Recently I looked into the definition of nudge put forward by Thaler and Sunstein. I found that it was seriously flawed.
As a philosopher, such observations are important to me. After all, concepts are what structure our understanding, research agenda and policy advice.
No doubt, the concept of nudge aims to capture an important idea. Yet, as it is used currently, it captures too little and too much with the result being that of conceptual confusion.
Pointing out the problems, I expected that most in the field would quietly take notice and adjust – much like if you claimed that a central empirical observation doesn’t replicate under certain conditions. Yet, it seems that conceptual inconsistences doesn’t trouble our community as much as replication. So here is my attempt to stir things up a bit – while paying my utmost respects to the scholars involved.
First we state the original definition, and then we state the problems:
“A nudge, as we will use the term, is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives.”
Problem #1: It’s not a definition
Assuming that a definition states necessary and sufficient conditions, the original definition of what a nudge is, is not a definition. It does not state what the necessary or sufficient conditions are; it only states what does not count as a nudge.
This elimination method only works, if it tracks the concept satisfactorily (and even then it’s not a definition). But as the original ‘definition’ on the one hand not only includes save-more-tomorrow pension programs but also hot-dogs, and excludes big-gulp bans, 5 cents tax on plastic bags and asymmetric dominance effects, it is definitely not satisfactorily.
As everything else in my 'notes-series' this post is written piecemeal. So sign up, if you don't wan't to miss the rest of this post.