Current Projects and Working Papers
of my recent work falls into two broad areas, intellectual property piracy and
the benefits that consumers derive from differentiated product markets. Lately I’ve been working on pricing of
digital products, the supply of new music since Napster, and globalization of
the music industry. Abstracts of
current and recent work appear below with links to the papers that are
available in draft form.
Recent technological changes may have altered the balance between technology and copyright law for digital products. While file-sharing has reduced revenue, other technological changes have reduced the costs of bringing creative works to market. As a result, we don’t know whether the effective copyright protection currently available provides adequate incentives to bring forth a steady stream of valuable new products. This paper assesses the quality of new recorded music since Napster, using three independent approaches. The first is an index of the quantity of high-quality music based on critics’ retrospective lists. The second and third approaches rely directly on music sales and airplay data, respectively, using of the idea that if one vintage’s music is better than another’s, its superior quality should generate higher sales or greater airplay through time, after accounting for depreciation. The three resulting indices of vintage quality for the past half-century are both consistent with each other and with other historical accounts of recorded music quality. There is no evidence of a reduction in the quality of music released since Napster, and the two usage-based indices suggest an increase since 1999. Hence, researchers and policymakers thinking about the strength of copyright protection should supplement their attention to producer surplus with concern for consumer surplus as well.
In the decade since Napster, research has focussed attention on whether file-sharing undermines the protection that copyright affords recorded music, and most observers believe that file-sharing undermines the appropriability of recorded music. For consumers, however, Napster’s effect on demand is less important than whether the diminished appropriability will reduce the supply of new recorded works. The legal monopoly created by copyright is justified to the extent that it encourages the creation of new works, but there is little evidence on this relationship. The file-sharing era can be viewed as a large-scale experiment allowing us to check whether diminished appropriability stems the supply of new works. We examine this hypothesis using a novel dataset on the supply of new recorded music derived from retrospective critical assessments of music such best-of-the-decade lists. We compare post-Napster album supply to 1) its pre-Napster level, 2) pre-Napster trends, and 3) a possible control, new song supply following the iTunes Music Store’s revitalization of the single. We find no evidence that recent changes in appropriability have affected the quantity of new, acclaimed recorded music coming to market. We reconcile this finding of a stable flow of new works in the face of decreased demand with a discussion of reduced costs of bringing works to market, which we substantiate with evidence of an increased role of independent labels since Napster.
3. Music for a Song: An Empirical Look at Uniform Song Pricing and its Alternatives (with Ben Shiller)
Economists have well-developed theories that challenge the wisdom of the common practice of uniform pricing. With digital music as its context, this paper explores the profit and welfare implications of various alternatives, including song-specific pricing, various forms of bundling, two-part tariffs, nonlinear pricing, and third-degree price discrimination. Using survey-based data on nearly 1000 students’ valuations of 100 popular songs in early 2008 and early 2009. We find that various alternatives – including simple schemes such as pure bundling and two-part tariffs – can raise both producer and consumer surplus. Revenue could be raised by between a sixth and a third relative to profit-maximizing uniform pricing. While person-specific uniform pricing can raise revenue by over 50 percent, none of the non-discriminatory schemes raise revenue’s share of surplus above 40 percent of total surplus. Even with sophisticated pricing, much of the area under the demand curve for this product cannot be appropriated as revenue.
See our short essay in VOX EU.
The Economist has a nice write up.
This paper is now forthcoming in the Journal of Industrial Economics.
4. Pop Internationalism: Globalization of the Music Industry, 1960-2006 (with Fernando Ferreira)
Advances in communication technologies over the past half century have made the cultural goods of one country more readily available to consumers in another, raising concerns that cultural products from large economies – in particular the US - will displace the indigenous cultural products of smaller economies. The debate over trade in cultural products has been better informed by theory than evidence. The goal of this project is to document some facts about the global music industry since 1960, using data on popular music charts from 11 countries (Austria, Brazil, Canada, France, Germany, Italy, Norway, Sweden, Switzerland, UK, US). Who buys whose music? How has it changed over time? And how do music trade patterns compare with trade patterns for movies? We find that while Hollywood fare dominates world trade in movies, music is quite different. While domestic musical repertoire is disproportionately popular everywhere, repertoire shares among imports are closer proportional to exporting countries’ GDP shares. Overall, different countries’ musical repertoires have world shares roughly equal to their GDP shares. Surprisingly, as the world has become better linked over the past half century, the US has grown less dominant in music relative to GDP. Since 2000 the trade in music has been more balanced – in line with GDP shares – than at any time since 1960, and domestic artists have increasing market shares in almost all markets. These facts are surprising against the backdrop of technological change that makes trade easier.
