Kristi Culpepper

Rarely do I feel compelled to share my political beliefs on topics, and that preference is not going to change with this essay. I think governments should be good stewards of taxpayer funds and there are universally recognized best practices for accomplishing that. Thus, I try to demonstrate that stewardship is mostly a mechanical — and not ideological — ideal.

In responding to Puerto Rico’s fiscal crisis, Republicans and Democrats should not be far apart. This conversation simply needs to be kept on a practical level.

An overview of Treasury’s legislative proposals

Here is the full version of Treasury’s proposal, along with the supportingfinancial and economic analysis. Briefly, Treasury’s proposal involves:

Territory-specific legislation that would allow Puerto Rico to restructure its outstanding debt. This is different than existing proposals that would merely include Puerto Rico in Chapter 9 of the federal bankruptcy code. This would also allow (but not require) the Commonwealth to restructureall of its outstanding debt, including its general obligation bonds.

A federally appointed oversight board to supervise the debt restructuring and introduce Puerto Rico’s government to the standard financial controls a government should have in place.

Increased access to Medicaid. And,

Access to the Earned Income Tax Credit.

Responding to Republicans’ legitimate concerns

Democratic lawmakers were on board with the concept of giving Puerto Rico bankruptcy powers even before the administration was. This does not put them at an advantage though, as they were previously touting a deeply flawed piece of legislation.

Republican lawmakers have expressed some completely legitimate concerns about going down this path in general. I am going to try to show how these concerns are ultimately able to be resolved.

My responses will likely not satisfy lawmakers (on either side) who have accepted political donations from Puerto Rico bondholders. (I am being deliberately gracious and not naming names.) There’s nothing one can do about people who choose to use their office that way. However, this discussion might help lawmakers who are trying to figure out how a potential bankruptcy fits within their intellectual framework.

Concern #1: “This is a bailout.”

Treasury emphatically reiterates in testimony and opinion pieces that its legislative proposals do not constitute a bailout. If your definition of a bailout stops at an explicit guarantee of the Commonwealth’s debt, then that’s accurate.

However, I’m sure that is not how most Americans define bailout. For most Americans, a bailout is anything that benefits an irresponsible actor that is not budget-neutral for the federal government. Half of Treasury’s proposal involves the provision or extension of government programs, which federal taxpayers will fund either directly or through tax expenditures. And the logic behind incurring these expenses is that doing so will provide budgetary relief for Puerto Rico’s government. So, yes, this is a bailout.

Moreover, this isn’t the first time Puerto Rico’s government has received what is tantamount to a federal bailout. For precedent, I would refer readers to Tax Analysts’ excellent research from January 2014. The problem with past efforts — as I will get to later — is that the federal government let Puerto Rico officials continue their feckless, debt-fueled gamble for redemption.

We have to move beyond the is-it or isn’t-it a bailout nonsense. It is now inevitable that Puerto Rico will need additional financial resources to operate. What matters is designing a policy response that is sufficiently comprehensive, such that taxpayers’ future financial commitments will be limited in scope.

Concern #2: “This changes the rules in the middle of the game.”

Any lawmaker who regards an emergency as an opportunity to moralize cannot in good faith be considered a fiscal conservative. Moralizing, when it cannot change the facts and circumstances, is very expensive. Sovereign debt crises have revealed over and over again that the worst things a government can do are restructure too little and restructure too late. But Puerto Rico is in an even more dire situation, as its attempts to restructure its debt independently will be an exercise in futility against a chaotic legal landscape.

This means the territory will devote its limited financial resources to negotiating and fighting with an army of creditors, who will sue the government and each other ad nauseam. Those are resources that cannot be spent on the real business of government in the form of operations and services. If you want to understand the actual cost of congressional inaction, ask what happens when those resources run out. The government will collapse. The exodus of fiscal crisis refugees will continue — and those refugees will increase the cost burden (in terms of additional services) on the mainland municipalities that absorb them. Puerto Rico will be left with the populations that do not have the option of relocating and require acute government support. What will the federal government do when these events unfold to their logical end? What additional cost will a bailout carry then, when there is not a functioning economy and tax base to offset expenses?

This fiscal and economic reality is the reason Chapter 9 came into existence in the first place. After the Great Depression, thousands of municipalities across the United States defaulted on their debt. These governments found themselves in roughly the same situation Puerto Rico finds itself now — trying to bargain with intransigent creditors to the point that they could not afford to function as governments. The main benefit of Chapter 9 is that it halts most legal actions and provides an impartial forum for the government to work out its issues with all stakeholders (which includes interested parties other than bondholders — something the media tends to forget).

When Chapter 9 was enacted following the Depression, did it “change the rules” on existing contracts? Of course it did. Dramatically. Was that a bad thing? No, it allowed American municipalities to continue to operate in perpetuity. Government in the United States is premised on the Enlightenment notion of a social contract. The crux of Chapter 9 is that the survival of that social contract is more important than the survival of private financial contracts.

One of the problems with how the entire “debate” over Puerto Rico’s debt crisis is framed is that most of the “how did Puerto Rico get in this position” explainers make it seem like Puerto Rico’s situation is an accident of the federal tax code. Tax provisions that had been advantageous to manufacturers on the island were phased out beginning in the mid-1990s, and now the island can’t support its debt. It’s nobody’s fault, and there are contracts that must be honored in full. The Commonwealth’s word is its bond!

All of the credit challenges that exist now for Puerto Rico existed when investors were hoovering up its bonds. Puerto Rico has been the bête noire of muni credits for as long as I have been in public finance. No one was lending money to the government because they thought it had a promising credit story. The bond market did not extend $72 billion of credit to a population that is smaller and poorer than an Appalachian state and has pension commitments that will begin swallowing one-quarter of its general fund resources in five years because that is rational behavior. Puerto Rico is one of few municipal market issuers (and, really, the only one issuing much debt) where the interest income from the bonds is tax-exempt at every level of government. And as the Federal Reserve pushed interest rates into the ground, Puerto Rico’s deteriorating credit became even more attractive to yield hogs.

Puerto Rico has ended up with so much debt for the same reason any bubble is created. Is it truly of profound moral consequence to ensure these actors are profitable? Puerto Rico’s debt is the municipal market’s equivalent to mortgage backed securities comprised of liar loans. And it takes two to tango. The appeals to morality and pearl-clutching over the sanctity of contracts from hedge funds who picked up Puerto Rico’s last bond issue are the worst. Their bond documents specifically addressed the possibility of Congress enacting restructuring legislation that would trump their legal protections(see page 9).

Some of Puerto Rico’s bondholders have actively managed misinformation campaigns, where they are trying to suggest that Puerto Rico’s debt burden is affordable and can be paid in full. Treasury has done a pretty good job of refuting these points mathematically. It doesn’t actually matter, however. These investors (and probably the monolines as well) are going to have a seagull-meets-jet engine reckoning whether Congress takes action or not. The money to pay them doesn’t exist. A government doesn’t need a bankruptcy regime to be functionally insolvent. That’s how credit works.

Concern #3: “Puerto Rico needs to release accurate financial information before Congress can consider assisting the Commonwealth.”

The Sun will consume all of its hydrogen, turn into a red giant, and explode before Puerto Rico’s government releases accurate and timely financial information. We are almost 500 days past the close of fiscal year 2014, and Puerto Rico says it is still waiting on agencies to submit required information to auditors. Lumesis recently quantified how well established Puerto Rico’s trend of poor disclosure is. And you’ll note that this did not stop investors from lending the Commonwealth money or the insurers from wrapping the bonds over the years. Actually, the bond insurers resolved that problem byproviding poor disclosure about their exposure to Puerto Rico to their own investors.

This is why Puerto Rico needs a federally appointed financial control board. Waiting on the government to change before considering federal legislation is the same as doing nothing (see costs of inaction above for how that turns out).

Concern #4: “Enacting territory-specific restructuring legislation will lead to state bankruptcy legislation.”

These slippery slope arguments neglect to mention the key difference between states and Puerto Rico. States are sovereign entities under the 10th Amendment to the US Constitution. They don’t need protection from the federal court system and Congress can’t force any kind of debt restructuring framework upon them. It’s not an accident that municipalities have to have explicit state permission to file for Chapter 9. This requirement was established following the US Supreme Court’s 1936 ruling in Ashton v. Cameron County Water Improvement Dist. No. 1, in recognition that municipal entities are political subdivisions of sovereign entities. Conversely, Congress can do pretty much whatever it wants to Puerto Rico.

There are three reasons simply including Puerto Rico by reference in Chapter 9 — as H.R. 870, the legislation sponsored by Puerto Rico’s non-voting member of Congress, proposes — is problematic. Some of this I covered in my earlier essay, A Financial Control Board Proposal for Puerto Rico.

First, much of the outstanding debt that Puerto Rico needs to restructure isprotected by Chapter 9. All H.R. 870 does is redefine “state” within Chapter 9 of the federal bankruptcy code to include Puerto Rico. As I explained above, states cannot file for Chapter 9, only their political subdivisions. The logic behind the legislation is that if Puerto Rico is considered a state, its general obligation debt would be shielded from bankruptcy. The Commonwealth’s other instrumentalities, however, could file for bankruptcy and adjust their debts.

The problem with H.R. 870 is that the bonds issued by some of Puerto Rico’s instrumentalities are of a variety that other provisions in Chapter 9 protect from being impaired — that is, “special revenue” debt. The pledge of what are considered special revenues survives bankruptcy. Special revenues are defined as:

(A) Receipts derived from the ownership, operation, and disposition of projects and systems of the debtor that are primarily used or intended to be used primarily to provide transportation, utility, or other services, including the proceeds of borrowings to finance the projects or systems;

(B) Special excise taxes imposed on particular activities or transactions;

(C) Incremental tax receipts from the benefited area in the case of tax increment financing;

(D) Other revenues or receipts derived from particular functions of the debtor, whether or not the debtor has other functions; or

(E) Taxes specifically levied to finance one or more projects or systems, excluding receipts from general property, sale, or income taxes (other than tax increment financing) levied to finance the general purpose of the debtor.

(James Spiotto, from The Handbook of Municipal Bonds).

The only occasion (that I know of) where a judge mentioned the possibility of a municipality being able to “cram down” special revenue debt was a digression in Judge Thomas Bennett’s 2012 opinion in The Bank of New York Mellon, as Indenture Trustee, et. al. v. Jefferson County, Alabama. (For background on Jefferson County’s bankruptcy, here is a piece I wrote on Business Insider.)

Bennett explained that the 1988 Amendments to the bankruptcy code (the legislation where special revenues protection was established) do not “override other provisions of the bankruptcy code that may be used by a municipal debtor to impair special revenue claims and liens in the plan process in connection with the debtor’s exit from bankruptcy, i.e., the power to value the special lenders’ secured claim downward under section 506 and use the ‘cram down’ power in section 1129 to restructure the secured claim.” (From link above.)

Essentially, if the bonds are under-secured going into the bankruptcy, a portion of the debt can be treated as unsecured debt (this is referred to as lien-stripping). I believe Jones Day made a similar argument in Detroit’s bankruptcy in connection with the city’s water and sewer debt, but it went nowhere because the bonds were not, in fact, under-secured. Regardless, this is an unprecedented and ambiguous part of the law. Puerto Rico needs legislation that reflects how the Commonwealth’s debt is structured, as opposed to the more straightforward debt structures of cities, counties, etc.

Second, Chapter 9 is ultimately driven by the municipality or its representatives (emergency managers or other appointees). The same Puerto Rico officials who drove the Commonwealth to insolvency, cannot even produce basic financial statements, and approach a government shutdown with little more than a shrug, cannot be trusted to navigate bankruptcy on their own.

Third, this strategy was originally proposed as a means to insulate Puerto Rico’s general obligation bonds from a debt restructuring. Puerto Rico’s non-voting member of Congress offers this as a matter of honor, since the bonds are protected by Puerto Rico’s constitution. Both the US Treasury andeconomists at the Federal Reserve Bank of New York have discussed at length why this is unreasonable. It is also worth noting that federal bankruptcy judges have dismissed state constitutional protections in other cases (with respect to pension commitments). The only durable protection appears to be a statutory lien on a specific source of revenue.

Concern #5: “What if we are giving Puerto Rico debt relief, only to have the Commonwealth run up its debt again? Aren’t we introducing moral hazard here?”

I see three outcomes here, and they depend on Puerto Rico’s political status when it re-emerges from this episode.

One, the Commonwealth emerges with a recast culture and attitude toward its finances, because it does not want to surrender control of its government again.

Two, Puerto Rico becomes a state. As such, Puerto Rico would be operating under a vastly different system of incentives and the debt restructuring legislation will be void.

Three, they run up their debt again. And they file for bankruptcy again. And something resembling market discipline emerges, which should have existed in the first place.

I am sympathetic with the concept of disruption that has taken off in the tech sector — see my earlier essay, In Defense of Disruption as a Cultural Phenomenon. However, I also think it is important to recognize that our social institutions have not evolved by accident. Some of what is labeled as disruption is, in fact, providing solutions to problems that do not exist.

The latter is precisely how I feel about the start-up Neighborly, which garnered headlines this week for raising $5.5 million from Jon Lonsdale’s Formation 8 and Ashton Kutcher’s Sound Ventures. Yes, that Ashton Kutcher. For coverage from someone far more entertaining than I am, readLauren La Capra’s hilarious piece at Fusion.

Of course, it is the job of venture capitalists to sort out the wheat from the chaff. In the case of Neighborly, however, we are talking about managing other people’s wealth through a brokerage firm— and more specifically, highly unsophisticated investors’ wealth — and offering them what is arguably unsuitable investment advice. So whether this turns out to be a financially viable recreational venture for celebrities — and it probably will not — is beside the point.

Artisanal Infrastructure?

I guess it was inevitable that someone would propose artisanal infrastructure investment for millennials. (Hat tip to @dougtee for coining the term.) I mean, someone’s already proposed artisanal stock options. Millennials apparently have a lot of kooky investment preferences. They don’t trust the stock market, so now there is a start-up to pitch them derivatives. They want to be more involved in their community, so now there is a start-up to pitch them micro-munis. With all this money they have to invest, it’s a wonder all we hear about is the mountain of student debt outstanding.

Here’s how Bloomberg described Neighborly back in August:

Many community-interest projects are still funded by traditional sources. When they are, start-ups such as Neighborly are helping individuals play a role. The company, which says on its website it is “democratizing the $3.8 trillion municipal securities market,” aims to allow users to buy portions of a municipal bond rather than the whole thing.

“Most people don’t think to go, ‘Hey, I’m going to go buy a municipal bond,’” says Jase Wilson, the chief executive officer of Neighborly. “It’s about breaking down the minimum denomination on one side so there’s more of a democratic access. It’s about demystifying it.”

Wilson says Neighborly seeks to sell municipal securities through its registered broker-dealer and tries to make it easier for investors to understand where their money is going. He says he hopes the company will help rekindle local involvement in the municipal market, which once created such projects as the Golden Gate Bridge. At the moment, the bond market that millennials’ grandparents invested in is gone.

“They would go buy bonds in their own community for a project they support,” Wilson says. “They would put [the bond certificate] under their bed and clip the coupon, and they would just get some ice cream. It was super local, super hands on, super regenerative.”

They’d clip the coupon and go get ice cream? Isn’t it a little strange that the tech sector would be nostalgic about physical coupons? Once created such projects as the Golden Gate Bridge? Did the market stop financing capital projects and no one told us?

It’s absurd to suggest that the municipal bond market has somehow become “less local” than it was when the Golden Gate Bridge was built. Heck, the Golden Gate Bridge and Highway District, the debt-issuing authorityestablished in 1923 to finance the construction of the bridge, still exists andstill issues debt. There are over 90,000 borrowers in the municipal bond market. Most of them are small governmental entities, and many of their bonds are picked up by local households and businesses that directly benefit from the improvements being financed. This is one of the many reasons why municipal defaults are so rare.

I have said in past essays that the United States is the City on the Hill when it comes to funding infrastructure. No other country in the world has this system of local funding of infrastructure, which evolved historically with the growth and development of our nation and fits our federalist ideals. It is certainly true that procurement structures like public-private partnerships are gaining in popularity, but they are hardly the threat to the multi-trillion-dollar municipal bond market some observers make them out to be. They are an opportunity to expand and diversify the investor base.

All this is to say that the municipal bond market already represents a system of crowdfunding. It is kind of difficult to disrupt crowdfunding with crowdfunding.

From TechCrunch:

Wilson started getting to know municipal bond traders, who were exchanging and issuing the debt that cities need to fund projects. A trader walked Wilson through underwriting and compliance, and then over to a fax machine with stacks of papers and contracts. The trader explained to Wilson that just by switching from corporate to municipal debt, he could double his income without doing any extra work. This was because public city governments ended up with extra layers of fees throughout all of the steps of municipal debt issuance that corporate issuers were more rigorous about holding accountable.

A fax machine. Hahaha. I’d love to know what firm this is.

As someone who has seen literally thousands of municipal bond transactions, I can’t even begin to tell you how false all of this is. First of all, it blows my mind that anyone would suggest that it is expensive to issue bonds of any stripe these days. In an environment where there is too much money chasing too few bonds (thanks to the Federal Reserve and a period of net negative issuance following the Great Recession), issuers hold all of the power. Basically every bond issue has been oversubscribed, even toxic credits. Competition among banks has been intense, and underwriting fees reflect that.

Second, not to point out the obvious to Mr. Wilson, but one major difference between corporate borrowers and government borrowers is that the expenses incurred by governments — including the costs of issuing bonds — is narrowly restricted by state laws and federal tax law. If you came up in a firm that tacked on fees for fun and that is your idea of how the world operates, the securities industry is probably not the ideal niche for you.

Cutting through all the crowdfunding rhetoric and getting to the mechanics, it seems that Neighborly is really only a standard brokerage firm that is marketing its services to (albeit civic-minded) millennials, and trying to persuade issuers of municipal bonds (i.e., governments) to use smaller denominations (rather than the usual $5,000 to $100,000) to accommodate their client base, which has relatively little money to invest but wants to participate in their community.

For top-rated credits, it probably doesn’t matter what denominations an issuer uses, and I think most issuers realize that. The larger denominations can be used to keep firms from marketing risky deals to unsophisticated retail investors (so pretty much Neighborly’s target market), as was the case with Puerto Rico’s last general obligation debt offering. Although I would be skeptical of a government that is especially enthused about placing a deal with the kind of retail investors Neighborly is bringing to the table regardless. That seems like a situation that is ripe for fraud.

Why municipal bonds are generally not suitable for younger investors

Practically speaking, the minimum denomination of bonds is not the limiting factor for investing in municipal bonds. Not even remotely close. The tax status of municipal bonds is. Millennials are not a native source of demand for municipal bonds for a reason — they do not have incomes high enough to benefit from the tax shelter the bonds provide. (For background, see my essay, Why Are Municipal Bonds Tax-Exempt?)

So what does this mean for them as investors? Municipal bonds have lower yields than other investment vehicles (e.g. corporate bonds) because investors are willing to accept lower interest rates in exchange for the tax benefits they offer. I mentioned this to the CEO of Neighborly during a brief exchange on Twitter — that by recommending municipal bonds to relatively low-earning millennials, his firm would basically just be telling them them to give up yield for nothing — and he told me that this was only true now, and these same people would be earning more in the future.

But that’s not how investing works. Usually when someone invests money, they want to make the most money given the amount of risk they are willing to make. Their savings aren’t supposed to be a form of tuition.

On top of that, there are good reasons that firms do not try to put investors with relatively small assets under management into high maintenance (in terms of due diligence) holdings like bonds. There’s no money in it, unless the firm is willing to charge exceptionally high fees, which the client by definition cannot afford. There’s also a lot of risk in compliance and surveillance. I imagine by the end of this venture, Lonsdale and Kutcher will realize they’ve donated to a charity.

So, Puerto Rico defaulted on a relatively minor bond payment last Monday, and it is a truth universally acknowledged that a government default must be in want of an avalanche of hot takes. Although it is difficult to absorb all of the instant authorities out there, these events do provide an interesting opportunity to observe how various money managers form opinions about different securities.

Brian Chappatta’s article, Puerto Rico Shows Perils of Muni Bonds Backed by Empty Promises, quoting a number of market participants and observers hyping the risks of appropriation-supported bonds versus other bond structures, has compelled me to write something about realistic versus sentimental perspectives on bondholder protections.

The plural of anecdote is not data

The article begins:

Municipal-bond investors are learning that when cash gets tight, promises are made to be broken.

Puerto Rico’s default Monday on bonds sold by its Public Finance Corp. underscored the risks of debt backed only by a legislature’s pledge to repay. Two days later, Chicago’s Metropolitan Pier and Exposition Authority’s rating was cut to near-junk from AAA by Standard & Poor’s because Illinois hasn’t appropriated the money to pay its bonds amid a stalemate over the budget.

Unlike with general obligations or debt that has a claim to specific revenue, buyers have little recourse if politicians walk away from appropriation bonds, a $197 billion niche of the municipal market. Vadnais Heights, Minnesota, and Menasha, Wisconsin, have already done so. In bankrupt San Bernardino, California, investors may recover one cent on the dollar.

“Appropriation debt is scarier than people want to think it is,” said Matt Dalton, chief executive officer of Rye Brook, New York-based Belle Haven Investments, which manages $3 billion of munis. He said his firm tends to avoid the securities.

Legislators “have to come around the table and appropriate each year, and as you can see, if they’re not on sound financial footing, you’re taking a lot of risk,” he said.

So S&P formerly rated Chicago’s Metropolitan Pier and Exposition Authority higher than it rates the United States of America, then noticed that (1) Illinois policymakers sometimes have difficulty agreeing on spending, and (2) the bonds are already in technical default — and the news story there is about the underlying debt structure? Come on, the super-downgrade is Onion material, not Bloomberg material.

