s wealth should be attributed to the state.[240] In a sense, the combination of a multi-year phase-in and phase-out period, along with equitable apportionment rules, would implement a limited tracing system for scenarios in which such tracing is practical and administrable and use the phased-residency approach only as the default and for scenarios in which tracing is not practical or administrable.

State income tax systems already use an approach similar to phase-in and phase-out rules for residency, although only for allocating income between states when a taxpayer moves partway through the year.[241] This one-year period is appropriate for income taxes because income taxes are assessed on a flow measure. By contrast, because wealth and mark-to-market taxes are assessed on a stock measure and so are inherently backward-looking, we have argued that the residency test for wealth tax assessment purposes should also be backward-looking. Overall, although our proposed phased-residency rules approach is by no means perfect—we are aiming only for rough justice—we view it as a substantial improvement over the binary approach.

It has been objected that our rule could violate the right to travel.[242] The “right to travel” is a right located in a constellation of provisions, particularly in the Privileges and Immunities Clauses of Article IV and the Fourteenth Amendment.[243] Our main contention is that the right to travel is simply not implicated by our residence apportionment rule. In the typical privileges and immunities case, there is a special disadvantage suffered by nonresidents: higher taxes (Austin),[244] loss of deduction (Lunding),[245] or lesser benefits (Saenz).[246] There is no such similar disadvantage triggered by moving. Indeed, those who move pay less tax than those who stay. Compare a Mark Zuckerberg who leaves California with one who remains. In the year after leaving California, the departing Zuckerberg would only be subject to tax on 75 percent of what the Zuckerberg who remains would be taxed, with this percentage declining thereafter. It is true that this is more liability than under current law, but, as already discussed, the current approach is not required, and reasonable temporal apportionment is permitted. Thus, taxing the departing Zuckerberg less neither violates the right to travel nor any other provision.

Even if our apportionment rule did implicate the right to travel and were subject to review under the Privileges and Immunities Clause, we think that it would pass the operative test, which requires a “substantial reason” for the difference in treatment and a “substantial relationship” between means and ends.[247] We have already explained at length why there is a “substantial reason” to apportion wealth accrued to a state. We have also explained why there is a “substantial relationship” between this reason and the use of time-based apportionment formulas, as is the current practice in many similar cases. Further, as already explained, the proposed rule permits petitioning for alternative apportionment, thus permitting even closer tailoring when possible.

B.      Implementing State-Level Mark-to-Market Income Tax Reforms

As we explained in Part II.B, taxpayers exploit the realization rule to shift business and investment earnings to low-tax places. That might not matter for our example of Lucky, the lottery winner, whose entire $100 million would be subject to ND’s income tax, even if they moved to another state in a subsequent year. Lucky would be taxed that way because income taxes are assessed on a flow measure and because the entire $100 million of lottery winnings would be realized in the year in which the award was won. But the majority of super-wealthy Americans earn their wealth and income not through lottery tickets, but through investment, entrepreneurship, or finance.[248] For sure, some super-wealthy Americans do earn most of their wealth and income in the form of wages and salaries that—like the lottery winnings—are treated as realized in the years those wages and salaries are earned. But this is the minority case. The majority of the wealth and income of super-wealthy Americans is accrued in forms that existing income tax rules treat as unrealized gains.[249]

We explained at the outset of this Part that the straightforward solution is to reform income taxes so that economic gains are taxed as they accrue instead of waiting for sale or other realization events. Here we note a few additional design considerations for implementing the mark-to-market solution.

1.       General Design Principles

First, it is important to acknowledge that mark-to-market reforms involve higher administrative and compliance burdens as compared to realization-based income taxes. Accordingly, in order to limit the extent to which taxpayers with limited resources and tax advice face these burdens, these proposals can be crafted so that they would only apply above some high thresholds for income or wealth. For instance, the proposed New York state Billionaire Mark-to-Market Act would only apply to taxpayers with a net worth in excess of a billion dollars, and the proposed Illinois Extremely High Wealth Mark-to-Market Tax Act would only apply to taxpayers with a net worth in excess of $50 million.[250]

We agree with critics that identifying which households qualify for these kinds of thresholds will typically have to be made through relatively simple methods that may imprecisely measure assets that lack a liquid market.[251] However, we do not see why a somewhat-imprecise measure of wealth, implemented to spare most households the administrative burdens of compliance, would meaningfully undermine the operation of the tax. This has long been the approach used (mostly successfully) for the design of minimum tax regimes.[252] Ultimately, any threshold is somewhat arbitrary, and misassigning a handful of taxpayers as falling below it, or vice-versa, has no obvious policy implication. Few households will fall in the range of uncertainty, making any behavioral effects of imprecision minor.[253] And for the very wealthy households who are comfortably above the threshold, our method measures wealth and the time value of money precisely.

Another set of issues concerns volatility in asset prices and the related challenges of refunding losses if a taxpayer pays tax on accrued gains in earlier years (after the asset prices increase), but the gains then disappear in subsequent years (because the asset prices decrease). To address these issues, two of us have previously proposed an approach that we called “phased mark-to-market.”[254] The essence of this approach is to treat all gains as realized in each year of assessment, but then to treat a percentage of those realized gains as unrecognized (i.e., not included in taxable income). For instance, a federal-level reform proposal that we co-designed based on this approach, the Babies Over Billionaires Act,[255] would annually deem realized all gains in traded assets but then would only recognize (i.e., impose immediate tax on) 30 percent of those deemed-realized gains. The next year, another thirty percent could be included in income, and so on. If the taxpayer’s assets lose money the next year, that loss could offset any older, unrecognized gains.

With only a percentage of deemed-realized gains being recognized (and thus subject to tax) in any assessment period, the volatility of the tax assessments would be greatly reduced, as the taxation of gains produced by market fluctuations would be spread over time. This is important at the state level because balanced budget constraints can create problematic fiscal volatility if tax revenues vary too much with economic cycles, creating numerous harms. Spreading out tax revenues across boom-and-bust economic cycles mediates these harms.[256] Spreading out the recognition of gains over time and across economic cycles also greatly reduces the need to refund losses, which is important both politically and to reduce administrative and compliance costs.[257] Of course, this approach would not completely resolve the volatility issues associated with mark-to-market reforms, and a sufficiently dire economic crisis could still result in the need to refund losses. But the approach would greatly alleviate these issues.

Additionally, for a state-level mark-to-market reform, this approach of combining full deemed realization with partial nonrecognition can be implemented in a manner that further mitigates the exploitative migration problem. The key is to make the partial nonrecognition an elective taxpayer option. In order to take advantage of partial nonrecognition of their deemed-realized gains, taxpayers would have to agree that they would continue to report and make payments on the unrecognized balance. Those obligations would continue regardless of whether the taxpayers remain in the state or migrate out of the state.[258]

We also recommend that states grant a credit or award basis for gains taxed in another state’s mark-to-market regime. Suppose Flora Founder has $0 basis in her shares in California, but they appreciate to $100 million while she is living in California, and Flora pays tax on the $100 million under California’s mark-to-market regime. Flora then moves to New York and sells her shares for $200 million. Absent a New York credit, Flora would pay tax on both the $100 million gain that arose in New York and the $100 million that was already taxed in California. A credit may not be constitutionally required, because any given state’s no-credit rule would meet the Court’s current “internal consistency” test for state tax rules.[259] But we think states in the position of New York in our example would not want to discourage incoming billionaires from mark-to-market states. In any event, since billionaires with substantial built-in gains are unlikely to realize those gains in a non-mark-to-market state (that is, in our example, Flora will likely not sell her stock while it would be taxable to New York), the absence of such crediting probably would not result in much (if any) double taxation.[260]

2.       Integrating Mark-to-Market and Wealth-Tax Reforms

Finally, there are advantages to integrating wealth tax and mark-to-market income tax reforms. On their own, wealth tax reforms cannot fully fix buy, borrow, die style tax planning, or the related buy, borrow, . . . and then eventually move to a state without an income tax style of tax planning.[261] To see why, consider a lucky investor whose investments massively increase in value over a couple of years. A mark-to-market reform can tax those economic gains immediately, but a wealth tax would only do so more slowly over time, thus leaving the taxpayer, at least temporarily, with massive untaxed gains. This can create perverse incentives for the taxpayer to engage in forms of tax gaming like exploitative migration designed to permanently exempt such gains from tax. A multi-year phase-in and phase-out period for a wealth tax could alleviate such perverse incentives but not eliminate them.

This might suggest that mark-to-market reforms are superior to wealth tax reforms, at least at the state level. But wealth tax reforms also have some advantages over mark-to-market reforms. For one, wealth taxes are likely easier to explain to voters and many politicians as compared to mark-to-market reforms. This is especially important when considering reforms to be implemented through ballot initiatives, but it may also matter in the legislative context. Additionally, although the phased mark-to-market approach we discussed previously can alleviate volatility issues, wealth tax reforms are still likely to cause less volatility as compared to mark-to-market reforms.

A full discussion of the comparative advantages and disadvantages of wealth tax and mark-to-market reforms is beyond the scope of this Article. Instead, we raise these issues to explain why a state considering implementing either a wealth tax reform or a mark-to-market income tax reform might wish to do both, while then making the wealth tax payments at least partially creditable (and refundable) against a portion of the mark-to-market payments.

This is only one possible approach for integrating wealth tax and mark-to-market reforms, which we offer here as an example. Another related approach might be to use an alternative minimum tax structure so taxpayers would pay the greater of the two forms of taxation.[262] The key takeaway is that integrating both reforms has the potential for designing a policy that could be described and explained as a wealth tax while retaining the advantages of mark-to-market reforms for better combatting tax gaming.

C.      Addressing Valuation and Liquidity Challenges

Critics of wealth and mark-to-market reform proposals sometimes argue that valuation and liquidity problems make them impossible to implement.[263] Even some supporters raise these concerns.[264] The critics argue that it is too hard to know what a taxpayer’s assets are worth before they are sold, and that before the sale, some taxpayers may not have the cash to pay any tax. These critiques typically fail to mention the century-long history of broad-based wealth taxes among the U.S. states.[265] In any event, we have written extensively elsewhere about how to address these complaints, building on an earlier literature describing so-called “retrospective” tax systems.[266]

Briefly, we think the best solution is a combination of notional equity interests with improved techniques for appraising an asset’s true value. At the time tax would be due, taxpayers who own illiquid or hard-to-value assets can give the government an IOU instead of cash. Taxpayers then pay off the IOU when the asset is ultimately sold and pay interest on the original tax debt at the asset’s internal rate of return. To streamline accounting and administration, the interest charge is calculated simply by awarding the government a proportional but nonvoting share in the taxed asset (hence “notional equity” interest), so that algebraically, the government’s interest will always grow or shrink at the same rate as the asset itself. Taxpayers who want to avoid this system can opt for a more traditional appraisal, but on terms that are more favorable to the taxing authority than under current practices.

This approach also helps states to solve the problem of how to implement a retrospective system at the state level. Retrospective systems generally impose tax at realization, plus an additional interest charge to account for the time value of money.[267] But before our proposal, it was unclear whether a state could successfully tax a former resident whose wealth appreciated during their residency but who realized those gains somewhere else, often in a place where the state lacked personal jurisdiction over the taxpayer.[268] Under prevailing personal-jurisdiction precedents, however, the states have jurisdiction to collect debts in the courts of other states.[269] Thus, if a state takes the trouble of getting its tax lien reduced to a judgment in its own courts and then follows the procedures of the Uniform Enforcement of Foreign Judgments Act,[270] it can enforce its judgments in the courts of another state. Forms of this Act have been passed in forty-eight states.[271] To be sure, states apparently do not typically go to this much trouble, but they do sometimes,[272] and the stakes relating to the extremely wealthy indicate that they should.

The notional equity interest, we argue, is indistinguishable from other collectible debts. In order to ensure future collections, the California and Vermont proposals have a belt-and-suspenders approach—that is, they take two approaches where one would suffice, with each approach backstopping the other. Thus, we do not just rely on the enforceability of the deferred tax liability. As with our phased mark-to-market proposal, taxpayers who choose to defer payment must agree to the state’s jurisdiction to collect payment of their liability in the future or else be denied the privilege of using deferral.[273]

Crucially, this is not a novel approach, which indicates it is well within the capacity of state tax administrators. In response to a similar challenge, many states give nonresident limited partners or shareholders of LLC a choice: Either consent to jurisdiction[274] (and to future information-reporting requirements for updating the state tax authorities) or else the entity must withhold for the nonresident partners.[275] Of course, the existence of an enforceable[276] threat of a continued tax obligation to their former state may also incentivize taxpayers to just opt for appraisal and immediate payment.[277]

*          *          *

In sum, wealth and mark-to-market taxes can significantly constrain exploitative migration by taxing wealth as it accumulates, rather than allowing taxpayers to pack it away and haul it to a tax haven. With thoughtful design, these systems can also allocate taxing rights to the places with the strongest claims on them, mitigate practical problems related to debt collection and rebates, and readily assign value even to hard-to-value assets.

V. Addressing Exploitative Money Moves

We have argued that reforming tax systems away from the realization rule can help states tax the wealthy more effectively. Yet while wealth tax or mark‑to‑market reforms would reduce opportunities for tax planning, some would still remain. In this Part we consider options states might pursue in shutting down these strategies. Again, our main objective is not to work out every detail, but instead to show that there are plausible solutions to these major tax minimization efforts and, thus, that the project of taxing the wealthy at the state level is worth pursuing.

A.      Income Splitting and Noncharitable Gifts

A standard tax-planning issue that looms larger when state borders are easier to cross is what tax lawyers call “income splitting” or “income shifting.”[278] Briefly, the idea is that a family or other group of close-knit taxpayers may not care much which one of them holds legal title to an asset because they all benefit from the asset. A parent might be planning to share some of their wealth with their children anyway, an unmarried couple might expect that each will share everything the other owns, or the lone shareholder of a small business may exercise the legal right to claim the business’s assets. In all these cases, the taxpayers can and often do arrange their affairs so that income is reported by the lowest-taxed member of the group. For example, parents who want to help their children buy a new house might gift taxable interest-bearing bonds to the kids instead of cash.[279] That way, the interest is taxed at the children’s (presumably) lower tax rate.

This strategy can be supercharged at the state level. Income shifting is only moderately effective at minimizing federal taxes because as the higher earner transfers more taxable income to the lower earner, the lower earner’s marginal tax rate rises.[280] At the state level, of course, earners in high-tax states can shift income without limit to loved ones or controlled entities in states with lower top marginal rates, reducing tax even if the relations have higher incomes. Further, most wealth tax or mark-to-market reform proposals build in large exemption thresholds, so that each additional taxpayer that wealth can be shifted to adds another, say, $50 million in exemptions.[281]

As a result, general federal tax rules aimed at limiting income shifting are too weak to constrain aggressive state tax planning. With certain statutory exceptions, we currently tax income to the taxpayer who exercises real control over the source of the income, regardless of legal formalities.[282] Determining “control” can be a complex and fact-intensive question, so it is costly for tax authorities to investigate, litigate, and prove.[283] Perhaps the rule is good enough for the more modest role it plays in limiting federal income shifting,[284] but we doubt it can realistically constrain more strongly motivated state tax planners. Further, by definition, it cannot prevent income shifting in cases where the taxpayers intend (or are indifferent to) changes in control, such as when parents actually want their children to have a new house or take over the family business. Other federal income-tax rules against income shifting, such as an extra tax on investment earnings by children under eighteen, are little more than speedbumps.[285]

Some experts have argued that the more important planning obstacle to these kinds of transactions is the separate federal tax on gratuitous transfers, but, as we mentioned earlier, it too is easy to sidestep.[286] For donors who don’t want to pay gift or estate tax, there is always the sale for “fair” market value.[287]

Given taxpayers’ powerful incentives to shift income and the ease with which these arrangements can be entered into, we would argue for strong presumptions, or even bright-line rules, that treat gifted property as belonging to the transferor for some minimum period of time after the gift.[288] For example, a state statute could provide that any asset transferred by a taxpayer to a related party is treated as still taxable to the transferor (or rebuttably presumed to be controlled by them), but that the taxable amount is reduced by the value of any consideration paid at the time of transfer.[289] Both wholly gratuitous and below-market transfers would then both be at least partly taxable to the original owner. This approach does have some constitutional questions, which we detail in the next Section.

