I like to write, and occasionally I idly consider the possibility that I should do it for a living. Then I remind myself: writing is a structurally terrible job. Whatever dreams you have of living off of big checks from Random House and Vanity Fair: crush them now. The economics of the media industry are historically not great for investors, but their saving grace is the enormous supply of people who really, really want to be Creative for a living: if you’re competing with broke amateurs and wealthy dilettantes for a mediocre prize, expect to suffer.

The old antecedent to bloggers waiting for a new trending topic to bubble up.

I should know: I’ve tried.

Way back before I was a fintech guy, I was an ad guy instead. I worked for a couple years as a copywriter for a Madison Avenue agency. Granted, this ad agency paid for a Madison Avenue lease for the same reason a certain variety of brokerage pays to locate on Wall Street even though the American finance industry is located in Midtown, Menlo Park, and Greenwich; bidding up for stale cachet is a way to select for not-quite-the-most-discriminating clientele.

Copywriting, in the early 2010s, meant SEO copywriting: churning out endless boilerplate that fit the right criteria (keyword in the title, keyword repeated in the body, keyword-rich links sprinkled throughout, 450 words or more). Once you get in your groove, you can do this quickly — I’ll never forget what a wonderful ATM eHow.com was, back when they paid $15 per article and were utterly indifferent to quality. In the right state of flow, I could crank out five or six articles in an hour. Good for me, not so good for the collective quality of information on the Internet.

What I eventually realized about eHow, eZine, and all the other content farms was that what really made money in media was not content itself but distribution: if you have a comparative advantage at directing people’s attention to your stuff, getting more stuff is just how you scale. Content is to media companies as iron is to steel mills.

While writing itself doesn’t scale (going from five articles an hour to six took practice; I don’t think I ever hit seven), distribution really does scale. When the SEO team at eHow figures out a way to get 1% more traffic to every article on their site, they benefit in proportion to how many articles they have; the writers don’t. And when their SEO team has done this a few times, they can afford to hire even savvier SEO mavens and repeat the process. They can also step up their ad sales game, and once again apply a force multiplier to the site. Content times pageviews per page times ad rates equals revenue, and there’s no way to multiply your content other than to multiply the number of hours people spend producing it.

eHow isn’t paying sufficiently energetic freelancers a surprisingly competitive hourly amounts any more. Because distribution is disruptable. What happened was something like this: as eHow and its competitors grew, search results got increasingly polluted with low-value pages. The search engines don’t like that; they’ve fenced in the commons of the Internet, so they have an incentive to maximize the value of that asset. Search engines tweaked their algorithms, found proxies for high quality (like “Is this site disproportionately linked by .edu domains?” or “Do journalists on trusted news sites tend to mention the brand name of this website a lot?”) and proxies for poor quality (“does this site have thousands of nearly identical articles on almost exactly the same subject? Are fifty of them by the same author?”), flipped a switch, and the content-farming business basically disappeared.

Not the first time that’s happened, and not the last, either. We saw a similar cycle a few years later in social media, where Upworthy went up in smoke and ViralNova imploded because they relied on gaming social media traffic in the same way. Supply chains are resistant to adjacent layers having high margins, in the same way that nature abhors a vacuum: search and social will do everything in their power to a) prevent the suppliers of content from producing stuff that’s bad for the quality of user experience, and b) capture the economic surplus from the good stuff. Whoever has the game theoretic upper-hand wins, and that’s usually whoever has the best economies of scale, and that’s whoever is in the more consolidated part of the supply chain, so it’s the fate of every marginal player in media to get ever-more-marginalized.

Now, the media landscape is a more efficient market. There’s mass-scale stuff with great name recognition, which does tend to have good economics for the content producers (I’ll get to that in a moment, but really it’s a footnote: the gini coefficient of popularity is even more extreme than the gini coefficient of wealth overall, so I’m talking about something like the top 0.1%, here). But there’s a lot more mediocrity. As you move down the long tail, the market is pretty good at matching supply and demand.

And that’s where you see the economics of content in their steady-state, which is okay from a business perspective but absolutely dire from a creative one. What a search- and social-driven media world wants is mass customization: instead of 3 channels, 300,000, each of which is perfectly appealing to about a thousand people. Thus: BuzzFeed listicles (the examples are boring, but the parody is spectacular), trade magazines, glurge, and worse. These are the kinds of media that there’s always demand for: funny articles that are just for you, useful news that useful to you, sad stories that play to your exact feelings, Kindle Erotica that plays to some of your other feelings — all for a hyper-specific value of you.

