How Lily Phillips Happens

The race to the bottom in vice markets

A colleague of mine—a much better journalist than I will ever be—once made an important point to me about which stories get traction and which ones don’t. There’s a sweet spot, he said, for how shocking a story can be. Too mundane, and no one wants to click it. But if the story is too shocking—too gory, too lewd, too gross—it won’t work either. There are some things people don’t like to talk about.

This is part of why I don’t write a lot about pornography. Porn is among the more common vices: at least half the population will admit to using it in the past year, and about one in four acknowledge past-month use. That’s less than consume alcohol, but approximately on par with the share of the population that were past-month cigarette smokers, and significant more than the 15% or so of Americans who regularly use marijuana.¹ And yet, unlike these other substances, no one wants to talk about porn. It’s just a little too gross to get the attention that other vices do.

But sometimes stories breakthrough. Thus, the debate last week about OnlyFans performer Lily Phillips, who appeared in the recently released documentary I Slept With 100 Men in One Day. The movie is here, but most of the discourse was generated by a short clip from about 42 minutes. The scene shows Phillips, post-coitus, overwhelmed and close to tears as she describes the obviously unpleasant experience.

The ensuing commentary touched on exactly the themes you would expect. Is Phillips a savvy businesswoman or a helpless victim? Is it possible to consent to having sex with 100 men? Is it moral to film it? Has something gone wrong with out culture to get us here?

I will admit to finding basically none of these questions all that interesting (which is why I am writing this post after the furor has died down). Debates over whether or not Phillips is a boss babe or a damsel in distress are often transparent in their prurience. And at the cultural level, the things that get said about pornography are much less interesting than all the consumption that does not get talked about.

To me, the more interesting questions are those of economics. Phillips was not, legally speaking at least, forced to have sex with 100 men on camera. She did it as a promotional stunt (and ostensibly will next attempt the anatomically implausible feat of 1,000 men in 24 hours).² Which raises the question: how do you end up in a situation where someone having sex with 100 men and shooting a documentary about it sounds like a savvy business move? After all, Phillips is an OnlyFans star—a porn entrepreneur, part of a group that now collectively make more than the NBA’s roster. As someone running (in effect) a small business, why was this the right choice?

The answers to these questions may seem at first glance to be obvious. But I think they say something important about how markets in vices—not just porn—work. Specifically, they tell us how the individual character of addictive products can create a race-to-the-bottom dynamic, yielding the sort of social harms that we see in other such industries.

From the Consumer …

Let’s take a step back and talk about sex. Sex (I think it is uncontroversial to say) usually feels good. The pleasurable experience of sex is not the only reason why we do it, but it’s a big reason why we do it. That’s for the best, evolutionarily speaking; if we didn’t have sex, there wouldn’t be more humans.

Of course, it’s quite possible to produce the pleasurable effects of sex without a partner, never mind a partner who has complementary reproductive function.³ Even better is if you can simulate many of the experiential components of sex by, for example, watching a video of other people having sex. Pornography—like other vices⁴—can in this regard be thought of as a short-circuit. Rather than doing the set of things you need to do to get sex (meet a nice partner, prove your reproductive fitness, etc. etc.), you can end run the whole process for something approximating the same reward.

Whether or not pornography should be described as “addictive” is a matter of some dispute, hinging on both concerns with the literature and quibbles over whether an act can be described as addictive.⁵ But it is sufficient to say that pornography shares many of the features of other addictive products. Most relevantly, it seems accurate to say that pornography is reinforcing—consuming it feels good, inducing you to consume more of it—and induces tolerance—the more of it you consume, the more you need to consume in order to experience the same reward.

“More,” though, is a multidimensional thing. One could, for example, move from consuming porn weekly to daily. But one could also consume porn for longer periods of time, or in novel contexts or—again, most relevantly—in novel forms. After all, if you spend enough time watching people having sex, it may start to lose its luster. You may look for them to be having sex in more novel ways.

Which is the naive explanation for why Lily Phillips had sex with 100 men on camera. After all, why watch a man and a woman have sex when you could watch a two men and a woman? Or three men and a woman? You can figure out the proof by induction.

… To the Market …

But I actually think something slightly more subtle is going on here—and is why Phillips not only had sex with 100 men, but appeared in a documentary about it. The point of the gangbang isn’t the gangbang—it’s the advertising that she is the kind of star who will engage in a gangbang. Phillips is signalling that she is the kind of porn producer (as opposed to consumer) who is willing to engage in extreme acts to satisfy her fans.

