Category: Uncategorized

Defined benefit pension funding is in trouble all over, not just in Minnesota and many other states but in the private sector and Europe, according to this article in the Financial Times. FT considers the article (originally published in November) one of its 10 best big reads of 2019, so it’s not gated and is well worth reading.

The problem with Minnesota pension funding is obvious to anyone with a minor amount of financial savvy, a combination of (1) a failure to consistently (rarely) make ARCs (actuarial required contributions) and (2) unrealistically high rate of return assumptions that go into calculating the ARCs. That, of course, is not a sustainable situation, so at some point an unhappy day of reckoning will come and compromises will need to be made (some combination of additional contributions and benefit reductions).

The FT article points out a more generalized problem with defined benefit pension funding – that is, bond market returns that are low or nonexistent and (although not explicitly stated in the article, I think) a low or no inflation environment. Low bond returns make it very hard, even for prudently designed and managed pensions, to generate returns to match pension liabilities. This is especially a problem in Europe with its negative interest rates in some countries! The article mainly focuses on the issue of low bond yields. But I think my point (low or no inflation) is also a big component of the problem.

Low inflation is a problem because defined benefit pension plan liabilities are typically in nominal terms, that is they are not indexed to inflation. (There are important exceptions to this, such as the old federal CSRS and some states who foolishly have indexed their benefit formulas to inflation indexes. Minnesota fortunately does not, although it provides an annual, fixed (1% or 2%) benefit increase to retirees.) That means that a modest or greater amount of inflation can typically provide a margin for error, as inflation erodes the value of the benefit payment liabilities and makes it easier for plans to pay benefits.

Actuaries, of course, are aware of this and so attempt to account for inflation’s effects in their estimates either implicitly (in various growth assumptions and investment returns) or explicitly (less sure the extent to which they do that). So the real issue is whether inflation is exceeding the net effects of those actuarial assumptions.

The effects on actuarial estimates can move in both directions (e.g., benefits are based on wages, so wage inflation can drive up future benefits (liabilities), but inflation that increases inflation-adjusted investment returns and concurrently erodes the value of nominal benefits can may it easier to pay unindexed, current retiree benefits). My strong intuition (without attempting to dissect Minnesota’s actuarial formulas) is that a modest amount of inflation (3% to 5%) would materially help out Minnesota’s funding IF the investment return on stocks and bonds can come close to the rate of return estimate, on an inflation adjusted basis. That would provide a substantial discount in paying retiree benefits. Greater longevity means that the plans are paying benefits for longer periods, so inflation above the actuarial estimates should be a big help.

Another way to gain an insight about these effects is to consider the 1970s. With regard to investment returns, the 1970s were a lost decade of sorts – just plain awful for stock and bond returns. One would assume low investment returns would be really bad for pension funding and it sort of was. But rising inflation, although temporarily depressing bond values, providing an important offsetting effects by discounting fixed liabilities (i.e., amounts of future pension payments set in the previous low inflation environment) and by ramping up the nominal bond yields on new money or reinvested bond principal. That made it a much easier environment for defined benefit pension managers to deal with than the current environment.

In any case, interest rate policies in Europe and (to a lesser extent) in America have clearly hurt significantly the financial viability of defined benefit plans, as the FT article makes abundantly clear. This is an unintended side effect of low-interest macro-economic policies, I guess – particularly if they affect long bond rates, not just the short term rates that central banks are typically assumed to control. And the adverse effects on defined benefit pension funding is an unappreciated side effect of declining and/or persistently lower than expected inflation, which typically is widely perceived to be a good thing so long as deflation is avoided. Modest inflation provides a useful financial lubricant of sorts.

Janet Hottzblatt has a good blog post at TPC. The post’s title focuses on how the Sanders and Warren campaigns are the most restrained of the candidates in proposing tax credits, rather than direct spending programs for their policy proposals.

