Even Bank Supporters Arguing for Restoration of Glass Steagall

John Dizard, who perhaps by virtue of being one of the Financial Times’ most original and insightful columnists, is relegated to its weekend “Wealth” section, has written a particularly important pair of articles. Note that Dizard’s ambit is not policy wonkery but apt and often cynical observations about behavior and trading patterns in less visible part of the financial markets, and sometimes the spending habits of the uber-rich themselves, and what they portend for investments and the economy.

The week before last, Dizard penned an important piece, In the shadow of quantitative easing, party like it is 1788, in which he pointed out that the meant-to-be-stealthy part of the bank bailouts, ZIRP and quantitative easing, had served to shift the risk of the next crisis off the regulated parts of the financial system and onto investors, particularly long-term investors like life insurers and pension funds. His piece last weekend, The US financial industry should listen to leftwing reformers, builds on his observations about where risks sit to argue that the financial services industry would do far better to listen to critics and go back to a clearer separation between commercial banking and trading activities* than we have now.

Dizard chooses to depict the case for breaking up big banks as a pinko argument, when the Bank of England has also fought fiercely for it; it had to settle for ring-fencing because the UK Treasury campaigned hard. And the arguments that the Bank of England made are sound. For instance, Andrew Haldane has pointed out that one of the results of deregulation has been homogeneity in strategies, and even in how banks model risk, ranging from approaches like VaR to the pervasive use of FICO in the US. The result resembles an ecological system with a dominant species. They are much more prone to collapse than ones with more diversity.

I’m in favor of Glass-Steagall type reforms too, but for different reasons. I’ve regarded from my very first days working with commercial banks (and that was Citibank in the early 1980s, meaning a top player at the time) that the managerial requirements for investment banks and commercial banks are diametrically opposed. And what has developed over time is a Rube Goldbergian compromise which results in the worst of all possible worlds (well, save for the inmates): it leaves the foxes, as in the “producers” running the henhouse, with the nominal leaders of these organizations regularly pleading ignorance as to the fact that there is gambling taking place in their ranks.

Here are the key sections from the first leg of Dizard’s argument:

As the promises to pension, life insurance and guaranteed investment contract beneficiaries are discounted at very low or negative rates, they eat through any reserves or capital the institutions have on hand. At the same time, the institutions earn less and less income from any new securities purchases. This has happened slowly, and, with the curve bending downwards in working populations, quickly.

Someone will have to explain to the pensioners and survivors that they are not getting what they are promised. I would suggest not applying for that particular job opening. The compensation for investment managers who are arithmetically certain to lose money will tend to decline over time.

When someone says I am not smart enough to understand persistent negative real rates (the “persistent” is important), I have to agree. There is no way I could project all the dreadful consequences. However, it would be difficult to match the stupidity of the excuses for the current consensus on central bank policy….

To their credit, Federal Reserve staff seem quite guileless about the shortcomings of the long-term projections generated by their central model.

Not that the market people are without their sins of oversimplification and formalism. They have stretched VAR models for risk far beyond their real utility. The market’s risk managers have the same motivation as the macroeconomists: their bosses want a short answer that supports their compensation plan or political platform.

Let us not put too fine a point on this: even though Dizard is talking about the impact on institutional investors, the parties that will take these investment shortfalls in the chin are life insurance policy-holders and pension fund beneficiaries, meaning individuals.

Now to Dizard’s defense of Glass-Steagall 2.0:

The US financial industry should accept the leftwing reformers’ demand to reinstate what is called the Glass-Steagall act…

So why should Wall Street go along with the demands of politicians such as Senator Bernie Sanders and Senator Elizabeth Warren, who do not even have control over the Democratic party? Because in the next crisis, the large banks and securities dealers will need a bailout not so much for themselves as for their customers.

Executives of big banks have already warned that their institutions will not even be able to accept more flight-to-safety deposits in a crisis, let alone extend much more credit.

