Friday, November 02, 2007

Municipal Bond Insurers: More dangerous than you realize

I've written before about bond insurers, and I reiterate my view that the big 4 (MBIA, FSA, FGIC, and AMBAC) will survive. But its certainly not impossible for any of them to falter. There is some non-zero probability that one of the four had weaker credit standards, and wound up underwriting policies on all the worst RMBS/CDO deals. I have no evidence that this was indeed the case, but then again, I can't prove it wasn't the case.
Anyway, as readers undoubtedly know, the big bond insurers were known primarily for insuring municipal bonds, and their CDO/ABS insurance business was merely a sideline. So just for the hell of it, let's consider what would happen if one of the big 4 bond insurers failed.
First, let's talk a bit about the municipal market. Its by far the least efficient of the major bond sectors. There are several reasons for this:
  • Munis are hard to short, so if munis become overvalued, its hard to bet against them.
  • The tax advantage of municipals can't be arbitraged. Or put another way, if municipals become undervalued, you could go long munis and short Treasuries, but because munis retain a tax advantage, such a trade would still have negative carry even before considering financing costs.
  • Muni demand is still mostly retail, either directly or through funds. Retail demand for bonds tends to be erratic.
  • Municipals from different states have different supply and demand conditions, causing trading levels to vary significantly.
  • No two municipals are exactly alike in structure, especially long-term municipals which are almost always callable. Since exact maturity, coupon, call price and call dates vary, two structures are never the same.
  • Deal sizes are too small for there to be frequent trading in all but a few issues. This makes it hard to determine the going level for a bond.
  • There is very limited credit information on issuers. Most don't have to report anything about their financials to anyone other than ratings agencies. Some have to make occasional financial reports available, but only if bond holders demanded it at time of issuance.

Despite all these limitations, the municipal market does have one key advantage over other sectors: bond insurance. Insurance on a municipal bond deal is pretty simple. The issuer pays an up-front fee and in exchange, the insurer agrees to pay timely principal and interest in the event that the issuer cannot. If the original issuer remains solvent, then the insurance policy never comes in to play. Since a large percentage of deals are insured, investors take the credit quality of municipals for granted. (Are alarm bells going off? Good.)

We've seen what happens when a group of investors thought they could take credit quality for granted, then suddenly they couldn't. For example, AAA Home Equity...


What if one of the major insurers defaults? Suddenly investors wouldn't take the credit quality of any insurer for granted. Could there be a repeat of the ABX price action in municipals? Well, not quite the same, since the credit quality of most municipal issuers is inherently pretty good. Moody's reports that there has never been a general obligation issuer (meaning a issuer with theoretically unlimited taxing power like a state, county or city) actually default on its obligations. Including Orange County CA, which defaulted on some pension obligations but never a general obligation.

But considering the inefficiency of the municipal market, having an insurer default would create serious and probably permanent repercussions in the muni market. Remember that unless the issuer is simultaneously bankrupt as well, the default of the insurer doesn't cause any interruption of cash flow. In fact, in most cases the ratings agencies issue a "underlying" rating, which tells investors what the rating would have been sans insurance. If the insurer was suddenly out of the picture, the bond would merely be downgraded to its underlying rating. Perhaps surprisingly to those not in the muni market, most insured bonds have an underlying rating of AA or A.

Anyway, so what would happen if an insurer was bankrupt? First of all, retail investors would likely call up their brokers and demand to sell any bond insured by the defunct insurer. The result would be that the (formerly) insured bonds would trade weaker than uninsured bonds with the same underlying rating. Don't believe me? Radian insured bonds are already trading weaker than their underlying, and Radian is still in business.

The other immediate effect will be a price searching process for insured bonds, particularly in sectors considered more risky, like health care and project financings. Much like what went on in ABS this summer, potential sellers of bonds will have to feel out where the bids are, and potential buyers of bonds will be extremely cautious and extremely cheap when bidding.