While this might at first seem strong medicine, we would argue it is effectively just a variation on the standard tax-policy decision about whether to tax married couples as a unit or as separate individuals. Modern tax systems strive to impose taxes based on a taxpayer’s “ability to pay.”[290] The argument for taxing married couples as one taxpayer is based on the idea that the couple’s spending decisions are made jointly, or at least that each partner’s spending decisions are so highly influenced by the other’s that it is difficult to say that one partner has a different ability to pay than the other.[291]

The ability-to-pay argument, if it’s persuasive, isn’t clearly limited to married individuals, but instead could extend to any group of people who closely share resources and take one another’s well-being into account.[292] Where to draw the exact line is a tradeoff, with equity and efficiency on one side and, on the other, the administrative difficulty of identifying nonmarital relationships that reflect a shared ability to pay.[293] As we’ve said, in the unusual context of state taxes on the wealthy, it would be easy for some families that in fact share resources to achieve a much lower tax rate than others. This would result in both inequity and inefficient expenditures of resources on tax planning. Accordingly, we’d argue that it makes sense to tax each wealthy family roughly the same regardless of whether some families happen to have children who live in low-tax states, or just have more children who can claim a $50 million exemption amount. Achieving this goal is worth some additional administrative expense. That is especially the case where the transferor demonstrates, by the fact that they have transferred the asset, that the transferee’s well-being is important to them. Taxing the transfer is, in effect, taxing based on the well-being of the transferor.

B.      Trusts

Trusts replicate all the state planning strategies offered by gifts, but with greater flexibility, and the benefit of generous IRS rulings that often allow for even better estate and gift tax results.[294] Unfortunately, courts have interpreted the U.S. Constitution in ways that tend to limit state options for taxing trust assets. In its 2019 Kaestner decision, the Supreme Court held that it would violate due process for a state to tax a trust based only on the fact that a potential future beneficiary of the trust resides in the state.[295] The possibility of future benefit was not a sufficient “definite link” or “minimum connection” to the taxing state.[296] To meet that standard, the Court held, the state must show that “the resident ha[s] some degree of possession, control, or enjoyment of the trust property or a right to receive that property before the State can tax the asset.”[297] That rule allows wealthy families to place unlimited value in trusts in low-tax jurisdictions, where they can grow free of any state tax, for enjoyment by later generations.[298] Beneficiaries can likely even borrow for their current enjoyment against these funds tax-free while living in high-tax states.[299]

Additionally, as we noted earlier, there is a division between state supreme courts over whether states can impose tax on a trust based on the residence of the grantor.[300] While older decisions have upheld that authority, the Minnesota Supreme Court ruled in 2018 that the Due Process Clause prevents Minnesota from indefinitely taxing a trust that was established during the grantor’s life in Minnesota.[301] The trust’s origin was not a sufficient in-state connection for taxation in the distant future. The Minnesota Supreme Court’s position would likely also imply that our proposal to tax gifts in the state of the gift giver would similarly raise due process concerns.

We suggest that basing tax on the grantor or gift giver’s continuing control or influence over the transferred assets could avoid this question. As Mitchell Gans has observed, Kaestner cited with approval earlier cases upholding state taxation based on continuing control by the grantor.[302] And before Congress developed a refined regime for trust taxation, the U.S. Supreme Court held that taxation of trust income could turn on, at least for federal law purposes, “actual command over the property taxed.”[303] That rule, the Court would later explain, was necessary to prevent income splitting: “[W]hat is in reality . . . one economic unit [cannot] be multiplied into two or more.”[304] We believe the Court would similarly recognize that states may, consistent with the Due Process Clause, take into account the economic reality that a seemingly out-of-state cache is actually part of the wealth of an in-state resident.[305] Further, we think the Court would allow states to do so based on practical reality, as it did in the federal income-tax context, instead of relying only on the trust documents’ formalities.[306]

A control-based test could likely capture most potentially abusive or tax-avoiding transfers. States could place the burden on donors to show that they don’t retain any meaningful influence over assets transferred to trusts or related parties, perhaps based on the federal rules for complete “termination[s] of interest” in intrafamilial transfers.[307] This would reflect the basic facts that transfer is itself an exercise of control,[308] family members are often indifferent about which family member has legal title,[309] and trust administrators compete for the right to manage trust assets, giving transferors considerable implicit influence over the administrator.[310] Even if the grantor does not retain formal control, they will retain significant influence over the trustee based on their ability to direct additional assets to the trustee, a fundamental economic concept known as “staged” financing.[311] Limiting the deemed “control” to the period when the grantor lives would further strengthen this inference.[312]

Taxing later generations after the grantor dies often requires a different technique,[313] but states can get there by tweaking a tool already in use in California. Once a grantor has passed, a family might use out-of-state trusts to allow assets to grow free from state tax until heirs need the funds.[314] California’s “throwback tax” (which gets a brief, seemingly approving, shout-out in Kaestner)[315] tries to address this problem by imposing California tax on a trust in the year when a beneficiary “vests,” or becomes entitled to, the trust assets.[316] The throwback tax imposed is not only for the year of vesting, but also for any prior years when the beneficiary lived in California. This is based on the plausible assumption that the beneficiary was effectively enjoying the trust’s accumulating wealth in those earlier years, such as by spending non-trust wealth in reliance on the future trust distribution.[317] The problem is that the throwback is relatively easy to dodge, since it triggers only if the beneficiary is living in California in the year of vesting.[318] If the beneficiary moves to Nevada, vests, then moves back to California, there is no throwback. A more sensible version of the rule would simply impose throwback liability in years when a beneficiary resides in California, regardless of when vesting occurred.[319]

C.      Pensions and Retirement Savings

The U.S. tax system already offers high-earning individuals an assortment of tax-favored tools for not only multiplying their wealth through gifts and trusts but also stashing money away until retirement age. And yet, Congress and creative tax planners are steadily inventing new ones.[320] States should decouple themselves from most of these overly generous policies. Most commentators have assumed that a federal statute, adopted in 1996, bars states from taxing retirement savings,[321] but in fact, state wealth and mark-to-market tax systems would not raise any issues under the federal law. We will now explain these points in a bit more depth.

Our focus here is not on the standard 401(k) and individual retirement arrangements available to most earners, but instead on the specialized benefits claimed by the very highest earners.[322] For example, many executives accrue large sums in “nonqualified” deferred compensation plans, so named because they are not bound by the relatively low contribution limits that are supposed to apply to plans that qualify for tax benefits under the Employee Retirement Income Security Act (ERISA).[323] Yet even though these plans do not meet ERISA’s requirements, they still grant the executive unlimited tax deferral.[324] This technique is not limited to executives at large companies. Any self-employed person (or group of persons, such as physicians or lawyers) can pay out business profits to themselves in the form of a nonqualified but tax-favored pension instead of a salary or dividend.[325] Partners in private equity funds get similar deferral benefits (plus others) from being paid with a “profits interest” rather than cash.[326] If impatient to spend their funds, the entrepreneur or partner can then borrow against the deferred payments without income-tax consequence.[327] We also believe this is the strategy Shohei Ohtani is taking a swing at, as we mentioned in the Introduction.

Even if this system of generous retirement benefits for the wealthy had some plausible justification, states still should not follow along. For instance, while courts have said that the difficulty in valuing deferred compensation requires deferred taxation, our ULTRA mechanism easily solves that problem.[328] Another standard explanation of retirement benefits, at least those available to middle-income earners, is to overcome the public’s tendency to save too little.[329] These subsidies arguably conserve overall government expenditures in the long run, if we think that those who fail to save would otherwise be supported through public programs.[330] This rationale does not support additional state retirement benefits, because under-saving by higher earners doesn’t likely vary much from state to state and so is best solved by federal policy.[331] Yet if states exempt retirement contributions from their own income tax, they are adding to the marginal incentive for savings.[332] And indeed, states did attempt to impose income tax on the retirement earnings of former residents, but Congress prohibited that approach in the 1996 statute we mentioned.[333]

State wealth and mark-to-market statutes are not meaningfully constrained by the federal limit on state retirement taxes. The federal law only prohibits states from imposing tax on the retirement income of nonresidents.[334] Nothing prevents the state from treating funds held in a retirement account, or in the form of a partnership profits interest, as part of the taxpayer’s wealth. Nor does anything prevent the state from imposing mark-to-market taxes on growth in the value of those funds during the taxpayer’s working years when the individual is a resident of the state. Federal law gives states broad authority to define “resident,” and so it would likely be compatible with our proposal for phased residency.[335]

We emphasize that this is an approach aimed only at the deferred earnings of those very wealthy households who would be the subjects of wealth or mark‑to‑market taxation. ERISA preempts states from taxing the ERISA-qualified savings held by most middle-class earners,[336] and we suspect that taxing retirement savings for lower- and middle-earning households would be politically infeasible, even if economically defensible. States might, though, close abusive loopholes in qualified accounts, such as the infamous claim by Peter Thiel and others that they can legally hold billions in Roth accounts whose annual contributions are limited to five figures.[337]

D.     Charities

Charitable and other tax-exempt organizations present several policy challenges for state taxes on the wealthy. Family foundations, for example, have been a key tool for dodging taxes in Europe,[338] and though the U.S. regime is less porous, it does have important holes that are especially wide at the state level.[339] The U.S. tax system has generally relied on a separate body of law, the law of charitable organizations, to police these kinds of abuses.[340] But it is unlikely that states can depend on that as an effective backstop to expanded state taxes on the wealthy. States should also consider the extent to which new taxes on wealth or capital earnings would affect how or how much donors give to charity.

Charities can be surprisingly effective tax shelters. A standard planning technique exploits the fact that the IRS allows a charity to pay “reasonable” compensation and reimbursable expenses to its employees and board members, even if those individuals are the same people who donated money to the charity.[341] A small charity whose board is controlled by a donor’s family can simply hire the family members to run the charity, paying them hundreds of thousands of dollars annually in the aggregate.[342] Indeed, as one of us has calculated, among the smallest private foundations, more than a third of the charities’ annual expenditures are these kinds of administrative expenses.[343]

Because contributions to a charity are not a realization event, donors who fund their family charity with appreciated assets can get a quadruple tax benefit: no tax on the appreciated gains, no tax on subsequent investment gains while the assets are held by the charity, no estate or gift tax on the transfer of the assets, and a charitable contribution deduction in the full (untaxed) value of the assets as a cherry on top.[344] In effect, the charity enables the tax-free purchase of millions of dollars’ worth of annuities, or annual payments that continue until the death of the beneficiaries. But even better than a standard annuity, the salary payments can move from the original donor to their heirs, and then so on in perpetuity. And this can be a highly tax-efficient method for transferring a stream of payments from a high-tax jurisdiction (say, where the family business was operated) to a low-tax jurisdiction (say, where the family retires).

Of course, the success of this strategy depends on how stringently tax officials enforce limits on “reasonable” payments. Over the last decade or so, regulatory oversight of U.S. charities has become notoriously lax.[345] Even more problematically, state wealth or mark-to-market taxes would put increased pressure on the IRS to enforce its rules. The higher the tax on investment income, the greater donors’ incentives would be to exploit tax-exempt entities. Even if donors’ primary goal is genuinely charitable, taxes on capital encourage donors to shift investment funds to charitable control so as to allow the donation to grow tax-free.[346] This, in turn, creates temptations for the donors, or the agents they hire to manage the charity, to take advantage of charitable resources for their own purposes.[347] We should not expect the federal authority to fully calibrate its enforcement efforts, tools, and priorities to make up for state-driven shifts in taxpayer attempts to exploit charity resources.

Accordingly, states should strongly consider taxing some assets that are nominally held by charities as though they were still owned by donors or those related to their donors, at least for donors who would be subject to wealth or mark-to-market taxation. In effect, the same rules we proposed for trusts could apply to charities as well. States could presume that donated assets are still under the sway of the donor unless the donor established otherwise. For example, a state could presume that a charity that pays compensation to its major supporters (or their relatives) is under their control. Board members who donate to a private foundation should be taxed as though they still directly owned the donated assets, because, in fact, they continue to exercise influence over those funds, even if, as a technical matter, the funds belong to the charity. States should also aim to counter incentives to hold assets longer inside charities by requiring “payout” over relatively short periods.[348]

Again, as with trusts, we think such treatment reflects both economic reality and good tax policy. Control over resources is an aspect of ownership, even if those resources are used for good social purposes.[349] The traditional policy argument for pretending otherwise—that is, for allowing charities to be tax exempt and granting a deduction for charitable contributions—is to stimulate production of positive externalities by the charity.[350] That is, both giving money to charitable causes and buying a yacht are “consumption,” or spending that satisfies the spender’s subjective preferences, but in order to encourage the former, we provide that form of consumption with extra tax benefits.[351] The question for states is whether there is any reason to provide additional tax benefits when the federal government is already subsidizing charitable activity. Perhaps the best argument for state-specific incentives for giving is that tastes for public goods might vary from place to place, or the marginal return on investments in public goods might be higher in some places than others.[352]

But there are better ways of making this connection. Tax deductions for charity tie the level of the state’s support with the tax rate imposed on wealthy donors when there is no necessary connection between those two. New taxes on extreme wealth, if they included deductions for charitable contributions (in the case of the mark-to-market tax) or exemption for charity-owned wealth (in the case of a wealth tax),[353] would greatly increase the state’s subsidy for charity. But it is unlikely that states with these new taxes would have a greater need for charity than those without. In addition, state charitable dollars might flow mostly out of state.[354] States that want to support charity without tying the value of the subsidy to their tax system have other excellent options, such as the government matching grant method employed in the U.K.[355]

For similar reasons, states should decline to follow the federal lead on so-called “split-interest trusts,” or trusts where one beneficiary is a charity but others are heirs of the grantor. By providing a vehicle for moving assets out of state without paying gift tax, these trusts can offer significant tax savings. Commentators think these trusts provide minimal benefit for charity while taking significant sums from the Treasury.[356] Given that the case for more traditional and effective state-level tax subsidies is already tenuous, it is difficult to see why states would want to pour in yet more money. States should not provide split‑interest trusts with benefits that are unavailable to other trusts.

*          *          *

Although state taxpayers currently have a long menu of exploitative money moves available for minimizing their tax bills, we have argued in this Part that states also have good counter-options to the most important of these modern strategies. In combination with a wealth tax or mark-to-market reform, these reforms should substantially curtail taxpayers’ ability to readily shift their taxable incomes from place to place.

Conclusion

We have argued that the conventional wisdom on state taxation of the wealthy is mistaken. Standard accounts assume that because redistributive taxes would drive the tax base to other places, meaningfully progressive state taxes are impractical, and especially that wealth or mark-to-market taxes aimed at the rich would be self-defeating. To the contrary, we have argued that, by removing the opportunities to exploit realization rules, wealth and mark-to-market taxes are exactly what would make truly progressive state taxation viable, especially if those big reforms were accompanied by other policies to close the most important remaining loopholes. With these fixes in place, families could continue to move from place to place, as healthy federalism likely requires, but exploitative legal maneuvers to relocate taxable income would no longer be commonplace.