First of all, there is not $197 billion of appropriation-supported debt outstanding. To arrive at this figure, one would have to count a lot of debt that does not truly fit into this category. I’d submit to you that some of the examples cited in the article do not count.

Second, even if that were true, defaults by issuers like Vadnais Heights — population 12,302 — and Menasha, Wisconsin — population 15,144 — are not exactly terrifying precedents for the marketplace. Both examples are more cautionary tales on economic development projects than an indictment of appropriation-supported debt in general.

One of the key considerations for investors in Vadnais Heights, which defaulted in 2012, should have been that bond proceeds were being used to finance the construction of a controversial, non-essential project. These factors used to be central to credit analysis, and now no one cites them. Why is that?

Did a city with a population slightly over 12,000 need a giant sports complex with two ice rinks and Minnesota’s second-tallest dome, covering a 100,000 square-foot turf field? (“If you build it, they will come. If you build it, they will come…”) I don’t know about you, but I am not particularly surprised by this default and more than a little reluctant to lump this example in the projects governments typically finance with general fund dollars.

Menasha defaulted on bond anticipation notes that the city issued in 2005 and 2006. Bond anticipation notes are a form of short-term debt that are supposed to be taken out through the later issuance of long-term debt. The primary investors were a Maryland retirement complex, a life insurance company, and a community bank.

These investors ultimately ended up with higher recoveries on their notes than the general obligation bond investors Chappatta cites later in the article. So Menasha is actually a counterexample, not an example, of Chappatta’s argument here.

Menasha’s largest employers are paper mills. In order to help the large corporations that operate these mills cut costs, the city decided to convert its utility from an electric plant to an industrial steam plant, thereby reducing the cost of power. (“If you help us cut costs, we will not downsize employment and your little village will survive.” Probably how discussions went, no?) The project was supposed to cost $12.7 million and be completed within one year, but was plagued by cost overruns. The city ended up borrowing $40 million, including general obligation bonds and the notes. Given the scope of cost overruns and mechanical problems with the facility, the city ended up not issuing bonds to take out the notes and closing the plant.

The story does not end here though. Menasha’s creditors sued the city and the case wound its way through the courts. It was eventually settled for $17.5 million of the $22.8 million owed. This is more (percentage-wise) than the voter-approved, unlimited tax general obligation bondholders in Detroit received in bankruptcy.

Incidentally, the settlement was funded through a sale-leaseback agreement with WPPI Energy and Menasha Utilities, not city tax dollars. Detroit had similar options, but Jones Day decided to turn the city into a fee machine rather than seriously consider them.

I could cherry-pick revenue bond deals that did not end well for investors (excluding hedge funds that specialize in distressed debt) as Chappatta has done with appropriation-supported debt in this article. Remember Jefferson County, Alabama? Or Harrisburg, Pennsylvania? These were not exactly small transactions, either. And yet folks like Mr. Dalton aren’t coming out of the woodwork swearing off revenue bonds forever. Why is that?

I also find it interesting that Chappatta uses San Bernardino considering paying its pension bondholders a penny on the dollar as an example.Stockton similarly wiped out Franklin in its bankruptcy too — a secured creditor.

“Legal protection” doesn’t mean what you think it means

It makes zero sense to make sweeping generalizations about the relative safety of debt based on the specific kind of pledge involved alone. And yet this is how the majority of folks in the municipal bond market think. What Chappatta is suggesting in this article fits the traditional way of looking at risk that old-school municipal bond attorneys frequently lay out in the press. But they are not the ones litigating Chapter 9 cases now, and that is a point that cannot be emphasized enough. There is a new landscape and even general obligation bonds involve very real risks in some distressed situations.

An investor’s first question should be “what kind of legal regime am I operating under here?” The legal and political contexts in which your bonds are issued are what make your rights and remedies relevant. If there is a lesson to be taken from the recent spate of financially challenged municipalities, this is it. General obligation debt is not the same security in different states depending on whether those states authorize Chapter 9, where it is treated as unsecured debt absent a statutory lien (which very few of the states that authorize bankruptcy offer) and may be adjusted.

Chapter 9 cases are now being litigated by corporate restructuring attorneys who have far less sentimental regard for general obligation debt than attorneys who specialize in municipal debt and, it seems from the article, many institutional money managers. The reason recent bankruptcies have been devastating for bondholders is that corporate restructuring attorneys will bulldoze anything and everything, and the law permits this. By the way, many of these attorneys are advising Puerto Rico’s government and Atlantic City.

There is also a big difference here in whether you are talking about debt issued by a state or local government, both with respect to structure and risk. Local governments, as political subdivisions of state governments, are constrained by state law regarding how much debt they can assume and how much and which taxes they may levy. Tax revenue may also feed into the same fund along with funds for operations, pensions, etc. (Chicago is an excellent example of this, in fact.) This can make a tremendous difference in the amount of risk involved.

This is where it gets strange to swear off a class of debt for its own sake, especially given the diversity of appropriation-supported debt. Which would you rather have: (1) Appropriation-supported debt financing an essential government property issued by a community that has broad ability to raise revenues or (2) general obligation bonds issued by a community subject to a tax cap? (1) Appropriation-supported debt in a state that does not authorize Chapter 9, where a potential default can be litigated aggressively in state courts, or (2) general obligation debt in a state that does authorize Chapter 9, where the borrower has access to a convenient legal shield? (1) Appropriation-supported debt issued by a government with no pension problems or (2) general obligation debt in a state where pensions are severely underfunded but have constitutional protections? It’s a big, big universe.

For many state and local governments that issue appropriation-supported debt, there is solid political will to repay the debt, hence the scarce defaults. Debt service may even be legally considered a continuing appropriation and thus insulated from late budget adoption (contra the Illinois example in the article). When budgets are not adopted for some reason or another, there are usually legal determinations about which government expenditures do and do not continue to be paid, and these are not the same thing from government to government. (Except in Illinois, where they aren’t even the same from budget cycle to budget cycle — ha.)

Using an appropriation-supported structure as opposed to a general obligation structure is not generally intended to provide a deliberately weaker security from the government’s standpoint. It can also be a political expedient to circumvent a constitutionally or statutorily required referendum. Many governments have extraordinarily well-established debt programs and legal protections that can be pursued in state courts. These could even include liens on a specific piece of property. That may not be the same thing as getting a court order to levy taxes, but it is not an “empty political promise” either. It is a remedy with quantifiable economic value.

How does one know these things exist? Well, relevant state law is usually summarized in offering documents.

Finally, not to point out the obvious, having debt issued under other structures/resolutions is its own incentive to appropriate funds for most borrowers. A good example of this is Rhode Island deciding to stand behindits embarrassing offer of support for the 38 Studios project.

Back to Chappatta’s article:

Detroit’s $18 billion bankruptcy illustrated the differing legal protections for investors. Buyers of its certificates of participation, which had neither the city’s taxing power nor specific revenue streams behind them, recouped 12 cents on the dollar, according to a report this week from Moody’s Investors Service. Holders of its general obligations got six times more.

Chappatta is only citing the unlimited tax general obligation bondholders, who recouped 74 cents on the dollar — and I guarantee you they felt completely and utterly screwed, not delighted, about it. The limited tax general obligation bondholders — who, last time I checked, were still general obligation bondholders — recouped only 34 cents on the dollar. Most municipal bondholders would be horrified by these recoveries.

Syncora, the insurer on the city’s pension certificates, also settled for physical assets in the city that may appreciate over time. Syncora is no stranger to project finance (Syncora acquired the operator of the Detroit-Windsor tunnel from bankruptcy), and made a somewhat clever play here in picking up assets related to those the insurer already held.

Importantly, these recoveries are the products of negotiated settlements between the bondholders and the city, which are both arbitrary and path dependent. Negotiated settlements do not say anything about legal protections. A legal precedent regarding the status of the bonds could only be created through a ruling that is upheld by an appellate court. Even then, it would be difficult for it to translate state-to-state.

Detroit’s bondholders were perversely subordinated to the city’s swap counterparties (!), which were treated as secured debt only because doing so was useful to Jones Day to have a cram down option in negotiating with bondholders and pension beneficiaries. The city could have chosen to challenge the swap counterparties’ status, seeing as how using casino revenues to secure interest rate swaps was not exactly a permissible use of casino revenues under state law. Then bondholders’ recoveries theoretically could have been higher. Such were their legal protections.

Bondholders were also practically subordinated to pension benefits, which were treated as another form of unsecured debt but arbitrarily had much higher recoveries. Such were their legal protections.

Had any of the parties settled earlier than the others, they could have received a higher amount. Had the grand bargain not have been involved, they could have received a higher amount. Had the judge not been a populist that often sided with pension beneficiaries, bondholders could have received a higher amount. The city could and often did play all groups of bondholders off of each other in the negotiation process. Theoretically, all bondholders could have been wiped out and stakeholders could still be fighting.

The bonds were the first to take the hit in Puerto Rico, where the government is reeling from $72 billion of obligations. Faced with a growing cash crunch, the legislature didn’t provide funds to cover bonds the commonwealth’s Finance Corp. sold to help keep the government afloat. When $58 million of interest and principal was due Monday, it paid just $628,000.

Let’s be intellectually honest here— Puerto Rico is already not making payments on its general obligation bonds by suspending the set-aside payments that it should be making.

The set-asides were the reason Puerto Rico’s $378 million July 1st general obligation payment was made as scheduled even though the territory was in the middle of a liquidity crisis — the trustee already had the funds on hand to make the payment.

From Puerto Rico’s perspective, resources due under both financial commitments have been diverted to operate and bondholder claims have been politically subordinated to services.

From the Bond Buyer:

Until early July a 1976 law required Puerto Rico’s government to equally set aside its interest and principal general obligation debt service on a monthly basis. The 1976 law required the Puerto Rico to set aside each month one sixth of all interest coming due in the coming six months. It also required the commonwealth to set aside monthly one twelfth of all principal coming due in the coming 12 months. The funds were set aside in a redemption fund.

In early July Gov. Alejandro García Padilla signed a measure annulling the 1976 law.

On Monday Puerto Rico posted a statement on the Electronic Municipal Market Access site that said it had “temporarily suspended” the set asides.

When the idea of stopping the set asides surfaced in June, Puerto Rico Representative Rafael Hernández Montañez told The Bond Buyer that Puerto Rico would make its GO payments. Hernández Montañez is the president of the Puerto Rico House of Representatives Treasury and Budget Committee and a member of the Popular Democratic Party with García Padilla.

Since then, Puerto Rican officials have not been as definitive that Jan. 1's $375 million GO payment will be made. After the governor signed the measure allowing the set asides to be suspended, his office sent out a press statement saying that, “the suspension of these deposits does not imply a breach with the bondholders on the date of payment.”

In mid-July Puerto Rico Director of the Office of Management and Budget Luis Cruz Batista told the El Vocero news site that the GO payment was not assured. Earlier this week, Melba Acosta Febo, the president of the Government Development Bank for Puerto Rico told El Vocero that she couldn’t guarantee the GO payment but that the government was working to improve the central government’s liquidity so it could make the payment.

“We recognize the GO’s priority over other debts and its difference from other debts, but we must also recognize that there is a duty of the state to maintain health services, education services, and some security to our people — we must look at those things,” El Vocero quoted her saying. “So it is a part of a negotiating process, in which both parties are agreeing to modify certain terms and in which conditions of some payments are very difficult.”

Puerto Rico’s general obligation bonds do have strong legal protections that range from constitutional provisions to clawing back revenues from other public entities. The wild card for Puerto Rico is the question I said bondholders need to start with, which is “what kind of legal regime am I operating under here?”

Section 8 of Article VI of the Constitution provides that the public debt of the Commonwealth will constitute a first claim on the public resources. A constitutional protection is obviously a very strong legal protection, but it remains unclear what a court would do if general obligation bond payments came into conflict with basic government services or how other constitutional provisions might be interpreted if the Commonwealth used them strategically. This could work out poorly for investors in the event of default (depending upon where they jumped in, of course) or it could work out very well.

So what happens in the event of non-payment? Well, it depends on when the bonds were issued.

Puerto Rico has some general obligation bonds (dated March 17, 2014) that are governed by New York State law. In the event of nonpayment, bondholders can sue the Commonwealth to enforce the provisions of the bonds in New York courts or US federal courts sitting in the borough of Manhattan; the City of New York, New York; Commonwealth courts; or US federal courts sitting in San Juan, Puerto Rico. Puerto Rico has also waived sovereign immunity for these bonds. Of course, the fact that bondholders have the flexibility to decide which court to bring suit in doesn’t mean the court is required to hear the suit. Fun, right?

Holders of Puerto Rico’s other general obligation debt would have to bring suit in San Juan. These bonds could potentially be subject to sovereign immunity, which restricts the scope of legal actions and monetary damages involved. Puerto Rico’s general obligation bonds are not subject to acceleration, a remedy found in other forms of debt.

The lack of a clear legal regime is further complicated by the fact that Puerto Rico owes money to a lot stakeholders, who will all be asserting their rights in a web of suits against the government and each other across jurisdictions. This is not a place devoid of investment opportunity, but to believe there is a clear hierarchy among all of these stakeholders in terms of potential recoveries is naive.

Political risk works both for and against bondholders. In situations like 38 Studios, there was intense political pressure to make good on a promise to appropriate funds for debt used to finance a bad project just to maintain the state’s good credit — even though those funds could have been directed toward more responsible priorities from taxpayers’ perspective.

Where negative political risk exists, it can be divided into soft and hard risks. Soft risks are usually one-off events, and I daresay disproportionately tied to economic development. You have a government where officials long gone endorsed a project that didn’t perform and now current officials do not want to support it. These risks are actually pretty easy to predict. If a government has to finance a project that should be producing revenue on its own, it is probably a bad deal. You’re welcome.

This is a story about supply and demand and bad public policy. Let’s start with the bad public policy.

The Impact of the Dodd-Frank Act on Municipal Bonds

Among the many changes imposed upon the capital markets by the Dodd-Frank Act is the requirement that the rating agencies use a uniform scale in assigning ratings to municipal and corporate securities. The litmus test for demonstrating this requirement has been met is whether similarly rated municipal and corporate securities have similar historical default rates.

Currently, municipal and corporate securities with similar ratings have very different historical default rates. According to the Bond Buyer:

Studies published by Moody’s Investors Service and Standard & Poor’s show that the default rates of munis 10 years after being rated BBB are lower than the default rates of corporates 10 years after being rated AAA.

According to a Moody’s report “US Municipal Bond Defaults and Recoveries, 1970–2013,” the 10-year cumulative default rate for munis rated Baa1, Baa2 or Baa3 was 0.32%. According to the same study, the rate for Aaa corporates was 0.49%.

S&P Wednesday released its latest default report. The study by analyst Lawrence Witte found that in the 10 years after municipal bonds were rated BBB-plus, BBB, or BBB-minus, 0.42% defaulted. A March 2014 study by S&P managing director Diane Vazza and several others found that in the 10 years after corporate bonds were rated AAA, 0.87% defaulted.

This is a meaningful discrepancy for state and local governments and the nonprofit organizations that borrow money in the municipal bond market to finance large capital construction projects. Although the rating agencies received considerable criticism for inaccuracy in the wake of the financial crisis, municipal yields remain highly correlated with ratings strata.

Why is this? The municipal bond market has many sectors, including state and local governments with a variety of pledged revenues, transportation, water and sewer infrastructure, higher education, charter schools, health care, mortgages, multifamily housing projects, student loans, tobacco settlement securitizations, industrial revenue bonds, and many more. Municipal bonds can also involve bespoke borrowing structures. Many bondholders are unsophisticated retail investors and even some of the so-called sophisticated investors are unsophisticated in reality. It will always be convenient for this crowd to outsource due diligence and surveillance. Ratings also commodify municipal bonds, which makes investment bankers’ lives easier.

So Dodd-Frank’s notion of a uniform rating scale seems intuitive in this context, right? If investors are going to continue to outsource their responsibilities, and if contracts are going to continue to be tethered to ratings events, then there might as well be uniform consequences. Taxpayers should not have arbitrarily higher interest costs than corporate borrowers.

This rule probably would have made sense if the municipal bond market was stuck in 2006 for all eternity and governments were still providing captive demand for the vestigial ratings arbitrage scheme known as the bond insurance industry. (For background, see What is the Value of Bond Insurance?) Alas, it is 2015 and the rule makes zero sense.

While “fairness” across borrowers seems like a social good, it is important to understand the limitations of historical default data as an indicator of “fairness.” Under some circumstances, Dodd-Frank could be seen as penalizing the rating agencies for, and economically precluding them from, being forward-looking. The perverse psychology associated with this transition was recently revealed through some market observers’ reactions to Moody’s cutting Chicago credits to speculative grade.

Credit fundamentals can and do shift, which is what is happening in the municipal bond market right now. Under Dodd-Frank, the rating agencies are forced to compare a municipal bond today to a corporate bond ten years ago. It doesn’t even make sense to compare a municipal bond today to a municipal bond ten years ago.

The Impact of Supply and Demand on Bond Structures

When Dodd-Frank was enacted in 2010, the supply of new municipal bonds hitting the market was peaking. Supply technically would have peaked earlier were it not for the Build America Bond program authorized under the American Recovery and Reinvestment Act of 2009. (For background, seeWhy Build America Bonds Won’t be Returning Anytime Soon.)

Then the issuance of new municipal bonds fell off a cliff. The Great Recession drained state and local coffers. (Tax collections tend to lag the economy.) Governments prioritized restoring public services over undertaking capital projects and political movements left policymakers reluctant to authorize new borrowing. Record low interest rates prompted municipal borrowers to refund debt at a rapid pace.

For all of these reasons, the municipal market has experienced net negative supply. More bonds were being replaced or retired than were being issued. In 2014, states’ total net tax-supported debt outstanding decreased for the first time in 28 years. Municipal bonds became scarce and consequently more valuable. Basically all new bond issues have been oversubscribed many times, meaning that there is more investor interest than there are bonds available for sale. There is too much cash chasing too few bonds.

Municipal borrowers realize that they have the advantage over investors now. One way this advantage has manifested itself is that borrowers are shedding various bondholder protections that have historically been included in contracts, even dating back to the Great Depression. Borrowers have no reason to provide any covenants, liens, or forms of credit enhancement that limit their operating flexibility or cost money when borrowers will have strong demand without them.

“Covenant-lite” debt typically becomes popular when money is cheap and market standards decline. This phenomenon is relatively new to the municipal bond market, however, and this is partly because municipal bonds had not been rated on the same scale as corporate bonds. For municipal bonds, the historical disparity in ratings created an artificial need for credit enhancement irrespective of where the rest of the market was in the credit cycle.

This change is significant because the municipal bond market has traditionally been known for its remote default risk. It is worth asking whether material changes in bondholder protections will result in default rates that are noticeably higher than they have been in the past. Some might argue that these bondholder protections were never necessary and that the market is finding a natural equilibrium. However, if that is not true, then our regulatory framework might present an obstacle for evaluating the introduction of new risks into the marketplace. It might even be paradoxically driving the new risks.

Anecdotal Evidence of this Phenomenon

The high-yield space illustrates the magnitude of this shift. FromBloomberg’s Municipal Market Brief (June 25, 2015):

The inflow into high-yield muni funds since the start of last year represented about 18 percent of total assets yesterday. The funds can include some lower-rated investment-grade securities.

Even as $1.2 billion was withdrawn from the funds in May and June while the market braces for the Federal Reserve to raise interest rates, covenants aren’t being strengthened …

Low- and non-rated borrowers have been jettisoning standard security provisions like mortgage liens, which give bondholders the right to foreclose on property in the event of default, and reserves the issuers can tap if they don’t have enough cash to service the debt …

Last month, fund managers clamored to get a piece of Albert Einstein Healthcare Networks’s $453.5 million bond deal even though the system had 99.5 days of cash and also didn’t include a reserve as security. Money managers placed more than three times as many orders as bonds available, allowing the healthcare provider to cut yields, said Chief Financial Officer David Ertel.

A debt service reserve fund is a portion of bond proceeds (that is, the money investors are lending to the borrower) that is set aside to make debt service payments in the event that pledged revenues are insufficient. Philosophically — since this is bondholders’ own money — a debt service reserve is equivalent to the investor partially insuring her own holdings from default in exchange for the yield on that amount (which represents a quasi-insurance premium). As the issue approaches its final maturity, the funds are applied to the amount outstanding.

For the above example, the absence of bondholder protections was not penalized. The Einstein deal came in 130 basis points above the benchmark rate, roughly the same as the similarly rated Boston Medical Center deal at 132 basis points above benchmark that involved a debt service reserve fund.

Another example provided is Peninsula Regional Medical Center in Maryland, which issued $127 million of bonds in January. The bond contract included language permitting the borrower to pledge accounts receivable for other loans, effectively subordinating bondholders. This is quite a development in the healthcare sector — which even historically has been among the most prone to default.

The Chicago Tribune published a self-recommending article over the weekend about Chicago Public Schools’ pending cash crisis. Information in the article — which was provided by Ernst & Young— directly contradicts claims officials made in investor presentations and offering documents when the school system issued bonds only two months ago:

To pay the huge tab owed to the Chicago Teachers’ Pension Fund, make payroll and cover other costs, the nation’s third-largest school system would have to drain its checking account, max out its line of credit and burn cash set aside to pay other debts, according to outlines of school finances presented in city documents.

