George Bernard Shaw famously quipped: "If all economists were laid end to end, they would not reach a conclusion". A common perception from the media is that economists are hopelessly divided on current policy issues. Regardless of the issue, noted economists can be found who stake out polar opposite views. With such divisions within the profession, it is no wonder that economists contribute little to the current policy debate.
How divided are we?
To what degree does this lack of consensus reflect the true state of the profession? To shed light on this, the Initiative on Global Markets at the Chicago Booth School of Business put together an Economic Experts Panel (see IGM Forum). The Panel consists of a diverse set of faculty from seven top departments, based on fields of interest, age, gender, and presumed political affiliation (e.g. including past chairmen of the Council of Economic Advisors under both Republican and Democratic administrations). Each week, the panel members have been given one or more statements on a particular policy issue and asked whether they agree or disagree with this statement, or are unsure.
We examine here the panel’s responses to eighty different topics, in each case measuring the fraction of the panel that did not disagree with the modal view (agreeing with this view or being uncertain).
Figure 1 provides summary data on the degree of consensus among panel members on these questions. If opinions were badly split, then only just over 50% would support the modal view, whereas with a full consensus 100% would support the modal view. Here, we find that very few issues generated significant disagreements within the panel. 32 questions out of 80 in fact generated a full consensus by this definition, and the average consensus across all questions was 94%.
We then attempted to classify these questions into three groups based on the amount of past academic research on the topic (a lot, some, or a little). Questions where there has been substantial academic research (or where the answer followed directly from established economic theory) are marked in green. Those with some or virtually no academic research are marked in yellow and red, respectively. Here we find that disagreements are much more prevalent when there is little or no past research available to guide panel members when expressing an opinion.
Figure 1. Degree of consensus on 80 questions
In Gordon and Dahl (2013), we examined these data more closely to see if there was any evidence in support of different ‘camps’ of economists. We found no detectable subset of the panel agreeing with each other but disagreeing with the rest of the panel. We did find that younger economists were somewhat more likely to disagree with the consensus views, though they did not agree with each other on when to disagree!
How do we reconcile the perceptions in the media of a sharp division of views among top economists with this evidence from the Booth surveys of a strong consensus on most issues? In order to understand the behaviour of economists as well as journalists, the first instinct of economists is to focus on the incentives each group faces. From this perspective, the contrasting perceptions quickly make sense.
Economists seeking fame and fortune
To begin with, what incentives do journalists face when seeking out economists to quote for any given article? To maintain an appearance of neutrality and balance, the reporter must report both sides of any debate. If an expert is quoted defending one side of a debate, then it is important to find an expert articulating the other side of a debate.
Similarly, lobbying groups try to gain credibility by quoting experts who support their positions. Experts supporting one side of a debate may be scarce, but readers have no knowledge of the effort needed to find a spokesperson for any given position.
What incentives do economists face when they are asked their views on a particular question? A short answer would be: fame and fortune.
One source of fame comes from being quoted in the media. What can economists do to gain media attention? For one, the media is likely to be attracted to those who not only are articulate but who also have strongly held views, thereby exaggerating any differences in opinion (journalists, along with Harry Truman, can find ‘two-handed’ economists frustrating).
In addition, though, the media often looks for spokespeople for different sides of a debate. Economists who predictably support a particular side of a debate then become attractive interviewees for journalists. Those representing views that are scarce within the profession will be yet more valuable contacts, creating an incentive for economists to serve as advocates for unusual views within the profession in order to gain media attention.
A second source of fame comes from within the profession. Economists who are quoted in the media as supporting views inconsistent with the academic evidence clearly suffer harm to their reputation within the profession. From this perspective, espousing the consensus views within the profession is a much safer response when interviewed.
However, economists who successfully challenge the consensus and ultimately shift opinions in their direction gain substantial professional recognition. Even if the challenge to the consensus ultimately proves to be unpersuasive, such challenges are an admired part of the scientific process. The payoffs from challenging the consensus can then well outweigh the risks to one’s reputation. This is more likely to be the case for younger economists who do not yet have an established reputation to lose. It is not surprising then that younger economists seek out their own issue on which to establish a reputation for changing the consensus views.
As with other individuals, economists can also be swayed by financial incentives. They might serve as a ‘hired gun’ for some political party. Certain views are also likely to gain more financial support from private foundations, more consulting opportunities, or job offers from partisan think tanks. The ‘market-clearing’ price for particular views will likely be higher the further these views are from the consensus within the profession, but there is always a ‘market-clearing’ price!
One role for the National Science Foundation, National Institutes of Health, National Institute on Aging, and other governmental organisations funding research is to counterbalance the financial incentives economists face to represent special interests. The more funding these government organisations have, the higher is likely to be the price faced by special interests to recruit spokespeople, given the availability of an alternative funding outlet that does not put one’s professional reputation at risk.
Academic journals, and to some degree the media, are increasingly trying to document such financial incentives when academic articles are published or individuals are quoted. There can be no presumption that the views of those receiving funding from interested parties provide any information about the consensus within the profession.
Of course, economists may also let their personal self-interest or personal party affiliation colour their professional opinions. It would not be surprising if high-earning economists, for example, focus on the costs more than the benefits generated from high government spending. Similarly, economists who support a particular party may find more persuasive those research results consistent with that party’s position, and selectively cite these papers.
Regardless of possible financial incentives, personal self-interest, or personal political affiliations, though, the stronger the academic evidence supporting some position the more costly it will be to cite opinions contrary to this evidence. This pattern clearly shows up in Figure 1. The main surprise from the surveys is not so much the strong consensus on issues that the profession has studied at length, but the lack of any clear ‘camps’ on issues where the academic evidence is thin.
These surveys then provide concrete documentation of a strong consensus among economists on a wide range of policy issues. A further advantage of these surveys is that they provide concrete and objective evidence for journalists about the range of views among top economists, obviating the need to find spokespeople for polar opposite views on any given topic in order to provide an ‘objective’ summary of the debate.
The example of global warming
The perception of polarised views, while particularly salient for economists, is certainly not limited to economists. The same apparent lack of consensus among academic scientists shows up, for example, in the media coverage of global warming. 53% of newspaper articles on global warming over a 15-year period in the US gave ‘roughly equal attention’ to arguments for and against human activities contributing to global warming (Boykoff and Boykoff 2004). This journalistic standard for balance prevailed even though a survey of 928 journal articles on the topic over roughly the same time period reveals that none disagreed that humans have contributed to global warming (Oreskes 2004).
Surveys can then provide useful documentation regarding the degree of professional consensus, regardless of field. There are clear gains from having organisations such as Booth School of Business or the Survey Research Center undertake these surveys on an ongoing basis.
Boykoff, Maxwell T and Jules M Boykoff (2004), “Balance as bias: global warming and the US prestige press”, Global Environmental Change 14, 125-36.
Dahl, Gordon and Roger Gordon (forthcoming), “Views among Economists: Professional Consensus or Point-Counterpoint?”, American Economic Review Papers and Proceedings.
IGM Forum (no date), website, available at http://www.igmchicago.org/igm-economic-experts-panel.
Oreskes, Naomi (2004), “The Scientific Consensus on Climate Change”, Science 306, 1686.