That matters because Phillips is in a highly competitive market place. There are (according to Statista, so grain of salt), about 4 million OnlyFans “creators,” as well as some presumably large number of performers using other sites. “Creators” gross a lot of money—about $5 billion in 2023—but it’s far from evenly distributed. Rather, it’s a superstar economy. From this essay by investor Matthew Ball:

Mathematically, the average OnlyFans creator grosses roughly $1,800 annually (of which $1,450 nets to the creator). But according to data previously displayed on OnlyFans’ internal creator dashboards, creators in the top 0.1% collect 15x as much as the average creator in the top 1%, and 100x that of those in the top 10%. The multiple versus those in the rest of the distribution was not shared but is likely to be in thousands and tens of thousands. According to one 2020 research report cited by the Washington Post last year, the top 10% of accounts collected to 73% of revenues (which would have been 365,000 and $3.9 billion in payouts last year, or $11,000 each), while the the top 1% were 33% (or 36,500 collecting $1.8 billion, or $49,000 each). Under such a distribution, the top 0.1% would hold 15% (3,650, $800 million, and $220,000 each), and top 0.01% would be 6.8% (365, $361 million, $1 million). OnlyFans does not disclose its distribution, and it should be noted the two distributions above are not consistent, but a handful of creators proved to make upwards of ten million per year – and these creators are not even the highest earning ones!

In other words, OnlyFans revenue follows our old friend the power law—a small fraction of sellers reap the large majority of profits. It’s harder to find data on this, but it’s likely that consumption follows the same distribution. After all, there is … a lot of free porn out there. Even among those who use, only some will pay to use. One suspects (although it’s hard to say for sure) that those who pay to use are among the most intensive users, and their preferences therefore drive much of the market.⁶

Which is to say, Phillips is signalling that she’s willing to provide a stronger product to consumers, and specifically to the small share of consumers who do most of the consuming. Here potency is understood to encompass the extremity of the act—in a viciously competitive market, she’s trying to win by showing that she has the best stuff. There is a sort of “race to the bottom” dynamic: buyers will flock to whichever seller is willing to degrade herself the most.

The Iron Law of Liberalization

If all of this is sounding very familiar to TCF readers, it should. This is is just an application of ideas that I’ve talked about before, e.g. with the Iron Law of Liberalization. The ILL applies because less restricted markets are more competitive, and more competitive markets produce better products, where better is often defined in terms of potency. What was true of weed and gambling is also, quite obviously, true of pornography.

… To Society

And, much like with weed and gambling, this race to the bottom has significant social consequences. I’ve talked about this with marijuana and sports gambling, where we have a lot of evidence. In the world of porn, it’s harder to know things (because people don’t like answering surveys about their porn use). There is some social science evidence—if you want a review from a biased source, see here—but I’d rather use the remaining space to reason about dynamics.

One is the obvious harms of repeated compulsive use of pornography—disinterest in sex, social isolation, lack of motivation, sexual dysfunction, etc. Much as with marijuana, one of the strongest arguments for regulating the porn market is that people who spend all day watching porn are not living very good lives, objectively speaking, and it’s not great that nothing is done about this.

Another is that pornography is an imperfect substitute for sex. Because of this, producers of the latter are always in competition with producers of the former. To put it less formally, Lily Phillips is in competition not only with other performers, but with other women as well, who “sell” sex for things like attention, commitment, and love. And those women, too, are affected by the race to the bottom dynamic in their negotiations about sex.

If porn affects men’s expectations for sex, and it affects their ability to engage in sex, and if it offers a substitute for sex, then at the margins porn will probably reduce relationship quality and quantity (a theme of Joseph Gordon-Levitt’s suprisingly subversive Don Jon). Bad things happening to men and bad things happening to women means bad things happening to the relationship between men and women. That’s problematic intrinsically, but also has all sorts of unfortunate social knock-on effects—less dating, less marriage, less childrearing, less comity between the sexes.

Yes Of Course You Can Regulate Porn

I’ve written previously about how and why to regulate porn; I won’t belabor the point except to say that literally any step in that direction would be an improvement, and frankly one I suspect would be welcomed by a large number of porn users. One of the arguments for the control of vice is the benefit of its consumers, helping them to do what they often struggle to do themselves.

The actual bigger point I want to make is that the progression I outlined for porn—individual dynamics affect the market, the market affects society—is true of vices generally. The reasons we should be worried about porn are the reasons we should worry about marijuana and tobacco and alcohol and gambling and heroin. Many analyses of these products stop at the individual level, identify that people enjoy using them, and leave it at that. But individual behavior—individual vice—can have cascading effects, with profound social consequences.

1. https://www.samhsa.gov/data/sites/default/files/reports/rpt42731/2022-nsduh-nnr.pdf.

2. Jenny Valentish offers some history of previously “world-record gang-bang” attempts. As it turns out, the advertised number of men is rarely the actual number of men who perform.

3. For historical example, see Genesis 38:9-10.

4. The way drugs work, to a first approximation, is by artificially stimulating chemical reactions in your brain (and the associated feelings) at much higher levels and without doing the work necessary to stimulate them without drugs.

5. I mean, do *I* think porn is addictive? Totally. But this is a moderate, centrist, facts-based substack. We defer to the experts.

6. Insofar as “porn addiction” remains a controversial concept in the literature, and insofar as it’s hard to get truthful survey responses, we really don’t have any good estimates of the prevalence of addicted use. See this paper for some attempts.