Working Paper: NBER ID: w29091
Authors: Joshua S. Gans
Abstract: This paper examines how a firm's choice of the type of experiment impacts on its potential exploitation of new technological opportunities. It does so in the context of the failure of successful firms (or disruption) where the literature has informally suggested that firms undertake errors in experimental choice (in particular, choosing experiments that involved biased signals). It is shown that firms will generically choose biased over unbiased experiments even when there are no differences in their relative costs. This is done to better inform decisions regarding the exploitation of technological opportunities. It is shown that these choices can differ between incumbents and entrants based on their fundamentals as well as because of the anticipation of competition between them.
Keywords: No keywords provided
JEL Codes: L26; O32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| firm type (incumbent vs. entrant) (L26) | type of experiments chosen (C90) |
| choice of experimental method (biased vs. unbiased) (C90) | ability to exploit technological opportunities (O33) |
| existing organizational structure and customer base (L22) | choice of experimental method (biased vs. unbiased) (C90) |
| entrants (M13) | choice of experiments that do not rely on existing customers (C90) |
| choice of experiments that do not rely on existing customers (C90) | better decisions regarding disruptive innovations (O36) |