Geithner (and Economics of Contempt): Caught in Haldane’s Pincers

By Richard Smith, a recovering capital markets IT specialist

Economics of Contempt took issue with a post I had written about a Financial Times op-ed by the Bank of England’s Andrew Haldane (and Robert May). He also attributed it to Yves…I really must get the byline habit.

To be candid, I find his piece a bit peculiar. He directs readers to focus on details, when a look at the broad-brush strokes will give a very different reading; and then the details don’t really back him up, either, if you accept his implicit challenge and get stuck into them.

Haldane’s article made a elegant and important argument: the robustness of a system has much more to do with its diversity, that is, the variability of its members, than the solidity of individual elements. Per Haldane and May, an ecology dominated by few, very large members is so unstable as to nullify the banks’ argument about the risk reduction diversification conferred by the of their businesses. And we separately know, per the experience of the Asian crisis, the global financial crisis, and to a lesser degree now with “risk on, risk off” trades, that in fearful markets, previously uncorrelated assets move together. The benefit of diversification is illusory when you most need it. In short, faith in “diversification” is yet another example of blindness to low-probability, high impact risks.

I went too fast for EoC, but his rebuttal seems to gloss over something important. The nub of his response is here:

Haldane was attempting to shoot down an argument about capital levels. It’s one thing to argue that big, complex banks are safer because risk is diversified across their balance sheets — I’ve seen plenty of people pushing that argument. But it’s entirely different to argue that because of this, big banks shouldn’t have to hold larger capital cushions. I’ve still yet to see anyone seriously push this argument, and none of Yves’ examples show anyone making this argument either.

Well, what does “safer” mean, when applied to a bank, then, in the context of diversification? I’d take it to imply something like “less likely to take relatively large capital hits from concentrated exposures”; which is to say, “able to tolerate lower capital levels, relative to less diversified institutions”. If you agree that diversification is good for banks, you assent to the proposition that bigger, more complicated banks use capital more efficiently (and I continue to maintain, contra EoC, that it’s very easy to segue from there into an argument that such banks can carry less capital, or don’t need more, even if that’s never spelt out, and still less, quantified. And why would banks be belaboring this point otherwise?).  But if you buy “diversification” you aren’t paying attention to the systemic risk angle.

Haldane highlights the misdirection and turns this argument back on the banks. Systemic risk arises from the fact that sufficiently large and ideally diversified banks end up holding similar proportions of the same damn stuff, at which point the diversification morphs, unpleasingly, into correlated losses. It doesn’t make sense to affirm that your bank is “safer”, whatever that means, because of diversification, while also admitting that in a crisis, diversification also means greater risk to the system of which your bank is a part. The diversification point advanced by Dimon and Diamond is a crock. Bank CEOs think systemic risk is someone else’s problem, and given that they’re pretty much back to BAU now, you can see how they might have got that impression. As recently as 1998 I-bank CEOs were herded into a room to get the message that they should help out LTCM, for their own good, hammered into their skulls. Times have changed.

Now, if you think diversification does not make the financial system safer, that means you need to compensate via other measures. That could mean more capital. In fact that was the PRECISE argument Timothy Geithner made to bloggers, and I presume publicly too, about Dodd-Frank. The shortcomings of Dodd-Frank did not matter if the banks had enough capital, and he was arguing Basel III would do the trick.

So you can talk capital; or you can talk structural measures, like breaking up banks. Do we see the industry itself  proposing either? If I remember correctly, the last year’s regulatory discussion did seem to involve banks and lobbies pushing back against the idea that banks needed to hold much higher levels of much higher quality capital. In fact that happened rather a lot. I will admit that Blankfein, quoted in my last, does nod to the systemic risk angle, but why are Dimon (mid 2010, just around the time Dodd-Frank was being finalized) and Diamond (late 2010, with the UK’s Independent Banking Commission cranking up to make regulatory recommendations that might include higher capital levels, or break-ups) praising their banks’ diversification, if not to keep up the lobbying push on capital requirements?

Well, if it’s not about regulatory capital, it’s to head off the proposal that the very large complex banks they have created should be broken up: that’s the other conclusion to which Haldane’s argument leads, in the Nature paper from which his FT piece is derived. It’s either more capital or breakup, whichever way up you hold it: a pincer movement. If I were a bank CEO I might start wondering whether it was a such a good idea to keep hawking the “diversification” point around.

