Category: Uncategorized

STRIB article

A couple of weeks ago, the Strib ran a front-page story on property tax increases on St. Paul homes. Shannon Prather, Kate Galioto, and Dave Orrick, Squeezed on Home Front (STRIB, 12/11/22). The headline is from the print story; the online headline differs slightly.

Both St. Paul and Minneapolis raised their levies for taxes payable in 2023 by quite a bit, as have Ramsey and Hennepin counties. St. Paul’s increase is magnified by the city shifting street and other assessments to its property tax to respond to a lawsuit invalidating frontage assessments. The combined effect is a 15% city levy increase. Disproportionately large increases in home market values add to their tax increases since property taxes spread fixed dollar levies over the tax base. The effects appear to be largest in neighborhoods with lower value homes, according to the article. So, the increases will affect lower income homeowners most heavily.

The article does not mention it, but the structure of the state homestead exclusion, which drops as values rise, contributes to the increases for lower-value homes with value increases. This is a side effect of the legislature’s denial of the exclusion to higher value homes.

The story provides the standard fare, based on interviews of homeowners with big increases (55%, 28%, and 19% are three examples cited), as to the adverse effects on their personal budgets. The third graph of the story quotes one homeowner:

“You are squeezing us out. Why don’t you buy us out, us older folks?” said Stickney, a runner for a title and escrow company. “It’s just impossible.”

Shannon Prather, Kate Galioto, and Dave Orrick, Squeezed on Home Front (STRIB, 12/11/22).

Why it interested me

This is a scenario that periodically plays out. I cannot remember how many similar stories that I have seen over the years. A lot. So, a story like this is not something I would ordinarily pay much attention to, but in this case three of its paragraphs highlight a systematic problem with how state homeowner property tax relief is delivered and thus perceived.

These are the first two paragraphs of the story:

St. Paul homeowner Roxanne Stickney will spend 10% of her take-home pay on her property tax bill next year.

The taxes on her two-bedroom East Side home will jump more than 55% in a year’s time to more than $3,000. The single mom’s frustration spilled out as she, along with dozens of other demoralized homeowners, addressed the City Council last week.

Shannon Prather, Kate Galioto, and Dave Orrick, Twin Cities property taxes jump, cutting deep into family budgets (STRIB, 12/11/22)

Take-home pay is not a complete measure of income. But Ms. Stickney must be paying well over 7% or 8% of her income in property tax, well over her state income tax rate.

A large portion of the story is spent interviewing local and state officials, a fair number of whom blame the problem on the state’s failure to increase aid to local governments, which help to hold down property taxes. The third paragraph (the one caused me to write this post) is buried near the end of the story, when it discusses potential state property tax relief programs:

Homeowners who see property tax increases of more than 12% can tap into a state property tax refund program. There’s also a state program that allows older homeowners earning less than $60,000 a year to defer tax payments.

Shannon Prather, Kate Galioto, and Dave Orrick, Twin Cities property taxes jump, cutting deep into family budgets (STRIB, 12/11/22)

I had two reactions to that paragraph, one mundane but revealing of a second more systematic problem with the political economy of “circuit breaker” style property tax relief. The first was that it reflected inattentive or incomplete journalism that illustrated the effective invisibility of the income adjusted property tax relief provided though the state Homestead Credit Refund or HCR.

Incomplete reporting

I don’t want to be too harsh. Tax issues are complicated and most journalists cover them only sporadically. Reporters typically are liberal-arts-educated generalists with little finance or economics training, putting the nuances of tax outside their wheelhouses. But the story has an important omission in the third paragraph I quoted above. It ignores the state’s main property tax relief program.

The paragraph identifies two state property tax relief programs that could help the homeowners covered by story – (1) special (“targeting”) refunds of up to $1,000 for those with property tax increases of more than 12% and $100; and (2) the deferral program for seniors who have owned their homes for more than 15 years and have incomes below $60,000.

This is accurate, as far as it goes, but is incomplete. These two programs are minor, at best. The special refund or targeting will pay an estimated $7.4 million in refunds for FY2023, based on the November MMB forecast (p. 9). That compares with more than $5 in billion property taxes paid by homes. In other words, a trivial amount. Yes, it is explicitly directed at the crux of the story, large property tax increases on homes, so it will help the homeowners with the increases for a year or two. The second program, senior deferral, has struggled to attract enrollees. In 2020 it was reported to have 341 enrollees. An even more minor program.

By contrast, the state has a very generous income adjusted, property tax relief program for homeowners, the state Homestead Credit Refund or HCR, which the story does not mention.  This program is estimated to deliver $616.7 million in state refunds to homeowners for FY 2023 under the November forecast. It provides relief when property taxes exceed specified percentages of household income (the highest threshold is 2.5%) and income is below about $125,000. It often pays half of the tax (even more for those with incomes below $60,000) up to a maximum of more than $1,500.

Applying this to the very sympathetic case in the first two graphs of the STRIB story, Ms. Stickey is very likely to have this program provide a refund of 50% or more of her property tax, considerably mitigating the impact of the increase. If she was getting a state refund for half or more of her property taxes, the lower numbers involved make for a less compelling a story. (Algebraically, the percentage increase does not change, but the dollar amount is more modest.) Unlike targeting, the HCR’s relief applies year after year. The failure to include this in discussing state relief programs is an unfortunate oversight, an error of omission. Especially when two minor programs are referenced. Targeting will likely provide additional temporary relief for Ms. Stickley further mitigating the short run effects of the increase.

Interestingly, when I looked at the article online. The Internet link included in the story was to a DOR page that described both the special refund or targeting credit and the regular HCR. So, if whoever generated the link had read the detail, they would have discovered the more generous and widely applicable program. I don’t want to be too critical given the time demands on reporters and decreasing numbers of them with the hard times the newspaper business has fallen on.

Invisibility of the HCR

More importantly, the story illustrates a more systematic problem with the HCR program.  It too often is overlooked or forgotten in the public conversation about homeowner property taxes.

The HCR is a circuit breaker style program; as property tax rises as a percentage of income, the circuit breaker kicks in and provide state relief. Hence, the description as a circuit breaker. The income thresholds that trigger relief range from 1% at the lowest income levels to 2.5% at the highest eligible incomes. The refund percentages (i.e., how much the state refund effectively pays above the threshold percentage of income) ranges from 95% (!!!) at the lowest income level to 50% for those with the highest eligible incomes (about $130,000 for 2023). The maximum credit is over $1,500. The maximum starts to phase out above $100,000 of income.

Describing its parameters make it obvious how very generous the program is for modest income homeowners. If the state’s goal is to provide relief to average income homeowners who have a difficult time affording their property taxes, it is a cost-effective way to do that. It only provides relief to homeowners (e.g., not owners of other property types) and only to those whose property taxes are high relative to their incomes. Providing state aid to local governments (extensively covered in the story) is a very expensive and untargeted way to provide relief to homeowners.

The political economy problem with the HCR is its relative invisibility. Hence, the story’s failure to discuss its effects (I assume anyway). I think that stems from a confluence of factors:

• It’s disconnected from the actual property tax bill. Because it is based in part on income, homeowners must pay their property taxes and separately apply to the state for a refund. Providing it directly on property tax bills would require either making property tax bills private or disclosing homeowners’ incomes to the public. In many cases, application will be done by their tax preparer as part of filing income taxes or by individuals using income tax preparation software. As a result, many individuals may not perceived it as reduced property taxes but just some state refund, perhaps of income tax.
• Many eligible homeowners likely do not claim their refunds. A recent DOR revenue estimate asserts 425,000 eligible individuals (that includes renters eligible for the renters property tax refund) do not file. I’m not sure how that estimate was done or how accurate it is, but it follows conventional wisdom for circuit breaker programs generally. (As an aside, if nonparticipation is really that high, it is borderline scandalous. It dramatically undercuts the effectiveness and equity of the program.) The circuit breaker/HCR structure is complicated, making it hard for people to know if they qualify. A typical homeowner may check to see if she qualifies in one year and if she doesn’t, then assume that will continue in future years. If her income drops, property tax increases, or both, she may become eligible but be unaware. Tax preparers may not check. Most taxpayers who do their own income tax use tax preparation software. Some software does not include the HCR and if it does, users may simply opt not to go through the necessary routine (questions) to see if they qualify because they did not in a prior year.
• It is a PR and political orphan. There is no natural interest group that publicizes the availability and parameters of the HCR. This is what the STRIB story illustrates in spades. No one, other than homeowners themselves, has a vested interest in publicizing or reminding the public about the HCR. By contrast, state aid to local governments has a legion of lobbyists and local officials banging the drum for aid program funding in the legislature and public conversations to justify their legislative asks for more aid or the lack of it to justify why they increased their levies. Reporters who heavily rely on information from public meetings, legislative discussions, and interviewing elected officials pick up on this. By contrast, nobody publicizes the HCR. As a result, it fades from view and public consciousness.

It’s worth noting here, repeating points I made in my post on city sales taxes, that property tax relief or reduction is not a principal purpose for state aid to cities, counties, and schools. It is a nice side effect, but the real purpose of those aid programs lies elsewhere. Their primary purposes differ somewhat for categorical aid (e.g., transportation aid for schools) and block grant aid (e.g., LGA). But in general, they seek to ensure an adequate level of service across the state (e.g., a quality education. social services, etc.) without regard to tax base and/or to adjust for the effects of spillover costs and benefits resulting from local government service delivery and taxation. But local governments and their lobbyists for obvious reasons tend to play up the property tax relief effects of state aid, rather than focusing on their primary purposes which are less politically appealing. Both Democrats and especially Republicans like tax cuts. So, no surprise proponents of state aid increases pitch them as property tax cuts. But state aid to local governments is an expensive and a very cost ineffective way to deliver property tax relief to homeowners. The HCR is more targeted and effective, as well as cheaper.

What can be done?

Since one of HCR’s fundamental problems is a lack of a natural mouthpiece or megaphone publicizing its benefits, one solution is for the state to take responsibility or to hire others to do so. That could be through advertising (beyond PSAs run when no one is watching), outreach to tax preparers, providing online tools on DOR’s website to calculate eligibility and estimated benefits, and similar. The legislature would need to appropriate money to do that and provide some direction as to how to do so. However, it is not something that government entities are typically very good at doing.

Another possibility would be to simplify HCR’s parameters to make program eligibility and benefits easier to understand. That could help homeowners more easily intuit its benefits without going through complicated calculations. A favorite idea of mine is to set one threshold – e.g., 2% of income to pick an arbitrary number – that applies to all income levels. This would allow saying something like: “If you pay more than 2% of your income in property taxes and your income is lower than $130,000, you can get a refund.” Making a change like that would shift benefits under the program around a bit (full disclosure: hurting those with lower incomes). But if it results in more people benefiting because they now file, that may be a reasonable tradeoff.

In any case, greater efforts to publicize the program are needed and simplification of its parameters (perhaps sacrificing some equity and progressivity) could help.

