FreeConomics Part III - or why is everything crap?

Last week I attended the Wealth of Nations conference - my talk was in the new media infrastructure panel, which I will write up a bit later, but I also attended the very interesting panel on Diginomics. Its taken a while to write up as (i) it needed a bit of thought and research and (ii) I have been a bit tied up on "real" work this week.

The core pieces of the session were output from (i) Thierry Rayna on the inefficiency of Web 2.0 economics (a theme that ties to our issues around Freeconomics), and (ii) a paper on the changing economics of media - especially regional print papers - by Gary Graham. There was a very interesting discussion afterwards. I've split this post into 3 sections - firstly, a discussion around Thierry's talk and a sbsequent chat with him, secondly Gary's talk about Media, third about the other issues discussed:

Web 2.0 Inefficiency

The core of the argument is that because of the non monetary nature of Web 2.0 services (ie its all given away free ) then inefficient "adverse selection" choices are resulting - of technologies, services, etc. Secondly, that because of the very low values of most of the content it is very difficult to adopt a simple market scheme for Web 2.0 output.

There are 3 main causes of adverse selection (from my notes - other can chip in if they have a different view of what was said):

- Skewed Incentives: Most contributors do not benefit from their contributions, as they become public goods very rapidly. Thus the rational decision is to "free ride", hence the huge skew between creators and consumers. Open Source software exhibits similar properties, but has avoided the consequences largely because of other benefits to the contributors.

- Search Costs & Cost of Production reduction - the "external" element of search costs (using Google) has dropped, but the "internal element" - winnowing the increased amount of chaff generated by Web 2.0's easier access to more stuff - has risen. Thus there has been declining overall net benefit in search transaction costs. Due to the lower production costs, too much crap is being produced, increasing the noise" and thus leading to a net decrease in social capital. In other words this has led to a "reverse tragedy of the commons" - instead of overgrazing we are over-polluting the commons.

- Unaccounted Externalities - the cost of this crap is by and large not being borne by the producers, it is being borne by the rest of the value chain - as there is so much crap, its increasingly hard to find the good stuff, and in effect it is being forced out the market. This is leading to Akerlof's Law in spades - Akerlof showed that in a market where quality is had to discern, crap drives out quality every time, which reduces overall value of the market. (And if you pay for the crap as per Knol, rather than just make it free, it is exacerbated)

In summary, too many of the wrong people are producing too much of the wrong stuff because it is all being subsidised to the nines.

However, it is extremely hard to actually price and thus monetise Web 2.0 content because of its low unit value and digital format. There is a "perfect storm" emerging from this format as it is:

- nearly instantly public (owing to the copy facility and near impossibility of maintaining copy control)
- infinitely durable (perfect copy every time, keep it so long as there is storage available)
- experience goods - you cannot find out the value until you try it.

Very few people will buy digital goods before sampling them, so they have to be sample-able to sell. In practice the digital format means that if I buy X, I lose nothing if I let you copy it, so there is no incentive for me to keep you from copying it, and because it is so easy, lots of copies inevitably emerge - so price collapses. Sadly, micropayments are not an answer as the rest of the transaction cost is still quite high, so pay per use is not viable yet for much of this content. (In fact, the only way much of this stuff can work is if you make it free to user because of this)

The Future of the Media

Traditionally, a newspaper did 2 things - it informed you, and made money from that. Increasingly however, the "inform" and "making money" bits have been separated. What the media has found is that people are willing to subsidise them (aka advertise) if they entertain, rather than inform. Hence free papers etc.

The economic discussion above demonstrates the theory of what is happening to the media. In the print world, infinite reams of blogcrap and sleb filled free papers are drowning out the quality content. In music (and soon video) the nature of the digital good reduces price of any social wealth producing (aka desirable) content to near zero. Owing to the resulting pricing and externality inequities, people adversely select and the result is that poor quality drives out good quality over time.

Gary's view is that the Internet Generation "News"paper will have to do what the craposphere does not - produce quality content, and charge reasonable prices for it.

So what will occur - the trends imply that two main trends will emerge in the value chain:

- Firstly, there will be customised printing close to source, done to order to try and reduce the distribution costs of paper products. This can push the economics of non subsidised (typically higher quality) products back within view of the free stuff.

- Secondly, there will probably be more customer involvement - from selection through to interaction - to create a more unique experience and compensate the customer in that way. The issue with this will be how to (i) get serendipity and (ii) prevent people from selecting such narrow/ easy to digest content that they become effectively uninformed.The final thought was to find ways of using "Internet Ju-jitsu" - use the nature of the web to create new incentives for people

The Q&A Session

There was a good debate on:

Micropayments - it was claimed that the transaction cost for each Paypal transaction is $2.35. That sounded high to me, and our analysis shows transaction costs of iTunes for eg is more than an order of magnitude lower than that. But the point stands - today, its hard to make money on something costing much less than an iTune, so that bundling methods (eg subscription) are the only way it will work. But with so much Freeconomic subsidy money in the system tight now ($150bn + if Chris Andersen has got it right) , the adverse selection bugbear rears its head.

Patents - good discussion on this, I loved Ludmila'a observation that a patent is merely a right to fight in court, and if you don't have money to fight the well funded incumbent its largely valueless (read our own Paul Lancefield's views on this - and why some trolls are your friends - over here).

Scalability of the "T Shirt & Concert" model for music, Thierry pointing out that ticket prices for concerts had to get very high to make up for lost recorded music sales owing to lower volume. And of course the higher the ticket price, the lower the volume of attendees.....

Should Incumbents just weather this - it is irrational economics, it will collapse, and the best is to stay out and let it run its course, surely. But what happens if they are wrong, or it is too big for one industry to avoid (or, as in media, it destroys far more value than it replaces it with).

An interesting start of a discussion around moving to services that are too complex to be offered by the "give it for free" startups (it was observed that many of these do not offer services, just subroutines as a main product), and these can be charged for. Another option was to package services up as cultural goods, and charge for increasing access (the Freemium model used by Flickr for example). A third discussion was to charge for the stuff that is not available for free today - privacy, quality, etc.

(In)Conclusion

No answers, lots of fascinating questions etc. Harsh though the piece on Diginomics is to the Web 2.0 faithful, I think the point about Web 2.0 potentially having a "reverse tragedy of the commons effect" and for the resulting FreeConomic based adverse selection to drive out quality for crap is a serious charge that needs addressing. Similarly, if there is a limit to micropayments that will impact Web 2.0 long term viability - and produce a huge benefit for anyone who can solve it!

Update - interesting corollary article by Kevin Kelly, arguing that people actually do want to pay:

I recently came across a UK survey, sponsored by British Music Rights (that means they represent musicians and music publishers) which suggest fans want to pay. According to this study what the respondents appear to want is an unlimited download service free of DRM that could be legally accessed for a monthly fee-- a way to pay that doesn't yet exist.

They are happy to pay if it is easy, fair, and beneficial.