Note: This post was originally posted on Dec 11 2020. However, it was updated January 24th 2021 to reflect current thinking.
Ambac is currently close to a pure-play special situations bet, and seems to have a 2.9x – 3.6x upside, with a ~20% downside. I’ve been working on this idea for a while now, and the situation is developing somewhat quickly at this point. This post is aiming to give the high level overview of my findings on the situation, without going too deep on the business itself. For more there, I would suggest reading either (a) the most recent 10-k and 10-q, or (b) reading one of the three recent–ish VIC posts about Ambac.
Ambac is incorporated as a financial services holding company, with all operations and “run-outs” sitting in subsidiaries.
Ambac has successfully lowered contingent liabilities over time, steadily increasing per-share book value (until COVID). In fact, the current adjusted book value is approximately $19.4 per share, and if this trend continues there is a chance for the company to continue organically increasing that value over time. However, this is all upside that isn’t material to the catalyst and/or thesis.
Some quick context: Ambac was founded back in the early nineties and was (for many years) one of the two largest monoline insurers (the other being MBIA of Bill Ackman fame). Its main business was insuring bonds for municipalities, governments, and corporations. It had a AAA rating, and by insuring other bonds, borrowers would adopt that rating, allowing insurees to pay significantly lower financing costs. Then during the housing bubble they got into the RMBS market. As the financial system melted down, they went underwater and the company was restructured in 2013 with the goal of reducing all legacy exposures, and running out the last remaining policies. They have kept on their books a few billion dollars in NOL to carry forward for any operating businesses, and are just recently beginning to find accretive deals to take advantage of those tax assets.
After emerging from bankruptcy, Ambac sued a variety of mortgage originators for both breaches of representatives and warranties in the Ambac-insured RMBS contracts, as well as for fraudulent inducement. These claims have been bounced around the legal system for ~8 years now, but the final trial was scheduled for Feb ’21, before being postponed once again due to COVID.
Now, the RMBS claims are the core of the thesis, but first let’s consider the downside.
Downside — holding co. assets
Worst case scenario for Ambac is that they lose in court, or that COVID causes contingent liabilities to be significantly more than expected (and reserved for). In that scenario, the run-off subsidiaries (Ambac Assurance group et al) would be once again restructured, and the holding co’s equity likely wiped out. However, the holding co would still have its other base assets: cash & equivalents, and its solvent operating subsidiaries.
The ex. subsidiary assets amount to approximately $10.10 per share:
| ($ in millions) | |
| Cash and short-term investments | $ 313 |
| Other investments (includes notes issued by subsidiaries) | $ 114 |
| Other net assets (includes accruals for tolling payments) | $ 38 |
| Total | $ 465 |
| Per share | $ 10.10 |
‘Other investments’ and ‘other net assets’ includes securities issued by Ambac Assurance, and in the worst case scenario these assets would likely become impaired to some degree. Theoretically, if contingent liabilities did increase significantly, these assets could be written down to zero. If we stay as conservative as possible in this scenario, the per-share value of holding co’s remaining assets goes down to $6.8 per share.
We still need to consider subsidiary value. There has not been much information released on the Xchange acquisition to date, and so we will have to assume that whatever Ambac paid for them was fair value, and doesn’t impact holding co book value. Everspan however, recently spun out of Ambac Assurance (the run-off subsidiary) to be held directly by the holding co. Insurance filings show that they have assets of $245M, with liabilities of less than $1M. This gives a capital surplus of ~$5.3 per-share. These numbers seem bizarre, and it would be great to corroborate them with other filings. However, for now they are the only data points we have to go off.
Therefore, if Ambac Assurance becomes insolvent (our worst case scenario), the residual value of holding co assets would still be ~$12 ($6.8 + $5.3). This equates to a downside of 22% based on current share price of $15.6.
One last note on this — this is a downside book value. There is always a risk the market doesn’t reward Ambac for these assets. However, because of their liquidity, it is harder to ignore the intrinsic value of these assets, as it would be much easier for Ambac to reward shareholders in the simplified structure than it would be to capitalize on a run-off monoline insurer.
The Catalyst
Back in 2008 Ambac got into the business of insuring RMBS. During the financial crisis, it became clear that many of these mortgages were sold on fraudulent claims– false information on FICO scores, falsified job histories, sometimes filed under the owners dog’s name. Much of this fraud was committed by consumers, but it was also facilitated by the mortgage originators, including a company named Countrywide. Ambac currently is still litigating ~$4B for fraudulent inducement, and breach of contract across a number of cases against a number of counterparties.
