Leadership Spotlight

Continued from previous page

Academic Research with Practical Application

A series of working papers have resulted from the collaboration—with more to come. For example, the first paper, for which Mitchell served as lead author, formed a profile of 401(k) participants and their trading behavior: "We find that most 401(k) plan participants are characterized by profound inertia. Almost all participants (80%) initiate no trades, and an additional 11% makes only a single trade, in a two-year period." They found that the most active traders are affluent, older, with higher incomes, and longer job tenure who use the Internet for trades, hold a larger number of investment options, and are more likely to hold active equity funds rather than index or lifecycle funds.

Most recently, the October 2006 paper elaborated with a surprising result, finding that the risk-adjusted returns of traders are no different than those of nontraders. Even more intriguing, the researchers found contradictory effects from trading. As they concluded: "Certain types of trading such as periodic rebalancing are beneficial, while high-turnover trading is costly. Interestingly, those who hold only balanced or lifecycle funds, whom we call passive rebalancers, earn the highest riskadjusted returns."

The research collaboration between the Pension Research Council and Vanguard is multi-faceted. A related study with Mitchell, Utkus, and Wharton doctoral student Tongxuan (Stella) Yang, looked at 401(k) plan design. It resulted in a 2005 paper published in Restructuring Retirement Risks by Oxford University Press. The team was trying to measure the effectiveness of employer matching in encouraging employees to contribute to defined contribution plans.

Explained Mitchell. "We were interested in the question of whether and how companies can help employees save for retirement. Employers often provide matches in the 401(k) setting but lack a clear idea of what impact the matches might have. For instance, when there's a larger match."

She continued, "Do more people join the plan, and do they contribute more?

To examine this question, they needed firm-specific data.

The reason is that tax law constrains how much workers can contribute to their pensions, with complicated rules that are linked to what other employees in their company do. For instance, when low-paid employees don't join the plan, or don't save much, that limits how much other employees can save. So we wanted to know what everyone in the plan was doing, to understand how these interactive constraints work and how companies can encourage retirement saving more effectively. Working with Vanguard helped us get a much better picture of what real-world employees and employers do."

The results were surprising—and a little depressing. The authors concluded that company matches do induce people to contribute but the effects are small. In the typical 401(k) plan, only 10% of non-highly-compensated workers are induced to save more by match incentives. The authors suggest that automatic employer contributions, regardless of employee participation, can be a more cost-effective and influential means of boosting retirement preparedness.

"This is just the tip of the iceberg—we have more papers coming out," Mitchell said.

Two and a half years into the project, Utkus said they'll keep going as long as they have questions. He explained that creating new knowledge is part of Vanguard's culture. The mission is to help plan sponsors and individual investors make better decisions, and the results are widely applicable. After all, who doesn't want to retire in comfort?

"When we do a formal working paper with Olivia or Takeshi, a lot of the math or statistics will be lost on our clients because they don't have PhDs," Utkus said. "They don't need to. But the introduction, the conclusions, the topline results will be very practical. We're doing high-quality, independent academic research, but on the other hand it has results people can relate to."

Describing her own criteria for getting involved in any project, Mitchell insisted, "I'm a researcher first and foremost, so my condition for getting involved in any project is that it be publishable. We take pains to make sure that we retain publication rights. We work with whoever we're working to make sure we have the right interpretation and caught the nuances that matter."

Bristol-Myers Squibbs Looks at New Product Diffusion

Often the process through which arcane research questions result in accepted business practice is more convoluted than the partner research at Merck, DuPont, and Vanguard. First a concept is picked up by an innovative, influential individual who takes a graduate class, attends a conference, or reads an academic journal.

The individual puts the idea into action, and successful results create believers. Management teams seek to replicate the results across departments—and some team members even take the idea to new employers. As the influence spreads, winning concepts are identified by market research firms, which disseminate the idea from client to client, and by the business press, which writes about the trends. The final stage of acceptance is when the idea becomes a standard business practice, a metric demanded by directors or a function embedded in commercial software.

That's what happened with Christophe Van den Bulte, an associate professor of marketing who has extensively researched social networks and the power of influencers.

He experienced that power himself recently. One afternoon he was contacted by a director at McKinsey, who wanted to discuss an unpublished working paper Van den Bulte had posted on his faculty web page. Two weeks later he had an email from Boston Consulting Group, and not long after that, from Arthur D. Little. The link to the unpublished paper had been found by influencers at each company.

"It's a trickledown process," he said. "Influential people with technical knowledge like to think at high level of abstraction, so they read academic journals and paper. Then they interpret what they find and pass on the conclusion to their peers and staffs."

It's not a coincidence that Van den Bulte both researches social networks and benefits from them.

Marketers at pharmaceutical company Bristol-Myers Squibb knew that opinion leaders can make or break the commercial success of a new drug. What they wanted to know was how to identify those leaders to deploy their representatives effectively rather than blanketing the entire medical community, and they hired pharmceutical marketing specialists Rivermark to help.

The citations in Malcolm Gladwell's bestseller The Tipping Point led to Everett Rogers' scholarly classic, Diffusions of Innovations. Those references led to University of Southern California public health professor Thomas Valente, who said he understood physician networks but not marketing. He pointed Rivermark to Van den Bulte.

Working with Rivermark and Valente, Van den Bulte began a small-scale analysis of physician networks in San Francisco. Using survey data from doctors nominating their most trusted contacts, he mapped connections between physicians.