Keep it simple, stupid

Mervyn King has come out in favour of breaking the banks into commercial and investment banks. We have been here before, and I must say I approve wholheartedly. Needless to say the idea has been dismissed in the past both by the government (who do not know what they are talking about) and the FSA (who also do not know what they are talking about and one might say that they have a vested interest in maintaining the status quo because that gives them more to regulate), but more on that later.

In the 1930's following the 1929 crash the US Senate decided that it would be wise to separate commercial banks, whose deposits were guaranteed by the FDIC, from investment banks. Hence we had Morgan Guaranty, later JP Morgan, which was quite distinct from Morgan Stanley, First National City Bank and Bank of America, which were noit permitted to undertake the range of activities that they do today as Citigroup and Merrill Lynch. Finance was much simpler in those days with a combination of bank lending and deposits by commercial banks, equity issuance and trading and bond underwriting and trading by investment banks. The only derivatives were commodity trades and limited equity options, and securitisation was unheard of. Listed debt securities were issued by governments, municipalities and large corporations.

Life was much simpler, but the system was effective. To see the effect in terms that a modern trader would understand, correlate the decades from the 1920's to the 200's in which a stock market crash of more than 10% led to a recession (1920's, 2000's) against the decades in which the Glass-Steagall Act was in force (1930's-1990's), and lo and behold we have a 100% negative correlation. We deduce, on behalf of the traders that a Glass Steagall Act heavily implies no economy busting stock market crashes, and it is easy to see why. If state guaranteed banks are unable to over expose themselves to market risk (some exposure is inevitable), then when the markets come crashing down, the banks do not find themselves short of capital and unable to lend.

Sadly, the Act was repealed at the end of the twentieth century. The situation we have at present is worse than that with capital starved banks continuing to pay vast salaries and bonuses to traders and debt repackagers because they generally make a higher return on equity than commercial lenders and their ilk.

Mr. King's comments are interesting because they were made on the same day that the BIS released its analysis of proposed changes, adopted in July, to Basel II capital rules. The report calculates how much more capital banks will have to hold under the rejigged rules, which are due to come into effect in January, and indicates an average increase of at least 11.5 per cent of overall capital requirements and 223.7 per cent of market risk capital requirements.

Some people might say that the new guidelines will cause the banks/investment banks to reduce or close some lines of business that were previously capital-lite, and perhaps to cause some banks to hand back their banking licenses which were so helpful when they were looking for a quick bailout.

But why bother with that level of sophistication? The previous trend towards "innovation" (read that as complexity) has been fraught with problems and the new Basel guidelines just give the bankers another bright-line test to game, and the FSA and their foreign counterparts a new game to police. I suspect that they like to play that game because it makes them feel important and it keeps them in work.

But there is another solution, and it is easier to police. Put all the plain vanilla banking in the regulated commercial banks and give them a state guarantee, but limit the games they can play. Put all the rest in the investment banks, and let them play all the games they like, subject to normal company legislation, standard non-bank financial regulation and shareholder activism, and limit the indirect exposure that guaranteed banks can have to investment banks, whether or not they are connected.

Less to police, simpler to monitor and proven to work.