In JPMorgan’s annual report to shareholders, chief executive Jamie Dimon wrote: “It is my belief that in a crisis environment, non-bank lenders will not continue rolling over loans or extending new credit except at exorbitant prices that take advantage of the crisis situation. Banks knew that they had to lend freely because effectively they are the ‘lender of last resort’ to their clients as the Federal Reserve is to the banks.”

Dizard uses the fact that Bernie Sanders’ plan to break up the banks is so thin as to be unrealistic to argue that banks need to embrace the need to radically restructure themselves rather than have solutions foisted upon them:

After all, in many ways, banks’ participation in the capital markets business is no longer so attractive. In still-liquid markets, dealing spreads have been arbitraged away by an excessive number and capital weight of participants. In illiquid markets, such as junk corporate credit, it is very difficult to do appropriate risk management.

Bank compliance culture has become so elaborate and impenetrable that traders are unable and unwilling to provide bids and offers. They do not want any echoes of Ms Warren’s recent comment that “I never hear the case why it is that some investment bank that wants to take risks on Wall Street, and wants those kinds of profits, should have access to your grandmother’s checking account”.

The only way to really end this sort of talk is by completely separating the capital structures, managements, directors, personnel and even premises of commercial and investment banking. That may be an immediate expense for the shareholders and the economy, but it is better than the possible alternatives.

As I indicated earlier, I’m in favor of Glass-Steagall type reforms, but have not pushed them hard because I don’t regard them at the top priority in addressing what ails our financial system. As Richard Bookstaber pointed out in his landmark book, A Demon of Our Own Design, tightly-coupled systems, meaning ones where processes can be triggered that move too quickly for humans to interrupt them, are failure prone. The most important step in dealing with a tightly-coupled system is to undo or reduce the tight coupling first. Taking other risk-reduction measures before you’ve reduced the tight coupling will increase, not decrease, risk.

That means the first orders of business are reducing the amount of over-the-counter derivatives (much of which has been shifted to the shadow banking system, often investors themselves) and getting rid of high-frequency trading, which serves no socially useful purpose** and increases systemic risk.

As for the managerial issues, traditional commercial banking (think retail branches) involves having a large number of staff perform activities with a high degree of routinization, a very high degree of accuracy, and strict controls. Think of branch tellers, or check and credit card processing, which in the stone ages of banking, was done manually!

Managers in banks tended to rise based on time in grade, and one’s power was strongly correlated with how large a span of operations you controlled. Crudely speaking, the more powerful managers oversaw more people. New product introduction was a deliberate affair, since any “innovation” would be rolled out across the relevant business units in a uniform manner. Thus management was also top-low, with many layers in the hierarchy and lower-level staff having very little discretion.

By contrast, power in investment banking is all about profits. Small units like Goldman’s risk arbitrage department under Bob Rubin were highly influential by virtue of how much they generated. Unlike bank managers, partners oversaw very narrow businesses, typically ones they had grown up in and had even built. Hierarchies in investment banks were flat, and junior people interacted regularly with partners. And “talented” people were promoted quickly. The idea of time-in-grade was anathema to investment banks.

But as investment banks and commercial banks started becoming more or less the same thing, the merging of the two managerial models produced what Occupy Wall Street has correctly called too complex to oversee. From ECONNED:

On paper, capital markets enterprises look like a great opportunity. The firms that are at the nexus of global money flows participate in a very high level of transactions. Enough of them are in complex products or not deeply liquid markets so as to allow firms to find ways to uncover and in many cases create and seize profit opportunities. New, typically sophisticated products often provide particularly juicy returns to the intermediary. And in theory, clever, adaptive, narrowly skilled staff can stay enough ahead of the game so that the amount captured off this huge transaction flow is handsome.

Once again, however, the real world deviates in important respects from the fantasy. Why? This business model is also a managerial nightmare. We have a paradox: “success” and profitability in the investment banking context entails giving broad discretion to individuals with highly specialized know-how. But the businesses have outgrown the ability to monitor and manage these specialists effectively. The high frequency, meaningful stakes, and large absolute number of decisions made at the operational level, the geographic span of these firms, and the often imperfectly understood interconnections among business risks make effective supervision well-nigh impossible.

