by David Epstein and Sharyn
O'Halloran
There is a standard story emerging
about how we should think of the Bear Stearns takeover. Not in regard to
its effect on capital markets, now or in the future, but the actual back-and-forth
negotiation of whether JP Morgan was going to take over Bear Stearns, and if so
at what price and under what conditions.
We tell the story three different
ways. First, the conventional wisdom as of now, which paints a picture of
frantic negotiations to avert a looming crisis, but which, we argue, leaves
some important questions unanswered. We then offer two different
game-theoretic analyses of the takeover which address those questions, albeit
in different ways. The end result, we believe, is an interesting example
of how the tools of game theory can help analysts understand the strategic
underpinnings of economic transactions.
1. According to the usual story, JP
Morgan went into the weekend of bargaining thinking it could offer $10 per
share of Bear Stearns. Over the weekend it started doubting that price,
and it was about to walk out on the deal when the Fed, anxious to avert a total
financial meltdown, negotiated a $2 per share price instead. This was
sufficient to let the deal go through, but then the shareholders revolted and
JP Morgan got a lot of bad press in the ensuing days. To improve its image
JP Morgan upped the bid back to $10, which was then accepted to everyone’s
mutual satisfaction.
This story makes sense as far as it
goes, but then why all the poison pills/lock-in provisions in the
deal? Why were these insisted on, and why did the Fed agree to
them? And in this story James Dimon, the JP Morgan CEO, is portrayed as
misjudging his initial offer, complaining about not being able to send Bear
Stearns into bankruptcy, and then reversing course under pressure. Did he
really make such basic errors?
2. The second story is an
incomplete information game theory story; unusually, it is more straightforward
in this case than the complete information story to come next. According
to this version, JP Morgan didn’t know how much Bear Stearns stockholders would
accept, so it started low and bid up from there, as any game theory text would
tell them to do. This strategy had risks for both sides; if bargaining
fell apart then the firm could implode, to no one’s benefit. This means
that the negative reaction to the initial bid was a truly costly signal, and JP
Morgan understood it as such, hence raising the offer.
This story makes Dimon look
more like a shrewd negotiator than the previous one; we might even be able to
excuse his complaining about not being able to send Bear Stearns into
bankruptcy as a bit of theater, aimed at moving the negotiations
along. But we still need an explanation for the lock-in provisions, and
the Fed’s eager acceptance of them.
3. Finally, the complete
information game theory story. Here we’ll more carefully define
preferences and the game being played, illustrated in the figure
below. Let’s assume that there are three options: continued operations by
Bear Stearns (O), bankruptcy (B), and a takeover by JP Morgan (T). Bear
Stearns management and the board had preferences O-T-B: they would have liked
to keep operating, perhaps with an injection of liquidity from the Fed, but
they preferred an orderly takeover to bankruptcy. The Fed and JP Morgan
both had preferences T-O-B, preferring the takeover most of all, and again
fearing the chaos of bankruptcy. Bear Stearns shareholders had preferences
O-B-T; they could try to tough it out as things were, otherwise they actually
preferred bankruptcy to a takeover, since in the former option they could
bargain with bondholders to get a higher share price than they would from JP
Morgan.
The game is played as
follows. First, the Fed decides whether or not to offer Bear Stearns the
opportunity to borrow short-term to cover its debts. If it chooses to
lend, then Bear Stearns management (BSMgmt) gets to choose whether
to try and keep operating or go into bankruptcy. Once the Fed denies Bear
Stearns the liquidity, then operation really isn’t a choice any more, and the
only question is whether a takeover can be arranged. JP Morgan has to
approve any takeover; if they reject then bankruptcy is the result. If
they accept, then Bear Stearns equity holders (BSEquity) have to
accept as well, or again we have bankruptcy.
And here, perhaps, is the solution
to the problem. Note that every player except the equity holders had
bankruptcy as their least favorite option. But since the equity holders
have the last play in the game, and they prefer bankruptcy to a takeover, the
whole deal would be in jeopardy. That’s why the lock-in provisions were so
important; they raised the costs of declining the deal so that the equity
holders would actually prefer the takeover to bankruptcy. Dimon was upset
because he realized those provisions were insufficient to close the deal,
forcing JP Morgan to pay more to get shareholders to agree than they had
planned. And the Fed favored these provisions as a necessary element of
getting a deal done and avoiding Chapter 11.
The insight gained from this
analysis is that the weakest link in the ratification chain drives outcomes in
takeover deals. Here, it was the equity holders, with the most to lose
from the takeover, whose preferences shaped the final bargain. Note too
the differences between outcomes here and outcomes in the usual bargaining games. When
any party to a bargain can simply walk away from the table and thereby enforce
the status quo, then in equilibrium all other players usually have to pay extra
to accommodate the party most reluctant to make the deal. In the brokered
takeover scenario depicted here, though, the Fed made the status quo untenable
by refusing to lend to Bear Stearns initially, thereby weakening the equity
holders’ bargaining position. So rather than make them better off, the
deal was structured to make them worse off; the lock-in provisions made
bankruptcy less appealing and coerced the equity holders to settle. Thus
bargaining in a brokered settlement looks more like politics, with coalition
formation and strategic agenda control, than the standard alternating-offer
bargaining scenario of economics.
Contrast this result, then with
Washington Mutual’s recent rejection of JP Morgan’s takeover bid of $8 a share.
Since this was not a brokered takeover, WaMu management simply walked
away from the deal and instead sold the equivalent debt to a private equity
fund. This maneuver was probably not in shareholders’ interest (the stock
price fell the day the deal was announced), but it let the current managers
keep their jobs. Here, the managers were the weakest link in the chain,
and since the JP Morgan offer did not make them unequivocally better off, they
preferred a white knight to selling.
Recent Comments