The main case was initiated September 2010. Ambac’s claim in this case is that Countrywide had refused to repurchase the faulty mortgages in the midst of the melt down (as it was contractually obligated to do). You can find this case here, under index number 0651612/2010 .
I’ll save you the long winded legal journey, but Ambac and Countrywide (now owned by Bank of America) have taken this case through a long list of jurisdictions, and have fought over a million tiny details. Late last year, the state court of New York set a date for the trial for this past July.
Then COVID struck and the date got pushed back to this February. In parallel, Countrywide found another opportunity to dismiss the claims of fraud in summary judgement. Their motion to dismiss can be found here (index number: 651612/2010 | Document number 2067). Countrywide’s claim here was that the “fraud damages” was not materially different to the “contract damages”, and therefore it should be dismissed.
On December 8th 2020, Countrywide won summary judgement to get the fraud claims dismissed. However, there will still be some trial to decide on the contractual damages. This trial was set for late February, but there has since been summary judgement to push the date back post-COVID due to safety concerns of the legal teams. Ambac has also filed an appeal to the summary judgment against them on the fraud claim.
Contract damages sizing and timing
Ambac’s claims have been supported by an analysis conducted by a Dr. Karl Snow. The full report is also available through the same e-filings page (most recent supplemental filings is document number 2072), and are summarized in presentation materials for the most recent hearing (document number 2092):
The allocation damages are for the fraud claims, which have been dismissed. The repurchase damages are the range of damages that Dr. Snow has calculated under a variety of legal theories.
This range doesn’t include the statutory interest, which could increase these numbers significantly. Depending on the year you assume accrual begins, and the statutory rate (which in NY appears to be 9%) could potentially 2.5x the total claim value. It is also worth noting that if the case goes to court, Countrywide would likely contend a number of the assumptions that go into these totals, potentially lowering final payout.
MBIA, which sued other banks over similar claims (most recently winning a case just a couple weeks ago against Credit Suisse) settled for 60 – 80% of total claim value. This could potentially accelerate recovery timeline, while slightly reducing the claims value.
The upside settlement value here is enormous. If Ambac wins everything in court, they could receive a $5.4B ($2.1 @ 9% interest for 11 years). The total received in settlement in that upside would likely be closer to $3.2B – $4.3B. This large potential upside is what gives this investment such a strong margin of safety. Even in a scenario where some number of contracts are removed from the damages calculation, or if for some reason interest is removed, payout should still be more than the current carrying value of $1.76B.
Additionally, due to the massive losses Ambac faced in ’08, they have $2.1 billion of NOL left to use (at the subsidiary level) that will materially lower their tax burden. In the scenario of a $1.8B payout they will not pay any taxes. In the $3.2B – $4.3B range, the total payout will only be lowered to $3.0B – $3.8B after tax.
Valuation
The current reported adjusted GAAP book value is $19.44 per share ($22.59 unadjusted). Per page 19 of the 2020 Q3 10-Q, this includes $1.76B carrying value for the RMBS settlement (which includes the other aforementioned ongoing litigation). This also doesn’t include claims for fraudulent inducement, which offer additional upside optionality.
If the case is settled for $1.8B and all other claims are declined, then the current book value still offers a 25% upside to the current share price. In a scenario where Ambac settles and gets everything they ask for, there will be an additional $1.7B ($3.4B less $1.7B) added to book value. This is a 2.9x multiple of where book value is today, and a 3.6x over current share price
Ambac also has the ability to win the other dismissed fraudulent inducement claim on appeal. While I only give this a ~5% chance (based roughly on the amount of civil cases that get overturned through appeal), the payout could be an additional $2.0B + interest. The other cases against Countrywide/BoA, as well as against other parties could also continue to pay out.
Why hasn’t the market noticed this?
Ambac stock was on a tear in November and December, even after the fraud claim was dismissed. It wasn’t until weeks later when Ambac released the 8-K announcing the court update that the market finally reacted. This delayed reaction, along with the mispricing of the situation, all point to most investors largely ignoring Ambac and their litigatory efforts. So why?
I think there are two main reasons. The first is Ambac’s 5% cap— they don’t allow share holders to hold more than 5% of the organization’s shares. That equates to an approximate value of just under $40M. This tempers the value of due diligence for larger institutional managers, who wouldn’t be able to make a meaningful allocation after investing in research.
This recovery process has also been a long time in the works. It has likely worn down many of the special situations desks that would have been watching this stock 3-4 years ago, as Ambac was only half way through its litigatory journey.
