Today witnesses the second “halving” event in Bitcoin’s history, during which the growth rate in the supply of bitcoin will drop by half, from an annualized rate around 8% to 4%. These halvings occur roughly once every four years, and ensure the supply of bitcoin continues to grow at an ever-decreasing rate. Three quarters of all the bitcoins that will ever exist will have come into existence by today. The other quarter will be produced over the coming century or so.
For the first four years of its existence, the bitcoin supply grew at 50 coins every 10 minutes (roughly), or around 2.6 million coins per year. On 28 November 2012, after 210,000 blocks were mined, the supply growth rate dropped by half, to around a 1.3 million new coins per year. As the stock of existing coins increases and the new supply decreases, the supply growth rate declines, reaching zero after 120 years or so, when the total supply will stabilize at exactly 21 million coins, the maximum number of bitcoins that will ever be produced.
This often-overlooked feature of Bitcoin is one of the most important drivers of bitcoin’s success, and reflects a deep understanding of monetary economics and the history of money by Bitcoin’s creator. Had bitcoin generation not been capped, the high inflation rate would likely cause the value of bitcoins to languish and drop, which would reduce the rewards for miners who secure the network, and make the currency no more than a quirky internet experiment among cryptography enthusiasts.
To understand the significance of the dropping supply rate, a familiarity with monetary theory and history is important. Money is defined as a medium of exchange; in essence, it is any good which is acquired not for the sake of consuming it or owning it but for the sake of exchanging it for another good. There is nothing in principle that stipulates what should or should not be used as money. Any person choosing to purchase something with the aim of exchanging it for something else is making it de facto money, and as people vary, so do their opinions and choices on what constitutes money. Throughout human history, many things have served the function of money: gold and silver, most notably, but also government-issued paper, copper, precious stones, salt, seashells, and even alcohol and cigarettes. People’s choices are subjective, and so there is no “right” and “wrong” choice of money. There are, however, consequences to each choice.
Human beings are differentiated from other animals (and certain economists) by our superior ability to think of and provide for our future needs, i.e. by having a lower time preference. Humans do not always want to consume everything they produce immediately. We have the foresight to store value we produce for the future, so we can consume it when we are unable or unwilling to produce. The fundamental concern of a person choosing a medium of exchange is that they would like their money to hold its value over time, in other words, they would like their money to be a good store of value.
This makes perishable goods a particularly bad choice for a medium of exchange, as they are likely to perish before the owner can exchange them for something else, which explains why nobody in their right mind would use apples as their medium of exchange. But even among non-perishable goods, market prices fluctuate over time, and so it is not a trivial consideration to pick the right money to maintain its value over time. The choice is quite complicated because a person’s choice of money itself, paradoxically, sows the seeds of turning it into bad money: choosing something as money raises its market value, incentivizing its producers to make more of it, which will generally bring its price crashing down.
To understand why, we need to differentiate between a good’s market demand (demand for consuming or holding the good for its own sake) and its monetary demand (demand for a good as a medium of exchange and store of value). Any time a person chooses a good as a store of value, they are effectively increasing the demand for it beyond the regular market demand for it, which will cause its price to rise. For example: Market demand for copper in its various industrial uses is around 20 million tons per year, at a price of around $5,000 per ton, and a total market valued around $100 billion. Imagine a billionaire deciding they would like to store $10 billion of their wealth in copper. As his bankers run around trying to buy 10% of annual global copper production, they would inevitably cause the price of copper to increase. Initially, this sounds like a vindication of the billionaire’s monetary strategy: the asset he decided to buy has already appreciated before he even completed his purchase. Surely, he reasons, this appreciation will cause more people to buy more copper as a store of value, bringing the price up even more.
But even if more people join him in monetizing copper, our billionaire is in trouble. The rising price makes copper a lucrative business for workers and capital across the world. The quantity of copper under the earth is beyond our ability to even measure, let alone extract, so practically speaking, the only binding restraint on how much copper can be produced is how much labor and capital is dedicated to the job. More copper can always be made with a higher price, and the price and quantity will continue to rise until they satisfy the monetary investors’ demand, let’s assume that happens at 10 million extra tons and $10,000 per ton. But at that point, monetary demand subsides, and some holders of copper will want to offload some of their stockpiles to purchase other goods, since, after all, that was the point of buying copper.
After the monetary demand subsides, all else being equal, the copper market would go back to its original supply and demand conditions, with 20,000,000 annual tons selling for $5,000 each. But as the holders begin to sell their accumulated stocks of copper, the price will drop significantly below that. The billionaire will have lost money in this process, as he was driving the price up, he bought most of his stock for more than $5,000 a ton, but now his entire stock is valued below $5,000 a ton. The others who joined him later bought at even higher prices, and will have lost more money.
