Le Citi Toujours Dormer…

Here’s another perspective on the Citigroup Crisis.  Courtesy of Brad DeLong at Grasping Reality with Both Hands: The Semi-Daily Journal of Economist Brad DeLong.

Le Citi Toujours Dormer…

Why oh why can’t we have a better press corps? Eric Dash and Julie Creswell write that:

To which the only appropriate response is: "Huh?" How can losses out of $43 billion of optimistically overvalued asserts eliminate $224 billion of value? Eric Dash and Julie Creswell don’t answer that question. They don’t even seem to recognize that it is a question that they should be interested in. That they were given this story to write, and that no editors said "wait a minute! this doesn’t add up!" is yet another signal that the New York Times is in its death spiral: not the place to go to learn anything about an issue.

Here they are:

The Reckoning – Citigroup Saw No Red Flags Even as It Made Bolder Bets: In September 2007… Citigroup’s chief executive, Charles O. Prince III, learned for the first time that the bank owned about $43 billion in mortgage-related assets…. [M]any Citigroup insiders say the bank’s risk managers never investigated deeply enough. Because of longstanding ties that clouded their judgment, the very people charged with overseeing deal makers… failed to rein them in…. Citigroup’s stock has plummeted to its lowest price in more than a decade, closing Friday at $3.77. At that price the company is worth just $20.5 billion, down from $244 billion two years ago….

The bank’s downfall was years in the making and involved many in its hierarchy, particularly Mr. Prince and Robert E. Rubin, an influential director and senior adviser…. For a time, Citigroup’s megabank model paid off handsomely, as it rang up billions in earnings each quarter from credit cards, mortgages, merger advice and trading. But when Citigroup’s trading machine began churning out billions of dollars in mortgage-related securities, it courted disaster…. 

From 2003 to 2005, Citigroup more than tripled its issuing of C.D.O.’s, to more than $20 billion from $6.28 billion, and Mr. Maheras, Mr. Barker and others on the C.D.O. team helped transform Citigroup into one of the industry’s biggest players. Firms issuing the C.D.O.’s generated fees of 0.4 percent to 2.5 percent of the amount sold….

Citigroup’s risk models never accounted for the possibility of a national housing downturn, this person said, and the prospect that millions of homeowners could default on their mortgages. Such a downturn did come…. The slice of mortgage-related securities held by Citigroup was “viewed by the rating agencies to have an extremely low probability of default (less than .01%),” according to Citigroup slides used at the meeting…. Mr. Maheras continued to assure his colleagues that the bank “would never lose a penny,” according to an executive who spoke to him….

“There is really no excuse for institutions that specialize in credit risk assessment, like large commercial banks, to rely solely on credit ratings in assessing credit risk,” John C. Dugan, the head of the Office of the Comptroller of the Currency, the chief federal bank regulator, said in a speech earlier this year. But he noted that what caused the largest problem for some banks was that they retained dangerously big positions in certain securities — like C.D.O.’s — rather than selling them off to other investors. “What most differentiated the companies sustaining the biggest losses from the rest was their willingness to hold exceptionally large positions on their balance sheets which, in turn, led to exceptionally large losses,” he said…. Citigroup has suffered four consecutive quarters of multibillion-dollar losses as it has written down billions of dollars of the mortgage-related assets it held on its books…

The narrative structure seems to be: They did not tell the bank to wash its hands! And it caught a cold! And then it died! The villains!

This does not seem to me to be satisfactory…

Look at it this way: A bank like Citigroup has a lot of assets–a lot of people have promised to pay it a lot of money in the future. Let’s collapse all those dates in the future at which people have promised to pay Citi down to one point in time four years in the future, and let’s collapse all the amounts promised in the pre-crisis situation in all their different configurations down to one number $1,263 billion. Citigroup thus has assets: in four years people will pay it $1,263 billion in cash.

Citigroup also has liabilities in the pre-crisis situation. It has borrowed–i.e., accepted deposits (that is what a deposit is: the bank has borrowed money from you, but they call it a deposit rather than a borrowing so that you will think that what you put in the bank is still there) and issued notes to the tune of $800 billion.