In other words, while it is wonderfully convenient for CEOs like Lloyd Blankfein and Jamie Dimon to profess that they didn’t know about chicanery that took place on their watch, the reality is it would be very difficult for them to keep an adequate rein on their sprawling, complex businesses even if they wanted to. And that makes it even more important to chop these organizations down to sizes that their putative leaders could conceivably control.

* Keep in mind that an absolute separation is neither practical nor desirable. Banks (like major coporations) have large treasury units which managed the bank’s own cash inflows and outflows. That means that inevitably that a bank will be engaged in trading if nothing else to manage its own money markets and foreign exchange operations. FX rates key off interest rate differentials between in money market instruments in various currencies, so an FX trader benefits from, indeed one can argue needs to trade Treasuries and other major sovereign bonds. Thus there is a bit of thinking that needs to go into how best to draw the line between commercial banking and securities trading, particularly since so much in the way of traditional banking has been moved into capital markets via securitization.

** Please spare me the canard that it makes markets more “efficient”. Where is the proof that having markets be more “efficient” than they were in 2002 has done one iota of good in terms of benefits to the real economy? In fact, academics are now increasingly arguing for the need for transaction taxes because trading costs are so low as to make gambling in the financial markets more attractive than real-economy investing. And that is before you get to studies that virtually without exception find HFT to provide junk liquidity: it provides more liquidity when investors don’t need it, when markets are functioning well, and drains it when investors need it most, when conditions are volatile.

Print Friendly, PDF & Email


  1. vlade


    while I entirely agree that the tight-coupling is an extremely important (although I disagree that OTC and HF are the only things to be addressed there – in general, the tight coupling is the liquidity requirement, and that goes with other stuff as well. OTC and HF are just (some) ways how to lever it), I’d say that it is because of the monoculture we have. Disaster (fire, diseases etc.) spreads much faster in a monoculture – every different ecosystem presents a natural firebreak.

    But, given the current system, the monoculture is the “natural” evolutionary point, because it’s supported by economics (the real one, as in how to get the most cash), the ideology, and, to a large part, regulators (exceptions like BoE are rare, them getting their way even more so).

    Personally, I don’t really care whether banks would or would not have trading/investment arm. If I had one goal, it would be to split the banks into smaller. Whether they do trading and retail at the same time, well, that’s ok. I doubt it, because it’s expensive to do both, but if someone would want to throw (their) money out of the window let them do so.

    To do that, I’d regulate that:
    – a bank has to be an unlimited partnership. no hiding behind shareholders.
    – a bank that is a UP as above, but grows over a certain limit (measured by assets and number of clients), automatically converts into “utility” model. It gets an automatic bail-out eligibility, but it gets 50%+ capital requirements, and a maximum RoC (with any return on capital over that target is a subject to 100% tax). Oh, and it stays unlimited partnership, with the bailout being in effect only after the partners were wiped out.

    Yes, of course, banks would scream that it means they loose efficiencies of scale, but given how often they can’t really deliver on those anyways, the answer is “so what”? Even if they could deliver efficiencies, efficiency for efficiency stake is not worth it.

    Incidentally, in the same vein I’d address asset managers, who so far benefited from banks grabbing the baddie headline, but for all terms and purposes are now very little-value-adds who extract large ongoing fees for badly managing other people’s money. The fact that a question remains whether they could ever manage it well, is not an excuse.

  2. alex morfesis

    We just need sherman act enforcement…banks refuse to make mortgages in trust deed state…that is collusion…they are all in on mers…with no alternative mers scams…collusion…the two major title insurance companies are the ones ignoring and encouraging (or insisting) that foreclosure firms mislead courts and jurists into converting delinquent homeownership into conclusion of ownership by any means necessary…sherman act enforcement would be a good start…and making derivative players and market disruptors(ie…those who short real estate) pay a real time honest premium for taxpayer subsidised fdic insurance…the major g/s deregulation beneficiary enterprises basically lied to the fdic in the years leading up to the 2007-08 crash and paid almost Nothing as insurance premiums, covering up their john paulson type shorting and claiming there was no reason to be forced to pay any premiums at all into the system…i guess that is what “taxpayers” are for…and on the 13th week they cooked the books…

  3. Ishmael

    Please spare me the canard that it makes markets more “efficient”.