Other HCR improvements

I can’t let the opportunity pass without listing my three pet ideas for improving the HCR. I pitched these ideas at various times to legislators when I was working and they were rebuffed or ignored, so I am under no illusion that they will be seriously considered but I still feel compelled to describe them and why I think they make sense. Their lack of political appeal will be especially true in the current environment of mega surpluses, because the first two changes are premised on the idea that the program is overly generous and could be cutback without harming its basic function.

My three ideas are:

• Limit the tax that qualifies for the HCR to that attributed to a specified amount of market value, i.e., limiting the qualifying tax on very high value homes.
• Redefine household income to add back depreciation deductions.
• Increase the maximum refund by a material amount but only if the first two suggestions are adopted.

Short explanations of my rationale follow.

Market value limit.  The HCR allows a homeowner to claim a refund based on the total tax paid on the home without regard to its value, whether it is worth $150K or $2M. The maximum limit on the refund (about $1,500) prevents this from getting out of hand. But allowing the tax on an unlimited amount of market value rewards individuals with a strong preference for consuming housing. So, one family that prefers spending a high percentage of their income on housing benefits at the expense of another that allocates more resources to other consumption. That is both unfair and market distorting.

It also helps people who have variable incomes like business owners and those working on commission; they buy homes based on their average incomes over multiple years and the HCR helps pay their property tax in years when their incomes are low. If their incomes were more stable, they would get less. I think this is a flaw; others disagree and think it is a feature.

[More explanation on the politics of this change: Individuals who purchased homes long ago in now hot real estate areas, causing their home values to rise a lot, present sympathetic political cases. Their luck in realizing outsized value increases (they won a mini lottery of sorts) is now “taxing them out of their homes” in political parlance. Legislators usually want to protect them. As an abstract policy matter, they probably don’t deserve it. Few elected officials would dare to tell them that, even indirectly by limiting their HCRs.]

I would address this by limiting the tax that qualifies for the HCR to a maximum market value amount – it could be some multiple of the state median homestead market value, the market value break point for the 1% class rate, or another amount. It should be indexed for inflation but should not vary by county or region to keep things simple, even though that favors homeowners Greater Minnesota where home values are systematically lower.

Obviously, this change would reduce refunds paid. The savings could be plowed back into the program by increasing the maximum credit (see below) or by making other enhancements. Or it could wait to a time when the state budget it tight and savings need to be found.

Redefine household income. Circuit breaker programs are based on income, so a definition of income must be devised. In a perfect world, a comprehensive measure of annualized lifetime (permanent) income would be used. This would filter out the year-to-year variability and the distortions caused by tax accounting. Of course, we’re stuck in a world where practicality dictates reality.

The HCR (and property tax refund for renters) uses household income, a very broad definition of income that includes most exempt income, such a tax-exempt bond interest, untaxed social security benefits, and similar, to determine eligibility and the refund amount. The idea is to come up with as broad a measure of a homeowner’s ability to pay property taxes as practical using tax accounting rules.

One flaw in this definition is that it fully allows owners of businesses, including real estate investors, who qualify as active under the passive loss rules to reduce their incomes with depreciation deductions. These deductions are (in common parlance) “paper losses.” Economic depreciation (i.e., the economic cost of a business’s machinery wearing out) should be allowed to reduce income but tax depreciation rules allow much faster deductions to encourage investment. Section 179 and bonus depreciation allow full deduction in the year a machine or equipment that may provide years of service is put in service. Allowing these paper losses mismeasures the ability to pay property taxes. It’s not practical to disallow only the acceleration (so true economic depreciation could be deducted), so I would disallow all depreciation as a better compromise than allowing the accelerated amount. This is, after all, a low and middle-income relief program, not an income tax.

As it stands now, I suspect that quite a few business owners (particularly of real estate) qualify for maximum HCRs on valuable homes. In my opinion, this is inconsistent with the purpose of the HCR. I do not have hard evidence for it but two anecdotes (the first of which I privately pointed out to a few legislators to no effect) provides what I think is good evidence for it.

1. In 2011, Governor Dayton’s budget proposed a state tax on homes with estimated market values over $1 million dollars. DOR prepared an incidence analysis of the taxes in his budget (i.e., an analysis of the economic burden of proposed taxes by income class). This analysis (Table 1, p. 5) showed that 9% of the tax on million-dollar homes would be paid by individuals in the very lowest income decile (income lower than $11,298). As the analysis notes (p. 2) that “reflects taxpayers with low incomes because they reported large business losses.” Obviously, almost 10% of the state’s most valuable homes are not owned by its lowest income residents. That anomaly occurs because we’re not measuring income very well, (DOR’s incidence database was used to do the estimate. It assumes tax depreciation is a real reduction in measuring business income and, of course, is stuck using annual rather than permanent or lifetime income, so year-to-year variability is a factor.) Much of that is likely attributable to depreciation deductions. To be fair, some of the effect may be due to the recession, since the income data used for that analysis were from calendar year 2008, extrapolated to 2013. 2008 was the first full year of the Great Recession. In any case, most of these individuals (i.e., for homesteads not second, third, or fourth homes) received the maximum HCR. One would assume that most have plenty of ability to pay the full taxes, despite either paper losses or transitory low incomes.
2. The ProPublica stories revealed a fair number of very wealthy individuals paying little or no federal tax because they used business losses (again, probably mostly accelerated depreciation or similar paper losses) to zero out their incomes or simply by living on borrowing against their portfolios. If they lived in Minnesota, they would get the maximum HCR on their homesteads. ProPublica’s stories, of course, revealed that some of them received federal tax benefits intended for lower income taxpayers.

As an aside, the 2022 House tax bill greatly simplified the definition of household income by adopting AGI as its income measure. I generally support simplification for a variety of reasons, but not in this case. Major simplification could be achieved by dropping some adjustments for retirement plan contributions and distributions, which involve more minor amounts but add a lot of complexity, but it is unnecessary to drop large categories of income, like untaxed social security, tax-exempt bond interest, and similar, that are easily known and simple. Those changes increase the cost of the program while decreasing its horizontal equity. Using the broadest practical measure of income become even more important if efforts are undertaken to increase participation or the maximum refund is raised.

Raise the maximum credit. My third idea is to increase the maximum HCR amount substantially, perhaps doubling or tripling it. This should only be done if both the market value limit and income definitional changes are made. Otherwise, increasing the maximums would be a bonanza for those faux low-income owners of very valuable homes. For example, wealthy owners of real estate interests (the Donald Trumps or Steve Rosses of Minnesota) would get multi-thousand-dollar refunds of their property taxes year after year. Proposals to do that were made various times during the last decade without any changes in income or qualifying tax. Fortunately, they were not enacted.

Another illustration: When the targeting refund program was reenacted in the early 1990s, it did not have a maximum refund limit. That caused at least one 5-figure refund to be paid to a homeowner with a very large tax increase. The legislature responded by imposing a $1,500 maximum refund limit in 1992 (art. 2 § 30), the first year after the sizable refund was revealed. That limit was further reduced to the current $1,000 in 1994 (art.4 § 3) and has not been adjusted for inflation.  It probably should be increased after almost 30 years of inflation.

Update 1/20/2023

A recent MinnPost column on St. Paul’s proposal to increase its sales tax for street improvements provides another illustration of the disconnect regarding the effects of the HCR. The column grudgingly accepts the idea of increasing the sales tax because of the desperate state of the city’s streets. It allows that the tax is regressive but waves that off as still being slightly better than using property taxes:

Even though Minnesota exempts food and clothing, sales taxes remain regressive (though marginally less so than the property tax).

Bill Lindeke, Like it or not, St. Paul needs a new sales tax to fix roads (1/20/23)

That statement may be close to accurate if you ignore the effect of the HCR. Here’s a quote from the Department of Revenue’s 2021 Tax Incidence Study that directly addresses the issue of the two taxes’ relative regressivity (its use of “PTR” and “property refunds” refers to the program that includes both the HCR and the similar program for renters):

Although residential property tax burdens (after PTR) are regressive, they are noticeably less regressive than either sales taxes or “all other taxes.” This is mostly due to the impact of property tax refunds. In their absence, the Suits index for residential property taxes would be -0.194 – much closer to that of state and local sales taxes (-0.196).

DOR, 2021 Minnesota Tax Incidence Study, p. 33 (3/4/2021) [Emphasis added and footnote omitted.]

Moreover, if the city were to raise property taxes to pay more for streets and roads, homeowners who qualify for the HCR and are not receiving the maximum refund (most) would see an increase in their HCRs pay part of the tab. Put another way, the state would indirectly pay some of the cost through the HCR. For many homeowners, it would be half of the cost. The HCR functions as a sort of indirect state aid paid to homeowners, rather than local governments.

Note: I’m not advocating St. Paul finance street and road improvements with property tax increases, just pointing out the HCR’s interaction with property increases and how that reality is typically missed by thoughtful observers. I may (or not) post some of my thoughts about St. Paul’s proposal from a more general perspective. A big element of the problem, not discussed in the MinnPost column, results from inflation’s erosion of gas tax revenues. Local governments receive a share of the state’s gas tax revenues. Its failure to keep pace with inflation (i.e., the legislature’s failure to increase its fixed-cents-per-gallon rate) has caused those revenues to lag dramatically. I covered that in this post. The failure of the tax to keep up with inflation has cost local governments billions of dollars in state aid for streets, road, and highways over the last few decades. That lost aid surely has been a factor in the deferred maintenance of St. Paul streets.

The November forecast has put the stadium reserve in front page headlines (STRIB) with the projection that by fiscal year 2027 its balance will be more than double the outstanding bonds. Its gargantuan size almost guarantees the 2023-24 legislature will politically need to resolve what to do with this largely accidental account. Its big forecast balance has led legislators, per the STRIB article, to call for paying off the bonds when they can be called beginning June 2023 (projections show there will be sufficient funds in the account to do so).

