Can anyone run Citigroup? (page 4.)

On the spectrum of aggressiveness, Pandit is thought (as is Morgan Stanley) to be relatively risk-averse. Bob Rubin, according to one Citi board member, actually regarded that trait as a negative when the CEO search process was on. Rubin, of course, grew up in the risk-taking world of Goldman Sachs, and his inclination over the years, so inside sources say, has been to push Citi's traders to become more aggressive. There is no indication, though, that he directly maneuvered Citi into CDOs, a product he has claimed very little knowledge of.

Wanting a sharp pair of eyes beneath him, Pandit has given the job of chief risk officer to Brian Leach, 49, formerly in a risk-management post at Morgan Stanley. Leach is in fact one of three Morgan Stanley alums who have become Pandit's inner circle.

A second, also in a risk-critical position, is John Havens, 51, a dead ringer for P&G's Mr. Clean, who is CEO of the Institutional Clients Group. That's essentially the old Salomon Brothers (brought to the Citigroup merger by Travelers), the corporate bank molded by Wriston, and the new world of Old Lane and other alternative investment engines.

The third Morgan Stanley amigo is Don Callahan, 51, who was once also at IBM and is now, as Citi's chief administrative officer, in charge of a big portfolio that includes operations and technology. Those two parts of the company have around 150,000 employees (think credit card processing, for example), which is more than Exxon, P&G, and 460 others of the Fortune 500 have in total. Pandit views those 150,000 in terms of a big, big budget: $20 billion a year. Callahan is intent on bringing that amount down and has given special attention to certain high-cost operations, such as Citi's U.S. retail branches. Pandit calls the whole effort "squeezing out as much money as possible." In general, the new regime is determined to reduce the expenses that got out of control in the Prince years or that simply built up over time.

Callahan also says that the 50 business reviews Pandit conducted in the past few months (and in which Callahan participated) show his fixation on the efficient use of capital. The executives sitting across from Pandit heard his inevitable question: "Why have you deployed capital this way?" And, says Callahan, "the one unacceptable answer was, 'Because we've always done it that way.' "

We've seen Pandit's imperatives: Raise capital, protect earning power by paying people as well as he can, and by all means get the risk assets under control. And here, now, are his conclusions about Citi's business model. Pandit says that when he began as CEO he was "absolutely, completely" agnostic as to what this model should be. But those business reviews and lots of thinking have led him to a conclusion: Citi will not be split into a consumer bank and a corporate bank. It will instead be a universal bank, which is pretty much what it is now.

True, Pandit says that some "nonstrategic businesses" will be sold, though he's not revealing which those are. But analysts have predicted that one of Weill's buys before the merger, insurance seller Primerica, will go. And maybe CitiFinancial (the old Commercial Credit) too.

Otherwise, Pandit figures that he has four strong products to sell around the world and the global organization to do it. The lineup starts with credit cards, a business that he and everybody else at Citi extol. CFO Crittenden pointed out recently that residential mortgages (from which Citi is retreating) had a return on assets before the credit crisis of less than 1%. For the same period, the ROA on U.S. credit cards (with their indecently high rates, of course) was running 5% to 7%. Globally, says Pandit, and particularly in emerging markets, Citi simply has the "leading" credit card. "In the Welchian sense," says Pandit, "this is a top business." He refers of course to Jack Welch, who wanted General Electric to be No. 1 or No. 2 in any industry it occupied.

The second Citi business on Pandit's list of favorites is wealth management. "Who would have known," asks Pandit, claiming he didn't originally, "that Citi has the second-largest wealth-management business in the world?" (That's second to UBS.) Citi's horses in this race are Smith Barney, the private bank, and the premium CitiGold account at retail branches. A central idea here, says Pandit, is riding the growth of the middle class in emerging markets, and capturing the wealthy besides.

The third strong entry, as Pandit sees it, is Citi's "one-of-a-kind" corporate bank, with its lucrative business in cash management, and the fourth is the investment bank descended from Salomon. That last, says Pandit, is obviously the business most under pressure these days, for both Citi and Wall Street. But a characteristic of Citi's business in emerging markets is that the company generates deposits it must gainfully deploy within local boundaries, and the investment bank is an entity that can play that game- using this money to fund trading positions, for example. So in the emerging markets, says Pandit, the need to be a universal bank is clear.

Whether being a universal bank is essential in the developed markets, adds Pandit, is more debatable, particularly in lush times. "When times are good," he says, "you can be anything, right?" But in period of stress, like now, he says it's nice to have both the deposits generated by a universal bank and businesses like cash management that are relatively unscathed by turmoil in the credit markets.

So how do you ride herd on a colossal company that's delivering four product lines, reaches into every corner of the world, and has a famously fragmented culture?

Under Pandit, as a Citi press release recently proclaimed, you move much more to a matrix form of organization, in which a significant number of executives report to two bosses. In Citi's case, certain functions (like information technology) and the management of product lines are to be centralized, while the company is otherwise to be run through four regions. The idea, said Citi's announcement, is to give authority to a region's leaders, allowing them to "make decisions on the ground."

Few business people would deny the logic of such a plan for a global enterprise. But it is certain in any matrix organization that there will be contention about how compensation and responsibility and decisions are to be shared. For example, there's a chapter devoted to the subject in Phillip Zweig's biography of former Citi CEO Walter Wriston. It has 30 pages and is called "Matrix Wars."

Much later, Chuck Prince edged Citi toward a more matrixed form of organization. But when Pandit subsequently examined how this plan worked, he found that decisions turned far too heavily on consensus, "like it was Iowa caucuses getting together, that kind of stuff." His plan to make the matrix work is to declare exactly who has the authority to make decisions and make sure the rules are followed- and to that we say, "Let us know how it goes."

Unfortunately for Citi, shuffling the organizational deck chairs is not easily going to make up for the damage recently done to its name. The depth of that injury is driven home by a report called "Banks Suffer Write-Downs in Reputation," published in March by Greenwich Associates and based on a survey of 293 financial officers at corporations in the United States, Europe, and Asia.

NEXT: Can anyone run Citigroup? (page 5.)

1 2 3 4 5