Sunday, January 13, 2013

Bank of America Successor Liability, the Article 77 Case, and Settlement Game Theory

Now that MBIA and Bank of America (together with its affiliates, BAC) have concluded arguments on MBIA's motion for summary judgment (SJ) regarding BAC successor liability before Justice Bransten, one can contemplate what BAC's next move will be in its legacy mbs liability saga.

This is no idle speculation as MBIA is a value investor's poster child.  MBIA is a stock that is trading at substantially less than adjusted book value (abv) (and if you believe MBIA's abv, specifically, the receivable MBIA has booked for anticipated mbs insurance recoveries, then you believe that MBIA is trading substantially below its intrinsic value).  As well, MBIA is subject to the possible near-term catalyst represented by Justice Bransten's ruling on BAC's successor liability.

Of course, whether that catalyst will be positive or negative for MBIA's share price is what makes a market.  So the prospects of a near-term MBIA/BAC settlement in advance of that ruling is an important inquiry for an MBIA shareholder (or short).

Therefore, it is not too early to assess what settlement strategy BAC may now wish to pursue, since i) as they say at the Casino de Monaco, "les jeu sont fait, rien ne va plus"...the bets are down, nothing more goes on the table...because nothing more is required to obtain Justice Bransten's ruling other than the simple passage of time, unless there is a settlement with MBIA that dismisses the case, and ii) BAC may have adopted a strategy to accelerate its disposition of legacy mbs liabilities, as it has recently settled claims asserted by Fannie Mae for $11.6 billion at what appears to be about 100 cents on the dollar of claims.  (see

It is also not too early to make this assessment because the two principal MBIA analysts, Harry Fong of MKM Partners and Mark Palmer of BTIG Research, already are out with their prognoses.  See Palmer and  Fong.  While both agree that BAC would be well served to engage in settlement discussions now that the question of BAC's successor liability is teed up and squarely in Justice Bransten's capable hands, Palmer analyzes the possibility of BAC settlement with MBIA in order to protect the advantageous Article 77 settlement, whereas Fong instead analyzes the possibility of BAC settlement with institutional investors and states attorneys general (collectively, the intervenors) who have objected to the Article 77 settlement, rather than settlement first with MBIA.

How long does BAC have until Justice Bransten can be expected to rule in order for any such settlement scenario to play out?  Fong sensibly points to the three month time period Justice Bransten needed to render her causation decision and suggests this as a guide.  Since it would make sense for Justice Bransten to issue her SJ opinions regarding primary liability (argued in the middle of December) and successor liability either at the same time or in close proximity, one might even expect Justice Bransten to take as long as four months to issue these twin opinions simultaneously.  But does it make sense for BAC to tempt the litigation gods and take four, three or even two months before it consummates any settlement it wishes to pursue before Justice Bransten makes her ruling?

So given that the two principal MBIA investment analysts have identified alternative BAC settlement strategies, one can engage in a little exercise of game theory analysis to conjecture which settlement strategy makes the most sense for BAC to pursue.  While I won't try to posit a Nash Equilibrium or use a von Neumann decision tree for this game theory analysis, I hope to explore what one might reasonably expect from BAC by taking into account BAC's incentive structure, as well as the incentives of MBIA and the Article 77 intervenors.

It seems to me based on this analysis that if BAC decides to puruse any settlement strategy before Justice Bransten renders her successor liability opinion, the rational choice would be to settle with MBIA first. 

In order to engage in an exercise of settlement game theory, one needs to assign outcome values to various scenarios involving a BAC settlement before Justice Bransten issues her successor liability opinion.  Doing this necessarily involves assigning hypothetical dollar amounts and litigation success probability values that are plausible, if not prescient.  Feel free to insert your own values if you disagree with the ones I choose.

In assigning these values, I have estimated that the likelihood of Justice Bransten finding that BAC has successor liability for Countrywide's liabilities at 75%.  I have done so based upon my reading of the BAC and MBIA SJ briefs (I have not yet analyzed the hearing transcripts), even though a finding of successor liability is normally an extraordinary holding.  For my views on the possible effect of an adverse ruling by Justice Bransten in the MBIA/BAC case upon the Article 77 settlement, see  Is Bank of America's Article 77 $8.5 Billion Settlement Hostage to its Case with MBIA?

