Streetwise Professor

June 23, 2012

Why I Was Not Won Over by Luigi Zingales, as Expressed by Mark Twain

Filed under: Economics,Financial crisis,Financial Crisis II,History,Politics,Regulation — The Professor @ 1:46 pm

Luigi Zingales is a brilliant guy.  I find his advocacy in the FT of a return to Glass-Steagall less than brilliant.

Zingales identifies four reasons for supporting G-S.  Three are wildly implausible, the fourth hardly persuasive enough to justify a return to 30s-era regulatory structures.

The first reason is that it makes sense to restrict commercial banks’ participation in “very risky activities.”  There are at least two problems with this, when evaluated in the context of the financial crisis.

First, the investments and activities that put commercial banks at risk during each of the last two crises-mortgages and MBS in Financial Crisis I and sovereign debt in Financial Crisis II (the one currently ongoing, particularly in Europe)-were blessed by regulators as low risk activities.  They almost surely would have passed muster under Glass-Steagall.  There were other major banking crises under Glass-Steagall.  It is painfully clear that G-S is not sufficient to prevent banks from taking on risk in sufficient amounts to put individual banks and the banking system at risk.

Second, and I realize I am an apostate on this, although moral hazard is a concern associated with deposit insurance, funding and liquidity risk is as dangerous and arguably more dangerous.  Runs are more likely, the less sticky the funding.  Concentrating riskier activities in institutions like stand-alone investment banks that rely on wholesale markets rather than stickier insured deposits for funding increases the risk of destabilizing runs and makes the system more fragile.   It therefore seems extremely dangerous to allow the institutions with the riskiest funding hold the riskiest assets, and preclude the institutions with the safest funding not hold such risky assets.  Look for other means-such as risk-based deposit insurance premiums or capital requirements-to control the moral hazard problem.

Zingales’s second reason is that Glass-Steagall was simple.  Well, yes, the original bill was (as compared to Frank-n-Dodd or the Volcker Rule portion thereof) but there did grow up a thicket of regulations and interpretations and decisions around it.  This is inherent in any law that attempts to draw lines between activities: such laws generate efforts to engineer products that cross the lines in substance but seem to comply to the letter.  This generates a process of litigation, interpretation, regulatory challenge, etc.

And does anybody believe that if Glass-Steagall were to be brought back, it would be the same today as in the 1930s?  Glass-Steagall was simple in part because it was the product of simpler times  Financial markets are much more complicated today, and splitting the baby would be far harder.  Any attempt to do so would almost certainly result in a Son of Glass-Steagall that is far more complex, and far longer, than the nearly 80 year old original.

I have to say that Zingales’s third reason is cracked:

The third reason why I came to support Glass-Steagall was because I realised it was not simply a coincidence that we witnessed a prospering of securities markets and the blossoming of new ones (options and futures markets) while Glass-Steagall was in place, but since its repeal have seen a demise of public equity markets and an explosion of opaque over-the-counter ones.

I realize this is a newspaper oped, with tight space limits, but Luigi provides no serious evidence to support his realization that coincidence equals causation.  With respect to the demise of public equity markets, trading volumes have skyrocketed, and although there has been an increase in “dark markets”, such things existed in the glory days of G-S, just under a different name (“third markets”, for instance), and there are good economic reasons why this is so.  Moreover, some problems associated with equity markets are attributable to other regulations, notably Sarbanes-Oxley.

With respect to OTC instruments, Zingales’s characterization of cause and effect is muddled and ahistorical.  OTC derivatives began to grow rapidly in the 1980s, while G-S was still in force, but their growth was constrained by regulation.  However, it was legal uncertainty under the CEA that was the main problem, not Glass-Steagall.  This was addressed by the CFMA of 2000, rather than the Gramm-Leach-Bliley  (GLB-wouldn’t it have been funny if there had been a fourth co-sponsor with a name starting with “T”?) of 1999, which repealed G-S.  Commercial banks and investment banks both did OTC derivatives before GLB, and would likely continue to do them under Son of Glass-Steagall.  What’s more, they are constrained-over-constrained, IMO-by Frank-n-Dodd.

Zingales continues in a rather bizarre vein:

To function properly markets need a large number of independent traders. The separation between commercial and investment banking deprived investment banks of access to cheap funds (in the form of deposits), forcing them to limit their size and the size of their bets. These limitations increased the number of market participants, making markets more liquid. With the repeal of Glass-Steagall, investment banks exploded in size and so did their market power. As a result, the new financial instruments (such as credit default swaps) developed in an opaque over-the-counter market populated by a few powerful dealers, rather than in a well regulated and transparent public market.

But the IBs that got into trouble (Bear, Lehman, Merrill, Morgan Stanley) never had access to deposits post-GLB.  Not before, not after.  So the cause-effect link between the repeal of G-S and the explosion in size of IBs that Zingales posits (resulting from the the access to deposit funding for IBs inside CBs) never happened.

No one knows the exact cause (global imbalances, Fed looseness), but the financial system was flooded with funding liquidity in the 2000s.  That is what permitted the IBs to lever up and risk up, not a sudden access to insured deposits.

And by objective measures, markets were more liquid-substantially so-post-G-S.  Spreads were smaller.  Volumes were bigger.  The number of participants was larger: hedge funds proliferated, for instance.  Really, I don’t think anybody noticed the markets getting less liquid post-G-S.  If anything, the concern was that there was too much liquidity, too much trading.

And insofar as CDS are concerned, it is hard to believe that G-S would have (a) prevented the creation of these instruments, or (b) resulted in them being traded by something other than big dealer institutions.  Again, note that stand-alone IBs were major dealers in these instruments, and that they were first developed by a commercial bank (JP Morgan) prior to 1999.  In many respects, the disintermediation of banks by capital markets, a process that started and became quite pronounced under the G-S separation led banks into other, non-lending forms of intermediation.

I would also emphasize that there are intense economies of scale in responding to expansive regulation like Dodd-Frank, meaning that regulation can contribute strongly to consolidation in finance, with its consequent effects on competition, the number of market participants, and systemic risks.

Zingales’s history, in other words, isn’t history at all.

There’s another point here.  If G-S was that important, why didn’t its repeal lead to dramatic changes in market structure, with the demise of stand-alone IBs and the dominance of integrated, full-service universal banks with access to insured deposit funding?  Zingales’s narrative is inherently contradictory: he focuses on the metastasization of IBs post-GLB, but if you believe his story, the repeal should have caused IBs to shrink or die or get absorbed by commercial banks or turn themselves into universal banks funded substantially by deposits.  None of that happened.   Indeed, there were numerous studies in the mid-2000s, notably by the Fed, the Senate Banking Committee, and the St. Louis Fed that all concluded that GLB’s G-S repeal had very limited effects on the structure of US financial markets: most of the major changes in market structure had occurred in the 1975-1998 period.

It is also worth mentioning that the universal bank model that Glass-Steagall eliminated developed in a world without deposit insurance.  Indeed, Glass-Steagall created deposit insurance, which means that at least then implicit subsidies in deposit insurance couldn’t have caused the combination of investment and commercial banking.  Why should we believe that deposit insurance would cause such an integration today?  Especially in light of the fact that the elimination of G-S boundaries did not result in the absorption of investment banking into commercial banks.  Given the role that deposit insurance plays in Zingales’s narrative, this is a major problem for it.

Zingales also claims that the financial system was more resilient under G-S, and points to the contrast between the ’87 Crash (which was not followed by a financial crisis) and the ’08 debacle (which was) as an illustration.   I really don’t see any connection between G-S and how the broader financial markets fared post-Black Monday.  Zingales notes that commercial banks didn’t have problems post-Crash.  Well, investment banks didn’t either, for the most part.  The main difference is that in ’87, most equities were held by real money investors, not financial institutions: ditto the Dot Com crash.

In contrast, in ’07-08, the mortgages that cratered were held primarily by highly leveraged financial institutions-commercial and investment banks; many of these were destined to go to real money investors, but banks and investment banks were caught with them before they could securitize them and market them to real money.  Today, in Europe, sovereign debt is held mainly by leveraged financial institutions.  Banks today don’t hold huge amounts of equities, and an ’87-type event, where the crash originated in the equity market would probably have a similar result today.  The equity crash in ’08 was a response to the crash that was occurring in mortgages.  The events are very different.

