Bayesian Investor Blog

I was tempted to call this post a response to Scott Alexander’s Problems with Paywalls, but after carefully rereading that I can’t find any clear disagreements with him. I mainly want to push a slightly more pro-paywall tone.

Paywalls have well-known benefits, such as incentivizing good content, and aligning the interests of writers with the interests of readers, at least compared to ad-based business models.

I’m annoyed by how the news media storyteller industry uses paywalls, but I mostly hope that industry goes away for reasons which would remain valid even if they had nice paywalls.

I’ve sometimes been tempted to generalize my dislike of paywalls to something beyond “this industry does them poorly”, but I keep remembering that paywalls seem okay for books.

What’s different about books? The most important factor seems to be that I can get pretty good ratings for books before buying, and I can often tell from reviews whether the book is appropriate for my desires and expertise. Whereas most paywalled stories are small enough that it’s not worth my time to check reviews to evaluate whether to pay for them.

A secondary consideration that distinguishes books from many paywalls is that libraries provide an alternative. I hesitate to link to a paywalled Matt Yglesias post. I don’t want political discussions to be limited to people who believe his posts are worth paying for, or segregated into those who pay for his posts versus those who don’t.

Whereas I’d be comfortable linking to his book, because people can borrow the book if they can’t afford to buy it, or if they object to giving money to someone they expect to dislike.

How to fix paywalls

My main suggestion for bloggers and storytellers is to charge money for early access, then remove the paywall after a moderate period of time.

Stock market quotes are my favorite example (this was more true 25 years ago than it is today). Stock quotes were readily available in 20-minute delay form. Real-time quotes required a somewhat high-end subscription.

The movie industry comes somewhat close to doing this, charging theater prices for early access, and cable TV prices or Netflix DVD prices for those who wait. This would match my recommended approach if movie copyrights expired after a year or two, or if copyright owners were willing to give away old movies.

For a journalist trying to make money via something like SubStack, I suggest implementing this approach via a paywall that lasts 2 days for short bits of information (e.g. the typical Tyler Cowen post), and 2 weeks for long essays (e.g. a Scott Alexander post). I recommend selling both subscriptions ($5 to $25 per year?) and individual posts ($0.10 to $0.50 each?). The latter wouldn’t raise much money directly, but I’d be more willing to link to posts if my readers could quickly see them with less commitment than is involved in subscribing. All of this likely requires using some fairly standard platform such as SubStack rather than having a hundred different sites with different payment mechanisms.

Who pays under this model? Often it will be people who are showing off their knowledge of current events at social events. That’s a lot like people who see movies early because they provide conversation topics. This accounts for a significant fraction of interest in stories, even if storytellers are reluctant to acknowledge it.

Some of the payment will be for convenient email updates.

In most cases, costs will be borne at least as much by those who can afford to pay as is the case with ads.

For people who really want to be informed quickly without paying for information, Wikipedia will often work. There’s a significant fraction of new information for which I find Wikipedia (or some other site that’s oriented toward data, not stories) more convenient than storytellers: hurricanes, COVID levels, forest fires, etc.

Why isn’t a pay for early access model being adopted more widely?

I expect that leading storytellers such as the New York Times promote and/or cling to inflated perceptions about the lasting intellectual value of their stories.

Less successful storytellers have less hope of surviving via any type of paywall. I’d be experimenting with better paywalls if I were in charge of a second-rate storyteller company. But I suspect anyone with enough sense to do that has fled those companies.

The Factual has a nice alternative: they’re a news aggregator that provides relatively fair, non-click-baity summaries of stories from other sites. They provide enough free info about each story that I used the site for a couple of months before feeling much temptation to subscribe. They charge for conveniences such as daily emails and the ability to click on links to the underlying articles. It’s not hard to bypass their paywall by Googling for the underlying articles, but it’s inconvenient enough that most people who have a job won’t bother.

I predict this model won’t earn enough money to tempt existing news companies to imitate them.

Another model that I hope gets tried somewhere is for readers to pay a small amount for each story. After reading the story, they have the option to alter the payment so that it becomes a refund to prior readers of the story. Any publisher that succeeds under this model would produce better stories than do current publishers. But adoption of this model will be slow due to how much thought it demands from readers. I ought to cite the post from which I got this idea, but I haven’t managed to find it.

Can peer-reviewed journals survive with one of these models? Not in their current form.

But they probably shouldn’t and can’t survive in their current form anyway. A more likely, and more desirable result is that they get replaced by much smaller blog-like organizations that either operate on a business model that’s similar to The Factual, or on a nonprofit model that’s more like Our World in Data or LessWrong.