Scenario 1 is the "settle first with MBIA scenario" that Palmer suggests.  This is a two party game between BAC and MBIA at the outset, and it estimates the incremental cost to BAC of a settlement with MBIA on terms likely to be satisfactory to MBIA before Justice Bransten issues her decision.  This value then needs to be subtracted from or added to, as the case may be, the avoided or incrememtal cost that this settlement move (and the dismissal of the MBIA/BAC case before Justice Bransten) will have on BAC's ultimate ability to resolve the Article 77 case and other cases against BAC.  As discussed below, I conjecture that Scenario 1 will result in a net benefit to BAC of $3.5 - $8.5 billion, as compared to a no-settlement strategy.

Scenario 2 is the "settle first with the Article 77 intervenors scenario" that Fong suggests.  This is a three party game among BAC, MBIA and the Article 77 intervenors at the outset, and it estimates the incremental cost to BAC of a final settlement with the Article 77 intervenors on terms likely to be satisfactory to the intervenors before Justice Bransten issues her decision.  Scenario 2 involves a three party game because the intervenors have to take into account their estimate of the probability that a BAC/MBIA settlement will not be concluded after their own settlement when they consider their strategy.  If the Article 77 intervenors settle, and subsequently BAC and MBIA do not settle and Justice Bransten's ruling is adverse to BAC, the Article 77 intervenors' outcome would likely have been substantially improved had they not settled.

Based on this analysis, I conjecture that because Scenario 2 will result in an additional cost to BAC as compared with Scenario 1, BAC's only rational choice would be to proceed with Scenario 1.

Scenario 1

Under scenario 1, we are trying to determine the net outcome to BAC in terms of the i) savings BAC derives from defending all of its mbs claims in the absence of an adverse successor liability decision, less ii) expense BAC incurs in settling with MBIA in order to avoid this adverse successor liability decision.

Taking ii) above first, let's assume that the delta between the bid and ask for a settlement between BAC and MBIA currently is approximately $1.5 billion.  I arrive at this figure as follows:  MBIA has paid out about $3.5 billion in insured losses and, together with statutory interest at the rate of 9% and reimbursable legal fees, MBIA's damages have "approached" (according to an MBIA earnings conference call) $5 billion in total breach of insurance agreement claims (let's assume $4.8 billion). Let's also assume that MBIA would be willing to settle at $3.8 billion, giving a discount for certainty and the time-value of a present payment.

Meanwhile, it is thought that MBIA has commuted various cmbs cds exposures at 10 cents on a $1 par insured, so that MBIA would be willing to commute BAC's $6 billion par cmbs cds at about $600 million.  This would result in an MBIA net settlement ask of $3.2 billion.

On the other hand, BAC has assigned a carrying value to its cmbs cds of approximately $1.3 billion, and would presumably wish to obtain at least that in connection with any commutation.  What BAC is prepared to settle MBIA's breach of insurance agreement claims at is hard to estimate, as BAC's prior monoline settlements have been clothed in ambiguity.  I would think applying about a 40% discount to MBIA's $4.8 billion damage is probably realistic, so that BAC might be willing to pay about $3 billion, in the absence of a holding by Justice Bransten that it has successor liability for Countrywide's liabilities.  If there is an adverse successor liability decision, BAC's discount would have to diminish.

This would result in a BAC net settlement bid of $1.7 billion.  This creates a delta between BAC's settlement bid ($1.7 billion) and MBIA's settlement ask ($3.2 billion) of $1.5 billion.  This implies that the cost to BAC of settling with MBIA before Justice Bransten has issued her opinion would be $1.5 billion, with BAC having to bid up that amount to hit MBIA's ask.

But all this assumes that Justice Bransten has not rendered an adverse successor liability decision; if this decision is issued, the delta will increase, as MBIA's expectation regarding its breach of insurance agreement claim will go up.

To finish consideration of Scenario 1 and proceeding to i) above, we have to estimate the costs BAC is able to avoid by eliminating an adverse holding of successor liability.  What BAC pays to MBIA to avoid an adverse successor liability decision must be netted against the savings BAC derives from having its Article 77 settlement and all of its other mbs cases not exposed to an adverse successor liability decision.  This is an inherently difficult estimation, but one can approach it by focusing on the Article 77 case in particular.

The avoided cost of eliminating a successor liability precedent in respect of the Article 77 case equals the incremental cost BAC would have to pay above $8.5 billion in the event the successor liability decision becomes precedent.  Fong posits an additional consideration, namely that a settlement with MBIA first on terms favorable to MBIA will increase the amount that BAC would have to pay to settle with the Article 77 intervenors.  However, for reasons I will discuss below, I disagree with Fong since I don't think a settlement favorable to MBIA affects the Article 77 settlement calculus.  To that extent, I throw my hat in the ring with Palmer rather than Fong. 

I just see no substantial risk incurred by BAC with respect to the Article 77 settlement posed by a MBIA setttlement on terms favorable to MBIA, because I don't see how such favorable terms would have any substantive effect on the Article 77 hearing itself.  Justice Kapnick will hold an Article 77 hearing to determine whether the trustee was reasonable and acted in good faith in approving a settlement at $8.5 billion, which many have characterized as charitable to BAC given that over $25 billion in losses (and counting) have been experienced by the investors in the trusts.

It is important to remember that Justice Kapnick will be acting as a judge in the Article 77 hearing, not some settlement arbitrator akin to an MLB arbitrator under the collective bargaining agreement who has to determine a fair salary based upon salaries of comparable players.  Justice Kapnick will not engage in an analysis of comparable settlements entered into by BAC to determine the adequacy of the $8.5 billion settlement at issue.  She will be looking at the presence or absence of judicial precedents regarding, among other things, successor liability.  In this respect, a favorable settlement for MBIA has really no precedential value for Justice Kapnick in the Article 77 hearing, while a decision by Justice Bransten holding BAC liable for Countrywide's damages has the potential to blow the $8.5 billion settlement wide open.

Moreover, if comparable settlement amounts were important evidence in the Article 77 hearing, then the damage would already have been done with the BAC/Fannie Mae settlement, which was much more favorable to Fannie Mae than the Article 77 settlement was to the institutional investors.  But the point is that Justice Kapnick is not going to issue a fairness opinion from a finnacial point of view.  She is going to make a ruling as to the legal reasonableness of the trustee's process and its exercise of care.  As to this, a judicial precedent speaks many volumes louder than any outlying settlement amount.

So I believe the avoided cost of a BAC settlement with MBIA is the entire avoided cost of the effect of an adverse successor liability decision on the Article 77 settlement, with no reduction on account of any effect the favorable MBIA settlement may have on the Article 77 settlement.  This is the crux of my disagreement with Fong.

So, to round out our Scenario 1 analysis, we come to estimating the savings, or avoided cost, to BAC of eliminating an adverse successor liability decision.  This is another difficult estimation, given that it is hard to quantify BAC's potential legacy mbs exposure...indeed, it is difficult to quantify even the number of mbs cases BAC may have to defend.  For example, AIG has just launched an action against the Federal Reserve to make clear that AIG did not transfer $7 billion in claims against BAC when it transferred a portfolio of mortgage loans to the Federal Reserve (Maiden Lane II) in connection with AIG's bailout.  See AIG sues NY Fed over right to sue Bank of America, others  It's hard to keep score without a reliable scorecard.

So frankly, as to BAC's ultimate benefit from avoiding a successor liability decision, your guess is as good, or even better, than mine.  I would like to put in an x, or rather a XX, in honor of all of the redactions I have been reading in the MBIA/BAC briefs, but I suppose one can posit a range of benefit of $7.5 - $15 billion, multiplied by a 75% probability of an adverse successor liability holding, resulting in an avoided cost of approximately $5 - $10 billion (rounding).  Counter-intuitively, the BAC/Fannie Mae settlement augers for the low end of the range, as it seems that BAC is willing to settle with little discount in the absence of a successor liability decision, at least with certain influential counterparties, so that the avoided cost might be less than one would think.  However, not all counterparties are created equal, and the absence of a successor liability decision has the effect of permitting BAC to stonewall such of its counterparties as it sees fit; taking away such a luxury might be inconvenient.

To finish the Scenario 1 analysis, the estimated $5 - $10 billion avoided cost to BAC is reduced by the incremental $1.5 billion cost to BAC to settle with MBIA, resulting in a overall net benefit to BAC from settling with MBIA before an adverse successor liability ruling of $3.5 - $8.5 billion, as compared to a no-settlement strategy.

Scenario 2 

Scenario 2 is different from Scenario 1 insofar as BAC is assumed to settle with the Article 77 intervenors before settling with MBIA.  However, at the time the Article 77 intervenors contemplate settlement, MBIA is an indispensible party to the intervenors' outcome assessment, since the possibility that MBIA will not settle and an adverse successor liability decision will issue significantly affects the intervenors outcome that would result from a non-settlement strategy.  This dynamic is illustrated by the following hypothetical settlement negotiation between BAC and the Article 77 intervenors:

"BAC: We will pay an additional $1.5 billion for all intervenors to drop their objections to the settlement.

