weThe Incentive Structure of Bitcoin

tl;dr

  1. The people using Bitcoin must realize that Bitcoin has well-defined constituencies and knowing what a healthy relationship looks like between each is a huge benefit for the things Bitcoin was originally created for.
  2. Due to the result of the Scaling War of 2017, we now have a painful precedent about how to handle contentious hard forks.
  3. We can use the distant past as well as modern day business models to clearly see the trap ready to spring.

/tl;dr

Topics covered:

  1. The proper perspective with respect to the Bitcoin network is the same as any free market enterprise.
  2. The current state of the Bitcoin network as it pertains to the free exchange of value.
  3. What the past tells us about where cryptocurrency networks are going given the business models of related technologies.

I have written other Medium articles outlining some of this material. I wanted to re-state a few things in light of the debate that I had with John Ruggieri. Some of the questions and comments that I received were very prescient. It certainly allowed me to think a bit differently.

First, it is really important to acknowledge who the major Bitcoin constituencies are. This is not a complete list and this list might change over Bitcoin’s lifecycle:

  1. Bitcoin users (“user” sounds harsh, but in today’s parlance, everyone knows what this is)
  2. Bitcoin miners (this includes mining pool operators)
  3. Bitcoin developers (this includes anyone involved in providing products/services related to the Bitcoin network)

These parties are very broad and they overlap quite a bit. I have been in each constituency at some point myself. It is very important to understand, not only how the Bitcoin network achieves consensus over value transactions, but how Bitcoin achieves consensus over changes to the core protocol. I would posit that the latter is much more important over the long term, but it is the former that is more important for casual onlookers to understand as their first pass. I am asking interested parties to take their learning to next level by understanding how Bitcoin evolves over time.

I should also say that what we have now, unfortunately, is not ideal if you are measuring the current state of the constituencies against free and open market systems. What we have is more like crony-capitalism overlaid on top of cryptocurrency that happens to have the Network Effect. I consider myself a staunch ally of the free market and I don’t believe that government regulations ever make things better in the long term. My critics will say that Bitcoin is just one crypto in a sea of cryptos and that the market will (and should) decide which one it prefers. So, who really cares what Bitcoin does?

My concern remains at the intersection of Bitcoin’s Network Effect and the capital investment of second tier networks like Bakkt or the Lightning Network. The market will eventually reject the bad ideas, but if we see a train wreck coming, why not point it out?

How it should be:

Let’s start with miners. Miners secure the network and make it possible to transact on the network. They are at the very core of what Bitcoin is trying to do; solve the double-spend problem and at the same time, minimizing counter-party risk. This a fancy way of saying Bitcoin is a peer-to-peer electronic cash system and miners are at the heart of it. As such, they receive coinbase rewards and transaction fees from Bitcoin users. As time goes on, the coinbase rewards decrease geometrically, therefore the importance of transaction fees goes up geometrically. If the miners expect to garner fees, they better deliver a good experience to their customers. The way the miners can do this is funding the build out of the Bitcoin infrastructure itself. This means funding wallet development, research and development for scaling Bitcoin and all the other things their users ask for. The miners’ return on investment of doing this is very clear. The miners get happy customers in exchange for well-earned transaction fees.

How it is now:

Significant funding from the miners did not start until late 2017, way too late to outbid the Wall Street banks via venture capitalists. Of course, I am not calling out any miners. In the early days of mining, it was just passionate individuals mining at home. The possibility of miners banding together in an informal way to fund Bitcoin development was just not there. By the time professional firms started manufacturing and mining coins en-masse, venture capital had already locked down key pieces of Bitcoin and some other currencies. Some newer currencies fully understand this deficiency and have answers for it.