So much for the broad-brush strokes; now for a dive into the detail, too, where, again, a close inspection gives grounds to challenge EoC. He thinks the diversification discussion, insofar as it moderates demands for extra capital, is all moot now, for the US at any rate:

Section 165 of Dodd-Frank already mandates that big banks hold larger capital cushions than other banks. That’s a statutory requirement, and not subject to regulatory discretion.

I don’t read it that way: let’s accept his invitation to plough through section 165 of Dodd-Frank. At first blush it does look as cut-and-dried as EoC says:


(1) PURPOSE.—In order to prevent or mitigate risks to
the financial stability of the United States that could arise
from the material financial distress or failure, or ongoing activities,
of large, interconnected financial institutions, the Board
of Governors shall, on its own or pursuant to recommendations
by the Council under section 115, establish prudential standards
for nonbank financial companies supervised by the Board
of Governors and bank holding companies with total consolidated
assets equal to or greater than $50,000,000,000 that—

(A) are more stringent than the standards and requirements
applicable to nonbank financial companies and bank
holding companies that do not present similar risks to
the financial stability of the United States; and
(B) increase in stringency, based on the considerations
identified in subsection (b)(3).

(A) IN GENERAL.—In prescribing more stringent
prudential standards under this section, the Board of Governors
may, on its own or pursuant to a recommendation
by the Council in accordance with section 115, differentiate
among companies on an individual basis or by category,
taking into consideration their capital structure, riskiness,
complexity, financial activities (including the financial
activities of their subsidiaries), size, and any other riskrelated
factors that the Board of Governors deems appropriate.
CERTAIN STANDARDS.—The Board of Governors may, pursuant
to a recommendation by the Council in accordance
with section 115, establish an asset threshold above
$50,000,000,000 for the application of any standard established
under subsections (c) through (g).

Oh, so we have to grind off to section 115. There we find the copy’n paste merchants reposing on a mossy bank, and thumbing their noses at us; what a drafting horror this Act is:

(1) PURPOSE.—In order to prevent or mitigate risks to
the financial stability of the United States that could arise
from the material financial distress, failure, or ongoing activities
of large, interconnected financial institutions, the Council
may make recommendations to the Board of Governors concerning
the establishment and refinement of prudential standards
and reporting and disclosure requirements applicable to
nonbank financial companies supervised by the Board of Governors
and large, interconnected bank holding companies,
(A) are more stringent than those applicable to other
nonbank financial companies and bank holding companies
that do not present similar risks to the financial stability
of the United States; and
(B) increase in stringency, based on the considerations
identified in subsection (b)(3).

Oh dear, we need subsection (b)(3):

(3) CONSIDERATIONS.—In making recommendations concerning
prudential standards under paragraph (1), the Council
(A) take into account differences among nonbank financial
companies supervised by the Board of Governors and
bank holding companies described in subsection (a), based
(i) the factors described in subsections (a) and
(b) of section 113;
(ii) whether the company owns an insured depository
(iii) nonfinancial activities and affiliations of the
company; and
(iv) any other factors that the Council determines
(B) to the extent possible, ensure that small changes
in the factors listed in subsections (a) and (b) of section
113 would not result in sharp, discontinuous changes in
the prudential standards established under section 165;

And finally we are at our destination, section 113. Subsection (b) is for foreign companies, but in subsection (a), we find the criteria for enhanced supervision – but applied to non-bank financials only; yikes. The late change to add big banks didn’t make it this far down into the act. Oops. What a mess. Happily section 115 already has the ultimate let-out for the Council, “any other factors that the Council determines appropriate”, so section 113 turns out not to matter very much for our purposes here, after all; it’s a classic Dodd-Frank wild goose chase.

Do we think the Section 115 let-out might be used? Heck, Geithner doesn’t even feel like having a stab at identifying systemically important institutions in advance of a crisis. You would think that was a pretty basic task for the Financial Stability Oversight Council (and Dodd-Frank would agree with you). So I don’t think interpreting Section 115’s “any other factors”, with as much elasticity as the occasion demands, is going to present any great problem, when the time comes. The “tailored application” of these rules looks as if it has the potential to turn that hard looking $50Bn asset size limit, above which extra regulatory requirements kick in, into something less demanding.

On the other hand, if the tour of Section 165 et al is any guide, figuring out what Dodd Frank actually means, in the heat of a crisis, might be excruciating (repeating Geithner’s discovery of the limits of his powers over AIG, Sep-Nov 2008, perhaps). It will probably depend which bit of Dodd-Frank gets tested the next time something blows up.

