Citi’s expensive TARP exit

By Felix Salmon
December 14, 2009
paying big to exit TARP:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Citi’s paying big to exit TARP:

The moves will result in a pre-tax loss of $10.1 billion that will likely be taken in the fourth quarter from accounting charges taken on the value of the repaid preferred shares and the cancelation of the insurance plan. The new stock offering, meanwhile, will severely dilute erode the value of existing Citigroup shares.

Once the repayment deal is completed, it will still take several more years to clean up the financial carnage. Citigroup has not posted a substantial profit in seven quarters, and the bank is expected to muddle through most of 2010 amid another wave of mortgage and credit card losses. And, like several big rivals, the bank continues to lean heavily on government support through a debt guarantee program that makes taxpayers liable if it is unable to pay back the loans.

There’s lots of talk this morning of Citi’s strong capital ratios once this deal is done. But capital ratios aren’t everything. Citi needs to become (a) smaller and (b) more profitable, while (c) being a force for good rather than evil with consumers. It’s actually doing a reasonably good job with (c): on things like overdrafts and credit-card fees Bank of America and even Chase are generally much worse, maybe because their US consumer-banking arms are so much bigger.

But Citi is still weighed down by those toxic assets that the government has put so much effort into trying and failing to get off the banks’ books. It’s selling some of them, slowly, but what it really needs to be able to do is recapitalize itself through earnings power. And there’s no sign of that happening any time soon.

Update: Citi calls to say that they’re reducing assets quickly, not slowly. They’ve put the assets they’ve marked for getting rid of into a vehicle called Citi Holdings, which includes business with about $900 billion in assets at the peak in the first quarter of 2008. That number was reduced by about $183 billion to $715 billion by the end of 2008, and today it’s $617 billion — a further reduction of almost $100 billion. More’s to come: selling Nikko Holdings will take another $25 billion off that total, and with the exit from the loss-sharing agreement with the government, Citi no longer needs Treasury approval to sell assets in the loss-sharing pool.

Still, it does seem that the low-hanging fruits — things like the Smith Barney brokerage — have largely been plucked, and that the pace of reductions in the assets at Citi Holdings is going to continue to decelerate in the quarters and years ahead, even accounting for divestitures like the upcoming IPO of Primerica. What’s more, core assets at Citicorp have been flat all year at $1 trillion, which itself is Too Big.

I’ll give this to Citi, though: at least they are shrinking, deliberately: they’re moving in the right direction. Which is not something you can say about most other too-big-to-fail banks.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

no sign of that happening? have you seen overnight rates? 0% overnights rates ARE a recapitalization of bank balance sheets via earnings power! (note the treasury curve – 30 year record steepness). as long as they can pay 10 bps on deposits, and have Fannie/Freddie buy mortgages at sub 5%, the banks are printing money!

i’m a mega bank bear, but if the Fed can leave rates low long enough, then there’s a chance that this “Extend and pretend” plan can actually delay writedowns long enough for banks to earn enough money to cover coming losses.

Posted by KidDynamite | Report as abusive

KidDyna, banks are not paying 0% on their bond borrowing, on their brokered CDs, preferreds, in fact on most all of their borrowing put in place prior to last year…

look up banks’ Net Interest Margin – the NIM is what is supposedly so wide that all these banks are going to “earn their way out”…. NIM is LOWER today than it was a year ago, and two years ago, and so on…

talk 2s/30s all you want, REAL Cost of Funds vs REAL quality assets have a historically narrow spread.

Posted by jswede | Report as abusive

Most banks fluctuate in a pretty small range of NIM because of competition, floating rates in a variety of products, government support, and hedging of interest rate risk. So WFC for instance has been 4.xx% for the last several years. Thus jswede is technically right. However KidDynamite gets at important point: large banks currently have enormously cheap funding supporting enormously cheap lending to their customers.

Without a lot of government support neither of these would make sense. We now know how much risk there is in banks – why would a capital-holder fund them cheaply without faith in government bailouts for depositors and creditors? We also now know how much risk there is in all forms of lending – why would a bank put out 5% mortgages in this climate without knowing that the government will buy them and bail the bank out if they get in trouble? Without lots of cheap government funding, all forms of credit would be scarcer and more expensive.

If you shift all the rates up due to a more neutral government policy, I think banks would probably be doing less lending at lower spreads because there’s simply a limit to what already highly-indebted customers can afford to pay to roll their debt. Cheap funding has been a huge boon particularly to companies like WFC that had enough trust from customers and faith in their own capital base to write a lot of new business at low rates/huge spreads this year. If WFC was paying 8% for 10-year debt, 5% for longer deposits, and 3% for whole-sale funding (to create an imaginary yield curve that probably would look funny), they simply wouldn’t find a lot of takers for the 8% mortgages they would have to be selling and they would be running at much lower NIM on whatever they could sell (maybe 2-3% instead of over 4%). This would make it much harder to earn out from under the pile of terrible Wachovia mortgages and only moderately better WFC 2005-7 mortgages. WFC needs to be able to get customers to refinance/take loans to move, both to avoid defaults and to generate the fee income that the bank depends on. We also can hope that they’re doing good underwriting on these loans, but that might be too much to hope for. At least they’ll do it better than Wachovia did.

Posted by najdorf | Report as abusive