I find her post interesting, however, for three reasons (unrelated to that point but what I suspect is more the real reasons for her post):

• She details the criteria that she would use in deciding whether it is preferable to implement or expand a social program through a refundable tax credit or a direct government spending program. Some of her criteria mirror points that I made in my 2018 House Research publication on the topic: do the recipients already file returns, are the criteria easily verifiable and reasonably simple, etc. She just says it shorter, clearer, and simpler than I did.
• She provides some interesting history about decisions during the Bush I and Clinton administrations as to why Congress used or expanded refundable tax credits, rather than opting for direct spending programs. I was unaware of some of this detail.
• Her post is one more modest confirmation of my perceptions about the underlying political dynamics that have led to the rampant expansion of the use of tax expenditures over the last three decades: GOP elected politicians have a professed opposition to expanding government, but paradoxically sometimes (often?) really do want government to address social problems. That is why some them ran for elective office, not just to be government naysayers, as the GOP (particularly the Tea Party iterations of it) official mantra seems to be. (In my view, most politicians, even Republicans, are preternaturally inclined to want to solve problems with government policies.) Using tax expenditures allows them to reconcile these conflicting urges and to reach compromises with the Democrats. But that sadly compels ignoring the practical factors – such as those Holtzblatt outlines – in deciding whether to use direct or tax expenditures for programs. A state of affairs that seems sure to continue for the foreseeable future. The ability to expand government and call it a tax cut is irresistible to R’s and so it is a path of least resistance for the D’s to achieve their ends as well.

My take:

• It’s interesting and probably revealing that the two most lefty of the Democratic candidates are the ones who are proposing to implement their policies using the fewest tax credits or expenditures. That is so even though they have the most ambitious policy agendas. Their preference for direct expenditures may simply reflect their policy views (it’s best to use direct expenditures for most stuff), which I happened to agree with. But a more cynical and probably realistic take is that it reflects an essence of their campaigns – to wit, they are more concerned about making points (and saying what they believe under the more favorable view) than thinking about what they can get through Congress. The latter will clearly require some Republican buy-in (given the Senate rules requiring super-majorities effectively) and that can only occur by using tax expenditures, I would assume.
• None of this matters. Political considerations (for the reasons I detail above) will inexorably drive the policy train down the tax expenditure track.
• All this will undercut the tax system’s core function, raising revenue to fund basic government operations. I fear if the continuing expansion of tax expenditures continues at its current pace, that it will bring down the whole tax system edifice, particularly given Congress’s propensity to under fund the IRS. I hope I’m just paranoid.

The pending tax legislation, poised to pass Congress as part of the 2020 budget deal, includes the SECURE Act, which makes a number of retirement tax changes. Among those are two consequential changes in the required minimum distribution (RMD) rules for IRAs and other qualified plans like 401(k)s:

1. The minimum age at which RMDs apply is increased from age 70.5 to age 72; and
2. “Stretch IRAs” (applying to IRAs inherited from someone other than a spouse) are limited to a 10-year period, rather than the recipient’s life expectancy.

As I noted in a previous post, these provisions will automatically decrease (#1) or increase (#2) Minnesota state income tax revenues, just as they will have those same effects on federal revenues. I assume Congress was using #2, the new limits on stretch IRAs, as a way to offset the revenue losses under other SECURE Act provisions. But I’m perplexed as to the policy rationale for increasing the minimum age for RMDs.

Congressional documents do not state a rationale. One might guess that it was based on a concern that more people are working after 70.5 and so should not be required to start taking IRA distributions until later. Or that some individuals with small IRAs should not be required to start draining them if they have other resources, such as wages or pensions. (My cynical view: the financial institutions who receive fees for holding and managing retirement accounts likely lobbied for the change. Because of those fees, they surely view policies that augment balances in those accounts, such as looser RMD rules, to be in their interest.)