Why it matters:

In the very long term, none of it does matter. The market will choose the option that serves the customer best, all things considered. There isn’t anything a government or Wall Street bank can do about that. Winston Churchhill is credited with saying that Americans always do the right thing after they try all the other things. I suspect that our path to good money will travel along the same lines. There might be opinions out there that believe that cryptocurrencies are a good idea, but a captive second tier network like the Lightning Network or Bakkt will be a good bridge from fiat money to something better like pre-2017 Bitcoin.

Bitcoin’s Current Incentive Structure:

When I say “Bitcoin”, I am referring to Bitcoin Core or the coin with the ticker symbol “BTC” as of February, 2019. The vast majority of Bitcoin network infrastructure is funded by venture capital, volunteers funding themselves or angel investors. This is very natural. In the United States, we have well-worn paths for these kinds of funding mechanisms. The problem is that Bitcoin and its infrastructure do not fit into the mold of a typical start up company.

Drawing parallels to past engineering efforts:

The best analog in recent history is the Internet itself. More specifically, the development of the Internet protocols themselves. Internet protocol, transmission control protocol, etc.. Ironically, this is the very example that some people use to defend the position that the proper way to scale Bitcoin is by using layers. TCP is to the Lightning Network as IP is to the current Bitcoin network. The trouble is if you dig at little deeper into this analogy it totally breaks down.

First, the idea is to “ossify” lower layers as new layers are built on top of Bitcoin. This means we stop building new features on lower layers in favor of rolling out new features on higher layers. If we draw from the example of IP, then I guess we need to hand wave over the ipv4 versus ipv6 issue. IP is supposedly “ossified” yet we are slowly, but surely rolling out ipv6. IP is not ossified to say the least.

Second, if you give credence to the IP/Bitcoin comparison, you must acknowledge the debates going on in the IP development community. By analyzing the traffic on usenet/mailing lists, the vast majority of conversations involve ipv6 and not ipv4. I do love the comparison of IP/Bitcoin because going from 32 bit addresses to 128 bit addresses fits with raising the block size. So, the plan is to make a clean break with the older Bitcoin mainnet and not look back, yet the best example of this is nothing but looking back to the IP layer.

So let’s do best case / worst case from Bitcoin Core’s perspective. Best case, the Bitcoin mainnet gets ossified and there is a clean break. We all happily transition to layer 2 networks. Worst case is there is no clean break and there is a massive effort in both Bitcoin and layer 2 networks.

Best Case:

The Bitcoin mainnet becomes irrelevant over time. There is no development there apart from bug fixes. Software degrades over time. Bitcoin software is no different. Bitcoin has users, but it isn’t gaining any features, why not just stop using it in favor of networks that are adding features? But, you might be asking: “but TCP isn’t really adding features, why hasn’t it gone away?” We know that IP is adding features with ipv6, so this doesn’t apply. TCP, on the other hand, does not get a whole lot of new features. Funny you should ask, TCP, in fact, is on the chopping block. Companies like Alphabet/Google have the juice to completely replace UDP/TCP with QUIC.

Worst Case:

The mainnet continues to get feature updates. If we scale on-chain, then we can’t get people off the mainnet. The second tier networks can’t compete with on-chain solutions. In other words, the network wouldn’t be rowing in the same direction. We’d have the same situation as we have on the Internet with IP. A long period of time where people are using both the new network and the old network with no clear end in sight.

Finally, I think I’ve made the case that the incentive structure of Bitcoin is completely out of whack. New forks of Bitcoin have been created to right those wrongs, but Bitcoin Core still has the Network Effect and the majority of users.

Even if we assign the purest of motives to the current method of scaling Bitcoin, we should still be extremely vigilant with respect to any changes to how users interact with Bitcoin. It is safe to assume the venture capitalists that invested in Bitcoin’s development want a return on their investment. The moment their interests are opposite to a current Bitcoin users’ interests, market forces should react and rectify the situation. But this takes time and fortunes will be won and lost during this period. As far as I see it, I can’t come up with very many scenarios where moving users to a layer 2 network without the ability to return to the original network is good for me, as an average Bitcoin user. Can you?