In the early days of file-sharing and the internet, experts worried that technology would gut the creative industries, leaving consumers with fewer (and lower-quality) choices than ever before. It’s clear now that this hasn’t been the case, says Joel Waldfogel, an economist at the University of Minnesota.
Pirating remains a problem, and traditional gatekeepers aren’t as powerful as they once were. But revenue for recorded music is growing, and there are more books, movies, television, and music than ever before. In his new book Digital Renaissance: What Data and Economics Tell Us about the Future of Popular Culture (Princeton University Press), Waldfogel argues that digital technologies haven’t killed creative industries, but they have created a renaissance of new cultural products that consumers like and that wouldn’t have made created otherwise.
The Verge spoke to Waldfogel about why pirating didn’t kill the music industry, the good news and bad news for creators, and the dangers to watch out for in the midst of this renaissance.
This interview has been lightly edited for clarity.
Let’s start with the case of music. When it comes to music, you say that the internet did make it harder to make money off of recorded music because of pirating, but it also made it easier to create music. Can you talk about the relationship between those findings?
In these industries, the fear was that if the internet and pirating threatened revenue, it’d threaten continued product creation, and we’ll be threatened as consumers. That’s a credible concern, and I tried to see what’s really happening to the creation of products and the enjoyment of consumers in the internet period.
The threat of piracy is real, but there’s good news: it’s clearly much cheaper now to produce and distribute movies, television, books, and so on. Creation and distribution have become trivially inexpensive. That’s one important counterpoint, at least, to loss of revenue.
The other thing, that’s in some ways even more important, is the unpredictability of success and how that plays out to potentially have big effects. Imagine that traditional gatekeepers were really good at predicting what was going to succeed. In that case, if the cost of distribution fell relative to revenue, they could greenlight more of the products they used to say no to. But all of those newly greenlit products would be worse than the former lowest threshold, so it wouldn’t be that helpful to consumers to get a lot of not-very-appealing stuff.
But the real world’s not like that at all. In the real world, it’s very hard, even for very sophisticated gatekeepers, to predict what’s going to win. So if some change occurs, it allows more products to be tested with consumers. A lot of them will be bad. But because it’s so hard to predict success, some of them will be really good, at least in the sense that they appeal to a lot of consumers. That lets a lot of new products come to market and be discovered that have never come through before.
Do we have numbers on how many more of these works are coming to market that wouldn’t have if it weren’t for digitization?
There are two kinds of numbers that jump out. One is the absolute number of new products, like a tripling of the number of new songs being brought to market. There’s huge growth in books. It’s good news for consumers, but most new products are unappealing and attract very few consumers.
To me, the much more interesting numbers come from the best-seller lists. What share of those were products that would not have made it through in the past? In music, I would say the growing success of music from independent labels is evidence. In books, it’s the growing success of self-published books. One astounding statistic in my mind is the USA Today best-seller list. It’s a fairly deep best-seller list of 150 titles per week.
The number of self-published books on that list went from about 0 percent to 15 percent in a few years after the Kindle was introduced. In the romance category, it’s nearly 50 percent now. These are books that circumvented traditional gatekeepers after demonstrating value.
What about the question of quality? You write that there’s no evidence that the rise of self-published books is leading to lower standards for literary culture. Why is that?
To answer this question, I looked at The New York Times Notables list, specifically at which of those books came to market as self-published books. There is no change in the share of self-published books that are NYT notable. It’s zero.
Basically, what you find is that self-publishing is displacing the usual mass-market stuff, but it’s not diverting our consumption away from the elevated stuff. There’s not a displacement of traditional elevated consumption.
We’ve been talking mostly about books and music. How has digitization changed the movie and television industries?
Two big things. One is the reduced cost of creation. I can make a cinema-quality movie with a $2,000 camera, and I don’t need film, and it can be inexpensive to shoot. That’s not a big deal if I’m spending $100 million, but it’s a big deal that if I’m an independent guy who went to film school and wants to make a movie.
The other, and perhaps more important, thing is the relaxation of distribution bottlenecks. That’s a fancy way of saying it’s easier to get things to people. It used to be that if you made a movie, you got it into theaters, and that was one of the important ways of generating revenue. Although we have 40,000 screens in the US, there was really only room for a couple hundred movies a year when you consider how many screens and weeks the big movies play. Movies beyond the ones that were going to get wide theatrical distribution didn’t have much of a way to make money.