If CPS resorts to these steps, the district has no clear way to close next school year’s $1 billion budget gap without teacher layoffs, increased class sizes and perhaps even a late opening of schools …

Even if CPS wins concessions from the City Council, state lawmakers and the [Chicago Teachers Union], the district won’t be able to close its annual $1 billion budget gap, according to the May 22 report by Ernst & Young, which spent four weeks meeting with school finance officials and analyzing budget documents. The report shows that even with a capital improvement tax, a separate, even-larger property tax increase, additional state aid, increased state funding of teachers pensions, concessions from the CTU and $150 million worth of budget cuts, CPS would still face an annual $350 million shortfall.

According to the Chicago Tribune, the investor presentation indicated that CPS would have $227 million left in its operating account at the end of the fiscal year, but Ernst & Young estimates that the school system could be out of cash as early as this summer. Prior to issuing the bonds, the Chicago Board of Education (CBOE) replaced Moody’s with Kroll for the sake of having three rating agencies willing to give the bonds investment-grade ratings.

This is one of several examples of nonsensical divergence in ratings of late. S&P rated the City of Chicago’s sales tax-supported debt AAA, when Moody’s has it at junk. I agree with the latter assessment, seeing as how the revenues are not legally segregated from the city’s general operating fund. This is the most basic bondholder protection, and it is absent. For an issuer in Chicago’s financial situation, this should be a major concern.

Another strange example is Wayne County, Michigan, which is currently under review by Michigan’s Treasury Department for financial emergency status. Wayne County could potentially go from financial emergency to consent agreement to emergency manager to Chapter 9 within a short period of time — for Detroit, this process took less than a year and was aggressively driven by the state. Per Municipal Market Analytics, S&P gave Wayne County an investment grade short-term rating (SP-1), while all three rating agencies have junk ratings on the county’s long-term debt. What’s up with that?

Another interesting tidbit from the Tribune article is that CPS has reportedly paid off some of the $228 million of swap termination payments it owes to bank counterparties, which CPS used in conjunction with previous bond issues. CPS declined to disclose to the Tribune the dollar amount of those payments because it could potentially jeopardize on-going negotiations.

These developments should concern CBOE bondholders for two reasons. First, the swap termination payments are likely to eliminate the school system’s $174 million debt service stabilization fund in its entirety. The CBOE had established this fund as a buffer to make debt service payments in the event that budgeted revenues do not materialize. However, those funds were not specifically pledged for the repayment of CBOE’s bonds and the board has never covenanted to maintain the fund at any particular level. Now the fund is being used as a source of liquidity for the swaps. This is a management decision that pits bondholders’ interests against those of its swap counterparties, and bondholders have lost.

(Pledges, covenants, legal segregation — these things matter!)

The second reason bondholders should be concerned is that, in some cases at least, the variable rate bonds they are holding do not involve a credit facility typically provided by a commercial bank. This means investors are directly exposed to CBOE’s credit in the event of a failed remarketing.

As I described in my earlier essay about ratings shopping, The Potential Insolvency of Chicago Public Schools and the Stickiness of Credit Ratings, CBOE bonds have a bizarre auction rate securities-ish structure. And — for no reason that I can figure out—CBOE continues to use this structure even though it has unraveled spectacularly.

Typically, when an issuer uses a variable rate structure, the issuer obtains some sort of credit facility to backstop the debt in the event of a failed remarketing. A failed remarketing occurs when there is not sufficient demand for the bonds. When a credit facility is in place, the debt does not carry the rating of the issuer. Instead, the debt carries the rating of the financial institution providing the credit facility.

I am assuming that CBOE could not locate a commercial bank that was willing to backstop its debt, and thus the liquidity/credit risk has been transferred to investors. CBOE has to pay penalty interest rates upwards of 9% on the debt, which only exacerbates its financial problems and decreases its ability to repay bondholders.

According to the Tribune, the use of this debt structure and the accompanying swaps was pushed by CBOE President David Vitale, a former vice-chairman of Bank One. Vitale will likewise be providing advice on how the school system will address a situation where it will be left with a large deficit even under the best of circumstances.

I expect that Chicago Public Schools will prove my thesis that a governmental entity can afford dysfunction or an unconventional debt structure, but it can’t afford both. An unconventional debt structure will require the use of resources that would otherwise prevent dysfunction from turning into a financial catastrophe.

Update 6/23/15: Yvette Shields at the Bond Buyer told me that Chicago Public Schools is now seeking a $200 million line of credit to address its deficit this year and $935 million to address its deficit next year. The school system’s deficit is expected to balloon by almost $1 billion more at the end of fiscal year 2016.

Photo: Reconstruction Finance Corporation clerks, Library of Congress

Back in January, I wrote a piece about how Jeffrey Gundlach was wrong about Puerto Rico’s default risk — “I don’t care if [Puerto Rico] securities flop around,” Gundlach said, they will find a way to pay.

The territory’s power utility, the Puerto Rico Electric Power Authority (PREPA) — which is already in forbearance for tapping its debt service reserve funds — is expected to exhaust those funds and have a monetary default on July 1st. According to the trustee, after reserves are drawn down (and trustee fees are paid), PREPA will be short $150 million. The monolines have wrapped most of the maturities coming due that day and will have to cough up the cash for investors.

According to Bloomberg:

Investors and bond insurers have tentatively agreed on a proposal for how they’d be willing to alter terms on the authority’s existing securities, according to a person with direct knowledge of the discussions.

Suspending principal and interest payments or asking bondholders to reduce the amount owed would help Prepa as it seeks to modernize a system that relies mostly on petroleum to produce electricity.

Obligations of the utility that mature in July 2042 traded Thursday at an average price of 54.6 cents on the dollar, the highest since June 2 …

Regardless of how PREPA’s commitments shift, investors in all Puerto Rico securities will be negotiating difficult terrain from this point on. This will likely be the first of many creative revisions, which will all ultimately amount to bondholders not receiving what they were originally promised.

I heard from a congressional staffer yesterday that Representative Jeffrey Duncan sent out a “Dear Colleague” letter asking for a financial control board for Puerto Rico to address its debt crisis. Establishing a federal financial control board for Puerto Rico could potentially be the best option for residents. I have started thinking through the mechanics of how such a board might work and the history of emergency interventions in the municipal market —no, this is not an unprecedented event.

Objectively speaking, any Chapter 9 legislation will be ineffective when applied to Puerto Rico

Representative Duncan’s letter was written in response to H.R. 870, the “Puerto Rico Chapter 9 Uniformity Act of 2015.” This legislation wasrecently discussed in a hearing of the House Judiciary Committee’s Subcommittee on Regulatory Reform, Commercial and Antitrust Law.

H.R. 870 has proven to be quite controversial and bondholders have spared no expense in lobbying policymakers to scrap it. The legislation should be scrapped, but not necessarily for moral reasons. As drafted, the legislation would be ineffective.

All H.R. 870 does is redefine “state” within Chapter 9 of the federal bankruptcy code to include Puerto Rico. Under Chapter 9, states may authorize their political subdivisions (municipalities) to file for bankruptcy. States themselves — which are considered sovereign entities under the U.S. Constitution — are not eligible to file for bankruptcy. The logic behind the legislation is that if Puerto Rico is considered a state, its general obligation debt would be shielded from bankruptcy. The territory’s public corporations, however, could file for bankruptcy and adjust their debts. Puerto Rico recently enacted legislation that would have accomplished the same thing — to the horror of municipal market participants, who interpreted the measure as a diminution of the government’s willingness to repay what it owed. The law was shot down when bondholders challenged it in court.

Rarely, however, does legislation that is one line long accomplish what you want. The problem with H.R. 870 is that the bonds issued by Puerto Rico’s public corporations are of a variety that other provisions in Chapter 9 protect from being impaired — that is, “special revenue” debt. Furthermore, this is an error that Congress should not resolve through amending those provisions, which were added to Chapter 9 through the landmark 1988 Amendments. The status of special revenue debt is important to the municipal bond market and municipalities’ ability to provide essential government services.

The pledge of what are considered special revenues survives bankruptcy. Special revenues are defined as:

(A) Receipts derived from the ownership, operation, and disposition of projects and systems of the debtor that are primarily used or intended to be used primarily to provide transportation, utility, or other services, including the proceeds of borrowings to finance the projects or systems;

(B) Special excise taxes imposed on particular activities or transactions;

(C) Incremental tax receipts from the benefited area in the case of tax increment financing;

(D) Other revenues or receipts derived from particular functions of the debtor, whether or not the debtor has other functions; or

(E) Taxes specifically levied to finance one or more projects or systems, excluding receipts from general property, sale, or income taxes (other than tax increment financing) levied to finance the general purpose of the debtor.

(James Spiotto, from The Handbook of Municipal Bonds)

If minutiae excite you, I’d also recommend this white paper on special revenue debt from Waller Lansden — the law firm that represented the trustee for the sewer warrants in Jefferson County’s bankruptcy and the trustee for water and sewer bondholders in Detroit.

Basically, if H.R. 870 is enacted, Puerto Rico still wouldn’t be able to adjust its debts.

It should likewise be noted that proponents of H.R. 870 are not the only ones with weak arguments. One of the claims frequently made by opponents is that Congress cannot (or should not) pass legislation that would retroactively alter the terms of outstanding debt. This is absolutely possible and how Chapter 9 came into existence in the first place.

Returning to Spiotto:

Prior to 1934, federal bankruptcy legislation did not provide a mechanism for municipal bankruptcy, insolvency, or debt adjustment. During the period 1929 through 1937, there were over 4,700 defaults by governmental bodies in the payment of their obligations. In 1934, the House and the Senate Judiciary Committees estimated that there were over 1,000 municipalities in default on their bonds.

Units of local government were dependent upon property tax. During the Depression, there was widespread nonpayment of such taxes. Bondholders brought cases for accountings, secured judgments and obtained writs of mandamus for levies of further taxes. The first municipal debt provisions of the Bankruptcy Act of 1898 as amended from time to time (the Bankruptcy Act) were enacted as emergency legislation for the relief of such municipalities …

Legislation permitting the adjustment of debt does not arise from thought experiments. It is born out of desperation and relates to existing contracts.

It is also difficult to make a moral argument in support of Puerto Rico’s bondholders. Investors have never purchased Puerto Rico’s debt based on the territory’s credit fundamentals, although some investors with large exposures are now trying to suggest otherwise. They purchased the debt because Puerto Rico’s bonds are the only bonds in the market (of any real size) that are triple exempt. And recently, Puerto Rico has attracted a surfeit of market tourists, who could care less about Puerto Rico’s fate. They are reaching for yield in a yield-less environment, where there is a lack of distressed debt opportunities in corporate bonds.

It is not the federal government’s responsibility to redeem opportunistic investment decisions or to reinsure the monolines. Fortunately, the monolines have taken to citing Puerto Rico and Detroit as cases that demonstrate the value of bond insurance. Who would have imagined that, following the financial crisis, the monolines would be advertising that that they continue to wrap toxic debt? But I digress.

Political hurdles to overcome

It is inevitable in the bond market that some governments will over-borrow and some investors will over-lend. Neither side is particularly virtuous and there has to be a process for resolving the conflicts that arise.

The practical wisdom behind Chapter 9 is the observation that if a government cannot honor its obligations, it is best for everyone involved if it at least does so in a fair and organized manner. A government’s raison d’être is to provide services to residents in perpetuity and it cannot do so effectively if it is fending off lawsuits from every direction. Value is also lost to investors as such situations suck resources from the government that could otherwise be used to offset its debts. Even if Chapter 9 cannot successfully be applied to Puerto Rico, this practical wisdom still abides.

Puerto Rico has two major political hurdles in the mainland U.S. to overcome as it advocates for such a process. The first is that the territory is literally no one’s constituency. Puerto Rico officials have tried to impress upon policymakers the gravity of their situation to little avail. Three congressmen appeared for the initial hearing on H.R. 870.

Some token technocrats from the U.S. Treasury Department have been dispatched to Puerto Rico to provide “technical advice” — but considering that the Puerto Rico government is contemplating raiding its state insurance funds and legalizing black market slot machines for operating cash, they probably aren’t getting very far on the advice front. Of course, I jest — everyone knows “technical advice” is code for “we aren’t giving you a dime, but we need to look like we did something.”

Federal policymakers likely will care about Puerto Rico if there is a string of major defaults. Such defaults might not have much of an impact on the municipal market as a whole, although they will be the largest the market has ever seen. However, large defaults could potentially produce a humanitarian crisis if they upend the territory’s financial system and interrupt the government’s ability to provide essential services. By then, providing something more than a symbolic gesture and avoiding a legacy of shame will be considerably more expensive. Policymakers can moralize all they want, but it’s remarkable how many man-made crises derive from simply waiting.

Puerto Rico’s other political hurdle is that it has an old school debt crisis but schizophrenic legal status/organization, which precludes the obvious solutions (like Chapter 9) and makes Puerto Rico easy to bully in court. Granting Puerto Rico statehood — and the federal resources that brings — could potentially be a good option in the future, but the territory requires immediate financial relief.

So what options does Puerto Rico have left?

Establishing a federal financial control board

This is not the first time the United States has had to grapple with a financially distressed entity that is stuck in not-quite-a-state limbo. Something similar happened with Washington D.C. in the 1990s. The federal government responded with the Revitalization Act, which was folded into omnibus legislation along with the Balanced Budget Act of 1997. By the time the legislation was enacted, it had broad bipartisan support.

From Jon Bouker of Arent Fox (Brookings):

When Congress granted home rule to the District of Columbia in 1973, Rep. Charles C. Diggs, Jr., then chair of the House D.C. Committee, declared that Washington’s residents had become “masters of their own fate.” …

However, a mere two decades later, the District’s limited home rule was in crisis. As the District government’s financial position reached its nadir in the mid-1990s, residents’ frustration and anger mounted as the District was unable to deliver efficiently the most basic services to its citizens, and the city’s congressional overseers began calling for a partial or even complete elimination of home rule.

After enjoying relative financial stability for most of the 1980s, the District began operating at a deficit in 1994 and by 1995 the accumulated deficit had ballooned to $722 million. To make matters worse, Wall Street dropped the District’s bond ratings to “junk” levels, prompting Moody’s to brand them as risky and “speculative.” As a result, the city was unable to pay its vendors, to render basic services, or to obtain a simple line of credit. District residents, tired of dealing with ineffective and inefficient services, underachieving schools, and high crime rates, fled to Maryland and Virginia suburbs in droves — 53,000 District residents, representing 22,000 households, left between 1990 and 1995. This flight contributed to the erosion of the District’s tax base and exacerbated shortfalls. It was a vicious cycle that was driving the city toward insolvency.

The Revitalization Act established a financial control board for D.C. with virtually unlimited powers and clear financial and economic objectives. The control board essentially replaced the local government until its finances stabilized.

Could a financial control board improve Puerto Rico’s situation? Is a highly polarized Congress capable of assembling a financial control board that is skilled, independent, and has the territory’s best interests in mind? Would a highly polarized Puerto Rico accept it? Financial control boards have historically only been utilized for local governments with far smaller debt burdens and far less complex organizational issues than Puerto Rico — can the Revitalization Act model be scaled?

Respectfully restoring accountability

It is possible that the federal government could create a financial control board without the consent of Puerto Rico’s government. If policymakers were to proceed down this path, however, it would be best to make any legislation contingent upon the Puerto Rico government enacting companion authorizing legislation because: (1) this is respectful; (2) such a provision would mirror the discretion given to states in authorizing municipal bankruptcy; and (3) the reform process will be a lot easier with the government’s cooperation than without it.

Under the financial control board, Puerto Rico would temporarily forfeit control over its finances, but it would gain financial aid contingent upon meeting specific economic and operational milestones. The control board would be able to represent the territory in court with creditors, if necessary, with the weight of the federal government and economic necessity behind it.

Providing affordable liquidity

The first step to curing Puerto Rico’s problems is to provide the territory with affordable liquidity. Obviously the political constraint here is that most Americans have a visceral reaction to the idea of a bailout. Having already covered Chapter 9, another Depression-era response to fiscal crisis provides useful guidance — when Arkansas defaulted on its highway debt in 1933 — which is to structure assistance as a loan.

A history lesson from Joe Mysak:

When Arkansas defaulted on its bonds in 1933, the politicians and investors talked about the same things we would talk about today. The state blamed underwriters for allowing it to sell too many bonds. Investors compared the willingness to repay debt with the ability to pay, and weighed the advantages of bonds backed by a pledge of taxing powers to those secured by a specific source of revenue …

The municipal market was a much different place in the 1920s and 1930s. States and localities borrowed about $1 billion in long-term debt each year … credit research was in its infancy, and the practice of bond law was only about 50 years old. The 10-year bond call wasn’t as common as it is today, and there were no prohibitions on the number of times issuers could refund their bonds. The market was less codified and more contentious …

“We have a state ranking 46th in per-capita wealth in 1929, ranking first in per-capita indebtedness,” was how state Senator Lee Reaves summed up the matter in a 1943 article for the Arkansas Historical Quarterly. “Under the best of circumstances it would have been difficult to meet payments on the mounting debt.”

Arkansas ended up with a debt burden of this magnitude when it assumed the debt of its many local road districts in 1927 to prevent them from defaulting. As the state’s borrowing costs climbed, the Arkansas General Assembly had the genius idea essentially to impair the debt through a refunding transaction and replaced bondholders’ first lien on automobile and gasoline taxes with … a general obligation pledge.

As you can imagine, the full faith and credit of a government that already could not service its debt was not what bondholders had in mind. (Hedge funds didn’t exist back then.) So they went to court, received an injunction blocking the use of said taxes for anything besides repaying the outstanding debt. And so the state’s financial woes persisted, albeit with some short-lived pockets of optimism.

Back to Mysak:

On April 1, 1941, $90.1 million of outstanding highway bonds was callable, an additional $45 million was callable on July 1. The state made plans for another, this time uncontested, refunding.

A syndicate of 250 banks said it would bid on the new Arkansas refunding bonds, in conjunction with the Reconstruction Finance Corporation, a new entrant in the municipal market and a creation of Herbert Hoover’s administration.

On February 27, 1941, to Wall Street’s shock, the RFC bought the entire issue single-handed.

Pause for a moment and consider this in the context of Puerto Rico. The federal government provided assistance to an issuer that represented over one-tenth of the market at the time. This is equivalent to multiple Puerto Ricos now.

The Reconstruction Finance Corporation’s intervention provided financing to Arkansas when the state would have otherwise had to pay punitive market interest rates, and it was structured to accomplish a variety of federal interventions. The RFC was modeled on the War Finance Corporation of World War I and kept in place from 1932 to 1957. (Kind of difficult to wrap your head around a Depression-era agency surviving until Elvis Presley, eh?) It supported everything from state and local governments to railroad companies.

Anyway, according to Mysak, “the RFC bid, which averaged 3.2 percent, saved Arkansas $28 million over the life of the bonds. The [RFC] later sold the securities to Wall Street banks at a profit of $4 million.” This strategy of structuring emergency government assistance as a loan was repeated with Washington, D.C. and with other local governments that have received state-level assistance.

When municipal bonds are involved, it is the U.S. Treasury — and not the Federal Reserve — that has to serve as lender of last resort. The Federal Reserve is only authorized to pick up short-term municipal debt, and financial support for Puerto Rico would likely need to be amortized over the medium term. This isn’t any different than it was in 1932 — in fact, the independent RFC was the brainchild of Eugene Meyer, who lead the Federal Reserve at the time.

A possible solution then would be for policymakers to establish and capitalize an independent agency for the purpose of purchasing notes from Puerto Rico’s government as a bridge to restructuring the territory’s overall debt with federal oversight. The debt could be refinanced later if desired.

Untangling the government’s existing liabilities

One of the advantages of Chapter 9 is that it allows a government to address its liabilities holistically, rather than one-by-one in a process that could drag out for several years. Something similar has to occur for Puerto Rico’s finances to stabilize.

The first project of Puerto Rico’s financial control board would be to figure out a comprehensive strategy. This will involve evaluating the territory’s tax structure; government operations (efficiently eliminating bloat and corruption); economic performance and the like. No more betting the house on municipal market tourists’ half-baked economic development fantasies. New policies cannot imperil the provision of government services, although they may include privatizing government activities.

This strategic plan will provide a rudimentary framework for evaluating the territory’s liabilities. For the sake of returning the territory to a genuinely structurally balanced budget, the financial control board will likely be in the position of demanding concessions from bondholders. The control board should be able to reverse-engineer approximately what the scope of these concessions needs to be. The ranking of Puerto Rico’s liabilities should probably follow the same requirements as Chapter 9. The federal government has an advantage in negotiating with bondholders in that it controls the tax status of the debt. There are many routes to a restructuring.

Puerto Rico’s unfunded pension liabilities cannot be addressed in the same manner as Washington D.C. Under the Revitalization Act, the federal government assumed the cost of the district’s pension liabilities. The municipal landscape was very different back then than it is now. The district’s financial problems peaked in the late 1990s, when most state and local government pension systems were either adequately funded or over-funded. The precedent doing so would set in the current environment is untenable, both practically and politically. The trade-off between funding pensions and retiring debt would have to be made by the control board.

This risk exists for bondholders with or without the financial control board. It will take strong leadership to broadcast to stakeholders that everyone will recover more through an impartial arbiter than from ad hoc restructurings.

Conditions for returning control to Puerto Rico’s government

As indicated earlier, returning control to Puerto Rico’s government should be conditioned upon the territory meeting certain financial and economic objectives. These could include: (1) a structurally balanced budget and manageable debt position; (2) market access at investment grade rates; (3) an expanding economy; and (4) a restructured government that is capable of stable, continuous management.

Because it demands large-scale paradigm destruction and major shifts in the problems and techniques of normal science, the emergence of new theories is generally preceded by a period of pronounced professional insecurity. As one might expect, that insecurity is generated by the persistent failure of the puzzles of normal science to come out as they should. Failure of existing rules is the prelude to a search for new ones.

