Earlier this month, the New York Times Magazine reached out to Future of Music Coalition with regard to a forthcoming feature. We like to help out with this sort of thing, because we know that music business structures and practices can be quite complicated, and think it’s important that journalists get the facts and context as correct as possible, whatever narrative they’re advancing. Last week, fact-checkers from the magazine followed up with FMC staff. There was a good deal of back and forth as we were provided short paragraphs, and later, individual sentences, from the article and asked to verify whether they were “true.” (Unfortunately, we weren’t provided with much context.)
Alas, what ended up running was rather disappointing. NYT Magazine chose to publish without substantive change most of the things that we told them were either: a) not accurate or b) not verifiable because there is no industry consensus and the “facts” could really go either way.
Steven Johnson’s article “The Creative Apocalypse That Wasn’t” frames itself as a data-driven response to concerns about the plight of creative workers in the digital age. But Johnson’s grasp of the limitations of the data he cites seems tenuous, and he ends up relying on some very dubious and all-too-familiar assumptions. In its sweeping dismissal of artists’ various concerns, the article reads as an exercise in gaslighting.
This is extra disheartening, because in our previous experiences with the NYT’s journalists and fact-checkers, we’ve found them to be rigorous and fair. Likewise, we’ve found Johnson’s books and TV work to be entertaining and often insightful.
We’ll focus our criticisms on Johnson’s thoughts about musicians in this piece, but it’s worth noting that friends in other creative industries—film and publishing—have also reached out to us noting similar objections about Johnson’s coverage of their fields.
Let’s start with Johnson’s core claims about data as it relates to music:
- “More people than ever are writing and performing songs for a living“…”musicians report their economic fortunes to be similar to those of their counterparts 15 years ago, and in many cases they have improved.” “More people are choosing to make a career as a musician or a songwriter than they did in the glory days of Tower Records.”
- Live music revenue can essentially substitute for lost recorded music revenue. “How can the average songwriter or musician be doing better in the post-Napster era?[…]Part of the answer is that the decline in recorded-music revenue has been accompanied by an increase in revenues from live music… Recorded music, then, becomes a kind of marketing expense for the main event of live shows.”
On both questions Johnson seems to fail to fully comprehend the limitations of the datasets that he uses, and the heterogeneity of the musician population. Oddly, he repeatedly seems to nod toward these complications and then proceeds to essentially ignore them.
Over the course of 15 years of research into the musician community, one of our core takeaways has been that there’s really no such thing as an “average musician,” even if it can be enlightening to highlight mean values from a group of respondents in a survey or study. Musicians are a highly specialized group, with a diverse range of business models. Performers and composers work at different scales in different contexts with different professional and creative goals. Technological changes and business developments that impact some kinds of musicians may have no direct impact on others. Qualitative and quantitative data around economics for artists thus resists simplified narratives. As we’ve previously noted:
This is an era that rewards simple explanations: TED Talks that prescribe neat solutions, the ability to learn “everything you need to know about X in one chart.” It’s nice when such things exist, but it’s easy to lapse into a preference for falsely totalizing narratives, and “expertise” is awarded on the basis of whether you can offer such a narrative (bonus points awarded if you can work in an affirmation of entrepreneurial progress that’s basically compatible with our prevailing neoliberal power structures). (read the full article at New Music Box)
Johnson suggests there are more musicians and composers now than ever before. To make his case, he starts with government data collected by the Bureau of Labor Statistics’ Occupational Employment Statistics (OES) program. To his credit, Johnson notes that:
The O.E.S. data goes back to the 1980s, though some of the category definitions have changed over time. This, and the way the agency collects its data, can make specific year-to-year comparisons less reliable.
