The idea of a power law, in my view, is fundamental to understanding the music market as well as the superstar phenomenon. For one thing, the distributions of streamed songs, album sales, and concert revenue are all closely approximated by a power law. And so are the numbers of Twitter followers, YouTube subscribers, and Facebook likes that musicians attract.
Figure 4.2 shows, for example, the number of times each of the top 2,500 artists were streamed in 2016, ranked from the most popular to the least popular, based on my tabulation of BuzzAngle data. The most popular artist, Drake, was streamed 6.1 billion times. He was followed by Rihanna (3.3 billion streams), Twenty One Pilots (2.7 billion streams), and The Weeknd (2.6 billion streams). Moving down a hundred places from Drake, the 101st-ranked group was the California band Los Tigres del Norte, which was streamed 0.5 billion times, or less than 10 percent as much as Drake. The sharp drop-off near the top is characteristic of a power law.
Figure 4.2: Number of Music Streams of Top 2,500 Artists in 2016, by Rank
Source: Author’s calculations from BuzzAngle Music data.
At the other end of the distribution, the country rapper Big Smo was streamed 25 million times, and most of the lowest-ranked artists were bunched around that level. A complete list of the 2,500 artists depicted in Figure 4.2 is available on my web page (www.Rockonomics.com), where curious readers and the statistically inclined can download a spreadsheet and analyze the data. (Oh, and don’t feel too bad for Big Smo and the other bottom-dwellers in the figure. Approximately three million artists are included on streaming platforms, so in the grand scheme of things, Big Smo does fairly well. With 25 million streams, he would net around $35,000 in royalties in 2016.) What this figure tells us is that the distribution of music streaming across artists follows a power law, which is what one would expect if popularity is determined in large part through social transmission.
A landmark experiment conducted by the sociologists Matthew Salganik, Peter Sheridan Dodds, and Duncan Watts demonstrated the importance of social influence on our music choices, as well as the tendency for social influence to increase the gap between the most and least popular songs.17 The researchers created parallel universes in which a total of 14,341 participants had an opportunity to download forty-eight songs from hitherto unknown bands. Participants were randomly assigned to a website in which they were either informed of the previous participants’ download rates for the songs or not informed of previous participants’ behavior. In the first website, participants’ preferences could be shaped by social influence effects, and in the second one preferences were determined independently. If participants were aware of how often others had downloaded various songs (social influence), there was a much greater tendency for superstar songs to emerge, and the frequency of downloads more closely resembled a power law relationship than when preferences were determined independently.
Further experimentation yielded another fascinating result: if participants were randomly divided into groups, allowing social influence to affect song selections resulted in different rankings of the forty-eight songs across the groups. That is, if a song happened to start off popular with early participants in one set of subjects, its popularity grew, and if it happened to start off unpopular in another group, its popularity faded by comparison. This is an example of what sociologists call cumulative advantage, as a small edge acquired over a rival, for whatever reason, snowballs into a much bigger advantage over time. Now, some songs had no chance of ever making it to the top. But among the set of better songs (as judged by the rankings of the songs in the experiment with independently determined preferences), almost any song could have ended up on top. It appears that quality gives a song a chance of becoming popular, but quality is not sufficient in and of itself, given the haphazard interactions in social networks.
It is likely that the sociologists’ experiment actually understates the true role of social influence on the popularity of music in the real world, for several reasons. The number of songs to choose from in the experiment was relatively small, so participants could listen to all of them if they were so inclined. Also, marketing, radio play, product placement, media attention, and—probably most important—informal interactions between friends all play a large role in the actual music market, but they were not a factor in an experiment with anonymous participants.
The importance of social network transmission and getting off to a good start is not news to music industry professionals. The multi-talented musician Jacob Collier told me: “The first set of comments posted on one of my YouTube videos can affect whether it goes viral or not.”18 And a good or a bad inaugural Pitchfork review can make or break a new band. Many recordings have bombed because they did not receive radio airtime when they were released. In other words, chance interactions through social networks create a tendency for a market to be dominated by superstars, and elevate the effect of good or bad luck.
Once you start looking at the world in terms of power laws, or extremely skewed distributions, you can’t miss them. Power laws have been used to describe the frequency of words used in the English language (and practically all other languages), the number of people living in cities, the number of electrical grid failures, the distribution of income, stock market returns, the pattern of musical notes in songs, the number of people joining protests, the frequency of web page links, and multiple physical phenomena.19 More important, the social or physical network mechanism that can generate a power law illuminates the process that leads to spectacular success or dismal failure. In many walks of life, an understanding of how reliance on interconnections generates a power law relationship can help us understand why extreme events—like a neighbor’s daughter becoming the next Taylor Swift, or a devastating blackout that lasts for days—sometimes occur.
