Motivated by the observation that infringing use of content reveals demand and that infringement may result from frictions preventing legal consumption, we contact infringers of expensive digital images and direct them toward a cheaper product. Two interventions are designed to
reduce search costs for (i) price and (ii) product information.
External Link(s)
Citation
Luo, Hong and Julie Mortimer. 2018. "Infringing Use, Search Costs, and Legal Consumption: Evidence from a Field Experiment." AEA RCT Registry. August 21. https://doi.org/10.1257/rct.3238-1.0.
Licensing outcomes of digital images, including whether or not a license is purchased and, if so, revenue generated.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
There are six groups in total: two control groups and four treatment groups. Emails to the two control groups offer only the option of licensing the infringed premium image currently displayed on the firm's (commercial) website. Emails to all the treatment groups add a cheaper micro-stock licensing option, in addition to the premium option. Two interventions are designed to provide the (i) price and (ii) product information.
Experimental Design Details
Randomization Method
Done by a random number generator in Stata
Randomization Unit
Firm
Was the treatment clustered?
No
Sample size: planned number of clusters
24680 infringing firms
Sample size: planned number of observations
24680 infringing firms
Sample size (or number of clusters) by treatment arms
2904 (control group, without premium-image price information); 2912 (control group, with premium-image price information); 2906 (treatment group, without premium-image price information and without image recommendations); 6522 (treatment group, without premium-image price information but with image recommendations); 2912 (treatment group, with premium-image price information but without image recommendations); 6524 (treatment group, with premium-image price information and with image recommendations)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)