    Now Yves you did not think I would disagree with you on anything you wrote above. Once again we are in total agreement and that is especially true about HFT. All I can say is HFT is another sign of the break down of the rule of law in this country. Or to paraphrase Caesar, some are more equal than others.

    I would add to it that we need to reverse, I am not quite sure what law it is but it passed under Clinton and Gingrich which basically relieved investment bankers, lawyers and CPA’s from responsibilities of performing due diligence before bringing companies public (it might have been the Private Securities Litigation Reform Act). Prior to that you would have never seen any of these crap companies taken public that are an everyday occurrence now. Back in the old days there was an unwritten rule that a company needed three years of profitability before it would even be considered a candidate to take public. The three bubbles were blown once that law was passed.

  4. Chauncey Gardiner

    Thank you for this post. Similar to US foreign policy, the financial system is a mess – a Gordian knot of complex derivatives and interconnected relationships that result largely in counterproductive short-term reactive policy measures. But is the system itself too difficult to unravel and reconstitute? I think not. Appreciate the suggestions both in this post, including the staging of policy measures, and those in recent articles such as Pam and Russ Martens wrote on November 30:


    As a society, we allowed the 2007-09 financial crisis to largely go to waste by failing to pursue criminal prosecutions and requiring broad and deep reform of the nation’s financial and regulatory system, market structures, and campaign finance. Their Shock doctrine worked to enable them to preserve their privileges. But it seems to me that based on a broad range of macro indicators ranging from corporate debt quality to the price of oil to recent bank credit rating downgrades and new Fed rules enabling Fed lending to “persons’, that the proponents of neoliberal-markets ideology who operate exclusively out of their perceived self-interest may be about ready to again fail to digest all that Fed ZIRP money.

  5. flora

    Very important post. Thank you.
    re: HFT in a tightly coupled system – a single triggering event can have swift and widespread catastrophic consequences that no one wants or can control. See flash crashes. I think of HFT as a kind of ”machine gun speed” trading.
    The “efficient markets” theory is too often used to justify what amounts to monopoly or collusion.

  6. Screwball

    After just finishing Michael Lewis’s “Flash Boys” I see no place, reason, nor argument for HFT.

  7. JustAnObserver

    I would agree, from a systems/compexity viewpoint, that reducing the tight coupling is the most important objective from the perspective of trying to stop local (single institution) crises escalating into systemic global ones. However I don’t think its possible to pursue this goal independently.

    To put it differently: TooBigToFail, TooBigToManage, or the presence of both “modes” of banking in the same financial institution may (almost certainly do) in and of themselves form part of the tight-coupling problem.

  8. Oregoncharles

    ” tightly-coupled systems, meaning ones where processes can be triggered that move too quickly for humans to interrupt them, are failure prone. ”

    Speed kills.

    Frank Herbert wrote about a version of this problem many years ago, in “Whipping Star” and a number of stories. the premise was that people had managed to make bureaucracy truly efficient – a “be careful what you wish for scenario”. So efficient that it moved far too fast, so the hero of the series worked for the Sabotage Agency, whose job it was to throw sand in the gears and slow things down.

    That’s speculation, but the basic problem is one we already know about from the highways – it’s the reason we have speed limits.

  9. flora

    Yves writes: “…. the managerial requirements for investment banks and commercial banks are diametrically opposed. And what has developed over time is a Rube Goldbergian compromise which results in the worst of all possible worlds ….”

    Yes. The old commercial bank system selected and promoted management based on prudence, long-term outlook, and soundness of the financial institution. What I know about investment banks in today’s go-go anything goes financial climate is that system selects and promotes based on aggressive risk taking. Combine the two systems and the more aggressive system dominates. Expecting individuals to self-police is asking the impossible. The current banking regulation-free climate increases aggressive risk taking. Yes, bring back sane regulations.

Comments are closed.