And it is quite plausible that these two events were the direct result of regulations, notably the very favorable capital treatment of mortgages and MBS (especially those transformed into AAA securities despite the quality of the underlying collateral) and sovereign debt.

Luigi’s last point is that a divided financial industry is less effective in exerting political influence because commercial banks, investment banks, and insurers have divergent interests.  But again, commercial banks, investment banks, and insurers existed post-GLB.  There was not a wholesale shift to universal banking in the US.  Certainly this could have happened, post-GLB, but it didn’t.  The greatest impetus to consolidation was the crisis itself, when IBs either blew up or were absorbed into commecial banks.  Thus, again, the structural change that could have happened under G-S didn’t happen to that great an extent, meaning that the political economy didn’t change that much either.

Moreover, to the extent that the activities of financial institutions are siloed, as under G-S, they are not competing with one another in the marketplace.  This also limits their need to compete with one another in obtaining favorable legislation or regulation.  Each can focus its efforts on obtaining favorable laws and regulations pertaining to its silo.  Commercial banks focus on favorable rules for commercial banking activities; insurers focus on favorable rules for insurance; investment banks on favorable rules for broker-dealer and securities underwriting.  There is no Galbraithian “countervailing power” mechanism operating that pits investment banks against commercial banks when they are functionally separated a la G-S, because the very separation of these activities means that the firms in one segment are not competing for rents against firms in the other segment.

In brief, it appears that Luigi is aiming his argument at a world that could have arisen after the repeal of Glass-Steagall.  A world in which there was a decisive move to universal banking in the US, resulting in the disappearance of stand-alone investment banks.  Some certainly predicted that outcome in 1999, but it didn’t occur.  There were some changes in market structure, but only modest ones.  It is passing strange that Zingales attributes the growth in investment banks to a law that, under the most straightforward interpretation of his argument, should have led to their demise.  Most importantly, the mechanism that Zingales posits-the availability of cheap insured deposits to fund risk taking by investment banks-didn’t work that way in practice.  Risk taking was funded through the wholesale markets which were awash in money.  That was the vulnerability of the system.

Zingales’s history about the evolution of equity markets, OTC derivatives markets generally, and CDS markets in particular, is also suspect.

So when reading this article, a Mark Twain quote came to mind: “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”

Print Friendly


  1. The major flaw of Dodd-Frank is that the bill is not needed reforms, complied into a comprehensive piece of legislation. Rather, it is the product of a populist desire to punish those who cause great economic pain and market disruption. Passing reforms and understanding their effects on the marketplace is much harder than offering an amendment to a bill that shows a given legislator has done his part to express his outrage.

    Dodd-Frank is doomed to fail because the timeframe for writing the regs then interpreting and deciding how to enforce them will be so protracted that the resulting uncertainty will stifle financial innovation and very soon will result in a massive wave of layoffs as financial firms pare down to wait out the regulators and then staff back up with b-school graduates 5 years from now (if we are lucky).

    To advoid the costs of Dodd-Frank there are two choices, scrap it or simply don’t enforce it (as they have done with Sarbanes-Oxley. To date, I haven’t heard of a single enforcement action under S/O). Non-enforcement is easiest, but it opens the door to selective enforcement. The most likey outcome is to water it down so it is meaningless and eventually scrtap it. Sadly, the result of scrapping D-F would likely be a strong push to bring back G-S.

    Until the regulatory uncertainty passes, Wall Street will be moribund, at best. Banks will not know how to position themselves in the marketplace and until banks get get their business models set, they will not be able to generate sufficient profits to fully recapitalize.

    Comment by Charles — June 23, 2012 @ 4:14 pm


    You’re spot on, Prof.

