Skip to content

Scalability

This is one of my favorite critiques,

Bitcoin only supports up to 7-transactions per-second. Do the math, that's not enough.

In truth, Bitcoin doesn't process any transactions per-second, it processes blocks (of transactions) every 10-minutes, on average. Blocks have a few thousand transactions, hence "up to 7 per-second", I guess. And while this is true for base layer transactions (final international settlement), 2nd-layer solutions like the lightning network can scale upwards to infinity.

But a Bitcoin critic is not interested in 2nd-layer payment rails or any actual engineering solutions to their criticism. The pattern of this criticism is quite dumb and once you see it it'll be hard to unsee. Take whatever perceived scaling limit that is less than the population of humans, and declare it as a fatal flaw in Bitcoin.

ah hah, this is less than 8-billion people, Bitcoin can't be money.

With this same logic, fiat couldn't scale, credit cards couldn't scale, not even coins could scale -- there are only 2-billion US quarters, "ah hah, this is less than 8-billion, quarters can't be money".

Rarely will you find a nuanced scaling criticism -- there are however, some interesting critiques and thought exercises to be found here.

For example, how will lightning channels scale globally?

Scalability of Lightning Channels

Opening and closing a lightning channel requires base layer transactions. Can all humans on earth open a lightning channel? Can we have 8-billion lightning channels?

Truth is, under a Bitcoin standard not every human will need to make base layer transactions. And they don't need to open lightning channels. The world does not need a lightning channel for every human. Only entities that we should call sovereigns will need to do base layer transactions and manage their own lightning channels. A lightning channel is effectively a liquidity provider for payments.

We see the same pattern with fiat, except with fiat each sovereign has the power of seigniorage; they can print as much money (liquidity) as they desire. Sovereign fiat currencies compete with each other, all trying their best to be the strongest fiat -- print too much and you hyperinflate, but print too little and you'll have no seigniorage powers. Under a Bitcoin standard, no one has the power of seigniorage. The fixed supply and the end of seigniorage is the entire value proposition of Bitcoin.

Under a Bitcoin standard, sovereigns will compete, and they will do so fairly under the rules of a fixed global money supply. If a sovereign lacks liquidity in their lightning channels, that is, they can't pay you, you simply go to a sovereign that can. This is more or less how the gold standard functioned before it was corrupted by fractional reserve paper money.

Sovereigns and Scale

Importantly, Bitcoin supports vastly more sovereigns than does fiat. In today's global fiat economy, there are approximately ~200 different central banks, each representing their own financial sovereignty. Base layer transactions under fiat happen infrequently between these sovereigns, typically through treasury bonds, where they buy and sell each others debt. In practice there is only ~20 or so sovereign currencies with real economic power, and only 7-8 that actually matter. Most other fiat currencies are typically backed by those top 7-8 fiat currencies (USD, EUR, JPY, etc).

Under a Bitcoin standard, money is decoupled from state, and the power of seigniorage is no more. Anyone with sufficient Bitcoin to do base layer transactions (who could open and close lightning channels) would be a sovereign. This would include nation states, corporations, family dynasties, and even individuals (see the Sovereign Individual). If this seems bizarre, consider that this is far more aligned with the history of money and civilization than our current ~50 year experiment in global fiat.

But how many sovereigns could exist under a Bitcoin standard? Answering this is pure speculation, but if we're only concerned with the mathematics of global scale, let's imagine a distant future where the very idea of fiat money is a footnote in history, an oddity of less advanced people who pretended governments could issue money. Let us imagine a Bitcoin standard. Let us imagine, the year 2140.

In the year 2140

If Bitcoin has survived this long, it will have proven itself true. The world will be a vastly different place, with wonders far beyond our present imagination. However, there are some things we can know with relative certainty. For example, the last Satoshi will have been mined. There will be zero inflation. What we today call "mining" will instead be a global settlement layer, where sovereigns compete in a zero sum game for transaction fees.