The reserve is a political construct, a gift that continues giving to the Vikings

I have written about the stadium deal and reserve fund (here, here, and here). The last of those posts was written 18 months ago and argues for repealing the reserve account. Its projected balance has increased since then but the financial and legal reality that underlies the reserve has not. For anyone who wants the gruesome details of my thinking, you can read Stadium Reserve – it’s time to move on. The following is a shorter summary:

• The stadium bonds are general fund appropriation bonds. Their only legal security is a general fund appropriation. Under the constitution, the bonds are not considered debt because they are not secured by a pledge of state taxes. That allowed passage of the stadium bill by a simple rather than 60% majority vote, which was correctly considered politically impossible. Payment of the bonds is a responsibility of the general fund, pure and simple. The reserve does not provide legal or financial security for the bondholders.
• To offset the general fund’s cost of paying the bonds, the 2012 legislature authorized electronic pull tabs. The operative political narrative was that would provide enough new gambling tax revenues to pay the one-half billion in stadium bonds. That provided a political argument that financing the stadium would not reduce general funds for ordinary state purposes (education, human services, etc.). Minneapolis city sales tax revenues also are legally obligated to pay the bonds but that obligation was deferred until the city’s convention center bonds were paid off in 2021. So, the initial annual drain on the general fund was bigger than now.
• The 2012 revenue estimate projected a bonanza of revenue from e-pull tabs, pretty much immediately – a lot more than needed for the stadium, even with the Minneapolis sales tax contribution deferred! That led to concerns by some legislators that those moneys would be used for other purposes. (The vast majority of insiders did not believe the estimate FWIW.) That estimate proved woefully wrong in the short run because it took multiple years for sales of e-pull tabs to ramp up. The data now show that they have done so.
• The reserve account was set up not to pay the bonds (that was a legal no-no according to bond counsel because it would make the bonds look like general obligation bonds). MMB has, however, treated the reserve as the way the bonds are paid. That probably was done to serve former Governor Dayton’s political purposes, to cover his support for the stadium. He (and legislative supporters) wanted to say the stadium did not divert regular general fund resources; of course, it did initially and thanks to the reserve continues to do so.
• This was a bipartisan deal, Democratic governor and GOP legislature. Republican and Democrats typically do not agree on how public money should be used and they certainly would not want the other party to decide if it controlled all of state government in the future. So, the bill had to make provision for how the estimated extra money – if it materialized – would be used. The deal did that by putting it into a reserve to pay for cutting lawful gambling taxes (that option expired before the revenue increase occurred) and stadium costs (the only legal option that remains). But importantly, the law’s set aside did not accurately measure “new” revenues; it sequestered any nominal increase in lawful gambling taxes off an artificially low baseline set during the Great Recession. The legal dedication (all money above a fixed dollar amount based on the February 2012 forecast) was done to make the accounting simple and easy for the state bean counters. No one was thinking about the effects a decade later when all of the cumulative growth in gambling tax revenues would be dedicated to the stadium. That’s where we are now.

Bottom line

1. The public share of the stadium, about $500 million, always was and is a state general fund and Minneapolis sales tax responsibility. The bonds are payable from the general fund and lawful gambling taxes have always been a general fund resource. They are in lieu of general sales taxes. The Minneapolis sales tax contributes after the convention center bonds were retired in 2021.
2. The dedication of tax revenues to the reserve, thus, sets aside more lawful gambling taxes than necessary to pay for the state’s share of stadium costs and related expenditures. That is obvious because the general fund is both paying the bonds and setting aside large balances in the reserve. If the law remains unchanged, the reserve will double how much taxpayers are subsidizing the Vikings and pro football. Spoiler alert: the law will be changed. The details of how that is done will determine how much extra the Vikings realize from the reserve dedication.
3. Changes to the reserve (e.g., as proposed by Governor Walz in 2021) likely were blocked by key senators (Bakk and Rosen) who were architects of the 2012 deal (Bakk, for the 2013 augmentation of the reserve) and were committed to the goofy reserve dedication. Both will be gone in 2023, clearing the way for commonsense changes that restore general fund resources to more traditional government uses (not pro sports owners).

Much of the reserve’s money was not generated by the stadium bill and it’s unclear why the stadium and Vikings should have a claim on them.

Those assertions go against the popular narrative that circulates in the media and halls of the capitol. There is a general perception that:

• The stadium bill’s authorization of e-pull tabs created a lot of new general fund tax revenues that otherwise would not have been collected; and
• Because the stadium deal caused that to happen, the stadium (i.e., Vikings since it is de facto their facility, notwithstanding its technical public ownership) has an implicit claim on or should be a beneficiary of those revenues.

First, the stadium reserve does not measure (in economic terms) the additional state general fund revenue that resulted from passage of the 2012 stadium deal. This is so for several reasons:

• It captures all growth in gambling revenue, not just that from e-pull tabs, over the 2012 February forecast baseline. That baseline was during the Great Recession. Economic recovery and inflation, thus, are big contributors to the stadium reserve balance – beyond whatever effect allowing e-pull tabs had.
• Taxes on purchases of e-pull tabs are not necessarily “new” money to the state. In many cases, money spent on e-pull tabs would have been spent on other taxable items (entertainment, restaurant meals or whatever). So even if the accounting were limited to e-pull tabs, it would overstate whatever new revenues the 2012 gambling expansion created. (Money that otherwise would have been spent on tribal gaming, trips to Las Vegas, illegal gambling is new state revenue. It’s not at all clear why the tribes should be expected to pay for a disproportionate cost of the stadium, To the extent e-pull tabs generated truly new state revenues, much of that likely came from tribal gaming interests. But that is another issue.) Do e-pull tabs materially increase Gross State Product? Doubtful. The reserve account dedication likely diverts a lot of revenue from other general fund purposes to the stadium, revenues the general fund would have had even if the stadium bill had not been enacted. Answering that question would require a sophisticated econometric analysis, obviously. But conventional economic wisdom suggests that neither pro sports subsidies nor a minor gambling expansion will generate a material increase in state economic activity (certainly not over and above paying for a half billion public subsidy for the stadium itself).
• Because the projected shortfall in estimated gambling tax revenue became quickly, the 2013 legislature allocated millions more to the account from cigarette and corporate tax revenues. (It did this, in my opinion, to provide political cover for the questionable – short-run – financial assumptions in the original 2012 deal.) Those moneys are still trapped in the reserve.

Second, there is no reason in principle that the stadium or the Vikings should have a claim on tax revenues from lawful gambling. The tax on lawful gambling is in lieu of the sales tax, a general tax on consumption. It’s not practical to tax gambling with a sales tax (re-bet moneys get taxed multiple times), so alternatives must be devised. The fact that the expansion was part of a political deal made a decade ago does not create a moral claim on the money. It would be different if the tax fell on users or beneficiaries of the stadium. It does not. Continued diversion provides an ongoing budgetary advantage for pro sports subsidies.

Nor was it what the 2012 supporters of the deal intended if they had presciently known what would happen. Reviewing the circumstances a decade ago will explain why I’m quite sure the legislators supporting the 2012 stadium deal did not intend to allocate materially more general fund money to Vikings than the explicit dollar commitments made in the bill. This was the political context:

1. The prior (2011) legislative session was dominated by an acrimonious budget deadlock. The state had a $5 billion budget shortfall. The GOP legislature sought to make up all of that with budget cuts while Governor Dayton insisted on some tax increases. It was only resolved after a multiweek government shutdown and by agreeing to deficit borrowing. A Minnesota Supreme Court decision was required to validate that borrowing. So, neither side got what they wanted and prudent budget practices (don’t borrow for current operating expenses) took it in the shorts.
2. Matters looked a little, but not much, better in 2012 when the Vikings stadium bill was passed. Neither side was willing to cut the budget or raise general taxes to finance the stadium. That’s almost a pro sports subsidy iron political rule. But they wanted to finance a stadium and keep the Vikings in Minnesota on a bipartisan basis, albeit by the narrowest of margins. That required turning to a budget ruse (i.e., the illusion that tax revenues from expanded gambling are truly incremental and that they would materialize almost instantly). It passed on close margins and was never a sure thing. Many important legislators had conflicted feelings about it. For example, the House Speaker voted against it but obviously implicitly supported it, since he allowed it to proceed through the legislative process, something he could have easily prevented.  
3. Moreover, how much to spend was contentious. The exact amounts were undecided until the 11^th hour. I recall sitting in the Governor’s retiring room in the wee hours of the morning of the last day the legislature could pass bills waiting for Governor Dayton, key legislators, and representatives of the Wilfs to agree on the precise numbers so we could draft the final agreement. (Aside: they likely misjudged how much the Wilfs were willing to pay, since they anted up a lot more than their required team contribution after the deal was done.) The idea that – had the course of future revenues been known – 10% or 25% more would be set aside for the Vikings is not credible. These same elected officials less than a year earlier had shutdown government because they opposed either more spending cuts or tax increases. Setting aside more gambling tax revenues than needed did either or both, since it prevented their use for general fund spending or tax cuts.
4. Given the amounts involved, the reserve was the result of miscalculation and/or inattention. Not a conscious, considered decision, mainly the common mistake of focusing only on the near term. The happenstance of the reserve should not create a de facto legal entitlement. But, of course, codification into law and the difficulty of making changes (always easier to play defense in the legislature) has precisely that effect.

These considerations led me to the natural conclusion that the reserve should simply be repealed. Subsidizing the Vikings stadium beyond the $500 million committed in 2012 should be an issue for the legislature to debate each time it sets the state budget. Arbitrary dedications of revenues (e.g., other than user charges and benefit taxes like the gas tax) go against good budgeting practice, which has all uses compete against one another for scare resources. Dedications can be an effective method for overcoming political opposition to raising taxes but budget rigidity is the price you pay. In this context allowing it to continue seems borderline silly.

Given the surplus, there is no need to do anything with the stadium reserve in the 2023 session (i.e., there is plenty of general fund money without tapping the reserve). In fact, waiting until 2024 or 2025 may have the salutary effect of holding off using its money until the state’s financial future is clearer. I’m still not convinced as to the permanence of the amazing jump in individual income and corporate tax revenues.

Should the bonds be paid off?

None of this answers the question of whether available general fund moneys should be used to pay off the stadium bonds when they become callable. Two considerations are relevant in that regard – financial and budgetary/political considerations.

Financial. From a financial perspective, whether to pay off the bonds (or to refinance them) simply depends on interest rates when the bonds are callable. If the state cannot borrow more cheaply at that point (i.e., if the relevant interest rates are higher than on the stadium bonds), it makes no financial sense to pay off the bonds. Since the state consistently issues new bonds, that would mean that it would be substituting higher interest rate debt for them. Instead, it should use the cash to pay for whatever it would have issued new bonds for and leave the stadium bonds in place. That is the least cost approach. Of course, the future course of interest rates is unknowable. So, that is a judgment that must be made when the bonds become callable. Current municipal bond interest rates would justify paying off or refinancing the bonds. Whether that holds in six months is unknown.

Statements by legislators, quoted in the STRIB article, that of course the bonds should be paid off do not make financial sense. The correct financial answer is that it depends on interest rates. Even that leaves unanswered whether the bonds should be refinanced if interest rates are lower. That’s where political and budgetary policy factors come into play.

