Economists are Biased. Who Would Have Guessed?!
[Excerpts taken from the blog “Ideology is Dead! Long Live Ideology!” by Mohsen Javdani and Ha-Joon Chang, posted at the website of the Institute for New Economic Thinking]
“Mainstream (neoclassical) economics has always put a strong emphasis on the positivist conception of the discipline, characterizing economists and their views as objective, unbiased, and non-ideological…the matter has never been directly subjected to empirical scrutiny. In a recent study, we do just that.”
“Using a well-known experimental “deception” technique embedded in an online survey that involves just over 2400 economists from 19 countries, we fictitiously attribute the source of 15 quotations to famous economists of different leanings. In other words, all participants received identical statements to agree or disagree with, but source attribution was randomly changed without the participants’ knowledge.”
“For example, when a statement criticizing “symbolic pseudo-mathematical methods of formalizing a system of economic analysis” is attributed to its real source, John Maynard Keynes, instead of its fictitious source, Kenneth Arrow, the agreement level among economists drops by 11.6%.”
“Similarly, when a statement criticizing intellectual monopoly (i.e. patent, copyright) is attributed to Richard Wolff, the American Marxian economist at the University of Massachusetts, Amherst, instead of its real source, David Levine, professor of economics at the Washington University in St. Louis, the agreement level drops by 6.6%.”
“We believe that recognizing their own biases, especially when there exists evidence suggesting that they could operate through implicit or unconscious modes, is the first step for economists who strive to be objective and ideology-free. This is also consistent with the standard to which most economists in our study hold themselves.”
“To echo the words of Alice Rivlin in her 1987 American Economic Association presidential address, ‘economists need to be more careful to sort out, for ourselves and others, what we really know from our ideological biases.'”
To read the article, click here.
Like this:
Like Loading...
The Replication Network 
Hello,
I tried to post the following comment (below), but WordPress would not allow me to. I seem to need a WordPress account, as if I had my own blog (I don’t), or a Twitter of Facebook account (which I object to using, not wanting to grant either the rights to track me).
In any event, I think the points in my comment are worth making, for they have clear implications for the goals of scientific integrity that you folks wish to uphold.
Thanks for your time.
Cheers, Michael Peirce, Ph.D.
————— Establishing bias is MUCH harder to do than this. Fwiw, a number of confounders were not considered or addressed, but needed to be. And these only begin to bring out the complexity of the task. (Also, in case you are wondering, I would love to skewer economists).
(1) Most sentences contain indexicals (‘you’, ‘this’, ‘the Xs’) and unstated but implicit temporal qualifiers and assumed local range restrictions, with the consequence that the same set of words (sentences) will express different things, on different occasions, or different time periods, and when coming from different mouths. For example, ‘The president is bald’ was true in October of 1959, when spoken by a US citizen, but false a year later, or when spoken by a citizen of Nigeria. In the present context, when Keynes used the term ‘symbolic pseudo-mathematical methods’ he was presumably referring to a subset of methods prominent during his day, while if we read the quote as coming from Arrow, we must shift the context, and the presumed referents of this phrase, to different subset, so that it ranges over work that accumulated over the next half century. Accordingly, survey respondents may not be judging one and the same proposition.
(2) Most declarative utterances, especially containing value judgment terms (e.g., “pseudo X”) in order to be interpreted properly, must be understood in a way that is attuned to what the speaker presumably meant and implied by them, their beliefs, and habits of speech, and so on, which often requires “reading into” the context. I don’t know enough about Keynes’ or Arrow’s habits of speech, but if Keynes consistently, and inflammatorily accused others of “pseudo” scientific work, while in contrast, Arrow was known to be more diplomatic and sparing, that could explain a wide divergence in respondents’ judgments, too.
(3) Keynes may have been talking about, or had in mind, the views of Austrian-school libertarian economists, while Arrow would presumably be referring to the views of a different group, in which case, right-wing respondents may have different views about whether the predicates applied to Keynes’ intended targets than left-wing respondents. The situation shifts when the views presumed to be targeted by Arrow are instead at issue. That’s not bias, that’s a difference of opinion.
In sum, shifting the author of a sentence will tend, also, to shift what has been said (the proposition expressed) by it. Consequently, it is too quick to leap to the conclusion that the difference is due to political or ideological biases.
>
LikeLike