If either of those were the rationale, Congress should have limited the age increase to individuals who still have material labor earnings or small IRA balances (e.g., less than $200k or something like that). (RMDs don’t apply to a 401(k) if you’re still working for the employer.) That would have prevented the main effect of the change – which seems nonsensical to me – giving holders of large IRAs who don’t need to take distributions one or two more years of tax deferral.

Most IRA holders don’t have the luxury of deferring distributions until RMDs apply; they need to take money out of their IRA to live on. The age increase will mainly be a boon to upper income IRA holders. By contrast, the limits on stretch IRAs move in the opposite direction, since this modestly reduces the income tax benefits of inherited IRAs, especially for younger beneficiaries. One would assume that disproportionately will hit heirs of those with big IRAs.

In other words, Congress seems to pushing policy in opposite directions with its two RMD policy changes. One rationale for RMDs is to prevent affluent individuals from just stockpiling money, tax deferred in their IRAs and, then, passing them on to their heirs. (Treasury regulations describe the purpose of RMDs as “to ensure that the
favorable tax treatment afforded a qualified plan is used primarily to provide retirement
income to a participant and a designated beneficiary, rather than to increase the estate
of a participant.”) The taxes would be deferred longer (without RMDs or with lower RMDs), and the heirs frequently will be in lower tax brackets to boot. Here, the age increase allows deferring more taxes by the IRA holder, while the limits on stretch IRAs require heirs to pay their taxes more quickly.

To me, a couple of interesting elements of the age change are (1) going to a full year (age 72) rather than a half year (70.5) and (2) the likely unintended effect on qualified charitable distributions for some in the year when they turn 72.

Going to a full year means that people born after July 1, 1949 with birthdays in the first half of the calendar year will get two extra years of deferral, while those with birthdays in the second half of the year will only get one more year. I’m not sure why the current regime uses 70.5 years (rather than 70 or 71), since the RMD is set based on the account value at the end of the prior year and distributions can be taken at any time during the taxable year or up until April 1 of the following year. Thus, it isn’t an administrative issue. A little mystery I have never tried to track down. But whatever the reason, Congress has decided to change it.

A quirky effect on qualified charitable distributions (or QCDs) is more concerning and something that I doubt the relevant congressional staffers thought about. The law (not changed by the SECURE Act) allows an individual who is subject to RMDs in a traditional IRA to have distributions (up to $100,000 per year) be paid directly to a charity. Making a QCD avoids the distributions from being included in the IRA holder/taxpayer’s income. That, as has been widely noted, is preferable to making an itemized deduction for a contribution, since it keeps the distribution totally out of your income (e.g., no need to forgo the standard deduction and doesn’t count toward any of the AGI limits, etc.). But in the first year RMDs apply, a QCD can only be made after the taxpayer meets the age requirement. See IRS Publication 590 which (mirroring the statutory language) says “You must be at least age 70½ when the distribution was made.” That will become 72 when the SECURE Act is signed into law. Thus, along with being able to defer taxes longer means that younger IRA holders will need to wait one or two more years before they can use QCD to make their charitable contributions.

Note how this could be a problem if your birthday is late in December. Assume your birthday is on December 30th, which is a Saturday in the year you turn 72. Qualified distributions are made by requesting the trustee of the IRA to cut a check to the charity. That check must be dated on or after the birthday on which you turn age 72 to qualify – i.e., on or after December 30th in this case. That effectively means the QCD will be made in the next calendar year (and tax year for a calendar year taxpayer, which virtually all individuals are). That in itself isn’t a big deal, since RMDs for the first year can be taken up until April 1 of the following year and still meet the requirement. (Unlike claiming an itemized deduction for a charitable contribution, it doesn’t need to be made in the taxable year. All you are doing is keeping the distribution out of your income.) Thus, the charity can receive the check in the next calendar year and the taxpayer can still exclude it from his or her income, IF the April 1st deadline is met AND the taxpayer gets a qualifying receipt from the charity. That is where the rub may come in, since charities typically issue those receipts on a calendar year cycle. They may have to make adjustments to accommodate people with birthdays very late in the calendar year who become subject to RMDs in that calendar year, so they can satisfy the IRS documentation requirements.