So there we are: EoC thinks the commitment to higher capital levels is a firm one. Having accepted his invitation to tour the detail, I think the opposite. I picture him rolling his eyes and sighing; just as I did as I trotted through the above entrancing paragraphs of Dodd-Frank. It is less horrible to wade through in its final 878-page embodiment than it was as a still-evolving 2,300-page late draft, but it’s still not a lovely read.

From a US political point of view the discussion is indeed moot, or nearly; EoC does have that right. In the UK, though, there are plenty more regulatory refuseniks: Adair Turner questions at length whether Dodd-Frank and Basel III are enough:

if global regulators were benevolent dictators designing regulations for a banking system to service a greenfield market economy, they would be wise to choose capital ratios far above even Basel III levels, something more like the 15% to 20% of risk-weighted assets which David Miles illustrates in his recent paper. And the empirical evidence is as compelling as the theoretical. Before the last 40 years or so, banking systems ran with much higher equity capital ratios, much lower leverage and yet economic growth was as high as it is today and investment as a percentage of GDP as high if not higher. (Slide 4)

There is simply no good theoretical argument or empirical evidence that we need to run banking systems with anything like as highly leverage as over recent decades. And today’s regulators are, in a sense, the inheritors of a half-century long policy error, in which we have allowed private sector banks to pursue their private interest in maximising bank leverage, at times influenced by a deep intellectual confusion between private cost and social optimality.

Tucker at the Bank of England chimes in too, in a slightly more muted style:

Bringing this together, policies on trading-book capital requirements, loss-absorbing capacity, resolution and shadow banking can all contribute towards addressing the stability-threatening problems of myopia identified in the literature. And the higher effective capital requirements that this package can deliver might potentially affect incentives by exposing debtholders to risk; and, via reducing return on equity, by harnessing the energies of shareholders in monitoring the terms of distributions to managers. In that sense, quite technical things might have significant effects over time on behaviour and the structure of the financial system. That assumes that lessons from this crisis endure. But, of course, the most pernicious forms of myopia surely come out of prolonged periods of apparent tranquillity in markets and the economy. That is when everyone gets suckered into believing the world is a safer place than it is, with elegant theories and eloquent practical stories for why the world has really changed. Which ultimately tips us from tranquillity to disaster.

I wonder if Tucker expected something like this from Geithner:

The core of the American financial system is in a much stronger position than it was before the crisis

In the US, one could just wait around gloomily to see how it all turns out in practice, though the Fed maverick Hoenig, who hasn’t quite given up on the break-up idea yet, reckons he already knows the outcome if nothing more is done:

In a 2009 article on too big to fail and the problems that resulted from the repeal of Glass-Steagall, Martin Mayer gave a very good description of where we currently stand. He stated:

“We know now that despite the violence of the shock, both the big banks and the cadre of bank regulators and supervisors—and academics—are shaking off the awful memories of 2008 and are setting up the same pins in the same alleys for the same players to try again. We will have to do this, at least, once more before we even try to get it right.”

I share Mayer’s concern that the United States is stuck in much the same game that came out so poorly this last time, but with the prospect for an even greater loss in the next game. We must make sure that large financial organizations are not in position to hold the U.S. economy hostage. But unlike Mayer, I refuse to accept that we must endure yet another crisis before we are willing to finally right the system.

Print Friendly, PDF & Email


  1. ndk

    As recently as 1998 I-bank CEOs were herded into a room to get the message that they should help out LTCM, for their own good, hammered into their skulls.

    This intervention was profitable in the end. I don’t think it conveyed the message that you think it conveyed. It was a triumph of cooperation over competition, rather than sacrifice for the greater good, and that cooperation is IMHO still widespread and malign.

    Otherwise, generally agree with you, and strongly agree with Hoenig (again) on the following:

    I share Mayer’s concern that the United States is stuck in much the same game that came out so poorly this last time, but with the prospect for an even greater loss in the next game.

    1. Richard Smith

      Well you are right, in the end, the conclusion drawn was that there was a safety net under everything. In that sense LTCM was a dry run for 2008.

      But I might still call it a last glimpse of a pretty warped version of enlightened self-interest; Bear famously repudiated even that on the spot – they didn’t even identify with the other I-banks.

