Why Citigroup Imploded

Brad DeLong and Prince Alwaleed both have theories on why Citigroup imploded as it did. Brad thinks its a combination of factors -- bad modelling of tail risks; bad modelling on the subject of house prices; an unprecedented spike in risk premiums -- all mixed up with the inherently levered-and-unstable banking business model. Add in some bad luck and lumbering size, and you get what we got.

Prince Alwaleed, on the other hand, thinks it's all Chuck Prince's fault.

I come down in the middle on this one. Yes, banking is inherently risky, and susceptible to bank runs and other tail risks. But all good career bankers understand that, on a very fundamental and intuitive gut level -- which is exactly why banking and bankers have historically been so boring.

What happened at Citigroup was that the good career bankers, as typified by John Reed, were pushed out. Sandy Weill and Jamie Dimon, for all their aggressiveness, did have a gut-level understanding of the banking business, and in hindsight it's pretty clear that Sandy's Citigroup did not take on the kind of crazy model-based risks that Prince's Citigroup seemed very comfortable with. Neither, for that matter, did Dimon's Bank One, or Dimon's JP Morgan.

But Prince was not a career banker, he was a lawyer, brought in to be strong where Weill was weak -- which was mainly in dealing with regulators around the world. But when Tommy Maheras and other big swinging dicks from Salomon Smith Barney started making enormous sums of money in structured debt products, Prince considered his role to be cheerleader rather than risk manager, and told them all to keep on dancing. I think it's fair to say that if Dimon had been in charge rather than Prince, a lot of those mistakes would never have happened.

What about Rubin and Pandit? They're investment bankers, and investment bankers often have a naive faith in the powers of dynamic hedging and risk management to be able to protect large institutions from enormous unexpected losses. At Goldman Sachs and Morgan Stanley that might even be the case -- but at Citigroup, with its super-senior CDO tranches and its liquidity puts and its off-balance-sheet investment vehicles, the only coherent way to manage risk is to be very conservative at every point in the chain.

At Goldman, you can have the mortgage desk taking one type of risk and the CDS desk taking another type of risk and a very clever risk manager overseeing them both, often the CEO himself, ensuring that the risks they take cancel each other out. At Citi, just as at all commercial banks, that could never happen: once the risk is on the bank's books, it just sits there, festering, until it either matures away or it blows up. That's why commercial banks are more boring than investment banks. And that's why it's fair to pin a fair amount of Citi blame on managers from the investment-banking side, such as Rubin and Pandit and even Prince, who ran the investment bank before he was elevated to CEO.

Prince Alwaleed says that he's looking for Citi's share price to soar quickly back to its former heights, just as quickly as it fell. With shareholders like that, I'm not optimistic about Citi's long-term prospects. This company is not amenable to quick fixes; rather, it has to get back to its commercial-banking roots, and start doing boring things like training up good loan officers and taking full responsibility for underwriting the asset side of its balance sheet. If it takes on more risk in an attempt to boost the share price, things are likely to continue to deteriorate.