— Thomas Kuhn, The Structure of Scientific Revolutions

On May 12th, Moody’s Investors Service fired the shot heard around the world. The credit rating agency downgraded the City of Chicago and bonds issued by related tax districts to speculative (“junk”) status.

In a sense, Moody’s was only validating the bond market’s opinion of the city’s creditworthiness — the bonds had already been trading at junk levels for several months. This should have been a straightforward event for the chattering class to process intellectually. Rating actions tend to lag the market rather than lead it.

Oddly, however, Moody’s downgrade sparked a debate over whether Moody’s was being “fair” to Chicago. And with Chicago attempting to convert a portion of its variable rate debt to a fixed rate tomorrow, this debate has received considerable amounts of publicity. How could Moody’s cut the city to junk when the other rating agencies rate the city so much higher? (That has obviously never happened before in an era of ratings shopping and superdowngrades.) Wouldn't having a diverse economy and large tax base cancel out the costs associated with machine politics? (It’s not like this is Chicago’s third fiscal crisis in the past century.)

This was probably the first instance in the history of the capital markets that a rating agency was accused of having too radical an attitude toward risk. How did we get here, folks?

A period of pronounced professional insecurity

Since the financial crisis, municipal bond market participants have been particularly defensive about the level of credit risk in the market as a whole. Commentary on any financially challenged issuer is reliably qualified with “the municipal market defies generalizations,” “these issuers are outliers,” or “remember that, historically, municipal defaults are small relative to corporate bonds.” But the parade of apologies for an issuer with Chicago’s level of financial dysfunction should signal that things have gone too far.

These observations began as a legitimate response to Meredith Whitney’s extremism. Whitney was never going to be correct — the amount of local defaults she predicted exceeded the amount of local government debt outstanding. It was mathematically impossible. It is still mathematically impossible. The financial media that turned her claims into clickbait have moved on to sensationalizing other sectors.

Five years later, however, many municipal market participants remain locked in an unproductive dialogue with an irrelevant personality. Consequently, they are now blind to the fact that what they are saying is no longer true. The market has more financially challenged issuers than can be counted on one hand. These governments are not outliers. They are a class, with similar characteristics and a universe of risks that differentiate them from other borrowers. And they are not small borrowers. It is a more meaningful trend that bondholders are receiving pennies on the dollar in court than it is that defaults remain rare.

There is a conversation to be had about how politics influences the perception of financial commitments and whether bond structures can further evolve to protect bondholders. If the general obligation pledge — absent a statutory lien, which few states have — lacks teeth in court, why isn’t it obsolete? Why is this bond structure still the foundation for credit analysis? Does the general obligation pledge allow governments to over-commit themselves financially in certain political contexts? I would submit to you that this absolutely the case with Chicago.

None of these issues will be substantively explored so long as market participants remain in “move along, nothing to see here” mode. These are the first fiscal crises where pensions have been a factor at all. In previous fiscal crises, capital markets creditors had the luxury of control. That is no longer a given.

So municipal credit analysis has a lot of growing up to do. Essentially half the market was insured several years ago. As Kuhn noted regarding paradigm shifts in science, there will probably be “a period of pronounced professional insecurity” in the municipal market until new approaches emerge. Until then, there will be a lot of failures.

I will explain my own philosophy regarding financially challenged issuers at the end of this essay.

Rating divergence is actually the status quo — that’s not the story here

Shawn O’Leary at Nuveen Asset Management points out that this divergence is nothing new in Chicago’s Rating Split: Is Moody’s Out of Step?:

Before looking at rating divergence specific to Chicago, it’s helpful to look at the long-term trend of rating divergence between Moody’s Investors Service and Standard and Poor’s. In our opinion, the ratings of the two rating companies have been largely incongruous for several years …

While S&P upgraded 1.01 municipal ratings for each Moody’s upgrade from 2002 through 2007, that ratio ballooned to a whopping 6.66 S&P upgrades for each Moody’s upgrade from 2008 to 2014. And, as noted by the unemployment rate at the time, S&P’s drastic spike in the pace of upgrades occurred at the height of the Great Recession and continued through 2014. This was long after the recession’s effect on municipal finances became apparent to most market participants. While Moody’s pace of upgrades slowed in response to the recession, S&P behaved as though the recession never happened.

We believe this broader context is important to understanding how S&P could rate Chicago’s general obligation pledge at A+ (now A-) while Moody’s rates the same pledge at Ba1. Specific to Chicago, however, we find that not only has the market long treated Chicago’s debt as being at a speculative credit quality — but that it is also well-founded given the credit fundamentals.

Moreover, it is worth noting that the credit rating agencies have published rating methodologies. S&P’s US Local Government General Obligation Ratings Methodology and Assumptions (September 12, 2013 — stashed behind their paywall) includes overriding factors for liquidity and structural imbalance that the rating agency has long decided to ignore for Chicago. Those factors would have put a junk-level ceiling on Chicago’s ratings. In other words, S&P is arbitrarily holding Chicago to different standards than the other local governments the agency rates. And it is obvious why.

What financial risks does Chicago pose to investors?

Let’s examine Chicago’s credit profile and you can decide whether or not the city’s bonds are speculative investments.

Unfunded pension liabilities

The magnitude of Chicago’s unfunded pension liabilities receives considerable attention, and rightfully so. From Nuveen:

Chicago’s combined annual debt and pension costs are substantially higher than any [of the ten largest US cities] when these obligations are indexed to total governmental revenue. Chicago’s fiscal 2015 debt service and annual pension costs account for 44.8% of fiscal 2013 governmental revenue. San Jose is the next closest city at 27.8%. The nine cities other than Chicago averaged 22.4% of revenue.

Last Thursday’s Bloomberg Municipal Market Brief also provided some good comparisons (refer to the StoryChart):

The next 10 years will be the most difficult for Chicago, as current statutes require the city to increase contributions to its four pension plans by 135 percent in 2016, and 8 percent annually through 2021, according to Moody’s. Pension payments will rise 3 percent, on average, through 2026 and around 2 percent through 2032 …

The median per capita aggregate unfunded actuarial accrued pension liability for the largest US cities and Puerto Rico is $3,350. The City of Chicago’s is $7,149.

Most municipal market analysts assume that the city will address its unfunded pension liabilities and relatively high debt burden by increasing residents’ property taxes by nearly 50%.

Chicago officials have been unwilling to raise property taxes for at least a decade. Offering documents indicate that this attitude continues. The city iscurrently in negotiations with its police and fire unions to postpone transitioning from a system of arbitrary contributions to actuarial contributions (i.e., contributions that reflect the true cost of benefits).

If officials lack the political will to raise taxes when their bonds are trading at 300 basis points (3%) over the AAA benchmark, will there ever be a resolution short of insolvency? This is a material risk that should not be shrugged off.

Borrowing money in order to borrow money

Bloomberg also notes that Chicago has the second-highest general obligation debt per capita among US cities at $3,047, following New York City at $5,500.

According to offering documents (available here), the city won’t be able to afford to make debt service payments on its outstanding bonds from available funds until 2019. The city has been borrowing money on a long-term basis to make debt service payments since before the financial crisis:

Since 2007, proceeds from general obligation bonds in the range of $90 million to $170 million per year have been used to make the city’s general obligation debt service. The city expects to use approximately $220 million of proceeds of long-term general obligation bonds to fund general obligation debt service in levy year 2015 for debt service paid in 2016. The city currently plans to eliminate the use of general obligation bonds to pay general obligation debt service by 2019.

As I described at length in my earlier essay, How Chicago Has Used Financial Engineering to Paper Over its Massive Budget Gap, the city has also been using long-term debt to: (1) finance everyday expenses and maintenance; (2) finance judgments and settlements, including police brutality cases and retroactive wage increases and pension contributions for unionized employees; (3) restructure the city’s existing debt to extend the the maturities on its bonds far out into the future, in order to avoid having to pay the debt as it was coming due; and (4) provide slush funds for the city’s 50 alderman to undertake projects in their respective areas (i.e., pork).

State and local governments typically only issue bonds to finance the construction of capital projects — buildings and infrastructure with long useful lives that will benefit residents for generations. Chicago has incurred literally billions of dollars of debt where residents have nothing to show for it.

Excessive reliance on short-term debt

Besides a sharp loss in population (as what happened in Detroit), excessive reliance on short-term debt is a solid indicator of financial stress. Chicago has essentially used its credit lines as permanent source of funding in the sense that they are usually carrying a large balance and have frequently been utilized for non-capital expenditures. The city recently expanded its short-term borrowing program to $1 billion. For the sake of comparison, the city’s general fund operating budget is in the neighborhood of $3.3 billion.

Forbearance agreements

Apparently the irony of assigning an investment grade rating to an issuer that is already in forbearance — i.e., its lenders and counterparties have conditionally agreed to delay declaring events of default and exercising their rights and remedies —h as been lost on the rating agencies, let alone an issuer that has $2.2 billion worth of forbearance agreements for variable rate debt, short-term credit facilities, and interest rate swap agreements. Chicago’s offering documents contain six pages of triggered events of default. It would be understandable if the city simply needed to replace a deal participant, but having to take it all out at (presumably) much higher interest cost in short order?

Absent these forbearance agreements, that $2.2 billion would become due immediately. The city usually has less than $1 billion of liquidity. The municipal bond market has not seen a liquidity problem of this magnitude for a local government borrower since the financial crisis. And S&P calls this situation “short-term interference.”

Although a portion of these products will be addressed by the city’s offering this week, Chicago’s lenders and counterparties have imposed some strict conditions on the remaining amount. The agreements are for a short period of time and if the city’s credit continues to deteriorate, the standstill agreements will end and the remaining amount will come due.

A large and diverse tax base, but…

Many of the analysts arguing that Chicago should still be considered investment grade cite the city’s large and diverse tax base. Chicago is a transportation hub and home to a number of major corporations.

The city’s population has been declining in recent years and only grew by 82 residents last year. That is not a typo. According to the Chicago Tribune:

Chicago’s population grew by only 82 residents last year, giving it the dubious distinction of being the slowest-growing city among the top 10 US cities with one million or more residents.

With a population of 2,722,389 residents as of July 1, 2014, Chicago still easily holds its place as the nation’s third-largest city … But cities on both sides of it are gaining.

New York maintained its ranking as the nation’s largest city, gaining 52,700 residents last year, for a gain of 0.6 percent that pushed its population to 8,491,079. Los Angeles added 30,924 residents, up 0.8 percent and bringing its population to 3,928,864.

Sun Belt cities with more than 1 million residents — places like Houston, San Antonio, Dallas and Phoenix — all continued to see dramatic gains in new residents …

“The boom of Chicago in the 1990s was due to immigration,” said Rob Paral, a Chicago-based demographer who advises nonprofits and community groups. “You take away the catalyst of immigration, and you see what we have. They’re going to different parts of the country, and there much less immigration to the US than there was decades ago.

“Texas, as an example, has been a magnet for a lot of lower-paying jobs and has the benefit of lower housing costs. If you’re making $15 an hour, the difference between making it where a house costs $100,000 and $300,000 is great.”

This last point brings us to property taxes. Some have also pointed out that Chicago has the lowest effective tax rate in Cook County, which means the city can withstand a large tax increase. This is true. According to Bloomberg, Chicago’s effective tax rate on residential property is 1.8% versus Harvey’s 8.9%. On commercial property, Chicago is not the lowest, but its 4.9% effective tax rate is much lower than Harvey’s 15.1%.

Discussing the level of property taxes in absolute terms fails to capture residents’ calculus in deciding on where to live, however. Just because Chicago has a lower tax rate does not mean residents will stay in the city if or when tax rates are increased significantly. In other areas, higher taxes will translate into more government services, better schools, and so on. In Chicago, they will be used to offset the costs associated with meaningless debt and unfunded pensions from a decade of fiscal mismanagement. That’s a huge difference and something to take into account.

The city has few assets left to sell

Chicago has already blown through the reserves it established from the Skyway and lease of its parking meters. It could try to hawk Midway Airport, but that has already failed three times.

The city’s other tax districts have their own problems

The Chicago Board of Education is also heavily indebted and its recent downgrade likewise triggered events of default. These will force the school system to pay penalty interest rates ranging from 9% to 13.5% and make swap termination payments. The board has significant unfunded pension liabilities and a $1 billion deficit.

Bonds are legally and likely politically subordinated to pension benefits

Perhaps the greatest irony of the debate over Moody’s downgrade is the other municipal bond market news it obscured. Last week, with a 6-t0–1 vote, the San Bernardino city council approved a plan of adjustment that would keep pension beneficiaries whole and basically wipe out bondholders. San Bernardino would be the third in a series of bankrupt municipalities to do so.

Some rulings in federal bankruptcy cases suggest that Chapter 9 could potentially be used to adjust pension liabilities. For that to happen, however, the municipality would have to want to adjust its pension liabilities. So far, when capital markets creditors have gone toe-to-toe with pension beneficiaries in court, they have walked away with massive haircuts.

Why is this happening? I see two (largely ignored) things driving outcomes in municipal bankruptcy cases where pensions are involved. The first is that courts neglect to situate claims in larger public policy contexts. Perhaps this is because some of the judges and law firms involved have mostly corporate restructuring experience and do not fully understand how public policy works. Perhaps it is just impossible the way Chapter 9 was drafted. Whatever the reason, the vocabulary of sacrifice in Chapter 9 cases has become quite mangled. The second is that Chapter 9 provides subtle opportunities for political rebalancing in regions where machine politics prevails. Let’s discuss these in turn.

The treatment of other post-employment benefits (i.e. health care) has been a land mine for capital markets creditors in Chapter 9 cases, whether they realize it or not. Bankruptcy judges have agreed with the municipalities that pension beneficiaries are “making sacrifices” when a plan of adjustment strips them of their health care benefits but leaves their income benefits intact.

It has become something of a farce that the courts fail to recognize and quantify the other forms of government assistance available to retirees in determining the scope of their sacrifices. In a post-Affordable Care Act world, a municipality shedding OPEBs is not an economic sacrifice — it is tantamount to transferring those commitments from local taxpayers to state and federal taxpayers. It does seem like a sacrifice from a contract perspective, however, which allows locals to say, “See what we gave up? Now it is your turn.”

This is sufficient logic for capital markets creditors to receive haircuts, which then provide the resources required for the city to honor its pension commitments. To the extent that there are future Chapter 9 cases, expect OPEBs to be the starting point for crafting a plan of adjustment going forward. It’s just too easy.

If one looks at these insolvencies from the standpoint of local politics, it is not difficult to see how capital markets creditors can end up in the crossfire between different interest groups. Market observers tend to talk about organized labor as if it were a unified whole. In reality, large city governments can have dozens of labor groups with different perspectives and connections. This can be a very big deal with respect to pension politics. Cities can have several different pension plans depending on employees’ affiliations with funded levels that vary according to influence. Pension contributions are appropriations and appropriations are political.

If a municipality can keep its income benefits whole by making a token sacrifice that “frees up” funds that would have otherwise gone to investors, then policymakers can realign the political interests of all of these groups. They are all finally back on equal footing with respect to what they are due from the government. This keeps the machine working like a machine. Considering that governments can wander down this path (and in Detroit’s case, with open contempt for its investors) and still have some borrowing options, it is not surprising that Chapter 9 cases remain tethered to local politics.

That’s a bit of a digression, but it is still instructive with respect to Chicago. Chicago bondholders could potentially be subjected to the same destructive politics, but they would be in a worse situation legally if the city’s credit continued to deteriorate. Chicago is not eligible to file for Chapter 9, which means bondholders would have seek remedies in state courts with resistance from all of the other stakeholders (who are local). The Illinois Supreme Court has elegantly subordinated their claims to pensions through its interpretation of the state constitution and has expressed indifference about the financial impact. Absent Chapter 9, there isn't even the possibility of sharing resources.

Regardless of what one thinks about ratings divergence, the divergence between the rating agencies and market participants with skin in the game makes perfect sense. Bondholders have been repeatedly hosed for giving municipalities the benefit of the doubt in these circumstances.

How I regard financially challenged governments

Chicago’s fiscal emergency is the confluence of two distinct, but related, problems: (1) the city has made extraordinarily bad decisions for over a decade about how to manage its resources; and (2) the city has made extraordinarily bad decisions for over a decade about which financial products to use in borrowing money. Both the city and its school system used excessive amounts of short-term debt, variable rate debt, and interest rate swaps. They have also waited until it is too late (expensive) to transition their debt portfolio to a more sustainable structure.

Ester Fuchs points out in Mayors and Money: Fiscal Policy in New York and Chicago, a classic text on fiscal crises, that municipalities can “afford” (i.e., financially survive) decades of mismanagement as long as the municipality can service its debt. While I agree with this observation (cynical as it is), I believe financial innovation has introduced some notable twists.

A municipality can “afford” either protracted fiscal mismanagement or an unconventional debt structure. Municipalities that are dealing with both, however, tend to be screwed. As its credit deteriorates, resources that would have cushioned the municipality against mismanagement are instead diverted to resolving broken debt structures (until they aren't).

In Chicago’s case, the city is going to have to learn how to function without most of the gimmicks that have helped it through the last decade. Many market observers underestimate how difficult that will be, even with steep tax increases. They also seem to believe city officials are capable of becoming intellectually honest overnight.

Chicago is hardly exploring new territory here. All of the recent insolvencies in the municipal bond market have combined protracted fiscal mismanagement with a reliance on innovative financial products (e.g., interest rate swaps and pension obligation bonds). This epiphany continues to elude many market participants, especially those who believe credit analysis is as simple as financial ratios.

Is ratings shopping financially rewarded? The answer might be more complicated than it seems. 

Photo Credit: Time http://topics.time.com/teachers/pictures/ 

The Chicago Board of Education (BOE) priced $300 million of bonds on April 21st in a follow-up to its March variable rate offering. Prior to its last offering, the BOE replaced Moody’s Investors Service with Kroll Bond Rating Agency as one of the three rating agencies whose ratings would be published in offering documents. Kroll is a relatively new rating agency in the municipal marketplace.

Standard and Poor’s rated the bonds A-minus (with a negative outlook); Fitch BBB-minus (with a negative outlook); and Kroll BBB-plus (with a stable outlook).

What might prompt the BOE to make this switch? Some market observers, myself included, have suggested that this is an example of ratings shopping.Janney Capital Markets has an excellent piece that explains how ratings shopping works:

Ratings shopping occurs when an issuer chooses to publish a rating from one agency that is higher than another. Sometimes an issuer will solicit ratings and then just choose to publish the highest. Oftentimes issuers will choose to publish a higher rating over a lower rating that was published during a previous financing. Rating shopping is not a new strategy and it is legal. From an issuer’s perspective, it makes all the sense in the world — especially when investors are not paying attention … By shopping for the highest rating, issuers can come out ahead as long as they are not excluding interested buyers by only publishing one rating.

Municipal market yields remain highly correlated with ratings. An issuer that has split ratings — meaning the issuer is rated at different levels by different rating agencies — could potentially have a financial incentive to publish the higher rating because this theoretically reduces the issuer’s borrowing costs.

The BOE has been riding the fence between investment grade and speculative grade (“junk”) ratings. It is likely that the BOE feared their rating would be cut to junk by Moody’s, as the agency has been the most aggressive in downgrading Chicago credits in recent years (although perhaps not enough). So the board opted to go with a rating agency that is trying to increase its market share. The issuer pays, after all.

Can the market spot ratings shopping?

If the board was shopping for ratings, its efforts were not directly rewarded.

The issue was oversubscribed, meaning that there were more than enough interested investors for the bonds to be sold. This is not surprising. Most municipal bond issues, regardless of credit, are oversubscribed now. Furthermore, several years of extraordinarily accommodative monetary policy has created an environment where yield-hungry investors (even ones who should know better) will throw their money at virtually anything.

That said, the yield that the BOE received indicates that market participants do not trust the rating agencies’ assessments. According to the Bond Buyer:

The top yield [on the BOE bonds] of 5.63% on a 25-year maturity landed 285 basis points over Municipal Market Data’s triple-A benchmark.

While the board’s rating remains in investment grade territory, its yields aren't.

Tuesday’s MMD scale put a mid-level, triple-B rated credit at a 3.78% yield on a 2039 maturity, underscoring just how severe a penalty the district paid. The Chicago Board of Education is rated between the BBB-minus level and A-minus.

An issuer with an A-minus credit rating borrowing at +285 is laughable on its face. This is hardly a unique event, however, despite the sharp and universal criticism that the rating agencies and issuers received for this very behavior during the financial crisis. Readers will recall that Puerto Rico was considered investment grade a little over a year ago. As with Chicago credits, Puerto Rico’s bonds were trading at junk levels for a long time before the rating agencies took action.

Although this makes the rating agencies seem reactive rather than proactive, they could legitimately reply that the market should not dictate their opinions.

Credit captivity and the indirect financial rewards of ratings shopping

All financial data and legal terms discussed from this point on come from the BOE’s 2015 offering documents.

The stickiness of ratings as issuers approach junk status is fascinating. When issuers are cut to junk, the number of investors that are willing or permitted to pick up their paper (under normal market conditions) shrinks. An issuer might also have a litany of contracts tied to its ratings — like interest rate swaps and credit lines — such that a downgrade could potentially have a serious impact on the issuer’s operations. An issuer thus becomes a captive of the rating agencies as its perceived creditworthiness deteriorates.

Both the City of Chicago and the Chicago Board of Education are captives in this sense. (I wrote about Chicago’s financial condition in an article on how the city used financial engineering to conceal the magnitude of its budget gap.) The persistent divergence between their credit ratings and the ratings that the yields on their bonds imply raises some interesting questions. Is there a resistance to cut issuers to junk because the rating agencies can profit from an issuer’s captivity? Or are the rating agencies perhaps self-conscious of the consequences of their downgrades? Or both?

