Disclosing economists’ conflicts

By Felix Salmon
November 8, 2010
paper out which gets boiled down quite effectively to its title: "Financial Economists, Financial Interests and Dark Corners of the Meltdown: It’s Time to set Ethical Standards for the Economics Profession".

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Gerald Epstein and Jessica Carrick-Hagenbarth have a 41-page paper out which gets boiled down quite effectively to its title: “Financial Economists, Financial Interests and Dark Corners of the Meltdown: It’s Time to set Ethical Standards for the Economics Profession”.

They took a group of 19 academic financial economists and looked for possible conflicts of interest — board memberships of financial institutions, consultancies, that kind of thing. They conclude:

In this study, we showed that the great majority of two groups of prominent academic financial economists did not disclose their private financial affiliation even when writing pieces on financial reform. This presents a potential conflict of interest. If this pattern prevailed among academic financial economists more broadly this, in our view, would represent an even greater social problem. Academic economists serve as experts in the media, molding public opinion. They are also important players in government policy. If those that are creating the culture around financial regulation as well as influencing policy at the government level for financial reform also have a significant, if hidden, conflict of interest, our public is not likely to be well-served.

The findings understate the severity of the situation in the real world, where most consultancies and substantially all paid speeches are kept secret. Financial economists tend to make a lot of money, most of it from the financial sector rather than their putative employers, and they’re very unlikely to disclose their income or their conflicts in public.

Paul Krugman is an interesting case in point:

Before I went to work for the NY Times I did a lot of paid speaking, mainly to investment bank conferences outside the US… My fee for overseas talks was usually $40-50K.

I do very little paid speaking now, and no consulting, because the New York Times has quite strict rules: basically I can only get paid for speaking to nonprofits that have no possible interest in influencing the content of the column. It’s a good rule – read Eric Alterman’s book “Sound and Fury” to see how speaking fees can corrupt pundits – though it meant that I took a substantial income cut to work for the Times.

Krugman is clearly comfortable with the idea that speaking fees can corrupt pundits, and thinks it’s a good idea that NYT columnists don’t accept them. Yet at the same time he had no problem accepting such fees as an economist, and I’m quite sure he never disclosed those fees when he was writing academic papers on financial subjects.

It seems obvious that when you’re regularly making significantly more than the median national annual personal income from giving a single speech, you’re prone to being captured by the people paying you all that money. And the secrecy makes things much worse. I once mentioned in passing on my blog a consultancy gig which I happened to know about and didn’t think was particularly secret. The consultant in question phoned me up extremely distraught, fearful that the employer, a hedge fund, would read my post and react to it with a whole parade of nasty possible actions. There’s no good reason for such secrecy on either the employer or the employee side — unless, of course, there’s something ethically suspect about the arrangement in the first place.

I don’t know what the solution to this problem is, but clearly more disclosure would be a very good idea. But it’s not going to happen: there’s too much money riding on the continuation of the status quo.

(Via Folbre)


Comments are closed.