© 2019 Neil Turkewitz

When Economic Theory Fails: A Critical Examination of Waldfogel’s Digital Renaissance

by Neil Turkewitz

Joel Waldfogel has released a book entitled: “Digital Renaissance: What Data and Economics Tell Us about the Future of Popular Culture,” examining the impact of technology on the production of creative materials. Hal Varian, chief economist at Google, goes so far to say that: “Is it the best of times or the worst of times for the entertainment industry? It depends on whether you ask the producers or the consumers. Distilling knowledge gleaned from years of research, Joel Waldfogel digs into this question with gusto and comes up with a definitive answer for all.” Spoiler alert: Waldfogel acknowledges that piracy has negatively affected revenue, but claims that production has increased without a decline in artistic value, and that the public interest has therefore been well-served by the cultural renaissance unleashed by the internet.

Now of course, “definitive answer for all” is quite a claim, and this particular definitive answer is notably self-serving for Google who is lobbying strenuously in capitals around the world — most urgently in the EU right now in opposition to Article 13 of the proposed Copyright Directive, to avoid any additional obligations with respect to its conduct in storing and distributing creative works. “This is the best of times” dovetails quite nicely with parties arguing for maintenance of the status quo, and deserves scrutiny.

I don’t know Joel Waldfogel, but my friends who do tell me that he is smart and honest, and sincerely searching for a better understanding of the digital marketplace. I don’t question this at all, and highlight that there are two distinct things I want to explore: (1) his actual methodology and conclusions; and (2) the way that his work is employed by organizations with a vested interest in promoting certain policies, and in defending the status quo. I also note that my observations are based on Waldfogel’s body of work which preceded this book, including in particular “How Digitization Has Created a Golden Age of Music, Movies, Books, and Television” and “Music Piracy and Its Effects on Demand, Supply, and Welfare” which this book appears to amplify and expand.

To begin, Waldfogel acknowledges that current market — or non-market, conditions have resulted in decreased revenue, thereby prejudicing the well-being of professional creators. He doesn’t engage in the ridiculous question of whether piracy affects sales/revenue. Like any normal observer, he can readily see and attest to the strong causal relationship. While he doesn’t engage in further discussion of this, it is important to note that this itself — even in the absence of any other inputs, raises questions about the moral, social and cultural implications of permitting practices that fail to sustain independent creators. As one observer noted: “Waldfogel, an economist, makes no grand claims about whether the Internet is destroying the social fabric or the body politic.” It is important to remember this when we get to seeing how his conclusions are employed in policy debates.

As an economist, Waldfogel is not interested (professionally speaking, not making any observation about his personal views) in these deeply moral questions. His line of inquiry is much more direct and limited — given the revenue decline, has there also been an observable impact on the cultural production of goods. In traditional economic terms, a decrease in demand (paid) would generally result in a decrease in production. But he observes that production hasn’t in fact fallen, and that if anything, it has increased. That raises a secondary question. Given that we are not talking about widgets, but rather with unique objects whose value is not exclusively economic, has there been a decline in the cultural value of production? Waldfogel seeks to answer that question by examining patterns of consumption, and suggests that the strong demand for current cultural products follows historical patterns of consumption, therefore suggesting that cultural value has not declined. But is consumption really a measure of quality? While Waldfogel creatively seeks to correlate consumption with other variables (like third party rating) to establish an objective basis for determining quality, ultimately his thesis breaks down if consumption is not a good indicator of quality, and it strikes me as palpably wrong. I understand the attraction as an economist, but consumption as a measure of cultural value makes little sense. We would end up valuing the mass produced rather than the unique. Is pop more culturally valuable than jazz? If we were only looking at consumption, then the answer would be yes. But does McDonalds really serve the finest food in the land?