Next, muni buyers would start questioning the value of bond insurance in general. If the insurance doesn't absolve the buyer of doing credit research, buyers will want to be paid for their efforts. If buyers won't assign a substantially lower rate to insured bonds, then issuers won't bother to pay for the insurance. All insurers, regardless of their actual financial strength, will see new business plummet.

All this will cause a very murky market to become murkier. That will create a wonderful buying opportunity for those with the cash and the confidence in their analysis. Wider bid/ask spreads tend to favor long-term buyers over other players, because wide bid/ask almost always indicates fear. You'll have to have the stomach and the job security to live through a bumpy ride.

What will be the long-term consequences? Possibly a decrease in retail buyers owning bonds directly. If retail investors don't feel confident in the quality of what they're buying, you'll likely see them rotate into funds.

Like I said in the beginning, this is a low probability scenario. But its not impossible.

15 comments:

  1. An article by David Einhorn posted at NakedShorts ( http://tinyurl.com/2phqel ) discusses the current credit mess and makes some points that appear to both support and not support your take on the municipal bond insurance issue. I'm not familiar enough with the topic or Einhorn to add anything useful but thought you might be interested.

    Here are a couple excerpts:

    "Without much fanfare the rating agencies abandoned this practice of AAA meaning AAA and BBB meaning BBB. Instead for each type of bond, they use a different rating scale with different so-called 'idealized default rates' for each rating. The idealized default rate for a municipal bond at a given rating is less than the idealized default rate for a corporate bond which is less than the idealized default rate for an asset backed security which is less than the idealized default rate for a CDO."

    IOW, municipal bonds are systematically under-rated while asset-backed securities are systematically over-rated. A couple paragraphs further down:

    "The municipalities purchase bond insurance to enhance their credits to the AAA level. Of course, since they are in fact, AAA to begin with the insurance provides no true benefit. I assure you that a quick peak (sic) at the balance sheets of any of these so-called AAA rated bond insurers will tell you that they are not likely to be there to pay more than a fraction of the claims they have insured in an environment where there are wide-scale defaults in the municipal bond sector."

    ReplyDelete
  2. As an owner of a number of individual muni bonds, (some of which are G.O.'s) I've always wondered the need for insurance for general obligation bonds. With the unlimited taxing power of ad valorem taxes, the municipality really doesn't need the insurance with these type of bonds.

    ReplyDelete
  3. David Einhorn's comments become even more trenchant WRT municipal bond insurance (but this seems more of an aside - not the main topic of his article). For example:

    "If municipal bonds are much safer than their ratings imply, it means that all kinds of states, cities and towns ...are systematically overcharged for borrowing. ...costing taxpayers $5 billion per year. [assuming the average municipality pays 20 basis points excess due to lower rating]

    The misrating of municipal bonds directly benefits the friends of the rating agencies on Wall Street, the banks who underwrite the deals. A lower rating - means bigger underwriting fees. Or alternatively, the excess cost can be shared with another great friend - I mean very large customer - of the rating agencies, the municipal bond insurers effectively rebate some of the 'overcharge' to the municipalities in exchange for a share of the savings through a scheme called 'bond insurance.' ..."

    I don't deal in individual muni issues and couldn't make enough sense of Einhorn's short description of "idealized default rate" modeling to even come up with a plausible hypothesis regarding the positioning of municipal bonds at the bottom of the risk totem pole under corporates, ABS and CDOs (at any given rating).

    Returning to our host's point though, if muni's do have the limitations he listed then the insurance could basically exist to compensate the investor for lack of information but that doesn't explain the lower rating because the municipality does have to provide the rating agency w/ all that; the triangle between rating agency, investment bank and insurance company doesn't make sense (or it's just good old capitalism at work, whatever).

    Still, if Einhorn's characterization is accurate and understood in the bond marketplace then the failure of a bond insurer or two would have no substantive impact because everyone 'knows' they can't cover the bet; it would be more like a canary in the coal mine, a sign that a system-wide breakdown was occurring perhaps?

