I’ve been interested in Bitcoin for a while now. In fact, I’ve covered it in the past in this blog some years back for those that remember these things… but an idea occurred to me today that I wanted to share with you before I leap into the workday (where, coincidentally, I’m working on guidance around Blockchain).
It relates to an article that I came across regarding the Bitcoin “mining monopoly” — specifically, the impact that monopoly both in mining equipment could have on the underlying security and resiliency of the Bitcoin ecosystem. It got me thinking about mining generally, and I think the Bitcoin ecosystem is developed and mature enough now that we can start to see some interesting emergent behaviors as a market. And I’m not sure what those behaviors portend is a good thing for where I’d like to see Bitcoin go.
Specifically, I think that the economic forces of mining favor two types of monopoly. First, monopoly related to mining hardware procurement and second, monopoly of the mining process itself. Let me explain what I mean by that, starting with looking at the process of mining itself from an economic point of view.
Mining Economic Equilibrium
The economics of mining are fairly simple to unpack — on the one hand, there is the cost to mine, including primarily:
- Space for operations
- Temperature control
And a bunch of other stuff I’m probably forgetting about here. In return for these expenses, you generate revenue – both in bitcoins mined and transaction fees. The degree to which the return you derive through mining exceeds your costs to do so is your profit. Pretty simple right? The issue comes about because I think it’s becoming more and more obvious that these things tend toward equilibrium; the consequences of that equilibrium are monopoly. I’ll explain that conclusion (the monopoly part) in a minute, but let me explain the equilibrium part first.
Meaning, miners will only participate to the extent that it is profitable, right? And there’s no shortage of willing miners, right? So more and more miners will tend to participate so long as it is profitable, causing it to be come less profitable (on average) per miner because the mining process is zero sum (meaning, if I mine a given coin, you don’t get one too – or if I solve a given proof of work, I get the transaction fee and you don’t.)
The point being, the return per miner will tend to drop until such time as it becomes unprofitable for that miner (based on the costs above) and they stop doing it. Miners that continue do so because they are able to operate at a profit in some way. What would cause that to happen? Because they can do so in a more efficient way. So efficiency is the limiting factor – the razor thin edge upon which profitability depends.
Monopoly: the natural state
So why does that mean that monopoly is mining’s natural state? Because only those suppliers that are able to supply hardware that operates most efficiently can work profitably. In the case of mining nowadays, that’s Bitmain. If someone can make hardware that’s more efficient (cost wise) than Bitmain, that will become the new standard that everyone gravitates to – because miners using the new, more efficient, platform will drive average profitability down until it becomes unprofitable for anyone not using it. It’s already to the point that you can’t profitably mine with general purpose hardware – try it (I did), you’ll find that the costs of power alone exceed profit significantly.
The same is true of mining pools. Pools can add efficiency, to a lesser degree than hardware, because they help even the odds of successful mining. Already the largest pools have a large stake of the hashrate. Take a look at the current distribution right now — AntPool (i.e. Bitmain) has about 20%. Now, I won’t draw out the obvious conclusion about the fact that the mining pool and the most efficient hardware vendor are the same company, but the point is that both hardware suppliers and mining pools will tend to converge over time. It likewise tends to happen quickly.
How’s that bad for Bitcoin? Remember the thing in Nakamoto’s paper that everybody in the blockchain bandwagon nowadays ignores about the security of the Bitcoin ecosystem? Specifically, “The system is secure as long as honest nodes collectively control more CPU power than any cooperating group of attacker nodes.” The tendency toward monopoly puts that at risk. Meaning, there is a danger in monopoly. Recall that one of the goals of Bitcoin was to prevent control from a government or other national authority. Is it antithetical to this goal of some power, by virtue of controlling >50% of the hashrate – i.e. of CPU power, could subvert the security of the model? I think it is. We’re close to being already there too.
Think about a corporate hostile takeover. Does an acquiring firm need to buy the whole company to control it? No, right? They just need a controlling interest. If a government wanted to control the Bitcoin ecosystem – say by acquiring Bitmain outright – what would the effect be? Bitmain’s a private company; would we even know it if that happened? I’m worried about this over the long haul. I don’t know that I have any advice on how to fix it, but I do suggest that we use it to inform future Blockchain systems.