Jeff Bezos, The Washington Post, and his plan to take over the media world (speculative)

Remember Sherlock Holmes and the dog which did not bark?  WaPo remains very much in the running to be the up-and-coming mega-web site which succeeds.  Perhaps the model is a Coasean one:

Much of the media world has been waiting with bated breath since Jeff Bezos bought the Washington Post for $250 million last year, eager to see some sign of the Amazon founder and CEO’s hand at work. The first tangible evidence appeared on Tuesday, when the newspaper announced a major national subscription partnership that will offer free digital access to readers of other newspapers in major U.S. cities.

While this may not be as dramatic as shutting down the printing presses to go web-only, or offering everyone a free Kindle with their subscription, it’s still a fairly dramatic departure from the approach taken not just by the Washington Post but by most newspapers with traditional management.

The partnership — which will see the Post provide free digital access to subscribers of newspapers like The Dallas Morning News, the Minneapolis Star Tribune and the Pittsburgh Post-Gazette — allows the Post to (theoretically at least) build a broader online readership outside of its core subscription area. As the Nieman Journalism Lab notes, the Post effectively ceded the national newspaper market to the New York Times by not launching a national edition, but the partnership could give it a way of achieving the same thing at much lower cost.

One possible model at work here is simply to buy the best content from everyone else, at cut-rate prices, relying on the willingness of outside sources to price discriminate and shed some marginal IP rights for some marginal revenue.  Before the rest of the world is fully aware of what is going on, suddenly you have one of the best news web sites.

But wait, doesn’t this article say the Post is giving free access to its content to other newspapers?  Here is where Coasean contracting, and symmetry of externalities, enters the picture.  WaPo giving free access to the Minneapolis Star-Tribune, or vice versa, end up being pretty much the same thing (over time, with renegotiations) in a world of Coasean contracting.  WaPo will end up becoming the hub and the others will be feeder spokes, with Wapo paying a fraction of the cost for the content it receives from each one.  (And I suspect there will be no easy “cross-access” of say the Minneapolis paper to the Pittsburgh paper, and so on, to limit the evolution of a rival hub.)  Furthermore, at least in the short run, the marketing work is being done by other newspapers, not by WaPo.

Over time the WaPo web site can buy bits of content from Le Monde and FAZ (translated by software programs, of course), The Guardian, The (London) Times, various local U.S. papers, London Review of Books, Boston Review, and who knows where else?  Probably only a few outlets, such as WSJ and NYT, will refuse to sell content to them at cut-rate prices.  If there is low marginal cost there will be price discrimination, so why not be the one buying on the low part of the demand curve and avoiding most of the costs?

Plus hire a few blogs while you are at it, see how that goes, and maybe over time reel in a few hundred of them.  Why not?  We’ve already seen some moves in this direction, with The Monkey Cage and Volokh Conspiracy.

How about some music streaming while we are at it?

How about calling it…”Amazon for News”?  And for other stuff too.  By the way, this hypothesis helps explain why Bezos doesn’t feel any great need to shake up the current WaPo newsroom.

In this model there is a cannibalization effect and the price and value of content end up falling.  Does that sound familiar?

Never underestimate how smart really smart people are.

For a further explication of what I take to be the Bezos business model, see my old MR post, “Luring Alex to Lunch,” still one of my favorites and a meditation on whether or not you should produce and write all of your own content.  (We don’t, and our model is sustainable.)  And thus, sometimes, I manage to lure Alex to lunch.  Here is how Alex feels about lunch.  That hasn’t changed.


'Le Monde and FAZ (translated by software programs, of course)'

Of course. So let's translate a portion of a front page article from today's FAZ using google's surprisingly good (well, by the standards of such things, that is) automatic translating service -

'For the first time since the controversial referendum in the Crimea to Moscow and Kiev have exchanged at ministerial level. Russian Defense Minister Sergei Shoigu was discussed in a telephone conversation with his Ukrainian colleague Igor Tenjuch "various aspects of the crisis in Ukraine and measures to de-escalate the situation in the Crimea." This was announced by the Ministry in Moscow on Thursday. The department heads would have agreed to continue contacts. In addition, Shoigu urged the Moscow-backed leadership of the Crimea to leave the Ukrainian Navy chief Sergei Gajduk leave the mainland. The Vice Admiral was soon released it.' (Erstmals seit dem umstritten Referendum auf der Krim haben sich Moskau und Kiew auf Ministerebene ausgetauscht. Der russische Verteidigungsminister Sergej Schoigu habe in einem Telefonat mit seinem ukrainischen Kollegen Igor Tenjuch „verschiedene Aspekte der Krise in der Ukraine und Maßnahmen zur Deeskalation der Lage auf der Krim“ besprochen. Das teilte das Ministerium in Moskau am Donnerstag mit. Die Ressortchefs hätten vereinbart, die Kontakte fortzusetzen. Zudem forderte Schoigu die moskautreue Führung der Krim auf, den ukrainischen Marinechef Sergej Gajduk aufs Festland ausreisen zu lassen. Der Vizeadmiral wurde bald darauf freigelassen.)