We think these insights should be valuable to several different sets of policy makers. To be sure, progressive state taxation may not be to every state’s taste. But for states where voters desire lower inequality, and where, on balance, progressive taxes could improve welfare, our proposals may be key to achieving voters’ preferences. We also have suggested that enabling state-level progressivity is likely important to the long-term health of our federation. Thus, our arguments should additionally be of interest to federal lawmakers and offer them potential reasons to further encourage, rather than block, state progressivity efforts. For instance, some commentators suggest that the federal deduction for state and local taxes encourages state progressivity.[357] Our analysis thus may be central to the ongoing debates over whether to renew limits on that deduction.[358]

Finally, we suspect that state-level experiments with new tools for implementing progressive taxation—like those we propose in this Article—will be needed to pave the way for these reforms to be adopted more widely. Just as early state-level income tax reform efforts led to the enactment of the modern federal income tax in 1913,[359] efforts for state-level wealth tax or mark-to-market reforms could lead to these proposals eventually being adopted nationally.

Copyright © 2025 Brian Galle*, David Gamage** & Darien Shanske***

          *     Professor of Law & Agnes Williams Sesquicentennial Chair in Tax Policy, Georgetown University Law Center.

          **     The Law School Foundation Distinguished Professor of Tax Law & Policy, University of Missouri School of Law.

          ***     Professor of Law, University of California, Davis, School of Law. The authors gratefully acknowledge thoughtful comments from Alan Auerbach, Joe Bankman, Dorothy Brown, Carl Davis, Gilad Edelman, Calvin Johnson, Stephen Land, Leandra Lederman, Bob Lord, Alex Raskolnikov, Julie Roin, Shayak Sarkar, Emily Satterthwaite, David Schleicher, Lee Sheppard, Kirk Stark, Neel Sukhatme, David Super, Josh Teitelbaum, and Jared Walczak; Vermont Representative Emily Kornheiser and expert legislative staff in California and Vermont; as well as attendees of presentations at the Experienced in Tax Conference, Georgetown Law School, University of Missouri School of Law, Indiana University Maurer School of Law, University of Georgia School of Law, the University of California, Irvine, School of Law – Taylor Nelson Amitrano, LLP Tax Symposium, the National Conference of State Legislators, the National Tax Association, and the Northern California Tax Roundtable.

          [1].     See, e.g., Emmanuel Saez & Gabriel Zucman, The Rise of Income and Wealth Inequality in America: Evidence from Distributional Macroeconomic Accounts, 34 J. Econ. Persps. 3, 10–11 (2020) (measuring growing wealth accumulation); Emmanuel Saez & Gabriel Zucman, Wealth Inequality in the United States Since 1913: Evidence from Capitalized Income Tax Data, 131 Q. J. Econ. 519, 551–54 (2016) (measuring wealth inequality); William G. Gale, John Sabelhaus & Samuel I. Thorpe, Measuring Income Inequality: A Primer on the Debate, Brookings Comment. (Dec. 21, 2023), https://www.brookings.edu/articles/measuring-income-inequality-a-primer-on-the-debate/ [https://perma.cc/GD8E-2R5S] (“[T]he preponderance of evidence suggests that income inequality has increased . . . .”).

          [2].     See, e.g., Greg Leiserson & Danny Yagan, What Is the Average Federal Individual Income Tax Rate on the Wealthiest Americans?, White House (Sept. 23, 2021), https://www.whitehouse.gov/cea/written-materials/2021/09/23/what-is-the-average-federal-individual-income-tax-rate-on-the-wealthiest-americans/ [https://perma.cc/5YQ9-CQ5W] (determining that the wealthy on average pay a lower tax rate); Lily Batchelder & David Kamin, Taxing the Rich: Issues and Options 4–8 (Sept. 11, 2019) (unpublished manuscript), https://ssrn.com/abstract_id=3452274 [https://perma.cc/X9BY-7LTF] (noting tax avoidance strategies of the wealthy); Jessie Eisinger, Jeff Ernsthausen & Paul Kiel, The Secret IRS Files: Troves of Never-Before-Seen Records Reveal How the Wealthiest Avoid Income Tax, Pro Publica (June 8, 2021), https://www.propublica.org/article/the-secret-irs-files-trove-of-never-before-seen-records-reveal-how-the-wealthiest-avoid-income-tax [https://perma.cc/L5Q2-R28Y] (investigating taxes paid by the twenty-five richest Americans).

          [3].     See, e.g., J. Baxter Oliphant, Top Tax Frustrations for Americans: The Feeling that Some Corporations, Wealthy People Don’t Pay Fair Share, Pew Rsch. Ctr. (Apr. 7, 2023), https://www.pewresearch.org/short-reads/2023/04/07/top-tax-frustrations-for-americans-the-feeling-that-some-corporations-wealthy-people-dont-pay-fair-share/ [https://perma.cc/C94F-XVYM].

          [4].     See, e.g., Louis Kaplow, The Theory of Taxation and Public Economics 123–36 (2008) (suggesting that income taxation is a more sensible way to redistribute income as compared to alternatives).

          [5].     See Carl Davis, Andrew Boardman, Neva Butkus, Eli Byerly-Duke, Kamolika Das, Aidan Davis, Michael Ettlinger, Erika Frankel, Dylan Grundman O’Neill, Matthew Gardner, Marco Guzman, Amy Hanauer, Mike Hegeman, Galen Hendricks, Spandan Marasini, Matt Salomon, Brakeyshia Samms, Emma Sifre, Miles Trinidad, Alex Welch & Jon Whiten, Inst. on Taxation & Econ. Pol’y, Who Pays? A Distributional Analysis of the Tax Systems in All 50 States 10–11 (7th ed. 2024), https://media.itep.org/ITEP-Who-Pays-7th-edition.pdf [https://perma.cc/F3GQ-J2TN] (finding that most state tax systems worsen inequality); see also Johannes Fleck, Jonathan Heathcote, Kjetil Storesletten & Giovanni L. Violante, Tax and Transfer Progressivity at the US State Level 3–4 (Sept. 12, 2021) (unpublished manuscript) (on file with author) (finding that even accounting for progressive state spending, state fiscal systems do not reduce inequality).

          [6].     See Saez & Zucman, supra note 1, at 18.

          [7].     See, e.g., Jéanne Rauch-Zender, Wealth Taxes and America Divided, Tax Notes State (Mar. 27, 2023), https://www.taxnotes.com/special-reports/legislation-and-lawmaking/wealth-taxes-and-america-divided/2023/03/24/7g760 [https://perma.cc/Y3R7-6C35]; Aimee Picchi, A National Wealth Tax Has Gone Nowhere. Now Some States Want to Tax the Ultra-Rich, CBS News MoneyWatch (Jan. 20, 2023), https://www.cbsnews.com/news/wealth-tax-in-8-states-california-new-york-connecticut-washington-illinois/ [https://perma.cc/V463-D5WQ]; Arthur Laffer & Stephen Moore, The ‘Hotel California’ Wealth Tax, Wall. St. J. (Mar. 5, 2023), https://www.wsj.com/articles/the-hotel-california-wealth-tax-high-taxes-resident-flight-new-jersey-massachusetts-new-york-texas-florida-utah-tennessee-cost-of-living-education-crime-silicon-valley-south-c39602ac [https://perma.cc/9FQJ-3J6K].

          [8].     See Wallace E. Oates, Fiscal Federalism 143–44 (Edward Elgar ed., 1972); see also, e.g., Robin Boadway & Jean-François Tremblay, Reassessment of the Tiebout Model, 96 J. Pub. Econ. 1063, 1063–64 (2012); Kirk J. Stark, Fiscal Federalism and Tax Progressivity: Should the Federal Income Tax Encourage State and Local Redistribution?, 51 UCLA L. Rev. 1389, 1406–08 (2004) (summarizing literature).

          [9].     We detail these points in Part II, infra.

        [10].     We review this evidence in Part III.A, infra.

        [11].     See Mike Ives, Jeff Bezos Says He is Leaving Seattle for Miami, N.Y. Times (Nov. 3, 2023), https://www.nytimes.com/2023/11/03/business/jeff-bezos-amazon-miami-seattle.html [https://perma.cc/G4EK-TQPC].

        [12].     For a counterexample of a more nuanced account, see Jonathan Levin, Fisher’s Texas Move is About Politics, Not Taxes, Bloomberg L. (Mar. 30, 2023), https://news.bloombergtax.com/tax-insights-and-commentary/fishers-texas-move-is-about-politics-not-taxes-jonathan-levin [https://perma.cc/9UGX-KPSE] (finding that one billionaire’s move was driven by politics, even if framed in the media as a response to taxes).

        [13].     We develop this argument in Part III.B, infra.

        [14].     See Edward D. Kleinbard, Stateless Income, 11 Fla. Tax Rev. 699, 701–05 (2011).

        [15].     See Julie Roin, Changing Places, Changing Taxes: Exploiting Tax Discontinuities, 22 Theoretical Inquiries L. 335, 378 (2021).

        [16].     See id.

        [17].     See, e.g., Richard Rubin, Wealthy N.Y. Residents Escape Tax with Trusts in Nevada, Bloomberg Bus. Wk. (Dec. 18, 2013), http://www.businessweek.com/ news/2013-12-18-wealthy-n-dot-y-dot-residents-escape-levy-with-trusts-in-Nevada-taxes [http://perma.cc/ K5CC-JCGV] (noting wealth planners describe state income taxes as “a huge issue” and reporting that New York state alone loses $150 million annually to just one of these techniques).

        [18].     See Evan Osnos, The Getty Family’s Trust Issues, New Yorker (Jan. 16, 2023), https://www.newyorker.com/magazine/2023/01/23/the-getty-familys-trust-issues [https://perma.cc/LM6M-NSHW] (discussing the “wealth-defense industry”); see also, e.g., Andrew Bernell, Eight Ways to Save Taxes when Selling a Business, Century Park Wealth Mgmt., https://www.centuryparkwm.com/eight-ways-to-save-taxes-when-selling-a-business/ [https://perma.cc/NND9-KGSU] (guiding readers on how to avoid taxes when selling a business).

        [19].     See Paul Kiel, The IRS Decided to Get Tough Against Microsoft. Microsoft Got Tougher, ProPublica (Jan. 22, 2020), https://www.propublica.org/article/the-irs-decided-to-get-tough-against-microsoft-microsoft-got-tougher [https://perma.cc/MM7T-6WNQ].

        [20].     See Paul Jones, Ohtani’s Dodgers Contract Draws California Controller’s Ire, Tax Notes St. (Jan. 10, 2024), https://www.taxnotes.com/featured-news/ohtanis-dodgers-contract-draws-california-controllers-ire/2024/01/09/7j274 [https://perma.cc/BCZ7-USU6]; Robert W. Wood, California 13.3% Tax Is Now 14.4% in 2024 but Some Say Texans Pay More, Forbes (Jan. 8, 2024), https://www.forbes.com/sites/robertwood/2024/01/03/california-133-tax-rose-to-144-in-2024-some-say-texans-pay-more/ [https://perma.cc/8UPB-N7TZ].

        [21].     Eric Boehm, Baseball Star Shohei Ohtani’s New Contract Is a Massive Tax Avoidance Scheme. Nice!, Reason (Dec. 15, 2023), https://reason.com/2023/12/15/baseball-star-shohei-ohtanis-new-contract-is-a-massive-tax-avoidance-scheme-nice/ [https://perma.cc/55TB-R37F].

        [22].     See infra Part V.C.

        [23].     We should add here that we do not necessarily agree that Ohtani’s particular scheme works, and apparently California’s Franchise Tax Board is not entirely on board either. See Michael McCann, Brendan Coffey & Robert Raiola, Ohtani’s $680M Deferment Hits Interest Income to Save on Taxes, Sportico (Dec. 12, 2023), https://www.sportico.com/law/analysis/2023/ohtani-interest-free-deferred-money-tax-1234756781/ [https://perma.cc/84TA-JKKJ] (describing the Franchise Tax Board’s policy position on deferred income and their claimed ability to tax Ohtani). Nevertheless, Ohtani’s tax planning is a salient example of the sort of tax avoidance schemes that often do work for many wealthy taxpayers.

        [24].     See infra Part III.B.

        [25].     See Daniel N. Shaviro, An Efficiency Analysis of Realization and Recognition Rules Under the Federal Income Tax, 48 Tax L. Rev. 1, 11–13 (1992).

        [26].     See Edward A. Zelinsky, For Realization: Income Taxation, Sectoral Accretionism, and the Virtue of Attainable Virtues, 19 Cardozo L. Rev. 861, 879–89 (1997).

        [27].     Goldburn P. Maynard Jr. & David Gamage, Wage Enslavement: How the Tax System Holds Back Historically Disadvantaged Groups of Americans, 110 Ky. L.J. 665, 686–87 (2022) (explaining “buy, borrow, die”); see also Edward J. McCaffery, Taxing Wealth Seriously, 70 Tax L. Rev. 305, 306 (2017) (coining the phrase).

        [28].     See David Gamage & John R. Brooks, Tax Now or Tax Never: Political Optionality and the Case for Current-Assessment Tax Reform, 100 N.C. L. Rev. 487, 500–05 (2022).

        [29].     See, e.g., Kathleen K. Wright, The Long Arm of California Stretches Even Farther, 107 Tax Notes St. 435, 435 (Jan. 30, 2023) (“The sourcing rules for income from intangibles have been one of the best tax planning tools for nonresidents of California.”).

        [30].     Tom Maloney, Anders Melin & Ben Steverman, Elon Musk’s California Exit Can Save Him $2B in Taxes, Bloomberg News (Nov. 29, 2021), https://www.bloomberg.com/news/articles/2021-11-29/how-much-tax-does-elon-musk-save-by-moving-to-texas-and-selling-tesla-stock#xj4y7vzkg [https://perma.cc/2QWN-JFTZ].

        [31].     Id.

        [32].     Id.

        [33].     See, e.g., Laura Davison, Musk’s Move to Texas May Yield Big Savings on Tesla Sale, Bloomberg News (Apr. 29, 2022), https://www.bloomberg.com/news/articles/2022-04-29/musk-s-2020-texas-move-may-yield-big-tax-savings-on-tesla-sale#xj4y7vzkg [https://perma.cc/EB9U-Q3UU].

        [34].     Assem. Bill 310, 2021–22 Reg. Sess. (Cal. 2021). For a more complete description, see Brian Galle, David Gamage, Emmanuel Saez & Darien Shanske, The California Tax on Extreme Wealth (ACA 8 & AB 310): Revenue, Economic, and Constitutional Analysis *7–16 (Ind. Univ. Legal Stud. Rsch. Paper No. 461, 2021), https://ssrn.com/abstract
_id=3924524.

        [35].     S.B. 5486, 68th Leg., Reg. Sess. (Wa. 2023).

        [36].     S. 8277B/A. 10414 (N.Y. 2021).

        [37].     H.B. 3475, 102nd Gen. Assemb. (Ill. 2021).

        [38].     H. 827, 2023 Leg. (Vt. 2024).

        [39].     S. 510, 117th Cong. § 2901(b) (2021).

        [40].     See H.R. 8558, 117th Cong. (2022).

        [41].     See David Weisbach, A Partial Mark-to-Market Tax System, 53 Tax L. Rev. 95, 95–96 (1999).

        [42].     See supra note 7 and accompanying text.

        [43].     See Rauch-Zender, supra note 7 (collecting valuation and other complaints from tax experts).

        [44].     See Brian Galle, David Gamage & Darien Shanske, Solving the Valuation Challenge: The ULTRA Method for Taxing Extreme Wealth, 72 Duke L.J. 1257, 1280–83 (2023).

        [45].     See id. at 1297–1313.

        [46].     See Fed’n Tax Adm’rs, Range Of State Corporate Income Tax Rates (2023), https://taxadmin.org/wp-content/uploads/resources/tax_rates/corp_inc.pdf [https://perma.cc/QH7B-P7RP] (listing over 40 states with state corporate income taxes).

        [47].     See Franchise Tax Overview, Tex. Comptroller (Dec. 2023), https://comptroller.texas.gov/taxes/publications/98-806.php [https://perma.cc/T9YT-BLRG] (describing the Texas Franchise Tax).

        [48].     See Oates, supra note 8, at xvii.

        [49].     See Wallace E. Oates, An Essay on Fiscal Federalism, 37 J. Econ. Literature 1120, 1120 (1999).

        [50].     See id. at 1122–23; see also Michael W. McConnell, Federalism: Evaluating the Founders’ Design, 54 U. Chi. L. Rev. 1484, 1493–95 (1987) (reviewing Raoul Berger, Federalism: The Founders’ Design (1987)).

        [51].     See McConnell, supra note 50, at 1495; see also Richard B. Stewart,  Pyramids of Sacrifice? Problems of Federalism in Mandating State Implementation of National Environmental Policy, 86 Yale L.J. 1196, 1215–16 (1977) (explaining spillovers in more detail).

        [52].     See generally supra note 8 and accompanying text (discussing fiscal federalism).

        [53].     See Stark, supra note 8, at 1393–94. An even broader version of the argument is that under full mobility, redistribution is not even possible because wages in high-tax states will be forced to adjust upwards (and vice-versa). See Martin Feldstein & Marian Vaillant Wrobel, Can State Taxes Redistribute Income?, 68 J. Pub. Econ. 369, 370, 373 (1998).

        [54].     See Saul Levmore, Interstate Exploitation and Judicial Intervention, 69 Va. L. Rev. 563, 571–72, 601 (1983); Brian Galle, Is Local Consumer Protection Law a Better Redistributive Mechanism than the Tax System?, 65 N.Y.U. Ann. Surv. Am. L. 525, 532–34 (2010).

        [55].     See Bev Dahlby, Fiscal Externalities and the Design of Intergovernmental Grants, 3 Int’l Tax & Pub. Fin. 397, 398 (1996) (explaining horizontal fiscal externalities).

        [56].     David Gamage & Darien Shanske, Tax Cannibalization and Fiscal Federalism in the United States, 111 Nw. U. L. Rev. 295, 302 (2017).

        [57].     Id.; see also David Schleicher, The City as a Law and Economic Subject, 2010 U. Ill. L. Rev. 1507, 1511–12 (discussing the inverse relationship between agglomeration and sorting that occurs in the Tiebout model); Joel Slemrod, Location, (Real) Location, (Tax) Location: An Essay on Mobility’s Place in Optimal Taxation, 63 Nat’l Tax J. 843, 850 (2010) (considering the assumed impracticality of levying taxes on mobile factors).

        [58].     See Reuven S. Avi-Yonah, Kimberly A. Clausing & Michael C. Durst, Allocating Business Profits for Tax Purposes: A Proposal to Adopt a Formulary Profit Split, 9 Fla. Tax Rev. 497, 509 (2009). That is not to say there are no complications. See, e.g., Susan C. Morse, Revisiting Global Formulary Apportionment, 29 Va. Tax Rev. 593, 607–35 (2010) (discussing difficulties of formulary apportionment).

        [59].     See John A. Swain, State Income Tax Jurisdiction: A Jurisprudential and Policy Perspective, 45 Wm. & Mary L. Rev. 319, 382–83 (2003).

        [60].     The state corporate income tax is a more complicated case.

        [61].     See Ian Crawford, Michael Keen & Stephen Smith, Taxing Goods and Services, in Tax by Design 148, 156 (James Mirrlees et al. eds., 2011).

        [62].     See, e.g., Dahlby, supra note 55, at 408–10; David E. Wildasin, Income Redistribution in a Common Labor Market, 81 Am. Econ. Rev. 757, 768 (1991).

        [63].     See Kirk J. Stark, Rich States, Poor States: Assessing the Design and Effect of a U.S. Fiscal Equalization Regime, 63 Tax L. Rev. 957, 957 (2010).

        [64].     John R. Brooks, Fiscal Federalism as Risk-Sharing: The Insurance Role of Redistributive Taxation, 68 Tax L. Rev. 89, 114–15 (2014); see Ruth Mason, Delegating Up: State Conformity with the Federal Tax Base, 62 Duke L.J. 1267, 1296, 1301 (2013) (arguing that varying local preferences justify state control over tax policy choices).

        [65].     See Stark, supra note 63, at 962–65.

        [66].     David R. Agrawal & Kirk Stark, Will the Remote Work Revolution Undermine Progressive State Income Taxes?, 42 Va. Tax Rev. 47, 54 (2022); see also Kevin Milligan & Michael Smart, An Estimable Model of Income Redistribution in a Federation: Musgrave Meets Oates, 11 Am. Econ. J.: Econ. Pol’y 406, 412 (2019) (modeling optimal degree of state redistribution when states vary in fiscal resources).

        [67].     Cf. Brian Galle, The Role of Charity in a Federal System, 53 Wm. & Mary L. Rev. 777, 828 (2012) (stating that federal redistribution may leave voters dissatisfied).

        [68].     See, e.g., Mason, supra note 64, at 1304–05; David Super, Rethinking Fiscal Federalism, 118 Harv. L. Rev. 2544, 2641 (2005). It is possible, though, that federal or federally-directed state experiments would be even more successful. Brian Galle & Joseph Leahy, Laboratories of Democracy? Policy Innovation in Decentralized Governments, 58 Emory L.J. 1333, 1398–1400 (2008).

        [69].     See Ajay K. Mehrotra, Making the Modern American Fiscal State: Law, Politics, and the Rise of Progressive Taxation, 1877–1929, 185–241 (2013).

        [70].     See Brooks, supra note 64, at 116. But see Stark, supra note 8, at 1408 (noting that redistribution may have significant spillovers that would justify federal provision).

        [71].     See, e.g., Oates, supra note 8, at 65–75; Brian Galle & Jonathan Klick, Recessions and the Social Safety Net: The Alternative Minimum Tax as a Countercyclical Fiscal Stabilizer, 63 Stan. L. Rev. 187, 210 (2010); Darien Shanske, How Less Can Be More: Using the Federal Income Tax to Stabilize State and Local Finance, 31 Va. Tax Rev. 413, 462–63 (2012). For recent empirical estimates of the optimal federal policy, see Milligan & Smart, supra note 66, at 422–28.

        [72].     See Stark, supra note 8, at 1393.

        [73].     See Gamage & Shanske, supra note 56, at 336–52; see also David Schleicher, Building Better Bailouts, in In a Bad State: Responding to State and Local Budget Crises 126–37 (2023) (discussing the flexibility the federal government has in addressing state budget crises).

        [74].     See Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 547 (2012).

        [75].     See Michael J. Graetz & Jerry L. Mashaw, Constitutional Uncertainty and the Design of Social Insurance: Reflections on the Obamacare Case, 7 Harv. L. & Pol’y Rev. 343, 363–64 (2013) (making this point about unemployment insurance taxes). But cf. Ruth Mason, Federalism and the Taxing Power, 99 Calif. L. Rev. 975, 1008–35 (2011) (arguing that federal tax incentives should face lesser constitutional constraint than conditional spending limits do).

        [76].     Galle & Klick, supra note 71, at 208–09; Barry R. Weingast, Second Generation Fiscal Federalism: The Implications of Fiscal Incentives, 65 J. Urb. Econ. 279, 285 (2009).

        [77].     Gamage & Shanske, supra note 56, at 331–34.

        [78].     Super, supra note 68, at 2650.

        [79].     See Roger H. Gordon & Julie Berry Cullen, Income Redistribution in a Federal System of Governments, 96 J. Pub. Econ. 1100, 1103, 1108 (2011) (stating that assignment of redistribution depends on balancing of mobility costs against other benefits); cf. Milligan & Smart, supra note 66, at 429 (finding that optimal fiscal decentralization depends on balance of horizontal and vertical externalities).

        [80].     E.g., Taylor Orth, Most Americans Support Raising Taxes on Billionaires, YouGov (Oct. 4, 2022), https://today.yougov.com/topics/politics/articles-reports/2022/10/04/most-americans-support-raising-taxes-billionaires [https://perma.cc/WR4X-2FS6] (reporting that nearly twice as many respondents support as oppose).

        [81].     See Slemrod, supra note 57, at 851, 855 (arguing that tax system design can reduce capital mobility, making it more likely that it is optimal to tax capital).

        [82].     For helpful overviews, see generally Brooks, supra note 64, at 102–09, and Francine J. Lipman, State and Local Tax Takeaways Redux, 101 Tax Notes St. 683, 684, 688–96 (2021).

        [83].     Stark, supra note 8, at 1390.

        [84].     See Davis et al., supra note 5, at 6.

        [85].     See Glenn W. Fisher, The Changing Role of Property Taxation, in Financing State and Local Governments in the 1980s: Issues and Trends 37, 50 (Norman Walzer & David L. Chicoine eds., 1981) (explaining that intangible property has “largely disappeared” from taxation, and some kinds of tangible property have been “eliminated or given favorable treatment”).

        [86].     Fleck et al., supra note 5, at 8–9, 18; see Darien Shanske, Revitalizing Local Political Economy Through Modernizing the Property Tax, 68 Tax L. Rev. 143, 178 (2014). We elide here a long scholarly debate on the relative progressivity of property taxes. For important recent data, see generally Daniel McMillen & Ruchi Singh, Assessment Regressivity and Property Taxation, 60 J. Real Est. Fin. & Econ. 155 (2020).

        [87].     Cong. Budget Off., Trends in the Distribution of Family Wealth, 1989 to 2019, 28 (2022), https://www.cbo.gov/publication/58533 [https://perma.cc/S259-SNFU].

        [88].     Katherine Baicker, Jeffrey Clemens & Monica Singhal, The Rise of the States: U.S. Fiscal Decentralization in the Postwar Period, 96 J. Pub. Econ. 1079, 1081–82 (2011).

        [89].     Public Spending on Education, Org. for Econ. Coop. & Dev., https://data.oecd.org/eduresource/public-spending-on-education.htm [https://perma.cc/5JQT-MBHC].

        [90].     OECD Revs. of Sch. Res., The Funding of School Education: Connecting Resources and Learning 61 (2017), https://doi.org/10.1787/9789264276147-en [https://perma.cc/3T2V-3UFF].

        [91].     Melanie Hanson, U.S. Public Education Spending Statistics, Educ. Data Initiative (Aug. 2, 2021), https://educationdata.org/public-education-spending-statistics [https://perma.cc/CA8P-QQNJ].

        [92].     State and Local Expenditures, Urb. Inst., https://www.urban.org/policy-centers/cross-center-initiatives/state-and-local-finance-initiative/state-and-local-backgrounders/state-and-local-expenditures [https://perma.cc/BEU5-DR36].

        [93].     See Baicker, Clemens & Singhal, supra note 88, at 1082.

        [94].     Urb. Inst., supra note 92. This percentage only relates to state-level expenditures.

        [95].     See, e.g., David Gamage, Preventing State Budget Crises: Managing the Fiscal Volatility Problem, 98 Calif. L. Rev. 749, 760 (2010).

        [96].     See Josh Bivens, Melissa Boteach, Rachel Deutsch, Francisco Díez, Rebecca Dixon, Brian Galle, Alix Gould-Werth, Nicole Marquez, Lily Roberts, Heidi Shierholz & William Spriggs, Econ. Pol’y Inst., Reforming Unemployment Insurance: Stabilizing a System in Crisis and Laying the Foundation for Equity 32–33 (June 2021), https://files.epi.org/uploads/Reforming-Unemployment-Insurance.pdf [https://perma.cc/ZHZ3-35BR].

        [97].     See Super, supra note 68, at 2648–50.

        [98].     See Brian Galle, The American Rescue Plan and the Future of the Safety Net, 131 Yale L.J. F. 561, 567–68 (2021).

        [99].     See Super, supra note 68, at 2609–11.

      [100].     Cf. Brooks, supra note 64, at 111 (noting this problem for safety-net spending generally). In theory there could be free-riding problems for a coalition that pushes for devolution, but wealthy interests can usually solve these problems by paying political operatives to coordinate their interests, and any opposition is likely to be diffuse and relatively disorganized. See, e.g., Ian Shapiro & Michael J. Graetz, Death by a Thousand Cuts: The Fight Over Taxing Inherited Wealth 21, 240–50 (2006).

      [101].     But see Kirk J. Stark, The Federal Role in State Tax Reform, 30 Va. Tax Rev. 407, 437 (2010) (arguing that progressivity of state tax systems should be “left to state political choice”). We might be more inclined to agree with Professor Stark’s argument if we thought that state political choices reflected actual preferences of state voters, as opposed to constrained choices shaped heavily by structural factors that lean against progressivity.

      [102].     See Edward D. Kleinbard, We Are Better Than This: How Government Should Spend Our Money 335–71 (2015) (proposing this approach for governments generally). But see Stark, supra note 8, at 1424–25 (noting some legal constraints on state ability to differentiate services by income); Super, supra note 68, at 2617–40 (describing structural features of state spending that make redistributive programs less effective).

      [103].     Roy Bahl, Jorge Martinez-Vasquez & Sally Wallace, State and Local Government Choices in Fiscal Redistribution, 55 Nat’l Tax J. 723, 737 (2002) (finding that “[s]tates that use income taxes more heavily are likely to weigh social services more heavily in their expenditure budgets, and vice versa”).

      [104].     Wildasin, supra note 62, at 766–67; David Gamage, Perverse Incentives Arising from the Tax Provisions of Healthcare Reform: Why Further Reforms Are Needed to Prevent Avoidable Costs to Low- and Moderate-Income Workers, 65 Tax L. Rev. 669, 701–08 (2012) (explaining and providing examples from the Affordable Care Act).

      [105].     See Henrik Kleven, Camille Landais, Mathilde Muñoz & Stefanie Stantcheva, Taxation and Migration: Evidence and Policy Implications, 34 J. Econ. Persp. 119, 122 (2020).

      [106].     A classic cite here is Paul E. Peterson & Mark Rom, American Federalism, Welfare Policy, and Residential Choices, 83 Am. Pol. Sci. Rev. 711, 725 (1989). For more modern reviews, see Jan K. Brueckner, Welfare Reform and the Race to the Bottom: Theory and Evidence, 66 S. Econ. J. 505, 514–18 (2000); Lucas Goodman, The Effect of the Affordable Care Act Medicaid Expansion on Migration, 36 J. Pol’y Analysis & Mgmt. 211, 213 (2017).

      [107].     See Isaac William Martin & Monica Prasad, Taxes and Fiscal Sociology, 40 Ann. Rev. Soc. 331, 336 (2014) (describing the tax-like effects of benefit phase-outs).

      [108].     See Saez & Zucman, supra note 1, at 22 (noting that transfer system provides a roughly flat amount per person).

      [109].     See Oliphant, supra note 3.

      [110].     David Scott Louk & David Gamage, Preventing Government Shutdowns: Designing Default Rules for Budgets, 86 U. Colo. L. Rev. 181, 198–200 (2015); William N. Eskridge Jr., Vetogates, Chevron, Preemption, 83 Notre Dame L. Rev. 1441, 1444–46 (2008) (listing nine additional veto-gates).

      [111].     E.g., Erik M. Jensen, Wealth Taxes Can’t Satisfy Constitutional Requirements, Bloomberg Tax (July 27, 2021), https://news.bloombergtax.com/daily-tax-report/wealth-taxes-cant-satisfy-constitutional-requirements [https://perma.cc/6WLV-8ZH6]; Philip Balzafiore, Mike Gaffney & Dylan Lionberger, The Constitutional Uncertainty of a Broad Mark-to-Market Rule for Derivatives, 172 Tax Notes State 2101, 2102–18 (2021).

      [112].     602 U.S. 572 (2024). For discussion of Moore, see generally John R. Brooks & David Gamage, The Original Meaning of the Sixteenth Amendment, 102 Wash U. L. Rev. 1 (2024).

      [113].     John R. Brooks & David Gamage, Taxation and the Constitution, Reconsidered, 76 Tax L. Rev. 75, 79–80 (2022); see also Lily Batchelder, Ari Glogower, Chye-Ching Huang, David Kamin, Rebecca M. Kysar, Kelsey Merrick, Darien Shanske & Thalia T. Spinrad, The Moores Lost Their Claim and Moore, 184 Tax Notes Fed. 1509, 1512–13 (2024) (arguing that mark-to-market taxes are constitutional after Moore); Brian D. Galle, What’s Next for Wealth and Mark-to-Market Taxes After Moore?, TaxProf Blog (June 25, 2024), https://taxprof.typepad.com/taxprof_blog/2024/06/galle-whats-next-for-wealth-and-mark-to-market-taxes-after-moore.html [https://perma.cc/NXM8-L55F] (same).

      [114].     For a prominent argument in favor of this constitutional diversity, see generally Jeffrey S. Sutton, 51 Imperfect Solutions: States And The Making of American Constitutional Law (2018).

      [115].     See Michelle D. Layser, Tax (Dis)Conformity, Reverse Federalism, and Social Justice Reform, 53 Seton Hall L. Rev. 413, 414, 420 (2022) (quoting Scott A. Moss & Douglas M. Raines, The Intriguing Federalist Future of Reproductive Rights, 88 B.U. L. Rev. 175, 180 (2008)).

      [116].     Andrew Appleby, Designing the Tax Supermajority Requirement, 71 Syracuse L. Rev 959, 963 (2021).

      [117].     Such as California, in which mere majorities of the people can and do pass tax increases. For the California supermajority requirement, see Cal. Const. art. XIIIA, § 3.

      [118].     Kevin Yamamura, Plans to ‘Tax the Rich’ Hold Risks and Rewards for California, Sacramento Bee (Dec. 27, 2011), https://www.mcclatchydc.com/news/nation-world/national/economy/article24721153.html [https://perma.cc/TV9X-WRXD].

      [119].     See Brooks, supra note 64, at 117–19.

      [120].     Arun Advani & Hannah Tarrant, Behavioural Responses to a Wealth Tax, 42 Fiscal Stud. 509, 531 (2021); see also Arun Advani, David Burgherr & Andy Summers, Taxation and Migration by the Super-Rich 15–18 (Warwick Econ. Research Papers No. 1427, 2022) (reporting “close to zero” international mobility response of the wealthy in response to U.K. tax changes).

      [121].     See, e.g., Michael Mazerov, State Taxes Have a Minimal Impact on People’s Interstate Moves, Ctr. On Budget & Pol’y Priorities (Aug. 9, 2023), https://www.cbpp.org/sites/default/files/8-9-23sfp.pdf [https://perma.cc/LY2G-KK7W] (summarizing literature and collecting new data). Notable studies that post-date the existing summaries include Katrine Jakobsen, Henrik Kleven, Jonas Kolsrud, Camille Landais & Mathilde Muñoz, Taxing Top Wealth: Migration Responses and Their Aggregate Economic Implications 5, 33, 36–37 (Nat’l Bureau Econ. Rsch. Working Paper No. 32153, 2024) (finding that Swedish and Danish wealth taxes impacts on those countries’ economies were “extremely limited”); Emily Eisner & Andrew Perry, Fiscal Pol’y Inst., Who Is Leaving New York State? Part I: Income Trends 2 (2023).

      [122].     David R. Agrawal, Dirk Foremny & Clara Martínez-Toledano, Wealth Tax Mobility and Tax Coordination 36–37 (Mar. 2022), https://ssrn.com/abstract=3676031.

      [123].     Cristina Enache, Spain Is Doubling Down on Bad Tax Policy, Tax Found. (Apr. 10, 2023), https://taxfoundation.org/spain-wealth-tax-windfall-tax/ [https://perma.cc/EU8M-H9V4]. For further explanation of variation in wealth tax rates across Spanish regions from 2011 through 2015, see Agrawal, Foremny & Martínez-Toledano, supra note 122, at 41–44.

      [124].     Agrawal, Foremny & Martínez-Toledano, supra note 122, at 7.

      [125].     Wealth Tax in Spain: Exact Percentages and How to Reduce It, Balcells, https://balcellsgroup.com/wealth-tax-in-spain/ [https://perma.cc/L4JK-2CAJ].

      [126].     See Advani & Tarrant, supra note 120, at 520–21 (explaining that a wealth tax can be compared to an income tax using an assumed rate of economic growth in the taxed assets).

      [127].     Agrawal, Foremny & Martínez-Toledano, supra note 122, at 3.

      [128].     Id.

      [129].     Id. at 59.

      [130].     Id.

      [131].     Joshua Rauh & Ryan Shyu, Behavioral Responses to State Income Taxation of High Earners: Evidence from California, 16 Am. Econ. J.: Econ. Pol’y 34 (2024).

      [132].     E.g., Editorial, California’s Tax-the-Rich Boomerang, Wall St. J. (Oct. 21, 2019), https://www.wsj.com/articles/californias-tax-the-rich-boomerang-11571697967 [https://perma.cc/ET6A-GBSA].

      [133].     Rauh & Shyu, supra note 131, at 34–35.

      [134].     Id. at 82 (explaining that of the 55.6 percent loss of revenue from taxpayer responsiveness, only 4.2 percent was due to out-migration).

      [135].     Id. at 85.

      [136].     Id. at 82.

      [137].     Cristobal Young, Charles Varner, Ithai Z. Lurie & Richard Prisinzano, Millionaire Migration and Taxation of the Elite: Evidence from Administrative Data, 81 Am. Soc. Rev. 421, 426 (2016). We put relatively little weight on the recent and supposedly contrary finding in Austin J. Drukker, Internal Migration and the Effective Price of State and Local Taxes, 32 J. Int’l Tax & Pub. Fin. 163 (2024), which reports a somewhat larger migration response in 2018 (but no response after that) to 2017 federal reforms that increased the net cost of state income tax. Among other difficulties, Drukker asserts that he can safely ignore state “workarounds” that allowed wealthy taxpayers to sidestep the impact of the 2017 reform because the IRS ultimately made some of the workarounds ineffective. But the IRS response did not come until after 2018, which is the only year Drukker claims to find a mobility response. See id. at 178 (citing Contributions in Exchange for State and Local Tax Credits, 84 Fed. Reg. 27513 (June 13, 2019)). In addition, as other researchers note, it is highly improbable that we would see an immediate real mobility response by high-earning households to tax changes, suggesting that any effects observed so quickly are likely to be changes of taxpayer’s reported address. Roberto Iacono & Bård Smedsvik, Behavioral Responses to Wealth Taxation: Evidence from a Norwegian Reform 25 (World Ineq. Lab Working Paper No. 2023/30, 2024).

      [138].     Young et al., supra note 137, at 439.

      [139].     See Ufuk Akcigit, Salomé Baslandze & Stefanie Stantcheva, Taxation and the International Mobility of Inventors, 106 Am. Econ. Rev. 2930, 2930, 2934–35 (2016); Henrik Jacobsen Kleven, Camille Landais & Emmanuel Saez, Taxation and International Migration of Superstars: Evidence from the European Football Market, 103 Am. Econ. Rev. 1892, 1903–12 (2013); Enrico Moretti & Daniel J. Wilson, The Effect of State Taxes on the Geographical Location of Top Earners: Evidence from Star Scientists, 107 Am. Econ. Rev. 1858, 1881–87 (2017).

      [140].     For example, Kleven, Landais, and Saez find that football players are two to four times more responsive to tax rates compared to other highly paid foreign workers in Denmark. Kleven, Landais & Saez, supra note 139, at 1913 & Fig. 4. Moretti and Wilson note that star scientists in the U.S. are three times more mobile than other professionals. Moretti & Wilson, supra note 139, at 1865.

      [141].     Akcigit, Baslandze & Stantcheva, supra note 139, at 2935. In addition, we believe that papers estimating mobility responses of “star” patent-holders overestimate mobility responses because of selection bias. Patents don’t happen overnight. A patent applicant controls when they will apply for a patent and can jointly choose when and where to file. They can do all their work in a high-tax state and then apply for the patent in a low-tax state. These inventors are thus an example of our point in Part III.B that money is more mobile than humans. Further, since Moretti and Wilson only observe location based on where patents are filed, they can only observe patent-holders with at least two patents. See Moretti & Wilson, supra note 139, at 1864. Thus, they see only the group of inventors who are motivated in this way to take their prior work elsewhere and never observe inventors who don’t have a second patent to file. See id. (arguing that their population patents frequently “over the period in which they patent”).

      [142].     Young et al., supra note 137, at 425; see Kleven et al., supra note 105, at 120; see also David Schleicher, Stuck! The Law and Economics of Residential Stagnation, 127 Yale L.J. 78, 111–49 (2017) (describing a set of legal and structural factors reducing American mobility).

      [143].     Young et al., supra note 137, at 428; see Advani & Tarrant, supra note 120, at 531 (noting that elasticity of relocation is highly sensitive to ease of mobility); Joshua Rauh, Taxes, Revenue, and Net Migration in California 43, Fig. A1 (Dec. 2022) (showing that nearly all of the relocation response to the 2012 California tax change was among near-retirees). Data also suggest that when entrepreneurs do leave, their business often stays behind, as they sell to new operators. Jakobsen et al., supra note 121, at 5, 23–24.

      [144].     See Marius Brülhart, Jonathan Gruber, Matthias Krapf & Kurt Schmidheiny, Behavioral Responses to Wealth Taxes: Evidence from Switzerland, 14 Am. Econ. J. 111, 135, 145–46 (2022) (reporting a revenue elasticity of -.78, with 24 percent of that caused by (reported) migration).

      [145].     See id. at 128; Advani & Tarrant, supra note 120, at 516 (making this point about the Swiss study).

      [146].     See Young et al., supra note 137, at 437–39.

      [147].     Because most states tax compensation earned in the state, no matter the residence of the earner, these cross-border tactics would largely only serve to reduce investment earnings. See David R. Agrawal & Kirk J. Stark, Will the Remote Work Revolution Undermine Progressive State Income Taxes?, 42 Va. Tax Rev. 47, 80, 83–84 (2022).

      [148].     Some early readers wondered whether the studies we survey, which mostly rely on data from before the global pandemic, reflect the rise of remote work. While it’s impossible to say for sure, we doubt remote work will matter much for the households who would face tax under our proposals, such as those with tens of millions of dollars in earnings. When it comes to the ultra-wealthy, we really aren’t talking about people who can do their job over Zoom. Whether it’s a public-company CEO, a private-equity manager, or the owner of the local car dealership, top-level managers and entrepreneurs are usually closely tied to their headquarters and the site of their business’s operations. See Young et al., supra note 137, at 425. For recent work on post-COVID-19 relocations finding little or no impact of taxes on mobility, see Eisner & Perry, supra note 121, at 12–14.

      [149].     See Agrawal, Foremny & Martínez-Toledano, supra note 122, at 56–57 (concluding that tax evasion, not actual relocation, is “the dominant mechanism” explaining the apparent shift of taxpayers to Madrid); Advani & Tarrant, supra note 120, at 531; Iacono & Smedsvik, supra note 137, at 16 (finding that the large apparent response to tax change over a single year was unlikely to be a “real response”); see also Bertrand Garbinti, Jonathan Goupille-Lebret, Mathilde Muñoz, Stefanie Stantcheva & Gabriel Zucman, Tax Design, Information, and Elasticities: Evidence from the French Wealth Tax 29–31 (Nat’l Bureau Econ. Rsch. Research Paper No. 31333, 2023) (finding that essentially all of the response to a French wealth tax reform was via misreporting). But see Jakobsen et al., supra note 121, at 18–19 (finding that Swedish taxpayer response to wealth taxes included sales of personal residences).

      [150].     Aaishah Hashmi, Is Home Really Where the Heart Is?: State Taxation of Domiciliaries, Statutory Residents, and Nonresidents in the District of Columbia, 65 Tax L. 797, 803–17 (2012); Edward A. Zelinsky, Defining Residency for Income-Tax Purposes: Domicile as Gap-Filler, Citizenship as Proxy and Gap-Filler, 38 Mich. J. Int’l L. 271, 274–78 (2017).

      [151].     Agrawal, Foremny & Martínez-Toledano, supra note 122, at 56–58; Advani & Tarrant, supra note 120, at 531.

      [152].     Agrawal, Foremny & Martínez-Toledano, supra note 122, at 57–58.

      [153].     Advani & Tarrant, supra note 120, at 524; see Garbinti et al., supra note 149, at 4–5 (describing French taxpayer responses to information reporting reforms).

      [154].     Rauh & Shyu, supra note 131, at 30.

      [155].     See Advani & Tarrant, supra note 120, at 524 (describing Norwegian enforcement efforts).

      [156].     Id. at 512.

      [157].     See Advani & Tarrant, supra note 120, at 524.

      [158].     The only available evidence we know of finds a tiny impact of wealth tax on in-migration. Jakobsen et al., supra note 121, at 32 (reporting a -0.05 elasticity of new arrivals to tax rates).

      [159].     See Mario Livio, Did Galileo Truly Say ‘And Yet It Moves’? A Modern Detective Story, Sci. Am. Blog (May 6, 2020), https://blogs.scientificamerican.com/observations/did-galileo-truly-say-and-yet-it-moves-a-modern-detective-story/ [https://perma.cc/88LP-XK69].

      [160].     Kleven et al., supra note 105, at 133.

      [161].     See supra notes 152–158 and accompanying text.

      [162].     See David Gamage, The Case for Taxing (All of) Labor Income, Consumption, Capital Income, and Wealth, 68 Tax L. Rev. 355, 387–98 (2015).

      [163].     E.g., Roger H. Gordon & James R. Hines, Jr., International Taxation, in 4 Handbook of Public Economics 1935, 1970 (Alan J. Auerbach & Martin Feldstein eds., 2002); Kleven et al., supra note 105, at 133–34.

      [164].     See Stephen E. Shay, J. Clifton Fleming, Jr. & Robert J. Peroni, The David R. Tillinghast Lecture: ‘What’s Source Got to Do With It?’ Source Rules and U.S. International Taxation, 56 Tax L. Rev. 81, 83–84 (2002); Slemrod, supra note 57, at 845–46.

      [165].     See Julie Roin, Can the Income Tax Be Saved? The Promise and Pitfalls of Adopting Worldwide Formulary Apportionment, 61 Tax L. Rev. 169, 177–78, 180–98 (2008).

