A few weeks ago, Kevin Vallier wrote a thoughtful, agonized blog-post (at BHL) on to what degree 'non-ideological' political theory is possible. Before I could pen a response, David Sobel created a robust discussion at Pea-Soup on the related question if it is "inappropriate to hold moral principles in a way that is immune to empirical falsification." Now, both Kevin and David flirt with treating 'ideology' and 'empirical' as contraries; admittedly, both introduce a great deal of sensible qualification and hesitate endorsing the conceptual opposition. Even so, David channels the ghost of Popper and treats the empirically 'falsifiable' as the opposite to the 'ideological.'[+]
Shortly thereafter, the influential economist, Raj Chetty, published a widely discussed op-ed piece in which he tacitly assumes that in virtue of being properly empirical his favored approach to economics is scientic and, therefore, cannot be ideological. In my response, I point out that even if one fully accepts that the new trend in data-mining economics, which exploits so-called 'natural experiments,' is based firmly in fact, it could still be ideological. For it presupposes background stability in one's institutions and norms. Much of the very best of contemporary economics is 'empirical' in this sense; it relentlessly explores the impact of policy within a given framework. (The idea that current, mainstream economics is somehow very formal and far removed from empirical reality is seriously outdated.) This status-quo bias of Chetty's mainstream approach remarkably friendly to existing background institutions and norms (recall also this post on Gul and Pesendorfer).
But was I correct in using 'ideology' in describing economics?
In Vallier's sense, economics can be ideological for it offers a systematically one-sided representation of moral reality. Even if all its claims are factual, its measures do not exhaust all other representations of the very same (moral) reality. A given social state of affairs has an income distribution, an economic growth rate, a natural unemployment rate, (etc.), but it may also have features of just-ness, happiness, environmental risks--some of which may be subjective commitments, but some are also very objective social facts. But one might say that this one-sided representation is built into the very nature of economics--all competent practitioners recognize at some level that Homo Economicus and MaxU are not the whole of social reality (even if these idealizations make for very lucrative and fruitful commitments). It would be not quite right to point to this feature as the way in which economics is ideology. Economists are, in fact, exploring, say, models of bounded rationality, imperfect information, etc.
As a discipline econonomics is also not insensitive to empirical evidence. But it does not follow that it is not ideological. Our very hard-nosed, data-driven contemporary economics is in a very bad position to explore alternative regimes with dramatically different institutions/norms. For example, one need not be a Marxist or an Austrian economist, to recognize that such explorations may well be useful in light of the fact that, say, the working of contemporary central banks clearly favor the interests of the (existing) financial services community. Maybe one day computer simulations of such alternative possible arrangements will find a central home in the discipline. But it is hard to see how one could make those simulations properly empirical. An inability to even think or explore alternative institutional arrangements is an instance of being ideological. Economists are like engineers who take many of the most important constraints for granted. That's okay, of course, for many purposes. But if one's science becomes the dominant way in which social reality is interpreted then this is engineering quality is a limitation.
The more important way in which economics is ideological is in its systematic blindness to its own normative commitments. Some such commitments are fairly innocent: econometricians tend not to regress on backward causation (not doing so often allows the data 'to speak,' as it were, more easily for 'itself.') While I have a fondness for 'funky' causes, I do not think economists are ideological in ignoring them. But it is less innocent if one understands one's own science as akin to, say, "medicine" (recall Chetty) without having a grasp of all the ways in which having "good health" is itself a complicated normative-empirical question and the ways in which the analogy is not appropriate in policy-oriented economics.
One non-trivial assumption that this analogy between medical health and economics imports into economics is considerable value-unanimity.* For example, Pareto-improvement is in some sense an extremely thin notion; yet some such unanimity is presupposed when we treat it as the policy ideal at the expense of other commitments. Even if there were value-unanimity in some community, it would be more appropriate to understand it as a sign of shared ideology than as a bed-rock empirical fact.
In analytical philosophy, we tend to treat 'science' as a paradigmatic 'rational' and non-ideological activity. I have even claimed that philosophy of science rests on quasi-transcendental assumption (QTA): if anything counts as knowledge it is fallible science, especially physics (chemistry, biology, whatever), so let's now articulate how this is possible or, more formally, justified (and develop, say, norms appropriate to this; recall). We are justifiably suspicious of practices that masquerade as science or that are now defended as science; there are several such practices that are best characterized as pseudo-science.
Contemporary economics is a genuine science notwithstanding its inability to secure a robust understanding of fundamental parameters that govern the most important economic phenomena . Even so, as I have recounted in the history of this blog, in recent decades many of its very best and leading practitioners become incompetent bunglers or worse when they engage in issues that ought to force them to reflect on the normative commitments and consequences of their own practice(s). We need a category that properly captures such facts about economics; 'ideology' may be unfairly judgmental, but for now it is probably the best concept we have.
[+] Sobel never uses 'ideological' in his post and opts for 'inappropriate.' Readers are free to judge my tendency to treat these terms as synonymous in this context as a mistake.
[*]This commitment and its association with health has its roots in the thought of the Godfather of Chicago economics Frank Knight (himself quite skeptical about the very idea of economics as science); see his "The Sickness of Liberal Society." (I thank Ross Emmett for the reference.) As I have argued, Milton Friedman made the analogies between health and economics (and associated commitment to a value unanimity)