What content producers like Buzzfeed recognize is that there’s a tradeoff: 99.9th percentile quality targeting an audience of a million gets less readership than 50th percentile quality targeting an audience of a thousand, so long as you can reach your target audience sufficiently well. And the labor pool at the 50th percentile is bigger, so it’s more important to find someone willing to pound the keyboard than to find someone who is a virtuoso at it. Just like in music: there are a lot more people playing weddings and bar mitzvahs than selling out stadiums.

Imagine how frustrating it is to get a job as a writer — literally, getting paid to do what you love! — and then spending all day trudging over a tiny tiny subset of a field you don’t necessarily care about. I don’t have to imagine: I once had to write five articles in a row about plaid cargo shorts. The only thing that made it fun was the challenge of a deadline.

Some kinds of media jobs have exit strategies: an astonishing number of tech bloggers turn into venture capitalists. The skills here happen to overlap: a good startup is also a good story, so someone with a nose for one has a nose for both. (This is also why venture capitalists are the best tech bloggers. That door revolves 360 degrees.) So, tech bloggers can become venture capitalists. But what do politics and entertainment bloggers become? Mostly alcoholics.

For most people who write for a living, there isn’t a good source of upside: you either hit it big, or you just keep doing more of what you’ve been doing until you give up. But that’s an optimistic view: since content creation is non-scalable, it’s a disproportionate share of the costs of any company that does it and has a competitive advantage at something else. And as everyone in business knows, you never want to be a cost center, even in a good business: sure, the chef at Google got enough stock options to get rich, but if they’d had to do a down round, you can bet his refresher grant wouldn’t have been as generous as the one they gave to the engineers.

If media companies grow when they win at distribution, and then shrink fast when the distribution model changes, the typical writer is always being buffeted about by economic forces they don’t control. Essentially, they benefit from a small-scale version of Baumol’s Cost Disease. When the economics of media are good, everyone’s wages get bid up even though they’re not getting more productive. When the economics go back to normal, cost disease is cured and you’re back to making about what you’d make in fast food or bartending, just with longer hours, no benefits, and pay on net-30 terms — if you’re lucky. All else being equal, you’re more likely to join Buzzfeed, or Vice, or Upworthy, or eHow close to the peak, so a typical new hire’s job is even more precarious than average.

But! What about the name-brand writers? Don’t we periodically hear stories about enviable comp packages for the best in the business?

We do, no doubt. It should be very suspicious to you that any examples you can name are people with name recognition; most writers don’t have that. I always wondered about those hefty media pay packages, though: don’t these people love what they do? How do they negotiate such great deals?

I got curious about this a while ago, so the next time I met a genuine New York Media Person, I asked him. The answer is: rich people buying trophy assets. What happens is that someone takes his company public and retires before 40, and then decides he’ll do some good for the world. And where better to start than to buy a magazine all his rich friends read? But just buying it isn’t prestigious enough, and anyway, there’s more money to burn: so let’s turn it around by stuffing the masthead with stars who have great name recognition.

After a while, they realize they’re wasting money and the magazine hasn’t turned around, sort of like how I once realized I’d spent the past six months subscribed to HelloFax even though I signed up to send just one fax. It’s just scaled by net worth. So they sell their trophy asset, probably to a newer younger rich person, and the cycle begins anew.

Ironically, these pay packages are a crude kind of economic justice. There are a vanishingly small number of journalists whose work is read fifty years after the fact. But there are a few. Distribution strikes again, though! You can’t sell subscriptions to a magazine that consists of three Tom Wolfe articles a year. The magazine, like many other good media distribution businesses, worked this out through bundling: sell a packages of some good stuff, some okay stuff, and lots of ads, and you’ll make more money than if you just sold the good stuff and the ads. But if it’s the good stories that make the magazine worth reading, there’s a sense in which good writers carried the whole endeavor. And yet, everybody had to get paid. It’s likely that the media industry redistributed some wealth from the best writers to cub reporters, even though the best were paid well. In that sense, the bidding war for marquee talent (and I use that term advisedly) is about making the economically irrational decision to pay them what they’re really worth, instead of the reasonable decision to pay them what you can get away with.

Being the next Tom Wolfe or Christopher Hitchens would pay well, yes. But a lot of people read Wolfe and Hitchens, some percentage of them said “That’s who I want to be when I grow up,” and some percentage of them have parents rich enough to subsidize New York or DC rent for however long it takes for them to make it. If you really are good — like, top hundred writers in your generation good — you have a shot. But there are a lot of people just like you competing to be the next-whoever, and that particular job has between one and zero vacancies. And anyway, it’s not as if Christopher Hitchens wanted to be the next anything; he just wanted to be himself. Wes Yang isn’t trying to be the next anybody either, and he’s doing a great job being Wes Yang.

