Cheers and thunderous applause greeted the co-chief executive officers of Primerica Inc. as they entered the Georgia Dome in June for their company's biannual conference.
As pyrotechnics burst on stage, Rick Williams waved to the audience, briefly hugged his counterpart and exited the stage. As Williams joined his wife and investors in the front row, co-CEO John Addison addressed the 40,000-person crowd, dominated by agents for the Duluth, Georgia-based insurance and financial-products company.
"Rick would walk out of his way to not be in the spotlight," Addison says with a chuckle several months after the June event. "And if I see one on stage, I'll walk toward it."
The two chief executives—who have lead Primerica since 2000 and oversaw its spinoff from Citigroup in 2010—describe their arrangement as a perfect pairing.
"Our skill sets are very, very different," Williams says. "I am the analytical, financial, make-the-trains-run-on time person. John is the motivational, intuitive half of the brain. I could never do John's job, and John wouldn't want to do my job."
Recent research supports Williams' notion that effective co-CEOs have complementary skills and challenges the belief that these arrangements often fail.
The analysis, published in the August issue of the journal The Financial Review, found that firms with co-CEOs have higher market-to-book values, which shows the market value of a company relative to its accounting value and suggests how much a company is worth. In addition, the tenure of co-CEOs was not much shorter than the average tenure of sole CEOs.
"These companies perform quite well," says Matteo Arena, assistant professor of finance at Marquette University and one of the study's three co-authors. "It's not an arrangement that's right for everybody, but it definitely works for most of the companies that have adopted it in the past."
The study upends widely held beliefs. What sets it apart from past research, Arena says, is that it looks at data from 10 years and follows 111 publicly traded firms. Past studies focused on a single firm or considered the performance of a number of companies during a brief period of time.
Firms including Apollo Group Inc., parent company of University of Phoenix, and Ralcorp Holdings Inc., the maker of Post cereals, have co-chief executives.
Like in past research, Arena and his colleagues found that these management arrangements arose during transitions when retiring chief executives mentor successors, because of mergers and acquisitions, among co-founders and in family owned firms. But unlike in past studies, the No. 1 reason they saw: choice.
"It's a choice made, believing that it's the best way for the company at the moment," Arena says.
One of the newest co-CEO arrangements began this year at Farmer Bros. Co., a wholesaler and distributor of coffee and tea to institutional clients. The Torrance, California, company's board tapped its chief financial officer and the president of its largest division to serve as co-CEOs on an interim basis. The move came after the financially troubled company said CEO Roger Laverty III would be retiring to spend more time with his family.
"What I didn't want to have happen is for us to spend our time being co-CEOS with each trying to become the sole CEO," CFO and co-CEO Jeffrey Wahba says. "It's worked out so well that we both went back to the board and said, 'If you end up deciding you want one of us to continue, our preference is that both of us continue in this role.' "
When Wahba and Patrick Criteser disagree, they work toward a compromise that satisfies each.
"We realized that turning around the company promptly is what is important," Wahba says. "There have been very few times when we've materially disagreed. When we do, we talk it out and come to a solution."
At 2-year-old SeatGeek, which forecasts ticket prices for sports and concerts, co-CEO Russ D'Souza handles marketing and his co-CEO, Jack Groetzinger, focuses on products. When D'Souza and Groetzinger first met with venture capitalists, they often were asked which one would become CEO and what the other would do.
"I don't think there needs to be a single person in 'power' or a single point of authority," D'Souza says of the New York City company. "But I also think what works for a 12-person company may not work for a 50-, 60-, 70-person company."
The arrangement is rare, says Gregg Passin, U.S. leader for executive remuneration at Mercer.
"Most companies—99 percent of companies—don't even entertain the question," Passin says. "The board should ask themselves, 'Why do we need this? What would we get from these two people that we wouldn't get from one person?' "
One advantage: Co-CEOs serve as mutual monitors. That's important because many of these firms have less-independent boards and they tend to have less debt, which means they don't have lenders monitoring them, Marquette's Arena says.
This organizational structure hasn't worked for every organization that's experimented with it.
Faced with weak advertising sales, Martha Stewart Living Omnimedia Inc. abandoned its co-CEO structure in 2009.
This year, Jude Thompson resigned from Papa John's International Inc. just one year after becoming co-CEO of the pizza chain.
And Research in Motion, faced with grim quarterly results and a bleak forecast for the rest of the year, has faced pressure to justify why Mike Lazaridis and Jim Balsillie serve as co-CEOs and co-chairmen. The BlackBerry maker has promised shareholders that a committee of independent directors will determine whether the co-CEOs need "significant board-level titles" and will recommend a governance structure by the end of January. The Waterloo, Ontario, company declined to comment for this article.
For Primerica, the arrangement appears to be working well. In November, the company reported that profit rose 2.5 percent to $40.6 million for the third quarter, compared with $39.6 million in the same period last year.
"A relationship like this—when it's the right personalities, the right skill sets, a shared vision, they actually like each other and they can put their ego aside enough to look out for the needs of the organization—works better," Addison says. "But if it's the wrong people, it's a disaster. I don't think there's an in-between. I think it works incredibly well or it's a disaster."