A reflection on the recent Warren/Litan Funding Flap
A few words on U.S. Senator Elizabeth Warren and this letter calling out Dr. Robert Litan over some industry funded research.
Industry funded research, IMO, is no different from any other research in the sense that it should stand and fall on its merits, not on its source. In my own field, for example, it is not enough to simply say that a study is funded by AT&T, or Google, or Consumer Federation of America and treat that fact alone as undermining its conclusions. A study by AT&T can be as methodologically sound -- or unsound -- as a study funded by a grant from a foundation or research coming out of a NIH.
The reason genuine disclosure is so important, however, is because most people rarely read studies in full and are rarely in a position to evaluate them. As a result, people tend to give greater weight to the parts they read (generally the executive summary and the conclusion, if that) based on the perceived neutrality of the source.
Which brings us to Dr. Litan. If Litan had appeared at the hearing as an expert witness for the Capital Group, noting his credentials as a non-resident Brookings Fellow, his appearance would be unremarkable and the source of his funding would have been subject to the usual cross-examination of expert witnesses. Had he deliberately hidden his funding, no one would question that he had failed to disclose a conflict of interest.
What makes this case interesting, and very typical of the policy world, is that Litan did disclose on the report that "funding for this study was provided by the Capital Group, which provides investment services worldwide." Litan also disclosed this in his testimony. At the same time, however, he maintained that the analysis and conclusions were his own (and those of his co-author, Hal Singer).
Because Litan's conclusions were so at odds with other research, and because of the acknowledged source of the funding, Warren asked further questions about Litan's relationship with Capital Group and the study. Litan responded that Capital Group paid for the whole study and that they had input into the initial outline, also saw a pre-release version, and provided some additional source material.
None of this actually goes to the merits of the study. Nor is Dr. Litan's conduct particularly nefarious, or even out of the ordinary. At the same time, Senator Warren is right to point out that the way Litan framed his appearance at the hearing and the nature of the initial disclosure were designed to emphasize the independence of the study and obscure the connection with industry. As noted above, this gives to the public and most policymakers an impression of independence that influences evaluation of the evidence which might otherwise be viewed with a more skeptical eye.
It would be nice to believe this is a clear cut case of right and wrong, but it is always somewhat grayer than that. Warren is quite right to point out the level of involvement is important, and that it raises questions for Brookings Institution if they allow a resident scholar to proffer work as primarily done through his affiliation with the Institution rather than presenting the work as being done in his role as expert economist for hire and noting his fellowship at Brookings as a credential of competence.
But finding a perfect formula for disclosure is difficult. I applaud Warren for shining light on a perrenial Washington problem. But there is never going to be a perfect formula for this.