Although the BOE’s debt is already trading at junk levels, formal downgrades are what has, and will continue to, jeopardize the board’s operations and destroy value for bondholders.

Downgrades from Moody’s and Fitch have already given the BOE’s counterparties the right to terminate the board’s outstanding interest rate swaps. The board is underwater (meaning it would be obligated to pay) approximately $228 million on these contracts. The last I heard was that the BOE is currently negotiating outcomes with its counterparties.

The board’s counterparties might be compelled to help the board navigate its financial situation, but that is not self-evident. There are resources available that the board could use to make these payments, including property tax revenues and the board’s debt service stabilization fund.

The BOE has reserved $174 million as a buffer to make debt service payments in the event that its budgeted revenues do not materialize. However, these funds have not been specifically pledged for the repayment of bonds and the board has not covenanted to maintain the debt service stabilization fund at any given level. Bond documents explain that the board has identified the fund as a potential source of liquidity to satisfy what it owes under its swap agreements. It is possible that the board’s counterparties will decide to take what they are owed before they have to fight other stakeholders for payments in court.

(One could argue that, perversely, the absence of a statute authorizing bankruptcy makes the latter a more attractive option, since any creditor-driven litigation in state courts is likely to be messy and unproductive.)

The BOE has approximately $1.1 billion of bonds and other borrowings that bear interest in a variable rate mode. Variable rate debt is extremely sensitive to an issuer’s perceived creditworthiness.

At least some of the board’s outstanding variable rate debt is structured akin to auction rate securities. It is unclear why any issuer would deem it appropriate to borrow in this fashion after the spectacular collapse of the auction rate securities market, but there you have it.

Variable rate debt has a nominal long-term maturity, but its coupon rate is reset periodically through the remarketing of the bonds. Investors have the option to put the bonds back to the issuer (technically, the tender agent) at any time with a specified notice. This is essentially forcing the issuer to buy its own debt, unless other investors are willing to step in to purchase the bonds themselves.

Typically, when an issuer uses a variable rate structure, the issuer obtains some sort of credit facility to backstop their debt — a letter of credit or a standby bond purchase agreement, for example — in the event of a failed remarketing. A failed remarketing occurs when there is not sufficient demand for the bonds. When a credit facility is in place, the debt does not carry the rating of the issuer. Instead, the debt carries the rating of the financial institution providing the credit facility.

Some of the BOE’s variable rate offerings do not include a credit facility. I would assume that this is because the board could not locate any financial institutions that would be willing to take on the board’s credit risk themselves. They would, on the other hand, be willing to transfer that risk to investors.

Because there is not a credit facility, the BOE would have to pay penalty interest rates upwards of 9% in the event of a failed remarketing. This would increase the board’s required debt service payments significantly. The increased debt service on these bonds would then crowd out the board’s ability to meet its other long-term financial obligations. Obviously, being cut to junk could potentially make this worst-case scenario a reality.

The BOE has $211 million of credit lines that are tethered to its ratings. Recent downgrades have already increased the interest rates the board pays. A cut to junk would be considered an event of default for some of this debt, which would give the banks various remedies, including increasing interest rates within a range of 9% to 13.5%.

This is all a very big deal for the BOE because the board depends on short-term borrowing to address its cash flow needs. According to the Chicago Tribune, state officials recently identified Chicago Public Schools as one of the 32 school systems in the state that had less than 30 days cash on hand.

If the BOE was ratings shopping, it is not a mystery why. Regardless of whether the board has a higher yield on its recent offering, it’s the rating agencies that could potentially set off a downward spiral.

The state Board of Education plans to give Chicago Public Schools a $33.3 million cash infusion from a $97 million pool that lawmakers created to help financially challenged schools. This will be a temporary source of relief.

Chicago Public Schools currently has a $1.1 billion funding shortfall. For the sake of comparison, this gap represents one-sixth of the school system’s operating budget. The BOE has little ability to transform its revenue situation. Approximately 43.8% of its budget comes from property taxes. State aid makes up 33.1% and federal funding makes up 17% — some of these funds are formulaic, but subject to appropriation, and some are discretionary, but they are mostly sources of funds that the BOE cannot control.

Mayor Rahm Emanuel has suggested that the state provide $700 million of income tax revenues to offset Chicago Public Schools’s pension obligations. Given the scope of the state’s funding shortfall and its own pension issues, this will probably be a difficult sell. On top of all of this, the school system is the subject of a federal corruption investigation, the scope of which remains unknown. It’s also not the first time this has been an issue.

How significant is the threat of bankruptcy?

It seems most conversations about Chicago credits keep coming back to this question. I believe there is a much stronger risk that the BOE ends up in bankruptcy than the City of Chicago. Let me explain why.

Governor Rauner has recently suggested that the BOE should be given a means of restructuring its obligations and legislation has already been introduced to authorize Chapter 9 filings in the state. For a municipal entity to be eligible to file for bankruptcy, the use of Chapter 9 has to be specifically authorized by the state in statute.

In my opinion, Detroit was the first strategic use of Chapter 9 that the municipal bond market has seen. Bankruptcy was hardly a last resort for Detroit. Even if one excludes the city’s prized art collection, which conveniently captured all of the headlines, the city had a large number of assets that could have been disposed of to meet creditors’ claims.

Some might argue that the BOE has several options for closing its funding shortfall. The logical extension of this observation is that a Chapter 9 filing would be strategic, not necessary. I am skeptical that these options will be sufficient if the board waits until it is cut to junk by more than one rating agency to pursue them. Furthermore, many of the solutions that have been proposed would provide non-recurring resources that would only delay the BOE’s insolvency.

As I have explained in previous essays, part of the political appeal of authorizing Chapter 9 is that it seems to present an opportunity for a municipal entity to adjust its pension liabilities, even when doing so is prohibited by statutory or constitutional provisions and has been fought by labor groups. Pension liabilities are one of the main forces driving the BOE’s current predicament.

Another part of the political appeal is the new phenomenon that bankrupt entities have not been locked out of the market. Market discipline might change, however, when/if the Federal Reserve restores some sanity to the credit markets.

A state bailout of the BOE could potentially require over a billion dollars, which the state doesn't have and state taxpayers are unlikely to embrace. It is worth noting that in all recent bankruptcies, states elected not to intervene. The Alabama legislature was so opposed to tax increases that it nixed one of Jefferson County’s primary sources of revenue, even though the county was insolvent.

Practically speaking, it does not matter whether a municipality is authorized to file for bankruptcy or not — the municipality can still be functionally insolvent. It is foolish to believe that the absence of a Chapter 9 authorization protects bondholders.

The reason Chapter 9 was created was to stop creditors from piling on an insolvent municipality in state courts to the extent that the municipality could not provide essential services to constituents. A government cannot be liquidated; it was established to serve a population in perpetuity. Chapter 9 allows a municipality to address claims through an organized process that (with the exception of Detroit) preserves value for everyone involved.

Illinois can fail to authorize Chapter 9 filings for political or well-founded moral reasons. But if the BOE continues on this trajectory, it will find itself sparring in state courts individually with unions (86% of the school system’s employees belong to unions), bondholders, swap counterparties, liquidity providers, you name it. And then the cost of saving the school system will only escalate. At some point, financial realities will eclipse the political climate.

Everyone should find the strategic use of Chapter 9 abhorrent and concur that it will not become any more attractive. Regardless of what financial market tourists say about the municipal bond market, there won’t be a landslide (avalanche/tsunami/insert miscellaneous hysterical metaphor here) of insolvent governments. Most state and local government issuers manage their finances effectively and value their creditworthiness.

But municipal market participants need to adapt to a new era, where it has become untenable for the costs of some governments’ years of over-borrowing, political dysfunction, and corruption to be redistributed to a larger population. One can no longer say that Chapter 9 cases are so rare and idiosyncratic that there are no legitimate precedents for anything. There are.

Chicago made headlines at the end of February after Moody’s downgraded the city’s general obligation bond rating to Baa2. Moody’s has cut Chicago’s rating five notches in less than two years. This downgrade, however, placed the city’s credit below the termination triggers on some of its outstanding interest rate swaps. The city has been working to renegotiate the terms of those contracts with its counterparties.

If Chicago’s general obligation rating falls below investment grade, the city’s credit deterioration will become a self-fulfilling prophesy. The city risks nearly $400 million of swap termination payments and the acceleration of its $294 million of outstanding short-term debt.

Unsurprisingly, some of Chicago’s bonds are already trading at junk levels.Chicago CUSIPs are listed here.

That said, the rating agencies and most other market participants still appear to be light years away from understanding the true scope of Chicago’s financial problems. The city has a very — well, let’s just call it unconventional — approach to borrowing money and probably should not be considered investment grade.

Some budget history

In order for you to follow my discussion of Chicago’s borrowing shenanigans, it is necessary to understand the fiscal machinery behind its bond issues. Please be patient with me here. This story will blow your mind shortly.

Chicago’s budget is divided into seven different fund classifications, but only three funds are relevant to our narrative: the Corporate Fund, Property Tax Fund, and Reserve Funds.

The Corporate Fund is Chicago’s general operating fund. This fund is used to pay for essential government services and activities (e.g. public safety and trash collection). Corporate Fund revenues are derived from a wide variety of sources, including: (1) local tax revenue from utility, transaction, transportation, recreation, and business taxes; (2) intergovernmental tax revenue, which represents the city’s share of the state’s sales and use taxes, income tax, and personal property replacement tax; and (3) non-tax revenue from fees, fines, asset sales, and leases.

Chicago’s property tax revenues do not go into its general operating fund. These revenues go into a Property Tax Fund, which is used to make debt service payments on the city’s general obligation bonds; make required employee pension contributions; and (to a minor extent) fund the library system. The fund also includes tax increment financing revenues that flow to projects in designated TIF districts.

The city used some of the proceeds from long-term leases of city assets to establish Reserve Funds. The Chicago Skyway reserve funds were established in 2005 in the amount of $975 million. The Metered Parking System reserve funds were established in 2009 in the amount of $1.15 billion. Of these funds, $475 million of the Skyway reserves were designated for budgetary uses. What remained was $500 million for the Skyway; $400 million for the Metered Parking System; and $326 million for a budget stabilization fund.

There has been a structural gap in Chicago’s Corporate Fund budget since at least 2003. Although most governments are required to balance their budgets on a cash flow basis each fiscal year, a structural budget gap can arise when recurring expenditures are greater than recurring revenues. Some of the city’s offering documents suggest that this gap is a legacy of the last economic downturn, but in reality the gap pre-dates the economic downturn by several years. The impact of economic downturns on tax collections tends to have a considerable lag anyway.

So, Chicago’s structural budget gap is a political, not economic, creature. Rather than cut expenditures to a level that could be supported by recurring revenues, the city mostly used non-recurring resources to fill the gap from one fiscal year to the next. This is not surprising. Most of Chicago’s Corporate Fund budget goes to salaries and benefits for its employees, and 90% of the city’s employees belong to around 40 different unions. Attempts to adjust expenditures tend to have well organized opposition.

Between fund transfers and drawing down its reserves, the city blew through its financial cushioning quickly. The $326 million budget stabilization fund was exhausted by 2010. From 2009 to 2011, the city used $320 million from the Metered Parking Reserves. The city’s budget gap was at its widest in the wake of the last economic downturn, at over $600 million.

Chicago’s dysfunctional debt program

Now things start to get interesting. Transfers from reserves and other funds have not been the only means Chicago officials (across administrations) have devised to subsidize the city’s Corporate Fund. The city has effectively been using its general obligation bond offerings and interest rate derivatives to accomplish the same thing.

State and local governments typically use the proceeds from their bond offerings to construct or renovate public buildings and infrastructure. These are projects that have long useful lives and will benefit residents for generations.

Dating back to at least 2003, however, Chicago has been issuing long-term tax-exempt and taxable bonds to:

(1) Roll over short-term debt used as working capital;

(2) Pay for maintenance activities that would otherwise be paid from the Corporate Fund;

(3) Pay for judgments and settlements that would otherwise be paid from the Corporate Fund, including wage increases and retroactive pension contributions for its employees; and

(4) Provide discretionary funds to each of the city’s 50 aldermen to pay for activities in their own districts.

The magnitude of tax-exempt bond proceeds used for judgments and settlements over this period is staggering. The Chicago Tribune estimated it at approximately $400 million:

In 2002, for example, the city used tax-exempt bonds to pay an arbitration award involving the Fraternal Order of Police. Rank-and-file officers rejected a city contract offer in 2001, but an arbitrator ruled in favor of the city’s wage proposal a year later.

The deal included raises of 2 to 4 percent a year, to be applied retroactively. In bond documents, city officials deemed the back pay the city owed an extraordinary expense and paid $164 million of it with tax-exempt bonds.

The city ultimately will need to pay bondholders $280 million to cover the loan …

Bonds also ended up covering the $28 million a jury awarded to Joseph Regaldo in 1999. The jury found that, years earlier, a Chicago police officer had beaten him in the back of the head and neck with a blunt object, which ripped apart an artery and cut off the blood supply to his brain. The injuries left Regaldo unable to walk, talk or care for himself.

The judgment won’t be paid off until 2019 at the earliest; by then, the total cost will have grown to $53 million.

City officials eventually switched to paying judgments with taxable bonds, which are even more costly in the long run.

That is, until 2012:

About $54 million from a tax-exempt bond helped cover a legal judgment awarded to African-Americans who were denied a chance to become firefighters by a 1990s entrance exam that favored white applicants. An additional $8 million in tax-exempt bond money went to pay legal fees related to the case, records show.

By using bond money, the city created an irony for many of those awarded damages, as their future property taxes will help pay interest on the debt. In 2033, when the city starts paying down the $54 million, interest will have more than doubled the total cost.

Stop and let that sink in for a moment. That police brutality case? Wage increases negotiated with labor unions? Not just financed, but financed with long-term debt.

So why haven’t the city’s 50 aldermen protested the use of bond proceeds for these purposes? It probably has something to do with the “Aldermen’s Menu,” which allows the aldermen to use a portion of the proceeds from the city’s general obligation bond issues to pay for whatever they want for their district.

It is unclear (to me) whether the city tracks how the funds from the “Aldermen’s Menu” are spent, but in the aggregate they are not a negligible amount. From 2003 to 2012, these projects have ranged from $54.2 million to $102 million.

The use of bond proceeds to provide slush funds for policymakers has a historic analog in the revenue bonds Harrisburg issued for its incineratorproject as the city was on its path to insolvency. Pennsylvania law limits the amount of debt local governments can issue to finance projects that are not self-supporting. Substantially all of the bonds tied to the incinerator project were issued to provide working capital, although city officials were able to locate financial advisory firms that were willing to certify the opposite to state regulators.

In order to win authorization for bond issues that outright defied state law, Harrisburg’s later bond issues included a “Special Projects Fund” for city officials to play with. They bought things like artifacts for a Wild West Museum. In Pennsylvania. Don’t think too hard, there is no why.

See? Reading the Sources and Uses provisions in official statements can be fun. I know, this makes you want to invite me to your next dinner party.

Our story gets even more interesting when you look at how Chicago’s past general obligation bond offerings have been structured.

First, the city has undertaken several large, non-traditional refundings to push the maturities on debt that is coming due out into later years.

A traditional refunding is akin to how a homeowner refinances a mortgage loan. A new loan is used to prepay an old loan to achieve an interest cost savings. A “scoop and toss” refunding, which is what Chicago has done, involves additional interest cost — even in a ridiculously low interest rate environment — because the debt remains outstanding for a longer period of time.

The objective of these deals was to provide budget relief for the city’s general operating fund in the short term, even if the structure means escalating debt service payments in the long term. These restructurings artificially inflated the city’s debt capacity, so the city could continue to use property tax-supported bonds to take out the city’s working capital credit facilities, which allowed the city to avoid balancing its Corporate Fund budget.

Chicago is far from the only government to restructure debt for budget relief. Quite a few state and local governments engaged in similar transactions following the last recession. What makes Chicago unique is, again, the magnitude of this activity. According to the Chicago Tribune, “since 2000, the city has used $3.6 billion in bond money to refund old debt as principal payments came due. Of that amount, half will end up costing taxpayers in the long run.” For the sake of comparison, Chicago has around $7.2 billion of general obligation bonds outstanding.

Second, Chicago’s past few general obligation bond offerings have involved considerable amounts of capitalized interest.

Capitalized interest is typically associated with project finance, not general obligation bond issues. Project finance is a sector where loans finance revenue-producing facilities and infrastructure. The debt is supported by those revenues, not taxes as with general obligation bonds.

With capitalized interest, a bond issuer borrows more money than a project requires for construction in order to pay the interest on the bonds while the project is being built. The idea here is that the project will eventually generate revenues that can support debt service payments and the cost added by capitalizing interest.

From 2010 to 2014, Chicago’s general obligation bond deals included over $235 million of capitalized interest, simply as a means for the city to avoid servicing its debt in the short term.

If you are a bondholder, there are two things you should take away from this segment of our narrative.

First, if you hold the tax-exempt portion of these deals and the Internal Revenue Service ever gets around to scrutinizing them, your bonds probably won’t be tax exempt for long. Many of these uses of bond proceeds are not eligible for tax-exempt financing under the federal tax code. These bonds have received extraordinarily aggressive tax opinions — including, incidentally, from the same law firm that drafted Illinois’s swap legislation, which I will get to momentarily.

Second, Chicago taxpayers are on the hook for billions of dollars of long-term debt and have little of tangible value to show for it. There is a good chance that residents do not understand the nature of their government’s borrowing activities, since these were complex offerings. (Well, unless they read what I have written here…) As debt service payments increasingly compete with other political priorities for funding, this revelation might eventually erode the city’s willingness to pay.

These transactions should never have happened.

Chicago’s interest rate derivatives portfolio

Perhaps a third thing a bondholder should take away from our narrative is that to the extent Chicago is slapped with future termination payments on its interest rate derivatives, the security for your investment will be diluted. Since Chicago’s property tax revenues are also applied to pension contributions and the debt/derivatives of several other overlapping taxing districts, this is not an insignificant factor.

The State of Illinois authorized local governments to use interest rate derivatives in 2003. Here is a link to the legislation. The bill restricts the notional amount of a municipality’s interest rate derivatives to the outstanding debt the contracts will ostensibly hedge. Since the notional amount of a swap, etc. says nothing about an issuer’s risk exposure, this provision is pretty much worthless. And since the legislation was drafted by the financial industry, that probably wasn't an accident.

The legislation allows the municipality to make payments due under the swap contract (which would include termination payments) from any source of revenue it has, including property taxes. This probably wasn't an accident either.

Chicago used interest rate swaps on its 2003, 2005, 2007, and 2009 bond deals, apparently as part of a synthetic fixed rate strategy. (I explain the mechanics of synthetic fixed rate deals in this essay.) The city also recreationally experimented with more exotic contracts — swaptions and the like.

The associated bond offerings were multimodal. Multimodal bonds are bonds that can be converted to any of a number of interest rate modes at the option of the issuer. Bond documents allow the bonds to be remarketed daily, weekly, or monthly as variable rate tender option bonds, or in term or fixed rate modes. Like capitalized interest, this structure is typically used only in project finance. The multimodal structure allows long-term debt to function as both interim and permanent financing to accommodate the life cycle of a revenue-producing project.

Because the underlying debt is multimodal, Chicago never required interest rate derivatives to hedge its interest rate exposure. The city could have virtually any interest rate exposure it wanted as the bonds were remarketed. Why don’t all government issuers use this structure, you ask? Because most governments value predictable payments over trying to handicap interest rate trends and basis risk. That’s a function of being tethered to a budget.

As I noted at the beginning of this essay, Chicago is now almost $400 million out-of-the-money on its outstanding swaps. This only matters to the extent that the city’s credit ratings continue to sink toward termination triggers. Only one rating agency has to break these thresholds — so, even though S&P and Fitch still somehow believe Chicago is an A+/A- credit, Moody’s is the only rating agency that matters. If the city doesn’t get cut to junk and interest rates normalize, Chicago’s interest rate swap situation will eventually repair itself. The tipping point here either way is probably the outcome of the state/city’s pension litigation.

Chicago Public Schools — which already takes more property tax revenues from Cook County than the City of Chicago — has a swap/pension nightmare of its own to muddle through. Between the city and school system, area residents are at risk of making/financing $660 million of termination payments. The payments would compete with $28.3 billion of city and overlapping debt and billions of dollars of escalating pension contributions for funding. Basically, if you are in Chicago, your property is about to become more expensive.

From the outside, it looks like Chicago also used its interest rate swaps as a means of drumming up non-recurring resources to fill its budget gap in 2010 and 2011. If so, this is another example of the city’s willingness to trade long-term costs for avoiding politically inconvenient spending decisions.

The city made amendments to outstanding swap contracts by layering on forward-starting basis swaps. (This is something else Chicago has in common with Harrisburg.) These transactions changed the city’s net interest rate exposure on those deals from fixed to variable and introduced basis risk to the portfolio after they became effective in 2014. There was not an event that prompted these amendments and the city remains underwater on the deals. The city did receive a series of up-front payments, however. Judging from the swap confirmations from Deutsche Bank, PNC, and Wells Fargo, these payments amounted to around $25 million.

Could Chicago file for Chapter 9 bankruptcy?

No. At least not right now. To be eligible to file for bankruptcy under Chapter 9, there must be a state law that specifically authorizes a municipality to do so. Illinois law does not currently permit municipalities to do so, except under a provision that relates to units of governments with populations under 25,000. Of course, Chicago would also have to meet the other eligibility criteria. The city does have a relatively large tax base.

