The Bear’s Lair: Glass-Steagall solves a problem caused by regulation

In the 1980s and 1990s I favored repeal of the Glass-Steagall Act that separated commercial banking from investment banking. Coming from a merchant banking background, where the distinction was ignored, it seemed a piece of FDR-era nastiness, that had decapitalized the investment banks and greatly worsened the Great Depression. However, the 2008 crash and subsequent developments have convinced me that I was wrong. Unless we are to throw out another Depression-era invention, deposit insurance, Glass-Steagall appears inevitable. The only question is how to put the legislation in without causing another ten-year depression.

The problem with ending Glass-Steagall was not investment banking itself, it was trading and the way it combines with deposit insurance. Before 1933, when banks that got into trouble were expected to fail, or be rescued by their peers, there was considerable moral obloquy attached to aggressive trading operations. Depositors frowned heavily on them, as did competitor banks expected to bail them out.

In London, Overend and Gurney failed to get a bailout in 1866 because its losses were largely due to over-aggressive trading in the discount market, while in New York in 1907 Knickerbocker Trust was refused a bailout by J.P. Morgan because its losses were due to aggressive trading in the copper market. Conversely, the entire London banking community rallied round to rescue Barings in 1890 because its losses were due to overaggressive lending in Argentina, a disaster that could happen to a gentleman, if an excessively dozy one.

Without deposit insurance, depositors had to be sure that the bank in which they deposited their money would use it soundly. Barings got money from depositors because it had survived for over a century the vicissitudes of international lending, including a very dicey period lending to U.S. states and municipalities that came unstuck in the Panic of 1837. Depositors wrongly assumed that Lord Revelstoke knew what he was doing in Argentine lending, and when Barings went under, other banks rallied round for a rescue because many of them must have felt “There, but for the grace of God, go I.” If Barings had gone bust, depositors would have started looking coldly at other banks’ more adventurous lending in other countries, and confidence in the London market would have collapsed.

Trading houses existed before 1929, but respectable banks were very careful to limit the business they did with them, by and large lending only against “repurchase” obligations on first class securities. Salomon Brothers and Goldman Sachs existed, but were small, relatively insignificant operations, financing themselves from repurchase agreements and “brokers’ loans” on their customers’ margin accounts.

Issue business in London and the small amount of corporate finance advisory work, was an altogether more upmarket operation. In New York it was mostly done by the big banks; in London it was done by the merchant banks, who were not as well capitalized as the joint-stock commercial banks, but engaged in only modest amounts of trading in such securities as bankers’ acceptances, on which they were backstopped by the Bank of England. Underwriting of large bond issues and especially stock issues was a problem for banks of limited capital. In New York that tended to send the business to the largest houses such as J.P. Morgan, while in London the merchant banks solved the problem by persuading the large insurance companies that they really should want to underwrite the securities they would generally expect to purchase. In both cases, the system worked.

After the 1929 crash, which was at the time blamed for the Great Depression (the lack of a depression after the similar 1987 crash pretty definitively proved this was rubbish) two pieces of legislation was passed that changed the financing business forever. The Glass-Steagall Act split the underwriting business from the big banks (trading could have continued in the big banks if there had been any, but there wasn’t much.) Since the U.S. insurance companies were not trained to underwrite issues and the new investment banks had little experience of trading, this resulted in a decapitalization of the market that reduced issue volume far below even the nadir of the 1920-21 recession. This, together with the dozy economic policies followed by both the Hoover and Roosevelt administrations, made the Great Depression last much longer than it need to, with recovery coming only in 1939-41.

