Over the past 10 years, commercial banks have set up all kinds of investment bank combinations in the hope they can turbo-charge their profits. That made superficial sense when you looked at the money the investment banks were making then. But the idea that commercial banks need investment banks should have been laid to rest by now, if only by the enormous losses which the investment banking divisions have inflicted on banks such a Citi and UBS, and brokerage Merrill Lynch.
Much of the losses at the commercial banks that have bought investment banking units were created by tiny teams of people. Merrill Lynch was brought down by a small team of CDO specialists in its fixed-income division. UBS suffered through a combination of its in-house hedge fund and mortgage-backed securities specialists. Both banks had tens of thousands of other employees working in many other units which have done very well over the past few years. AIG was brought down by its numerically small financial products team, which was composed partly of former bankers from the failed 1980s junk bond firm Drexel Burnham. One of Drexel Burnham's leading lights, convicted felon Michael Milken, is (with delicious irony) speaking at the CLSA forum in Hong Kong this week.
Investment banks get three things from commercial banks. Firstly, they get control. Commercial bankers and their boards often donÆt understand what the investment bankers get up to and may be afraid to challenge them, especially when they are making money. Secondly, investment banks get ultra cheap funding from commercial banks. In other words, much of their profits came from simple carry trades. Thirdly, they get access to the commercial bankÆs loan book, which they use as raw material. They get paid handsomely for their expertise in packaging simple loans into complicated products û for example, changing mortgage loans into asset-backed securities, then into collateralised debt obligations (CD0s), and finally into squared CDOs.
It's worth looking in detail at this unfortunate mingling of investment banking and commercial banking operations because it appears to be at the root of the current crisis. This is what might happen: bank X makes housing loans, which its in-house investment banking team repackages into asset backed securities (ABSs). It sells off most of them. The ones that are too dangerous, the bankers package up as CDOs and sell off in their turn. And the ones that are left over from that packaging are again repackaged and sold off as squared CDOs (ie they are composed of other CDOs rather than ABSs, like ordinary CDOs). Of course, the investment bankers can also re-package the assets for other clients, reaping fees in the process.(CDOs are a technique for spreading risk through diversification and over-collateralisation û a theory which failed after house prices collapsed across the US, something previously thought impossible.)
It was in order to access these loans that many banks in the US were buying mortgage providers, and may have been one reason why HSBC got involved in Household.
The really damaging thing about CDOs for the banks is that they depended on bad assets, since the purpose of CDOs is to create the impression that junk products are better than junk if you package enough of them together. In other words, the bank is actually incentivised to source bad assets (itÆs not a simple matter of slack underwriting standards), for example by making loans to inappropriate home buyers through subprime mortgages. The bank charges these homebuyers very expensive interest, repackages the assets into supposedly prime assets and sells them off to investors.
The bank benefits from the (for example) 10% interest it's getting from the NINJA (no income, no job) borrower and the 3% it is giving the investor (who thinks he has bought triple-A or double-A products thanks to the miracle of financial engineering). The problem arises when the banks keep part of the CDOs on their own books û often the supposedly safest tranches since investors want the higher yield offered by the tranches that were considered æslightlyÆ less safe. When the supposedly safe tranches on their own books turn toxic, the commercial bank has to provide capital to offset them.
This approach to business is not good for the bankÆs sense of risk. ItÆs gone from a prudent assessor and warehouser of risk, to a bank actively seeking out less credit-worthy borrowers to whom it can make loans, supposedly secured by constantly rising asset prices. The bank has become like a usurer or loan-shark, deliberately saddling borrowers with loans they canÆt repay in order to reap the interest income (which is so high it rapidly outweighs the principal) and the fees from letting other predatory investors take a bite of that interest. And its sense of risk management has become completely distorted.
ThatÆs one aspect of it. The other aspect is that investment banks rely on very high levels of leverage to increase their return on equity û that concept so sacred to bankers. ROE is simple: it pretty much depends on your levels of debt. With debt becoming much tighter in the future (both internally from the commercial bank, if it has any sense, and from other external funding sources), itÆs impossible to see investment banks maintaining their current levels of profitability.
The conclusion is surely clear. Commercial banks should go back to what they used to be good at: assessing risk and providing useful loans to businesses and homebuyers at a decent profit. And stay away from the predatory lending business.