Intervenors:  That sounds enticing, but how do we know that after we accept that offer, BAC will then decline to settle with MBIA.  We believe that if there is an adverse successor liability decision, we will be entitled to much more than an additional $1.5 billion; try something more like an additional $8.5 billion. By settling with us first, BAC will have lessened the risk posed by waiting to see whether Justice Bransten might rule in BAC's favor.  Your ticket to see Justice Bransten's decision has just become much cheaper if we settle with you first.

BAC: You can't take the risk that we won't settle with MBIA, since if you decline to settle with us first and then we settle with MBIA, you will be left out in the cold with no successor liability decision, and our $1.5 billion offer will be off the table.

Intervenors:  We might not have Justice Bransten's opinion in that event, but we will still have all of MBIA's briefs and other work product of MBIA's counsel relating to BAC's successor liability.  We can make these arguments ourselves in the Article 77 hearing.  Not a penny less than an additional $4.5 billion."

And so on.  This negotiation may drag on not only because of the strategic complexity posed by a three party game, but also because of the difficulty of achieving collective action among all Article 77 intervenors to agree to a particular settlement amount.

The basic point is that if BAC tries to settle with the Article 77 intervenors without having first eliminated the risk posed by an adverse successor liability decision, BAC will be reducing the amount of its own avoided cost because it will have to provide additional settlement value to the Article 77 intervenors (represented in the above dialog by at least some portion of the $3 billion difference between the BAC bid and intervenor ask at the end of the dialog).  Why would BAC want to share any portion of this avoided cost with the Article 77 intervenors by not settling with MBIA first?

How much of BAC's avoided cost would it have to forgo by pursuing settlement first with the intervenors?  This depends largely on the extent to which BAC values the opportunity to see Justice Bransten's successor liability ruling, understanding that it would be prudent to reduce its potential downside from this course of action by settling with the Article 77 intervenors first before it embarked on that path.   

I cannot provide a reliable answer to this question.  But because I believe that it would be cheaper for BAC to simply settle with MBIA first and eliminate the successor liability precedent risk than it would be to settle with the Article 77 intervenors in order to see how Justice Bransten will rule, it seems that BAC's only rational choice would be to settle with MBIA first. 

*  *  *

The role of the NYAG as an intervenor in the Article 77 hearing may prove to be something of a wildcard.  Remember that the NYAG has stated that it will bring "template" suits against mbs issuers along the lines of its actions against JP Morgan and Credit Suisse.  I have posted that BAC is incurring the risk that the NYAG will be able to assert offensive collateral estoppel against BAC in any template action it might bring against BAC based on any rulings adverse to BAC in the MBIA matter, and the damages sought in any such NYAG action would dwarf the MBIA damage claims.  See Bank of America's Idiotic Litigation Strategy and Why the New York Attorney General is Licking His Chops . The NYAG has an interest in seeing that the Article 77 hearing is fully litigated, to enhance the institutional investors recovery and render less necessary its own independent template action.  One wonders if BAC is able to settle with the Article 77 intervenors before the hearing, this will enhance the likelihood that the NYAG will bring its template action against BAC.

*   *   *

Wouldn't it be an interesting development if each of MBIA and the Article 77 intervenors announced publicly that they would not settle their respective actions unless the other action was settled as well?  Would this have the effect of increasing the amount of BAC's settlements? How would it affect the likelihood of settlement?  Would MBIA and the Article 77 intervenors have sufficient identity of interest in order to coordinate their actions without any defection (if this situation reminds you of the Prisoner's Dilemma, remember that defection occurs because the prisoners are unable to coordinate their actions)?  If I am right that BAC believes it should settle with MBIA first, then the Article 77 intervenors would favor such an alliance more than MBIA...which seems to make sense.

NB:  this blog is not intended to be investment advice, and should not be relied upon by anyone to constitute investment advice.  Investing is a tough game, and everyone must do and "own" their own work, because you will certainly own your investments.

Disclosure: long MBI.  Follow me on twitter.


  1. "...the New York State Department of Financial Services (“NYSDFS”) has denied MBIA Corp.’s request to make the January 15, 2013 scheduled interest payment on the Surplus Notes."