Leaving aside the fact that how Russian names are transcribed is different in German and English, it is fascinating just how important context can be. For a concrete example, Russian Defense Minister Sergei Shoigu was not discussed in a telephone conversation - he had one with Igor Tenjuch. The sort of error that no human would make, but which software is currently unable to recognize as incorrect - we are still a long way from AI. And that first sentence is incoherent, mainly because word level translation is pretty much doomed to failure, when words are used in a way that assumes knowledge on the part of the reader (Russia and Ukraine have just had their first ministerial level contact, that being the telephone conversation - 'austauschen' is not exchangeable to English in such a simple fashion).

And it may come as a surprise, but even the Post's writers are copy edited - by editors, as again, the best software in this area is laughable. Though there is a solution to the cost of using people to actually ensure that what is presented is held to a certain standard - simply lower the standard. Which has pretty much become the American solution to all sorts of challenges over the last generation.

Or use a computer and pretend that it is a perfectly acceptable substitute.

So have google do the first pass and have a low-paid English major - with no Russian or other foreign language skills - do the copy edit. As an intern years ago my job was to take technical documentation that a native German engineer translated into bad english, and convert bad english into good english.

Don't forget, machines*people > machines + people

Interesting point - and of course, a native speaker is generally required to make a translation worth reading.

But why even bother with the effort of trying to correct such a wretched first pass? Instead, just cut out the weakest part of the chain - which is the software.

I'm a professional translator and having anyone who isn't familiar not only with both languages but also the journalistic style conventions in both languages do PEMT (post-editing machine translation) is a recipe for disaster at worst, and barely-readable content at best, in almost everywhere it has been tried. It is not that hard or expensive to get articles translated by someone who knows what they're doing. There are applications for machine translation, but producing content for a general public that can choose to read something else with a mouseclick isn't one,

Maybe it will be Coasean, peace be upon him, but the market for mega-news site is far from established. The economics of selling content on-line will never match those of traditional print services. Thirty years ago WaPo had the resources to staff a massive news gathering operation. There's no evidence that can be duplicated on-line.

Instead the world may be a lot of small local news gathering operations and large aggregators that work like syndicates of the old days. The "brand" of the syndicate was never important and will be less important in the digital age. You pay a fee to a service that beams a news feed to your platforms.

"the world may be a lot of small local news gathering operations and large aggregators that work like syndicates of the old days...You pay a fee to a service that beams a news feed to your platforms."

That doesn't sound very different from Tyler's thesis to me. The only real difference is that the direction of flow is reversed: content flows to the aggregator rather than to the local news organizations. And why not? The reason for the previous syndication structure was that news could only be consumed at the local level. That is no longer true.

I may be wrong, but Tyler seems to be focusing his interest on the site aiming for the hipster, pseudo-intellectual crowd. The folks who need to tell people they were reading Le Monde. That will be an over served market with lots of barely profitable, maybe even money losing suppliers.

The digital version of AP will probably let anyone put their content in the stream and have some sort of mechanism to pay suppliers based on retail consumption. I'm just guessing here, but I'm thinking news content, outside of opinion and analysis, will be a commodity with no room for the reseller to add margin.

Well, I'm quite confident that very few hipsters read the paper of record for the home city of the DAX and EZB.

The New York Post is America's paper of record.

You don't seem to even know what the FAZ is, much less what DAX means or where the one the world's most influential central banks is located.

But as a Dutchman noted years ago, Americans tend to think that Paris equals Europe.

Nonethless, the Bild is probably the paper of record in Germany in all too many ways, though not of Germany's financial capital. Few of those ways actually disagreeing with basic premise that the FAZ is on board with, of course.

prior_approval: The set of things i don't know is very large, but I know one thing you don't know. That's one of us is a humorless pedant and it is not me.

The Post? Huh?

Overall, the whole business idea, at least as described by Tyler, sounds unconvincing. And confusing. e.g. Why does the WaPo get cut-rate content? Is that a part of this deal? What'd stop competitors from doing the same?

Anyways, with so much of good free content, does anyone really want to pay for paid content, except of the best sort. And that's hardly likely to be from some cut-rate writer at a small newspaper.