The OES themselves make this limitation even more explicit in their FAQ:
“Although the OES survey methodology is designed to create detailed cross-sectional employment and wage estimates for the U.S., States, metropolitan and nonmetropolitan areas, across industry and by industry, it is less useful for comparisons of two or more points in time. Challenges in using OES data as a time series include changes in the occupational, industrial, and geographical classification systems, changes in the way data are collected, changes in the survey reference period, and changes in mean wage estimation methodology, as well as permanent features of the methodology.
Johnson writes, “According to the OES, in 1999 there were nearly 53,000 Americans who considered their primary occupation to be that of a musician, a music director or a composer; in 2014 more than 60,000 people were employed writing, singing, or playing music. That’s a rise of 15 percent.” Let’s look at those years of topline data in full.
|Year||“Music Directors and Composers”||“Musicians and Singers”|
It’s valuable to have some trend data, but the chart above also generates more questions. Why the unexplained jump in music directors and composers around 2008? This could be explained by: (a) the tiny sample sizes that the BLS uses to extrapolate these kinds of counts for the whole population, which makes the estimates of the number of people in just one particular profession somewhat volatile; (b) some kind of change in methodology — the fact that BLS is reporting a three-year moving average suggests a huge jump from the 2008 number to the 2009 number; (c) some combination of the two; or (d) something else.
Is it reasonable to suspect that simply comparing 1999 to 2014 might be an analytical mistake? We think so. (Update 8/23/15: New Zealand Statistician Thomas Lumley drills down on this point in a helpful blog post, where he concludes: “A lot of primary and secondary school teachers got reclassified into this group; it wasn’t a real increase. When the school teachers are kept out of “Music Directors and Composers”, to get better comparability across years, the change is from 53000 in 1999 to 47000 in 2014. That’s not a 15% increase; it’s an 11% decrease.”)
Johnson writes,“According to the O.E.S., songwriters and music directors saw their average income rise by nearly 60 percent since 1999.” Johnson here forgets his earlier observation that OES categorization can be overly broad. A full 55% of the people included in this category (actually called “Music Directors and Composers” and carrying the following description: conduct, direct, plan, and lead instrumental or vocal performances by musical groups, such as orchestras, bands, choirs, and glee clubs. Includes arrangers, composers, choral directors, and orchestrators”) in 2014 are elementary or secondary school band/choir/orchestra directors; certainly an important part of the music ecosystem, but probably not the best proxy for say, a professional songwriter or a classical composer, let alone for the plight of musicians generally. Note also that Johnson seems to move back and forth between looking at the two OES categories individually or pooling their numbers, depending on which approach better supports his thesis.
Beyond that, remember that average wages don’t tell you anything about the distribution. Again, musicians are a heterogeneous group and a small handful of extreme outliers (not uncommon in music) can radically distort an average. Average income can stay the same as distribution changes radically over time.
Johnson rightly notes that OES data fails to capture self-employed musicians. For this reason, he moves on to data from the US Economic Census and proprietary data from a firm called Economic Modeling Specialists International. Unfortunately, EMSI’s methodology is proprietary and thus unverifiable. He writes:
From 2002 to 2012, the number of businesses that identify as or employ ‘‘independent artists, writers and performers’’ (which also includes some athletes) grew by almost 40 percent, while the total revenue generated by this group grew by 60 percent, far exceeding the rate of inflation.
Johnson doesn’t specify here what study he’s pulling from but our best guess is that it’s the US Economic Census. (update 8/23: though it’s not the Economic Census’ spinoff, the SBO (Survey of Business Owners) as we originally suspected when we first posted this Friday; our thanks to the Census Bureau for helping clarify). The bureau offers some warnings about methodological changes and categorical changes over time with the Economic Census similar to the OES. But Johnson also omits that this same study found that the number of paid employees working in that category declined from 58,828 to 41,117, a decrease of 28%.
