30 August 2010

The tricky pitfalls of "independent" university-based research

Selections from a report at the Wall Street Journal:
As retirement investments, annuities have had their share of critics. But these days, when there's an argument, people in favor of the pension-like products have the ultimate trump card: an authoritative report by experts from the University of Pennsylvania's prestigious Wharton School. In it, finance professors address seven common annuity questions as "myths," knocking each one down in turn, and say that experts worldwide have decided they're "the best way to go."

The study says one more thing: According to small, pale print on the cover page, it's co-sponsored—that is, paid for—by New York Life, one of the nation's biggest sellers of income annuities. The study's lead author, David Babbel, a Wharton professor of insurance and finance, says its conclusions weren't influenced in any way. "Once New York Life commissioned the study, they didn't know what we were going to come up with," says Prof. Babbel. "That's just how it is."

...The universities facilitating these studies and the professors writing them say they don't allow sponsors—which typically pay anywhere from $5,000 to $50,000 for research—to do any meddling, and that academic freedom is always maintained. And the companies footing the bill say they're attracting attention to important financial issues...

Hans Stoll, director of Vanderbilt's Financial Markets Research Center, says business schools have been accused of not being close enough to the real world. "The connection to business is desirable. I don't think we want to sever that on the altar of conflict of interest."

Still, ethics experts say that appearance issues alone create grounds for serious concern. Over the years research for hire has defended some of the industry's most controversial products, including subprime mortgages and risky derivatives...

To be sure, few investors know that any of this research exists. But the companies sponsoring it send the results to financial advisers and brokers, who recommend products to their clients or quote the studies to confer an extra level of gravitas to their own marketing.
More at the link.   I'm more familiar with this process in the fields of the biological sciences.  One key point to remember is that even though the funding sponsor may stay "hands off" and not influence the data collection or interpretation, they may be doing what some Big Pharma companies have done - fund ten studies, find the one with the most favorable results, and then promote those results in their advertising while ignoring the other nine with suboptimal or negative results.

3 comments:

  1. There is also the old trick of "We won't interfere in your research at all, but if you find negative results then we won't ever fund you again and we will ensure that our friends in the industry never fund you either."

    ReplyDelete
  2. Or yet another trick of first find someone who is in support of your position, and then fund their research.

    ReplyDelete
  3. DaBris is correct, nolandda is partially correct. The fact is that everyone in academia cares more about their reputation (present value around $0.39) than they do about future sponsorship of their research.

    ReplyDelete