And who thinks music streaming from your staid newspaper is going to be the next success? When so many other dedicated, & likely better, sites are still struggling at it?

"some cut-rate writer at a small newspaper"

So a writer's salary and the circulation of his paper are what determines if the content is worthy of purchase? That must make Tom Friedman among the very best.

Excuse me, I forgot Ruth Marcus, Nick Kristof, and E.J. Dionne. The ghost of William Allen White glides soundlessly from the room.

I think the biggest obstacle to understanding what the digital content age will be is the failure to understand what newspapers were at various stages of history. The big mono-dailies are new. From Franklin forward through the WW2, Americans had lots of small papers from which to choose. Most had an explicit cultural or ideological flavor. The great consolidation of the industry turned the newspaper into a megaphone for the ruling class. That model is collapsing and we are returning to something similar to the 19th century model.

Bezos has a certain genius, but it is mostly in shifting his expenses to other people, often against their will. He's also had a talent for attracting the best supply chain minds on the planet. It is hard to see how either of those skills apply to the new business. The fact that he laughed Ezra Klein out of the room, however, suggests he is learning. Vox looks like a state of the art buggy whip factory.

+1 for your "hard to see how either of those skills apply to the new business." I don't see any synergy either.

The model will change yes, but I doubt it's going to be very close to the 19th century model. Basically, a lot of functions a newspaper used to perform have slowly become redundant or offered by other channels gratis. e.g. Craigslist, Adsense, OkCupid, Blogs, Wikipedia, Forums, YouTube, cheap cellphone cameras, Facebook etc.

Their biggest challenge is going to be inventing new functions that people are actually willing to pay for.

Aggregation in itself is an attractive function, especially with credibility on price and content established.

I used to shop at amazon for books, best buy for consumer electronics, newegg for computer components, sports authority for baseballs... Now I just go to Amazon, lured by the fact that they are likely to have everything I want at a good (often the best) price, and free two-day shipping with my Prime membership.

Currently I surf several media sites. But, lately I have been busy with work and family so only go to a handful. If I could go to just one, and be reasonably assured I'd get 95% of the content I was interested in, all included with my Amazon Prime membership? Done.

Small-town newspapers now have a web presence that may or may not be economically viable but it does serve the same function it always has and more. Residents of Kearney, Nebraska or Ladysmith, Wisconsin, and their kids that have moved away, want to know about the local crimes, high school sports, business activity, political campaigns, etc. that have a bearing on their lives. The big media outlets can't and won't cover this stuff. Regional newspapers have made a big mistake in decimating their local newsrooms in favor of cheaper product provided by news services. People can get that on TV.

not sure why you like ronald coase so much, specially given your harsh criticism of the economics field and the difficulty of directly applying his work. My favorite economists are the developers of the options pricing models because their work gives a definitive answer and has many direct applications.

Like Google and Facebook, Amazon is at the forefront of the 21st century digital infrastructure. Best of all he got rid of Erza Klein.

After reading the wikipedia page I still don't the significance of the coase theorem. Something to do with apples and property lines. What is the likelihood you're going to know the marginal benefit of your own apple consumption, your neighbor, and society. Seems like it hard to aply this in the real world.

ummm, I am quite unsurprised by this comment.

Seems like there's already a lot of competition in the media biz. If Bezos wants to differentiate himself from pre-digested, sanitized, CNN, or reality denying FOX, he could try reporting actual facts.

Your market is already served by Democracy Now!

I'd say Tyler serves as a pretty useful indicator of bubble-itis. Tesla, drones, autonomous cars, and now even the Washington post.

When he gets around to biotech we are done.

This is predictable as a network model. And has been done.
Ever read the HuffPost? An aggregation site?

The more interesting change has been the videoizing of the NYT.

Boo to that. I go to a newspaper site to read, not watch videos.

I am sure if I said this to my 15-yo niece, she'd have no idea why I made that distinction. Fogie I am.

Heh. Good 2004 post.

The Washington Post, otoh, has some problems that Bezos probably can't fix. This is really more like Carlos Slim buying the NYT -- it's basically a positional good.

My theorem is that Tyler overstates the important of this upcoming battle for king of the wonk website universe by a factor of about 1,000. Something about Bryan Caplan & bubbles would be appropriate here.

"My theorem is that Tyler overstates the important of this upcoming battle for king of the wonk website universe by a factor of about 1,000."

It would be difficult not to.

1) The WaPo is a liberal rag that is as interesting as any mainstream outlet, i.e., ZERO content of interest.
2) You then hook up with LeMonde, a French socialist rag and other leftist foreign papers,
3) You hook up with US newspapers that get their non-local news from The NY Times, Reuters and the AP - all left wing organizations.

