Why Does Bitcoin Have Small Blocks?

We thought this debate would get settled once and for all in 2017, when Bitcoin Cash forked off to pursue big block scaling, and Bitcoin activated SegWit in preparation for the Lightning Network. But it only took a week of rising fees to bring back this ghost from the past: Bitcoin.org domain owner Cobra has posted a poll which inquired about potentially increasing Bitcoin’s block size until the year 2025.

Unsurprisingly, more than half of the 5515 voters on Twitter were not in favor of the proposal. And only 19.7% though it was a good idea, as the others only wanted to see the results without expressing a clear opinion on the issue.

It’s likely that the issue of increasing the block size will come up during every cycle when the supply for block space pushes fees through the roof and temporarily makes transactions in the next few blocks rather expensive (or at least not suited for small transfers).

But Bitcoin wasn’t created for the purpose of moving money at a reduced cost, and it’s likely that Visa, Mastercard and PayPal can compete in this regard in the blink of the eye – the network effects are on their side anyway, so it wouldn’t hurt their business if they charged a lot less for the service.

Yet why is talking about increasing the Bitcoin block size so taboo? Do bitcoiners want their network to be unusable by poor people? Is this part of a conspiracy with the miners who collect the fees, so they make more money while pricing out billions of people? Did lizard people infiltrate Bitcoin protocol development to prevent this Heaven-sent financial invention from benefiting the billions who need it? And am I in favor of this exclusionary thesis?

The Bitcoin Big Block Utopia

It sounds simple, right? If the Bitcoin blockchain doesn’t scale, then we increase the block size to fit more transactions. We either make the blocks big enough to be able to handle just as many transactions as Visa, or else allow the free market to determine the block size with a dynamic supply-demand mechanism.

Why use second layers, sidechains, and all of those complicated protocols that require years of development and can only be used by people who own a computer science degree when we can make Bitcoin scale on chain? After all, Satoshi Nakamoto did increase the block size a few times before setting the 1 MB limit. We should do the same, right? And it’s very simple: we change the 1 with a 2, then the 2 with a 4, and adapt according to the existing demand on the free market.

This way, the transactions will always be affordable, the blocks will almost never get filled, and supposedly the increased amount of transactions that we think will happen will subsidize the miners enough to keep the operations profitable.

A Bitcoin block size increase been proposed since 2014 with Mike Hearn’s Bitcoin XT, it’s been promoted in 2015 by Gavin Andresen’s BIP 101, and it also received variations with 2016’s Bitcoin Classic and Bitcoin Unlimited clients. They ended up deploying and supporting Bitcoin Cash, and on August 1st 2021 it will have been 4 years since the fork got launched. So since it worked provably well, why won’t Bitcoin also implement it?

Also, Ethereum has voted in favor of increasing the gas limit to reduce the fees – the direct equivalent to Bitcoin’s blocksize increase. Why don’t we also do it to catch up?

Why Big Blocks Don’t Work

In the process of finding a good answer for this question, we must first ask ourselves: what is Bitcoin and why was it created? Well, it’s a decentralized global network which enables pemissionless peer to peer payments. Come again? It’s a decentralized…

Oh right, decentralization. And there are multiple layers to it: the ability to run a validating node by using a regular home computer, the proper incentivization of miners (which also concerns security), and the avoidance of consensus issues that may split the network.

The first one is the most obvious: if the blocks are too large in size, then the blockchain grows exponentially. This means that all node operators will have to store these large blocks, so in time they will have to either prune their storage (drop older blocks from their registration to fit the blockchain to their storage size) or else buy more storage devices to successfully store the entire history of transactions from the Genesis block to date.

If we are reckless about the block size and allow it to get too big, then we will end up with more centralization. Fewer people will afford to run nodes from home, the validation will become more industrialized, and there will be a lot less privacy and individual sovereignty as users must trust third parties with validation (as opposed to willingly choosing to do so).

This industrilization of validation had already happened with the younger Ethereum network, which scaled so poorly that it relies on Geth, Infura, and Amazon Web Servers. Admittedly, not even the Ethereum Foundation is running an archival node.

And the Bitcoin Cash camp will claim that big blocks are not centralizing, as their blockchain is still small after years of enabling big blocks – well, this is only because people know that their network is not Bitcoin and there is very little demand for a lesser secure fork. If their blocks ever filled, then their transactions would have been just as expensive as Bitcoin’s.

Now let’s move on to miner’s incentives: it’s important to guarantee that in the next 7 years, Bitcoin mining can become sustainable only through fees. This is the part for which big blockers don’t really have a plan – they can either spam the blocks until they fill and move their coins around, or else increase the block rewards while messing with the supply schedule.

By the year 2028, the mining reward will be 1.5625 bitcoins per block (currently, it’s 6.25). And if the price doesn’t magically quadruple (something for which we have no guarantee), then the miners need to collect more fees. Does this mean that the fees need to be high? Not necessarily – but it’s about the blocks getting filled with transactions all the time, so the subsidy covers the electricity costs involved in securing the Bitcoin network.

Bitcoin mining must stay both secure and decentralized. And if we don’t pay for this security, then we should get prepared to do a lot of altruistic home mining (aka unprofitable mining) in the coming years, just to compensate for what we weren’t willing to pay for in fees. Due to the poor scalability of the blockchain, the plan is to migrate towards second layers and sidechains, so that most of the base layer operations would be channel openings/closings or peg-ins/peg-outs.

It’s not that we don’t want Bitcoin to scale, but the issue is that increasing the block size is a bad idea. It’s like stacking 56k modems in an attempt to get high speed internet, instead of researching to get a broadband connection.

And now let’s get to the consensus issue. Let’s say that the developers do create a well-balanced and secure client which effectively increases the block size. However, the users would have to accept that it’s the right version of Bitcoin and migrate on this new chain. But what if they don’t?

Hard forks are not out of the question by default – in the event of a catastrophic phenomenon which destroys the usability or security of Bitcoin as we know it, it’s fairly easy to copy the UTXO set to preserve ownership of the coins and hard fork to create a new chain that fixes the fatal issue. But scaling with bigger blocks is a controversial issue that won’t get enough support. What such an attempt would accomplish is to once again split the community.

There is also the economic argument: if you adjust a money system according to the needs you have right now, while thinking that you’ll figure it out along the way as the technology improves, how different are you from a fiat central banker who generates more debt that future generations will pay? A system like Bitcoin should get governed by principles and algorithmic rules that all participants agree to respect, not human politics.

Enjoyed this article? Make a donation!

If you want to support my work and get added to the Hall of Patrons, here are the ways in which you can contribute:

Donate BTC on the base layer, pay potentially-high fees: 3KsLGzuABH4iknGMrB5gr3dU4HGm4P9mDn

Donate bitcoins via Lightning, pay the smallest fee ever: https://tippin.me/@TheVladCostea

Donate dirty fiat via Patreon (recurring payment) or PayPal (one-time payment).

If you send a donation, please write me an e-mail so I can place your name in the Hall of Patrons. Thank you for reading!