The controversy surrounding the financial guarantors has been one of the most interesting, intellectually stimulating debates to emerge in the financial services industry in recent years. On the one hand, the bears say the companies are goners. They’ll soon be hit with a tsunami of claims as the subprime mortgages that underlie the collateral debt obligations they’ve insured default in wave after wave. The bulls (of which I am one) suspect that the level of eventual claims won’t be nearly as high as the latest hysterical predictions suggest and that, in any event, the companies won’t be wiped out even under very severe circumstances. Smart investors are lined up on both sides of the issue.
As it happens, things came to a head of sorts this past week when major industry players, key critics, and regulators testified in front of the House subcommittee on capital markets and insurance. Witnesses’ testimony was revealing. Here are some quoted highlights, with my take:
Eric Dinallo – NY Superintendent of Insurance:
The primary goal of insurance regulation with respect to financial oversight is to ensure that the insurer maintains an adequate level of solvency and is able to honor policy holders’ claims.
In the case of bond insurance, there are two groups of policy holders. There are the municipal governments who bought insurance. . . Then there are the banks and investment banks who bought insurance for the structured securities.
While we are doing all that we can to protect all policy holders and strengthen the bond insurers, if it becomes clear that it is not possible our first priority will be to protect the municipal bond holders and issuers [emphasis added].
Wow! Where exactly does it say in New York State insurance law that the chief regulator gets to pick and choose which class of policyholders actually receives the protection they paid for, and which ones deserve to have their claims dismissed? What is it about municipal bond holders that makes them more entitled than the retirement savings plans of large employers? Nothing obvious, in my view. Regardless of which type of investor, they all bought bonds that were guaranteed by a monoline with a stated equity value and claims-paying capability. But now the New York State Insurance Superintendent believes he can go back and retroactively change some of these contracts at will. Maybe he can, but it doesn’t seem fair to me, or sensible, for that matter.
To see why, take the Dinallo proposal to its logical conclusion. By his thinking, any time a guarantor faces a claim, it could just place the policy within a new, separate entity with insufficient equity to pay the claim—all in order to “protect” the vast majority of the policy holders. Obviously, this couldn’t happen.
William Ackman, Managing Member, Pershing Square Capital
. . by allowing investment banks that are bond insurers counterparties to artificially prop up ailing insurers, this approach creates even greater risk for the global capital markets.[emphasis added]
Ackman is of course the guarantors’ chief antagonist. He makes no secret of the fact that he’s short Ambac and MBIA common and owns credit default swaps on their debt. But even taking his bias into account, he says some curious things. Here, he’s referring to the back-stop plan that the banks and the guarantors have been working to craft the last few weeks. If a plan were to materialize and be effective, it would be because the rating agencies would have to be comfortable with the level and form of capital commitment it contained. That’s the way these things work. There’s nothing “artificial” about it.
Banks are trying to take advantage of the lower regulatory capital and rating standards for insurers where magically one dollar of capital in an insurer is deemed to be worth about eight dollars of capital in a bank.
This is a line of argument Ackman has repeated a number of times in recent weeks. I know he’s much too smart to actually believe it, so I have to assume it is simply part of his efforts to undermine confidence in the monolines.
The leverage constraints established by regulators, the rating agencies, and investors are of course set based on inherent risk. The monolines’ insurance risk carries lower risk of loss (municipal bonds in particular almost never default) compared to the credit risk maintained by banks, so hence they can employ more leverage. It’s a pretty simple concept.
Policyholders’ surplus is based largely on holding company executives’ estimates of future losses. Since the task of determining capital reserves is a self-graded exam, one would not expect executives to give their companies a failing grade...
Bill, this time you’re going too far! You must have watched too many presidential debates, and have fallen into the habit of saying things that just aren’t so. Under U.S. GAAP, companies must use accruals to try to match revenues to expenses in the time period each is incurred. Under these rules, lenders and insurers establish reserves when future losses are probable and can be estimated. But they don’t have carte blanche in setting reserve levels. Independent auditors require companies use pretty specific formulas in setting accrual levels, and monitor the process very closely.
The establishment of reserves is just one area where accruals are utilized in determining net income. Ackman of course knows this, and is long several companies, such as Target and Sears, whose accounting depends heavily on accruals. Apparently, he is comfortable that the auditors of those firms are “checking” the self-graded examinations taken by management, but is not comfortable with the same at Ambac and MBIA. Interesting, isn’t it, that KPMG is the auditor for both Ambac and Sears?
Michael Callen, CEO, Ambac
We therefore see the current issues facing the financial guarantors not as a question of ability to meet obligations, but rather a challenge to maintain the stability of ratings that have supported our business in the past . . .
Investors and the media don’t seem to understand the connection between the claims-paying ability and solvency of the monolines and possible changes in their investment ratings. In particular, a downgrade from AAA will not lead to insolvency. More likely, the company would stop writing new business, but would continue to collect premiums and pay claims on business it’s written in the past. In the event of such a runoff, it’s very likely that the net present value of net future cash flows would exceed the company’s stock price. Again, the possibility of insolvency would be remote. Ambac currently has $14.5 billion in claims-paying resources; the vast majority of investors and analysts believe that will be more than sufficient to handle future claims. There are a few analysts and investors, such as Ackman and Sean Egan, who believe future claims will exceed the companies’ claims-paying resources--but they are in the minority.
The principal debate has not been over current claims-paying resources, but rather how much new capital the companies will need to raise to keep their triple-As. Here, the challenge has been to get the rating agencies to calm down and stop mindlessly ratcheting up their estimates of worst-case losses.
According to Moody’s, both a triple-A rating and a double-A rating require Ambac to be able to demonstrate claims paying resources sufficient to cover any potential claims at a 99.99% confidence level. The only difference is this: to get a triple-A rating, Ambac must not only be able to cover potential claims at a 99.99% confidence level, but also to have a 30% capital cushion left over while double-A requires a 15% cushion.
Again, I think that many investors and many in the media have gotten confused by all the announcements coming out of the rating agencies. In a way, I don’t blame them. The debate about keeping the triple-A centers on the rating agencies’ estimates of stress-level future losses, not most probable loss estimates, and then having a 30% capital cushion on top. The agencies have dramatically altered their stress assumptions in the past six months (by way too much, in my view), yet many of the monolines have built sufficient capital to maintain their triple-A rating. Then the agencies move the goalposts again, and the process starts all over again.
Gary Dunton, CEO, MBIA (Letter to Regulators)
The [Ackman Open Souce] model and the data appear to have some major deficiencies, as follows:
- The model is described as a ‘security by security’ analysis, while it actually uses an average of 1,267 randomly selected securities to estimates losses. . .
- Assumptions driving loss estimates are proprietary to Global Bank’s [Ackman’s unnamed source of bond-by-bond data] trading model. This is a particularly opaque approach to enhancing transparency.
- The model does not take into account of the structures of CDOs and our contracts that provide us protection. . ..
- The model doesn’t lay into account of the tax impact of losses. . .
- The analysis of RMBS transactions employs a steadily increasing default rate which increases by the trend established in the three most recent months.
Gary Dunton’s letter to insurance and securities regulators was a response to a letter Bill Ackamn sent the same group, in which he explains why he expects Ambac and MBIA to each generate $12 billion in future losses.
Ackman claims he derived those estimates using a proprietary “Open Source Model” provided to him by an unnamed global investment bank. CNBC, Bloomberg, and other members of the media were breathless when Ackman made public the conclusions his Open Source Model spit out. Now that MBIA CEO Dunton has pointed out that the model isn’t exactly a model of transparency, and ignores such insurance basics as subordination and tax effects, the media suddenly goes silent. I don’t get it.
In fact, Ackman has tried to pass off a big complex model, that generates huge loss estimates, as a key pillar supporting his bearish case. Upon further review, however, it turns out the model isn’t nearly as powerful, nuanced, and accurate as he claims.
What disturbs me most about Ackman’s tactics in his battle with the monolines is his, well, let’s call it his lack of consistency. He publishes this open source model and lauds it as a great tool for predicting future losses--but then won’t release the key assumptions embedded in it, on the claim they’re proprietary to the investment bank. Yet he then turns around and accuses the monolines of being too opaque when they don’t disclose every detail about the deals they have insured. The monolines don’t release that data for good reason. Those details are proprietary to the their risk and pricing models; management is understandably eager to protect its intellectual property rights. Bill, you can’t have it both ways. What’s good for the goose is good for the gander.
A reduction in the level of uncertainty surrounding the operations of the monolines, and the debt they have guaranteed, is central to fixing the current dysfunctional credit intermediation process and, in turn, for the expansion of valuations of financial services companies. This is a crucial debate. It is sure interesting to watch and listen to the major players.
Tom Brown is head of BankStocks.com.