The type of social dynamic that generates power laws—where people’s beliefs, knowledge, and behavior are influenced by the beliefs, knowledge, and behavior of their associates—can help explain the occurrence of critically important macroeconomic phenomena, including bank runs, financial crises, and housing bubbles. In particular, this type of social dynamic, where individuals’ views are influenced by others’, can create a self-fulfilling prophecy. For example, the widespread belief that home prices will rise causes more and more people to jump into the housing market before prices rise, in turn causing home prices to be bid up even further. That upward movement can last for a period of time before collapsing under its own weight. Indeed, this social dynamic underlies Citigroup CEO Chuck Prince’s famous July 2007 description of the forces that would soon lead to the worst financial crisis of our lifetimes: “When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you’ve got to get up and dance. We’re still dancing.”20 Sad to say, his bank was not dancing much longer, and needed to be rescued by the U.S. government. The same type of dynamic that creates fads in music also can inflate and burst financial bubbles.
What About the Long Tail Hypothesis?
Some have predicted that the decline in the cost of producing music and distributing it over the Internet will lead to greater variety in music consumption, and move music away from a superstar market and toward a niche market. In his blockbuster book The Long Tail, Chris Anderson argued:
Our culture and economy is increasingly shifting away from a focus on a relatively small number of “hits” (mainstream products and markets) at the head of the demand curve and toward a huge number of niches in the tail. As the costs of production and distribution fall, especially online, there is now less need to lump products and consumers into one-size-fits-all containers. In an era without the constraints of physical shelf space and other bottlenecks of distribution, narrowly-targeted goods and services can be as economically attractive as mainstream fare.21
Anderson applied this logic to forecast that the market for books, movies, music, and es
sentially all retail products will become less concentrated and more diversified. The implication is that the shape of the distribution would tilt away from superstars, and a power law would become less powerful in predicting relative performance.
Echoing this theme, the economist Paul Krugman wrote a short New York Times blog post in 2013 in response to my Rock Hall speech, raising the question, “Is this (still) the age of the superstar?” His argument: “Basically, the music business has been hugely disrupted by the Internet. I’d be very curious to know whether that hasn’t changed the calculus: radio play matters much less, the audience has fractured, performers can build a following on Pandora and YouTube.” He ended his post by commenting, “So I wonder if even Alan Krueger is now behind the curve here—and in any case I’d love to see how the trend for the music business looks since 2003.”22
Although it is possible that music will become more of a niche market someday, this has not happened yet. Figure 4.1 makes clear that the music industry has become even more unequal over recent decades insofar as touring revenue—the main source of musicians’ income—is concerned. Since 2003, the share of concert revenue going to the top 1 percent rose from 54 percent to 60 percent.23 And although it is difficult to compare record sales over time because of rapidly changing formats, Harvard Business School’s Anita Elberse finds that song sales became more skewed toward the very top performers from 2007 to 2011.24
So to answer Paul Krugman’s question, this still appears to be the age of the superstar.
The Internet has, to some extent, changed the process by which superstars become superstars—Justin Bieber and Jacob Collier were discovered through their YouTube channels. But it has not leveled the distribution of income or enabled a large number of musicians to earn a middle-class livelihood.25 The Internet may well lead to more music being produced by more musicians, but even in the age of the Internet, a handful of artists remain much, much more popular than everyone else.
The multitalented Questlove, leader of The Roots, recently lamented the highly skewed nature of the music business when I saw him in Miami Beach:
In my world I’d just like to see a balance. A place where Donald Glover and Tyler Perry can exist on opposite ends of the spectrum. The same for Issa Rae and Whitney Cummings, and for Cardi B and Roxanne Shante. I think oftentimes—especially in music, especially in hip-hop—it’s like just one person and no one else. And whoever is the most digestible gets that spotlight and that attention. Meanwhile, there are zillions and jillions of artists who are just as worthy of getting these things. My position is more or less, while the spotlight is still warm, to show people options.26
To a considerable degree, he is fighting an uphill battle against the power curve.