One might assume that it is concerns like this that led to the use of half year (70.5) rule. But 70.5 age was set well before QCDs were even a thing.

Congress could fix this by allowing QCDs to be made any time a qualifying RMD can be taken (i.e., at any time during the taxable year when the taxpayer becomes subject to RMDs). Why Congress required taxpayers to meet the minimum age requirement before making a QCD when a similar requirement does not apply to RMDs themselves is unclear to me.

Dated: December 19, 2019

The STRIB reprinted a column by Justin Fox (Bloomberg economics writer) with more evidence of traditional retailers’ woes. This follows up on my recent post on the use of the Bloomington/MOA tax to help finance MOA’s proposed waterpark, which it appears to hope will get more open wallets into MOA’s stores, hotels and restaurants.

Fox reports that in-store retail jobs have dropped by over 400k since 2017 – despite a growing economy and retail sales. The obvious explanation is more online sales and less buying in stores. Employment growth in online retail jobs has not totally offset the decline of in-store jobs and many of those online jobs (no surprise) are in “couriers and messengers” with smaller growth in warehouse employment.

Some observations:

• These jobs data underline the challenges faced by traditional retailers, including MOA.
• The data reflect some fundamental shifts in the character of jobs (clerks versus stockers, pickers, and drivers).
• Most retail jobs have never been very good (high paying) so it is hard to get too choked up about this, BUT
• Conversion of sales from bricks and mortar to online sales appears to yield productivity increases, as one would expect, and that should lead to compensation (wage) increases as employers share some of the productivity dividend with their workers, BUT
• The modern (post 1980s) pattern by businesses is to not share much of their productivity dividends with lower wage workers, but rather to channel them back to capital (shareholders or owners) or more highly compensated employees (i.e., management, particularly top management).
• This is just one more data point in what appears to be the inexorable growth in inequality.

Retail malls face daunting challenges – the specter of Amazon and online shopping is dramatically reducing purchases at malls and big box stores.  Bankruptcies, including outright liquidations (Herbergers and Pamida to name two affecting Minnesota), of malls and retailers show the harsh reality of this phenomenon.  This WaPo article describes the challenges and how they have resulted in closures. According to a Credit Suisse study it cites, 25% of malls are expected to close by 2023.

The WaPo story describes how more successful malls are adapting.  Thriving malls, according to the story, are doing so by reinventing themselves as “integrated lifestyle hubs” whatever that is. It appears to mean figuring out alternative way beyond a standard shopping experience of attracting customers to show up and spend money – e.g., providing yoga studios, fitness centers, microbreweries (what else?), health clinics, and so forth.

One assumes that this challenge must be hitting our local giga-mall, the Mall of America (or MOA). Moreover, because MOA is so big and, thus, must lure many more customers to its premises, the challenge is much bigger than adding a few yoga studios, microbreweries, and other trendy stuff can solve.

Rather than its years-ago plan of simply adding more of same in phase II, MOA’s answer is to build a big waterpark. The theory is that adding a large waterpark as part of Phase II will attract enough people to MOA who will spend money at its hotels, restaurants, and shops to keep most of them afloat. [I have thoughts about whether this will work, but I have no comparative advantage in judging such things so won’t spend any time on it. I think, though, we can safely say the move is risky.] This is a variation on what Triple Five (MOA’s owners) are doing at their just opened giant mall in NJ.  The WaPo story discusses the Ghermezians (Triple Five’s owners) and their strategy as it relates to the NJ mall and its plans for another big mall in Miami, so I’m not totally making this up.

As with all things related to MOA, one can safely predict that its response will both:

• Be risky
• Involve government reducing MOA’s costs and risk.