That said, if state lawmakers wanted to give Chicago the ability to adjust its pension liabilities — its pensions have an aggregate funded ratio of 37% — amending a statute is a lot easier than amending the state constitution.Article 8, Section 5, of the Illinois Constitution says: “Membership in any pension or retirement system of the State, any unit of local government or school district, or any agency or instrumentality thereof, shall be an enforceable contractual relationship, the benefits of which shall not be diminished or impaired.”

Multiple federal bankruptcy judges have ruled that the federal bankruptcy code supersedes state constitutions, which theoretically provides a path for municipalities to adjust their pension commitments. The rulings have not been challenged in an appellate court, however, so there isn’t a bona fide legal precedent yet.

Illinois’s Constitution describes pension commitments as contractual in nature. For an obligation to be considered secured in bankruptcy, there has to be a property interest.

This logic also applies to the city’s general obligation debt though. For most of the city’s outstanding general obligation bonds, the city has pledged a specific property tax levy. Illinois is not one of the handful of states that provides general obligation debt a statutory lien. So it would seem, in my very non-legal opinion, that the bonds would be considered unsecured debt.

As I noted earlier, the city’s general obligation bond offerings have provided little of tangible value to taxpayers. If the city were authorized to file for bankruptcy and actually did so, there could potentially be political pressure to adjust general obligation debt before depriving pension beneficiaries their incomes.

It seems unlikely that the state or federal government would “Lehman” Chicago. It is the third largest city in the United States and a vital transportation hub. It seems reckless, however, to dismiss this possibility in its entirety over the medium term.

In a recent interview with Bloomberg, Jeffrey Gundlach made some surprising comments about Puerto Rico bonds:

He said investors with an appetite for risk should consider buying debt from Puerto Rico. Even though the securities will swing in price as the commonwealth and its agencies stagger under $73 billion of debt, Gundlach said they will find a way to pay.

“I don’t care if they flop around,” he said.

Like most people with a fixed income background, I have a lot of respect for Gundlach. I think his decision to start an infrastructure fund — which is discussed earlier in the article — will be both prescient and profitable. But I think he is way off on Puerto Rico.

In a sense, debating whether Puerto Rico will be able to service its debt going forward is purely academic. The territory is already borrowing funds to make its debt service payments — much of the bond proceeds from Puerto Rico’s general obligation bond offering last March were used for this purpose. So, technically speaking, the territory is not fully servicing its debt right now.

One of the risk factors that is cited in Puerto Rico’s offering documents is that the territory’s future required debt service payments escalate sharply from here. If Puerto Rico loses market access — which has already occurred with the investors who traditionally invest in tax-exempt debt — the government’s ability to kite payments is over.

Overborrowing as a pathology with a predictable course

In my mind, Puerto Rico is as obvious a candidate for sovereign default as they come. The territory is undergoing a species of debt crisis that has many historical precedents and a familiar trajectory. It is well understood that a borrower in Puerto Rico’s position is unlikely to recover economically while honoring its debts on time and in full.

Puerto Rico’s situation is accurately captured in Revisiting Sovereign Bankruptcy, Chapter 2: Pathologies in Sovereign Debt, incidentally co-authored by Lee Buchheit, whose firm is one of Puerto Rico’s advisors:

Overborrowing requires creditors in the private or official sector that agree to provide the needed financing. Overborrowing is often facilitated by herding behavior, which leads creditors to take on too much risk during periods of global optimism …

There is evidence that policymakers are often reluctant to restructure their debts and suboptimally postpone unavoidable defaults (e.g. Borensztein and Panizza 2009; Levy Yeyati and Panizza 2010; IMF 2013). Delayed defaults can lead to the destruction of value because a prolonged predefault crisis may reduce a country’s capacity and willingness to pay. Its capacity to pay is reduced because procrastination prolongs the climate of uncertainty, high interest rates and restrictive fiscal policies that are ineffective in avoiding default but amplify output contractions. Delayed defaults reduce its willingness to pay because electors that have suffered long periods of economic austerity are less likely to support a creditor-friendly debt restructuring.

Because policymakers are often replaced after a debt default, late restructurings may be caused by self-interested agents that have incentives to gamble for redemption, even when delays entail economic costs for society as a whole. Myopic policymakers who do not take into account the long-run costs of excessive debt accumulation may also decide to delay a default in order to have continuous access to external resources. Short political horizons may also create incentives to undertake policies that increase the vulnerability of the financial sector to government default. This generates short-term benefits in terms of a higher capacity to borrow, but at the expense of higher future default costs if the accumulated debt turns out to be unsustainable (Acharya and Rajan 2013).

This is exactly what is happening in Puerto Rico right now. Hedge funds and other institutional investors have stepped in as private lenders of last resort.

Although I would not say that we are experiencing a period of global optimism, the current interest rate environment is fostering the same herding behavior described above and is probably the only reason some otherwise sophisticated investors feel comfortable shrugging off the real risks involved in investing in Puerto Rico’s debt. And since this environment is the product of extraordinarily accommodative monetary policy, it will be similarly transitory.

So hedge funds will continue to stuff Puerto Rico like a giant debt piñata until a shift in capital flows busts it open.

There are at least two things that could cause Puerto Rico’s private lenders of last resort to vanish (and possibly abruptly). The first would be if interest rates stateside were to normalize. The second would be if any compelling distressed corporate opportunities presented themselves. As my friend @groditi has pointed out to me, one factor that has driven hedge funds to cross over to Puerto Rico’s debt is an overall lack of distressed situations in corporate debt. Perhaps this will occur with bonds for energy-related companies.

Why monetary default has been characterized as inevitable for Puerto Rico

As far as I can tell, essentially three kinds of distressed investors are attracted to Puerto Rico’s paper:

(1) Investors who believe they can profit by purchasing the debt and then convincing traditional investors to return in the short or medium term;

(2) Investors who believe the government can successfully restructure the territory’s public corporations while protecting its general obligation, municipal, and COFINA debt; and

(3) Investors with unique theories about economic development and how Puerto Rico might go about broadening its tax base.

All of these scenarios strike me as unlikely.

Scenario (1)

In my opinion, traditional investors are unlikely to return to investing in Puerto Rico’s debt anytime soon. These investors never purchased Puerto Rico’s debt based on the territory’s credit fundamentals. (Some institutional investors  with large exposures are now trying to suggest otherwise, however.) They purchased the debt because Puerto Rico’s bonds are the only bonds in the market (of any real size) that are triple exempt.

Recent developments have since forced them to evaluate Puerto Rico’s creditworthiness. It is clear investors are not being compensated for the magnitude of financial risk and uncertainty involved or protracted headline risk (which is a big deal for many institutional investors).

It is also reasonable to fear that the federal government will rethink subsidizing Puerto Rico’s debt (through its tax exemption) given that Puerto Rico’s singular tax status has provided an obvious incentive for overborrowing.

Scenario (2)

Taken all together, Puerto Rico currently has around $71 billion of outstanding debt. Given a population of around 3.5 million people (roughly the same population as Connecticut), that amounts to around $20,000 of debt per capita and a debt-to-GDP ratio around 70%. Puerto Rico’s outstanding debt literally doubled from 2003 to 2012 as the territory borrowed to finance its deficits.

This debt burden is not even remotely comparable to any US state. According to Moody’s 2014 State Debt Medians report (subscription required), the median debt per capita for US states is $1,054. The median debt-to-GDP ratio is 2.4%. The only viable analogy here is with distressed sovereign borrowers like Greece.

Principal and interest payments consume about 15% of Puerto Rico’s budget and will swell by $287 million in FY 16 (which begins July 1, 2015). This burden will increase as Puerto Rico’s interest cost on future borrowings climbs. The general obligation bonds Puerto Rico issued last March have been hovering around 10%, at a time when yields for other sovereign borrowers have been declining (and some have even turned negative).

The notion that Puerto Rico can successfully “ring-fence” its core obligations has been aggressively promoted by some market observers, who suggest that this approach could decrease Puerto Rico’s debt burden and improve liquidity for the Government Development Bank (GDB).

My first rejoinder to this is that it is impossible for any government to shirk financial relationships with entities that provide essential services. This is a fairly straightforward concept in credit analysis.

Furthermore, Puerto Rico has also shown zero ability to forecast its burn rate even two months in advance. Its liquidity position was off 17% from thepredictions it made just in October. Its projections regarding the feasibility and financial impact of major restructurings are thus questionable.

When Puerto Rico went to market last March for $3.5 billion (which was the largest junk offering in the history of the muni market), officials told investors the offering would carry the territory for several years. Now they are back, one year later, with a $2.9 billion offering. It’s remarkable such claims have escaped the attention of regulators. Perhaps this will change when the retail investors that remain stuck with Puerto Rico exposure get burned.

The hedge funds that have been picking up Puerto Rico's bonds realize the territory's dependence on the market is a permanent situation. Some have suggested that Puerto Rico's next offering be increased from $2.9 billion to over $4 billion so the territory will not have to return to the market for 2 years (optimistically). One told Caribbean Business that they should do this "or they should start restructuring."

My second rejoinder is that, from a management standpoint, this effort seems futile. Let’s say, hypothetically, that the territory could restructure its public corporations as self-sustaining enterprises. According to recent disclosure documents, Puerto Rico’s total good faith and credit debt is around $18.6 billion and its total general fund-supported debt (which includes appropriation-supported debt and tax and revenue anticipation notes) is around $23 billion. For comparison, this is still around six times the median debt per capita for US states.

The government’s core debt obligations seem even less affordable when one takes into account that Puerto Rico’s pensions have a funded ratio of less than 9%, which means the government is close to having to cash flow benefits payments.

The structure of Puerto Rico’s required debt service payments also presents a major problem. Puerto Rico has ascending debt service payments but an economy that has been in decline since 2006, a rapidly shrinking labor force, and unemployment rate above 13%. Its debt burden and the tax base that supports that debt burden are moving in opposite directions.

These problems have been apparent to most analysts for a long time. Back in October, Moody’s published a report noting that “Puerto Rico’s [Sovereign Expected Default Frequency] measure indicates the highest sovereign probability of default globally.” (Moody’s Analytics, October 6, 2014)

Scenario (3)

Some investors believe the government could still identify new sources of revenue — a value-added tax, etc. These proposals strike me as increasingly unfeasible politically. Puerto Rico’s debt burden is so large that all of the revenue generated would have to be devoted to debt service. Puerto Rico’s maximum annual debt service for FY 16 is $1.16 billion and $1.18 billion if one also takes into account guaranteed debt.

Residents are already choosing to vote with their feet and seek new opportunities in the mainland US. There are more Puerto Ricans living in the mainland US than there are living on the island:

The Puerto Rican population stateside reached over 4.9 million in 2012 and has surpassed that of the island since 2006. Puerto Rico's population peaked at 4 million in 2009, but it hastily dropped to 3.6 million in 2013 and is estimated to fall to 3 million by 2050.

It is difficult to imagine there is broad consensus among residents that they should stick around just to offer up the fruits of their labors to hedge fund investors. But this is precisely what investors who are arguing that Puerto Rico is money-good assume.

The government’s willingness to pay appears to be incrementally shifting in response to political pressure.

The role of bond insurers

One aspect of Puerto Rico’s future borrowing activity that intrigues me is the role of bond insurers. I have already written about their underwriting standards here.

A handful of bond insurers have significant exposure to Puerto Rico. From the Wall Street Journal:

MBIA, through its National Public Finance Guarantee Corp. subsidiary, insures $4.8 billion of Puerto Rico bonds, Assured Guaranty backs $5.4 billion through three separate subsidiaries and Radian Group insures $452 million through Radian Asset Assurance, according to Moody’s.

MBIA’s National Public Finance and Assured Guaranty Re Ltd., one of the three Assured subsidiaries, have the highest exposure to Puerto Rico bonds as a percentage of their “qualified statutory capital,” or assets available to cover unexpected losses, according to the Moody’s analysis. National Public Finance insures $4.8 billion but has only set aside $3.3 billion for unexpected losses, a ratio of 145%. Assured Guaranty Re backs $1.5 billion but has only set aside $1 billion, a ratio of 147%.

(NB: These figures are a bit dated as the article was written in the middle of last year. But the point still holds.)

Some have suggested that the bond insurers may continue to have a role in Puerto Rico’s future financings to reduce the risk of default. It strikes me that this has an interesting precedent in Harrisburg, Pennsylvania.

FSA continued to insure Harrisburg’s incinerator borrowings long after it became evident that the city was not actually financing project improvements and the city’s offerings were essentially robbing Peter to pay Paul. Why did the insurer do this? Because the moment Harrisburg lost market access — which would have been immediate absent credit enhancement — the city would default and the bond insurer would have to begin making payments. In this sense, the insurer had backstopped so much of the city’s debt that the insurer’s financial fate became intertwined with the city’s. This is what enabled a city with fewer than 50,000 residents (and scant taxable property due to its being the seat of government operations) to rack up hundreds of millions of dollars of debt, which eventually rendered the city insolvent.

(Puerto Rico is not eligible to file for Chapter 9 bankruptcy. But if it were, the territory would already meet the criterion for insolvency.)

So it is possible for a lot of sophisticated players to end up unfavorably involved in Puerto Rico’s debt crisis.

As a closing note, I anticipate many readers will react to this with “investors and the bond insurers are all just betting on a federal bailout.” The federal government engineered a quiet bailout years ago. While considerable, it is not sufficient to alter the territory’s financial predicament.

I have always wondered what was responsible for significant differences in bids for infrastructure assets. A paper from the EDHEC-Risk Institute,Unlisted Infrastructure Debt Valuation and Performance Measurement: Theoretical Framework and Data Collection Requirements (July 2014), answered a lot of my questions on this topic.

The paper notes three reasons why long-term infrastructure debt is difficult to price effectively:

(1) Market prices cannot be observed;

(2) Originators and investors possess unique cash flow data; and

(3) Covenants and lender control-rights function as embedded options.

Many market participants use pricing models that are insensitive to structure — and, in particular, exclude (3).

When Non-Recourse or Limited Recourse Loans are Nothing of the Sort…

One aspect of project finance loans that distinguishes them from other forms of lending is that they are structured on a non-recourse or limited recourse basis. These loans, however, have a “tail” that provides a sort of pseudo-collateral:

[The tail is] the difference between the original maturity of the project loan and the life of the infrastructure project. The free cash flow of the firm available during that period acts as a form of collateral. In certain states of the world — corresponding to a breach of covenant — lenders have control rights that allow them to restructure a loan and use its tail to maximize their recovery rate. The value of the loan’s tail, as well as the relative size of liquidation and restructuring costs of the project company, can thus be expected to have a significant impact on performance.

This is why leases associated with improving and managing infrastructure projects are typically long-term. The length of the contract plays a role in managing financial risk overall by insulating the lender from a continuum of failures.

So how does structure translate into investors submitting radically different bids on projects?

I have always assumed this was driven by the lack of empirical observations regarding cash flows and varying attitudes toward holding illiquid assets. But another factor might be the sophistication of the models employed. According to the authors,“if embedded options and covenants found in [project finance] debt are not taken into account, infrastructure debt valuation is likely to be off by an order of magnitude.”

How Business Risk Determines Loan Structure

Infrastructure projects are typically divided into two categories, depending on the source of their cash flows.

“Merchant infrastructure” (demand-risk model) generates revenue through the sale of a service. Utilities and toll roads are good examples of merchant infrastructure. These projects involve a high degree of market risk.

“Contracted infrastructure” (availability model) projects receive a contracted revenue stream in exchange for providing a service. These projects involve hardly any market risk. This model is often used in the privatization of social infrastructure, such as schools and hospitals.

Lenders treat merchant and contracted infrastructure differently:

Merchant infrastructure projects are generally structured with a rising DSCR and a longer tail. A rising DSCR [debt service coverage ratio] implies that lenders get paid faster than equity owners, and a longer tail increases the value of lenders’ security. In other words, lenders demand an increasing mean DSCR and a longer tail to protect themselves against a higher and increasing DSCR volatility, which results from higher revenue risk.

In contrast, contracted projects are structured with a flat DSCR and shorter tails, as lenders demand less protection against default due to lower expected underlying revenue risk.

These structures give lenders different kinds of options regarding when to intervene in a project’s operations. They also determine under what conditions the lender will attempt to renegotiate the terms of a project loan or exit the arrangement:

With technical defaults [breaching a covenant, as opposed to failing to make a debt service payment], lenders can only try to maximize the recovery rate of the original debt given the tail, whereas hard defaults give them more options, including exiting the relationship with original equity investors and taking over or selling the project or its debt.

The new debt service is determined by taking into account what each party would lose in the absence of a workout … If there is no space for renegotiation upon a hard default, the project is taken over by lenders (which may seek a new equity investor). Conversely, restructuring must take place as long as the value of either debt or equity post-restructuring is higher than in the absence of renegotiation i.e. as long as lenders can get more than their exit value and equity holders more than nothing, a new debt schedule is agreed and maximizes recovery in the tail.

Shifts in the implied default risk are reflected in how the debt is structured:

The rising DSCR profile of merchant infrastructure implies that the project’s likelihood of default decreases faster in time. If a loan survives the first few years after the construction stage, the increasing mean DSCR more than offsets the increasing DSCR volatility, making it less likely that the project will default in the future. For contracted infrastructure, [the] flat DSCR profile implies that the probability to default barely changes in time, though it stays at a very low level due to lower DSCR volatility.

I have previously argued with respect to municipal bonds that there are trade-offs between credit risk and duration risk. This is also true in project finance: “To reduce the credit losses upon default, investors have to extend the maturity of their loan further in the tail, and have a higher interest rate risk due to a higher duration.”

Typical Covenants and the Introduction of Optionality

This paper provides an excellent overview of the various financial and non-financial covenants that are included in project finance loans.

Minimum Debt Service Coverage Ratio — The project’s DSCR is the ratio of the special purpose entity’s free cash flow to scheduled debt payments. If the DSCR falls below a specified threshold, equity dividends are locked-up to insulate lenders from losses. As the paper observes, this also gives equity holders an incentive to fix whatever operational issues are contributing to decreased cash flows.

Non-financial default triggers — These would be events like, the special purpose entity losing its legal ability to operate the infrastructure; construction delays or cost overruns; and cross-default events.

Step-in events — These would be specified events that give lenders the right to intervene in the project’s management and operations. Lenders could be given the option to accelerate debt or reschedule debt payments.

Cash sweeps — These are compulsory prepayments intended to reduce losses from lower-than-expected cash flows. The contract will specify how much of the cash flow after debt service (if any) will go to equity holders, with the remaining being used to retire the debt.

Cash clawbacks — Wherein equity investors agree to lend additional funds to the special purpose entity if it encounters any problems, up to the amount of the dividends the equity investors have received from the project to date.

Reserve accounts — These are common in all types of lending, and involve setting aside funds to cover debt service, taxes, and other expenses.

Prepayment option — This may seem counter-intuitive, but project debt can be restructured if the special purpose entity is in financial distress or if the project does exceptionally well. This option allows the special purpose entity to refinance the debt at a lower interest cost in the latter event.

...And why Stockton's confirmation might make Chapter 9 filings more popular.

One of the most frustrating elements of Detroit’s bankruptcy was the fact that the court permitted the city’s pension systems to manipulate the assumptions used in determining the city’s pension liabilities. The city used assumptions negotiated with labor as opposed to those used by its actuaries. This creative pension math resulted in beneficiaries receiving above-par recoveries when some bondholders received pennies on the dollar. (I explain the mechanics here.) This injustice was magnified by the pension administrators’ history of corruption.

Pension math did not elude the court in Stockton’s bankruptcy, however.

In his Opinion Regarding Confirmation and Status of Calpers, Judge Christopher Klein laid bare the California Public Employees Retirement System’s (Calpers) attempts to obstruct Stockton’s restructuring process. Klein also made a strong case that Calpers has no standing to weigh in on any municipal bankruptcy. Of Calpers, he wrote, “the bully may have an iron fist, but it also turns out to have a glass jaw.”

Calpers has a long history of using the court system to intimidate local governments that are considering adjusting their pension liabilities. In San Bernardino’s bankruptcy, Calpers threatened to litigate its position all the way to the US Supreme Court. While a behemoth like Calpers can afford the legal fees associated with such a fight, local governments cannot. San Bernardino predictably acquiesced.

Calpers has been so defensive of its perceived territory that the pension system even filed an amicus brief on behalf of retirees in Detroit’s bankruptcy.

Klein is no doubt aware of Calpers’ attempts to derail other California bankruptcy proceedings and squash future filings. And he is certainly not the first judge exasperated by Calpers’ antics.

Some background on Calpers’ relationship with local governments

Most people recognize Calpers as the pension system for California state employees. However, Calpers may also contract with local governments to administer their pensions.

As Klein observes, California municipalities enjoy considerable flexibility with respect to how they establish and administer pension systems for their employees. A municipality could: (1) establish and administer its own system using individual or group life insurance or annuity contracts; (2) join a county system or partner with other municipalities; (3) contract with a private entity; (4) potentially have a union-administered pension system; or (5) contract with Calpers. Municipalities are also (theoretically) free to move from one structure to another.

Stockton has contracted with Calpers, but some large municipalities —such as Los Angeles, San Diego, and Fresno — have not.

From a local government’s perspective, Calpers is just another government contractor. According to Klein, Calpers has contracted with 1,580 local public agencies and 1,513 school districts to service their pensions. Surprisingly, only 32% of Calpers’ members are state employees — 31% are local government employees and 37% are school employees. It does not take a rocket scientist to understand why Calpers so vehemently challenges the adjustment of pension benefits through Chapter 9. Servicing local government pensions is a significant source of fee income for Calpers.

Although municipalities have several options regarding how their pensions are administered, Calpers’ authorizing legislation gives the system some significant competitive advantages.