    Comment by markets.aurelius — June 23, 2012 @ 5:00 pm

  3. Thanks, Markets. Excellent link, btw. Very informative, and does a great job at exploding the conventional wisdom. To me it all comes back to fragile funding of maturity transformation. But isn’t that always the way with financial crises? Historically, pre-deposit insurance, deposits were a fragile funding source. Now, not so much (except in the Eurozone now, but that is a whole other story.) The IMF paper shows pretty conclusively that the intermediation of global financial flows hinged on a very fragile form of maturity transformation, and that the IBs were at the center of this.

    This is the reason for my apostasy on the whole ring fencing idea. It is predicated on a complete misdiagnosis. Moral hazard arising from deposit guarantees weren’t, and aren’t the problem. All of the regulatory and legislative prescriptions based on this diagnosis are the worst kind of nostrum. Indeed, all of these prescriptions, especially those that increase reliance on collateral (especially Dodd-Frank clearing and OTC margining mandates) are going to increase the reliance on fragile funding mechanisms, and put risk on the balance sheets of those with the runniest funding. This is nuts, IMO.

    The piece also shows that Glass-Steagall is a complete irrelevance. As I pointed out in the post, much of the transformation of the American financial system occurred while G-S was in place. Indeed, the increased reliance on capital markets and the resulting marginalization of banks was arguably the result of G-S, though I just advance that as a hypothesis, rather than a conclusion. But how anyone who claims even a passing familiarity with the facts could attribute the dramatic growth in investment banks to the elimination of G-S is beyond me. I mean, this claim is idiotic on its face.

    Charles-yes, the regulatory uncertainty associated with Frankendodd is extreme. Unfortunately, non-enforcement doesn’t really address that issue, because it would still hang there, like a Sword of Damocles, hanging over the heads of everyone in the industry. That is, there is always the risk-a regulatory tail risk, if you will-that the populist urge or something else will trigger a frenzy of enforcement.

    Look at how Gensler is hyping the Morgan CIO episode for all it is worth. There will always be the risk of some demagogue or power hungry official deciding to wield that sword with abandon.

    The ProfessorComment by The Professor — June 24, 2012 @ 1:18 am

  4. [...] Professor is not convinced by Luigi Zingales’s version of the return-to-Glass-Steagall popular [...]

    Pingback by I Read Stuff, You Read Me « Rhymes With Cars & Girls — June 25, 2012 @ 6:16 pm

  5. Non-enforcement isn’t the answer, but neither is enforcement.

    Comment by Charles — June 26, 2012 @ 6:01 pm

  6. I surely am not an expert but, I personally would not discount Mr. Zingales so readily. Here is why:

    You quote IBS in trouble (Bear, Lehman, Merrill, Morgan Stanley) on access to Funds, what about the big/integrated banks that contributed but were rescued from failing? Say maybe because of political ties or because they have deposits.

    What if those IB’s in question, to stay competitive with unified banks, had to take risks that they should not have taken without the blanket of deposits, either directly or via use of complex derivatives? I guess we can call this as moral hazard pass-thru (from one organisation with deposits, due to market competitiveness to to others without deposits). I assume the inherent moral hazard would change how the integrated banks value the risk, due to their size, effect the market fundamentals, thus gets built into how the markets value their risks. (if these entities were smaller, they probably would not be able to change market fundamentals. Quite like how the LTCM crisis lead to the volatility Smile…

    I assume IB with high returns, would not want to gobble up a deposit bank and as they still would need to deliver competitive returns. Or due to other integration pains, like who will be the CEO after integration.

    In either case, I agree with Zingales, on keeping the players smaller; not sure if G-S by it self would do the trick, yet, Extensive Vertical integration, in a market with low number of players, would/could shield inherent risks which otherwise would have been exposed, leading to what might seem like Black Swan events…

    Comment by Yucel — June 30, 2012 @ 1:26 pm

  7. [...] admit) have pumped for the restoration of Glass-Steagall.  Luigi Zingales is a prominent example. I expressed my criticisms of Zingales’s advocacy of Glass-Steagall a little more than a year a….  Certainly the bloviations of Warren and McCain [...]

    Pingback by Streetwise Professor » Doctors Warren and McCain: Screwing Up the Diagnosis and the Prescription — July 11, 2013 @ 6:39 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress