Too Big to Fail and Justification

September 25, 2008

Posts in the last few days triggered so much good discussion in the comments that I thought that it would be a shame to leave it buried there. So, here is, with minimal editing, the discussion that Martha Poon and Zsuzui Vargha had here last week, following the bailout of AIG.

Whether a firm is “too large to fail” is in fact the outcome. I want to add that it’s not only about justifications but also calculation, as Yuval suggested at the end of the original post. In order to establish how “big” the firm is in terms of market worth (along the lines of Boltanski and Thevenot), the regulators have to trace or get a vague sense of what the network of contracts looks like, estimate the scenarios, assess the ripples. Another kind of calculation is about credibility. Regulators are always called upon being consistent, because markets have to be calculable, and calculability can only be maintained if actors’ responses are not random. This is both what Max Weber already suggested, and also the lesson from socialist economies with “soft budgetary constraints”. So, regulators have to prove that what they are doing is consistent-why they are saving AIG when they are not saving Lehman.

The justifications [for saving one institution but not saving the other] become inconsistent as they pile up on each other. At some points, however, the actors do look back on their decisions and try to justify how their current super-interventionist measures fit well with their earlier anti-regulatory position. In the most grand terms, Bernanke and Paulson try to say it’s a qualitatively different situation than the ordinary state of markets-it’s a state of emergency. Such a statement allows them to discard free market dogma, gives them carte blanche, and makes them problem-solving world-saving heroes. I wonder how accountability will or will not develop after the crisis is over. Bush managed to avoid it after 9/11.

Rapid change during this crisis makes the trial and error process of policy-making much more visible than otherwise. We literally see how the regulators are shifting justifications within 24 hours, from the case-by-case, now admittedly ad hoc way of addressing the crisis, to the “systemic” view of intervention.

What do you think about the following description? That this shift of frame means that the actors have given up calculating the consequences of each failing bank, it’s too complicated and they can’t identify the losers in advance, and they can’t bail them out as companies (that would really go against their anti-interventionist position). They are now calculating in terms of product market categories (what kind of debt should the government buy), which is not specific to the individual company. So they went from a firm-centered view of where the crisis is, to a market-centered one.

11 Responses to “Too Big to Fail and Justification”

  1. typewritten Says:

    Extremely interesting. One point: although the “regimes of justification” sociological framework is indeed quite to the point here, maybe it is also possible to think how the too-big-to-fail situation overflows it. Meaning: is not this, to some extent, a case of decision without dispute, of decision without justification? One striking point is how a very reduced group of guys (Bernanke, Paulson) take only 24 hours to say that it’s the end of the world and that they need this considerable amount of dough, and that’s it. It took Roosevelt quite a while to “justify” his line of action during the Great Depression. A theoretical note: in the Boltanskian-Thevenotian literature, there is an harsh discussion about the difference between “legitimate trials of strength” (the ones involving grammars for settling disputes) and “trials of strength, period” (the actor-networkish way). Maybe the discussion applies here (this looks like “a state of exception”!).

    (JL, master of the Maoist unconscious, pointed that out in a discrete email).

  2. yuvalmillo Says:

    Good points! Thanks. And, yes, the spirit of the Basque Master is evident. Well, now that politics and markets are openly intertwined, maybe he should join us…

  3. Zsuzsanna Says:

    typewritten: So, can we apply that discussion in this way: the public management of the bailout decision was a “trial of strength, period”. But it now seems they might not pass the trial. The rescue package wasn’t passed immediately, and as time passes more and more actors want to open it up and demand justification. To understand why the government might eventually fail the trial, we need to use the concept of “legitimate trial of strength”. The reason they are failing is because the other actors want a legitimate trial.

    I wonder why it wasn’t enough for Paulson and Bernanke to ride the shock wave. At the beginning it looked like they can pull it off. My guess is that it has to do with disputes about how esoteric finance is. Paulson-Bernanke seem to suggest it is esoteric, so only they can fix it, while politicians and journalists are caught saying it’s not esoteric, it’s simply immoral. That’s why they are interested in accountability.

  4. Zsuzsanna Says:

    Correction: what I meant is that Paulson and Bernanke _bet_ on finance being divorced from “society”, but they lost their bet.

  5. danielbeunza Says:

    Zsuzsi, right on. What I have not yet seen is any discussion of why the plan might not work for technical reasons. And please note the following: the plumbers and technicians… such as my colleagues in finance, are coming back again and again to the ethical regime… moral hazard and so on.

  6. Peter Says:

    I like this discussion, but I keep coming back to the initial formulation by Yuval – that given AIG and Lehman, why was one considered ‘too big to fail’ while the other was allowed to fail. Zsuzsanna’s argument is based on differing orders of worth being mobilized (or rather, varied levels of success in mobilizing them); Martha’s point is that there is inconsistency in the application of these orders, or at least we ought to think seriously about how to operationalize them so we can tell one from another.

    But I the premise is a bit flawed, or maybe that the justifications were cooked into the problem earlier in the process. AIG wasn’t just ‘too big’, but too central. We could compare AIG with LTCM, for instance, arguing that they were too Jenga’d* to fail. It wasn’t about assessing capitalization or size, but of likelihood of systemic effects versus localized effects.

