Why muni investors shouldn’t worry about Jefferson County
There’s a distinct whiff of the faux-naive coming off Mary Williams Walsh’s article about the fate of Jefferson County’s general-obligation bonds:
People who own what is considered the safest type of municipal bond may be in for a surprise.
This safe debt, called a general-obligation bond, is said to be the next strongest thing to Treasuries because it is backed by a “full faith and credit” pledge. That means the government that issued it will pay it on time, no matter what.
But now Jefferson County, Ala., has stopped paying such debt, breaking with convention and setting up a fundamental test of what full faith and credit truly means.
While we’re repeatedly told what “conventional wisdom” has to say on such matters, or what certain fund managers might once have been taught, not once is any authority actually cited saying that GO bonds are “the next strongest thing to Treasuries”.
Which is hardly surprising, since the muni market hasn’t been considered particularly strong for some time now. Back in February I blogged Michael Corkery’s article on how individual investors were abandoning the market, precisely because they had become aware that it was anything but safe. When 60 Minutes is running alarmist pieces on how hundreds of billions of dollars of muni bonds could default, no one’s sitting back with a smug expression and saying “well, your muni bonds, backed by actual cashflows, might default, but my muni bonds, backed by some amorphous ‘full faith and credit’, are perfectly safe”.
And in truth no one ever said that — it’s hardly been a closely-held secret that Jefferson County has no tax-raising abilities. In fact, that’s one reason why revenue bonds came into being: as in most other markets, there’s a strong case to be made that secured bonds are safer instruments than unsecured bonds.
Indeed, if you carry on reading through the end of the NYT piece, you’re eventually told that Jefferson County’s MBIA bonds were wrapped (that is, insured) by MBIA. Obviously, if everybody thought they were perfectly safe, then no one would have demanded them to be insured.
But the biggest problem with the NYT article is its deep premise: that Jefferson County is some kind of harbinger, and that if it can default on its GO bonds, then lots of other municipalities can as well. This meme is both very pervasive, and very dangerous, as Bond Girl explained in a long blog entry in October:
Some people mistakenly characterize Jefferson County’s financial problems as a canary in the coalmine for the municipal bond market, which suggests that they still have no idea what transpired there (or how long Jefferson County has been in financial distress). Portraying Jefferson County as a typical municipal credit is akin to portraying Enron as a typical corporate credit. With Jefferson County, various financial firms – but primarily JP Morgan – exploited an existing culture of corruption and made taxpayers the victims of one of the largest frauds in the history of the financial markets.
I’ve been saying at least since at least April 2009 that munis are one of those asset classes which is safe until it isn’t — that once you get some unknowable number of municipal bond defaults, suddenly no one will have access to the market any more, and the whole market could implode:
If five or six munis default, things get much, much worse. At that point, the cost of default for a wrapped muni issuer plunges, and possibly even goes negative. Once a few munis default, no one’s going to lend to any muni, even the ones which are current on their debt. So why bother staying current? Why not just default and let the insurer, rather than your local taxpayers, take most of the pain?
In other words, there’s a very serious, and pretty much impossible to hedge, risk of snowballing muni defaults.
But the point here is that Jefferson Country is sui generis and emphatically not one of those five or six munis. It’s a case unto itself, cut through with corruption and fraud, and is in no sense an example for any other municipality to follow. Jefferson County’s GO bonds are not an example of the safety of GO bonds more generally: as Bond Girl says, they’re more akin to the way that people lost their money with Enron or Madoff. The germane risk in Jefferson County is fraud risk, not GO-bonds-defaulting risk.
And remember that no one knows what the recovery value is likely to be even on Jefferson County’s bonds: it could yet turn out to be quite high.
Now I’m no great believer that bondholders should be senior to everybody else, to the point at which bond coupons are more important than vital municipal services. But let’s not look to Jefferson County to blaze a trail on that particular front. What’s going on in Jefferson County is an interesting datapoint, but it’s in no sense the beginning of some kind of muni-default snowball.