Bitcoin's Death Spiral

More than two years ago in my first post on Bitcoin I wrote about the difficulty of maintaining its decentralized nature. Nearly a year later I wrote Economies of Scale in Peer-to-Peer Networks, a detailed explanation of why peer-to-peer currencies could not maintain decentralization for long. In a long and fascinating post Mike Hearn, one of the original developers of the Bitcoin software, has now announced that The resolution of the Bitcoin experiment is that it has failed.

The fundamental reasons for the failure are lack of decentralization at both the organizational and technical levels. You have to read Mike's post to understand the organizational issues, which would probably have doomed Bitcoin irrespective of the technical issues. They prevented Bitcoin responding to the need to increase the block size. But the block size is a minor technical issue compared to the fact that:

the block chain is controlled by Chinese miners, just two of whom control more than 50% of the hash power. At a recent conference over 95% of hashing power was controlled by a handful of guys sitting on a single stage.

As Mike says:

Even if a new team was built to replace Bitcoin Core, the problem of mining power being concentrated behind the Great Firewall would remain. Bitcoin has no future whilst it’s controlled by fewer than 10 people. And there’s no solution in sight for this problem: nobody even has any suggestions. For a community that has always worried about the block chain being taken over by an oppressive government, it is a rich irony.

Mike's post is a must-read. But reading it doesn't explain why "nobody even has any suggestions". For that you need to read Economies of Scale in Peer-to-Peer Networks.

A summary of this post made it to Dave Farber's IP list, drawing a response from Tony Lauck, which you should read. Tony argues that Bitcoin is not a peer-to-peer system. In strict terms I would agree with him, but my argument does not depend on its being a strict P2P system. I would also point out two flaws in what Tony says here:

Decentralized control of the network depends on the rational behavior of the owners of the hashing power, but this is not concentrated for protocol reasons, rather it is an historical artifact of the evolution of the network due to ASIC supply chain issues and geography (low cost electricity and cold climates for inexpensive cooling). The highly concentrated operators of mining pools serve as representatives of the hash power, who can switch to other pools in less than a minute if they believe the operators are misbehaving.

First, I have never argued that the failure of decentralization was for "protocol reasons". It is for economic reasons, namely the inevitable economies of scale. Tony in effect agrees with me when he assigns "ASIC supply chain issues and geography (low cost electricity and cold climates for inexpensive cooling)" as the cause. Economies of Scale in Peer-to-Peer Networks addresses both low costs:

If there is even one participant whose rewards outpace their costs, Brian Arthur's analysis shows they will end up dominating the network.

and supply chain issues:

Early availability of new technology acts to reduce the costs of the larger participants, amplifying their economies of scale. This effect must be very significant in Bitcoin mining, as Butterfly Labs noticed.

Second, while in principle the "owners of the hashing power" can switch pools, it is a fact that the mining power has been controlled by a small number of pools each much larger than needed to provide stable income to miners for a long time. Mike Hearn's argument that "Bitcoin has no future whilst it’s controlled by fewer than 10 people." is sound at both the technical and organizational levels.