I find Waldfogel’s premise — that consumption patterns tell you something about quality, to lack credibility. While he does some interesting things with correlating data points, there are simply too many holes and assumptions, and he would be well served by talking to folks in the creative community to understand the complexities of artistic production in the current environment. The idea that more equals better, or more importantly, greater societal benefits, is simply flawed. While there may be more musical works being created (and consumed) today than ever before, that really tells us nothing about quality — even assuming that Waldfogel’s data about the nature of consumed content is accurate. As has recently been explored, the long tail is largely mythical, and the head has grown in size. I would have thought that this would have given Waldfogel some pause in using number of works produced as evidence of any form of social or cultural benefit. And consumption is affected by so many factors — technological developments expanding the ability to consume being first among them, that using consumption as a referendum on quality is fundamentally flawed. And of course, the fact that consumption of music has increased as the cost of access has declined (in many cases to zero) only manifests that the most basic principle of economics in fact works!

I do agree with Waldfogel about one thing — copyright isn’t really an incentive to produce, or at least its role is marginal. It is more an incentive to invest in production — a person’s ability to invest his or her own time, or a third party investing financially based on the possibility of financial returns. It’s what separates professional from amateur productions. It’s what creates jobs. It’s what allows artists to hone their craft…to take a year off to write, to spend time in the studio. To create great art. I understand that this poses complexities for economists, and I don’t have any great ideas about how to measure quality. But this points out a limitation of empiricism rather than a limitation on the lessons of observation. There are times when talking to people reveals greater truths than economic theory can test. All I can say is that if acceptance of an economic explanation requires us to reject common sense, then it is probably the economic theory that is failing to accurately reflect reality.

In some of his earlier work, Waldfogel has argued that because costs have fallen, it has allowed for investment in more products. I assume he repeats that in his new book. But that’s wrong. Costs of production haven’t materially decreased, and production costs (if you are only counting material costs) are, and always have been, the least economically relevant aspect of product development. Costs of distribution have certainly decreased, but perhaps not as much as one would think given the need for investment in new data systems. And once you figure in the greater complexity and cost of promotion/publicity, not sure that “costs” writ large have decreased at all, at least for a “like product.” Of course, defining “like product” is highly problematic from an economist’s standpoint given that it requires an assessment of quality beyond a willingness to consume. Quite simply, I think Waldfogel’s assumption that there is more investment (or equal investment over a broader range of products) is completely wrong. At least on the music side, the global label groups, IFPI & WIN published figures on investment by major and indie labels which shows a decrease in investment (but a higher percentage of investment relative to gross label revenue).

The biggest problem with Waldfogel’s analysis is that his theories about quality not being degraded from the demonetization of content gives political cover to those that believe copyright protection is unnecessary to advance the public interest (and to provide the incentives contemplated in Article 1, Section 8, Clause 8 of the Constitution). And everything I see and hear tells me that he is wrong. I speak with artists every day about what’s happening in their world, and it is about as far from a “golden age” as you could get. I mean, it is golden for some, but not for the artist community as a whole. And as we have seen with Hal Varian, Waldfogel’s arguments get simplified and weaponized in political battles about copyright, intermediary liability and internet governance. See for example how Ariel Katz sought to employ Waldfogel’s theories in his own advocacy against a proposal for site-blocking: “For example, even in the recording industry, record labels’ revenue crisis has not slowed down the flow of new music, nor did it lower its quality. On the contrary, even though record labels’ revenue has declined, the flow and quality of new music have not. And across the entire creative industries, digitization and related technological advancement have ushered in a golden age of music, movies, books, and television.

Waldfogel did not consider or weigh the long term social and cultural impact of undermining the economic foundations of a professional creative community. He looked exclusively at one thing: did the decrease in revenue lead to a decrease in the quality and quantity of cultural production. He found — wrongly as I have argued — that it had not. But he did not “find,” nor do I believe he would think himself qualified to find, that society is best served by maintenance of the status quo in terms of intermediary liability and the relative absence of legal accountability in the internet ecosystem.