believe we need a broader approach. In fact, the two-hundred-year tradition is a good deal broader. Let me take a few pages and say a little more about the broader approach. According to Friedman, positive economics has to do with "what is," while normative economics has to do with "what ought to be." Positive economics is a social science, and as such is subject to the same checks on the basis of evidence as any science. By contrast, normative economics has a moral or ethical aspect, and as such goes beyond what a science can say.
It is true that economics cannot rely on experimental methods to verify its hypotheses, in many cases. (Experimental economics is now a growing field, but still somewhat limited in scope). However, the same is true of some of the sciences, ranging from astronomy to ecology. These are observational sciences, and so is positive economics.
Let us illustrate the distinction -- and some of its pitfalls -- by an example. A person might say "Everybody ought to be paid the same hourly wage, because it is just that each person should be rewarded in proportion to her labor." This is clearly normative economics -- it has to do with what should be.
Now, a "positive" economist might observe that this rule would be inefficient, in the following sense. Some occupations require more training, more effort, or more talent than others; or they are more responsible. If these occupations are better rewarded, they will be better performed, and overall productivity of labor will increase as a result. This increase in productivity will be more than enough to pay the higher wages for the skilled, talented, effortful and responsible occupations, with something left over that might make everyone better off. This is positive economics so far as it goes, and like any proposition of positive science, it might be either true or false.
But suppose it is true, and suppose the "positive" economist goes on to say that, because equal wages are inefficient, wages should not be equalized. That would be normative, not positive economics, and it would be fallacious. The normative economist can respond with perfect logic that justice is more important than a reduction in output which may well be quite modest. There is a hidden assumption, and the assumption is that people "ought" to value production over equality of reward, and there is no logical reason why they should, and some people do not.
What the positive economist can do is spell out (so far as the evidence and reason permit) the consequences of a rule such as equal pay. This is important, since rules always have consequences that are hard to anticipate, and if we understand the consequences we may want to reconsider our support for the rule. But the rules we propose, and the consequences we are interested in, depend upon our values. Not all consequences are equally important! In the last analysis, positive economics is the servant of normative economics. It is a powerful servant, but a poor master, and "positive" economists who lose sight of this and try to give "scientific" answers to questions of "what ought to be" are not only mistaken but, I believe, ineffective. (I don't believe Milton Friedman ever made that mistake). It is possible for economics to be too "scientific."
Some modern (or postmodern?) economists believe we need a broader approach. In fact, the two-hundred-year tradition is a good deal broader. Let me take a few pages and say a little more about the broader approach.
Reasonable Dialog