The old magazine model is mostly dead. As far as I can tell, the reason the magazine bundle persists is that there’s still a cohort of print subscribers who like it, and media ad sales are structured around it. But if you were designing a magazine de novo, you wouldn’t want to bid against economically irrational players for name-brand talent. You’d want to shoot for quantity, with just enough quality to attract a few prestige advertisers and a lot of affordable, prestige-seeking junior journalists. You might, for example, hire a star reporter from Politico, but still use cat videos as your bread and butter.

There are other models besides magazines, sure. Instead of a rich patron, why not Patreon? If you look at the top media people on Patreon, something interesting pops out: Jordan Peterson is a writer and lecturer; the Chapo Traphouse gang make podcasts, but Peterson and Team Chapo don’t have an audience so much as they have a community. They’re not in the media business at all; they’re pastors. This is not a bad business, it’s just a different business. There are about as many clergy working in the US as journalists, and they’re a bit better-paid. If you add in secular pseudo-clergy, you get a bigger but harder-to-estimate number. And the secular flocks really matter. Who has a bigger day-to-day influence on more people? Tim Ferriss or Rick Warren? I’d bet on the former. While there are some professional clerics who are also good writers, there are a lot more people who ostensibly work as writers (or podcasters, or vloggers), but who are really in the business of building a community and helping people solve their personal problems.

Interestingly, this category of writing also avoids the pathologies I mentioned before: the one distribution network that technology can’t disrupt is a real-word community that happened to agglomerate online. If you want the kind of job where you get to spend time writing, it’s possible, but never forget what business you’re really in.

Business Writing and the Writing Business

Being a writer is a bad job, but is being a good writer helpful for other jobs? I’ll give it a definite maybe. On the one hand, I thought this piece was inspiring when I first read it, but in retrospect it’s a tad cargo-culty. The reason investors think being a good writer correlates with being a good investor is that they correlate very well indeed when your sample size is Warren Buffett.

There are a couple problems here: First, Warren Buffett co-wrote many of his annual letters with Carol Loomis, and she is a great journalist. You can find Buffett’s early partnership letters online, and they’re decent but nothing special compared to the Berkshire stuff. The ideas were mostly his, but the flair was mostly hers. Second, there are plenty of other great investors who don’t write exceptional letters. If you run down a list of the top-performing investors of the last decade, you’ll see a few other wordsmiths but mostly people who are good at investing. And third, Warren Buffett’s letters are not really business writing. They’re journalism, with a one-company beat and one deadline a year.

Actual business writing, the kind you use at work, is a totally different genre. Internal memos are fundamentally about bullet points, not paragraphs; most business questions are in multiple-choice format, not essay. (If you find yourself writing an essay, it’s often a bad sign — you only need a structured narrative to explain why something went wrong.) To the extent that you can exercise creativity it’s in deciding what the question is and what the answers are, but even that’s limited — as companies grow, they become more structured and less risk-seeking, and part of that means prizing a repeatable process over a questionable improvement.

And in one crucial sense, business writing and other kinds of writing have inverted incentives: if you’re writing to entertain, you want to fill time enjoyably; if you’re writing to inform someone, you want to get to the point, ASAP. You’re writing Cliff’s Notes, not the classics.

The overlap between the genres isn’t zero, of course, it just isn’t very big. Some of the skills overlap quite well: if you can write good subject lines for cold emails, you can write good titles for blog posts; if you know which bullet has to go first, you have an eye for a good first sentence. But if your internal email takes the reader on a fascinating journey across intellectual disciplines, makes them laugh, gives them a quote they’ll repeat to sound smart to their friends — it’s a complete failure at actually getting them to quickly take action, so it misses the point.

Writing about business, but not as part of your business, is different. It’s a hybrid between research and marketing. Writing is a form of research because long essays are the single best way to organize your thoughts. You can write an essay to figure out a strategy, but it’s best to memorize your conclusion and throw away the verbiage, unless you happen to work at an organization that does its strategic thinking in the form of long memoranda.

Writing is a form of marketing because it’s a way to show off what you know and how you think, and that’s a good way to meet good people. And it’s a hybrid between research and marketing because the best ideas are the prospect of alternating periods of introspection and conversation. Think something up, write a thousand words; pitch your idea to a friend, get it torn to shreds; think up a rebuttal, two thousand words; repeat until you’ve either successfully trashed your idea or successfully made your case.

That’s basically what I’m doing here: casting a net of fun ideas, using it to catch some interesting people who drift by. A cynical ploy, but it’s working so far.