Political/budgetary considerations. Pure financial considerations rarely determine state financial decisions, political and budgetary policy decisions more often dictate. Some possibilities in the context of paying off the stadium bonds include:

• Doing so might be politically necessary to repeal the reserve account. That is not true legally, but the political and popular narrative has linked the two. Incurring higher state borrowing costs may be a reasonable price to free up gambling tax revenues for other general fund purposes.
• Similarly, paying off the bonds will free up the Minneapolis sales tax from its legal dedication to pay the bonds. That could allow it to be used for other purposes or even repealed (highly unlikely). So, city interests may favor paying off the bonds. Of course, whether paying off the bonds will free up the city sales tax to be used for other purposes is ultimately a question for the state legislature to resolve and inevitably will involve unrelated issues in the political horse trading that is the legislative process.
• The Vikings likely favor paying off the bonds, since that will help the half billion dollars the public paid for the stadium recede more quickly into faint memory. That will make their inevitable future requests for more public money (e.g., maintenance, improvement, or a new facility) an easier political sell. So long as bonds are outstanding, that is a harder political case to make (“@#$%&!, we’re paying $30 million/year on the bonds, and you want more?”). They’re sure to also argue for continued dedication of some gambling taxes to pay for maintenance and improvements as the price for giving up the current dedication.
• Keeping bonds in place (i.e., refinancing them if interest rates are lower or leaving them in place if not) is consistent with the principle of matching the benefits provided by the facility with the taxpayers who pay for it.
• The stadium is publicly owned because (among other reasons) that is only way tax exempt bonds could be used to finance the public share of its cost under federal tax law. If a long-term goal is to free the public from the burden of owning it (by transferring it to the Vikings, along with the ongoing obligation to maintain it), paying off the bonds could hasten that possibility. A transfer cannot be made until the bonds are paid and likely some additional interval would need to lapse to prevent the public ownership to appear little more than a ruse to qualify for federal tax exempt bonds. That could enable the IRS to challenge the bonds’ tax exemption.

Some combination of these considerations is likely to play a more important role in deciding whether to pay off the bonds than whether it makes sense financially. I have no insight into how the 2023-24 legislature will resolved them, but it seems certain to do so.

Addendum

Opinion columnist and media company behaving opaquely – assertion on stadium financing

12/22/22 update. I did not include this in my original post, but on reflection decided to add it because it continues to irritate me. D.J. Tice, a STRIB columnist and editor, included an erroneous reference to stadium funding in his December 4^th column (Who’s Afraid of the DFL?) on an unrelated topic, the ongoing lawsuit challenging the adequacy of public education for minority children. He falsely attributed “heavy spending” on the stadium to the 2013 DFL-controlled legislature.

My jaw dropped when I first read his column in my print newspaper, attributing heavy stadium funding to the DFL’s total control of the legislature. When I was writing this post, I went to the online version and the sentence I remembered was not there. I assumed that I had falsely remembered. I’m old and my memory is far from what it was, and no correction noted its omission. (There was a correction for a more egregious mistake, attributing an assertion in the case to the plaintiff rather than the defendant.) Going back to the print version, I discovered I wasn’t forgetful. He and the STRIB deleted the mistake without fessing up to doing so. I consider that to be bad journalistic practice. It is not good form to delete mistakes and hope no one notices, at least in my book.

The relevance to the merits of stadium financing or the reserve are marginal, but it is somewhat revealing about partisanship and pro sports subsidies. So, I thought I would catalogue it. (I use my blog as a sort of personal archive.)

Here are the details.

Print version

This is the sentence I remembered from the print version. You must trust me since it is no longer online. (I do have an iPhone picture of the column that I’m too lazy to crop to isolate the sentence and, then, insert as a jpeg.) The italics are mine to highlight the deletion:

Remembering the array of tax hikes on businesses and higher incomes enacted in 2013-14, the last time DFLers ruled unchallenged in St. Paul, along with regulatory expansions and heavy spending on the Vikings stadium, Senate Office Building, and much more, at least some business leaders are bracing for more of the same next year.

D,J. Tice, Who’s Afraid of the DFL, Star Tribune (12/4/22, print version only)

Online version

Remembering the array of tax hikes on business and higher incomes enacted in 2013-14, the last time DFLers ruled unchallenged in St. Paul, along with heavy spending, at least some business leaders are bracing for more of the same next year. 

D.J. Who’s Afraid of the DFL Star Tribune

The online version conveniently wipes out the erroneous stadium example, along with the Senate Office Building and putative regulatory expansion. There is an editor’s note preceding the online column that says nothing about the deletion:

 (Editor’s note: An earlier version of this column referred to a legal brief filed in the Cruz-Guzman case, which asserted that plaintiff’s counsel had admitted certain charter schools were “killing it” academically. That reference did not accurately describe plaintiff counsel’s position and has been removed.)

Id.

My take

The 2013-14 legislature did not authorize or spend any money on the Vikings stadium. What it did do – as my posts outlined – was to put more money into the stadium reserve. That spent no money; it just set it aside, allowing a future legislature or governor (under his or her authority to manage the reserve) to spend it. So far, doing so has not caused any additional general fund money to be spent on the stadium. The only money spent is money that would have been spent under the original 2012 appropriations whether or not there was a reserve account.

Characterizing what the 2013 legislature did on the stadium as “heavy spending” is flatly false. Tice was reaching for examples of spending excess and his biggest example (the Vikings stadium) was enacted by the 2012 GOP legislature. That legislation (signed by a DFL governor) enacted the general fund appropriations that fully funded the stadium.

Two footnote observations about Tice’s column:

1. His two concrete examples of heavy spending would cause me to question his premise, i.e., tagging the DFL for heavy spending. His cited and deleted examples are the stadium and the Senate Office Building. The stadium spending enacted by a legislature totally controlled by the GOP dwarfed DFL funding of the Senate building by 5X. Subsidizing a venue used 90%+ by a commercial for-profit entity owned by billionaires is hardly a traditional government function. A legislative office building, by contrast, is.  (Additional context: The DFL funding of the Senate Office Building was used as a hot campaign issue by the GOP legislative caucuses in the 2014 election, when the DFL lost control of both houses of the legislature. Conventional wisdom is that it helped the GOP flip DFL seats. The stadium was not much of an issue in either the 2012 or 2014 campaign. It would have been consistent with their limited government principles and tight control over state spending for Republicans to make stadium funding a caucus campaign issue. Hard to do, I know, when the legislature you controlled passed it. I assume that the fact that it wasn’t a campaign issue that garnered attention is what caused Tice to mistakenly assume the DFL was responsible for it. At least, that’s best rationale I can come up with to explain such an obvious mistake, along with some inherent bias on his part that being parsimonious with government money is a GOP-monopoly. Their tight-fistedness is selective, obviously. It sure doesn’t extend to tax expenditures which they have been lavish in doling out.)
2. I agree with Tice to the extent he asserts the Minnesota Supreme Court should not use the Minnesota Constitution’s education clause to dictate how much to spend on or how to design education policy for minority children. (Disclosure: when I was still working, I was part of a team of lawyers representing the legislative defendants in the case.) But that would be a problem for either a DFL or GOP legislature and it is a big policy problem that needs to be addressed by nearly all accounts. I just don’t think the court is institutionally well positioned or qualified to do it. I guess a GOP legislature is more likely to create a constitutional crisis/confrontation if the court orders funding.

The failure of the 2022 legislature to enact a tax bill paused a persistent, recent phenomenon: enactment of special laws authorizing city sales taxes. According to this STRIB article (Trey Mewes, Minnesota communities looking for local sales taxes stuck without a special session, 6/11/22), the negotiated 2022 tax bill agreement included 9 new city sales tax authorizations. The bill died when the governor and legislature could not agree on spending and bonding deals. The bill’s authorizations would have been in addition to the 11 new taxes authorized by the 2021 legislature. Cities with newly authorized taxes in the two bills included suburbs with large retail sales tax bases, including Bloomington, Edina, Maple Grove, and Roseville, a first. An earlier Strib story (Kim Hyatt and Shannon Prather, More Twin Cities suburbs see a local sales tax as a funding fix, 2/5/2022) covered the requests by the suburbs. (Edina and Maple Grove’s legislative authorizations passed in 2021. Edina’s is on the ballot for voter approval.)

Before 2019, the legislature had not authorized Twin Cities suburbs to impose general city sales taxes. The 2019 authorizations were for suburbs with modest tax bases (Excelsior, Rogers, and West St. Paul). In any case, the prevalence of city general sales taxes has steadily grown. Fifty-one Minnesota cities now impose sales taxes, a dramatic increase from 20 years ago when only ten cities did. Whatever the partisan composition of the legislature turns out to be after the election, that growth will likely continue. Legislatures controlled by both parties have regularly authorized these taxes; unlike many tax issues, it’s bipartisan. (Warning: that typically means it’s either of little consequence or a bad idea in these polarized times.)

The 2023 legislature should reexamine its practice of using special laws to authorize the taxes. I would allow any city of some minimum size to impose a city sales tax (perhaps subject to voter approval) and use the revenues for any permitted city purpose, while modifying the local government aid (LGA) formula to account for a city’s ability to generate sales tax revenues.

The current ad hoc practice is a historical accident that evolved from the general prohibition in the 1971 Minnesota Miracle legislation. That prohibition reserved general sales taxes to the state general fund, which pays intergovernmental aid to schools, counties, and cities. That avoided problems with the unequal distribution of the local sales tax bases and the political accountability concerns that come from the inherent tax exporting by communities with concentrations of retail facilities, like shopping malls.

Allowing a few city taxes to finance important regional projects on an ad hoc basis was thought to overcome those concerns because legislative review would ensure that the projects had broad regional benefits or spillovers similar to the distribution of the tax burden. The legislature did that for a handful of cities. That concept held for over two decades, but eventually typical legislative and political dynamics took over with authority extended to more and more cities, making the system less tenable. Even if the legislature carefully reviews the funded projects to ensure they provide regional type benefits, doing so would duplicate the similar purpose of the LGA formula. The LGA formula ignores the effects of city sales taxes, compromising the validity of its measures of city need and capacity. Over time, the projects became more mundane (typical city facilities like parks, trails, police and fire stations, etc.) and less uniquely regional.

Policy issues

While sales taxes provide cities with welcome revenue diversity, their growing ubiquity raises process and state intergovernmental aid issues:

• Should the legislature continue approving the taxes on a city-by-city and project-by-project basis, isolating the spending and projects from the city’s regular budgeting processes?
• Should LGA, the state aid program for cities, account for their effects in determining a city’s need for and capacity to provide municipal services? Or should other mechanisms be implemented to correct for the unevenness of the ability of cities to raise sales tax revenues?

These issues have been festering for years, but recent developments suggests that the legislature should explicitly consider them sooner rather than later. The legislature typically rewrites the LGA formula once a decade when new census data allows re-calibrating its need measures, key factors in how the formula allocates state aid among cities. I assume that will happen in the 2023 session, a good time for the legislature to consider issues related to authorizing and equalizing city sales tax revenues. (The 2022 House tax bill rewrite the LGA formula, based on city recommendations. The rewrite was not included in the agreed upon conference committee bill. So, it remains for the 2023 legislature to take up.)

This post provides some background and my casual thoughts on the issue. They are casual thoughts because I neither worked on these issues during my career nor am I an expert on them. But I sat through many legislative hearings on the issues, which caused me to develop opinions. Dangerous, I know.

