Schuster, Mitchell and Brown (2018)
|Schuster, Mitchell and Brown (2018)|
|Title:||Sampling Increases Music Sales: An Empirical Copyright Study|
|Author(s):||W. Michael Schuster, David M. Mitchell, Kenneth Brown|
|Citation:||Shuster, W.M., Mitchell, D.M. and Brown, K. (2018) Sampling Increases Music Sales: An Empirical Copyright Study. American Business law Journal, Forthcoming.|
|Key Related Studies:|
|About the Data|
|Data Description:||The study consists of sales data over a ten year period based on 450 sampling songs obtained from the Billboard Hot 100 Chart (being a chart ranking song popularity based on sales, streams and radio airplay in America). Sampling and sampled songs were determined using the WhoSampled.com website. Several factors were examined to determine the effect of digital sampling on the sales figures of the sampled songs, including: chart entry; year; genre; nature of the sample (e.g. lyrical or instrumental); and whether the sample was used throughout the entire song. A fixed effects model was used to correlate the effect of being sampled on sales of the original work, and an ordinary least squares analysis was used to assess the impact of other factors (such as genre, pervasiveness etc.) on sales.|
|Data Type:||Primary and Secondary data|
|Secondary Data Sources:|
|Data Collection Methods:|
|Data Analysis Methods:|
|Cross Country Study?:||No|
|Government or policy study?:||No|
|Time Period(s) of Collection:||
This Article presents an empirical study of digital sampling’s effect on the sales of sampled songs and — based on the collected data — argues that a reassessment of fair use in this area is needed. To conduct the research, a group of previously sampled songs was identified and sales information for these works collected. The analysis found, to a 99.99% degree of statistical significance, sales of sampled songs increased after being repurposed in a new work. The Supreme Court instructs that the most important consideration in analyzing fair use is the effect on the market for the original. Evaluating a set of songs sampled by works appearing in the Billboard Music Year End Charts for 2006-2015 found increases in sales for the earlier works post-sampling. These gains were enhanced where the sample included lyrics from the original or was pervasively used throughout the new song. Findings of this nature favor judicial determination that sampling constitutes a fair use (and thus no license is needed), but they are not conclusive. In evaluating fair use, courts must also review the influence that a new work has on extant licensing markets to create derivatives of the original. Thus, to the extent a market for licensing copyrighted songs properly exists, influence on this market should be considered in addressing sampling and fair use. This paper argues that the current sample-licensing market is a product of aberrant anti-sampling caselaw and a want of relevant information (such as the data in this study) which incented risk-averse actors to purchase unnecessary licenses. Evaluating digital sampling and copyright law through this lens warrants a whole-cloth review of whether the practice is a fair use. Based on evidence provided herein, this article asserts that sampling should be a presumptive fair use in certain instances that maximize the new work’s capacity to increase sales of the original. Such a presumption furthers the goal of encouraging creative activity without hindering the copyright owner’s capacity to financially gain from their work. Lastly, while the genesis of this investigation was application of the fair use doctrine to sampling, the findings are applicable outside this limited purpose. The article applies its results in the realm of private law through a law and strategy lens. Forward thinking music firms should reframe their approach to sample licensing to achieve a competitive advantage. This goal can be achieved by deviating from the norm of zealous sample policing, and instead offering costless sample licenses to maximize the free advertising and increased sales associated with therewith.
Main Results of the Study
When a song is sampled and incorporated into a different song, the sales of the original song increase (in effect acting as a form of advertising for the original work). Increased sales peak eight weeks after the sampling song enters the chart. Increased sales for the original work are also positively associated with (over the course of 5 weeks after chart entry): the use of lyrical samples (+ 9.1%); the sampling song being a different genre from the original (+8.1%); the sample being used throughout the sampling song (+ 17.7%), and; to a lesser extent the sampling of an older song (+ 0.3%). Songs which had previously been sampled in other works also benefit from increased sales upon subsequent sampling (suggesting that prior exposure does not exhaust any benefits to sampling). Over a 20 week period, a sampled/original work can expect to benefit from an additional $1,263.60 from sales revenue.
Policy Implications as Stated By Author
The study suggests that “forward-thinking” copyright owners and firms should create business models with encourage cost-free sampling in order to maximise their income. Certain limitations may be implemented, including whether usages may be only cross-genre, lyrical or instrumental-only, or limited to certain number of seconds. Should such a practice become commonplace, the authors envisage a Creative Commons style catalog of free-to-sample songs.
In regards to fair use, the study suggests that a presumption of defence should be applied in cases where this is invoked (in keeping with a more utilitarian-based model of maximising innovation, and inadvertently the “advertising effect” for the sampled song).