And with some degree of confidence, we can predict that fiat money will be a relic of the past. Money will be decoupled from nation states, which, if they're still around, would be far weaker and less influencial than today.

And, the easiest prediction of all: Bitcoin will simply be known as "money".

Let us now imagine a global population of 12-billion humans, living in relative peace, though always with the typical human dramas that define humanity.

USD as Unit of Account

I've included USD in the below examples; which is a bit funny since USD is obviously not a great measure, however, these are wild speculations meant as a thought exercise in the mathematics of scale. Forgive me.

Let's imagine that the market cap of money stands at $300-trillion, the equivalent of a $14.285-million BTCUSD. This is arguably much smaller than one would expect in the year 2140, but we'll start here to set some baseline numbers for our scaling calculation. We'll have the following values in sats,

1 sat = 0.1428 USD

1 USD = 7 sats

Meanwhile, the transaction fees are averaging 1,000 sats, which corresponds to $142 per-transaction. Each block has at least 3,000 transactions, resulting in a block reward of $428-thousand.

1,000 sats ($142) is the fee to go from 2nd-layer savings to sovereign individual. In other words, this is the fee to become a sovereign. As a sovereign individual, one is free to hire their own team, investing in their own infrastructure. Sovereigns either profit, sustain, or go bankrupt (assets absorbed into other sovereigns). Competition amongst sovereign individuals is zero sum on the base layer, but positive sum in local environments as economies grow and new sovereign individuals are created.

Now let's assume the economy is much larger by the year 2140. Let's assume a $100-million BTCUSD. In other words,

1 sat = 1 USD

This means 1,000 sats per-transaction fee is now $1,000. The cost of becoming a sovereign individual is now quite expensive, and will grow continually more expensive as the economy grows. This benefits families and multigenerational wealth strategies while not precluding individuals from joining the ranks of financial sovereignty. Anyone could save and become a sovereign, adding liquidity to a lightning channel in order to run their own business.

Each block would be worth about $3-million, with network security at an all time high, despite zero block subsidies (it's no longer mining, just international settlement).

But is all this really scalable?

There would only be approximately 160-million transactions per-year.

This equates to tens-of-millions of financial sovereigns, out of billions of people. Most individuals will not be sovereign. Many families and dynasties will be sovereign. All individuals will be associated with one or more sovereigns, either by family or by work.

Let's imagine there's only 50-million sovereigns at any given point in time. This would easily be scalable with base layer transactions (using today's technology), and there'd be an average population of only 240-persons per-sovereign (and with a very high per-capita income). However, under a Pareto distribution, we could expect 80% of the population to be working for the top 20% of sovereigns, that is, 9.6-billion people in 10-million sovereigns, averaging ~960 people per-sovereign, with a long tail of 40-million sovereigns; mostly family, individual, and small business sovereigns. The largest sovereigns might have thousands of people, maybe even some with tens-of-thousands. The smallest, are just individuals.

And due to the nature of the game (the economic game), it's unlikely a given sovereign would grow too large (before disintegrating into smaller sovereigns). People would be free to work with different sovereigns; you're working for Bitcoin after all. Sovereigns need workers and people, they're always hiring (otherwise why waste the money on even being a sovereign).

In any reasonable analysis, this is far more scalable (and far more fair) than the ~200 or so sovereign nation states acting as totalitarian regimes and slave labor coordinators for billions of people. Bitcoin is an alternative to fiat debt slavery. And a natural distribution of sovereigns avoids the entrenched oligarchies we see in modern fiat.

A Bitcoin standard, even with the technology as is, is already scalable enough to replace fiat. This would also put an end to our globalist oligarchy, much like modernity was the end of nepotistic monarchies.

A future Bitcoin standard could easily scale to tens-of-millions of sovereign individuals and dynasties, where today's nation states would either be relics of a strange past, or economically small players amidst tens-of-millions of other sovereigns, competing for wealth and resources like everyone else.