Background

General prohibition since 1971

In the marathon (5-month) 1971 special session, the legislature enacted a major restructuring of Minnesota state and local financing often referred to as the Minnesota Miracle. It dramatically increased state taxes to provide aid to schools, counties, and cities.  As a tradeoff, it (ch. 31, art. XXI, § 1, subd. 18) prohibited cities and counties from imposing new or increasing existing local sales and income taxes. There were no local income taxes or general sales taxes in 1971, just a few special sales taxes. It was thought to be more equitable and efficient for the state to collect those taxes and distribute some of their revenues as state aid than for cities and counties to impose their own taxes.

Greater reliance on state aid has an obvious downside of reducing local accountability for taxing decisions. Locals will be more willing to spend state moneys since they aren’t responsible for taxing their voters to raise the money. But local sales taxes have political accountability issues of their own. Local sales taxes are often paid by purchasers who do not live in the taxing jurisdictions and, thus, have no say in their imposition and may benefit little from the services they finance. The growth of big box stores and shopping malls in the 1970s and 1980s heightened these concerns.  I don’t know how much those concerns entered the 1971 legislature’s calculations, but they loomed large in the legislative debates on city taxes that I witnessed. The growth of online retail and reversal of the Quill case has lessened those concerns somewhat. When a customer buys online, rather than driving to a shopping mall in another city, the local tax in the city where she lives applies and typically is collected. But many city sales tax revenues are still generated by nonresidents’ purchases.

City-by-city special exemptions grow

The legislature could always circumvent the 1971 prohibition by enacting a special law authorizing a city to impose a tax. That began almost immediately in 1973 with an authorization for Duluth. But the legislature maintained a tight rein for two decades. By 1990 only three cities, Duluth, Minneapolis, and Rochester, had general city sales taxes under special laws. Each was a sort of special case with little commonality of policy rationale, probably reflecting legislators relenting to the reality of both local and internal legislative politics.

But in the 1990s, the legislature began authorizing more city taxes, a pattern that has escalated in this century. These authorizations almost always specify how the revenues may be used (identifying projects) and put time limits on the taxes. Duluth, which is allowed permanently to use its revenues for any purpose, is the exception. In nearly all cases (Minneapolis, St. Paul, and Duluth are exceptions), the voters also must approve the taxes before the city can impose them. Only occasionally do the voters fail to do so.

The chart shows the authorization of taxes imposed by decade. It reflects only the first time a city’s tax was authorized (some have been reauthorized one or more times) and includes only taxes imposed. City councils occasionally decide not to impose the taxes or the voters reject them. The 2021 authorizations are shown in the 2020 bar. These authorizations (other than St. Peter’s tax which was approved in November 2021) are pending voter approval at the 2022 general election. According to MinnPost, the others are not being submitted.

Modest revenue yield

City sales tax revenues remain a relatively small source of revenue, compared with property taxes. Property taxes yield about 20X more city revenues. The chart below shows the growth in revenues over the last 15 years from about $82 million in 2006 to $164 million in 2021, the most recent year for which revenue is available on the DOR website.

Revenues are growing faster than property taxes and that differential is likely to increase if the legislature authorizes suburbs, particularly those with large shopping malls like Bloomington and Roseville, to impose taxes. According to the STRIB article, both cities were authorized to impose taxes in the 2022 bipartisan tax agreement that died at the end of the session. Edina, home to Southdale and much retail, was authorized to impose a tax in 2021. Maple Grove, another 2021 authorization, also has above average retail. I would not be surprised if general sales tax revenues equal more than 10% of city property tax revenues by the end of this decade and fairly quickly exceed $200 million per year.

DOR, Local Sales and Use Taxes (September 2022) lists the current local sales taxes that DOR administers. That includes all the general city sales taxes. Cities also impose a variety of special local taxes, such as lodging and food and beverage taxes, some of which the cities administer themselves and some of which DOR does. This post focuses only on general city sales taxes.

Revenues used for capital projects

As noted above, the special laws authorizing city sales tax routinely specify the purposes or projects for which the tax revenues must be used. Duluth is allowed to spend the proceeds of its tax for general purposes and similar treatment was provided to Minneapolis for a very short period, while all other cities’ tax revenues must be used for projects identified in the special laws that authorize the taxes. The nature of these projects varies widely. They range from the Vikings Stadium (Minneapolis) to more mundane projects elsewhere. Common uses are community and convention centers, parks, and trails, wastewater treatment facilities, and roads. This House Research publication lists the uses for the taxes authorized and imposed through 2019. The taxes authorized in 2021 largely financed community centers and recreational facilities.

The list below (compiled from the House Research publication) shows the types of projects that are permitted uses of four or more of the city taxes that are now imposed. Some taxes fund multiple types of projects.

• Park and recreational facilities (15)
• Sewer and wastewater facilities (10)
• Streets, roads, and other transportation (10)
• Civic, convention, and community centers (9)
• Trails, sidewalks, bikeways etc. (8)
• Water treatment and supply (6)
• Libraries (5)
• Police and fire facilities (4)

The enacting laws typically only contain cryptic descriptions of the projects. However, even those descriptions make it easy to conclude that many other cities fund similar projects with standard revenues, such as property taxes and fees, and that the sales tax revenues are partial substitutes for those revenues in many cities with sales taxes. Thus, these are not uniquely different projects; many cities finance similar projects with property taxes and other revenues. Virtually all cities provide sewer, water, police, and fire services and most provide parks or recreational facilities.

Legislature requires regional benefit

In general, the legislature has focused on whether proposed projects have regional benefits to address political accountability concerns. The apparent thinking is that it’s okay to expect nonresident purchasers to pay if substantial benefits of the projects also go to nonresidents. Thus, the legislature has typically focused on whether the funded projects are regional, and cities have pitched their proposals that way, suggesting significant benefits go to nonresidents or “spillover” outside the city. The comments in the February Strib article by the Gary Carlson, a city lobbyist, illustrate that.

The reality is, of course, more complicated and the existence of regional benefits may be more of a rationalization than a compelling justification. The political argument for the taxes is clear – sales taxes are more popular (less unpopular) than property taxes, the alternative. And exporting tax burdens (“tax that guy behind the tree” as the old saying goes) is always politically popular. Consider:

• The class of nonresidents taxpayers and beneficiaries of the projects may not overlap very much. Nonresident sales taxpayers may not use or benefit from the regional projects, while other nonresidents do and may make few taxable purchases.
• In some cases (e.g., wastewater and water treatment facilities and some recreation facilities), it would be easy to closely align payer and beneficiaries through user charges, forgoing sales taxes. That is the textbook way to finance projects. Even though user financing may be practical (i.e., users can be identified and charged), it may not be politically feasible. The users often cannot afford or are unwilling to pay. Thus, cities wish to export the burden to a broader group that includes nonresidents and/or make the charges less visible than monthly fees or explicit user charges.
• Some of the authorized projects appear to impose costs on nonresidents, rather than to confer benefits. Take flood control, an authorized use of two taxes, which inevitably diverts water, potentially flooding someone else’s property, perhaps, outside the city (e.g., upstream). Similarly, financed facilities may divert commercial activity from other cities to the city in which the project is located.
• Projects vary substantially in how extensive or widely distributed their regional benefits are. Some finance basic city facilities like police and fire stations.
• There are spillover or regional benefits of basic city services, such as transportation and public safety expenditures. Nonresident workers, shoppers, and others directly benefit from city services. Moreover, the indirect burden of property taxes affects nonresidents as well. Adjusting for these effects is a key purpose of the LGA formula. But the current LGA formula ignores city sales taxes (more or less) – assuming cities with taxes do not have them and the once-a-decade development of the formula’s need measure may indirectly capture some effects of cities’ spending of sales tax revenues.

Legislative Issues

I see two interrelated issues. First, should these taxes continue to be creatures of special law, requiring the legislature to separately approve each tax and the projects it funds? Second, should the LGA formula be modified to account for the taxes or should some other general mechanism for revenue sharing be designed? I list a few of my thoughts on each issue.

Should the legislature approve each tax and project?

Cities (and other local governments) are creatures of the state and state law. Most presume that they should be treated equally, i.e., given similar legal authority, perhaps, adjusting for common differences, such as population as reflected in the different classes of cities. That pattern is typically followed. The reason that treatment of local sales taxes deviates is largely an accident of the Minnesota Miracle’s prohibition and an ever-growing list of ad hoc, special law exemptions. That practice could be justified or ignored when the exemptions applied to only a few projects or cities. They no longer do.

The obvious question is whether some policy problem or problems are best addressed by the legislature reviewing and deciding on each tax and project or if it is better to generally allow all or a set of cities to impose the taxes. If one concludes individual review is needed, a second question is whether the legislature has been doing that and/or is institutionally positioned to do it well?

Local sales taxes present two main public policy problems:

1. Uneven revenue raising ability. The ability to raise sales tax revenues varies widely among cities. Retail sales, the tax base, can be highly concentrated. Relative to their populations or other measures of city spending need, some cities have large potential tax bases while others have little. Two prototypical examples of high tax base cities are suburbs that host regional shopping malls and small Greater Minnesota cities with big box stores. Some cities have virtually no retail. Sometimes that is a city zoning choice, but often just an artifact of the local market for retail locations. The uneven distribution is also true, but to a lesser extent, for property taxes, the main own-source city tax. But LGA addresses and mitigates that unevenness.
2. Political accountability. Local sales taxes raise accountability concerns. Cities with retail heavily patronized by nonresidents can export a share of their tax burden to nonresidents who cannot vote and may benefit little from the funded services. Also, funded projects are likely considered separately from regular budgeting for capital facilities that rely on user fees and property taxes, undercutting the budget process. That might be justified if the projects were particularly regional or unique; they rarely are any more.

My thoughts on whether the legislature or a set of general law rules can best address these two issues follow along with thoughts on the institutional capabilities of the legislature for addressing them at the end of the section.

Uneven distribution of tax base

• The legislature has not addressed this problem in considering city requests to impose sales taxes with one exception. For Rochester’s 2011 reauthorization, the legislature required the city to share $5 million in revenues with surrounding communities. That occurred largely (in my opinion) because the chair of the House Taxes Committee represented some of the communities and used his position and power to protect them or extract concessions on their behalf. (He assigned a jocular eponymous title to the measure: “Greg Davids’ Good Neighborhood Program” revealing its self-promotional essence and casting himself as a sort of Santa Claus to his constituents’ cities.) More generally, when city requests were evaluated during my tenure as a legislative staffer, I do not recall data being presented on the requesting city’s relative sales tax base – i.e., how it compared to other Minnesota cities (average, above, or below). Bottom line: to the extent this is a policy problem (I think it is a real problem), the existing ad hoc practice has not addressed it. Aside from that failure, the legislature is poorly situated it to address it. My thoughts on that are below under institutional considerations.
• A general solution is more likely to work. That could be done by integrating local sales taxes into the LGA formula or by setting up a separate formula. For example, a separate formula could rely on “power equalization” (i.e., the state keeps some of the revenues from very high tax base cities and redistributes it to low tax base cities, so revenues raised are more equal for all cities on a rate-adjusted basis. That would ease the state aid cost and might deter high tax base cities from imposing the tax mainly to export their tax burdens (looking at you Bloomington, Roseville, and Edina). For reasons of simplicity, I think integrating local sales taxes in the LGA formula is the best approach, but I’m not an expert on this.
• A first step is to gather data measuring the relative distribution of sales tax bases across cities. That would help assess whether a problem exists and how extensive it is. That should be easy for either DOR or legislative staff to do. To my knowledge, it has not been done or at least made publicly available.