The problems associated with counting musicians also include questions of definition. As Kristin Thomson has explained:
For many populations, there are commonly-understood criteria, or demographic characteristics, or certifications that make this sorting process relatively easy. But unlike – say – doctors or lawyers, there are no formal exams for musicians to pass, nor accrediting organizations to join. There is nothing stopping a young guitar player from declaring herself as a “musician”, packing up the van and hitting the road, nor is there any formal accreditation needed for an eager wordsmith to hang out his shingle as a “songwriter”.
And of course, many musicians have day jobs, and some don’t even bother reporting their music-related earnings, meaning again that the data is incomplete. This leads us to conclude, as we’ve said before, there is no reliable way to measure the size of the musician population. Anyone who tells you they know for sure has an agenda.
On to the questions associated with live revenue. Johnson cites a $30 billion annual live revenue gross, while other sources have estimated lower figures. He writes:
It’s true that most of that live-music revenue is captured by superstar acts like Taylor Swift or the Rolling Stones. In 1982, the musical 1-percenters took in only 26 percent of the total revenues generated by live music; in 2003, they captured 56 percent of the market, with the top 5 percent of musicians capturing almost 90 percent of live revenues. But this winner-takes-all trend seems to have preceded the digital revolution; most 1-percenters achieved their gains in the ’80s and early ’90s, as the concert business matured into a promotional machine oriented around marquee world tours. In the post-Napster era, there seems to have been a swing back in a more egalitarian direction. According to one source, the top 100 tours of 2000 captured 90 percent of all revenue, while today the top 100 capture only 43 percent.
Johnson doesn’t cite the sources of his data in this paragraph. As best we can tell, the account of 1982-2003, and musical one-percenters, comes from Princeton economist and White House advisor Alan Krueger. As best we can tell, the claim of a “swing back in a more egalitarian direction” references a report by a “media strategist” named Liam Boluk.
Both Krueger and Boluk appear to take their data from Pollstar. Pollstar works hard to generate good data, but their dataset is always going to be incomplete. That’s because it consists of voluntary submissions by venues and concert promoters. Some promoters, including the world’s number one promoter LiveNation, have chosen, at least some of the time, to withhold reporting data from PollStar for competitive reasons. (Pollstar’s 2014 data also included nine different productions of Cirque Du Soleil!) Some event producers, like operas and symphonies, jazz clubs, etc, don’t typically submit their data to Pollstar because they’re not really competing in the market that Pollstar covers. Many smaller/grassroots venues might not see any benefit in reporting to the magazine (and in some of these venues, bands are struggling to make it work on the same five-dollar cover charge as in the 1980s). Overall growth in the live marketplace as measured by Pollstar’s data could be attributed to rising ticket prices, more shows, or simply more venues reporting that did not report before. (Krueger offers a lenghty explanation of his attempts to account for the limitations of Pollstar’s dataset , but Boluk provides none.) That amounts to a weak case for the democratization of live revenue.
That weakness is compounded by the fact that gross revenue numbers in any part of an industry tell you nothing about how individuals working in that industry are faring.
As ticket prices climb, so do production costs and audience expectations. Touring costs are never reported publicly. Usually the only people who know about the cost side are band members and—if they have them—their booking agent, their manager(s) and/or their accountant. Sometimes, even well-attended tours—even the expensive tours that drive massive Pollstar grosses—actually lose money.
Gross tour revenues tell us nothing about how much of that gross actually ends up in artists’ pockets. And nowhere does Johnson address the costs of touring (which is curious, as he’s quite eager to mention the decreases in costs of recording and distribution). And, unlike revenue streams derived from making recordings or compositions, touring costs aren’t very scalable. The more shows you play, the more money you have to spend.
As our own research shows touring income is certainly an important revenue stream for musicians who tour, though it varies by genre. But importantly, that’s not everyone. Johnson puts it this way:
“the growth of live music isn’t great news for the Brian Wilsons of the world, artists who would prefer to cloister themselves in the studio, endlessly tinkering with the recording process in pursuit of a masterpiece. The new economics of the post-Napster era are certainly skewed toward artists who like to perform in public.”
