Deepak Lal: The banking conundrum

THE CREDIT MARKET CRISIS - II

In the run-up to the Asian financial crisis, a former student (not by any means the brightest) who had joined a Wall Street investment bank rang to thank me for having given him a reference for what had turned out to be an excellent job. He said he had just received a very large bonus for doing one of the simplest jobs imaginable. He had been put on the Thai desk to trade in the country's bonds. This job he said was a no-brainer. Given the interest differential between Thai bonds and US interest rates, all he had to do was to buy Thai bonds and earn a tidy arbitraging profit for the firm. When I asked him if there wasn't a risk that the baht might be devalued, he said that was a minor risk as the Thai government (and with implicit IMF support) was committed to a fixed exchange rate. Of course, the baht was devalued. I only hope he unwound his Thai bond positions before then, but I have not heard from him since!

This example illustrates the problem of perverse incentives in the current global banking system, which has led to a number of global financial crises caused by imprudent bank lending. The 1980s' debt crisis was due to the imprudent sovereign lending by offshore branches of Western money centre banks (of their deposits of OPEC surpluses) to fiscally imprudent Third World countries, at negative floating interest rates. The Asian crisis of the 1990s and the linked LTMC crisis were caused by imprudent lending based on bankers' gambles on the non-occurrence of low probability events (changes in exchange rates and/or interest rates). The current sub-prime mortgage crisis, which resembles the Third World debt crisis of the 1980s, in so far as in both cases abnormally liquid banks funnelled cheap credit to borrowers who would never have been considered to be creditworthy by any standard of prudence. In all cases this imprudence ended in tears, particularly for the many innocents caught by the subsequent real adjustments that had to take place in the affected economies. But most bankers, like my former student, in the good years when the money was being shovelled out had received large bonuses, which more than made up for any temporary loss they may have suffered when their gambles turned sour and they were sacked "" only to be rehired by someone else when the next manic banking spree started. Clearly there is some systemic fault running through the world's banking systems. This article seeks to identify this, and suggests some cures.

The former economic counsellor at the IMF, Raghuram Rajan, has suggested (Financial Times, January 9, 2008) that the system of annual bonuses by which traders are compensated needs to be changed. For these rewards provide an incentive for traders to gamble on events in the tail of the probability distribution of risks. Until these low-likelihood events occur, their gambles seem to pay off, allowing them to show higher returns than the market average, and thereby earn large annual bonuses. By the time the low-probability event occurs, leading to losses, they have already taken their bonuses and run. Regulation of traders' remuneration packages is then recommended. But this form of "incomes policy" only makes sense if the high-risk trading activities actually damage the financial system. Currently it does, but only because the previously separated activities of commercial and investment banking are now combined along with mortgage lending in "universal" banks. Because the traders' activities can affect the balance sheets and thereby the "money creating" (or destroying) activities of the "clearing bank" bit of the consolidated "universal" bank, they can affect the overall monetary system.

This could be avoided even in universal banks if the depositors in their "clearing" component had to bear the risks of the gambles being taken by the "investment" or "mortgage" components of the bank. But though this discipline on reckless lending did exist in the free banking period in Scotland and the US in the 19th century, today the deposit insurance, which is ubiquitous, and politically impossible to rescind, precludes this method of avoiding moral hazard.

One of the complementary laws passed along with the creation of the FDIC in the US during the New Deal was the Glass-Steagall Act of 1933, which separated the investment and commercial halves of the banking industry. This was done to prevent the speculation indulged in by the former "universal" banks, which had contributed to the collapse of broad money during the Great Depression. With the financial "Big Bang" in London in the early 1980s and the repeal of the Glass-Steagall Act in the US in 1999, the banking industry reverted to the universal banking of the earlier era.

This "deregulation" was justified by monetary theorists, beginning with Gurley and Shaw and culminating with Eugene Fama, who argued that there was no real distinction between clearing banks and non-bank financial intermediaries. This, as explained in my last column, illegitimately removes the crucial distinction between deposit-taking banks, which can create money out of "thin air", and deposit-taking non-bank financial intermediaries like mortgage lenders, which can't (see T Congdon: Playing with Monetary Fire, Economic Research Council, London, December 2007, for an elaboration and discussion of the baneful effects of dethroning the pivotal role of broad money "" including bank deposits "" in determining macro-economic outcomes). Thus the only "regulation" which is needed is to go back to the future: the commercial deposit taking role of banks, which affects broad money, once again, needs to be separated from the investment banking and mortgage lending of the financial system. True banks and hence the macro-economy can then be protected from the imprudent gambles which no doubt will continue to be taken by the masters of the universe in the newly restored and distinct non-bank financial intermediaries "" but they will then be gambling with their own and their clients' money and not those of taxpayers.