Block size is a contentious issue. Programmers and businessmen proposed various sizes, without fully clarifying the economic impact. Let’s put the horse in front of the carriage by laying out the economic effects of larger vs smaller block size first, so that Bitcoiners can then choose appropriate technical solutions.
An economic good is something that is both useful and scarce. Air is useful, but generally, air is not scarce. Hence, air is not an economic good and there exists no price for it. On a submarine, air is scarce. In this situation, air becomes an economic good. It gets a definite price, which in turn determines the value of the machinery that produces the air, the salaries of the people that make the machinery (at least in capitalism) and so on. Understanding scarcity and economic goods is prerequisite for understanding the effect of transaction scarcity, or its absence.
Bitcoin price as a variable of hoarding
Currency derives its value from hoarding. Every time somebody spends currency, they value having the currency less than having the acquired good or service. When people hoard currency, they value holding the currency more than having any other good or service or alternative currency. They could exchange it, but they prefer not to, thereby manifesting the currency’s value compared to all other goods, services and alternative currencies.
All benefits of a currency manifest themselves in the question of whether it is being hoarded or not. If a token can hardly be transferred to others, people will not even accept it. If a token can be multiplied easily, people will not hoard it. If a token is perishable, people will not hoard it. But if it ticks all the boxes, people value having it and will try to hold on to it as long as they can.
Security is crucial
Transaction fees reduce the value of a coin by a fraction, but a lack of security can reduce its value completely (theft). Hence, the critical variable for Bitcoin hoarding is its security, which depends on network hash power. The more hash power, the more people trust the network and the more people are willing to hoard the coin over longer periods instead of spending it, thereby increasing its value relative to all other goods and currencies.
Cryptocurrency security requires an ongoing amount of expenditure for electricity, hardware, maintenance and network fees. These costs are currently covered by block rewards, but rewards will become irrelevant. Miners must eventually rely on transaction fees to run their operations. The amount of fees collected corresponds to the amount of security that can be provided.
Competition and cartels
Everybody is allowed to mine Bitcoins, so there will never be a mining monopoly. The original meaning of monopoly is an act of government that prevents certain people to engage in certain businesses. As long as no “world government” interferes, there is competition in Bitcoin mining.
Imagine a mining cartel, where miners agree to only process transactions with fees above X. As long as there is a single miner outside of the cartel, the effective fees will be below X. Why? Because the external miner will include as many paying transactions as possible in each block he manages to win. Ceteris paribus, this miner would be more profitable than cartel miners, because he does not leave money on the table. His mining power would therefore grow relative to the cartel miners’, reducing their profits, at which point the cartel becomes silly. In reality, the mere threat of “external” competitors prevents cartels.
Competition means that no miner excludes paying transactions if there is space for them. Doing so would only make his competitors richer. Users would know that, and since most people do not wish to pay more than others for the same good, transaction fees drop. But how far?
Unlimited block size
In case of unlimited or quasi-unlimited numbers of transactions, meaning that in almost all cases transaction space exceeds transaction demand, the transaction fee tends towards one satoshi.
Again, a miner competes with all other miners and so would hurt himself by leaving paying transactions on the table for the next miner to collect, no matter how small. Hence, every paying transaction is included immediately. As before, users would take note and, since nobody likes paying more than everyone else, everyone will start paying less and less, until the minimum price that must be paid is reached: one satoshi. Even then, no miner will exclude one-satoshi fees, since doing so would only increase the profit of the winner of the next block. However, such very small fees would severely restrict the overall size of mining operations, especially compared to the value they are protecting.
Most people understand that security costs money, one way or another. At the same time, nobody likes paying more than the market rate and most people don’t like to subsidize free-riders, at least not indefinitely. Users will start worrying how much security can be bought with one-satoshi-fees, conclude “not much” and stop hoarding.
Limited block size
By limiting the block size to Y transactions per time step, people who wish to transact within a definite period of time would need to be among the highest Y bidders. The more restricted the number of transactions, the greater the competition of users to get their transactions included, the higher the fees collected by the mining community, the greater their hash power, the greater security. Today, the fee is around 6000 satoshis, magnitudes larger than the minimum on unlimited blocks, yet users are not running away, as manifested in market price and turnover.
Obviously, there is another limit. If transactions become too costly, people may choose other currencies with more suitable (lower) security and fees. So what’s the right transaction limit?
The “right” size
Unhappy users can easily sell out and use a different coin. Bitcoin miners, on the other hand, are the ones shouldering the greater risks, because of their large, up-front, non-convertible investments. They have the most to lose, so it is up to them to set the balance between security and transaction cost to keep maximum interest in Bitcoin.
Finding the right size is trial and error, so increases from the current size will be incremental. Miners know the size is right when their profits are stable or rising and wrong when profits are shrinking. It’s a narrow path.
It should be clear why miners have refused big-block proposals and that proposals with automatically increasing transaction numbers are anathema to them. Miners stick with the Core team, because it promised to publish code with small transaction number increases limited by hard forks. Raising limits by hard forks requires broad mining consent, which is how it should be, since miners have most at stake.
Having fees kick in these days is awesome. The sooner people experience fees, the sooner the right transaction limits can be found. Users who value security will take comfort in fees. Users who think other coins provide similar or better security at lower costs should go there and see if others agree. The Bitcoin price will signal this preference to miners and they can adjust size and technology if necessary.
I suggest that Bitcoin miners compile and publish various security rankings for the largest cryptocurrencies to get a handle on the only real threat they face in terms of coin competition.