This model is applicable for all consumable commodities such as copper, zinc, nickel, brass, or oil, which are primarily consumed and destroyed, not stockpiled. Global stockpiles of these commodities at any moment in time are around the same order of magnitude as new annual production. New supply is constantly being generated to be consumed. Should a saver decide to store their wealth in one of these commodities, their wealth will only buy a fraction of global supply before bidding the price up enough to absorb all his investment, since he is competing with all the consumers of this commodity who use it productively in industry. As the revenue to the producers of the good increases, they can then invest in increasing their production, bringing the price crashing down again, and robbing the saver of his wealth. The net effect of this entire episode is the transfer of the wealth of the misguided saver to the producers of the commodity he purchased.
What I just described is the anatomy of a market bubble: increased demand causes a sharp rise in prices, which drives further demand, raising prices further, incentivizing increased production and increased supply which inevitably brings prices down punishing everyone who bought at a price higher than the usual market price. Investors in the bubble are fleeced, while producers of the asset benefit immensely. For copper and almost every other commodity in the world, this dynamic has held true for most of recorded history, consistently punishing those who choose these commodities as money by devaluing their wealth and impoverishing them in the long run, and returning the commodity to its natural role as a market good, and not a medium of exchange.
For anything to function as a good store of value, it has to beat this bubble trap: it has to appreciate when people demand it as a store of value, but its producers have to be constrained from inflating the supply significantly to bring the price down. Such an asset will reward anybody who chooses it as their store of value, increasing their wealth in the long run as it becomes the prime store of value because those who chose other commodities will either reverse course by copying the choice of their more successful peers, or they will simply lose their wealth.
There has only been one example of such a commodity throughout history: gold, which maintains its monetary role due to two unique physical characteristics that differentiate it from other commodities: Firstly, gold is so chemically stable that it is virtually impossible to destroy, and secondly, gold is impossible to synthesize from other materials (alchemists’ claims notwithstanding), and can only be extracted from its unrefined ore which is extremely rare in our planet.
The chemical stability of gold implies that virtually all of the gold ever mined by humans is still more or less owned by people around the world. Humanity has been accumulating an ever-growing hoard of gold in jewelry, coins, and bars that is never consumed and never rusts or disintegrates. The impossibility of synthesizing gold from other chemicals means that the only way to increase the supply of gold is by mining gold from the earth, an expensive, toxic, and uncertain process in which humans have been engaged for thousands of years, with ever-diminishing returns. This all means that the existing stockpile of gold held by people around the world is the product of thousands of years of gold production, and is orders of magnitude larger than new annual production. Over the past seven decades with relatively reliable statistics, this growth rate has always been around 1.5%, never exceeding 2%.
Source: US Geological Survey.
It is this consistently low rate of supply of gold that is the fundamental reason it has maintained its monetary role throughout human history, a role which it continues to hold today, as central banks continue to hold significant supplies of gold to protect their paper currencies. Official central bank reserves are at around 33,000 tons, or a sixth of total above ground gold.
Source: World Gold Council
To understand the difference between gold and any consumable commodity, imagine the effect of a large increase in store of value demand that causes the price to spike and annual production to double. For any consumable commodity, this doubling of output will dwarf any existing stockpiles, bringing the price crashing down and hurting the holders. For gold, a price spark that causes a doubling of annual production will be insignificant, increasing stockpiles by 3% rather than 1.5%. If the new increased pace of production is maintained, the stockpiles grow faster, making new increases less significant. It remains practically impossible for gold miners to mine quantities of gold large enough to depress the market significantly.
Only silver comes close to gold in this regard, with an annual supply growth rate around 5%, higher than that of gold for two reasons: Firstly, silver does corrode and can be consumed in industrial processes, which means the existing stockpiles are not as large relative to annual production as gold’s stockpiles are relative to its annual production. Secondly, silver is less rare than gold in the crust of the earth and easier to refine. This explains why the silver bubble has popped before and will pop again: as soon as significant monetary investment flows into silver, it is not as difficult for producers to increase the supply significantly and bring the price crashing down, in the process taking the savers’ wealth.
But commodities are not the only pretenders to monetary status. It is perfectly possible to create an artificially scarce asset to endow it with a monetary role. Governments around the world did this after abandoning the gold standard, as did Bitcoin’s creator, with contrasting results.
Government-issued paper currencies were at one point backed fully by gold, making them no more than receipts for real gold, and ensuring their supply cannot be inflated. After the gold standard was abandoned, paper monies have had a higher growth in their supply rate than gold, and a collapse in their value compared to gold. The total US M2 measure of the Money Supply in 1971 was around $600 billion, while today it is in excess of $12 trillion, growing at an average annual rate of 6.7%. Correspondingly, in 1971, 1 ounce of gold was worth $35, and today it is worth more than $1,300.
The “stable” and “strong” currencies of the developed countries have had growth rates similar to those of silver, but with a much higher variance, including contractions of the supply during deflationary recessions. Developing country currencies have at many times experienced supply growth rates closer to those of consumable commodities, leading to disastrous hyperinflation and the destruction of the wealth of holders.
I conducted a review of the annual supply growth rate of the broadest available measure of the most important global currencies over the period from 1984 to 2013 for a recent paper and found that they are at least double the average growth rate of gold. This helps explain why global central banks still maintain reserves in gold. While it is useful to have the major global reserve fiat currencies for settling international payments, it is useful to have gold reserves to protect from the erosion of the value of these currencies, which have all fared badly compared to gold in the post-Bretton Woods period.
Source: author’s calculations from data from St. Louis Federal Reserve Bank and World Gold Council.
The problem with government-provided money is that those in charge of it will inevitably fail to resist the temptation to inflate the supply of money. Whether it’s because of downright graft, “national emergency”, or an infestation of inflationist schools of economics, government will always find a reason and a way to print more money, expanding government power while reducing the wealth of the currency holders. This is no different from copper producers mining more copper in response to monetary demand for copper, it rewards the producers of the monetary good, but punishes those who choose to put their savings in copper.
Should a currency credibly demonstrate its supply cannot be expanded, it would immediately gain value significantly. In 2003, when the US invaded Iraq, aerial bombardment destroyed the Iraqi central bank, and with it, the capability of the Iraqi government to print new Iraqi dinars. This led to the dinar drastically appreciating overnight, as Iraqis became more confident in the currency given that no central bank could print it anymore. Money is more desirable when demonstrably scarce than when liable to being expanded.
And yet people the world over continue to be forced to use government money via coercive tax and legal tender laws. Gold is not an easily accessible option for most people given high transaction costs involved in moving it around, and the fact that the enormous central bank reserves can act as an emergency excess supply that can be used to flood the gold market to prevent the price of gold from rising during periods of increased demand, to protect the monopoly role of government money. As Alan Greenspan once explained: “central banks stand ready to lease gold in increasing quantities should the price rise.”
While it is technically possible to produce a new asset with a lower growth rate, no such asset has succeeded because its producers can not credibly commit that they will never, under any conditions, expand the supply at a fast rate. But in 2009, Satoshi Nakamoto succeeded in making bitcoin the second asset in human history with an ironclad guarantee that the supply will never increase at a high rate (after the first few formative years).
Bitcoin takes the macroeconomists, politicians, presidents, revolutionary leaders, military dictators, and TV pundits out of monetary policy altogether. Money supply growth is determined by a programmed function adopted by all members of the network. There may have been a time at the start of this currency when this inflation schedule could have been conceivably changed, but that time has well passed.
The supply growth rate was very high in the first few years, similar to that of consumable commodities. By 2013, after the first halving, it dropped to a rate similar to that of moderately inflationary paper currencies (10-15%), where it remained until this year. Starting today, it will drop to growth rates similar to those of the world’s strong reserve currencies and silver (4-6%). Around the year 2023, bitcoin’s inflation rate will drop below that of gold. From then on, it will become increasingly negligible.
Source: blockchain.info for data up to 2015. Author’s projections from 2016
Being new and only beginning to spread, bitcoin’s price has fluctuated wildly as demand fluctuates, but the impossibility of increasing the supply arbitrarily by any authority in response to price spikes explains the meteoric rise in the purchasing power of the currency. When there is a spike in demand for bitcoin, bitcoin miners cannot increase production beyond the set schedule like copper miners can, and no central bank can step in to flood the market with increasing quantities of bitcoin, as Greenspan suggested central banks do with their gold. The only way for the market to meet the growing demand is for the price to rise enough to incentivize the holders to sell some of their coins to the newcomers. This helps explain why in less than 8 years of existence, the price of a bitcoin has gone from $0.00076 in the first recorded transaction, to around $650 at the time of writing, an increase of 85,000,000%.
The real significance of bitcoin is that it has given everyone in the world the chance to save their wealth in easily-accessible sound money that cannot be inflated by any authority in the world. While initially the most promising potential for bitcoin seemed to be in offering cheap instant global payments for everyone, it is beginning to look more likely that its use as a store of value and hedge against inflation is the more important role, at least for the time being. Bitcoin is currently capable of processing only around 300,000 transactions per day. It remains to be seen whether various scaling options, such as Segwit and Lightning Network, will increase capacity significantly.
The world needs a liquid and easily-accessible sound money far more than it needs a low-fee small-payments network. Bitcoin could evolve into a global online reserve currency or base money, with further layers of intermediation and payment clearance on top of it, and in the process profoundly improve the options people worldwide have for saving value. On-chain transactions would become increasingly expensive and be used for important and large payments, while less important and smaller transactions are processed by intermediaries, with hourly or daily balances reflected on the block-chain to save on transaction costs.
While having bitcoin intermediaries processing payments may seem to fly in the face of bitcoin’s original vision, these intermediaries will not be able to engage in fractional reserve banking, which can only survive if backed by a lender of last resort with the ability to inflate the money supply when needed. No such lender can exist in Bitcoin, but that is a topic for another day.