So Citigroup in this pre-crisis situation has assets of $1,263 and liabilities of $800 billion and thus is worth $463 billion right? Wrong. The $800 billion is essentially due tomorrow–if the depositors and creditors want to get it out rather than roll it over, they can do so. The assets are in the future and are uncertain. Future assets are worth less than current assets: this is the time discount on safe assets. Risky assets are worth less than safe assets: this is the risk discount. With a (safe) time premium of 4% per year and a risk premium of 2% per year, Citi’s assets are worth 6% less on the open market today for each year they are in the future. Compound this over four years and you get a risk factor of 0.792. In this pre-crisis situation, Citi’s assets could only be sold on the open market for $1,000 if they had to be sold immediately. But you still have a healthy bank: assets with a present value of $1,000; liabilities with a present value of $800; a net worth of $200 billion.

But things can go wrong:

So now you have the post-crisis balance sheet of Citi:

As best as I can guess, things (1), (2), and (4) have gone wrong:

Citi’s (and everybody else’s, it seems) risk models were wrong: assumed that the tails of distributions were much too thin–never mind what they were doing making calculations based on tail densities about which you inevitably have no information at all anyway).

We have gotten bad news about housing prices, independent of the erroneous distributional assumptions in the risk models. This seems to have cost Citi another $30 billion or so in impairing the future value of its assets.

The Federal Reserve has pushed short- and medium-term safe interest rates down far to diminish the magnitude of the liquidity premium–the preference for cash now rather than cash later. This would have raised the value of Citi’s assets but for the explosion of the risk premium. The risk premium on other investment banks’s assets has gone from 2% to 6% or so. Citi’s premium has gone up to 8% because it right now bears the additional risk that the government will step in and nationalize it, confiscating much of the shareholders’ equity stake in the process.

Should Citi’s management have planned for and guarded against this explosion in the risk premium? I certainly did not expect it–I did not think we could see this big a rise in the risk premium outside of a real cousin of the Great Depression, and I thought that modern tools of macroeconomic management would keep such a thing from happening. I never expected to see the unemployment rate hit 15% in my lifetime. I still don’t.

It’s in the nature of a bank to get into trouble and be on (or over) the edge of failure in a financial crisis. Banks exist to provide liquidity and safety: to turn the long-term highly-risky investments in plant, equipment, and infrastructure that are our social capital into the short-term liquid largely-safe assets that savers largely want. This means that banks are–if they are doing there job–long duration and long risk, and their values crater whenever there is a financial crisis because a financial crisis is a sharp fall in the value of long-duration and high-risk assets.

A bank that has not lost massive amounts of value in the past year and a half is either extremely nimble or extremely lucky: even the nimble and lucky JPMorgan Chase has lost 60% of its shareholder value in the past year and a half.

The question of how much duration and risk a bank should assume per dollar of capital is a knotty one–if you match durations and assume no risk, then your stock value never crashes. But shareholders are paying you to be a bank, not to be a not-bank. Rubin has a lot of big wins in his career: as a risk-arbitrage trader, as head of Goldman Sachs, the 1993 budget, the 1995 Mexican rescue, the 1997-98 East Asia crisis–that suggest that his judgment is generally good, and that he takes aggressive but appropriate risks that are in the interests of his principals. You got to know when to hold ’em and know when to fold ’em, but even if you do you may still break even.

The article says that Bob Rubin is racking his brain right now trying to think of what he could have known or could have learned back in 2005-6 that would have told him that Citigroup had taken on too much subprime mortgage risk, or that its internal risk-management controls were deficient:

Asked then whether he had made any mistakes during his tenure at Citigroup, [Rubin] offered a tentative response. “I’ve thought a lot about that,” he said. “I honestly don’t know. In hindsight, there are a lot of things we’d do differently. But in the context of the facts as I knew them and my role, I’m inclined to think probably not.” Besides, he said, it was impossible to get a complete handle on Citigroup’s vulnerabilities unless you dealt with the trades daily. “There is no way you would know what was going on with a risk book unless you’re directly involved with the trading arena,” he said. “We had highly experienced, highly qualified people running the operation”…