Here is another attempt to get a discussion going about VSS and the current financial crisis. I want to focus on one of the angles into the current debates that seems most interesting for us: the question of how one anticipates or prevents massive catastrophic disruptions to the financial system.Â
As we all know, one of the commentators to get a tremendous amount of attention in the past months is Nassim Taeb, who has written about the pervasive tendency to underestimate the importance of preparing for or anticipating so-called black swan events. These are, in our terms, massive, catastrophic disruptions that annihilate our existing understandings of risk distributions (a question that is very much related to our discussion on Bell Curves and Power Laws over on OAT). Taeb has recently written a short column in the Financial Times outlining “Ten Principles for a Black Swan-Proof World.” He writes, for example, that if a private company is too big to fail it should not exist, and that instruments like derivatives — which are too complicated to understand — should be abolished.
Effectively, Taeb is arguing that we should create a world in which there will not be any black swans; thus the “black-swan proof world.” Much of the response has focused on two points: first, that Taeb does not deal much with the details of how any of these principles would be implemented in practice (in other words, how do you actually determine when something is too big to fail?); second, as Ezra Klein has written, it does not address the costs of black swan proofing the economy. As Klein provocatively puts the point (in terms that seem to me very familiar), we have to ask whether we actually want to live in a black-swan proof world. Â
Of course, our game will not be to weigh in on these costs and benefits. But I do think that we have something interesting to say about the technologies of power that would be at work in either case. A “black-swan proof world” suggests a paradigm of safety or prevention — one that simply declares that catastrophic events cannot be allowed to happen. It would involve erecting a series of control mechanisms such as strict limits on the kinds of financial instruments that can be created, strict divisions between different kinds of financial activities, particularly between heavily regulated and lightly regulated activities (perhaps in the form of Glass-Stegal II, perhaps in some other form), and limitations on the kinds and scales of economic organizations that can emerge.
The alternative — or at least one alternative — is a regulatory apparatus that recognizes that black swans are inevitable, and, perhaps, might begin from the proposition that the potential diseconomies of “black-swan proofing” the world are themselves enormous and difficult to anticipate.  This would be a paradigm of preparedness and resilience. It would take as its central problem the creation of mechanisms to ensure that black swan events do not become unmanagaeable. Thus, for example, current proposals for systemic risk regulation — which I have been discussing with Onur — seem to work along these lines. They propose to use network analysis to assess the interdependencies of risk exposures among major actors in the economy — effectively identifying the vulberable, vital nodes that would blow up in the case of a black swan event. AIG would be an example.
Although I need to think about this question more, the lines between “prevention” and “preparedness” may be blurry. This is particularly true in cases — like the current crisis — where the catastrophic event is actually a product of the system that it is disrupting (rather than an exogenous shock like a hurricane or a nuclear bomb). But I think that we can, nonetheless, identify two very different logics of control and intervention. In a mode of prevention, the instinct is simply to abolish dangerous activities, such as certain types of derivatives, or the growth of certain kinds of institutions. For preparedness, the question would not be preventing these activities but instead recognizing when a number of interlocking activities lead to systemic risk. In this view, for example, the problem in the current crisis was not the collapse of the housing bubble, or the securitization of mortgages, or the failure to assess the risks of securitization, but the fact that so much else depended on the collapse of these markets not happening. A paradigm of resilience would thus be very concerned with the distinction between “small black swans” — local catastrophes that it is very costly to prevent — and “big black swans” — global catastrophes that are the products of total network breakdowns. The vulnerability would be understood to arrive not from individual activities and firms but from systemic effects, which would, in turn, be seen as the proper target of surveillance and intervention.
In reading Taleb’s “10 principles”, it’s not clear whether he has a coherent regulatory framework in mind. I would call him “precautionary” rather than “preventive” (see below). He seems in large part to want to question the very legitimacy of the regime of expertise that has so far been in charge of designing the response to the crisis. “People who were driving a school bus blindfolded (and crashed it) should never be given a new bus. The economics establishment….lost its legitimacy with the failure of the system.” His prescriptions are more about shutting things down, banning extant practices, than about what a new system would look like.