What I find interesting about this is the creativity of MOA’s advisers in devising a way to use decade-old (2008) special legislation to put together a financing plan that snags government subsidies and reduces the credit risk to Triple Five’s enterprises, including MOA.  How they did that can only be described as financial alchemy.  Whoever came up with this should get the Rumpelstiltskin award (to mix two metaphors involving gold).  It’s a story worth outlining.

In spring 2008 (after the Great Recession had begun but before its existence was commonly recognized), the legislature enacted special legislation to help Bloomington finance and finally develop Phase II of MOA.  (Why Phase II took so long to develop is a separate story with many twists and turns. The delays probably helped MOA dodge a bullet of holding too much obsolete retail space in the current environment.) This legislation was controversial (both in the legislature and in Bloomington).  Controversy centered on whether the public should provide more subsidies to MOA’s owners beyond those approved in the 1980s and 1990s.  Opponents (including Rep. Ann Lenczewski who chaired the Houses Taxes Committee and represented Bloomington) felt that more public assistance was inappropriate.  (MOA’s increased property tax, through TIF, had paid for the Phase I’s parking ramp and built up a war chest of development money, so the city, county, and school had to that point had not received any expanded property tax base as a result of MOA construction nearly 20 years earlier.)  By contrast, construction trade unions (who generally favoring spending more on projects that their members work on) and legislative supporters of the mall supported more TIF for Phase II by allowing an extended TIF period. Because that would have exceeded what is permitted under Minnesota law generally, it required enactment of special legislation for Bloomington.

A compromise was reached that became law in 2008. One element of this compromise – inserted by opponents to permit the city to make MOA pay more of the development’s cost – allowed the city to impose a sales tax that would apply ONLY TO MOA.  2008 Minn. Laws, ch. 366, art. 5 § 28.  This was done by limiting the taxing authority to the area of the two TIF districts in which MOA is located. Imposing the “tax” would have raised the price of taxable purchases at MOA and was for obvious reasons opposed by MOA.  In effect, it would have been little different than expecting the owners and their tenants to pay more of the mall’s infrastructure costs (parking ramps and so forth), rather than taxpayers.

This “tax” is a tax in name only.  MOA could have reached the same result by charging more rent to its retail tenants (e.g., a lease can have a “gross receipts” clause that varies the amount of rent based on how much revenue the tenant generates, adding such a clause to MOA’s retail leases could easily include a sales calibration to rent which replicated the effect of a sales tax).  Although Bloomington approved the special law to use its other provisions, it did not impose the tax. I assume it made that choice because MOA opposed paying more of the cost of Phase II’s parking ramps or whatever else was planned for the TIF money.

The Great Recession put most development of Phase II on hold shortly thereafter.  Legislation enacted in 2013 provided a wholly different funding mechanism (i.e., not a standard TIF extension) for Phase II that relied on the Fiscal Disparities taxes paid by MOA, rather than diverting city, county, and school property taxes as TIF does.  This mechanism will be used to finance the second big parking ramp and other infrastructure to serve MOA.  But the MOA sales tax remained on the books, essentially a vestige of the 2008 legislative negotiations.

But as MOA formulated its waterpark plan in 2017-2018, the MOA sales tax provided an opportunity for it to snag some government subsidies. Bloomington apparently is more than happy to oblige MOA, since the costs will be born by the feds.  As I noted in my earlier post, Bloomington choose to skip tapping potentially a small state subsidy under the Minnesota income tax. For that, other Minnesota governments should be thankful.

The proposed waterpark will be, for all intents and purposes, a private facility. The normal way to finance it would be for MOA or a new for-profit entity formed by Triple Five to finance it with some combination of equity and debt.  I’m sure that is the way the waterparks in Wisconsin Dells have been financed and I suspect that to the casual observer, the MOA waterpark will appear little different functionally than those at the Dells.

