Glossary / Markets & investing
Stock-to-flow
Also known as S2F.
- Definition
- Stock-to-flow divides an asset's existing supply by the amount produced each year, and PlanB's 2019 model used that single number to forecast bitcoin's price.
The ratio itself is uncontroversial: about 20 million coins exist and roughly 164,000 new ones are issued a year, giving bitcoin a stock-to-flow near 120, higher than gold's. The forecast built on top of it is not. The model predicted a six figure bitcoin by the end of 2021, the price peaked near 69,000 dollars that November and fell to about 16,000 a year later.
How it works
Status: the stock-to-flow ratio is a sound description of supply. The price model published on top of it missed its own targets by roughly 88 percent within twelve months and should not be used to forecast anything.
Stock is what already exists, flow is annual new production, and the quotient answers a simple question: how many years of production does the current stockpile represent? Gold scores high because the above-ground stock is enormous relative to what miners pull out each year. The World Gold Council puts the above-ground total around 216,000 tonnes against annual mine production near 3,600 tonnes, which is a ratio close to 60.
Bitcoin's version is unusual because the flow is written into the software rather than set by geology or price. Before the April 20, 2024 halving, roughly 328,500 coins were issued a year and the ratio sat near gold's. At block 840,000 the subsidy dropped from 6.25 to 3.125 coins, annual issuance fell to about 164,250, and the ratio doubled in the space of one block. It doubles again at the next halving, and again after that, and eventually the flow rounds to zero.
That step-function behaviour is the tell. Because the schedule is public back to 2009, any model mapping the ratio to price predicts a staircase whose steps are known years in advance, which is a strange property for a forecast of a traded market to have.
Where you see it
The stock-to-flow price model is quoted most often in cycle-top arguments and in social media charts with a rainbow-coloured price band.
Its published record is the reason to be careful. PlanB's floor model set a minimum of 98,000 dollars for November 2021 and 135,000 for December of that year. Bitcoin traded near 16,000 dollars in November 2022, roughly 88 percent below that December floor, and did not reach 100,000 until December 4, 2024, three years late. The statistical criticism arrived earlier and independently: fitting one trending series against another produces impressive correlation coefficients whether or not any relationship exists, and the model has no demand side at all.
What survives is the underlying observation. New supply is now tiny relative to the stock, so a shift in demand meets a supply curve that cannot respond, which is a genuine structural difference between bitcoin and anything a producer can make more of when the price rises. That is a reason to understand issuance. It is not a price target.
Stock-to-flow vs the halving
The halving is an event in the protocol; stock-to-flow is a ratio that the halving happens to change.
Confusing them produces the recurring expectation that the halving day itself should move the price. Nothing about block 840,000 was new information: the schedule has been auditable since the genesis block in January 2009, and every trader had the date approximately right months in advance. The halving alters issuance, and issuance alters the ratio. Whether markets price a known, pre-announced supply change in advance, on the day, or not at all is a separate question that the ratio cannot answer.