Relinking the Normative to the Positive in Economics

Most economists believe that economics is basically a positive value-free science, and that because there is a clear difference between the language of description (“is” statements) and the language of prescription (“ought” statements), there is strong separation between positive economics and normative economics. This view reflects a number of methodological errors and misconceptions about the nature of language and science (cf. Blaug 1998; Mongin 2006; Boumans and Davis 2010), and what should rather be said is that economists’ explanatory frameworks, which are value-laden in many different ways, largely determine how they formulate economic policy. This is especially the case with respect to the explanation of the individual one employs. Indeed, policy’s impact is ultimately on people, and so its formulation depends first and foremost on how they are explained. Thus, the standard form of policy recommendation in economics, Pareto efficiency judgments, rules out making interpersonal comparisons, only because the utility function conception of the individual defines people as collections of subjective preferences, which are supposed to be incomparable. Of course, restricting economic policy recommendations to Pareto recommendations significantly limits the space of policy evaluation and eliminates a whole range of fundamental normative concerns – fairness,rights, justice, equality, and so forth – from the domain of economic policy. That is, the standard “description” of the individual determines what is allowed and not allowed in policy “prescription.” Most economists overlook this evident bias, because they are committed to some form of the preference conception of the individual, and suppose it to be value-free. This has had odd effects on recent efforts to reintroduce some of these missing normative concerns into economics through the “back door” by way of social preference analysis of such things as fairness and equality. It’s a “back door” strategy because introducing a concern with fairness and equality as arguments in utility functions means they are not valuable per se, but only as a means to increased preference satisfaction. What is wrong with this is that such things as fairness and equality have normative value irrespective of whether they increase preference satisfaction. They constitute fundamental normative concerns people have that are independent of the question of preference satisfaction. Therefore, they ought to also be objects of economic policy, because they also concern the well-being of individuals; and to act as if they are not, or are subsidiary to preference satisfaction, simply biases economic policy – hardly what one would expect from a “positive value-free science.”