Matt Yglesias has a way of identifying great questions quickly. Responding to Erik’s widely discussed post on the low Kindle price yesterday, he writes:
That said, this seems arguably upside down as a business model. The marginal cost of distributing a digital copy of a book, song, TV show, or movie is $0. A company with a lot of confidence in its tablet could acquire the copyright of media, treat that as similar to the fixed R&D costs involved in developing new products, and then make its money by selling high margin tablets (rivalrous physical objects) that come with streaming “cloud” access to the firm’s entire library of content. It would then be in a position to pay writers, musicians, etc. to create new content that would be available to tablet owners.
Erik’s point was that Amazon’s Kindle and its content were complements and so for every Kindle sold, Amazon would earn more from content sales. Yglesias’s point is that the Kindle’s cost money whereas content is free to distribute. So why not give away the content and make more money on the Kindle sales? This would better match usage with costs: we want consumers to economize on the number of Kindles they buy but not on the content they use.
I’m sceptical that Kindles can ever be free precisely because Amazon will want people to economize on them and to use a Kindle price to self-select more desirable customers. That said, my own research that I conducted this year while at Microsoft (all conclusions my own, etc) has identified a reason why the durable gadget (like a Kindle or iPad) is likely to have a very low or zero price with platform owners making a return on the content. When there is a large installed base of a device, content providers will jack up prices for content on those platforms (either retail if they choose them or wholesale otherwise) because they don’t risk people throwing away their gadgets as a result. Therefore, they might be a slow unravelling of higher content prices and lower device prices. This will leave platform owners doing otherwise strange things like most favored customer clauses and revenue sharing to stem the tide. The theory — and it is just that — is driven by the fact that there is always a consumer who is just willing to buy a Kindle at the current price and will experience regret if book prices climb higher than they expected today. If marginal consumers anticipate this, attempting to earn money from devices will unravel quickly.
That said, if Amazon could commit to free content forever, that would allow them to implement a high Kindle price strategy. They’d be very vulnerable to hold-up by content providers in the future so I suspect that commitment isn’t on the cards.
Digitopoly 
I think Matt Y misses a factor in between device and content. Amazon has a huge advantage in its EC2 system. You could argue that Netflix and even (I believe) Apple can deliver for free only because they are taking advantage of Amazon’s infrastructure. Price = marginal cost only in a competitive environment, and I don’t think it is for EC2.
I also think I’ll be following this blog with interest.
If you charge something for a zero-cost product it’s all profit. All you need is a massive distribution channel.
Cheap Kindles does that.
It also stillbirths competitors. Why compete on engineering awesome consumer consumer electronics? That sounds pretty hard.
It feels like there’s a durable goods monopoly problem lurking here. Why would I buy a durable good at a substantial markup in an environment when the producer is coming out new and improved versions of the durable good at similar or even lower prices?
I’m not sure why this post ignores the fact that if you use one of Amazon’s new Kindles to access the web Amazon gets to see and record every single keystroke you send and all the results you get back and what you do/buy/watch/etc.
The fact that the above information is an advertiser’s dream has been widely discussed on the web since the announcement so it seems odd to overlook it in a discussion of Amazon’s Kindle economics.
The flaw in this is so obvious that I’ve re-read the article three times, figuring I mist have missed something.
Here’s the flaw: the people creating the content want to get paid too. The model that says the cost of each copy is $0 ignores royalties to those content creators.
I just looked at my Kindle; right now I’ve got about 475 documents on there. At a guess, the usual document pays its creator something in the neighborhood of $1 (although I’d guess the distribution is actually multimodal, with a bump around $0 — I’ve been catching up on things like Adam Smith and Mark Twain.) Still, a good bit of it is purchased content with a living author receiving royalties, and that’s often a fair bit more than $1 per document. So, let’s say at a guess my Kindle represents about $300 in author royalties over four months.
The Kindle cost $139, about four months ago.
Now consider if the price of the Kindle were set to allow “free” content under this model. If the content creators aren’t to be screwed over, the Kindle price would have to be enough to pay those effective royalties — which means either the Kindle will need to cost $1800 for a two-year effective lifetime, or the royalties paid to the content creators would have to be reduced to almost zero.
One thing that distinguishes Apple from Amazon (and enables the former to implement an ‘expensive devices/cheap (or at least homogeneously priced) content’ is Apple’s ability to sell updates of their devices on an almost annual basis. This has a design/branding aspect, but also improved features for the consumption of multi-media content. Insofar the Kindle is a device mostly used for reading text, there’s less scope for that approach. A bit like razor & razorblades. With content pricing, it’s very difficult to increase prices once customer expectations are set (as in the AppStore), so providing it really cheaply as per the Yglesias suggestion could be an expensive mistake if consumers didn’t upgrade in their devices.
The ‘content ecosystem’ issue has also been mentioned – it’s not in Amazon’s interest to syphon revenue out from their content suppliers, while it may be easier to do that with their hardware suppliers, whose output is also largely commoditised.
I really like this blog.
Which would you rather have, a one-time payment of, say, $500 max, or recurring income of $20-100 a month for perpetuity? On a marginal basis that would be ($500 – $200 = $300) additional payment in order to forgo Amazon’s 30% margin on digital sales for years ($6-30 a month), not to mention having to pay the creators.
Bezos is a businessman; Yglesias is not.
Price discrimination is surely something importante here. I think the end result could be similar to cable TV, where you pay for plans, and some premium books or plans will be directed to high willigness to pay costumers.
on the other hand, it is true that, in the long run, books must be sold at an average margin sufficient enough to pay the fixed costs involved in their production.