I have no knowledge of the financial condition of Triple Five, a private entity, but assume that its credit must be stretched (maybe to the max) by the amount it has spent on its mammoth NJ mall that just opened and its Miami plans.  The media has reported that Triple Five pledged 49% of MOA to secure its borrowings for the NJ mall.  That likely made the cost of it borrowing to finance the MOA waterpark unaffordable (relative to projections of waterpark revenues), if lenders or bond market were willing to make credit available at all.

So how can MOA successfully borrow money to finance the waterpark? Enter the 2008 MOA tax, which I had always considered a dead letter, as an enabling mechanism. There were essentially two steps involved:

1. Have the waterpark owned and operated by a nonprofit, but not just any nonprofit but a 501(c)(3) entity, a charity under the federal tax code that qualifies to have its facilities financed with tax-exempt bonds. I’m not sure exactly what the “charitable” purpose is (and haven’t made any effort to find out), but the federal tax law is malleable and expansive when it comes to the types of charitable purposes that qualify.  (You don’t need to be a classic charity that helps the poor; in fact, I doubt poor people will get a break when they try to use the waterpark. They probably would rather not have poor people show up at all, since they won’t be prospective shoppers with fat wallets.) If a state or local government is willing to issue debt to finance the facility, the interest on the debt will be exempt from federal tax. That will, as I described in my previous post, lower the interest rate on the bonds, making it easier for the project to cash flow after paying its debt.  The government subsidy is courtesy of the lower federal income tax revenues resulting from the foregone tax on the bonds’ interest.
2. Pledge the MOA sales tax as additional security for the bonds.  Although revenues from the waterpark itself will be pledged to pay the bonds, that probably won’t do the job. Without some unknown amount of equity and/or a track record or some assurance that the waterpark will generate enough revenues, a pure revenue pledge is unlike to be enough to induce bond buyers to fund the project. In their minds, it’s too risky. MOA could pledge its revenues and/or equity but remember much of that has already been pledged to the NJ Mall and doing so would cause the bonds to lose their tax-exempt status under the federal tax law. The latter is so, because the project would be secured (too much) by a private business, not a government or qualifying nonprofit. That is where the goofy MOA sales tax comes in. Recall its proceeds are really not functionally different than mall revenues but the federal tax law considers it to be a real tax, so pledging it does not contaminate the project with private business revenues or security interest, as pledging MOA lease payments would.

Voila, problem solved – at least in theory – (1) the federal subsidy through tax exempt bonds lowers the interest rate; (2) compartmentalizing the waterpark in a separate, unrelated nonprofit entity insulates MOA’s credit from being on the line; (3) pledging the MOA tax, effectively means the mall and its tenants, are standing behind the waterpark, making Bloomington happy; and (4) the MOA tax appears to the federal tax law and to MOA’s and Triple Five’s creditors to be government support, rather than MOA’s support.

Some of my random observation on all of this:

