Description

And Ambac’s stock price came back down again. I was planning on writing this up a few days ago when it was trading in the $14s, but got side tracked a little. Still given the Ballantyne deal at today’s price it is still cheap.

There are other write ups on the VIC on AMBC that I would recommend you read.

In short, AMBC is a bond insurance in run off … kind off. It went bust because of the financial crisis and came out of bk in 2013.

It is not too hard to get an overall idea as to what AMBC does, so I will try to stick to the relevant value drivers going forward.

Management

I really do like this management. They have been buying stock recently.

What is more impressive is how they have taken down risk. At the end of 2015 they had $108 billion in financial guarantees in force, currently that number sits around $43 billion post the Ballantyne deal. All the while they maintained the adjusted book value per share. In short, the quality of the book value has improved significantly.

They also brought AMBC against expectations out of rehabilitation and executed the AMPS transaction, in addition to the Cofina transaction. All value creating transactions.

The quality of management in this business is important and I believe we have the right one. It is my expectation that management will continue to generation value for shareholders like this. Claude Leblanc thinks the same as he has been buying shares on a fairly regular basis lately.

RMBS recoverables

The most interesting part here is the RMBS litigation against Countrywide/BAC. The total RMBS subrogation recoverable is $1.74 billion, which includes three lawsuits against Countrywide, one lawsuit against Nomura, two against US Bank and one more against multiple defendants. The three Countrywide suits have noted damages of $3.15 billion total. What is on the books of AMBAC is only a conservative estimate of actual paid and future expected damages. What this does not included on AMBAC’s balance sheet is the fraudulent inducement claims and the statutory interest, which is a 9% simple annual interest on the actual damages. In case of damages from a trial the statutory interest is a fairly cookie cutter issue. The only issue the judge decides on really is what the actual damages were in what year. Given that Countrywide refused to buy back reps and warranty loans starting 2008 that is about 11 years to this this day. Let’s just assume that most of the damages happened before 2011 and let’s say total damages were only $2 billion rather than the $3.15 billion claimed by AMBAC. That gets us 8 years of 9% interest on $2 billion. Which brings us to the fraudulent inducement claim. It is hard to assume how much that number should be and the claim has to be obvious, meaning it has to have been “repeated and malicious” behavior on part of the defendant. I bought subordinated RMBS at the height of the financial crisis and it was quite obvious that the RMBS issuers had only cared about loan quantity rather than quality. Countrywide’s MBS were a disaster. AMBAC’s expert found that more than 70% of the loans in a multiple thousand loan sample had reps and warranty issues. Anyway, there are many examples of egregious behavior back then. My favorite is still the gardener that actually made less than $20K a year and got to buy a $1.5 million house in California. Anyway, I have no idea how much the fraudulent inducement number would potentially be and given that no one has ever dared to take these cases to an actual trial it is hard to assess. On the other hand it is not hard to come up with examples of fraudulent behavior on the part of Countrywide. There is a reason these RMBS lawsuits tend to get settled before going to trial. If this goes to trial there is a fairly good argument for fraudulent inducement and the number could come in pretty large, even more than 10 years after the facts it would be difficult to win one of these trials for BAC.

Beside BAC not wanting to take this to trial, also AMBC management wants a settlement I believe. However unlikely, a zero on these RMBS recoverables drops book value down to minus $100 million. My point is more that it is likely that the settlement will come in higher than the $1.6 billion subrogation recoverable range that is booked for the Countrywide trials on the balance sheet. I do not expect a $3.15 billion number, but somewhere in the middle seems reasonable. Every $490 million means an additional $10 per share in book value assuming the warrants get exercised.

And then there still are the lawsuits against First Franklin/BAC/Merrill, Nomura and against US bank. In short it seems likely to me that the RMBS subrogation recoverable asset value is quite conservative on AMBAC’s balance sheet.

So where are we regarding the Countrywide trials? At the end of last year the judge ruled conclusively in favor of AMBAC on pre-trial motions. (FYI, the last judge, Bransten, retired at the end of last year. The new judge is Sherwood.) Trial was set for February 2019, but BAC appealed. From reading the transcripts it is clear that Sherwood is aware of the lengthy duration of this trial, which started in 2010, and is eager to get this to trial. From the transcript of the last judge, her ruling makes sense to me. Still one never knows. The case was presented to the appeals court in May and now we are waiting for the outcome. Once the ruling on the pre-trial issues is in, either AMBAC appeals or we should be ready to set a new trial date. AMBAC talks about a trial near the end of 2019, the beginning of 2020. I would expect a settlement before the end of the year.

One more issue to discuss is that it isn’t just the potential for getting excess dollars out of a settlement or trial against Countrywide/BAC, but also that the transformation of that RMBS subrogation assets on the balance sheet into cash will remove a significant amount of risk. Most of the cash will be used to pay down debt.

AFG

AFG is the holding company for AAC. AAC carries the liabilities, AFG does not carry financial guarantee liabilities. It currently holds about $469 million in cash ($9.27 per share). If we assume the warrants get exercised then we get and additional $81.3 million ($1.61 per share). From a downside protection perspective I kind of look at those dollars as my downside. Also AMBAC has a huge NOL that sits partly at the level of AFG. AMBAC is looking to buy a business, likely at the AFG level, so it can start using these NOLs.

Puerto Rico

Recently AMBAC settled the Cofina bonds which resulted in a decline of the net par of $619 million. This left us at the end of last quarter with $1.282 billion net par exposure to Puerto Rico. Currently other parts of the Puerto Rico exposure are being negotiated. As with Detroit there is a lot of noise and it looks like everyone is so far apart that a solution looks impossible. Here is a good example of the discrepancy between reality and perception at times. The Cofina seniors traded in the 30s and those bonds recovered 93 cents on dollar. The juniors traded in the single digits and settled at 53 cents. It is not as if a default results in a total loss for AMBAC, something some people seem to believe at times.

Overall AMBAC is conservative about reserving and it did increase reserves for Puerto Rico exposure during the last quarter. Overall I feel pretty comfortable with reserving by AMBAC.

Adjusted book value

I find a decent way to value the company is using AMBAC’s adjusted book value. A better way would be to focus on the liquidation value … but who truly knows what that is? I read some old write-ups that put AMBAC’s value above $40 per share and years later it still trades in the 10s. AMBAC does have an adjusted book value it publishes. It looks that most people like to value bond insurers based off their adjusted book value, although there is an argument that can be made that that is not entirely the right way to look at it for AMBAC. My view is that the insurance intangible, net unearned premiums and the unrealized investment gains are real assets, although one could argue that some of that money is unrealized and at risk …

Just to be conservative I would say to go with the adjusted book value. Well at the end of the quarter the adjusted book value per share was $27.52, but post the Ballantyne transaction we now have an adjusted book value per share of $30.52. Once the Countrywide lawsuit is settles and cash is collected I do not see why AMBAC should not trade at 0.8 times the adjusted book value, which gets me to $24.42 per share. If we get more favorable settlements above the current RMBS subrogation recoverables, then one would quickly get a higher number.

Catalyst

BAC/Countrywide settlement or trial. The closer we get to a trial the higher likelihood for a settlement. The first catalyst is the coming appeal court ruling.

Reduction of noise related to PR.