Are So-Called Normative Statements Practically the Same as Suitable Formulated So-Called Positive Statements?

Evidence from a Survey of Economics Professors

Originally published in SSRN GMU Working Paper Series

In this paper, we investigate whether a statement like “the reform ought to be favored” is practically the same as “the reform would yield positive net consequences.” The first statement, featuring “ought,” is one that many economists would deem “normative,” the second, “positive.” But if the two statements are practically the same, where does that leave the positive-normative distinction? We advance the idea that any statement that is apt to be deemed “normative” can be recast as a statement that is practically the same but apt to be deemed “positive.” We generated empirical evidence on the idea by creating a set of surveys, fielded to economics professors. We sent out 3000 and received 574 completed surveys (19.1 percent). The key aspect of the surveys was that most of them contained both a “normative” formulation and a “positive” formulation, enabling investigation and assessment of practical sameness between the two formulations. When we sum all of the intra-individual-respondent differences between his or her “positive” answer and his or her “normative” answer—over all forms of the “positive” formulations, “for humankind (worldwide, present and future)”—we arrive at a sum of differences remarkably close to zero, a result consistent with the contention that so-called normative questions are practically the same as suitably formulated so-called positive questions. An online webpage contains numerous appendixes, allowing full replicability and including a kappa analysis and other ways of viewing our results.

Additional details

Find the paper at SSRN.

To speak with a scholar or learn more on this topic, visit our contact page.

Mercatus AI Assistant
Ask questions about this research.
GPT Logo
Mercatus AI Research Assistant
Ask questions about this research. Mercatus Chatbot AI More Details
Suggested Prompts:
Ask us anything. We use OpenAI's ChatGPT 4o base model to answer any question about Mercatus research.