    ReplyDelete
  4. I love comments like "..wide-scale defaults in the municipal bond sector." Do you realize what it would take for this to happen? After Katrina, there was not a single default in Louisiana. So for wide-scale defaults, we're talking about once in a 10000 year occurrence. About half of all muni bonds issued have insurance. Investors have a choice as to what to purchase - many choose AAA bonds. PB

    ReplyDelete
  5. Great blog... just ran across it for the first time and your posts are excellent.

    As someone who is looking to take a long position in Ambac (ABK) stock, can you give me your OPINION (yes I realize it's an opinion so feel free to say whatever is on your mind :) ) of the following? Also, I invest in stocks but would like to hear your thoughts given that you approach it from a credit point of view.

    Do you think the market doesn't understand the bond insurers' business? Or is there a real genuine concern over the viability of Ambac, MBIA, etc? One of the reasons I'm bullish on these companies is because they primarily insure the interest and principal payments. These companies should not face a liquidity crisis since they can easily cover those interest+principal payments. Some people are analyzing these firms as if they have to take billion dollar losses in a short period of time (like hedge funds, banks, etc) when in fact the mark-to-market losses aren't real. What am I missing here?

    I'm just a newbie investor but what is happening doesn't seem to make any sense. For instance, why is Ambac's CDS trading in deep junk territory? Based on your bond market experience, does this seem like overreaction and blood on the streets?

    Thanks...

    ReplyDelete
  6. Doug Kass has written and talked extensively about these companies. He disagrees. He believes the chances for bankruptcy for all of these companies, especially MTG, is very high.

    ReplyDelete
  7. Anon,

    do you have any free website links that have views of Doug Kass (or any other bear) on how the debt insurers can go bankrupt.

    I just don't see how they can go bankrupt given that they primarily insure the interest and principal payments. All these payments are also spread out over a long period of time. How can they face a liquidity issue?

    Either I'm missing something or all these guys are wrong. What am I missing? Anyone care to elaborate how these bond insurers can go bankrupt?

    ReplyDelete
  8. Some Insurers levered 90:1

    ReplyDelete
  9. In CDOs, their exposure is not the same as municipal insurance contracts. A CDO insurnace contract is similar to a CDS, where the insurer has to buy the security at par given a default event.

    I believe the CDO CDS are structured as a "pay as you go" CDS, which means the insurer doesn't buy the whole CDO given a default event. But still, its more cash flow problematic than classic muni insurance policies.

    For some reason, ABK stock was up sharply today. No such action in MBI. Dunno why.

    Anyway, I think there is a reasonable chance one of the insurers built a bad portfolio. Not a high probability, but some probability.

    ReplyDelete
  10. So how can you be in the financial guarantee business if everyone thinks you are going out of business? Aren't 1 yr prob. of defaults on ABK at around 50%?

    ReplyDelete
  11. Supposedly, Bear brought an AZ Trans deal today uninsured. They couldn't find anyone who would bid the bonds tighter if insurance was attached. Therefore from AZ's perspective, the insurance policy was worthless. AZ Trans is AA rated on its own.

    ReplyDelete
  12. I am retired and have a significant portion of my assets in a Cal Muni ladder. all AAA and insured, mostly GO. How do I react to the MBIC and Ambac downgrades in a rational way? How serious is this?

    ReplyDelete
  13. Anon:

    Do you have a broker that you trust? If so, ask him/her to find the "underlying" rating on all your pieces. If you have trouble, e-mail me at accruedint at gmail.com.

    ReplyDelete
  14. Thanks for your input. I will get the underlying rating. BTW is there a site where I can analyze my portfolio to get more insight into the quality of my portfolio?

    ReplyDelete
  15. Unfortunately, I don't know where on the web there is good muni research. You can go to www.munios.com and find many municipal official statements, but not all.

    ReplyDelete

Comment rules:
All comments must contribute to the conversation
All comments should be civil
No comment should include any personal attacks, however minor, on the author or other commenter.
Do not hawk your own website unless its a specific reference to the article
If you post anonymously, please give some identifyer
I will delete any comment which doesn't fit this criterea