      [166].     See, e.g., 26 U.S.C. § 1001(b).

      [167].     See Zelinsky, supra note 26, at 879–89. For more discussion of the policy plusses and minuses of the realization rule, see Galle, Gamage & Shanske, supra note 44, at 1268–73.

      [168].     See Shay, Fleming, Jr. & Peroni, supra note 164, at 137–38, 144.

      [169].     Galle, Gamage & Shanske, supra note 44, at 1285, 1315.

      [170].     See Roin, supra note 15, at 346.

      [171].     This strategy has some tradeoffs. In general, a benefit of the realization rule is that it gives taxpayers the option to delay sales, capturing the time value of their unpaid taxes, see generally J.B. Chay, Dosoung Choi & Jeffrey Pontiff, Market Valuation of Tax-Timing Options: Evidence from Capital Gains Distributions, 61 J. Fin. 837 (2006) (discussing tax-timing benefits), and, in the United States, also escaping tax entirely if they die while holding the appreciated asset, David M. Schizer, Realization as Subsidy, 73 N.Y.U. L. Rev. 1549, 1611 (1998). So, the option to sell before leaving a low-tax state would probably only be appealing if the taxpayer were planning on selling the asset soon anyway.

      [172].     We should keep this concern in context. As discussed above, see supra text accompanying notes 137–154, even studies that find significant and observable losses of revenue do not find that such responses undermine the absolute revenue raised. Consider again a 2012 income tax increase in California, one argued to be particularly costly by Rauh and Shyu. See supra note 131. That provision raised considerably more than the state projected. See Mazerov, supra note 121, at 56 n.108.

      [173].     See N.C. Dep’t of Rev. v. Kimberley Rice Kaestner 1992 Family Tr., 588 U.S. 262, 277–79 (2019) (describing a version of this strategy).

      [174].     See Enrico Moretti & Daniel J. Wilson, Taxing Billionaires: Estate Taxes and the Geographical Location of the Ultra-Wealthy 17–21 (Nat’l Bureau Econ. Rsch. Working Paper No. 26387, 2020); Jon Bakija & Joel Slemrod, Do the Rich Flee from High Tax States? Evidence from Federal Estate Tax Returns 4, 24–25, 34 (Nat’l Bureau Econ. Rsch. Working Paper No. 10645, 2004) (finding that estate tax influences reported, though not necessarily real, location of estates, but that “the welfare cost and revenue loss from any tax-induced migration would be small”).

      [175].     Moretti & Wilson, supra note 174, at 23 (reporting that effect of state estate taxes rises sharply with old age). But see Advani & Tarrant, supra note 120, at 532 (concluding that the literature implies that “the location choices of older, wealthy individuals are relatively inelastic” and that the Moretti and Wilson finding is caused by the desirability of San Francisco for the tech industry).

      [176].     John Buckley, Transfer Tax Repeal Proposals: Implications for the Income Tax, 90 Tax Notes 539, 540 (2001).

      [177].     26 C.F.R. § 1.1001-1(e).

      [178].     See Jonathan G. Blattmachr & Mitchell M. Gans, Wealth Transfer Tax Repeal: Some Thoughts on Policy and Planning, 54 Nat’l Tax J. 569, 571–73 (2001).

      [179].     See id.; Jay A. Soled, Reassigning and Assessing the Role of the Gift Tax, 83 B.U. L. Rev. 401, 410, 412 (2003) (suggesting that the gift tax might mitigate income shifting generally). More technically, it results in gift tax, which is, roughly speaking, the tax imposed on wealthy households when they transfer property during life rather than at death.

      [180].     Soled, supra note 179, at 423 (noting the many improvements that would be needed for gift tax to effectively constrain income shifting); see Jay A. Soled & Mitchell Gans, Sales to Grantor Trusts: A Case Study of What the IRS and Congress Can Do to Curb Aggressive Transfer Tax Techniques, 78 Tenn. L. Rev. 973, 975–86 (2011) (describing strategies for avoiding gift tax).

      [181].     Mitchell M. Gans, Kaestner Fails: The Way Forward, 11 Wm. & Mary Bus. L. Rev. 651, 657 (2020).

      [182].     See id. at 656; Dwight Drake, Transitioning the Family Business, 83 Wash. L. Rev. 123, 173–77 (2008) (describing this strategy).

      [183].     See Drake, supra note 182, at 180–83.

      [184].     See Walton v. Comm’r, 115 T.C. 589, 603 (2000) (rejecting the IRS regulation that had limited use of annuity payments to grantor); Soled & Gans, supra note 180, at 988–90.

      [185].     Gans, supra note 181, at 656; see Peter Spero, State Taxation of Trusts, 46 Est. Plan. 20, 20 (2019) (“A common use of out-of-state trusts is the avoidance of taxation of income by the state in which the settlor resides.”).

      [186].     See Jonathan G. Blattmachr & Martin M. Shenkman, State Income Taxation of Trusts: Some Lessons of Kaestner, 29 J. Multistate Tax’n & Incentives 20, 22–23 (2019).

      [187].     See Grayson M.P. McCouch, Adversity, Inconsistency, and the Incomplete Nongrantor Trust, 39 Va. Tax Rev. 419, 420–24 (2020).

      [188].     See Gans, supra note 181, at 671–74.

      [189].     See Jeffrey Schoenblum, Strange Bedfellows: The Federal Constitution, Out-of-State Nongrantor Accumulation Trusts, and the Complete Avoidance of State Income Taxation, 67 Vand. L. Rev. 1945, 1950 (2014).

      [190].     N.C. Dep’t of Rev. v. The Kimberley Rice Kaestner 1992 Family Tr., 588 U.S. 262, 270–71 (2019); see Gans, supra note 181, at 656–66.

      [191].     See Paul D. Callister, Charitable Remainder Trusts: An Overview, 51 Tax Law. 549, 559 (1998). Briefly, the charitable remainder trust allows a taxpayer to move assets into the trust, have the trust sell them—this step is tax-free even in a high-tax state, assuming its rules parallel federal law—then receive a stream of payments over time funded by the sale. Id. The taxpayer can then move before the stream is exhausted, so that in effect the sale is taxed partly at the rates of the low-tax jurisdiction. If any money is left at the end of the stream of payments, it goes to charity, but if that is undesirable, it’s often possible to arrange matters so that this residuum is zero or close to it. See Wendy Gerzog, The Times They Are Not A-Changin’: Reforming the Charitable Split-Interest Trust Rules (Again), 85 Chi.-Kent L. Rev. 849, 874–75 (2010) (explaining that the minimum value of charitable remainder interest is based on actuarial tables, but rich donors usually outlive actuarial projections, reducing value of the remainder).

      [192].     Kathleen K. Wright, Planning Opportunities in Taxation of Nonresident Partners, 2 J. Multistate Tax’n & Incentives 244, 249 (1993).

      [193].     Carolyn Joy Lee, Bruce P. Ely & Dennis Rimkunas, State Taxation of Partnerships and LLCs and Their Members, 19 J. Multistate Tax’n & Incentives 8, 17 (2010); Michael W. McLoughlin & Walter Hellerstein, State Tax Treatment of Foreign Corporate Partners and LLC Members After Check-the-Box, 8 St. & Local Tax Law. 1, 30 (2003). That is, states, for the most part, do not apply any analogue to 26 U.S.C. § 751 and § 864(c)(8), which, roughly speaking, recharacterize sales of a partnership interest as sale of the underlying assets. See Emily Cauble, Taxing Selling Partners, 94 Wash. L. Rev. 1, 21–23, 26 (2019). But California has a version of the FIRPTA provision, 26 U.S.C. § 897(g), taxing sales of partnerships that hold U.S. real estate. Cal. Rev. & Tax Code § 25125(d) (allocating a portion of partnership sale proceeds to California, based on share of partnership tangible assets or other business factors in California). California’s Franchise Tax Board has also recently announced that it will source an apportioned share of 26 U.S.C. § 751 income to California. See California Franchise Tax Board Legal Ruling 2022-02, 2022.

      [194].     States probably have constitutional authority to apply a different rule, see McLoughlin & Hellerstein, supra note 193, at 18, but as we note below it is unlikely that anti-abuse rules would have much bite. See infra notes 195–202.

      [195].     See Edward D. Kleinbard, The Right Tax at the Right Time, 21 Fla. Tax Rev. 208, 229 (2017) (offering the example of a celebrity restaurant owner).

      [196].     See Victor Fleischer, Two and Twenty: Taxing Partnership Profits in Private Equity Funds, 83 N.Y.U. L. Rev. 1, 44 (2008).

      [197].     See Mitchell A. Kane, A Defense of Source Rules in International Taxation, 32 Yale J. Reg. 311, 336–37 (2015).

      [198].     See James R. Repetti, William H. Lyons & Charlene D. Luke, Partnership Income Taxation 76–106 (7th ed. 2023) (describing the deferral system); Andrea Monroe, Saving Subchapter K: Substance, Shattered Ceilings, and the Problem of Contributed Property, 74 Brook. L. Rev. 1381, 1382, 1403 (2009) (same).

      [199].     See Roin, supra note 15, at 361–62.

      [200].     E.g., N.Y. Comp. Codes R. & Regs. tit. 20, § 9-2.3(a)(3) (2023).

      [201].     Monroe, supra note 198, at 1382.

      [202].     Leo N. Hitt, Rethinking the Obvious: Choice of Entity After the Tax Cuts and Jobs Act, 16 Pitt. Tax Rev. 67, 98 (2018); Mary LaFrance, The Separate Tax Status of Loan-Out Corporations, 48 Vand. L. Rev. 879, 881–82 (1995).

      [203].     David Kamin, David Gamage, Ari Glogower, Rebecca Kysar, Darien Shanske, Reuven Avi-Yonah, Lily Batchelder, J. Clifton Fleming, Daniel Hemel & Mitchell Kane, The Games They Will Play: Tax Games, Roadblocks, and Glitches Under the 2017 Tax Legislation, 103 Minn. L. Rev. 1439, 1448 (2019); Hitt, supra note 202, at 98.

      [204].     See Alexander Arnon & Zheli He, Effective Tax Rates on U.S. Multinationals’ Foreign Income Under Proposed Changes by House Ways and Means and the OECD, Penn Wharton Budget Model (Sept. 28, 2021), https://budgetmodel.wharton.upenn.edu/issues/2021/9/28/effective-tax-rates-multinationals-ways-and-means-and-oecd [https://perma.cc/N4XK-78JS].

      [205].     See Swain, supra note 59, at 385–86.

      [206].     Pub. L. No. 115-97, sec. 13001, 131 Stat. 2096 (amending 26 U.S.C. § 11). Because distributed corporate profits are taxed at lower capital-gains rates, whereas pass-through income from an active business taxed as a partnership would typically be ordinary income, the federal tax disadvantage to a C Corp is smaller than it might appear. And particularly for multi-national entities that can drive their effective federal rates to 10.5 percent (or lower), the C Corp can be appealing. But there is still a net disadvantage for many businesses.

      [207].     See Joseph Bankman, The Structure of Silicon Valley Start-Ups, 41 UCLA L. Rev. 1737, 1738 (1994) (explaining why private equity firms prefer C Corps for their portfolio companies); Victor Fleischer, The Rational Exuberance of Structuring Venture Capital Start-Ups, 57 Tax L. Rev. 137, 175–84 (2003) (same).

      [208].     26 U.S.C. §§ 531, 541; see Calypso Music, Inc. v. Comm’r, T.C.M. (CCH) 2000-293, at *2 (2000).

      [209].     Schleicher, supra note 142, at 86–106.

      [210].     Roin, supra note 15, at 378.

      [211].     See Part III.B.

      [212].     As the maxim goes, past performance does not guarantee future results. Also, it takes time for empirical studies to fully take account of more recent developments. For instance, it could be that changing technology, COVID-19-induced social changes, or other developments might result in greater mobility responses than what the empirical literature has found in the past. Further, when it comes to evaluating the behaviors of individual members of the super-rich, the number of such people may be too small to analyze effectively with empirical methods, and yet some of these super-rich may be so wealthy that their individual mobility decisions may have significant tax revenue implications even apart from any larger trends.

                 It is also possible that even small real migration responses could cause excess economic harm due to externalities, such that even relatively small real migration responses could potentially be more concerning than much larger tax gaming responses. However, there are also reasons for thinking that the opposite might be true, that—all else equal—tax gaming responses might be more problematic than equivalently sized real migration responses. The relevant literature does not offer conclusiveness on these questions. For discussion, see David Gamage, How Should Governments Promote Distributive Justice?: A Framework for Analyzing the Optimal Choice of Tax Instruments, 68 Tax L. Rev. 1, 63–67 (2014).

                 Overall, again, we urge caution in interpreting the implications of the empirical literature.

      [213].     See supra notes 34–35, 123–126 and accompanying text; see also Julie Zauzmer Weil, Billionaires in Blue States Face Coordinated Wealth-Tax Bills, Wash. Post (Jan. 17, 2023), https://www.washingtonpost.com/business/2023/01/17/wealth-taxes-state-level/ [https://perma.cc/VW6C-FVQG] (discussing the proposed rate in California).

      [214].     Roin, supra note 15, at 378.

      [215].     See, e.g., Greg Leiserson, Taxing Wealth, in Tackling the Tax Code: Equitable and Efficient Ways to Raise Revenue 89, 91 (Jay Shambaugh & Ryan Nunn eds., 2020).

      [216].     See Roin, supra note 15, at 346–57 (suggesting mark-to-market taxation can reduce exploitative mobility).

      [217].     See, e.g., Rauch-Zender, supra note 7.

      [218].     See Mark P. Gergen, How to Tax Capital, 70 Tax L. Rev. 1, 5–6 (2016) (noting that a wealth-type tax on securities would largely eliminate profit-shifting and use of tax havens).

      [219].     See Mitchell A. Kane & Adam Kern, The Use and Abuse of Location-Specific Rent, 76 Tax L. Rev. 277, 286–87 (2023). To be more specific, the two foundational principles of multijurisdictional taxation are source and residence, and there are debates about which of these principles should apply in specific contexts. But when source-based taxation is infeasible, as is generally thought to be for income earned from intangibles and wealth taxes, then residence stands alone as the foundational legal and normative principle. See Jerome R. Hellerstein & Walter Hellerstein, State Taxation 20.04[1] (3d ed. 2024) (“The states’ constitutional power to tax residents on all of their personal income from whatever source derived is well established.”); John A. Swain, State Income Tax Jurisdiction: A Jurisprudential and Policy Perspective, 45 Wm. & Mary L. Rev. 319, 344–45 (2003).

      [220].     See Andrew Appleby, No Migration Without Taxation: State Exit Taxes, 60 Harv. J. Legis. 55, 92–93 (2023).

      [221].     See id. at 61–63; David Elkins, A Scalar Conception of Tax Residence, 41 Va. Tax Rev. 149, 174–76 (2022).

      [222].     See Edward Fox & Jacob Goldin, Sharp Lines and Sliding Scales in Tax Law, 72 Tax L. Rev. 237, 292–95 (2020).

      [223].     See id.; Appleby, supra note 220, at 93–94; Elkins, supra note 221, at 178–86; Edward A. Zelinsky, Apportioning State Personal Income Taxes to Eliminate Double Taxation of Dual Residents: Thoughts Provoked by the Proposed Minnesota Snowbird Tax, 15 Fla. Tax Rev. 533, 553–56, 572–81 (2014); see generally David Gamage & Darien Shanske, A New Theory of Equitable Apportionment, 85 State Tax Notes 267 (2017) (defending use of single sales factor apportionment). We thank Kirk Stark for first sending us down this road.

      [224].     Galle et al., supra note 34.

      [225].     Brian Galle, David Gamage & Darien Shanske, Vermont Mark-to-Market Tax: Section by Section Summary (U. Mo. Sch. L. Legal Stud. Working Paper, Paper No. 2024-02, 2024).

      [226].     A.B. 259, Reg. Sess. (Cal. 2023) (proposing to enact a new California Revenue and Taxation Code section 50313).

      [227].     See, e.g., Lee A. Sheppard, Is the Proposed California Exit Tax Constitutional?, 178 Tax Notes Fed. 785, 790–91 (2023); William Hays Weissman, Is Physical Presence Necessary to Tax Individuals?, 107 Tax Notes St. 689, 692–93 (2023).

      [228].     See Mobil Oil Corp. v. Comm’r of Taxes, 445 U.S. 425, 445 (1980) (citing First Bank Stock Corp. v. Minn., 301 U.S. 234, 241 (1937)).

      [229].     See Amanda Parsons, The Shifting Economic Allegiance of Capital Gains, 26 Fla. Tax Rev. 308, 315–16 (2023); Wei Cui, Taxation of Non-Residents’ Capital Gains, in United Nations Handbook on Selected Issues in Protecting the Tax Base of Developing Countries 107–08 (Alexander Trepelkov, Harry Tonino & Dominika Halka, eds., 2015); see also David Gamage & Devin J. Heckman, A Better Way Forward for State Taxation of E-Commerce, 92 B.U. L. Rev. 483, 491 (2012) (“More broadly, the Court’s modern due process jurisprudence allows states to reach out-of-state actors who ‘purposefully avail’ themselves of the state’s economic market. Modern due process jurisprudence thus imposes a very light burden on a state’s ability to exercise jurisdiction over out-of-staters that do business within a state.”).

      [230].     See Mobil Oil Corp., 445 U.S. at 444–46.

      [231].     See Ford Motor Co. v. Beauchamp, 308 U.S. 331, 336 (1939).

      [232].     See Jerome R. Hellerstein & Walter Hellerstein, State Taxation § 11.05[2] (3d ed. 2024). And there is ample precedent that a state can tax all of a resident’s intangible wealth, not just income. See Curry v. McCanless, 307 U.S. 357, 368 (1939). For worldwide income, see generally Okla. Tax Comm’n v. Chickasaw Nation, 515 U.S. 450 (1995). Tangible wealth located in another state cannot be taxed. See Frick v. Pennsylvania, 268 U.S. 473, 488–92 (1925). For critique of this distinction, see generally Boris Bittker, The Taxation of Out of State Tangible Property, 56 Yale L.J. 640 (1947).

      [233].     See Adams Exp. Co. v. Ohio St. Auditor, 166 U.S. 185, 224 (1897) (“It would certainly seem a misapplication of the doctrine . . . to hold that, by merely transferring its principal office across the river to Jersey City, the situs of $12,000,000 of intangible property, for purposes of taxation, was changed from the state of New York to that of New Jersey.”).

      [234].     See Int’l Harvester Co. v. Wis. Dep’t of Tax’n, 322 U.S. 435, 441–42 (1944).

      [235].     See J.P. Morgan Tr. Co. of Del. v. Franchise Tax Bd., 79 Cal. App. 5th 245, 270, 275, 277 (2022). Other states have reached similar conclusions. See Robert Willens, Gain from Sale of Goodwill Classified as Business Income, Tax Notes St. (Jan. 23, 2023), https://www.taxnotes.com/featured-analysis/gain-sale-goodwill-classified-business-income/2023/01/20/7fvc8 [https://perma.cc/NG4T-HUM9]; see also VAS Holdings & Invs. LLC v. Comm’r of Revenue, 489 Mass. 669, 676, 678, 687–90 (2022) (following International Harvester in permitting sourcing intangibles to Massachusetts but finding the statute in question did not allow such sourcing).

      [236].     See Willacy v. Cleveland Bd. of Income Tax Rev., 159 Ohio St.3d 383, 391–92 (2020).

      [237].     See 4 U.S.C. § 114 (1996).

      [238].     See Cal. Code Regs. tit. 18, § 17951-5(b).

      [239].     RTC 19057(a). Interestingly, California has twenty years to collect a liability. RTC 19255(a).

      [240].     See Gamage & Shanske, supra note 34, at *8. This is the approach of the Vermont bill. See Galle, Gamage & Shanske, supra note 225, at 2.

      [241].     For discussion of the complex factual analysis that goes into assessing tax residence, see Daniel P. Kelly, Deferred Compensation—Delayed, but Not Forgotten, 27 J. Multistate Tax’n & Incentives 14, 17–19 (2017).

      [242].     See, e.g., comments of Kathleen Wright in Jéanne Rauch-Zender, supra note 7.

      [243].     U.S. Const. art. IV, § 2, cl. 1; id. amend. XIV, § 1; see Saenz v. Roe, 526 U.S. 489, 498 (1999).

      [244].     Austin v. New Hampshire, 420 U.S. 656, 665–66 (1975).

      [245].     Lunding v. N.Y. Tax Appeals Tribunal, 522 U.S. 287, 314–15 (1998).

      [246].     Saenz, 526 U.S. at 504–05.

      [247].     Supreme Ct. of N.H. v. Piper, 470 U.S. 274, 284 (1985) (citing Toomer v. Witsell, 334 U.S. 385, 396 (1948)).

      [248].     David Gamage & John R. Brooks, Tax Now or Tax Never: Political Optionality and the Case for Current-Assessment Tax Reform, 100 N.C. L. Rev. 487, 499 (2022).

      [249].     Id. at 501–02.

      [250].     See S. 8277B/A. 10414 (N.Y. 2021); HB 3475 (Ill. 2021).

      [251].     See, e.g., Charles Delmotte, Beyond the Wealth Tax, 76 Ala. L. Rev. 326, 355–58 (2024).

      [252].     David Gamage & Ari Glogower, The Policy and Politics of Alternative Minimum Taxes, 77 Nat’l Tax J. 467, 479–81, 487 (2024).

      [253].     See Fox & Goldin, supra note 222, at 249, 259–60.

      [254].     See David Gamage & Darien Shanske, Phased Mark-to-Market for Billionaire Income Tax Reforms, 176 Tax Notes Fed. 1875, 1875–76 (2022).

      [255].     H.R. 7502, introduced by U.S. Rep. Jamaal Bowman, D-N.Y., in April 2022.

      [256].     See David Gamage, Preventing State Budget Crises: Managing the Fiscal Volatility Problem, 98 Calif. L. Rev. 749, 754–91 (2010).

      [257].     See Gamage & Shanske, supra note 254, at 1877.

      [258].     Id. at 1878.

      [259].     For a discussion of the Court’s deferential approach to state taxation, see generally Hayes R. Holderness, Individual Home-Work Assignments for State Taxes, 98 Wash. L. Rev. 53 (2023). Nevertheless, there might perhaps be a dormant Commerce Clause claim against states like New York, in our example above, based on the argument that it is taxing all of the income from the sale of an intangible. The argument here would be that the Court often has “requir[ed] the state seeking to tax the entire tax ‘pie’ to yield to the state that was seeking to tax only a ‘slice’ of that ‘pie.’” See Jerome R. Hellerstein & Walter Hellerstein, State Taxation § 8.02[1][a][i] (3d ed. 2024). However, as the authors of this leading treatise acknowledge, the Court in 2015 appeared to “abandon[] this preference in favor of internal consistency, paying equal respect to states’ claims to all of a tax base and states’ competing claims to only a portion of a tax base, as long they do not seek to tax both bases simultaneously . . . .” Id. (interpreting Comptroller of the Treasury v. Wynne, 575 U.S. 542 (2015)). Therefore, whether such a dormant Commerce Clause claim could succeed is speculative, and probably ultimately not that important, as in the event that states like New York failed to provide a credit, taxpayers would likely just use self-help, as we explain above.

      [260].     Flora can either defer selling altogether and rely on borrowing to meet her cash needs or can instead sell before moving to New York. While the latter incurs a federal tax, that would also be the case if she were otherwise going to sell after moving to New York anyway.

      [261].     See supra notes 27–28 and accompanying text.

      [262].     For related discussion of alternative minimum tax designs, see generally Gamage & Glogower, supra note 252.

      [263].     See, e.g., Natasha Sarin, Lawrence Summers & Joe Kupferberg, Tax Reform for Progressivity: A Pragmatic Approach, in Tackling The Tax Code, supra note 215, at 317, 344; Rauch-Zender, supra note 7 (collecting valuation complaints from tax experts).

      [264].     See, e.g., Roin, supra note 15, at 349–50.

      [265].     See Fisher, supra note 85, at 50–51.

      [266].     See, e.g., Galle, Gamage & Shanske, supra note 44, at 1297–309.

      [267].     See Schizer, supra note 171, at 1596–97; Leiserson, supra note 215, at 105–06, 112.

      [268].     Schoenblum, supra note 189, at 1995.

      [269].     Milwaukee Cnty. v. M.E. White Co., 296 U.S. 268, 277, 279 (1935).

      [270].     See generally C. Joseph Lennihan, Cross-Border Collection of State Tax Assessments: A Primer, 19 J. Multistate Tax’n & Incentives 6 (2009) (discussing state tax lien process).

      [271].     Id. at 44; see also Enforcement of Foreign Judgments Act, Unif. L. Comm’n, https://www.uniformlaws.org/committees/community-home?communitykey=e70884d0-db03-414d-b19a-f617bf3e25a3 [https://perma.cc/XG7Q-9PQ7] (noting that forty-eight states have passed the uniform law). California did not adopt the uniform law but has a similar statutory method to enforce sister-state judgments. See Cal. Civ. Proc. Code §§ 1710.10, 1710.15, 1710.20.

      [272].     There are a number of cases involving New York trying to enforce judgments in Florida. See, e.g., N.Y. State Comm’r of Tax’n & Fin. v. Hayward, 902 So.2d 309, 310 (Fla. Dist. Ct. App. 2005).

      [273].     See Galle, Gamage & Shanske, supra note 44, at 1315–16.

      [274].     See Charles W. Rhodes & Cassandra Burke Robertson, A New State Registration Act: Legislating a Longer Arm for Personal Jurisdiction, 57 Harv. J. Legis. 377, 400 (2020) (“The Supreme Court has long acknowledged that non-resident defendants can consent to personal jurisdiction, which, when given in accordance with the Constitution, waives other potential constitutional challenges to the state’s adjudicative power.”).

      [275].     See, e.g., Cal. Rev. & Tax’n Code § 18633.5(i)(1)-(2) (stating that LLCs which opt for pass-through taxation and limited liability partnerships must pay a tax on distributive share of California income if the nonresident does not consent to jurisdiction); Bruce P. Ely & William T. Thistle, II, An Update on the State Tax Treatment of LLCs and LLPs, 94 Tax Notes St. 319 (2019) (listing nonresident partner-withholding state tax treatment for each state). Importantly, courts have generally upheld the assertion of state nexus with limited partners if the underlying business has a nexus with the state. John A. Swain, State Income Taxation of Out-of-State Corporate Partners, 18 Chap. L. Rev. 211, 213–14 (2014); Jerome R. Hellerstein, Walter Hellerstein & John A. Swain, State Taxation §§ 6.12, 20.08[2][a][ii], 20.08[2][a][iii] (3d ed. 1998) (describing taxation of corporate nonresident partners, limited partners, and S corporation shareholders).

      [276].     See Galle, Gamage & Shanske, supra note 44, at 1312–13, 1316.

      [277].     See id. at 1307–09.

      [278].     For an overview, see Brant J. Hellwig, The Supreme Court’s Casual Use of the Assignment of Income Doctrine, 2006 Ill. L. Rev. 751, 765–90.

      [279].     See Helvering v. Horst, 311 U.S. 112, 114, 117 (1940). Of course, Horst ultimately holds that the income belongs to parents in that situation. Id.

      [280].     See Schoenblum, supra note 189, at 1949 (making this point about income shifting with trusts).

      [281].     See Advani & Tarrant, supra note 120, at 524.

      [282].     James Edward Maule, Gross Income: Tax Benefit, Claim of Right, and Assignment of Income, BNA Portfolio 502-4th, §§ IV.B.1.b.(1)(a), IV.B.3.b.(2)

      [283].     Id. § IV.A. For extensive examples, see id. § IV.B.1.b.(1).

      [284].     For a skeptical account, see Soled, supra note 179, at 417–20.

      [285].     See Samuel D. Brunson, Grown-Up Income Shifting: Yesterday’s Kiddie Tax is Not Enough, 59 Kan. L. Rev. 457, 482–83 (2011).

      [286].     See Blattmachr & Gans, supra note 178, at 396. Annual gifts of more than about $19,000 per donor-donee pair are taxed at a 40 percent rate, and this payment does not reduce federal income tax. But the transferor’s estate would owe a similar amount if the assets instead passed at death, so the gift tax is not an obstacle to transferring any wealth that the older generation expects to hold until they die. In fact, paying gift tax is usually an excellent estate-tax planning strategy because it, in effect, “freezes” the value that’s subject to the transfer tax, as the donor pays tax on the low value the asset has in year one, instead of the potentially much higher value it has in year twenty. See Drake, supra note 182, at 160–61. Tax advisors also hint in published comments that gift-tax returns are very rarely audited. See Austin W. Bramwell, Considerations and Consequences of Disclosing Non-Gift Transfers, 116 J. Tax’n 19, 24 (2012) (describing how the practice of disclosing sales “starts the clock” in the hope that “the sale may never attract IRS scrutiny”). Thus, many donors would be happy to pay the gift tax as a “price” of (really, an additional tax savings from) transferring assets to a low-tax jurisdiction.

      [287].     See supra note 184.

      [288].     Cf. 26 U.S.C. § 2036 (treating assets held by trust in which grantor still retains an interest as still within the taxable estate of the grantor).

      [289].     An alternative solution would be to track such transfers using the notional equity interest concept we described earlier and to presume that all spending by the recipient was a distribution from that interest. Cf. Brunson, supra note 285, at 486–89 (proposing that parents should pay income tax on any appreciation of property gifted to children). Of course, if the purchase price is not spent by the seller, the payment will, itself, be part of seller’s taxable wealth.

                 If lawmakers are willing to tolerate some additional complexity, the rule could further provide that transferred assets are treated as owned by the individual whose tax rate would be higher if they owned the asset. This would account for situations where parents transfer a large portion of their wealth to later generations. This approach would also mitigate concerns that the transferee might consume the transferred wealth after transfer, so that, arguably, the wealth should not be taxed in the future. See Jay A. Soled & Mitchell Gans, Related Parties and the Need to Bridge the Gap Between the Income and Transfer Tax Systems, 62 Ala. L. Rev. 405, 422 (2011) (noting this potential problem with including gifts in the estate of the giver).

      [290].     Joseph M. Dodge, Theories of Tax Justice: Ruminations on the Benefit, Partnership, and Ability-to-Pay Principles, 58 Tax L. Rev. 399, 401 (2005).

      [291].     See Boris I. Bittker, Federal Income Taxation and the Family, 27 Stan. L. Rev. 1389, 1392–95 (1975); see also Lawrence Zelenak, Marriage and the Income Tax, 67 S. Cal. L. Rev. 339, 348–58 (1994) (considering both shared control and shared consumption as rationales for joint taxation of married couples).

      [292].     See Bittker, supra note 291, at 1398–99.