Political accountability

• The legislature has focused its attention on this problem and has tried to address it by limiting the taxes to regional projects with benefits outside of the city. How well it has done that can be questioned by simply looking at the wide collection of projects.
• Sales taxes are the politically preferred tax by most accounts, especially compared to the main city tax source, property taxes. This attractiveness is likely a big factor in cities’ requests, as much as or more than their reluctance to pay for projects whose benefits spillover outside of the city’s borders.
• Limiting usage to regional projects has drawbacks. Yes, it does distribute benefits of the taxes more widely. But by treating these projects as special cases outside the regular budget process, city councils will not weigh their relative benefits against other uses for the money. More succinctly, limiting the revenue to regional projects goes against textbook budgeting principles under which all uses compete for scarce resources. Because the sales tax is probably the least unpopular tax, dedicating its revenues to a subset of projects inevitably favors those projects.
• Scanning the funded projects suggests that recreational facilities (community centers, parks, trails, and similar) are a prime use. My jaundiced view is that they may be good uses, but I would guess that they are typically regarded more as frills or less important than core city functions like public safety or streets. If the benefit is truly regional, another approach would be to make them county responsibilities funded with county revenues (perhaps, county sales taxes). That might better align decisionmakers and payers. I suspect, however, that the prime benefits are often concentrated in the cities and the city business community and other key constituencies in the city are behind them. But that may just be my inherent skepticism.

Institutional/decisional considerations

Institutional factors bear on whether the legislature itself, on a case-by-case basis, should be making these decisions. Alternatives are (1) general authority for all or a defined group of cities to impose the taxes or (2) delegating case-by-case authority to an official or public body (e.g., an appointed board or commission). The latter option is rarely used for local government authority in Minnesota and so I do not discuss it. A couple of exceptions are the Met Council for certain land use authority and administrative law judges for annexation decisions. I think the basic choice is between the current model and general law authority.

Advantages of the current model:

• It allows the legislature to evaluate the unique circumstances that each city and project presents and to tailor limits to that situation. General rules that are applied rigidly, a necessity for a statewide law, inevitably are over and under inclusive. Some projects/taxes will be allowed and others disallowed, contrary to how an all-wise Philosopher King would act. But the legislature is not a Philosopher King.
• Legislative decisions are made publicly with an opportunity for those affected who live outside the city to make their arguments to the legislature. (They can do the same to a city council, but they are unrepresented while in-city proponents and opponents are.) I rarely saw that done while I was a legislative staffer. Hearings were mostly dog-and-pony shows put on by the city and local boosters of the projects. Occasionally, interests from outside of the city did testify but not often.

Disadvantages of the current model:

• Special laws – laws that apply only to one entity or situation – are presumptively disfavored. The constitution prohibits them in most situations (laws related to local governments are an exception, which is why the current system is legal). Minn. Const. art. XII. The prohibition dates to a 19^th century constitutional amendment. I’m not sure about the exact rational beyond favoring equal treatment over potential political favoritism. But institutional factors tend to cause legislative bodies to make poor decisions in resolving specific cases. Legislative politics – both partisan and personal – can get in the way of good decisions. In an inherently political body like the legislature, that is a fact of life. Some tax chairs and committees work assiduously to be even handed and fair, others less so. Inevitably irrational factors will occasionally pollute the decision making – e.g., a desire to reward friends or favored colleagues, to help one’s own party or to look bipartisan, to approve projects of a favored type or types, to get another member’s support for an unrelated proposal, and so on.
• The process is expensive. Cities essentially must mount lobbying efforts and navigate the complicated, bi-cameral legislative process. Some pay outside lobbyists. Even if they don’t, it can require multiple trips to St. Paul and many communications with legislators.
• Along with higher costs, the process creates uncertainty, making planning more difficult.
• It’s slow, delaying projects.
• As noted, it typical isolates the projects from the competition provided by the local budget processes.
• The legislature, because of its decision-making dynamics, is poorly situated to resolve issues when good policy suggests redistributing tax revenues is appropriate. For a city with a much higher than average tax base, a natural solution is to use some of their revenues to supplement the revenues of those with below average tax bases. I cannot imagine the legislature doing that on a case-by-case base as cities request special law authority. It would be distasteful, and it is unclear what to do with the money (other than share it with surrounding cities as in the Rochester case, a second-best solution since they may not deserve it under objective criteria). For below average tax base cities, supplementing their revenues creates a budget cost and makes it harder to enact their requests and makes them look like supplicants, rather than entities willing to pay their own way. A general power-equalizing tax or aid system that applies to all cities is more plausible and fairer, if also unlikely.

My take

It is time to enact a general law allowing any city (with a minimum population requirement of some size) to impose a general sales tax whose revenues can be used for any city purpose. In a perfect world, I would limit general sales taxes to counties, jurisdictions with broader geographic reaches that provide fewer discretionary services (health and human services mandated by state law are a prime county service) than cities. Limiting cities’ authority to property taxes makes sense, because city services are much more property-related than those provided by counties. But that ship has sailed and politically cannot be called back to port. Given that, treating sales taxes as a general revenue source will lead to better city budgeting. If there truly is a concern that city projects with regional benefits will be underfunded, ad hoc authorizing of city sales taxes is probably not the way to address that problem.

LGA formula treatment

Block grant aid’s purpose

Block grant state aid, like LGA, is intended to equalize cities’ ability to deliver municipal services while affecting as little as possible local service decisions (i.e., which and how much service to provide). A prime goal is to put cities and their residents on more equal footing without disturbing local decision making. LGA is often promoted as a way to deliver property tax relief. That really is. at best, a side effect of its core purpose. A more elaborate statement of the policy problem LGA addresses will provide context.

Point 1:  A prime purpose of autonomous or semi-autonomous local governments is to permit a diversity of local choices and preferences to flourish. That is a way to better satisfy people’s preferences/tastes and our democratic values. Some cities (i.e., voters and taxpayers) may prefer a package of basic services, while others want more extensive and/or higher quality services. Democratically controlled autonomous local governments help to efficiently satisfy those preferences. Local voters decide on the services they want and people can opt to live in cities whose packages of services and taxes align with their preferences. Economists refer to this ability to pick residential locations based on preferences for public services and tax levels as the Teibout hypothesis (named after Charles Teibout, the economist who first published a conceptual articulation of it).

But municipal services are not exactly private goods and services where society is indifferent about people’s ability to pay for or buy them. We accept that most people cannot buy luxury goods as a fact of life to take an extreme example. It’s okay that most people cannot afford to buy a Mercedes; we’re fine with them buying a Ford or taking public transit. But for public goods like municipal services, we expect that everyone, even the poorest, should have some basic level of service at an affordable price. That brings us to the second consideration.

Point 2: Econ 101 principles say that higher prices reduce demand; for cities price is determined by the tax rate needed to provide services. The classic supply and demand curve shows that higher prices mean few goods or services are purchased. For public goods, like municipal services where taxpayers/voters are making decisions, that price is determined by the tax level or rate needed to finance the services. That price or tax rate varies across cities for a variety of reasons including:

1. Ability to raise revenues. Some cities have large tax bases, others less so. A below average tax base means taxpayers must pay higher tax rates (prices) for the same services than those in a city with an average or above average tax base. Some cities are able to export their tax burden to nonresident/nonvoters. How that is done is obvious for sales taxes but commercial-industrial property taxes are another common example.
2. Spillover effects of services. Some cities, such as urban centers with large numbers of daily visitors or extensive tax-exempt properties that consume services, provide more services to nonresidents and other non-taxpayers. When above average amounts of services go to nonresidents or exempt organizations, all else equal, taxpayers will pay higher tax prices for similar services than in a city with fewer free rider beneficiaries.
3. Cost to provide services. It is typically more expensive to provide services in areas with more dense or poorer populations. Fighting fires in high-rise buildings is more expensive, for example, than in low-rise, less dense neighborhoods.

A little thought reveals that these factors can distort municipal decision making. Higher or lower tax prices affect voters’ willingness to pay and preferences for services. Some of that is due to ability to pay (the local tax base), but spillovers and higher costs are factors too.  Block grant aid programs, like LGA, seek to minimize those distortions by giving aid to cities to mitigate the effects of the three factors and to put cities on somewhat more equal footing in making those decisions. By contrast, matching grant style aid is explicitly intended to change preferences or spending decisions. It assumes that the recipients otherwise would not spend enough on the targeted services.

Given that generalized description of LGA’s purpose, the challenge is how to write a formula that satisfies the somewhat contradictory purposes. LGA attempts to do that by measuring a city’s (1) need and (2) its capacity (tax base) to meet that need, yielding a need-capacity gap for some cities. Then, the formula allocates the legislative appropriation of aid among cities based on their relative scores – how large their relative gap between need and capacity is compared to that of other cities. The challenge is to measure the two factors.

Capacity

Measuring capacity or tax base is somewhat straightforward. The property tax is the one general tax that all cities can use to pay for services. Thus, it has been thought that capacity is simply a matter of determining how much property tax base the city has, adjusted to neutralize variations in assessment practices. Because that assumption is so widely accepted, it hasn’t changed over the years. (In the early 1990s, outside consultants advocated using a different measure, the income of city residents. As a theoretical matter, residents’ income is a better measure because it eliminates the distortions created by variations in business property in city property tax bases, a way taxes are exported to nonresidents and focuses more on residents’ ability to pay. The legislature did not agree for a variety of reasons, mainly practical reality. That is as close as anyone has come to seriously suggesting changes in the measure of capacity.)

Need

Measuring need is more controversial. The obvious goals are to be objective and to avoid measures that a city controls, so its behavior does not affect how much it gets. It would be simple to use population, the more people, the more need. But that would ignore a myriad of factors, such as density, poverty, age of infrastructure, geographic size, and similar, which may affect how costly it is to provide services, objective need for certain types of services, the extent to which benefits spillover outside the city are present, and similar potentially relevant factors.

For the last 30+ years, the legislature has gone through a periodic exercise of identifying factors outside of the control of cities that statistically explain (using regression analysis) varying levels of city spending to measure need. To be considered, a factor must pass some theoretical test of relevance to city spending/need and other factors must be controlled for (statistically held constant). The hope, of course, is to avoid measuring spurious correlation. Using spending by a large group of cities and various statistical controls, it is hoped will average out or mask individual city’s preference for services. But there are obvious limits and there can be no purely objective measure of need. In other words, an element of art, as well as statistical science, goes into making these determinations, and the distributions must pass a political acceptability test, since the legislature is an elected body and must enact the formulas into law. The identified need measures are assigned weights and used to create factors that, in combination with the capacity measure (i.e., adjusted property tax base), distribute aid among cities.

