The Pleb Economist #3: On the political economy of blocksize1
@jakoyoh629 recently posted a link to Jameson Lopp's Treatise on Block Space Economics (#872466), a presentation which he made in the 2024 Lugano's Plan ₿ Forum.2
I found the presentation interesting enough to warrant some further thoughts and reflection. I hope to use this space to jump start some discussion on block size, which we're all aware is a sensitive issue, as Lopp made so clear in his presentation:
Today I'm gonna talk about block space and block size. Those of you who have been around since the 2017 era scaling wars may have some trauma and PTSD, but I would urge you to not allow your emotions to override your logic with the ideas that I'm gonna be talking about today.-- Jameson Lopp
Let me first summarize Lopp's points and then add some further thoughts of my own.
Lopp's Main Points
-
The old blocksize question was whether Bitcoin should be optimized for low cost of full system validation (small blocks) or for low cost of transacting on chain (large blocks).
- My comment: The small blockers won this war, and I think that was the correct choice. Because the block reward decreases over time, the only way to sustain mining in the long-run is to have a non-zero fee rate. We can still debate as to the exact tradeoff, but the basic economic argument that transactions need have a cost won out, with the added benefit that node-running is still economically feasible for individual users.
-
The new blocksize question is how to maximize the number of people who are able to use Bitcoin without trusted third parties, in a permissionless manner, and without disrupting the game theory and economic forces that make the system sound.Blocksize impinges on this question in two dimensions.
-
First, if blocksize is too large, individual users will not be able to economically self-validate the entire chain. They will be forced to trust large enterprises to validate and enforce Bitcoin's rules.
-
On the other hand, if blocksize is too small, individual users will not be able to economically transact on the chain. They will once again be forced to use trusted intermediaries to transact on their behalf. The prevalence of custodial wallets and exchanges highlights this problem.
Thus, Lopp calls this a "Goldilocks Problem" because it isn't about small blocks vs. large blocks: it's about getting the size "just right". -
-
All of the historic blocksize proposals suffer from major flaws.
-
BIP 100 proposed for dynamic block size via miner vote. Problem is that it gives miners too much power. Also, its scaling limits are too high and could result in 50MB blocks in just 2 years.
-
BIP 101 proposed doubling block size every 2 years for 20 years. This was overly simplistic and results in a max block size of 8 GB which is absurdly large.
-
BIP 102 proposed a one-time doubling, which would have been safe but not a long-term fix.
-
BIP 103 used laws of technological scaling to propose a 17.7% per year growth. Using technological scaling laws is a good idea, but this proposal doesn't account for demand. (e.g. If demand grew at a lower rate, you'd end up with non-full blocks which undermines thermodynamic security.3)
-
BIP 104 proposed automatic adjustment of blocksize to keep blocks 75% full. The problem here is that full blocks are needed for positive fee rates, which are needed for thermodynamic security.
-
-
Ultimately, Lopp does not have an answer, but he doesn't think we should settle for one time compromises or solutions that kick the can down the road. Instead, he believes that the solution must be an automated blockspace algorithm (not based on miner voting) that takes into account both demand and obeys technological scaling laws, inclusive of growth in computing speeds, data bandwidth, and storage costs. Lopp believes this would require a hard fork, but he believes that to be safe if implemented over a 2-year timeframe.
-
Lopp ends with a word of warning. Many companies moving into Bitcoin, including most of TradFi, do not care about self sovereignty. They are happy to let people use custodial solutions. Thus, those who care about Bitcoin self sovereignty need to be willing to fight for the future of Bitcoin, and not cede control of the protocol to the above groups.
My thoughts
My thoughts center around the issue is that this is more than just about economics, it is about political economy.
To summarize, it seems that Bitcoin must optimize over the following four dimensions:
-
The cost of validation. In other words, the cost of running a node. This includes not just storage space, but also bandwidth. Other things equal, lower cost of validation is better.
-
The network throughput. In other words, the blocksize or the cost to make an on-chain transaction. Other things equal, more throughput is better.
-
Thermodynamic security. This refers to the hash rate. More hash rate isn't necessarily better, but the hashrate has to be high enough (and decentralized enough) that dishonest mining is infeasible or irrational.
-
Required trust in third parties. This can refer to trusting others to validate the blockchain, but it can also refer to trusting others to take custody of your coins. Other things equal, lower trust requirement is better.
Bitcoin's technical specifications, of which the blocksize is one of them, places it on some point in this four dimensional space.
Now, suppose you aren't satisfied with Bitcoin's parameters. The traditional economic approach to this problem is to let the market provide the solution. If enough people aren't happy with Bitcoin, an alternative can be created with different parameters, and each individual can self-select into the protocol that meets their requirements.
This, of course, was the promise that sold many people on altcoins. But the history of altcoins has cast doubt on their ability to make actual improvements on bitcoin, for two reasons. First, the incentive to run scams is much greater than the incentive to make real technical improvements. Second, if your goal is to make a technical improvement instead of running a scam, the size of Bitcoin's pre-existing network makes it more rational to develop on Bitcoin layer 2 rather than creating a brand new protocol.
Thus, one can't simply rely on market competition to drive technical improvements in Bitcoin, and especially not improvements that respect the ethos of self-sovereignty. Improvements to Bitcoin must come from within the Bitcoin community.
And here is where problems emerge. Everyone in the Bitcoin community has different preferences and priorities over the four dimensions. There isn't a single point in the possibility space that would satisfy everyone. If there was, then optimizing along the four dimensions is a mere technical problem to be solved. As it stands, it's not just a technical problem, it's a political one.
As I discussed here (#849906), politics is provably hard. When you have to make one decision for a whole group but everyone in the group has different preferences, there isn't a single method for decision-making that will prove satisfactory, even ex-ante. Any decision making process is likely to result in some rough-and-tumble, some hurt feelings, and some accusations of unfairness.
It is in this spirit of provably hard politics that I respectfully disagree with Lopp's assertion that we can find a dynamic algorithm that solves the problem satisfactorily. Although such an algorithm may prove useful, and I absolutely agree that technological scaling laws need to be considered, I just don't think an algorithm is going to solve the political fights that must inevitably be had.
In other words, I think political fighting is going to be a necessary feature of Bitcoin technical improvements, and that we're going to see this cycle play out again and again. Rather than think of this as something out of the ordinary, we should see the politics as entirely predictable if not desirable; and maybe if we can predict it coming, we'll be prepared to handle it in a more productive manner going forward.
Thoughts?
Previous Pleb Economists
-
Pleb Economist #2: Politics is Provably Hard #849906
-
Pleb Economist #1: The Art of Economic Communication #839278
Footnotes
-
The Pleb Economist is a (not quite) weekly column where I share my thoughts as a mainstream economist who is slightly jaded by mainstream economics. I also have a somewhat libertarian and rebellious bent, as well as a deep conviction in the utility of Bitcoin. ↩
-
I'd forward @jakoyoh629 some sats, but forwarding results in both of us earning only CCs, so instead I decided to zap his original post some more. ↩
-
Thermodynamic security refers to the incentives that drive a robust hash rate and thus prevents a 51% attack. If miner revenue is too weak, this would reduce thermodynamic security as not enough hash power would be incentivized to mine. ↩