First, Calpers’ pension rights are portable and tied to template plans, meaning that employees can easily transfer them if they move to another employer that has also contracted with Calpers.

Second, Calpers’ board members are all selected on a political and not professional basis. The board consists of seven public officials and six current or retired employees. The California Constitution includes provisions preventing the legislature from changing Calpers’ board composition or influencing investment policies.

Participating municipalities become disenfranchised in determining the benefits they provide to employees. Whether this is an issue or not depends on the political persuasion of their leadership. As Klein notes, “once a municipality agrees to a Calpers contract, the Calpers board gets into a position to block changes in the municipality’s pensions by saying a local change would adversely affect the system.” Since labor representatives comprise 46% of the board, municipalities are unable to amend plans in a manner that employees reject.

This scheme has historically proven to be a major obstacle for distressed municipalities trying to improve their financial situations.

Calpers’ confiscatory termination policies

Trigger warning: If you are a California taxpayer, this section will probably make you want to break things.

In Stockton’s bankruptcy, Calpers argued that under state law any cut to pension benefits would be tantamount to terminating the contract between the city and Calpers. Calpers also argued that state law gives Calpers a lien on the city’s revenues to satisfy the termination payment Calpers requires when a city leaves the system. Calpers argued that this unique lien gives the pension system standing to argue against the city’s plan of adjustment and should prevent pension benefits from being impaired.

Klein suggests that Calpers’ termination policies were intended to be a poison pill, which is spot-on. This might not be completely obvious to local government officials, however, until they are exposed to the mechanics. 

So let’s get into the mechanics.

In the event that a municipality’s service contract with Calpers is terminated, either voluntarily or involuntarily, Calpers moves the municipality’s plan assets from Calpers’ general investment pool to a special termination pool.

Although Calpers does not bear any financial risk for benefits payments, Calpers invests funds in the termination pool on a more conservative basis than it does funds in its general investment pool. This is an important detail. The difference in investment strategy is what makes Calpers’ termination policies a poison pill.

When the municipality’s contract is terminated, Calpers determines the amount that the municipality’s plan is underfunded and demands that the municipality deposit that amount in the termination pool. Even if a municipality’s plan was 100% funded when it was invested in the general investment pool, it will be dramatically underfunded when it is moved to the termination pool.

This occurs because a lower discount rate is being used in the calculation to reflect the more conservative investment strategy. Per Klein, Calpers uses 7.5% for funds in the general investment pool, which is based on 2.75% for inflation and 4.75% for investment return (net of expenses). The assumed return for the termination pool is the yield on 30-Year Treasuries — 2.98% as of June 30, 2012.

So Calpers claimed that if Stockton intended to impair its pension payments, Calpers would terminate the service contract. In this event, Stockton’s pension liabilities would balloon from $211 million on an actuarial basis to $1.6 billion under the termination methodology. Not something the city could just cut a check for.

Now remember that Calpers has zero financial risk here. If a municipality were to fail to make the contributions required to pay benefits at the levels it has committed to, then the benefits would be automatically reduced. There is literally no justification for the termination pool to be invested differently and, by extension, for the plans to be valued on a different basis.

The only reasons these termination policies exist are to (1) remove any incentive for a municipality to switch from Calpers to another pension administrator (which would have a financial impact on Calpers in the form of lost fee income and reduce labor’s bargaining power) and (2) sabotage any attempt to adjust pension benefits in Chapter 9. Perversely, this framework seems to reward a lack of funding discipline so long as a municipality toes the party line.

Calpers’ golden handcuffs have been an issue for municipalities that are not even remotely close to financial distress. There are many reasons a municipality might wish to change its pension administrator — officials believe they can find better investment performance elsewhere, another entity has less onerous terms, etc.

One example that particularly demonstrates the absurdity of Calpers’ exit costs is Villa Park. Villa Park is a small but wealthy city within Orange County. The city does not have a police or fire department and only has 7 employees. Its required employer contribution for its pension fund is $110,000 a year. When Villa Park recently considered leaving Calpers, Calpers estimated the city’s termination cost at $3.6 million. For the sake of comparison, the city’s total general fund budget was $2.7 million. Calpers’ termination policies left the city with two options: (1) maintain the status quo and stay with Calpers, or (2) leave Calpers and become insolvent.

So why doesn't a municipality sever its relationship with Calpers and tell the pension system where it can shove its termination costs? Because California law gives Calpers a lien to enforce the payment of termination expenses. Per Klein, this lien applies to all of the municipality’s assets, subject only to a prior lien for wages.

As Klein points out, the provisions in California law that create this lien are oddly nonuniform. California law selectively protects Calpers. The provisions do not apply to any other municipal pension system.

Calpers and Chapter 9

Klein determined that Calpers did not have standing to interject itself in California municipal bankruptcies for essentially three reasons:

(1) The lien on participating municipalities’ assets does not survive a Chapter 9 filing (federal law supersedes state law);

(2) As the lien does not survive, Calpers cannot be construed as a creditor beyond the fees it is owed for administering the municipality’s pensions (which would make Calpers a very minor creditor); and

(3) The real creditors with respect to pensions are the city’s current and retired employees singly, and the city’s employees had already negotiated a settlement with the city.

It strikes me that Calpers’ termination policies combined with Klein’s confirmation logic (assuming it is upheld) make it very convenient for participating California municipalities to adjust their pension liabilities through Chapter 9. This is especially true if it becomes a trend that bankrupt municipalities maintain market access.

The insolvency test for Chapter 9 is not a balance sheet test. A municipality is only eligible to file for bankruptcy if it is insolvent on a cash flow basis. This means that the municipality has to demonstrate that it currently cannot pay its obligations as they come due.

This criterion is what prevents most local governments with substantial unfunded pension liabilities from filing for bankruptcy for the specific purpose of restructuring their pension commitments. Calpers’ termination payments are due upon exiting the system, however. And these termination payments are usually so large that paying them would render the municipality insolvent outside of bankruptcy.

The termination payments were intended as a mechanism to prevent governments from cutting pension benefits and force other stakeholders, like bondholders, to take cuts instead. But since the termination lien does not survive bankruptcy, Calpers’ termination policies might actually provide a means for municipalities to do the opposite of what state law intended. This seems like a viable solution in a situation where no middle ground exists.

And if municipalities elected to restructure their debts in ways that are favorable to bondholders — thus removing the question of market access altogether — this might become a popular solution. Since California law does provide a statutory lien for general obligation bonds and special revenue bonds are shielded in bankruptcy, this does not seem all that far-fetched.

There is considerable debate about whether the federal government should ease the terms of outstanding student debt or forgive student loans altogether. There is also a lot of misinformation. I am writing this article to provide some context and clarification on the government’s lending activities. I am not writing this article to debate whether federal student lending serves a public good or represents an essential government service. I will leave that to you.

Unless otherwise specified, the figures I cite come from the US Treasury’s Fiscal Year 2014 Q4 report to the Treasury Borrowing Advisory Committee. The report has many fantastic charts, so I highly recommend perusing it.

The federal government has been providing financing for higher education since 1965 with Title IV of the Higher Education Act. Origination of federal student loans by private lenders ceased with the Health Care and Education Reconciliation Act of 2010. (Yes, these provisions were tucked into the Affordable Care Act legislation.)

All federal student loans are now made directly by the Department of Education and funded through the issuance of US Treasury bonds. So the federal government borrows money from investors in the bond market and then uses the proceeds to make student loans. US Treasury bonds are increasingly becoming a form of asset-backed security — and, as I will get to later, poorly structured ones at that. According to Treasury, direct lending for higher education accounts for around 8% of federal debt held by the public. It is estimated to be approaching 12% within 10 years. Treasury is borrowing these funds at a market rate, but the interest rates on the loans are established by law.

The amount of student debt outstanding has doubled since 2007 to around $1.3 trillion. The federal government’s direct lending represents more than 85% of origination volume. (Pause for a moment and reflect on the fact that the federal government backs more than 85% of student lending, but is responsible for only around 15% of total infrastructure investment in the US.)

Myth #1: Student lending is a profitable enterprise for the US Government

It actually costs taxpayers billions of dollars.

This myth is debunked on a regular basis, but often resurfaces when policymakers are looking for a justification for easing the terms on student loans. Most recently, this argument was made by Senator Elizabeth Warren in trying to build support for her student loan refinancing legislation, which would have also been offset by raising taxes on wealthy households.

The “budget savings” some policymakers cite is an artifact of an accounting methodology that even the Congressional Budget Office acknowledges is nonsensical.

Here is a good explanation of CBO’s methodology from the Washington Post:

The Congressional Budget Office is required by law to use a bizarre and faulty method for determining the cost of government loans. Just like any institution, the CBO determines the cost of loans by “discounting all of the expected future cash flows associated with the loan or loan guarantee — including the amounts disbursed, principal repaid, interest received, fees charged and net losses that accrue from defaults — to a present value at the date the loan is disbursed.” To do that, it needs to settle on a “discount rate,” which is usually the expected rate of return on the loan in question. Banks and other private institutions generally estimate that by finding loans with similar risks and maturities to the one being evaluated, and then using those similar loans’ rate of returns.

The CBO does not do that. It discounts all government loans using the returns on Treasuries of a similar maturity. So a 30-year student loan would be compared to a 30-year Treasury bond. But Treasuries are the safest bets in the world. The US government does not have a very high risk of defaulting, not least since it prints its own money. Student loans are much riskier.

Of course, saying that student loans are riskier than Treasuries is kind of like saying Alpha Centauri is farther away than London. First, approximately 76% of federal student loans are originated with zero credit underwriting — here’s a loan for tens of thousands of dollars, no questions asked.

Second, according to Treasury, the default rate (270 days without repayment) on outstanding student loans is about 9%. However, “behind the default rate is a shadow book of potential future defaults, reflected in the volume of loans in deferment and forbearance. Those loans add 23% to the 9% that are already listed in default.”

It should be an obvious point that student loans are costing taxpayers money as long as the loans are made at anything below a market rate. (One could consider this a form of negative arbitrage.) And it should be an obvious point that — assuming a private lender could be found for such loans — a market rate would be multiples of the interest rate on Treasuries.

Based on the methodology required by the Federal Credit Reporting Act of 1990, CBO estimates that student lending programs will result in a profit of $135 billion over the next ten years. However, using fair-value accounting, the programs will result in an $88 billion cost to taxpayers. Any modifications to existing programs (reducing interest rates, more generous repayment options, extending maturities to increase the chances of repayment) could increase this cost dramatically. For example, Treasury estimates that “the gross cost of maturity extension in order to increase the probability of repayment would be approximately $220 billion.”

Another point to be made here: much of this borrowing — which is driving up the country’s debt and generating this cost to taxpayers — benefits for-profit schools. Per Treasury, students at for-profit universities are twice as likely to utilize federal student loans. Furthermore, many of these schools have predatory business models: approximately 68% of the students at these schools either drop out or do not finish their degree programs within 6 years. As of 2011, loans for private, for-profit institutions had a 13.6% default rate.

There could potentially be broader social costs as student debt burdens crowd out other forms of borrowing (mortgages, auto, small business, etc.), although this is still somewhat controversial.

Myth #2: Student loan growth is driven by English majors (Kidding, but really, isn't this the stereotype in most articles on this topic?)

Treasury identifies a number of drivers behind the dramatic rise in student debt:

· There are more college-aged people (the cohort aged 20–24 has grown by 9.4% over the last 10 years);

· More students are choosing to attend college and stay in school longer to pursue advanced degrees;

· A larger percentage of students are taking out federal student loans (48% in 2012 versus 33% in 2002);

· Tuition has increased as state governments reduced subsidies for public universities;

· Outstanding balances are declining at a slower pace because (1) more loans are in deferral and forbearance, (2) the tenors of the loans are longer than they have been historically; and

· The number of for-profit institutions has been growing rapidly, along with enrollment and borrowing

Myth #3: Student lending necessarily contributes to economic and social mobility

This depends entirely on how much a student borrows, how long they stay in school, and if they graduate.

According to Treasury, “today an average of 40% of students at four-year institutions (and 68% of students in for-profit institutions) do not graduate within six years, which means they most likely do not benefit from the income upside from a higher degree yet have the burden of student debt.” Furthermore, student loan debt cannot be extinguished in bankruptcy in almost all cases and the federal government can withhold other benefits to offset amounts owed.

It is remarkable how porous historical accounts of the bond insurance industry have become. Articles written about the industry almost never discuss ratings arbitrage, for example. Ratings arbitrage was the raison d’être of bond insurance before the financial crisis.

These holes in the narrative are significant because what is missing would inform investors about the value of bond insurance in the market now. (This might not be an accident.)

In 2007, roughly half of new municipal bond issues carried bond insurance. (This equates to hundreds of billions of dollars in volume.) The bond insurance industry was decimated during the financial crisis due to their heavy exposure to toxic mortgage-backed securities. 

In 2014, only about $18.6 billion of new municipal bond issues were wrapped. Most of this volume is attributable to two firms, Assured Guaranty and Build America Mutual (BAM). Assured Guaranty wrapped $9.9 billion, giving it a market share of 53.5%. BAM wrapped $7.5 billion of bonds, giving it a market share of 40.4%.

One might look at these figures and assume that this is a story about how the industry made a lot of bad bets and two companies are slowly building it back again. That is happening, but the story isn't that simple.

Before the financial crisis, the rating agencies rated municipal bonds on a scale that was unique to the municipal bond market. The rating scales made comparing municipals to other asset classes impossible. Municipal bonds were rated below corporate credits with objectively similar credit profiles. 

The rating agencies aggressively defended this practice. They claimed that it was necessary to show “granular” differences between municipal credits since municipal defaults are extremely rare. But this did not reflect how the market used the ratings. State and local governments were unnecessarily penalized with higher yields until Congress forced the rating agencies to adopt universal scales.

These artificial disparities created a captive market for bond insurers, which is how they ended up wrapping half of the market. Insurers were writing policies to cover the risk of default on thousands of credits that had remote chances of defaulting. And somehow they decided to ruin that gig by backstopping toxic waste.

This arrangement had a perverse effect on the bond market. The rating agencies wanted to use a separate scale to show granular differences in municipal credits. The bond insurance companies (that the rating agencies kept in business) actually removed all granular differences in perceived creditworthiness. It didn't matter to investors if they were buying a state with a AA underlying rating and the ability to levy taxes or a barely investment grade conference center. Everything was AAA. Nothing about this structure was healthy.

It wasn't until the destruction of the bond insurance industry during the financial crisis that investors started to look at what they owned and bonds started to trade accordingly. It is unlikely that investors will forget this experience anytime soon.

So what value does bond insurance provide to investors in the current environment? The insurers have been using high-profile defaults like Detroit (and probably Puerto Rico soon) as evidence of the value of their product. And it is true — the investors that had bond insurance in Detroit were glad they had it, because they would have been mostly wiped out by a rogue bankruptcy court otherwise. 

But these credits are outliers and that is unlikely to change. It is also worth noting that in cases like Detroit and Harrisburg, the impression that investors were covered facilitated the governments borrowing past the point where they should not have had market access. (This kind of behavior is why downward spirals take longer than one would otherwise expect. Not to mention the fact that these deals involved fraud.)

With the paucity of new supply in the market, most new bond issues are massively oversubscribed. When deals are bound to be oversubscribed, borrowers have no incentive to include strict covenants or pay a premium for bond insurance. Investors will be fighting each other for the bonds anyway. Also, as my friend @StevePendergast observed, “compression of rates makes [the] incremental value of a wrap less important and [the] premium more expensive.

There might still be opportunities for ratings arbitrage of a different stripe, where an issuer has been arbitrarily rated below where it should be. But given how picked over the markets are from a long period of extraordinarily accommodative monetary policy, this seems difficult to do. 

(This is part six in a series on infrastructure investment.)

Public-private partnerships (also referred to as P3s or PPPs) are perhaps one of the most unnecessarily controversial topics when it comes to infrastructure. This is partially due to the unitiated not understanding that this term refers to a variety of arrangements and not just one. There is also a lack of care on the part of public officials to explain how these arrangements can be constructed well or poorly.

Practically speaking, a public-private partnership is a contractual arrangement between a governmental entity and a private firm to design, construct, operate, and maintain an infrastructure asset over some established period of time. At the end of this period, the responsibility to operate and maintain the asset reverts back to the government. The government typically maintains ownership of the asset during the whole process.

Philosophically speaking, a public-private partnership is a means for governments and the private sector to share the risks and rewards associated with infrastructure investment. The risks these counterparties are comfortable assuming and the rewards they seek will determine the specific terms of the contractual agreement.

Availability versus demand risk models

One of the best discussions of the types of public-private partnerships that I have seen is in Moody’s Investors Service’s report, The Global P3 Landscape(September 8, 2014 — research subscription required). Please forgive me for quoting heavily from this report below.

There are broadly two types of public-private partnerships: availability and demand risk. Not unlike the types of municipal bonds I described in an earlier post, these categories reflect the source of the payments that a private investor receives from an asset. The risks associated with these approaches vary with different phases in the project cycle.

Moody’s:

What are availability payments? Once an asset is built to the specifications required by the government and the government accepts the project, the private developer is entitled to payments from the government as long as the asset is made available to the public at the standard required by the government. Availability payments are sized to cover operating and maintenance costs as well as debt service costs and equity returns.

Also important, availability payments are not subject to swings in demand, such as traffic levels in the case of toll roads, for example, and are adjusted only for lack of performance or availability to the public. The payments are usually subject to annual appropriation by the sponsoring government.

P3 availability-payment projects carry higher risk in the early years of construction, usually commensurate with Baa-level credit risk. They carry lower risk when the project reaches a steady state of operation, in line with A-level credit risk.

This would put the projects on par with many local governments credit-wise.

Moody’s:

What is demand risk? In this model, the government grants the private developer the right to collect fees from the public for the use of a road, a bridge, a subway, an airport. This right is sometimes called a concession, since the government is conceding the asset to the developer for a defined period. The private developer takes on revenue risk under this fee-for-service model, because it relies on fees for use of the asset to pay for operations and to pay back debt issued to build the asset.

One of the advantages to the government of using the demand risk model is that the developer has more of an incentive to complete the project on time or ahead of schedule. This is because the developer will not start collecting revenues until the project is operational. The availability model provides no such incentive because the payments the developer is to receive from the government have been predetermined. The ability to push off construction risk to the developer is a major advantage, since completing a project on time is one of the largest risks associated with a project in terms of cost overruns.

Another consideration of using the availability model is its potential credit impact. Some argue that availability payments could be construed as a form of debt because they involve a contractually obligated stream of payments.

Prevalence

The use of these models differs geographically. Almost all of the public-private partnerships in the United Kingdom and Canada use the availability model. Latin American projects mostly use the demand risk model. Both types can be found in the United States and Europe.

This can partially be explained by the kinds of projects involved. Some projects, like transportation infrastructure, can work well with a demand risk structure. Others, like social infrastructure (hospitals, school buildings, etc.), may only work with the availability model.

By Moody’s classifications, the United Kingdom, Canada, Australia, India, and Chile have mature markets for public-private partnerships. The United States, Brazil, Peru, Mexico, and Colombia are considered to have expanding markets for public-private partnerships.

The UK has the most expansive market for public-private partnerships globally. The UK began using public-private partnerships in the 1990s with roads and hospitals, and then moved on to schools and waste projects.

According to Moody’s, “More than 725 projects [in the UK] have reached financial close since the 1990s, one of the highest numbers in the world … Excluding the UK, Europe had over 500 new and primarily availability-payment P3s reached financial close from 2002–13 … Just over half of these projects are in Spain and France.”

Canada has the most mature public-private partnership market in North America and primarily uses the availability model. Mexico primarily uses the demand risk model.

Why doesn't the United States have a mature P3 market?

The United States has been late to the game. Moody’s:

Many US states have yet to authorize the use of P3s for transportation projects, the classic beginner market for P3s. Virginia, Florida, Texas, Indiana and Colorado are leading the charge, while P3 authorizing legislation has failed in New York and Kentucky. That said, the momentum of states authorizing the use of the P3 model has notably increased over the past five years: 33 states and Puerto Rico have P3 authorizing legislation for transportation projects, and 39 states have some form of P3 authorizing legislation, either for transportation or social infrastructure projects.

There are a number of reasons the United States does not have a mature market for public-private partnerships, although this is in the process of changing.

First, the United States has a long tradition of funding projects through the public sector as opposed to the private sector. Although there were toll roads even in the 18th century, the bias has been strongly in the other direction.

(For an excellent discussion of how the public versus private debate played out when the United States itself was considered an emerging market, I would recommend Carl Smith’s book, City Water, City Life: Water and the Infrastructure of Ideas from University of Chicago Press.)

Although this bias may be unconscious, I suspect it contributes to the negative press public-private partnerships often receive and the media and other observers’ general unwillingness to study the mechanics of risk transfer in these projects. The Indiana Toll Road is an excellent example of what I am talking about here.

Second, unlike the rest of the world, the United States has a multi-trillion-dollar municipal bond market. State and local governments can borrow with a relatively low cost of capital. State and local governments are limited in their capacity to borrow, however, as the revenues used to support these projects are crowded out by other political priorities, such as pensions, health care, and education. This is the major driving force behind governments seeking private investment.

Third, the United States has approached authorizing public-private partnerships in an ad hoc manner when other countries have standardized procurement processes and contracts.

It is also worth noting that there is a traditional preference for design-bid-build project delivery versus design-build project delivery models at the state and local levels. For states to incorporate public-private partnerships into their programs, they would have to become comfortable with design-build delivery models.

Advantages of public-private partnerships

So far, I have focused on risk and risk transfer in public-private partnerships. Now let’s talk about why they might be a compelling component in a capital construction program.