      1. ndk

        But I might still call it a last glimpse of a pretty warped version of enlightened self-interest; Bear famously repudiated even that on the spot – they didn’t even identify with the other I-banks.

        Of course, it’s incredibly pernicious. Capitalism is fragile, and cooperation is one of the ways it can be undone. It’s one degree removed from monopoly.

        WRT Bear’s reticence and what was or wasn’t learned, we can say one thing for certain: look at Bear today.

        1. Parvaneh Ferhadi

          It might appear that Capitalism is fragile, but I am not so sure that that’s really the case.
          Capitalism as a system seems pretty resilient to me, as it has not only survived all the crises it itself created, it used them to concentrate even more wealth and power to the already wealthy and powerful that are running and profiting from this Capitalst system.
          What becomes fraglie are the societies that run Capitalism as the socio-econmic model and this means almost everyone, except a few holdouts.

          1. Mel

            I don’t know how much Capitalism developed before the year 1500, but since then it has lived in the sunshine of that big externality, the “New World”. Since 1500, every systemic shortfall — in gold, in food, in employment, in war materiel — could be fixed up by going out and getting gold, or new food crops, or more farmland, or more manufacturing capacity, from the stock originally held by the people we call “First Natons” here.

            Arguably, the American Frontier closed sometime in the 20th century, and Capitalism is feeling the shock.

  2. Foppe

    I am curious: why is EoC treated with so much reverence by you? I tried following his blog for a while, but what struck me was that whenever I did, I felt he’d tried to pull a jedi mind-trick on me: there are no (systemic) problems with banking, and every problem that was there has been resolved by Chris Dodd. I’m sure the guy is a capable technocrat who loves his details, but every time his reading of the details seemed to lead him to the oddest conclusions.
    So what am I missing?

    1. Richard Smith

      *Very* nicely put.

      The post is more deadpan than reverent, I hope. The point is, with EoC you are getting a glimpse of a quasi-official mind set, and that, plus his often persuasive technical points, plus the jedi stuff, which is always fun to puzzle out, makes him an interesting read, if somewhat demanding.

      1. DownSouth

        What are you trying to say? That he has all the technical jargon down pat, which he wields as if it was Excalubur, but doesn’t have enough horse sense to come in out of the rain?

        That seems to describe about 99% of economists.

        1. Richard Smith

          Yes, that is a lot of it, but it’s even more pointed. He is a law pro as well as an economics graduate, and is equipped with a sharp eye (and a sharp tongue).

          That critical faculty makes his devotion to the Democrat end of the financial/regulatory world, where I think is he pretty solidly connected, all the more puzzling.

        2. MikeJ

          If you’re looking for analysis of financial legislation, with a close look at the statutory language, then EoC knows his stuff.

          It’s easy to bemoan legislation such as Dodd-Frank as being toothless and a sop to the banks, but he can show you where there’s actually some teeth (within his corporate finance worldview, of course).

    2. Yves Smith Post author

      Richard is not as, um, enthusiastic in his attacks as I am. My preferred implement is a machette, Richard likes to pull out a scalpel.

  3. Jim A

    Diversification helps when it is broader than the risks in question. Look at insurance. Insuring five million dollars of real estate spread throughout a town is less risky than insuring one five million dollar property. After all, the chances that a bunch of different properties will burn down is minimal. But if the town is in a flood plain, there is still the risk that a large proportion of your policys will have claims. To minimize the chance of losses greater than the companys ability to pay, the insurance has to be MORE diverse than the breadth of the possible hazard.

    In engineering, failsafing is NOT designing things so that they cannot fail. Rather it is designing them so that WHEN they fail, they do so in a safe, obvious way that minimized the risk to life and property. For example on cranes, the hook is not as strong as the cable. This is because a failure of the hook is less dangerous than a failure of the cable, AND because the stretching of the hook provides an obvious indication that the crane has been overloaded.

    The financial “innovation” of the last decade or so has seen the creation of numerous instruments and organizations that can only fail under catastrophic conditions: eg. a sustained, national fall in real estate prices. The illusion of safety thus created, has encouraged greater risk taking: eg. greater leverage, poorer underwriting etc. The result is that when they fail, they fail catastrophicly. This has been the opposite of “failsafing” in the engineering.

    1. Just a thought...

      Nicely explained. Too many people miss that sometimes the application of the theory is the error not the underlying theory.