So, if I am anything other than a prog/socialist/fascist Obama voting reader, I am supposed to get excited about this?


'and other leftist foreign papers'

Yet another MR commenter with apparently no idea of what the FAZ is.

Even by American standards, the FAZ is not leftist. Unless one thinks the media outlet of German banking and industry can be called leftist - not that there aren't MR commenters quite capable of just that, of course.

"Never underestimate how smart really smart people are."

Oy. That sounds like famous last words. Besides the fact that even really smart people are subject to Hayekian knowledge constraints, they're also prone to hubris as a result of all those smarts, as well as the danger that their smarts will simply allow them to engage in more effective self-deception. The ability to self-deceive in some areas of life might be quite valuable, but running a commercial empire, I'd wager, is not one of those areas.

"The ability to self-deceive in some areas of life might be quite valuable, but running a commercial empire, I’d wager, is not one of those areas."

Right. Ego and hubris impede success. That's what held back Bill Gates, Larry Ellison and Steve Jobs from making some really big money.

Sometimes they clearly do. Do you recall the names Jeff Skilling, Dennis Kozlowski, Dick Fuld (how could you forget this one, even if you wanted to?) and Alan Schwartz? Did overconfidence play no role in these folks' failures as business leaders?

Hey, some do, some don't.

My point was simply that all generalizations are wrong.

This is most easily visible in the art world. If an artist produces it, it must be art. Mark Rothko, Jackson Pollock, Jasper Johns. In the small and incestuous world of the intelligencia, everything they come up with must be smart because it's the product of smart people. A generation later their ideas and very names have passed into obscurity.

I was happy (why? I dunno) to hear of Einstein's failed theses driving to work this morning. At least he had the good sense to self-edit and abandon things he could not prove.

The market for online edgy/opinionated news is saturated.

We have too many news sites out there, but we don't have good enough ways to get the best news.

So yeah, a backwards AP is what really makes sense: Less ways of getting news, but far more customizable news, and of average higher quality. Crowdsourcing aggregation, is already quite popular: HackerNews, Reddit and such are popular, so why not throw Amazon's computational power and predictive algorithms at the problem?

When all is said and done, people will still want news, and therefore, the journalist will never die. But the entire system that makes sure that journalist makes a living is bound to change drastically, and Bezos sure has the money to make it happen.

Very good post.

In the same way that Amazon is an online hub for many small retailers, and the go-to online shopping site, perhaps Bezos can do the same thing with news.

A news hub where small papers have no choice but to make their content available to WaPo and the go-to place for consumers to find the news.

Pretty brilliant.

MR drinking game: You have to take a drink every time they say "Coasean."

There is nothing here to suggest anything about content sharing with the partner distributors; or cut price content buying and aggregating from other organisations. The deal as it stands is to gain audience for WaPo, and it is a clever and cheap way of doing that. Maybe Bezos has challenged the WaPo to become profitable; and they have realised that print rates are not replicated in digital advertising, so to maintain revenues sufficient to cover a broadly unchanged cost base of journalists and editors, they need to have a bigger audience to show to advertisers.

I would wait before speculating about what Mr Bezos might be planning on the content production, buying, sharing or aggregating side.

I thought we weren't supposed to overestimate how smart really smart people really are.

A good strategy for getting involved in an undefined opportunity like this is just to try stuff out, analyze the results and adjust quickly. This is pretty much what happened with Amazon. If their is demand-pull you can fail forward and allow your customers to design your product. If the demand is not there that is a different story.

I can see media outlets evolving toward rich guys' vanity projects, like pro sports teams. Or maybe we're already there. So absent somebody with loads of excess cash to pay j-school majors, journalism seems headed back to where it started: an "office" job for literate folks from blue collar backgrounds. Not that that's a bad thing.

‘Le Monde and FAZ (translated by software programs, of course)’

So, this model you suggested Tyler, it exists already - and with high quality human translation. We're called Worldcrunch, and our translations of top articles from the foreign press (Le Monde, Kommersant, Caixin,...) can already be found on,, Business Insider, etc etc. - check it out. We partner with leading foreign-language newspapers to hand-pick their most interesting journalism and present readers a world of fresh perspectives, all translated into English for the first time.

As prior_approval points out, context is key, which is why our translations are done by multilingual journalists who understand the nuance of the stories they translate, and bring that over into English.

Our business model is around advertising, subscription, and syndication. I welcome you all to take a look at what we're up to, and shoot me a line if you're interested to connect - or @garrettgoodman on twitter.

Comments for this post are closed