• The unclear color of the money. This only works (we’ll know whether it does work, if bond sale goes through) because of the chameleon-like character of the MOA tax. Its color or true nature depends upon your angle of view: from the federal tax law’s, MOA’s, the bond market’s, and MOA creditors’ perspectives, it’s a tax; from the perspective of Bloomington and financial realists, it is a private contribution or payment by MOA and its tenants.  For the former, that means the MOA tax provides an independent, government source of funding for the waterpark other than MOA itself.  For the latter, it means that the city is not taxing its constituents to pay for a private waterpark and keeps the viability of the waterpark from being tangled up with the city’s finances.  Which is correct? I would tend to the latter view, but Bloomington is inextricably linked to MOA’s success, as I note below.
• Clothing retailers at MOA won’t care.  I’m curious what MOA’s tenants think of this scheme. They certainly have a stake in the ongoing success of MOA and so, I’m sure, would like to see the water park succeed – success being defined as attracting sufficient customers to ensure the ongoing viability of the overall development and their own operations. It is worth noting that the MOA tax will hit those tenants very unequally: because of Minnesota’s clothing exemption, clothing retailers will be spared. Those who sell taxable goods will be more directly affected.
• Nobody will notice (okay, most won’t notice). The structure of the sales tax, as an add-on at the cash register, makes it an effective way to disguise its price-raising effects to shoppers. Research on the behavioral effects of taxes shows that taxes that are built into the explicit prices of commodities (e.g., excise taxes like the gas tax and cigarette taxes that show up in the sticker price of the product) have much larger effects on purchasing behavior, as contrasted with a tax, like a retail sales tax, that only shows up at the cash register – probably after the consumer has already decided to buy the item. That should provide small comfort to MOA’s tenants who may have their sales taxed, if the water park’s finances go South.
• Is it worth it? The real public policy question here is whether the government (whether city, state, or federal) should go on subsidizing MOA. Answering that question would typically be done with a thorough and careful cost-benefit study, something that has never been done throughout this process (from 1984 on).  (Full disclosure: As a legislative staffer, I worked on MOA state legislation from 1985 through 2018 and so was complicit in that failure.) Given that this “sales tax” is really little more than an odd sort of rent increase at MOA, it’s hard to get too excited about that issue here, although issuing more tax-exempt bonds for private projects is never good in my mind. I totally understand why Bloomington is going along (see the next bullet for more on that).
• It ain’t easy to dismount a tiger. The current dilemma Bloomington faces illustrates how long-ago decisions effectively bind a city or put it on a course that is very difficult to deviate from. It’s like the old Chinese proverb of how difficult it is get off the back of tiger. The state’s 1977 decision to build the Metrodome effectively required Bloomington to redevelop the Met Stadium site, a prime piece of real estate that was close to the airport and on 494. The city in 1982 had two basic choices, a classic fork in the road:
  1. It could have allowed the private market to redevelop the site without government assistance. That likely would have resulted in some mix of retail, hotels, and offices and would have generated immediate tax base as the site gradually built out. Given how prime a site it was, the development likely would have been high value and would have contributed to Bloomington and its 494 strip being a classic “Edge City.”  (Interestingly, Wikipedia does not list Bloomington as an Edge City.  I had always assumed it was.)
  2. The other path, chosen by the city as encouraged by then Governor Perpich, opted for a signature development, MOA. That approach was perceived to help give Bloomington a national reputation (sort of like a corporation buying naming rights, I guess), making up for the loss of hosting the Twins and Vikings at Met Stadium.  That loss, by most accounts, stung Bloomington public officials and business community as a big loss of prestige and visibility for the suburb. But that required the city (and county and school) forgoing increased property taxes from the site for well over 30 years because TIF was used, recycling the increased property tax back into the development. That period has now ended (thanks to the 2013 special legislation which shifted the ongoing property tax costs to the fiscal disparities program spreading costs over the entire metro area), but that doesn’t end Bloomington’s costs and responsibility.

Opting for a mega development, like MOA, is much higher risk – obviously because of the public investment (cost), but also because a big single use development is less certain to generate a return, long run. In investment terms, it is not diversified the way a market-driven, but more modest development of mixed uses would be.  It’s like investing all your money in one growth stock, rather than buying a diversified mix of stocks and bonds. Your chances are greater of striking out or hitting a home run, but it’s safer to try and hit a single.

Another consequence of choosing path #2 that probably was not recognized by Bloomington in 1985 is the ongoing effects of opting to fund a signature development like MOA. A mammoth development like MOA in a modest sized city will always be an implicit city responsibility. If MOA falls on hard times and requires extensive repurposing or new investment, there is little question that city will feel it is practically and politically necessary to help pay for that and will undoubtedly ask for the state’s help as well, at least if past experience is any guide. The city’s behavior with regard to the waterpark is instructive in this regard. It illustrates a sobering, but often little recognized cost of not just allowing the private market to determine the course of real estate development.