Local sales tax revenues and LGA

So, how does the ad hoc authorization of city sales taxes affect the LGA formula amounts and how should it? Potential distortions can arise in (at least) two ways:

• Sales taxes do not count in measuring capacity; the formula’s capacity measure is a city’s adjusted property tax base. The theory is the sales tax revenue is dedicated to a regional project and is unavailable to pay for basic city services, such as public safety, administration, or streets. That, of course, is true but it is also true that other cities fund similar projects (e.g., water treatment, parks, public safety buildings, etc.) with general fund revenues. When sales taxes were limited and/or funded a few unique projects (Minneapolis convention center, Rochester flood control), that was not a major flaw. As sales taxes become more common and projects more ordinary, it is harder to justify.
• Sales tax-funded spending could distort need measures.  This is a much smaller issue because when the last need analysis was done sales tax revenues were small. (I don’t know but some testing of the potential effects might have been done.) Recall that the statistical analysis is attempting to find background factors (demographics, physical conditions, etc.) that explain city spending and, thus, measure the need for city services. But in conducting that analysis, it is also necessary to control for tastes or preferences for services; tax base can be a factor in that regard. Cities with larger tax bases would be expected to spend more, unrelated to objective need because the tax price is lower. So, this needs to be controlled for. For those cities with taxes, sales tax revenues and a city’s relative sales base would muddy that analysis. It’s easy to ignore that when only a few cities have special sales tax authority and revenues are low. Much less so when they are more common, and revenues are higher. So, it probably wasn’t a problem, but we cannot be as sure as the taxes become more common. It makes the statistical analysis more difficult and the results less reliable. Probably not a big deal, but certainly something to be avoided or minimized in the best of all worlds.

My take

As city sales taxes become more and more common and pay for standard infrastructure, such as sewer, water, roads, and parks, it is difficult to justify not incorporating them in the LGA formula. It means that the formula’s measure of a city’s capacity or tax base is less accurate, and it complicates the task of developing good measures of city need to use in the formula. A likely result is that the formula allocates too much aid to cities with taxes, especially those with above average sales tax bases and those funding more ordinary projects.

This is what one would expect when two separate and uncoordinated mechanisms are used to do the same task. In this case, the task is compensating cities for the fact that the fiscal system requires them to provide spillover benefits to nontaxpayers and the two mechanisms are the LGA formula and the ad hoc legislative authorization of special sales taxes for purported regional projects. The net result will almost inevitably be overlapping and inconsistent results.

How to integrate city sales taxes into the formula is not a simple or straightforward matter. It would be especially problematic if sales tax authority continues to be approved ad hoc matter by special legislation. Thus, I think the 2023 legislature should carefully consider incorporating the effects of city sales taxes into the LGA formula. That would be much easier and straightforward to do that with a general law authorization for cities to impose sales taxes for any city purpose (just like property taxes) – subject to some limits, e.g., so micro cities with big box stores cannot impose the tax or similar. The time to do this is when new calibrations of need are estimated using the 2020 census data. Transition rules will be needed so that cities with special law taxes for dedicated projects can only use the authority when their special law authority expires and imposing those taxes will not affect their LGA (at least fully).

I am under no illusion that this will happen for two reasons:

• The LGA lobby, city interests generally but especially the Coalition of Greater Minnesota Cities, will oppose it. That is so, because it is likely to help make the case that a smaller pool of or appropriation for LGA is sufficient. Second, it will tend to reduce relative LGA for cities that disproportionately benefit from city sales taxes and now get generous amounts of aid. The classic cases are regional centers, the Coalition of Greater Minnesota Cities prime members.
• Despite protestations to the contrary, many legislators like to pass special legislation for specific cities and counties. For those in control (committee chairs), it gives them a little extra power and leverage. For average members, it helps make them look responsive to local community requests and because referendum approval is typically required, it does not risk (e.g., for GOP members) alienating anyone by supporting tax increases or specific projects. More fundamentally, it maximizes legislative control.

Postscript – LGA pet peeves

Many years of listening to the LGA legislative debates caused me to develop a few pet peeves about the legislature’s consideration of LGA. Two stand out:

Selling LGA increases as property tax cuts

It is no surprise that LGA proponents do this. The reasons are political and obvious:

• It appeals to Republicans. Recall that the GOP’s three fiscal priorities in order are (1) tax cuts, (2) more tax cuts, and (3) even more tax cuts. As a result, to gain their support it is useful, if possible, to cast your proposal as a tax cut.
• Democrats’ DNA, going back to the Minnesota Miracle, favors property tax cuts over any other type. (The Minnesota Miracle was a very big state tax increase to fund expansion of education funding and other stuff. But it was the then property tax revolt that spurred it, along with the need to expand education funding when the baby boom went to school.) For many years, the central DFL mantra was to control and cut property taxes. Naturally, a proponent of a program would, if possible, cast it as a property tax cut.
• The general populace likes tax cuts. It’s human nature to want to purchase anything, including public services, at a discount.

If I am correct about the purposes of LGA, characterizing it as a tax cut is misleading. Any tax cut is a side effect of its core purpose, and a good aid design minimizes that effect as much as possible. We want aid to equalize the ability to provide municipal services, not cut taxes. If the goal is to cut property taxes, there are better ways to do that. The fundamental challenge with the state cutting property taxes is that the property tax is not a state tax (other than the general or state C/I tax). Local governments impose it. So, the only way to do so is either to induce or require local governments to reduce their levies (e.g., by providing state aid) or to provide state refunds or credits directly to taxpayers (e.g., through the property tax refund, a homestead credit or etc.) and even that is not fool proof since the locals can tax it away. Paying aid to local governments is indirect and inefficient. Some/much of the aid will go for local spending. (That is why public employee unions lobby for LGA.) State paid refunds or credits are more cost effective and can be better targeted, even if they are still somewhat indirect.

Using LGA to manage property tax burdens

A corollary of the DFL focus on controlling and cutting property taxes led for many years (the 1970s through the 1990s) to legislative efforts to manage property tax burdens. Each session estimates of the likely increases in property taxes by property type and region were done. If those estimates showed undesirable increases, efforts would be undertaken to mitigate the effects using various policy levers, including state aid to cities, counties, and schools. Other levers included the classification system, state credits, and levy limits. Because LGA, as compared with school and county aid, targets smaller areas and groups of taxpayers, it became the preferred type of state aid to manage or mitigate property taxes generally (for all property types in an area). Again, that is a bastardization of LGA’s real purpose and it increased the LGA appropriation unnecessarily from an objective perspective. Thankfully, this process largely ended with the 2001 property tax restructuring, but not totally.

Providing more confirmation of the proliferation of SCOTUS amicus briefs, the Onion (yes, America’s Finest News Source) filed one in a Novak v. City of Parma. It will likely be the most widely read brief of the term (certainly for a cert petition) if Twitter is any guide and compete for having the most social value. As the Readers Digest used to say (maybe still does), “Laughter is the Best Medicine” and certainly beats ponderous boredom (i.e, repeating arguments already made to please whoever is paying or to pad the author’s resume or cv), the stock-in-trade of too many amicus briefs.

It’s worth reading if you, like me, like parody. For example, I had missed that the Onion had broke the Mar-a-Lago nuclear secrets story back in 2017 until I read the brief (fn 2).

Opinion below for anyone who wants to see what can happen when you parody people in power without a sense of humor or proportion.

As persuasive as the brief may sound read alone, I understand cops’ qualified immunity is a tough nut to crack. With the current SCOTUS, making free exercise of religion First Amendment claims (not free speech) might have a better chance. For the Onion, parody as religion doesn’t seem too much of a stretch.

Back in 2020, many of us expected the pandemic to decimate state revenues. (Okay, I can only speak for myself but there were forecasts, including official ones. Remember MMB’s May 2020 interim budget projection? Most other states reached similar conclusions.) Of course, we were wrong. Things turned out not to be that bad. After a short dip, revenues recovered and the feds delivered several generous servings of aid (some of which undoubtedly substituted for stuff that otherwise would have been paid with own-source revenues and others remain unspent). Now, the state has fully restocked reserves and a $9 billion surplus.

Old story. But over at EconoFact, Jeffery Clemens has a nice piece, The Resilience of State and Local Government Budgets in the Pandemic (3/29/2022), that summarizes what happened and why from a national perspective. He hits all the relevant points I am aware of. Here’s his conclusion:

The nature of the pandemic recession introduced unprecedented uncertainty into the process of estimating both revenue inflows and expenditure needs for states and local governments. The federal government provided state and local governments with unprecedented flows of aid that, in the end, substantially exceeded states’ needs. This dynamic helps explain why, despite initial fears of fiscal distress, states have more recently contemplated issuing fourth stimulus checks, reductions in tax rates, and gas tax holidays. The overshooting of federal fiscal assistance may reflect a hazard of relying primarily on ad hoc relief packages in response to crises. Had the American Rescue Plan’s assistance package reflected the resilience of household incomes or the unemployment rate’s rapid improvement, total federal aid would have overshot state and local needs far less. This points to a potential benefit of tying aid to economic conditions through formulas, ideally legislated in advance of a given crisis, so that aid adapts automatically to conditions on the ground. An automatic stabilizer of this sort would have the benefit of easing the strains on Congress’s bandwidth as it responds to other pressing dimensions of complex crises, like the need to erect testing infrastructure and spur vaccine production in the context of pandemics. Automatic stabilizers have the further benefit that they are responsive to changes in need even in the face of Congressional inaction.

Jeffery Clemens, The Resilience of State and Local Government Budgets in the Pandemic (3/29/2022)

His point about the benefits of formulas and automatic stabilizers resonates with me (as pie in the sky that it is with the Congress we actually have), since I’m in the middle of reading Noise A Flaw in Human Judgment. (I highly recommend it. Not as good as Thinking Fast and Slow or Nudge, but good.) It makes a strong case for rules-based decision making, Ad hoc decision making is simply a recipe for bad decisions, more often than not, even without the handicap of political gridlock/partisanship/polarization/etc.

More importantly, reading his full account should cause one to pause before thinking you know what will happen to state revenues when the hot economy simmers back down and Uncle Sam’s birthday money has been spent. Chances are good that spending demands and revenue flows will not be budget balancers’ friends.

What the GOP thinks about the wisdom of one-time rebates or refunds of state surpluses appears to depend upon the state you’re in.

Minnesota

Governor Walz proposes to use $1.5 billion of Minnesota’s over $9 billion surplus for a one-time rebate (maximum of $500 single and $1,000 married joint). The reaction from the GOP:

Republicans say the Walz plan is a gimmick. They argue the state should cut taxes permanently by lowering income tax rates and eliminating any tax on Social Security.