My guess is that—contrary to the long tail hypothesis—the availability of just about all music on the Internet is unlikely to change the tendency for musical preferences to be largely socially determined and, as a consequence, highly skewed toward a relatively small group of musicians. In fact, the number of musical choices facing individuals greatly expanded—and became even more bewildering—with the advent of streaming services, which is likely to lead us to rely even more on our social networks for clues in selecting songs and artists. And the rapidly growing set of curated recommendation systems that use Big Data to help us discover new music is also likely to reinforce network effects, unless there is a surge in demand for curation systems that recommend songs that are both unpopular and likely to stay that way.
Gloria Estefan: Music of the Heart
Gloria Estefan is the most successful crossover artist of all time. With her husband, Emilio, and the Miami Sound Machine, and as a solo artist, she has sold over one hundred million records worldwide and won seven Grammys. She has performed at the Super Bowl and Olympics, for popes and presidents. I interviewed Gloria Estefan on October 4, 2018.
You came from a musical family and starred in the movie Music of the Heart, about music education in schools. Did a school music program influence you?
Very much so, but I also come with the genetics of music. My mom was the diva of her family and school. On my dad’s side, there were also a few artists: there was a classical pianist, a salsa band leader, and a classical violinist. I sang since I talked.
In fifth grade I joined the school band at St. Michael the Archangel School. At the time I wanted to play saxophone, but they would only let boys play the saxophone. So I played the clarinet. To this day I have a deep fondness for the clarinet and reed instruments, and the wooden sound. But it was very important because I didn’t like being the center of attention. I played from fifth to eighth grade. It was helpful in a myriad of ways. The first time I left Miami was to go to Tampa with the school band for a state competition. You learn cooperation. You learn math, because music is math. People think for some reason that art and music are less important than math and other subjects, but it is quite the contrary. Music is really symbiotic with other courses.
Every time I see a school music program cut it really hurts. That’s why I focus on trying to fill in the gap when government takes funds away, and through private donations try to put music back in the schools.
You earned a degree in psychology from the University of Miami. What do you think it is about music that produces an emotional connection in people?
I’ve thought about that a lot because if we had that secret then it would go far beyond a hit record on the radio. Music, just like colors, can change your mood.
Music is something that unites us. I’ve had the opportunity to travel the world, and despite whatever linguistic difference the people in the audience had, I saw literally the same emotional reactions in the same parts of the show. It was totally about the music. It wasn’t about whatever language was being sung.
Most definitely there is a very strong connection from music to our emotional well-being. And likewise, music can have a negative effect as well. Sometimes very hard core, violent music can drive people to do very tough things. Music is a very powerful force.
How has the economics of the music business changed?
The economics of music has changed a lot. It is like we’ve gone full circle back to the 1950s, where it’s become about singles—and people not getting paid for their work. Artists had difficulty getting paid in a different way in the 1950s because a lot of the pop music was literally pilfering R&B and African American music, then making it pop and not paying the writers or the original artists that created the music.
Now, it is just really tough for artists to get paid, period. Before you could be a recording artist and never tour. Now, recording artists aren’t going to get paid. And the Catch-22 is you can tour, but who will go see you if nobody knows your music.
And the fragmentation of the market has hurt a lot because even though the Internet has freed people from having to find a record company—and GarageBand and the technology that now exists has put that ability to create music in the hands of pretty much anyone, and you can upload your tune to iTunes or Spotify or wherever you want to put it—the challenge is getting ears on it!
Before, for example, when we first did “Conga” on the Johnny Carson show, you were guaranteed a top ten hit the next day….You have so many options today that it causes fragmentation.
Nowadays, artists are not even thinking about albums, they’re thinking about EPs, four songs, or just one song. It is very difficult for a young artist to grow the kind of audience that is going to last through the decades that we’ve been lucky enough to have.
What has led to your business success?
A lot of artists have suffered at the hands of their handlers. It is not usual to have an artist that is very much about the business. But my husband and I both are. We both had careers, so when it came the moment to decide do we go for a music career full-time, it was a risk.
We did it because we loved it, not to become rich or famous. I joined the band as a hobby and for fun. Emilio and I were kind of destined to be together. And we were making a lot of money in the late 1970s and early 1980s here in Miami as a gig band. We made our own record. We got it on the air. We could do the disco music, the Latin music. We were in high demand. But we made a conscious decision that we had to stop being a gig band because we couldn’t continue to do weddings and bar mitzvahs and Quinceañeras and try to have a career where you went out and did concerts.
That happened simultaneously. We became very well known in Latin America. We would go down there and do a concert in a 50,000-seat stadium and then come back home and do a wedding for 200 people in Miami.
Rockonomics Page 10