The other major initiative after 1929 was the institution of deposit insurance, initially on $2,500 of bank deposits but since then increased to $250,000. This changed the incentives in the banking business completely. No longer did depositors rely on the bank itself or at worst on a syndicate of its friends, to provide them with security against loss in a downturn. Now they simply rely on the deposit insurance fund, and have no need to worry at all about the bank’s leverage or business practices. This has pushed towards the consolidation of banking, since depositors no longer need to look the banker in the eye when depositing their money with him. It has also enormously increased bank leverage, from about 4 or 5 to 1 in the 1920s to 30 to 1 today (and control of leverage has been further weakened by the absurd Basel system, which rates many of a bank’s assets at zero, thereby hugely favoring wasteful government lending.)

However, deposit insurance has also allowed the big banks to engage in gigantic trading operations. These are a cesspit of moral hazard. If they are successful both the traders and top management walk home with enormous bonuses If they go wrong, the deposit insurance scheme or taxpayers as a whole are forced to bail the banks out. The bailouts even extend to the trading businesses themselves; the $180 billion bailout of the insurance company AIG was undertaken simply to pay off the bank holders of AIG’s credit default swaps, gambling contracts by which the banks had been able to profit from the collapse of the U.S. home mortgage market.

In an ideal world, the solution to the problem is not Glass-Steagall but a Volcker Rule that works, preventing banks from trading for their own account. There is no problem in principle in banks underwriting large securities issues, providing the trading they undertake pursuant to that underwriting does not spread to proprietary trading. In practice, experience in the last few years has pretty clearly demonstrated that it is impossible to design a Volcker Rule with teeth.

There is in any case no advantage to having the corporate underwriting role, let alone the advisory role, carried out by large commercial banks. Ideally, a commercial bank culture should be rule-bound and risk-averse; equally there is really no need to pay fancy salaries to the drones who run those institutions. Underwriting requires an acceptance of risk and a large securities business; in an ideal world it should indeed be carried out by large insurance companies and pension funds, who can thereby earn a modest additional commission by underwriting securities they will buy anyway. Advisory work is ideally carried out by the “best and brightest” who should not be forced to sully themselves with the bureaucracy and office politics of a large commercial bank.

If the pension funds and insurance companies won’t underwrite a securities issue, this is a pretty good signal to the lead manager that the issue is poorly designed, mispriced, or that the borrower/issuer itself is unattractive. Failure of the issue in such circumstances is a very useful market signal. The worst possible result is for retail “wire houses” to stuff the issue into their unsuspecting retail clients, which happened all too often in the pre-1975 U.S. investment banks.

The optimal position therefore is for the banks to prevented from underwriting, as under Glass-Steagall, but with the expectation that the investment banks so formed will sub-underwrite deals among the institutions rather than underwriting them themselves, endangering their limited capital. Investment banks would then naturally form into three groups: client service operations, like the old Morgan Stanley, which did little trading and subcontracted underwriting, retail brokerages, which did little trading or underwriting and acted as investment managers for some of their clients and hedge funds, who engaged in trading and fancy investment schemes, and maybe some underwriting but who nobody asked for an opinion on new issues or other corporate advice.

As for the commercial banks, their deficiencies could be rectified in two ways. First, the very large ones could be broken up, by imposing a maximum limit for bank assets of perhaps $500 billion. Second, and more important, deposit insurance should be capped at 95% of the deposits insured (perhaps with an even higher limit than the current $250,000 to allow passive investors, such as the elderly, to keep their savings in commercial banks without much risk.) With deposits subject to the first 5% of losses, depositors would agitate for de-leveraging of the banks, at least from the current 30 to 1 to a more reasonable 10 to 1. With no trading or underwriting, banks with that lower level of leverage would be more or less completely secure.

However, the most essential reform to make the banking system work would be to eliminate the current Fed zero-interest-rate policies. With real risk-free interest rates negative, all kinds of idiotic speculations are encouraged and asset values are driven up to unsustainable heights.

Only with a responsible Fed can we have a responsible banking system; regulation alone will not achieve this. However, there are a few simple regulations that are also necessary. With those in place, the great bulk of the current massive over-regulation of the banking system could be swept away, to the great benefit of the economy.

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)