Whereas Geithner, Summers and company are indeed engaged in an attempt to envision and intervene in what they see as a “vital, vulnerable system.” Here it is worth looking at Geithner’s 3/26 testimony before the House Financial Services Committee, in which he (a) defines what the system does (“financial systems serve an essential and basic function….they exist to allocate savings and investment to their most productive uses.”); (b) names the system’s basic elements (“financial institutions and markets”); (c) diagnoses the failure of the system (“The system proved too unstable and fragile,
subject to significant crises every few years, periodic booms in real estate markets and in
credit, followed by busts and contraction….”), and (d) prescribes a governmental solution based on the concept of regulating “systemic risk” according to the norm of resilience to shock:
“Financial institutions and markets that are critical to the functioning of the financial
system and that could pose serious risks to the stability of the financial system need to be
subject to strong oversight by the government. Our financial system and the major
centralized markets must be strong and resilient enough to withstand very severe shocks
and the failure of one or more large institutions.”
Taleb, in contrast, focuses his energy on the claim by finance experts that they could rationally manage risk. He wants to shrink the system, or even deconstitute it. “Nothing should ever become too big to fail.” In other words, it is not that the government limit itself to being ready to save institutions that have become systemic risks, but that we should prohibit their creation (“Nothing should ever become too big to fail”). It is perhaps akin to those who say that the response to the risk of a nuclear melt-down is not (in Wildavsky fashion) to create resilient safety systems but rather to stop building nuclear power plants.
Also: at a global scale, it’s interesting how certain practices associated with civil defense and emergency management have been proposed for dealing with the financial crisis:
http://www.nytimes.com/2009/04/03/world/europe/03summit.html?partner=rss
So in this story from the NYT, we see (1) the notion of the IMF as a “first responder”:
“The most concrete step was a $750 billion reinforcement of the resources of the monetary fund, which has emerged from years of waning relevance to become the first responder in this crisis, lending billions of dollars in emergency loans to dozens of countries.”
And (2) the idea of creating an early warning/ monitoring system:
“the leaders agreed to create a new Financial Stability Board to monitor the financial system for signs of risks”
There will even be exercises: the FSB will “manage contingency planning for cross-border crisis management, particularly with respect to systemically important firms” and “collaborate with the IMF to conduct Early Warning Exercises.
Thanks for this Andy. A couple responses.
First, on this “global first responder” theme — I think that this is right, and very interesting. Actually my post on Vulnerability and Development relates to this question. Onur has been looking quite specifically at the practices of stress testing and other techniques for monitoring systemic risk in the financial system, and specifically at the genealogy of IMF and World Bank practices in this domain. It seems that one of the interesting developments is a shift from assessment focused on currency shocks to broader financial system risk assessment. Hopefully Onur will weigh in soon.
Second, on the first post, relating to prevention and precaution, a couple things:
(a) It would be interesting to hear a bit more about this distinction. To my mind a precautionary ethic suggests a policy of prevention, precisely in the sense that you take steps to stop something from happening even in the absence of a cost-benefit analysis. So I don’t quite grasp what the important distinction between them is.
(b) I agree that Geithner is talking the game of vital systems security, and that the precautionary position is distinct from his, so the interesting question is where the differences lie. I think that both sides recognize that finance is a vulnerable vital system — there is no one saying that it isn’t. The real distinction comes up in how proposes to manage threats to that system, and that is what I was trying to get at in my post. It seems that Geithner et al are simultaneously asking questions of economy: how is it that we do vital systems security in a way that does not destroy capitalism (or at least capitalism as we understand it)? How is it that we do it in a way that does not disrupt the very characteristics that make this system vital (risk taking, entrepreneurialism, the flow of capital). This is very much a Flynn-esque position, and it is actually quite close to the better regulation discussions in Europe: we have to take into account the costs of precaution. Geithner never says this (at least I have not heard him say it in these terms), but I think it is implicit in the entire strategy.
(a) Yes, precaution seeks to prevent the bad thing from happening. The distinction that Ewald makes (in “The Return of Descartes’ Malicious Demon”) is that whereas in a solidaristic system (such as welfare) one seeks to manage risks through prevention (eg. classical public health) – a precautionary logic does not apply to “risk” strictly speaking (insofar as a risk is inherently calculable); rather, it applies to what is uncertain – to what one can apprehend without being able to assess. He writes: “The precautionary hypothesis puts us in the presence of a risk that is neither measurable nor assessable – that is, essentially a nonrisk.†This is akin to the Frank Knight distinction between risk and uncertainty. Along these lines, Taleb is saying that black swans are not “risks” in that their occurrence is not amenable to statistical calculation. And so we have to try to eliminate the potential source of their occurrence. That, to me, is precautionary.
(b) Aha, liberalism!
Agreed that Geithner’s regulatory initiatives are oriented to stabilizing the system without disrupting the ongoing functioning of capitalism as it is currently structured. What critics on the left (eg. Krugman) are saying is that the size and scale of the financial system (what should be an adjunct to the economy) is the problem – and that Geithner is mistakenly trying to preserve this.
Analytically, I think it is worth asking how exactly the “financial system” is being constituted as a structure that has certain vital nodes that can be recognized by experts. How do you know what counts as a “systemic risk” and therefore needs extra supervision? There is a certain post-decisional regret at play here, perhaps – “we should have realized that Lehman Brothers was in fact too big to fail.”
On Ewald — I have always found his terminology on this point confusing. Obviously “classic public health” would also mean a modulated intervention that does not eradicate but weighs the costs and benefits of intervening. As we have recognized many times, there are clearly different logics of public health; and probably “prevention” is a term that seems to refer to too many different things. In any case, we can agree that what we are talking about is strategies that aim to simply stop certain kind of events from ever happening by doing things like banning certain kinds of securities or creating firewalls between different activities. These are the things that Taeb talks about in that short article, and these are the things that are circulating as the serious candidates for regulation of a precautionary sort.
On the second — definitely, liberalism. Scarcity cannot be eradicated, it has to be left to have its own life. I think that Onur should weigh in on how vital and vulnerable nodes are being identified. It has to do with network analysis and complex techniques for evaluating counter-party risk.
This is a quite interesting conversation that I am quite interested in as it fits into the broader confines of my dissertation project and recent discussions Stephen and I have been having in the last few months. As I am intervening relatively late (sorry for this, but I have been sick over the weekend) and the discussion seems to take its own course, I will try to weave through some of the points of exchange carefully.
First, on Taleb and prevention versus precaution: Although I think the notion of “black-swan-free” world and Taleb as a locus of a broader controversy over the future of the financial system is worth visiting and thinking over, probably we should not beat over him as an exemplary figure too much. Part of the disagreement between Stephen and Andy over whether Taleb represents a logic of prevention or precaution seems to be due to the over-reactionary and rather arrogant style and attitude of Taleb himself which has alienated and isolated him from much of the regulatory debates over the future of the financial system. Perhaps a figure such as Stiglitz (a figure closer to Krugman) would be a better exemplary that would allow us to differentiate between predispositional characteristics of actors (reactionary tone of Taleb) and the different substantive logics through which they problematize existing security mechanisms. In short, it seems to me Andy’s skepticism of Taleb is emanating from Taleb’s rather distracting attitude and not the substance of the logic he as Stephen points out represents.
Stiglitz, similar to Taleb but in a much more sound manner, has been asking for the re-institution of the Glass and Steagell Act of 1933 which was brought down in 1999 by the Clinton administration allowing the functions of different classes of legal entities (commercial banks, hedge funds and investment banks; insurance was added to this scheme of separation in 1956) to be conducted under the confines of the same firm. Along with other left leaning economists and political commentators, he has also been strongly emphasizing the excessive level of financialization of the US economy and a necessary return to an economy centered around the so-called real economy. I am not sure if he makes the argument that institutions too-big-to-fail should be dismantled, but it would not be surprising to hear this from him as well… So, I think, as you guys also pointed out, we can settle on the prevention camp (or as I would call them the camp of abstinence following Latour’s notion in the Politics of Nature, since prevention as an analytical category should be reserved for security strategies of various kinds much familiar to VSSers and probably not to broader problematizations of security and risk) having the following characteristics:
1) Separate different juridico-financial entities and regulate them based their status with law (Glass & Steagell)
2) Don’t let financial firms get too big to undermine market discipline through “moral hazard” (Trust/monopoly busting)
3) Restructure the economy so that the real sector (read material production) becomes once again the gravitational center of the overall economy at the expense of the financial sector. (19th century economy)
The response to this critique by actors who are playing the game of vital systems security has been that (1) a regulatory frame based on juridico-financial distinctions is bound to fail since as the 1970s junk bond market proves they argue, financial innovation constituting the very inner logic and engine of the financial activities will always find ways to drill holes in the regulatory firewalls of a regulatory scheme similar to that of the New Deal. (2) Size of an institution, implied by the notion of too-big-to-fail, is not a good way of thinking regarding the relationship between a firm and its systemic importance vis-a-vis the broader system. Rather than big-ness, the interconnectedness (and other network-analytics such as network centrality and connectivity density) should be the basis of a framework based on financial system security: Hence the notion of too-interconnected-to-fail. This they argue has proved itself with the failure of Lehman (so from this perspective the criticism over the failure of Lehman is not really a matter of post-decisional regret, but rather it is a controversy over what constitutes a systemically vital node and how to asses it.) (3) Banking is no longer as in the pre-1970s a domain of rich financial capitalist preying on innocent citizens, but as a socio-technical (infrastructure) system it is a development and planning mechanism that not only re-allocates capital in more efficient ways, but also increases the resilience of the overall US economy. I think the response of this side to the former on whether the financial system is excessively large or not would be that the problem is not one of size (though they might acknowledge the optimum size might have to be smaller), but rather the mode in which security is conducted and consequently what the potential systemic vulnerabilities are. As I will touch upon below, I think this way of thinking does not so much concern itself with how large a system is per se, but how resilient a system is and in case of an emergency whether a market crash can be contained or not, since regardless of the size of the financial system a weak financial system in the face of a uncontainable market crash might easily bring down the broader economy through creating a credit squeeze.
As Andy’s quotes of Geithner reveals, according to Geithner et al. this system, however, was too fragile as previous regulatory schemes mostly developed since the late 1980s onwards under the guidance of Greenspan’s Fed seems to have confused flexibility, as a quality of the inner structural properties of a system, with resilience as a normative rationality. Financial system being a source of economic flexibility against exogenous shocks such as the crash of the dot-com bubble or 9/11 reflects Greenspan’s repeated responses in the sessions of various congressional oversight meetings questioning whether one needs further regulation of the broader financial system and the shadow banking system more specifically throughout the 1990s. Greenspan’s defense of the financial system against these attempts of further regulation (probably in a mode of abstinence) and the source of his belief in the flexibility seems to be a misrecognition that as long as the actors engage in risk taking practices in sound and rational manner free of speculative and irrational behaviors and these practices are distributed over and inserted in the firms of certain size and above (aka Basel II), one would accumulate “system security” through the standardization of risk management practices configured to ensure the safety of the firm against the risk of a default.
What is interesting however in the case of financial system security is that this approach, basically a distributed preparedness scheme, seems to have failed as a rather relatively non-significant market contraction in the housing market first leading to a market crash in the broader first circuit of securitization and finally ending with either the collapse (Lehman) or the near-collapse (Bear Sterns and AIG) of systemically important institutions resulting with the break down of the entire price signaling mechanism of the markets and causing “mark to market” accounting mechanism to destroy balance sheets of and, therefore, the confidence in broader financial institutions in the absence of a “market” and therefore “market prices”. The diagnostic category reserved for the end result of this process of contagion of a local market crash into an much broader market crash seems to be referred by experts specializing on financial emergencies is “panic” (one should probably note that this conceptualization should not be confused with the earlier uses of the word, since “panic” seems to be reconceptualized in the last 30 years or so in the light of information theory of prices and behavioral economics based on agent based simulations).
So, I think Stephen’s emphasis between different forms of black swans is right on target as the overall crisis itself seems to be a series of unlikely events taking place consecutively, and this nuanced differentiation seems to be missing from almost all actors who are on the prevention/abstinence side of the debate. By either putting the blame on the unsound business practices of actors or emphasizing the inevitability of such an event, they are basically obstructing the different moments of the overall crisis. In an interesting way and probably not coincidentally either, they sound much like the New Dealers who had put the blame for the recession on the stock market crash of 1929 and could not have identified mitigating and exasperating factors that helped the mutation of a market crash (a local black swan) into an economic depression (a global black swan) such as the crisis management and intervention of the Fed and the government. Thanks to the work of Milton Friedman and Anna Schwartz (see the chapter The Great Contraction, 1929-1933 in A Monetary History of the United States, 1867-1960, published in 1963) among many other neo-liberal leaning actors, we now know that what had allowed a market crash to spill over into the broader economy initially in the form of a liquidity contraction turning into insolvencies causing thousands of banks to default and naturally the collapse of the overall economy was the misleading and inappropriate monetary practices of the Fed.
I think the trajectory of monetarism as a set of governmental practices focusing on the continual adjustment of monetary flows by changing the flow rate/intensity of money supply on the basis of ex-post aggregate indicators is quite illuminating in allowing us to make sense of what the historical significance of current crisis is and how one might think about the recent uproar about “systemic risk regulation”. In a way then, we can see the rise of monetarism as a moment of reflexive rationalization in which the governmental conduct itself rather than the object of the conduct, i.e. the economy, is taken up as a problem and a source of malaise for collective life. And yet as we have witnessed in this crisis the Fed even under the guidance of a figure such as Bernanke who has spent his life studying the Great Depression and building on Friedman’s monetarist history and lessons could not prevent contagion to spill over. Hence vital system security in the case of financial system security seems to be inserted in the form of systemic risk regulation and other proposals for reform for making the system more resilient, as opposed to simply flexible, at the limit at which we have come to realize monetarism was simply a moment of critique of fiscal policies of the New Deal within a broader cycle of reflexivization finding itself expression now in the debates over systemic risk regulation.
Then the interesting research question for me is what the genealogy of this new mode of problematization focusing on systemic risk regulation on the one hand and increasing the inner resilience of a system in the face of exogenous/endogenous shocks on the other hand is. So far in my not so extensive sniffing around, I have realized these issues have been debated by what seems to be a relatively not so large group of actors located within the Fed and other central banks around the world, the IMF and the WB along with the Bank of International Settlements and the Financial System Stability Forum in substantive ways. While the problem of resilience as expected is taken up by economists and bankers in these debates, the question of systemic risk regulation seems to be handled through an alliance between economists and “systems analysts” or “complexity scientists” who have been developing simulation models based on agent based modeling and network analysis to understand either ways to prevent small liquidity problems to turn into system wide liquidity deadlock, to solve liquidity deadlocks through finding ways to change the lending behavior of market participants, or identifying which actors are systemically important for the wider system through measures such as network centrality and inter-connectivity density.
So I think it would be safe to point out we have two main genealogical threads: 1) History of the problem of resilience within the domain of financial system security in IMF and other global financial institutions (one should keep in mind Geithner learned many of his tricks by working at IMF in the late 1990s under the task group focusing on financial system resilience and stability) and 2) the History of the technical rationalities, embedded within techniques such as stress testing and network analysis, upon which such a problematization can be substantiated in the form of either a systemic risk regulator that would allow central banks to be prepared for a financial emergency or vulnerability reduction reform that would deem to re-engineer a financial system to address financial system weaknesses that might lead to broader economic crises (including the way in which how much capital buffers, i.e. stockpile, one needs for each financial object, how much interconnectivity or size can be allowed for one institution to attain etc).
Finally, I think the most interesting aspect of the financial system security is that we observe the limitations of distributed preparedness and therefore liberalism in ensuring the care and security of the overall system. I believe we had also found this tension in the OEP archives as on the one hand OEP was shifting from being a centralized planning agency into an agency of distributed preparedness in the mid-1960s while at the same time developing planning capacity (the work of Economic Impact Group and the Strike Analysis System based on input-output modeling) that looked much different from the distributed preparedness strand of VSS, but nevertheless also distinct from other “totalitarian” and teleological planning schemes for which the same calculative tools were developed. This is why I think what we might be witnessing might not be so much about liberalism per se, but a new form being tried to be deployed and inserted by the liberals such as Geithner et al. This, I would argue, although akin to the analytical axes of readiness and preparedness in the conceptual structure of VSS, is a distinct axis in which one tries to reduce the vulnerabilities of a system ex ante through deploying anticipatory modes of veridiction and is directed at the system rather than neither the experts’ predisposition for emergency action (readiness) nor the apparatuses of security necessary for crisis intervention (preparedness). But this is likely to open a can of worms as to what our analytical categories are referring to and probably worth a blog entry on its own.