Charles Lane writes an opinion piece for the Washington Post today, taking issue with posts at the Monkey Cage, and arguing that the NSF should not fund political science (or the social sciences more generally). My take (other Monkey Cagers may differ) is that his argument starts in the right place, but ends up in the wrong one.
Perhaps it was frivolous to spend $301,000 on a study of gender and political ambition among students, as Flake charges. Or perhaps a report on economic sanctions was a good taxpayer investment, as McCarty and his fellow department chairs insist. The relevant question, however, is whether society could have reaped equal or greater benefits through other uses of the money — and how unreasonable it would be to ask the political scientists to rely on non-federal support. If this research is as valuable as its proponents say, someone other than the U.S. Treasury will pay for it. If anything, Flake’s amendment does not go far enough: the NSF shouldn’t fund any social science. The private sector chronically underinvests in basic scientific research; the costs and risks are relatively high, and the benefits relatively hard to commercialize. Government support compensates for this “market failure,” enabling society to reap “positive externalities” — economic, environmental or military. Federal funding for mathematics, engineering and other “hard” sciences is appropriate. In these fields, researchers can test their hypotheses under controlled conditions; then those experiments can be repeated by others.Though quantitative methods may rule economics, political science and psychology, these disciplines can never achieve the objectivity of the natural sciences. Those who study social behavior — or fund studies of it — are inevitably influenced by value judgments, left, right and center. And unlike hypotheses in the hard sciences, hypotheses about society usually can’t be proven or disproven by experimentation. Society is not a laboratory.
Lane’s argument has three parts. First – that we should think about the opportunity costs of funding political science, as opposed to funding other kinds of research. Second, that if the research is “valuable”, then someone else than the government will pay for it. Third, that there is no “market failure” in the social sciences because there is no way to test social science propositions.
Lane is right to say that we should think about funding allocation in terms of opportunity costs. However, as Seth Masket has observed, he is wrong to suggest that social science findings are no better than value-judgments gussied up with pretty numbers.1 Indeed, if you think about it for a bit, Lane’s apparent belief that there’s no way to establish reliable truths about politics and society is a radically postmodern one. But what’s most interesting, perhaps, is Lane’s suggestion that we should think about market and market failure in the social sciences.
As it happens, there is a market for ‘political science,’ even if it’s one that many political scientists don’t usually compare to their own research. It’s mostly supplied by think tanks, on the right, left and center of the political spectrum, as well as for-profit consultancy firms. These think tanks, to a greater or lesser extent, are market oriented (albeit towards a quite idiosyncratic ‘market’). If there isn’t obvious funding for research on a particular issue, think-tanks will avoid it. If funding dries up for an issue, then think tanks will drop it. Finally, the arguments and findings of think-tank sponsored research usually have to fit into some range acceptable to the sponsor. This is not to say that think tank fellows are hacks, or cut their opinions to suit their sponsor’s measures. It is to say that some kinds of opinions (those which can attract substantial funding) tend to be over-represented in think-tank research, while others are systematically under-represented.
If think tank funding reflected voters’ best interests, this wouldn’t be a problem. Sadly, it doesn’t. Businesses are responsible to their shareholders rather than to the general public, and their funding decisions are likely to reflect this. Unions are responsible to their members – not to society as a whole. Foundations have their own politics and priorities, and individual funders are pretty quirky. Research by think tanks reflects the priorities of this disparate bunch of funders, not the broader priorities of the US public. Again – this is not to dump on think tanks. Much of their work is good; some of what is not good is at least interesting and provocative. Furthermore, there is a lot that the profession of political science can, and should learn from them (e.g. how better to engage in public debate). But we shouldn’t rely on think tank sponsored research alone, since it usually indirectly reflects funders’ priorities. Still less should we rely on research emanating from professional research consultancies, which are typically purely market driven, and hence prepared to find more or less what their sponsors want them to find.
In short, there is a market for political science – but one that’s imperfect in at least two ways. First, some kinds of research will be systematically underprovided, as per Mancur Olson’s arguments about collective action. For example, large scale social science research, which is of benefit to US society as a whole, but not to individual groups or tendencies within it, will be provided suboptimally, or not provided at all by the ‘market.’
Second, there is a broader problem of truth on the market. If Lane got his druthers, and all social science was privately sponsored, some points of view (on the right, left or center) would be over-represented, and some under-represented. People interested e.g. in sponsoring research on the cost-effectiveness of economic sanctions are likely to have strong interests, whether they be in loosening sanctions (so as to get market access), or strengthening them (because they have strong political objections to the regime being targeted). These interests are likely to be reflected in their choices over who they fund, and to what end.
This would hurt US democracy. First and most obviously, it would limit the information available to policy makers. All that they would know about on important social questions was whatever was provided by interested (and often self-interested) private actors. Second, and more subtly, this information would be even less useful than it is in the current system. NSF funded research does two important things. First, it provides widely available datasets on many issues of public importance, which are not systematically skewed to support one interest or another (the NSF likes projects that have social and political relevance – it does not like projects that seem designed to support pre-cooked conclusions). Secondly, it provides funding to expert researchers to work with this data. This not only provides valuable findings, but it helps keep others honest. If someone wants to do sponsored research e.g. on why there aren’t more women involved in politics, and they use their own idiosyncratic data, rather than broadly available datasets, without good reason, they’re likely to get serious criticism from other researchers. If someone uses commonly available datasets (such as those sponsored by the NSF), but skews their techniques so as to reach a predetermined conclusion, then it’s much easier for others to identify the flaws, and to show how better specifications would lead to different results. In short, if we didn’t have a disinterested body, such as the NSF, meeting the public need for objective research on important social questions, then interested actors would (a) have the field to themselves, and (b) would have much greater incentive to cook the books.
Lane is right to think that we should look at the marketplace for the social sciences. He is quite wrong, however, in arguing that social science can’t aspire to objectivity, and hence is blind to the actual market failures that we would see in the absence of NSF funding. The opportunity costs of abolishing funding for the social sciences are very, very high, precisely because the social sciences provide the best and least biased (albeit still imperfect) knowledge we have about the functioning of politics, markets and society. If we didn’t have this funding, we would see other actors rushing in to fill the gap – the problem is that what would fill this gap would be far, far worse, than what we have already. This doesn’t let political science off scot-free – as John and I have argued elsewhere, the discipline needs to do a much better job in communicating its findings to a broader public than it does at the moment. More attention to reproducibility, along the lines that Victoria Stodden is pushing, would be nice too. Even so, I’m pretty sure that Charles Lane would miss publicly funded social science much more than he realizes, if it suddenly weren’t there any more.
1 Lane’s claim that “society is not a laboratory” runs directly against my favorite quote from notorious self-interest serving special-interest-flunky David Hume. “Mankind are so much the same, in all times and places, that history informs us of nothing new or strange in this particular. Its chief use is only to discover the constant and universal principles of human nature, by showing men in all varieties of circumstances and situations, and furnishing us with materials from which we may form our observations and become acquainted with the regular springs of human action and behaviour. These records of wars, intrigues, factions, and revolutions, are so many collections of experiments, by which the politician or moral philosopher fixes the principles of his science, in the same manner as the physician or natural philosopher becomes acquainted with the nature of plants, minerals, and other external objects, by the experiments which he forms concerning them.”