Narratives, Nomenclature, and the Limit or License of Labels: The Effect of Labeling on the Supply and Evaluation of Project Descriptions on



Labeling is a commonly utilized but rarely theorized construct in research on markets and categories. Extending research into categories, features, and their interplay (Askin & Mauskapf, 2017; Pontikes & Hannan, 2014), I consider how the absence or presence of labels affects the production and evaluation of candidate-supplied narratives. Particularly, I examine how labels may either impose limits on the language candidates use to solicit resources, or give license to non-conformity. I study a sudden change to one of the first crowdfunding platforms in the United States,, in which prospective borrowers, in addition to writing descriptions of their loan purpose, began selecting a purpose label. Using embedded topic models (Dieng, Ruiz, & Blei 2019), I show how the frequency and nature of narrative themes changed after labels were introduced. Through regression models on coarsened-exact matched loan requests, I examine the effect of labels on lender assessments of borrower eligibility. In all, I demonstrate labels are a key part of the nomenclature configuring the supply and evaluation of narratives.


Meeting ID: 979 6026 8681
Passcode: 751919