    *Referring to that block game Jenga, where you build a tower and then start to pull pieces from the bottom until it falls. Bigger pieces may be more likely to be load-bearing pieces, but the distinction is less about size than about centrality…

  7. yuvalmillo Says:

    Peter: I love the Jenga reference! I love this game. I cannot wait until my kids are old enough to play it. In SNA, by the way, such a systemically central actor would, typically, have low dyadic constraint. That way, the removal of which, in the extreme would separate the network to different components.
    But, as you imply, and as the discussion shows, the classical structural theory would not work here. That is because an important dimension of the centrality of actors like AIG comes not from direct interactions with them, but also from the awareness ‘that they are there’. Let’s say, for the sake of argument, that not each and every actor in the financial market has direct or even second-degree connections with AIG. Yet, it would be safe to say that had AIG collapsed, every actor in the market would have been affected. The candidate theories for the dispersion of such a ‘shockwave’ are varied: information asymmetry (Akerlof’s market for lemons), status differences (Podolny’s pipes and prisms), and Bordieu’s field (which I suggested). Of course, it would be important to see what theory can offer a good explanatory and interpretative value, but the point I was making was that is that moral hazard and systemic risk challenge distinction between structure and action and that maybe a more market-informed sociological view (read SSF) is necessary if we want to address them properly.

  8. Peter Says:

    wow, Yuval, you speak in academese even for me! I appreciate your drawing this out (well, you, Martha, Zsuzsanna, and the soon-to-be-joining-you-but-hasn’t-come-to-embrace-it-yet Typewritten) – the argument is something like the following:

    1) ‘too big to fail’ = a vague enough rhetoric that multiple orders of worth can be employed, politically

    2) ‘too big to fail’ = central v. peripheral, not size; in this case, potentially challenges SNA because it’s not a network in the conventional sense of being connected via bilateral, hub-spoke relations – the metaphor fails a bit

    3) the question is, in a reconceived version of centrality, how is ‘crisis’ transmitted?

    Ok, so how to we move from interesting to empirical? What should we be looking for as evidence of these arguments?

  9. panik Says:

    There is a dimension to the choice between a Lehmans vs. AIG rescue that is not completely captured by the “too big” or “too central” calculus. Even from the opposite side of the Atlantic to where the decisions were ultimately made it was more or less open knowledge that Lehmans was in big trouble and would probably go down or at least be radically reconfigured in some way. It is now known that the UK government, at least at the level of PM and Chancellor, were made aware of this by the US authorities way back in August or even earlier, especially since Barclays were the main candidate for a possible rescue and it was felt that before such a move the BoE, and hence the UK government, would need to give such a massive move the nod.

    The fact is that plan B if a bailout could not be arranged was winding up and it seems that plenty of prep work was done ahead of such an eventuality to unwind Lehman positions and as far as possible insulate important counterparties from such an outcome. This was probably facilitated by the deep due diligence that Barclays were doing on Lehmans and which it must have shared with the US authorities as the rescue plan was eventually dependent on the US government putting a floor under Lehmans liabilities in order to insure against any skeletons lurking in the cupboards. I think that in this way the US authorities must have felt that enough firebrakes were by then in place to prevent any serious contagion, thus making it easier to pull the plug without the rest of the system taking an irreparable hit.

    By putting Lehmans in a kind of quarantine it was possible to reintroduce some modicum of calculability; a bit like disentangling objects from their natural setting to study them in a laboratory. Maybe it is exactly because of this (apparent?) calculability that pulling the trigger was easier in the Lehmans case.

  10. yuvalmillo Says:

    Panik, this bit of forensic finance, if you may, sounds compelling (as long as the empirics fit, which we do know at this stage). However, this, on its own, does not solve the Lehman Bros – AIG discrepancy. That is, it is possible, and even plausible, that ‘in the right circles’ there were rumours and fragments of information about how exposed AIG was, through its default contracts, to bad debts, and, consequently, moves were made to de-centralise it. Fine. Still, both the actions on the two cases and the justification discourse around them were very different. Still didn’t hear a good explanation to this discrepancy. Perhaps it can be attributed to the cognitive perception that there is an accumulation of cases.

  11. panik Says:

    Yuval, OK, I accept that my posting does not address the justificatory element of the discussion. You are also right that without the evidence we can’t really say much, although both Brown and Darling in interviews the last week-10 days have pretty much said what I am basing the argument on. Two points I want to make in reply to your posting.

    The first one is clarificatory. While the unwinding of positions no doubt has as a de-centering effect, the point I am making is slightly different. By actually ‘cutting out’ from a big chunck of the networks it is part of an entity such as Lehmans, it is much easier to assess – if not necessarilly measure – what the policy makers don’t (and maybe can’t) know. It does, however, improve your calculative handle on the sittuation.

    The second point relates to – dare I say it – the technicalities of unwinding. While I have little knowledge on the market mechanisms for default contracts, I have little doubt that the specificity of such contratcs (they are either OTC or bilateral, have different durations and terms, exist in many widely different jurisdictions, and so on) is much greater that for assets and liabilities that are traded on organised open marketplaces. Such marketplaces, even though they are loath to use them, have well established mechanisms for unwinding transactions in an orderly way. This is because they have to use such ‘undo’ buttons even in their everyday operations for purely functional reasons not related to a major default. The ability to de-center an invetsment bank that will have a big chunck of its assets and liabilities in organised marketplaces that have the necesary mechanisms to do unwinding is very different from the ability to do so with products such as those provided by AIG. While AIG counterparties might have sought to find ways to act on the rumours and scraps of information to insulate themselves, such effocts could not be compared in efficacy to a concerted aproach that brings together all marketplace participants, marketplace operators and infrastructure providers, central banks, regulators and so on. In fact, one of the future issues that regulators will no dount look at once the dust has settled will be how much this crisis had to do with the increasing use of OTC arrangements and internatlisation of trades at the expense of trading on organised marketplaces.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: