Network Effects: A Value Lost

Why we get less out of more

My Grandma was a small-town cinephile. Born in the twenties, she came of age in southeastern Wisconsin as the Hollywood stars of the golden age peaked two thousand miles west. She watched Bogart and Cooper in their prime with the “girlfriends” she had known since grade school. She used the money she made working “men’s jobs” in the Second World War to see double features, relishing westerns, musicals, comedies, and, above all, mysteries. Grandma’s love of movies flowed through two generations, thanks in no small part to her enormous catalog (complete with an extensive numerical filing system) of films she taped over decades.

When I came home from college, she could no longer drive but still loved movies and would gladly go to whatever film I wanted to see (even paying for popcorn with my ticket, which my broke college self very much appreciated). She continued to love the cinema through her eighties. “I read someone say that George Clooney is the nearest thing to Cary Grant today, and I agree,” she told me. As much as she loved movies, she was not a snob. Tropic Thunder remained one of her favorites, next to Casablanca and North by Northwest. That is not to say she loved every movie; she hated Shutter Island, disliked rom-coms as a genre, and passionately disliked Frank Sinatra for reasons that were never entirely clear. 

Towards the end of her life - she passed away several years back - she began to lose interest in cinema. She seemed lukewarm on many new films. She would go to Marvel films, but she didn’t like them. She declared the Transformers series unintelligible because “the good guys and the bad guys look the same,” making it impossible to follow the plot. Though she never got into streaming (“I would have gotten better at computers if I’d have known I was going to live this long, “ she would say), she increasingly enjoyed watching movies on her television and, from the last of the movie rental places, on DVD. 

Moving to the small screen was harder for us to enjoy together, though. What should we watch? With a near-infinite number of possibilities, my preferred movie was unlikely to be hers. With only a handful of new movies at the theatre for us to see, we could always agree on a new release. But all of Blockbuster meant thousands of options. Graciously, she valued time with her grandchildren over her movie taste, giving me wide latitude (not as wide as my younger brother’s, however - I’m reasonably sure, though I cannot prove, that Grandma allowed him to watch movies at her house that would never have been allowed in our home).

 Today, I often think of her as I wade through my streaming services in search of something to watch with my wife. The ability to find the perfect thing to enjoy at home was one of the great promises of the internet. By having an enormous back catalog of past works, we would not depend on the current crop of movies, music, books, etc., and could all enjoy our unique taste. Though he’s already been rewarded handsomely for his observation, credit must be given to Jeff Bezos for recognizing that while best-sellers sell the most books, most books sold are not best-sellers. Therefore, Bezos realized, a bookstore that could sell any book could outsell all other stores. Indeed, it has.

The “fat tail” (a statistical term meaning the parts of the distribution far away from the average, made famous by Nassim Taleb) of consumer preferences now defines entertainment. Rather than enjoying the most popular thing on average, we can enjoy the thing made just for us. Just as those in the mood for a human, not a robot, brawl can stream Raging Bull and avoid Transformers n+1, so too can those who prefer folk to hip hop find a trove of resources on music streaming services by searching for Joni Mitchell and clicking “more like this.” Television, of course, works the same way, even allowing you to set up multiple accounts within your household so that you do not have to be bothered with suggestions aimed at your spouse. More concerningly, news has followed the same trajectory, with seemingly every American listening to a different podcast and preferring a specific opinion writer. 

From a business perspective, this has worked marvelously (pun intended). We consume more media than ever and by an enormous margin. Grandma may have enjoyed a double feature on a lazy Saturday afternoon now and again (approximately four hours of screen time), but the average American now spends over 7 hours in front of a screen each day, at least half of which is on our phones. We’ve given up calling music, movies, television, and writing “art” and now describe it as “content,” much the way we call edible materials we shovel into our mouths “food” and not “delicacies.” 

The strange part is that we’re unhappy. The “decline” of film has created a cottage industry online, with those like the Critical Drinker amassing millions of subscribers, subscribers who often prefer watching take-downs of Hollywood films more than the films themselves. Similar stories are told by music critics, such as Ted Goia and Rick Beato, neither of whom find much to like in today’s Top 40. Critiquing “the media” has become an overused trope of both the left and right to justify their displeasure with unfavorable coverage. But none of this explains why we care. From my home in Boston, Amazon Prime costs about $15 a month, coming with millions of movies for free. The ones I want are never free, generally costing $4 to rent, for a total of $19, which is almost exactly the price of a single movie ticket in Boston. And, with this $19, you can see nearly every movie ever made. What does it matter if, as the Critical Drinker says, Hollywood is utterly out of ideas and recycling plots with increasingly diverse characters? It’s hard to believe we’ve all exhausted the back catalog of media we like. It’s even harder to believe that we’re unhappy with the media available to us when we spend nearly half our waking hours consuming it.

But what if the problem isn’t the media but how we consume it? Like so many people, I found rediscovering old interests to be one of the pandemic’s exceptionally few bright spots. For me, one of those interests was the blues. I had liked listening to the blues before COVID-19 sentenced me to house arrest, but now I had a lot of time and slowly began to work my way from Robert Johnson to BB King. I’ve savored this journey, and it has shaped my musical interests ever since. This pleasure would not have been possible without streaming services, and I appreciate them for it.

Nonetheless, this experience alerted me to their fundamental limitation: they cannot provide someone to share the experience with. I continue to enjoy the blues online, but I have no one with whom to discuss them offline. My gracious wife attends blues concerts with me and genuinely seems to enjoy them, but she discovered sea shanties during COVID-19, and that’s where her passions lie.

As they often do, economists have a word for this sort of thing, “network effects.” Goods with network effects become more valuable as more people buy them. Perhaps not surprisingly, this term came out of literature developed in the latter half of the 20th century to understand markets for new technologies. Jeffrey Rohlf, an MIT graduate student at the time, published the first paper to use this term in 1974 based on his study of telephones. Rohlf’s key observation - an insight worth trillions in today’s economy - was that technologies such as the telephone are more valuable to users as more users are connected to each other. In essence, if you have the only phone in the world, you have an ugly paperweight. But, if everyone on the face of the globe has a telephone, then it is an indispensable device that you will likely need to check 96 times each day.  

For probably obvious reasons, understanding network effects is critical to understanding our current economy. Facebook, Instagram, X, and TikTok all benefit from network effects for the same reason as the telephone: the more people on X, the wider the potential reach of your message, and the more valuable it is. This has led economists to hone their understanding of the topic by creating and refining mathematical models of network effects over the half-century since Rohlf’s paper.

Despite all the work done on network effects - nearly 7,000 papers in 2023 alone - we still understand network effects too narrowly. While we recognize that platforms like Netflix have network effects, we think television series like Stranger Things do not. But ask yourself, what if Netflix came out with a show you adored - your only 10/10 rating -  but no one else ever watched it? Suppose (here, it will become evident that I’ve spent far too much time with economists) that you can trade 1 point on a 10-point scale for 10% of your friends to have seen and enjoyed the show. So, instead of a perfect 10/10 show that only you watched, you can have a 9/10 show that 10% of your friends enjoyed, or an 8/10 show that 20% of your friends enjoyed, etc. What is your optimal balance?  

I cannot tell you your optimal ratio, but I am confident it’s not 10/10. I say this confidently, partly out of experience. I love the show (and book) Jonathan Strange and Mr. Norrell (for the record, I do not claim to be a television critic), but of the 2-3 people in my network who have watched it, none could remember anything about it. This has been a tremendous disappointment for me not only because I adored the show but also because I thought I finally had a socially acceptable response to the most ubiquitous question of American middle-class life, “What shows have you been watching lately?” Instead, my response would fall flat, killing the conversation like a loud burp at a Downton Abbey dinner party. Consequently I’ve given up Strange as a conversation line and head straight for the safety of the cold, dark Succession saga, which (for the moment) often leads to a conversation more fun and stimulating than the show itself.

Compare this to I Love Lucy, which, at its peak, 75% of American homes watched each week. For context, the Super Bowl is far and away the biggest television event of our time, with a current viewership of roughly 50% of American households. Now, I can’t say I’ve ever watched a whole episode of Lucy, but I can say confidently that I would make every effort if three out of four of my friends watched it each week. For this same reason, many people with no interest in football watch the Super Bowl, leading to the bizarre half-time shows that do not make even the slightest attempt to cater to core NFL demographics. 

Mercifully, sports are one of the few areas where network effects continue alive and well. This is not because media platforms haven’t tried to change our consumption habits but because of the nature of sports. While, of course,  today’s sports fans can go back and watch any game they want, doing so misses the thrill of sports. Not knowing what will happen ahead of time is an enormous part of the fun, as is the fact that the competition is constantly changing - players are growing (or aging), teams change, the weather differs, etc. All of this makes a back catalog of sports games much less valuable than a back catalog of music. Consequently, interest in sports remains astonishingly high—a rare bright spot on the media’s horizon.

Sports also benefit from having in-person events, which have continued to flourish. For those skeptical that entertainment can have network effects, ask yourself why anyone would pay to watch a football game live when they could get a much better view for free at home. (This is particularly true of football, which, like soccer, no matter where one sits, much of the game will happen in an inconvenient place). The obvious answer is that it is more fun to enjoy with someone. The crowd's energy and enthusiasm are infectious, making everyone in the crowd’s experience richer. 

Perhaps no one knows this better than Taylor Swift. When the Beatles played Shea Stadium in 1965, they required one outfit, one instrument each, and a small, rickety stage. To play the same stadium, Tswift requires an astonishing 27 semis to set up each show. It takes so long that she sets up one show before she’s completed the one before it. As others have noted, Swift seems to have a keen understanding of the importance of network effects to her fans: not only does she create a jaw-dropping live show, but she’s honed a fanbase that likes each other enough to create friendship bracelets for strangers. Being a Swiftie has huge network benefits, and Swfit’s capitalization on them has made her one of the greatest pop stars of all time.

In fairness, today’s platform masters have thought a lot about people’s sense of isolation, and their solution is online communities or the metaverse. The logic is immaculate: given all these options, there is no statistical chance your physical environment will match your unique preferences, so an online world will enable you to find your tribe(s). So, one day, Mark Zuckerberg believes, you will strap on your (Meta) headset and head into the (Meta) 3D world where you will find your group of people dressed like you, reading like you, watching movies like you. And, if you’re too unique, and those groups aren’t the same group, you can move to different virtual locations. You will discuss movies with your movie friends, clothes with your clothes friends, etc.

Virtual community is vastly better than no community, but it takes a tremendous optimist to say in 2024 that this substitution will lead us all to longer-term happiness. Rather than rehash the empirical debate over current social media use and links to depression, let me suggest three reasons to anticipate a ceiling on how happy virtual communities can make us.

The first is commitment. Sociologists make a big deal of commitment because being committed to a group is integral to deriving meaning from being in that group. A group you cannot leave (say, skin color) carries a lot more meaning than a group you can leave (like hair color). A group where no one is committed doesn’t tell us much about the people in it. By contrast, a group that’s fixed for life tells us a good deal. But, for all of our work fighting robots and adding verifications, it’s tough to have commitment in a virtual group - you can always have another account or another email, so group members cannot verify that you are being consistent. Consequently, affirmations and criticisms from the group simply mean less.

Second, you still have to live in the real world sometimes. Though I’m not “very online,” those who are constantly remind us how frustrating it is to have to relate to people different than yourself in the real world. The nice thing about watching I Love Lucy in the 1950s was that you could discuss it with your neighbor or colleague, or even better, you could watch one show to relate to them both. Today, we have no such luck - our colleagues likely watch something totally different. Or, worse, don’t get our references, can’t understand our slang, etc. This can lead to an unfortunate situation, often experienced by Gen Z, in which our physical environment always disappoints while our online world never quite satisfies. 

Finally, even if you find your community and, by good fortune, have people in the physical world around you that you like, you still have the regrettable problem of knowing exactly what your enemies are thinking all of the time. If you have a movie community that likes your movie, then you have the frustration of knowing that there is another community that hates your film - and they will talk about it, and you will see it. Bizarrely, as Redditors can attest, they will even bomb your conversation to tell you about the views you specifically said you did not want to hear. This can range from the frankly hilarious (see Tswift fans posting snake emojis on Kim Karsdashian’s Instagram) to the wildly offensive (e.g., the enormous rise in antisemitic content on X after October 7). 

The resulting loneliness is remarkably similar to C.S. Lewis’s conception of hell. In The Great Divorce, Lewis illustrates the alienation that he believed comes from being distant from God by describing a hell in which the damned can have anything they want just by asking (clicking?) but have no one to share with, living alone in palaces surrounded by all their hearts’ desires. While housing prices mean that none of us are living on palatial estates (except the boomers, of course), the sense of having entertainment Ceaser would envy, but no one to share it with feels all too real.

Now that I’ve dragged you from Taylor Swift to hell (a shorter journey for some than others), I’d like to tell you how to make it better, but before I get there, I’m afraid I have two more disappointing things to tell you: First, AI will make this worse. Up until now, media platforms were limited by their catalog. Though unlikely, you could exhaust your interests by consuming them all. I am learning there are only so many cozy British mystery movies. Going forward, GenerativeAI will eventually be able to create new things exactly like the old, however, effectively Marvel-izing past works of art to create infinite content. I do not know when Clapton.AI will make its first debut, cutting and pasting Slowhand’s famous licks into a new song transposed to a different key, but I’m confident the day will come.

Second, as consumers’ tastes are increasingly different, investing in things that appeal to the average makes less and less sense. This phenomenon is well documented in film by those like the Critical Drinker, but I find it even more visible in books. Publishers increasingly offer less advertising and editing, preferring to “buy lottery tickets” and see what works rather than investing up front in an unknown author or idea. The result is that we are not just pulled to the tails of the distribution but pushed there by the declining quality of the center. 

I fear that the result will be an even crueler hell for us millennials than the one Lewis pictured. After all, we were raised by the Baby Boomers, the last generation to enjoy mass culture and, therefore, the most prominent “brands” on the planet (witness the sale of Bruce Springsteen’s catalog). Rather than investing in new ideas, investors have a huge incentive to turn the AI crank, generating more and more content from the corpse of boomer creativity. As I watched the de-aged Harrison Ford in the latest Indiana Jones (does the number even really matter at this point?), I could feel my entire generation thinking, “I’m going to come and go, and my parents will stay 30. I’m going to have to live in their world forever.”

Unfortunately, this is not a problem that lends itself to an easy solution. We all use online platforms, and we won’t give up our favorite books and songs to return to some sort of top 40. But we mustn’t deny the loss we’re experiencing as well. Coming back to our friends in economics, understanding that “normal” goods also have network effects will help economists as they chart our future. Increased choice has a declining marginal utility, and we may now be in the negative. 

Ultimately, Grandma’s solution is the correct one. Sacrificing your preferred good so you can enjoy something with someone often creates a better result overall. In economists’ terms, this is a sort of Nash equilibrium: by all pursuing our first choice - our preferred work to be enjoyed alongside everyone else - none of us gets exactly what we want. We get the good, but not the community with which to enjoy it. But, if we choose our second best thing, something not as good but everyone by our offline friends, we can trade the value of the good for the value of the community and perhaps enjoy more value overall. Clearly, this only works to a point. Going back to the above question of movie trade-offs, almost by definition, it’s not worth watching a 0/10 movie that all your friends enjoyed. But you may find that the works you thought were merely “good” are the ones you remember best. 

Of course, we’re all haunted by the memory that, once upon a time, Hemingway was a contemporary author, the Beatles were a boy band, and Saving Private Ryan was a summer blockbuster. We can hope, as the optimists say, that in 50 years, the wheat will have been separated from the chaff, and the best of our time will be remembered fondly - Swifties seem especially well situated for this. But more realistically, as publishers and platforms tell us, the economics have changed, and so has the quality. As economists always remind us, the solution lies in being very precise on what we value. My suggestion is this: the value you get from creative works is partly the ability to discuss them with others. Consuming accordingly will make you happier, something Grandma, of all people, seems to have known all along.