See our short essay in VOX EU.
Foreign Policy has a nifty write up.
While private monopolists are generally assumed to maximize profits, the goals of public enterprises are less well known. Using the example of Pennsylvania's state liquor retailing monopoly, we use information on store location choices, prices, wholesale costs, and sales to uncover the goals implicit in its entry decisions. Does it seek to maximize profits or welfare? We estimate a spatial model of demand for liquor that allows us to calculate counterfactual configurations of stores that maximize profit and welfare. We find that welfare maximizing networks have roughly twice as many stores as would maximize profit. Moreover, the actual network is much more similar in size and configuration to the welfare maximizing configuration. An alternative to a state monopoly would be the common practice of regulated private entry. While such regimes can give rise to inefficient location decisions, little is known about the size of the resulting inefficiencies. Even for a given number of stores, a simple characterization of free entry with our model results in a store configuration that produces welfare losses of between 3 and 9% of revenue. This is a third to half of the overall loss from unregulated free entry.
A growing empirical literature examines the relationship between music file sharing and legal purchases of music, but existing studies examine the period before consumers had attractive legal digital a la carte options. The iTunes Music Store has grown quickly since its appearance in 2003, and digital music now accounts for a third of US recorded music sales. Using a new survey of University of Pennsylvania undergraduates, we ask how music file sharing and sales displacement operate in the iTunes era, when the alternative to file sharing is purchasing individual songs, rather than entire albums. We find large amounts of file sharing in this population. Respondents have more stolen than paid music, but the music obtained via file sharing is, for the most part, low-valuation music which the respondents would likely not have purchased. The rate of sales displacement implied by the relationship between stolen and purchased music across respondents is between -0.15 and -0.3. That is, an additional song stolen reduces paid consumption by between a third and a sixth of song. Perhaps surprisingly, this is about the same as the CD sales displacement rate found for the pre-iTunes era using a similar empirical approach on a similar study population.
Intellectual property piracy is widely believed, by authorities in both U.S. industry and government, to be rampant in China. Because we lack evidence on the rate at which unpaid consumption displaces paid consumption, we know little about the size of the effect of pirate consumption on the volume of paid consumption. We provide direct evidence on both the volume of unpaid consumption and the rate of sales displacement for movies in China using two surveys administered in late 2008 and mid-2009. First, using a survey of Chinese college students’ movie consumption and an empirical approach parallel to a similar recent study of U.S. college students, we find that three quarters of movie consumption is unpaid and that each instance of unpaid consumption displaces 0.14 paid consumption instances. Second, a survey of online Chinese consumers reveals similar patterns of paid and unpaid movie consumption but a displacement rate of roughly zero. We speculate on the small displacement rate finding relative to most of the piracy literature.
8. The Challenge of Revenue Sharing with Bundled Pricing: An Application to Digital Music (with Ben Shiller)
Bundling can increase revenue and profits relative to selling products on a standalone basis, and this is an especially attractive strategy for zero-marginal-cost information products. Despite the clear benefits of bundling, it has one major problem: bundling produces revenue that is not readily attributable to particular pieces of intellectual property, creating a revenue division problem. The Shapley value provides a well-motivated solution to this problem, and we use unique survey data to create measures of bundle value and, in turn, to estimate Shapley values for each of 50 bundle elements. We then evaluate feasible revenue sharing schemes, including equal sharing, proportional sharing, and the modified Shapley value of Ginsburgh and Zang (2003, 2004). We first document that the Shapley value is highly incentive compatible (all bundle elements fare better inside the bundle than they do outside on a standalone basis). We then evaluate the feasible schemes according to both their incentive compatibility and their similarity with the Shapley value. We find, not surprisingly, that the feasible schemes are less incentive compatible than the Shapley value. Among the feasible schemes, equal sharing performs worst while the GZ scheme performs best and substantially better than proportional schemes in current practice.
In 2009 Princeton published my fun book, Scroogenomics.
Four years ago I finished a book, The Tyranny of the Market: Why You Can’t Always Get What You Want. More information on that here.