MPR, Walz pitches big one-time checks (3/17/22)

Georgia

Georgia is a much bigger state with a smaller surplus ($3.7 billion but that is a year-end closing amount, not a forecast). The governor is Republican and both houses of the legislature are controlled by Republicans. Governor Kemp proposed giving $1.1 billion of the surplus back as one-time income tax refunds, which the legislature just passed. The maximum refunds are one-half what Walz proposes ($250/$500), but only will go to income taxpayers (Walz’s would go to everyone on the notion that everyone pays sales tax). According to the Atlanta Constitution:

Sen. Clint Dixon, R-Buford, one of the governor’s floor leaders, told colleagues Wednesday that the refund backed up Kemp’s belief that “when government takes in more money than it needs, the surplus funds should go back to taxpayers. They know best how to spend their hard-earned money.”

Georgia Senate give final passage to $1.1 billion income tax refund (3/16/22)

Why the difference?

To be fair to Minnesota Republicans, I am sure they would contend that Minnesota’s bigger state government and surplus relative to Georgia’s justifies turning up their noses at one-time rebates in favor of permanent cuts. But I would speculate that difference is determined more by the two parties’ positions in their respective states.

The Georgia GOP is in power, expects to be in power after the 2022 election, knows it will have to put a state budget together for 2023, isn’t sure that the revenue surplus is a permanent deal, and does not relish the idea of starting the budget process underwater. Under the circumstances, a gimmick in the eyes of Minnesota Republican is fine policy and consistent with Georgia Republicans’ philosophy.

By contrast, Minnesota Republicans expect to control one or both houses of the legislature and hope to control the governorship after the 2022 election, but the latter is at best an aspiration or hope. Moreover, their position on rebates is a legislative session negotiating position. As result, treating rebates as gimmicks is some combination of a negotiation position and a campaign slogan. If they are lucky to run the table in November 2022 election, they’ll worry about governing then.

Put more starkly, considering rebates to be gimmicks is a negotiating posture and/or campaign fodder. Rebates under the circumstances, as evidenced by actions of the Georgia GOP, are consistent with prudent Republican-oriented governance. That Republicans in both states would limit their tax cuts to income taxpayers (cutting out more than a third of taxpayers), while Walz would extend the largess to everyone is more reflective of a real difference in partisan philosophy.

Jane Mayer, Is Ginni Thomas a Threat to the Supreme Court? New Yorker (January 21, 2022).

Supreme Court justices are subject to the thinnest of formal judicial ethics rules. It is almost accurate to characterize them as being subject to something like an honor code with only self-reported violations possible. (Abe Fortas might disagree. Of course, his problems, which were serious and legitimate issues, likely would not have surfaced without LBJ submitting his appointment to be chief justice. See Gerard N. Magliocca, The Legacy of Chief Justice Fortas for an account of how transformational these events were for the Court. He was the last justice to have an ongoing close relationship with a sitting president – there were others. William O. Douglas, for example, regularly played poker with FDR and served as a de facto advisor; Roosevelt considered him as a potential running mate. See this WaPo retropolis article (posted after I originally wrote this) for more details about Fortas’s resignation and his wife’s role in that decision.)

This article, typical for Jane Mayer, goes into extensive detail about the activities and positions of the wife of Justice Clarence Thomas, Ginni Thomas, and how they create (at a minimum) appearance problems for him. As Mayer puts it, Ginni Thomas “has held so many leadership or advisory positions at conservative pressure groups that it’s hard to keep track of them. And many, if not all, of these groups have been involved in cases that have come before her husband.” In addition, Mayer documents activities of her lobbying business which regularly has dealings with individuals and organizations with business before the Court. Trump reportedly considered her a “wacko,” which says something (not sure what). She associates with some of the sketchiest characters on the political right (at least in my opinion – Steve Bannon, Sebastian Gorka, and less high-profile individuals of similar ilk). The article is interesting reading if you’re into those sorts of things.

Some of my off-the-cuff observations after reading Mayer’s article:

• Given the consistency and predictability of Thomas’s positions, it’s hard to imagine his wife’s involvement with radical right wing groups and individuals has much impact on how he votes, writes opinions, or judges generally. The cases they are involved in (based on the article) appear to be the typical issues the right is pushing before the Court (guns, abortion, religion, etc.). One assumes that they’ve got Thomas’s vote and if they haven’t, their positions are hopeless. The graph below from 538’s story on Breyer’s retirement shows just how consistently Thomas is the most conservative justice.
• One data point confirming that observation is the recent action of the Court dispatching Trump’s executive privilege claims related to the events of January 6^th. Justice Thomas was the only dissenter. (Mayer documents Ginni Thomas’s extensive involvement with the rally. The Court’s decision was filed after her article was published. Quixotic dissents like this seem much more likely to be motivated by principle than conflicts, although maybe Justice Thomas just wanted to keep the peace at home? Seems implausible.)
• Judicial ethics in the Supreme Court inevitably apply in a different context than for a typical court. The Court’s docket is almost totally discretionary with the Court selecting cases largely to write law, annotate the Constitution, clarify important statutes, and so forth (acting more like a legislature or constitutional convention than a court in high-profile cases). Technically, of course, the Court is deciding cases or disputes between parties, but for the handful of high profile, “political” cases each term it’s really doing something different. (I think it is largely these cases that readers of the New Yorker are interested in and that Mayer is addressing.) The parties and their lawyers are necessary bit players, but whether some involved lawyer (especially on one of the many of the mostly meaningless amicus briefs) is a friend or associate of one’s wife seems less than consequential or concerning. Note, that is not true for the Court’s more mundane docket, including the decision as to which cases to take (e.g., tax cases and many others that resolve circuit splits, statutory construction issue without political overtones, and so on). There it is a big deal and application of standard judicial ethics is appropriate. The Court should be subject to some higher authority but constitutionally that’s not realistic, thanks to the prevailing imperial view of the Court and the limited reach of impeachment. Major political events, like the Fortas nomination, are what moves the dial.
• None of this excuses Ginni Thomas’s brazen behavior, which is unbecoming the spouse of a justice and inappropriate in my opinion, and the ugly appearances it creates. She should have used Marty Ginsberg or John O’Connor as her model for the lawyer-spouse of a justice, instead of staining her husband’s already shaky reputation (curtesy of the events surrounding his confirmation). The fact that Justice Thomas does not appear to be troubled by her actions is itself troubling.

Personal postscript

Reading Mayer’s article reminded me of my brief and inconsequential brush with Supreme Court ethics. In 2007 when Davis v. Kentucky was before the Court, I wrote an article about the case for State Tax Notes. The case involved the constitutionality of a state exempting the interest on in-state municipal bonds from tax, while taxing out-of-state bond interest. Kentucky, like most states, followed that pattern. The plaintiff taxpayers, holders of non-Kentucky bonds, contended this violated the Commerce Clause, a plausible argument since the Court had long held taxes that discriminated against out-of-state businesses and transactions per se violated the Commerce Clause. Had it applied that rule to municipal bonds, it would have compelled most states to change how they taxed municipal bond interest, requiring them to exempt or tax it uniformly.

It occurred to me that because some of the justices likely invested in municipal bonds, the decision could affect their personal tax situations. Both Maryland and Virginia had discriminatory taxes while the District of Columbia did not, but I knew from media reports that not all the justices lived in the district. The Internet made it easy to check financial disclosures. Sure enough, six justices owned municipal bonds, some from multiple states and some from one state only. It was obvious some of them could be affected adversely or positively by a decision for the plaintiff taxpayers. (The ambiguity resulted from not knowing exactly how state legislatures in Maryland and Virginia would respond to an adverse court decision.) So, I pointed out this odd effect at the end of the article. Because the effect was indirect, depending upon intervening actions by state legislatures in response to a decision, it did not raise a true judicial ethics issue.

The Court ruled in favor of Kentucky, upholding its discriminatory tax treatment. This was not unexpected because the Court in the previous term had signaled it was going down a path of carving out some sort of special Commerce Clause treatment for certain state actions and enterprises.

As expected, the ethical issue I raised generated no apparent response. Out of curiosity, I later checked the justices’ financial disclosures for 2008. Chief Justice Roberts between 2006 and hearing the case in the 2007-08 term divested his small municipal bond holding. I suppose that might have been caused by my article, but more likely was just his financial advisor finding a more appropriate place to invest. Checking his most recent disclosure, though, he is back to holding municipal bonds.

Update 2/22/2022

NYT is out with a long article (Danny Hakim and Jo Becker, The Long Crusade of Clarence and Ginni Thomas, 2/22/22) that reploughs much of same ground Mayer did, but of course with additional details and some new information about Ginni Thomas’s activities – in particular, more White House meetings, involvement with Council for National Policy or CNP, and so on. It has more personal and background information about the Thomases than Mayer’s article.

Nothing really new or substantively different on the legal ethics issue, in my judgment. They do tag Thomas with speaking at two “fund raising” events (one for the Eagle Forum and the other for the Manhattan Institute), a no-no under the federal code of judicial conduct, which does not strictly apply to the Supremes.

One colorful quote about Ginni Thomas’s calls:

“When the Batphone rings and it’s Commissioner Ginni Gordon, otherwise known as Ginni Thomas, of course you have to show up,” said Dan Bongino, a former Secret Service agent turned popular pro-Trump radio host, after receiving one of Thomas’s Impact Awards in 2017. 

Ibid.

Update 3/30/2022

The Plot Thickens

The revelation of Ginni Thomas’s texts with White House Chief of Staff Mark Meadows, as reported by WaPo, are eye-opening, but will likely have no legal effect. They make two things clear:

• Ginni Thomas is worse than I had thought possible. She has been sucked into the Right Wing Conspiracy Theory Fever Swamp (she appears to actually believe this nutty stuff, including Q Anon crap!) and was actively lobbying the administration at the highest levels to further the 1/6/2022 efforts. This Bulwark piece presents an unvarnished view that is probably close to accurate.
• Because his wife’s involvement, Justice Thomas should have recused himself in the SCOTUS case involving the subpoena of materials from the National Archives. I guess it’s possible he didn’t know about the extent of her involvement, but that strains credulity more than a bit. In any case, he should recuse in future cases involving January 6th. This Nina Totenberg NPR story cites ethics experts to that effect. Given his apparent stubbornness and that it is solely his decision, I would not count on it.

Supreme Court justices are pretty much bullet proof, if they have thick enough skins. Thomas appears to have the skin of a hippo, so I expect this to have little effect.

4/6/2023 update

This ProPublica story about Justice Thomas taking luxury vacations with and at the expense of billionaire Harlan Crow provides an unseemly footnote. Crow appears to have indirectly funded Ginni Thomas’s salary by contributing half million to an organization that employed. The story doesn’t contain any evidence that Justice Thomas violated any rule of judicial ethics, even if he failed to file required disclosures, but it doesn’t look good, to say the least.