Now, every phone, television, iPad, and computer is a distribution venue. In the old days, it only made sense to make a movie if I could get enough people near a bunch of theaters who wanted to see it. If I made a movie that would attract an audience of 100,000 in the whole country, that’d be a total non-starter because it’s a few people per theater. Whereas if I can distribute it digitally, I can find those people without having done a very expensive distribution through physical means. So it’s not an economic waste of time to target a much narrower swath of moviegoers, and we really have seen large growth in the number of new movies created.
So far, we’ve been talking about how digitization has been good for the consumer. What about creators? Even if we have lots more creators and lots more cultural products, the industry isn’t very sustainable for any one creator.
There are winners and losers here. There are some people who are able to make money creating things that in the past made none. There are some people who once made more, and now they make less because of competition. There’s still stealing going on.
I am sympathetic to the concerns of creators, but the book is about the consumer standpoint, so the relevant measure is whether there are a lot of new cultural products of value being produced, and that answer is yes. Having said all of that, I think there are some hopeful examples of how business models can change to help creators.
One example is bundled subscription sales, which is basically Netflix and Spotify. In the US, when music is played on the radio, the owner of the sound recording doesn’t get paid. But when you get played on Spotify or Pandora, there is a payment. In the old days, if I got a CD — and I’m old enough to be a CD guy — when I played it the thousandth time, there was no additional revenue.
If you think about the way we used to sell music in the recent past, it was a dollar per song, so you only bought a song if you were pretty confident that it was worth a dollar. Now, you pay $10 a month for access to everything, and it creates revenue where there was none. Recorded music revenue actually rose substantially for the past two years, precisely because of bundled sales. I’m enthusiastic about that.
You write that recorded music revenue started going up in 2016 because people started paying the subscription fee for Spotify, right?
Right. It really took off in the US. It takes a while for people to become comfortable buying digital products as services. You’re used to the ownership model, which is either a physical product or a permanent download you couldn’t sell. But now people are becoming much more comfortable with paying $10 for access.
My next question is one that I’ve personally wondered for a long time. Why is it that Spotify can license basically every song, but Netflix’s offerings have become worse and worse over time?
That’s an interesting question. In video, like with Netflix as they transform from being a re-renter of physical products to a streamer of old products to a streamer of new products, the studio that created movies started to worry about them and said, “I don’t know if I want to make my stuff available. We’ll make our own exclusive content.” So the basis of competition has moved toward “I want to have a bunch of exclusive content to make people subscribe to my service,” whereas in music, it’s really been different. In music, essentially every service has the same products. There are little differences here and there, and for a while, Spotify didn’t have Taylor Swift, but by and large, they compete not on the basis of differential content.
In music, it looks to me more like a race to be big first, to get a lot of subscribers, and have some appealing services like playlists and other aspects that get better as you have more subscribers.
One interesting thing you addressed in the book is digital globalization, or the fear that local cultural products would suffer because ones from America or the UK would become so popular. Why hasn’t that happened?
It’s become cheaper to make things, but another aspect that’s underappreciated is the extent to which digitization makes it possible to sell things to export. People are actually consuming products from elsewhere, and small countries that used to bear the hegemony of Anglo products are finding themselves more successfully exporting. How many Netflix viewers thought they would be watching products in Spanish and Norwegian? There are more doing that than we had expected.
Clearly, you’re pretty optimistic about digitization and culture. But you also note things to watch out for, such as the threat of new technological gatekeepers.
The worry is that a small number of decision-makers will get to choose what gets greenlighted and what we have access to. Traditionally, that’s stuff like traditional movie studios, retailers, even. Music labels used to worry about Walmart having a big share of the market and therefore potentially outsized influence.
Well, now Spotify is playing the role of radio stations and record stores. It’s got a big share of the market. It’s not alone — Apple and Google and Amazon are trying — but we’re really talking about a very small number of players who have an enormous influence on what succeeds. I don’t see evidence of sinister behavior, but we have to be on the watch for this, in part because a lot of this behavior is harder to observe than it used to be. For example, Amazon doesn’t release statistics on sales of its products, so it’s very hard to know. We should be vigilant because we’re allowing a lot of power to concentrate in a small number of hands.