      [293].     Cf. id. at 1399 (considering administrative costs of identifying correct taxable groups). We recognize that in some circumstances there may also be arguments other than administrative cost against taxing households as a collective. For example, Tony Infanti argues that basing tax on marriage privileges marriage and individuals whose relationship can be recognized by the state. Anthony C. Infanti, Decentralizing Family: An Inclusive Proposal for Individual Tax Filing in the United States, 2010 Utah L. Rev. 605, 642–45, 664 (2010). We have no quarrel with this claim and would be open to definitions of “related party” that would include, say, both married individuals as well as other “economically interdependent” persons. Id. at 658.

      [294].     See supra notes 187–192.

      [295].     See N.C. Dep’t of Revenue v. The Kimberley Rice Kaestner 1992 Family Tr., 588 U.S. 262, 270–74 (2019).

      [296].     Id. at 269.

      [297].     Id. at 273.

      [298].     We agree with Professor Gans that while states could likely include the value of vested trust benefits as part of the wealth of an in-state taxpayer, it is uncertain whether that approach would satisfy Kaestner with respect to unvested beneficiaries. See Gans, supra note 181, at 667–70. Older cases do suggest that states could at least determine the in-state person’s tax bracket based on the out-of-state trust, cf. Ari Glogower, A Constitutional Wealth Tax, 118 Mich. L. Rev. 717, 734, 768–69 (2020) (citing Pollock v. Farmers’ Loan & Tr. Co., 157 U.S. 429, 576 (1895)) (making this point for business income), but if this option were pursued very aggressively (say, with special brackets applying to in-state wealth to replicate the tax burden that would apply if the trust wealth were directly taxable), it would probably invite modern courts to carefully reconsider those precedents.

      [299].     While the Court held open the possibility that a state might be able to reach trusts whose in-state beneficiaries borrow against trust assets, see Kaestner, 588 U.S. at 274 n.9 (noting the possibility of basing one’s taxing jurisdiction on the beneficiary’s right to assign assets, as would be necessary in a loan secured by trust assets), as a practical matter, money is fungible. Thus, it will typically be hard for a state to show that an individual with significant resources borrowed against any particular asset.

      [300].     States currently take a variety of approaches in defining a trust as a taxable resident, ranging from taxing the location of the grantor to the location of the trust administrator, or to the location where trust business is conducted. For surveys, see generally Jonathan G. Blattmachr & Martin M. Shenkman, State Income Taxation of Trusts: Some Lessons of Kaestner, 46 Est. Plan. 3 (2019); Schoenblum, supra note 189, at 1957–61.

      [301].     Fielding v. Comm’r of Revenue, 916 N.W.2d 323, 330–32 (Minn. 2018); see Gans, supra note 181, at 671–74. For detailed consideration of the law before 2018, although one we think is obsolete in light of later Commerce Clause developments, see Schoenblum, supra note 189, at 1969–89.

      [302].     Gans, supra note 181, at 683–87 (citing Kaestner, 588 U.S. at 272 & n.7.).

      [303].     Corliss v. Bowers, 281 U.S. 376, 378 (1930); see Curry v. McCanless, 307 U.S. 357, 370–71 (1939).

      [304].     Helvering v. Clifford, 309 U.S. 331, 334–35 (1940).

      [305].     See T. Ryan Legg Irrevocable Tr. v. Testa, 75 N.E.3d 184, 197–98 (Ohio 2016) (holding that Ohio grantor’s control over the non-Ohio trust gave Ohio jurisdiction to tax trust assets) (citing Curry, 307 U.S. at 370–71); Blattmachr & Shenkman, supra note 186, at 28. But see Schoenblum, supra note 189, at 1992–94 (arguing that a New York law treating some incomplete grantor trusts as grantor trusts under state law would be unconstitutional because the trust would not have “physical presence” in New York) (citing Robert L. McNeil, Jr. Tr. ex rel. McNeil v. Commonwealth, 67 A.3d 185, 192–198 (Pa. Commw. Ct. 2013)). The Supreme Court has since essentially eliminated the physical presence requirement. See South Dakota v. Wayfair, Inc., 585 U.S. 162, 175–76 (2018); see Spero, supra note 185, at 25 (suggesting that states will likely broaden trust taxation to take advantage of the elimination of physical presence requirement).

      [306].     See Clifford, 309 U.S. at 336–37 (holding that “all considerations and circumstances” are relevant to the question of who is taxable on trust income, not “mere formalism”).

      [307].     26 U.S.C. § 302(c)(2). We note that most litigation in the complete termination of interest context is over an inexplicable exception for continuing creditor interests. See Drake, supra note 182, at 175 (citing 26 U.S.C. § 302(c)(2)(A)(i)). We would recommend omitting that provision.

      [308].     See Helvering v. Horst, 311 U.S. 112, 117 (1940).

      [309].     See Clifford, 309 U.S. at 335–36.

      [310].     At a minimum, any control rule should treat a trust as “related” to the grantor if one or more of the trustees is related to the grantor, which somewhat unbelievably is not currently a feature of federal law. See Soled & Gans, supra note 289, at 422–24.

      [311].     Paul A. Gompers, Optimal Investment, Monitoring, and the Staging of Venture Capital, 50 J. Fin. 1461, 1464 (1995). For evidence that living donors exercise greater control over entities they have funded, see Brian Galle, The Quick (Spending) and the Dead: The Agency Costs of Forever Philanthropy, 74 Vand. L. Rev. 757, 782–90 (2021).

      [312].     Cf. Fielding, 916 N.W.2d at 331 n.6 (distinguishing other cases upholding state jurisdiction over a trust because they “look back, . . . years . . . , to find contacts by persons other than the Trust . . . ”).

      [313].     In some cases, though, states might treat contingent beneficiaries as though they were in control of the trust, such as where the beneficiary is consulted on or consents to trustee decisions. See Gans, supra note 181, at 687.

      [314].     Schoenblum, supra note 189, at 1950.

      [315].     N.C. Dep’t of Revenue v. The Kimberley Rice Kaestner 1992 Family Tr., 588 U.S. 262, 278 n.13 (2019).

      [316].     Cal. Rev. & Tax. Code § 17745(b); see also N.Y. Tax Law § 612(b)(40) (imposing a similar throwback tax in New York).

      [317].     See Spero, supra note 185, at 34 (discussing how throwback tax relates to beneficiaries’ benefits from living in the taxing state).

      [318].     See Schoenblum, supra note 189, at 1991 (describing the year-of-vesting rule). But cf. Osnos, supra note 18 (describing the family’s efforts to establish legal facts sufficient to show they were living in Nevada when they preferred to live in California).

      [319].     Of course, an improved rule would also close other blatant loopholes, such as the electing small-business trust scheme. See F. Ladson Boyle, Jonathan G. Blattmachr & Mitchell M. Gans, Planning Opportunities with ESBTs: Saving State and Local Income Taxes, 129 J. Tax’n 20, 21 (2018) (recommending the ESBT for high-wealth clients).

      [320].     E.g., Terry LaBant, For Snowbird Tax Savings, Avoid Homing Pigeon Instincts, 43 Est. Plan. 11, 14 (2016); see Michael Doran, The Great American Retirement Fraud, 30 Elder L.J. 265, 276–79, 294–326 (2023).

      [321].     See, e.g., Roin, supra note 15, at 368.

      [322].     For brief overviews of these “qualified” retirement savings, see Brendan S. Maher, Regulating Employment-Based Anything, 100 Minn. L. Rev. 1257, 1268–70 (2016); Eric D. Chason, Deferred Compensation Reform: Taxing the Fruit of the Tree in its Proper Season, 67 Ohio St. L.J. 347, 360–63 (2006). At a high level of generality, qualified plans either allow workers to contribute pre-tax dollars to a taxable investment account (the “standard” option) or, equivalently, to contribute after-tax dollars to an account that will then be tax free (the so-called “Roth” account).

      [323].     David I. Walker, The Practice and Tax Consequences of Nonqualified Deferred Compensation, 75 Wash. & Lee L. Rev. 2065, 2068 (2018). Annual contribution limits for most workers are usually under ten thousand dollars. See Internal Revenue Service, Publication 590-A: Contributions to Individual Retirement Arrangements 9 (2024), https://www.irs.gov/pub/irs-pdf/p590a.pdf [https://perma.cc/JBN5-VAQ8].

      [324].     Walker, supra note 323, at 2081.

      [325].     See Gregg D. Polsky & Brant J. Hellwig, Taxing the Promise To Pay, 89 Minn. L. Rev. 1092, 1134–37 (2005) (describing and criticizing cases in which lawyers and physicians benefitted from deferred tax on their compensation); see LaFrance, supra note 202, at 889–90 (noting that self-employed individuals can create shell corporations for this purpose).

      [326].     Victor Fleischer, Two and Twenty: Taxing Partnership Profits in Private Equity Funds, 83 N.Y.U. L. Rev. 1, 3–4 (2008).

      [327].     Edward J. McCaffrey, The Death of the Income Tax (Or, the Rise of America’s Universal Wage Tax), 95 Ind. L.J. 1233, 1263–64 (2020).

      [328].     Galle, Gamage & Shanske, supra note 44, at 1328–30.

      [329].     See Richard A. Posner, Economic Analysis of Law 498–99 (7th ed. 2007) (analyzing the rationale for social security system).

      [330].     Id. Savers might also be failing to maximize their own welfare, and this could justify government intervention, although it is unclear if tax incentives are an effective intervention. See Brian Galle, The Problem of Intra-Personal Cost, 18 Yale J. Health Pol’y L. & Ethics 1, 17, 19–32, 41–42 (2018). For a good in-depth discussion of these rationales, see Andrew Hayashi & Daniel P. Murphy, Savings Policy and the Paradox of Thrift, 34 Yale J. Reg. 743, 747–54 (2017).

      [331].     Admittedly, the fiscal externality from retirement savings is larger in states with more generous safety-net programs. But this probably isn’t an argument to encourage savings by families who have more than $50 million.

      [332].     See Mason, supra note 64, at 1340–41.

      [333].     See Roin, supra note 15, at 368 (citing Pub. L. No. 104-95 (1996) (codified as 4 U.S.C. § 114)).

      [334].     See Pub. L. No. 104-95, 4 U.S.C. § 114(a).

      [335].     See J. Brian Knopp, New Federal Statute Bars States from Taxing Pension Income of Nonresidents, 6 J. Multistate Tax’n & Incentives 68, 71 (1996) (noting that the statute provides that residency is determined under state law).

      [336].     See 29 U.S.C. § 1144(a).

      [337].     See Justin Elliot, Patricia Callahan & James Bandler, The Lord of the Roths: How Tech Mogul Peter Thiel Turned a Retirement Account for the Middle Class Into a $5 Billion Tax-Free Piggy Bank, ProPublica (June 24, 2021), https://www.propublica.org/article/lord-of-the-roths-how-tech-mogul-peter-thiel-turned-a-retirement-account-for-the-middle-class-into-a-5-billion-dollar-tax-free-piggy-bank [https://perma.cc/S7ZC-7V8Q]. ERISA does not prohibit state taxation of individual retirement arrangements. See 29 C.F.R. § 2510.3-2(d) (2017) (excluding IRAs from the scope of “pension plan” covered by ERISA).

      [338].     OECD Tax Policy Studies, Taxation and Philanthropy 103–06 (2020).

      [339].     See generally Brian Galle & Ray Madoff, The Myth of Payout Rules: Where Do We Go From Here?, in Giving in Time: Temporal Considerations in Philanthropy 235, 240–45 (Ray Madoff & Benjamin Soskis eds., 2023) (describing the use of private foundations for federal tax minimization).

      [340].     Galle, supra note 311, at 794–95.

      [341].     See Galle & Madoff, supra note 339, at 240.

      [342].     Ray Madoff, The Five Percent Fig Leaf, 17 Pitt. Tax Rev. 341, 346–47 (2020).

      [343].     Galle & Madoff, supra note 339, at 241 fig. 1.

      [344].     Myron S. Scholes, Mark A. Wolfson, Merle Erickson, Michelle Hanlon, Edward L. Maydew & Terry Shevlin, Taxes and Business Strategy: A Planning Approach 485–502 (5th ed. 2016).

      [345].     See Jasper Craven, There’s Never Been a Better Time to be a Scammy Nonprofit, New Republic (Aug. 29, 2022), https://newrepublic.com/article/167244/theres-never-better-time-scammy-nonprofit [https://perma.cc/NEJ5-2A5P]; Maya Miller, How the IRS Gave Up Fighting Political Dark Money Groups, ProPublica (Apr. 18, 2019), https://www.propublica.org/article/irs-political-dark-money-groups-501c4-tax-regulation [https://perma.cc/7885-KJDL].

      [346].     See Daniel Halperin, Is Income Tax Exemption for Charities a Subsidy?, 64 Tax L. Rev. 283, 287–88, 306–08 (2011).

      [347].     See Galle, supra note 311, at 768–69.

      [348].     See id. at 769–70, 790–93.

      [349].     See John D. Colombo, The Marketing of Philanthropy and the Charitable Contributions Deduction: Integrating Theories for the Deduction and Tax Exemption, 36 Wake Forest L. Rev. 657, 661 (2001); Miranda Perry Fleischer, Theorizing the Charitable Tax Subsidies: The Role of Distributive Justice, 87 Wash. U. L. Rev. 505, 517 (2010).

      [350].     Mark P. Gergen, The Case for a Charitable Contributions Deduction, 74 Va. L. Rev. 1393, 1397 (1988).

      [351].     See id.

      [352].     Galle, supra note 67, at 821–22; see Richard Steinberg, Voluntary Donations and Public Expenditures in a Federalist System, 77 Am. Econ. Rev. 24, 32 (1987) (discussing joint voter decisions over taxing and charity).

      [353].     A wealth tax cannot have a meaningful charitable contribution deduction. In a wealth tax, all money spent before the end of the year is removed from the tax base, so that effectively all consumption, whether given to charity or otherwise, is deductible. But exempting charity-owned assets from an annual wealth tax would still be quite valuable and would encourage donors to shift assets to charitable control. Cf. David L. Sjoquist & Rayna Stoycheva, The Property Tax Exemption for Nonprofits, in Handbook of Research on Nonprofit Economics and Management 303, 309 (Bruce A. Seaman & Dennis R. Young eds. 2010) (explaining the theoretical case for why a property tax exemption would encourage nonprofit ownership of real estate).

      [354].     The Constitution prohibits states from favoring charities that provide in-state benefits. See Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U.S. 564, 583–95 (1997).

      [355].     Miguel Almunia, Irem Guceri, Ben Lockwood & Kimberly Scharf, More Giving or More Givers? The Effects of Tax Incentives on Charitable Donations in the U.K., 183 J. Pub. Econ. 104114, at 3 (2020). In the U.K. system, donors file a form with the government to report their donation, and the government then makes a matching payment to the donee charity. Id.

      [356].     See Wendy Gerzog, Toward a Reality-Based Estate Tax, 57 B.C. L. Rev. 1037, 1058–59 (2016).

      [357].     Stark, supra note 8, at 1417–22; see Galle & Klick, supra note 71, at 223–35 (reporting evidence that the SALT deduction is correlated with higher safety-net spending and personal income-tax revenue).

      [358].     See Howard Gleckman, Buckle Up. 2025 Promises to Be an Historic Year in Tax and Budget Policy, TaxVox (June 7, 2023), https://www.taxpolicycenter.org/taxvox/buckle-2025-promises-be-historic-year-tax-and-budget-policy [https://perma.cc/48S2-BZXX].

      [359].     See supra note 69 and accompanying text.