But Johnson misses a lot in reducing this question to artists’ personal preferences, a matter of whether they “like” being on the road.. Some artists’ ability to tour is limited by health, age, and whether they have kids. Some, like professional songwriters, session players, etc, are just working in a different part of the industry.
Are musicians making more or less from touring? In 2012, we asked musicians whether their gross revenue from touring was up or down over the past five years. It was an even split. For 28%, their income from live performance has gone up. For 20%, it has stayed the same, and for 28% it has gone down over the past five years.
(Again, remember that these are gross numbers, not a measure of what ends up in artists’ pockets.)
Johnson notes FMC’s research identifying new revenue streams, but fails to note that we point out that not all musicians have access to these new streams, and not all new revenue streams are going to produce meaningful income for those who participate in them.
For example, sync licensing—where revenue is generated when an underlying composition and/or sound recording is attached to a visual product, like a movie, TV show, advertisement or video game—is important for some music creators, but it isn’t a magic money machine. Nor is it new income, as Johnson weirdly suggests—syncs are as old as television and motion pictures with sound. Sure, there are new opportunities like online advertising, video games, web series and the like, but it’s also a very competitive market, and an artist’s mileage will vary. As will their take-home: musicians and composers who retain their copyrights are in a position to make the most money from syncs, as there are no labels or publishers to take a cut. However, that also means that they have to do all the pitching themselves, which can be time and labor intensive, and dependent on relationships with music supervisors. Success here depends on a number of factors, from a track’s “replaceability” to the rightsholder’s negotiating skills. And not every musician is making work that’s appropriate for sync placement. So, while syncs can represent an important revenue stream for musicians and composers, it’s hardly a panacea for challenges in monetizing music.
There are other problems. Johnson writes, “The vast machinery of promoters and shippers and manufacturers and A&R executives that sprouted in the middle of the 20th century, fueled by the profits of those high-margin vinyl records and CDs, has largely withered away.” Here again, Johnson steps away from data and toward conventional wisdom. There are probably fewer A&R jobs and physical media doesn’t do the numbers it once did (though physical sales did account for $6.82 billion in 2014 according to IPFI—nothing to sneeze at!) But in our digital age, promoters are certainly as active a part of the music landscape as ever (ask any music journalist and she’ll point to her overflowing inbox). Social media tools may allow some kinds of musicians more direct connections to potential fans. But again, remember that these tools are not relevant to every kind of musician.
There’s also a more fundamental problem with Johnson’s framing, in that he reduces all contemporary critiques of aspects of the digital economy to Lars Ulrich’s senate testimony about Napster. In doing so, Johnson conflates several distinct lines of criticism into one unified techno-dystopian line of argument to position himself against: “The thrust of this argument is simple and bleak: that the digital economy creates a kind of structural impossibility that art will make money in the future.”
The debate today is structural, but it’s not just about whether art will make money, but whose art, what kind of art, and how much of the money generated by art ends up with artists, and what they’ll have to endure to get it. It’s about how much agency artists get in defining the terms of the digital landscape. It’s coming from artists who hated Napster passionately and artists who really had no problem with unauthorized file sharing. It’s coming from artists who’ve recorded for major labels, indie labels, or no label at all.
Let us be clear: our problem with Johnson’s article isn’t that he fails to conform to some doom-and-gloom scenario for artists working today. Indeed, there are a lot of new opportunities for artists, and those opportunities are worth celebrating. Most frustrating to us is that Johnson reinforces a false binary between pro-technology optimistic futurism and anti-technology digital pessimism. And that simply doesn’t describe the state of the contemporary debate about art and the digital age.
If you want to know how musicians are faring, you have to ask musicians, preferably a whole lot of them. You’ll get different answers from different musicians, and they’ll all be correct in terms of their own experiences. But your overall understanding will better reflect the complexity of the landscape.
(Image via Shutterstock.)