  4. Allen C

    “It’s perfectly consistent to argue that big banks are historically safer due to diversification”

    All correlations approach one during crises in an over-levered, fraud infested system. Excess leverage combined with fraud leads to asset price bubbles, that lead to crises. Securitization provides a great example. A diversified portfolio of crap is remains largely crap no matter how you slice it.

    The payment systems, commercial banking, consumer banking, and sovereign finance should have NO associated derivative products or accounting schemes. The bankers are perfectly capable of screwing up traditional banking. The derivatives and accounting schemes lead to more complex shell games.

    The reason why gold IS money is due to its incorruptibility (assuming proper assay). The proper structure of any financial system must assume criminal participation from politicians, to regulators, to bankers, to speculators, to debtors.

    It seems that we forget every 70 odd years that mankind in general is unable to be trusted with money and power.

  5. F. Beard

    I have a short, sweet conjecture: Make the money system ethical and stability will take care of itself. How?

    1) An ethical system would be decentralized with government money for government debts and private monies for private debts.

    2) An ethical system would be diversified with different private monies and private money types. Some different private money types include common stock, futures contracts, store coupons, and even silly PMs.

    3) Leverage in an ethical system would be limited by competition.

  6. ex-PFC Chuck

    every 70 odd years . . mankind in general is unable to be trusted with money and power.

    That is a (now) obvious, but profound insight.

  7. Schofield

    I like the analogy of Jim A that a banking and credit system should have fail-safe features built into it. You have to go back to basics though and figure out why you need a fail-safe system in the first place. I would argue that human psychological functioning has a limitation in that for a significant portion of the population they seriously struggle to reconcile self-interest with other-interest. In other words they are prone to stretch economic circumstances beyond breaking point (the blowing of financial bubbles). This results in our need to seriously reconsider Adam Smith’s paradigm of Neo-Liberalism that the Invisible Hand of Self-Interest automatically provides the best guarantee against excessive risk taking. Clearly the Financial Crash showed us that selfishness in the form of greed and power seeking and not risk protection was at the forefront of the financial industry’s psychological drives. This is unlikely to change until a fail-safe system is also introduced into the democratic system to stop the financial capture of politicians and government by the financial industry and other corporate organizations. This latter requirement has to take precedent over financial industry reform with fail-safe mechanisms since these mechanisms are so vulnerable to future dismemberment by a corrupt political and economic system.

  8. readerOfTeaLeaves

    Perhaps if EoC read more widely, including a brief synopsis about slime mold behavior, or about the phase changes that induce plasmas, then he might have some new, better mental models for thinking about the dangers of systemic risk.

    EoC might also find a little background on the search term ‘monoculture’ instructive. You can plant millions of acres of the same genetic strain of corn, and thereby claim that you have ‘diversity’ (meaning, ‘geographically distributed’) in your corn. That, however, is a faulty assumption; what you have is one strain that is going to be wiped out when the pests figure it out. See also: Potato Famine, Ireland.

    1. ScottS


      My tour of Ireland told me that the Irish planted a variety of crops, but the English took virtually everything and only potatoes were left for the Irish to subsist on. I.e., it was less of a monoculture problem than an imperialism problem So I decided to look up the Potato Famine to back me up. Wow, what fascinating parallels to current problems!

      The rest of this post from: — I encourage everyone to read it in its entirety.

      The middlemen leased large tracts of land from the landlords on long leases with fixed rents, which they then sublet as they saw fit. They split the holding into smaller and smaller parcels to increase the amounts of rents they could then obtain, a system called conacre. Tenants could be evicted for reasons such as non-payment of rents (which were very high), or if the landlord decided to raise sheep instead of grain crops. The cottier paid his rent by working for the landlord.[20] Any improvements made on the holdings by the tenants became the property of the landlords when the lease expired or was terminated, which acted as a disincentive to improvements. The tenants had no security of tenure on the land; being tenants “at will” they could be turned out whenever the landlord chose. This class of tenant made up the majority of tenant farmers in Ireland, the exception being in Ulster where there existed a practice known as “tenant right”, under which tenants were compensated for any improvements made to their holdings. The commission according to Woodham-Smith stated that “the superior prosperity and tranquility of Ulster, compared with the rest of Ireland, were due to tenant right.”[19]

      Landlords in Ireland used their powers without remorse, and the people lived in dread of them. In these circumstances, Woodham-Smith writes “industry and enterprise were extinguished and a peasantry created which was one of the most destitute in Europe.”[17]

Comments are closed.