When Citigroup Inc's board of directors was looking for a chief executive in 2007, it called Michael O'Neill, who had turned around Bank of Hawaii Corp a few years before.
O'Neill spoke to a Citigroup board member on the telephone, but the board was reluctant to consider O'Neill seriously because he had never run a bank of Citigroup's size or complexity, according to a person familiar with the situation.
Five years later, O'Neill, 66, is effectively running Citigroup, which is more than 130 times bigger than Bank of Hawaii was last decade, as measured by assets. He joined the Citigroup board in 2009, became chairman this year and has played an increasingly powerful role, as most vividly shown by his ousting of Vikram Pandit as chief executive in October, after months of tension.
O'Neill, who hand-picked new CEO Michael Corbat, has an uphill task ahead of him. Citigroup is groaning under $171 billion of assets it wants to shed, has high expenses, and its profitability lags behind that of such competitors as JPMorgan Chase & Co.
And O'Neill faces the same question that kept him from being a contender for the Citigroup CEO spot: while he can fix a smaller bank, can he revamp a behemoth as complicated as Citigroup?
"It is an awfully complex situation," said current Bank of Hawaii CEO Peter Ho, who worked for O'Neill. "He relishes simplicity."
O'Neill, who declined to comment for this story through a spokesman, has provided some clues about his plan to turn the bank around. On a conference call with investors the day that Pandit stepped down, he said that he will follow his typical playbook.
A dozen people who have worked with O'Neill over the years say that plan usually involves the ruthless pruning of underperforming operations and deciding which ones are worth additional investment.
As he evaluates a bank, he delves deeply into details to look for value or risk that others may have missed. His decisions may differ from those of competitors, but he has a strong track record of being right, these people said. For example, when he was at BankAmerica in 1998, he exited a type of mortgage lending that brought heartburn to competitors a few years later.
Given Corbat spent three years after the financial crisis helping to dispose of Citigroup's bad assets, he may be a perfect chief executive - and ally - for O'Neill.
Still, it remains far from certain that O'Neill can use the methods he has previously used to turn Citi around. It not only dwarfs Bank of Hawaii but is also seven times as large as BankAmerica was in 1997, when O'Neill was that bank's chief financial officer.
Citigroup is also vastly more complicated and has been exiting businesses for years.
At this stage, O'Neill could still decide to shed a big business, like investment banking. But Citigroup executives who have considered such steps in the past decided that the damage to other businesses, like corporate lending, would be too great.
"Take the process at Bank of Hawaii, and multiply it by 10,000, and you've got what Citigroup will be like," said one former senior Citi executive. It may take years for O'Neill's effort to bear fruit, barring a major global economic turnaround, he added.
Aloha shirts and bar graphs
In 2000, when O'Neill took the CEO's job at Bank of Hawaii, he had to fix a regional bank that was struggling after over-expanding around the Pacific Rim.
Employees, already bruised by deep cost-cutting, were wary of an outsider from the U.S. mainland, said Lori McCarney, the bank's former chief marketing officer.
But O'Neill listened to employees, laid out a strategy and put his own money behind it, investing $10 million in the bank's stock. The former U.S. Marine even starred in a series of commercials, dressed in a flowered aloha shirt.
At investor meetings in New York, O'Neill would show a chart with about 30 bars - about three-fourths represented businesses that were profitable, while the remainder were losing money or breaking even.
"It was a pretty clear graphic picture of what to expect," said Al Landon, a risk manager hired shortly before O'Neill's arrival and who later became his successor.
Within six months of being named CEO, O'Neill announced that the company would shed its holdings in California, Asia and most of the South Pacific.
O'Neill dug into various businesses to understand what was happening. He would suddenly show up in employees' offices or summon them for short discussions, recalls Ho, who was then a senior vice president fairly low down in the chain of command.
"He clearly understood the issues in my sphere, wanted to get my opinion on them, and wanted to make sure that I understood that he understood what was going on several layers down," said Ho, who became CEO of the bank in 2010.
It was in such no-nonsense meetings that O'Neill convinced Bank of Hawaii employees that the company would rationally sort out which of its businesses.
When O'Neill departed in 2004, he left the bank "with a very strong credit culture and they got through the credit crisis better than most banks," said Fred Cannon, who was director of investor relations at BankAmerica and is now director of U.S. research and chief equity strategist at Keefe, Bruyette & Woods.
Former colleagues say O'Neill also has a knack for seeing risks that other executives may miss.
In 1997, when he was chief financial officer at BankAmerica, he pushed the lender to auction off its manufactured housing finance business, which was then booking strong profits as the second-biggest lender in a booming market.
"It was a smart analysis of the numbers using good judgment," Cannon said. "Mike sold that business for a very good price." That deal happened in 1998, and two years later, most players in the manufactured housing finance industry suffered big losses.
BankAmerica, led by then CEO David Coulter, agreed to merge in 1998 with Charlotte, North Carolina-based NationsBank to form today's Bank of America Corp.
A crazy quilt
At Citigroup, O'Neill will have to take these skills to an entirely new level.
Sandy Weill built the bank higgledy-piggledy over the course of decades, starting with a small consumer lender in the 1980s and ending with the largest bank in the United States.
But the bank was slow to integrate its many businesses into a single company, resulting in a crazy quilt of computer systems, overlapping operations and confusing organizational charts, making it hard to figure out whether any business was earning a meaningful return.
The bank has made progress but is still working on these issues, which would complicate a Bank of Hawaii-type analysis on profitability of various units.
Moreover, unlike Bank of Hawaii and BankAmerica, shedding assets will be much harder at Citigroup.
The bank has been disposing of extraneous and risky assets housed in its Citi Holdings unit for nearly four years. It had cut down assets to $171 billion by the end of September from $649 billion in mid-2009, but executives expect the pace of disposals to slow as few obvious buyers remain. What's left is likely to be a source of more losses in the future.
Despite these challenges, veteran bank analyst Mike Mayo said he can think of no better person than O'Neill to be chairman of Citigroup.
"He almost wrote the book on how to decide which businesses should be sold or kept," said Mayo, whose 2012 book "Exile on Wall Street" details his numerous arguments with senior bank executives, including former Citigroup leaders Chuck Prince, Robert Rubin and Pandit.