One thing I have always found interesting about opponents of public-private partnerships is that they tend to portray project risks as one-sided. They fixate on the risks associated with public-private partnerships and neglect to mention that there are risks associated with traditional procurement methods.

One risk is that the public sector (smaller governmental entities especially) is not aware of the latest innovations.

From Patrick Sabol and Robert Puentes’s paper, Private Capital, Public Good: Drivers of Successful Infrastructure Public-Partnerships:

While the public sector brings significant expertise to projects; many private sector firms have access to technologies, materials, and management techniques that exceed the capabilities of an individual governmental agency or department. PPPs are one way to harness the ideas and breadth of experience the private sector brings to projects by fully incorporating them into the procurement process.

Public and private sector collaboration from the outset of an infrastructure project, whether greenfield or brownfield, can lead to a number of innovations. These can come in the form of new materials, faster project delivery, increased use of technology, operational efficiencies, or enhanced building techniques. An open PPP procurement process, at minimum, provides the possibility for new ideas that the public sector may have never considered.

Another risk is traditional methods offer limited control over the pace of construction. The increased costs associated with project delays and blown budgets are a good reason to differentiate between up-front costs and value for money:

PPPs are rarely the lowest-cost way to procure infrastructure for several reasons. For one, the transaction costs for PPPs are usually higher than traditional bid-build contracts, which average around 10 percent of the entire value of the project. Plus, private sector borrowing costs are generally higher than those available to the public sector, as governments are able to access the tax-exempt municipal bond market. Despite these limitations, a well-structured PPP can deliver better value for the public dollar. This value can be derived in a number of ways.

Driven by the need to deliver profit to investors and shareholders, the private sector is less tolerant of cost overruns and project delays than the public sector. Therefore, transferring construction, operational, and/or demand risk to the private sector can result in quantifiable savings for the public sector, as taxpayers and ratepayers do not bear the costs if the project takes longer than expected to complete, goes over budget, or underperforms. The company or consortium that assumes responsibility for the infrastructure asset may also opt to invest in more durable materials or efficient technologies that drive down lifecycle costs. These might not be the cheapest options in the short term, but they do have the potential to drive savings over the long term through decreased energy usage, lower maintenance costs, or enhanced resiliency.

These are all important considerations because the kinds of assets we are talking about have long useful lives.

The fact that developers are profit-driven and have to answer to shareholders also reduces or eliminates the potential for corruption. Although true financial distress is rare for state and local governments, when it does occur it is usually attributable to fraud and incompetent managment. Jefferson County, Alabama, and Harrisburg, Pennsylvania, were both left insolvent due to corruption in undertaking and financing major construction projects.

State and local governments also tend to cut capital investment following recessions. This has the effect of reducing the useful life of existing infrastructure. Having a private entity handling operations and maintenance for an asset insulates it from particularly lean budget cycles.

Furthermore, public-private partnerships provide an opportunity for governments to spin off the responsibility of operating and maintaining facilities that are auxiliary (as opposed to essential) government functions. Some examples would include housing and dining, athletic, and recreational facilities at public universities; parking lots and garages; and publicly owned water and electric utilities. These operations are perhaps already better suited for the private sector and can divert scarce resources from government budgets.

(This is part five in a series on infrastructure investment.)

Legislation to create a national infrastructure bank (NIB) was first introduced in Congress in 2007 and has been repeatedly reintroduced. The concept has also been promoted by the Obama administration. Several OECD countries have infrastructure banks and some US states have experimented with them on a smaller scale.

Proponents of this legislation believe that an NIB would (1) increase infrastructure investment over time without requiring significant new federal appropriations and (2) protect infrastructure investment from the political process.

Many elaborate defenses of these proposals have been written, but they all have the same core components. In this post, I am going to pick apart the the model provided in William Galston and Korin Davis’s paper, Setting Priorities, Meeting Needs: The Case for a National Infrastructure Bank.

Corporate structure and governance

They begin:

Most proposals would establish the bank as an independent government-owned corporation (GOC) … A GOC is a government agency, established by Congress to provide a market-oriented public service, whose revenues make it financially self-sustainable.

While some proposed legislation would place the bank within an agency, such as the Department of Transportation, we recommend making it independent so as not to narrow the scope of infrastructure projects it could support. Establishing it as a GOC — as opposed to a regular government agency within a department, as is the case with TIFIA — would also provide it with greater budgetary flexibility …

That is not to say it would have complete budgetary freedom. Under the Government Corporate Control Act, it would be required to submit a budget to the President, who in turn is required to include that budget with the executive branch budget he submits to Congress. That budget would include “estimates of the financial condition and operation of the corporation for the current and following fiscal years” as well as information on “income and expenses, sources and uses of money, an analysis of surplus and deficit,” a statement on the overall financial condition of the corporation and a summary of its major activities …

The point of invoking all of this machinery, of course, is to distinguish the NIB from government-sponsored enterprises (GSEs) like Fannie Mae and Freddie Mac, which became major liabilities in the wake of the financial crisis.

The authors suggest that a GOC would not involve the same perceived bailout risk of a GSE. I am not sure this is a fair claim to the extent that the NIB is funding an essential government service. A quasi-governmental agency that is providing an essential government service will always be perceived politically as worth saving. This political risk would be compounded if the NIB offered loan guarantees, as others have proposed, because its financial failures then become contagious if the entity is not bailed out. (Providing loan guarantees in any context has always struck me as bad policy. Moral hazard, anyone?)

It is also worth noting that the Transportation Infrastructure Finance and Innovation Act (TIFIA) program mentioned above is oversubscribed relative to the funding it has received and its counterpart for water infrastructure, WIFIA, effectively hasn't been funded yet.

Continuing:

While there is no uniform governing structure for government-owned corporations, infrastructure bank proposals build on a common model. Most provide for a CEO and board of directors, some nominated by the president, others by the leaders of the two parties, confirmed by the Senate, serving staggered terms of about six years. We recommend such a structure, which would give Congress some oversight authority but would sufficiently insulate its operations from political whims and create enough of a buffer so that elected officials would neither determine strategic and project choices nor be called on the carpet for unpopular or controversial decisions.

The idea that lawmakers would not try or be able to interfere with the operations of an NIB seems naive. This kind of situation plays out on a regular basis with other ostensibly independent entities — the Federal Reserve, for example, will never have a respite from asserting its autonomy. In fact, the nomination of a community banker with only a bachelor’s degree and a CPA to the panel that sets monetary policy is evidence of how such organizations can be manipulated politically.

If Congress can legislate an entity into existence, it can legislate an entity out of existence. That simple fact also carries a lot of weight. And, as we've already established, an NIB would be on-budget…

Continuing:

A strong and permanent professional staff would be vital to the success of the bank. We recommend adopting the European Investment Bank’s model by creating a division of the bank responsible both for analyzing the viability of proposed projects and for advising those seeking support. As is the case with the EIB, staff should be capable of providing expert advice on both the financial and technical aspects of the project. Financial advice would not have to be limited to those projects receiving bank support, but instead could help facilitate financing for projects which do not meet NIB criteria or are denied financing for other reasons …

The sector-specific expertise of the professional staff would far exceed that of the bond ratings agencies, generating more reliable estimates of potential risks and returns … The analytical staff would provide policy entrepreneurship, identifying unique opportunities to leverage private capital for public needs. Once such possibilities are identified, bank executives could convene the parties whose participation would be needed and work together to broker agreements with them.

I do not think the authors appreciate the size of the bureaucracy they are describing here and how much it would cost. They are talking about building — from scratch — an agency with staff that understands the physical elements of all varieties of infrastructure projects; the legal nuances affecting project finance in each of the fifty states and local jurisdictions (since this seems to be anything-goes from a contract perspective); regional clashes and overlap; etc etc.

In a sense, these proposals would duplicate swaths of state government. This expertise already exists at the state level, which is comme il faut. States need in-house staff to manage their large capital construction programs.

The remark about the rating agencies is also inaccurate. Rating agencies already have divisions devoted to evaluating and monitoring project finance, including sector-specific analysis. As I will discuss in my next post, private investment is much more common elsewhere in the world, where countries do not have equivalent municipal bond markets. And the rating agencies are global enterprises that cover these markets. Whatever reputation issues the rating agencies may have, investors would likely trust their opinions more than the staff at a start-up quasi-governmental agency.

Lending activities

Continuing:

To achieve its objectives, the bank would require an initial tranche of public funds. Annual appropriations of $5 billion for the first five years would create a pool of $25 billion. Administrative costs could be covered by application and transaction fees.

To achieve leverage, the new entity would have to attract private investor-depositors as well. Brookings financial expert Douglas Elliott estimates that the up-front public capital could bring in as much as $125 billion from private sources, generating a lending pool of $150 billion.

Proponents of an NIB are essentially copying the structure used for the drinking water and wastewater state revolving loan funds that are regulated by the Environmental Protection Agency. These funds are capitalized by state governments with a fractional grant from the federal government. The state governments use the funds to make loans to local government entities and utilities to build wastewater treatment plants, water mains and the like, with borrowers’ payments returning to the fund to be used for future loans. Although these funds are an important resource for financing infrastructure projects, they were established decades ago and have not yet become self-sustaining as proponents claim an NIB would be.

An NIB is essentially one of these revolving loan funds with a miniature, un-elected legislature attached to select which projects receive financing. And proponents have skipped ahead to leveraging the program, which is philosophically analogous to what some states have done through the issuance of asset-backed securities against their drinking water and wastewater loan portfolios.

For a state to leverage its revolving loan fund, the fund has to be relatively mature with an established income stream. This will also be true for an NIB, regardless of how investor participation is structured. Private capital is not magic. The federal government does not deposit $25 billion into a fund and the infrastructure fairy turns that into $150 billion. Investors expect a return for their money. The program would not require larger federal appropriations because the costs associated with these projects are being pushed off to state and local residents. And in addition to project costs, they are paying for their share of a superfluous layer of bureaucracy.

This structure would provide a service that is already provided by the municipal bond market in a less expensive way. Bosworth and Milushevaexplain how an NIB fails from an economic standpoint:

Given that [state and local] governments already have access to low-cost tax-favored borrowing, an infrastructure bank is not really an attractive financial innovation in the US context. Because of the size and liquidity of its municipal bond market, the US situation is different from that of other countries that have used infrastructure banks as a means of providing sub-governments with access to financial markets. It is difficult to see how the NIB could function as a lender except by offering subsidies in excess of those implied by tax-exempt bonds. Yet it will face a cost of funds that normally exceeds the tax-exempt rate. Thus it is hard to visualize the NIB operating as a bank making loans to S&L governments, unless it envisions limiting S&L access to the market for tax-exempt municipals. [KC: At which point, the federal government would need to pony up enough money to replace the entire market until some grand equilibrium was achieved.] It might be able to compete with private lending sources because, even if it renounced the notion of an explicit federal guarantee, investors will attach an implicit guarantee to its activities — as occurred with the government-sponsored agencies in the mortgage market. Yet past experience suggests that implicit guarantees have a substantial probability of becoming real costs.

This isn't to suggest that the municipal market in the US is an obstacle to this approach. This just clarifies my earlier point that the US has a problemfunding infrastructure, not a problem financing it. An infrastructure bank as a financing mechanism is indistinct from how the bond market functions. It isn't the infrastructure bank that works well in other countries — it’s the fact that other countries have relatively higher taxes to fund infrastructure.

(This is the fourth post in a series on infrastructure investment.) 

There will be considerable high finance discussion in this post, but I have tried to make it as readable as possible. As I said in my first post, a grasp of the larger framework is necessary to understand where infrastructure policy succeeds and where it fails.

One must imagine Sisyphus happy

Federal lawmakers love to propose and create new tax credit bond programs despite the fact that they are highly unpopular with both government issuers and investors (so basically the entire market). Tax credit bond proposals date back to 1968 with Senator William Proxmire’s guarantee scheme. Seriously, that’s how long this game has been carrying on.

I don’t understand why members of Congress continue to invent investment vehicles that will never amount to anything more than a niche product, but I suppose everyone needs a hobby. Besides, one does have to be a little in awe of any policy concept that has remained completely detached from reality for nearly half a century.

Since I don’t enjoy spending my limited time on this planet swatting down zombie legislation —well, unless it has been bitten in an interesting way — I am going to outsource explaining why tax credit bond proposals are terrible policy to Bond Dealers of America and George Friedlander. This post will be about why the only tax credit bond proposal that had been broadly regarded as effective — Build America Bonds (BABs) — won’t be coming back.

BABs were a bond structure created by the American Recovery and Reinvestment Act of 2009 (federal stimulus legislation). Essentially, state and local governments were permitted to issue bonds to finance infrastructure projects on a taxable — rather than tax-exempt — basis and receive a direct payment from the federal government equivalent to a 35 percent interest rate subsidy. The subsidy was intended to compensate state and local governments for the lower interest cost they would have received if they issued the bonds on a tax-exempt basis.

(Many people considered 35 percent an overly generous subsidy from the federal government’s perspective. Considering that the federal government is responsible for less than a quarter of total infrastructure investment, I am not one of those people.)

The provisions that allowed state and local governments to issue these bonds expired at the end of 2010. The program was initially considered a great success. The White House and some in Congress have introduced legislation that would make the program permanent, sometimes with a decreased subsidy and with similar monikers like America Fast Forward bonds.

To understand why this program — in whatever form it may be reincarnated — is unlikely to be utilized by state and local governments going forward requires a bit of market history.

An overview of pre-financial crisis innovations

The BAB program was politically sold as a means to create and preserve jobs just like every other piece of Great Recession-era federal legislation. Anyone who understands the municipal bond market’s mechanics and history of financial innovations, however, recognizes that this was hardly the purpose of the program. The purpose of the BAB program was to prop up the municipal bond market amid a period of intense distress within the financial markets generally, during which the investors that had traditionally participated in the municipal market either had been sidelined by their individual capital constraints or evaporated altogether.

State and local government borrowers tend to issue long-term bonds so their debt burdens do not crowd out their programmatic priorities or require them to levy additional taxes. This is similar to how a homebuyer would prefer to finance a house with a loan that accommodates their cash flow constraints and reflects the long useful life of the asset rather than a short-term note.

As I mentioned earlier, the municipal bond market functions as a closed system due to tax policy. But a handful of innovations that were forcefully marketed in the early 2000s created shadow demand (opening the market to investors) essentially by converting the long-term debt that governments want to issue into short-term (and, in theory, more liquid) debt that some investors want to buy.

In the early 1980s, municipal issuers began using two types of variable rate debt — auction rate securities (ARS) and variable rate demand obligations (VRDOs) — as an alternative to traditional fixed rate bonds. ARS are a type of security that has a long tenor (typically 20+ years) but bears interest at a rate established at short-term intervals through an auction process (or at a penalty rate specified in bond documents in the event that there are not sufficient bidders to cover the amount of securities offered for resale).

VRDOs are similar to ARS in the sense that they reset at short-term intervals, but investors’ liquidity stems from a put feature. With VRDOs, investors may put the bonds back to the trustee or tender agent (effectively force the bonds to be purchased at face value plus accrued interest) with little notice. Since most issuers would not easily be able to accommodate an event where substantially all of the debt suddenly could not be successfully remarketed, the issuer pays a commercial bank to support the debt through a letter of credit or standby bond purchase agreement.

From 2002 to 2007, 29% of all municipal borrowing was in the form of ARS or VRDOs.

There certainly has been no shortage of commentary regarding the grotesque denouement of the ARS market, or as some of us fondly called it, the Valentine’s Massacre. (In fact, I am going to assume you already know how the market collapsed and not repeat it here.) Few commentators, however, mention why these complex instruments became popular in the first place.

Some municipal market issuers wanted or required variable rate exposure (issuers of student loan asset-backed securities, for example). Also, although it complicates government issuers’ budget processes, variable rate debt exposure can be perceived as a positive from a portfolio management standpoint. Issuers who actually wanted a fixed rate coupled their variable rate debt issues with interest rate swaps to achieve a synthetic fixed rate. (This, of course, meant that they assumed a litany of other risks.)

But the main reason these structures suited the municipal bond market was because they increased demand for the securities (which had the effect of lowering borrowing costs for the market as a whole) by allowing borrowers to issue debt with long-term nominal maturities and investors to buy debt that mimicked short-term debt with respect to interest rate risk and liquidity. Issuers and investors both had the commitment structure they desired until the financial markets ceased to function properly. Banks also had significant new sources of fee income (for the credit facility and task of remarketing the debt, for example).

So we have VRDOs, ARS, and interest rate swaps. Now let’s add a fourth innovation.

Whereas the use of variable rate debt instruments “found” demand for the long-end of the yield curve by disguising short-term investors as long-term investors, tender option bond programs (TOBs) created shadow demand for long-term municipal debt through the use of leverage. TOBs were similar to structured investment vehicles (SIVs) in strategy, in the sense that they involved issuing short-term securities at low interest rates and purchasing long-term securities at higher interest rates, with the spread minus expenses going to investors as a profit.

(Actually, this is not-so-plain vanilla but ultimately still kind-of-plain vanilla banking.)

Here is a (very) rough sketch of how a TOB program works:

A [hedge] fund manager invests contributed capital in high-quality assets [i.e., municipal bonds] pledged as collateral to a dealer who forms a trust and issues two classes of securities: senior short-term floating-rate notes [VRDOs], and junior notes called residual certificates. The proceeds from the sale of the short-term notes are used to purchase additional long-term municipal bonds thus leveraging the residual certificate holders’ exposure to the long-term municipal bonds held in the trust.

The interest rate paid by the trust on the [VRDOs] is reset weekly and is equal to the short-term municipal yield benchmark less a market-determined spread … The junior residual certificates are retained by the hedge fund [or are sold to investors who do not have interest in the underlying collateral] and pay an interest rate equal to the difference between the interest rate on the underlying long-term municipal bonds and the interest paid to the short-term, senior noteholders, less the trust’s expenses…

Hedge funds used interest rate swaps to hedge both the long-term and the short-term interest rate risks.

Ideally, this arrangement would have protected the fund from a situation where market yields for municipal bonds increased (thus causing the value of the bonds held in the trust to fall) or where short-term yields increased (thus reducing investors’ net interest). However, during the financial crisis, the relationship between municipal bond yields and swap rates became unhinged. The funds suffered significant losses as declines on the municipal debt were not offset by gains on the swaps.

According to the Bond Buyer, “at their peak, [TOBs] had grown to anywhere from $300 billion to $500 billion, soaking up as much as a quarter of annual municipal bond issuance some years.”

All of these sources of shadow demand were annihilated at the same time traditional investors in municipal bonds began to hoard money. Structurally, the BAB program provided a bridge to a normal market in the future by once again introducing alternative sources of demand for municipal bonds, such as foreign investors and pension funds, who would otherwise not benefit from a tax exemption.

If you accept that this was the purpose of the program, then BABs were successful. This is different than saying that a BAB-like structure could replace the market in its entirety, which is what some have proposed.

Not a substitute for the tax-exempt bond market

First, during the two years the BAB program was in place, many issuers opted to issue tax-exempt bonds. BABs represented one-third of municipal issuance from 2009–10. In 2009, $409.7 billion (par amount) of new bonds were issued, of which $64.1 billion were BABs. In 2010, $433.3 billion of new bonds were issued, of which $117.3 billion were BABs.

Of the $181.5 billion of total BABs issued, 46% was issued by California, New York, and Texas. In fact, $37.8 billion of the total was issued by California alone.

The BAB program mostly diverted the supply (new bond issues) generated by the market’s largest borrowers. During periods where there is considerable supply in the market (i.e., the opposite of what is happening in the market right now), having an issuer like California go to market is like having a blue whale jump into a swimming pool. Other issuers watch the calendar to avoid being the pool themselves.

California can plausibly market its debt to foreign investors and pension plans because the state is one of the largest economies in the world. California’s debt is even a substitute for sovereign debt in that sense. But a small school district in Kansas is not going to be pitching to investors in France anytime soon. (Although wouldn't that be entertaining?) Smaller issuers need a structure that reliably gives investors a reason to purchase municipal bonds over alternative investment vehicles, and tax incentives accomplish that quite well. As I mentioned in an earlier post, there are tens of thousands of issuers in the municipal market.

Federal appropriation risk

Second, another practical problem with broadly transitioning to a BAB-like structure is that it would expose the entire municipal bond market to federal appropriation risk.

When the BAB program was in place, it was impossible for an issuer to receive a decent rating on a BAB issue without including in the bond indenture a pledge for gross debt service and not debt service net of payments from the federal government. This pledge would protect bondholders in the event that the federal government chose not to make good on its promise to provide the interest rate subsidy.

This feature was somewhat problematic from a credit analysis perspective, because it effectively gave issuers two debt capacity ratios: one for if the federal government appropriated the amounts that it promised in a timely fashion and one for if it did not. None of this is an issue with tax-exempt debt.

BABs are one instance where rating agencies’ methodologies have turned out to be prescient. (Whatever their problems with structured finance outside the municipal market might have been, the rating agencies have done a mostly good job of covering the municipal bond market.) The federal government has arbitrarily reduced its subsidy payments to the governments that issued BABs on multiple occasions. This has prompted many issuers to refinance their outstanding BAB issues into traditional structures. It is also a back office nightmare for state and local governments that access the market routinely.

Much of the “success” of the BAB program was due to native bond issuance in the market being pulled forward. The program was expiring and financial conditions in 2010 were still uncertain enough for state and local governments to take advantage of the market while this structural support was still in place.

Probably the only positive legacy of the BAB program was that it increased the number of crossover buyers in the municipal market as they became familiar with government credits. (A crossover buyer is an investor that would traditionally invest in taxable corporate bonds, but will “crossover” to the tax-exempt market to exploit special situations.)