Elasticity and inelasticity of bitcoin's supply and demand

[jehst]

Demand:

1) My experience as a seller on localbitcoins.com tells me that there is a baseline daily demand of bitcoin coming from people who are price-agnostic such as Joe Schmoe who is buying $2000 worth of bitcoin per day for.. something. Maybe remittances. I never ask. But I will tell you that every day he needs $2000 worth of bitcoin no matter what the price is. He doesn't care what the price is. This component of total demand, measured in USD, is inelastic.

2) The second big component of bitcoin's demand is demand from people who are speculating. If bitcoin is in a downtrend and the market is pessimistic this demand can easily dry up to almost nothing. Speculators can take their money and buy some other asset. If everyone is super optimistic about bitcoin, then this demand can skyrocket thousands and thousands of percent and has practically no ceiling. This component of the total demand is extremely elastic.

Supply:

1) There is a fixed supply schedule. Currently, roughly ~3600 coins are created per day in the form of block rewards. Miners choose whether or not to sell part of this, all of this, or none of this. But the maximum amount of new bitcoin that can be put on to the market per day from miners has a ceiling of 3600 coins. If a miner keeps his bitcoin for more than 24 hours or a week after he mines them, he becomes no different from a speculator.  So although there's a fixed supply schedule, this component of the total supply is elastic because miners choose whether or not to put mined coins on the market. They can choose to be speculators.

2) Speculators are the second big component of supply. Most bitcoin lays dormant at any given time and most bitcoin are distributed between a small number of big holders. The amount that can move onto the market on any given day from speculators is absolutely enormous. 15 million coins can be put up for sale tomorrow. (In reality, this never happens. Even during the bubble that would've made Satoshi a billionaire and Roger Ver a quarter-billionaire, we didn't see the dumps. Likely because these holders are too smart to dump all at once and suffer extreme slippage.) In any event, this component of bitcoin's supply is extremely elastic.

In the end, the supply is totally elastic and the demand is partially inelastic. We have two extreme scenarios:

1)  The price is bouncing along the price floor created by price-agnostic Joe Schmoe because sentiment is very low and miners are selling everything they mine. Speculators are heavily shorting and selling. (It's impossible to calculate what the price floor is, but there is some price floor created by inelastic demand from Joe Schmoe.)

2) When sentiment is very high, we have the supply of coins from miners drying and the  supply of coins from speculators also drying. It's hard to get coins at any reasonable price. Speculators borrow heavily to buy.  

The Halving:
The next question is what should happen during a halving? It depends on which situation we are closer to. If we are near the floor created by inelastic demand, we would expect the halving to have a big impact. If most mined bitcoin are being sold, and shorters are as short as they can get, then the supply will drastically be reduced when miners have only half as much coin to sell.

But if the price is high and sentiment is high, then the change in supply might be negligible because the supply of coins from (greedy speculating) miners would already be low. It cannot go negative.

Conclusion:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant.

[jaysabi]

Interesting write-up based on your observations, but the conclusions aren't particularly meaningful:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is. True of literally any asset. The floor of any asset is 0 and the ceiling is hypothetically infinite.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant. The halving could be a significant event but it could also be an insignificant event. This is a truism.

As far as the halving goes, it seems more people than not (based on what I've observed on these boards) expect that the halving will result in a rise in price. If this is the case, you should expect the bulk of the effect of the halving to be priced in before the actual halving occurs, not after.

[jehst]

Interesting write-up based on your observations, but the conclusions aren't particularly meaningful:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is. True of literally any asset. The floor of any asset is 0 and the ceiling is hypothetically infinite.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant. The halving could be a significant event but it could also be an insignificant event. This is a truism.

As far as the halving goes, it seems more people than not (based on what I've observed on these boards) expect that the halving will result in a rise in price. If this is the case, you should expect the bulk of the effect of the halving to be priced in before the actual halving occurs, not after.

Three comments

1. My conclusions become more useful as we approach the halving. If the price still seems to be scraping along in the lower to mid 200s or lower in Spring 2016, then I'd be excited about the halving. If the price starts rising significantly before the halving ("pre halving hype"), I'd be more likely to sell into the halving or even before it. Here in Oct. 2015, it's too early to judge.

2. Most of the people who will be involved in the next big rally don't even know what a halving is. They don't own any bitcoin right now. If there's a truly large pre-halving hype, it will attract newcomers. If the rally smaller and more incestuous (which we can measure in some ways by looking at trade volume, forum statistics, reddit subscriptions, etc) then I agree that it would be priced in.

3. The elasticity is important because when supply falls and demand is inelastic, price rises in a non-linear fashion. So if the supply of oil drops 10% because of a war, we don't just see a 10% rise in the price of oil. We sometimes see a doubling in the price of oil because of a small change in the supply. So another (hidden) conclusion I have come to is that the halving could actually have a very large effect on the price if pre-halving sentiment is very bad and we are scraping along the bottom. But I'm doubtful that this will happen because the halving has been so hyped.

The most likely outcome, IMO, is that we begin with an incestuous pre-halving hype and start rising in anticipation of the halving.

[odolvlobo]

There are some classes of users you overlooked.

Demand:

New users buying bitcoins at any price.

Supply:

If Joe Schmoe is buying bitcoins at any price for remittance, then someone is also selling an equal amount at any price. In short, there is also a class of users who generally sell them immediately at any price.

Merchants accepting bitcoins are immediately selling them for local currency.
People selling bitcoins at any price because they need the money.
Early adopters selling at any price because they can.

[Q7]

I think there are few other scenarios which I think can also add up to the demand part. For instance, right now we still have about 6-7 months before the halving occurs so what if let's say a service or bitcoin infrastructure (for example money remittance business) were to be developed within this time span that sudden causes demand to explode exponentially, it would change everything. Whether the halving kicks in or even before that happens, anything is still possible.

[Amph]

Interesting write-up based on your observations, but the conclusions aren't particularly meaningful:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is. True of literally any asset. The floor of any asset is 0 and the ceiling is hypothetically infinite.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant. The halving could be a significant event but it could also be an insignificant event. This is a truism.

As far as the halving goes, it seems more people than not (based on what I've observed on these boards) expect that the halving will result in a rise in price. If this is the case, you should expect the bulk of the effect of the halving to be priced in before the actual halving occurs, not after.

i'm not so sure, take a look at the 2012 halving, the price rosed not only before but after the halving, and on a very generous scale

it's not mathematical that the price increase will occur before or in the same date, it can happen at anytime around the date of the halving

[jaysabi]

Interesting write-up based on your observations, but the conclusions aren't particularly meaningful:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is. True of literally any asset. The floor of any asset is 0 and the ceiling is hypothetically infinite.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant. The halving could be a significant event but it could also be an insignificant event. This is a truism.

As far as the halving goes, it seems more people than not (based on what I've observed on these boards) expect that the halving will result in a rise in price. If this is the case, you should expect the bulk of the effect of the halving to be priced in before the actual halving occurs, not after.

i'm not sos ure, take a look at the 2012 halving, the price rosed not only before but after the halving, and on a very generous scale

it's not mathematical that the price increase will occur before or in the same date, it can happen at anytime around the date of the halving

Developmentally, the bitcoin community is eons past the point where it was at the last halving. I would not draw any conclusions based on how it went last time, because there is very little about bitcoin's public presence now that is similar to how it was 3 years ago.

[gentlemand]

I think sentiment is all. The halving is a technical detail. It's up to the humans to get excited about it. There may well be some self fulfilling prophecy involved. Ultimately there isn't a realistic supply constriction if demand doesn't grow.

[mirana12345]

Interesting write-up based on your observations, but the conclusions aren't particularly meaningful:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is. True of literally any asset. The floor of any asset is 0 and the ceiling is hypothetically infinite.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant. The halving could be a significant event but it could also be an insignificant event. This is a truism.

As far as the halving goes, it seems more people than not (based on what I've observed on these boards) expect that the halving will result in a rise in price. If this is the case, you should expect the bulk of the effect of the halving to be priced in before the actual halving occurs, not after.

i'm not sos ure, take a look at the 2012 halving, the price rosed not only before but after the halving, and on a very generous scale

it's not mathematical that the price increase will occur before or in the same date, it can happen at anytime around the date of the halving

You can't really count on halving to increase the price at that date. I think that price will be tanked way before the actual halving, so that when the days of halving come,
people that expect the rise will be dumped onto by people that bought much earlier and at lower price.

But like i said, there are no rules about this, there's just too many possible scenarios that can happen, but the halving - technical point or not , is definitely a part of the equation.

[jehst]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

[mirana12345]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

¸

Trust me, there will also be speculative investments as well. Looking at the whole bitcoin architecture and possibilities , there's just too much money to be made regarding bitcoin,
so people will never abandon it and it will most probably never even return to double digit price. On the other hand, top price could be anything really, from what i can see, the bigger the price overtime,
the stronger infrastructure builds around it.

[jaysabi]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

The premise that there will always be price-agnostic bidders defies the laws of economics, and there are too many competing cryptos that fulfill the same function as bitcoin for this to be true indefinitely. What you have is a tremendously small sample size (relative to the whole bitcoin market) that you are projecting into an absolute truth.

[jehst]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

The premise that there will always be price-agnostic bidders defies the laws of economics, and there are too many competing cryptos that fulfill the same function as bitcoin for this to be true indefinitely. What you have is a tremendously small sample size (relative to the whole bitcoin market) that you are projecting into an absolute truth.

Litecoin has been "competing" with bitcoin for years and no one is using it for any non-speculative purposes. I'm not saying that the existence of bitcoin's price floor is an absolute truth but it will be true for the next 1-2 years, which is what I'm interested in.

[NorrisK]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

The premise that there will always be price-agnostic bidders defies the laws of economics, and there are too many competing cryptos that fulfill the same function as bitcoin for this to be true indefinitely. What you have is a tremendously small sample size (relative to the whole bitcoin market) that you are projecting into an absolute truth.

Litecoin has been "competing" with bitcoin for years and no one is using it for any non-speculative purposes. I'm not saying that the existence of bitcoin's price floor is an absolute truth but it will be true for the next 1-2 years, which is what I'm interested in.

I wouldn't call litecoin competing to bitcoin. I would rather look at it as living next to it or being in symbiosis.

Litecoin is great for its cheaper price and faster blocks, while still being secure.

[knowhow]

Well some people trying the luck witht the halving that will happen,the question is demand and supply doesnt affect bitcoin at the moment,since as you stated around 3600btc coins daily ,but well poloniex one of the biggest exchanges has around 800btc to 5k btc trading daily soo whales present at it .
The demand and supply doesnt work at the bitcoin,the big value and market capitalisation,is too big to this works at the moment,even when the mine ends the demands and supply will keep being impossible ,bitcoin market need more interest and new investors.

[Possum577]

Conclusion:

1. Bitcoin has a floor but practically no ceiling. But it's impossible to calculate where the floor is.
2. The effect of the halving really depends on market sentiment at the time of the halving. It could be large or insignificant.

Your conclusions are correct:
1. Bitcoin has a floor, it's zero. And the sky is technically the limit
2. The halving will depend on market sentiment at that time...so predictions are meaningless

I feel that you're missing some categories in your Demand section. First, I don't think the people willing to buy at any price represent a large portion of demand. Maybe it's in your wording, but people that need bitcoin regardless of price are still limited by the price. They can only afford $xxx to spend so if the price of bitcoin rises they can exchange that fiat for less bitcoin, therefore their impact on demand goes down (because they can't buy as many bitcoin).

PS: If you want all 21 Million you'll need mine...what are you offering for mine?

[jaysabi]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

The premise that there will always be price-agnostic bidders defies the laws of economics, and there are too many competing cryptos that fulfill the same function as bitcoin for this to be true indefinitely. What you have is a tremendously small sample size (relative to the whole bitcoin market) that you are projecting into an absolute truth.

Litecoin has been "competing" with bitcoin for years and no one is using it for any non-speculative purposes. I'm not saying that the existence of bitcoin's price floor is an absolute truth but it will be true for the next 1-2 years, which is what I'm interested in.

Fair enough on the 1-2 year time span, but that's not how you presented your point initially. You said there would always be price-agnostic buyers, an absolute which cannot be true.

As far as competition goes, every alt coin that exists is competing with bitcoin, and they all in turn are competing with fiat. All bitcoin is is a method of value transfer, and the same is true of any other cryptocurrency. To the extent any alt coin has a market cap above zero, it means there is some market for it, so it is succeeding to that extent. (A failed coin has a market cap of 0, because no one is willing to trade anything of value for it, whether that be another coin or any physical object.) Litecoin is used for the same purpose as bitcoin: to transfer value from one person to another. Same as Doge, Dash, Clams, etc. They're all competing over the same market of value transfers.

[johnyj]

If you buy 100 coins and hold it for 10 years, they will disappear from market for 10 years, thus make the supply decrease by 100 bitcoins

If you buy 100 coins and sell them after 10 days, they will disappear from market for just 10 days, then another buyer hold it for another 10 days, and they will disappear from market for another 10 days...

So if the interval between buying and selling are days instead of years, you will need 365x more people to make the same amount of coin disappear from market for 10 years

It is clear, the supply shrink caused by high frequency trading is magnitudes lower than long term holding. But high frequency trading increase the market liquidity, some trader might trade forever for 10 years thus reaching the same result (but contributing large amount of fees)

Suppose that an exchange's daily average volume is 20K coins, then all those transactions on the exchange have the effect of removing 20K coins from circulation. However, the exchange might have a cold wallet holding 200K coins, they seldom move, causing a large decrease in supply

[OROBTC]

I think the main thing that I can contribute is the idea that there IS a non-zero floor. Because there will always be price-agnostic bidders who are using bitcoin for non-speculative purpose. There is a floor and it's above zero.

The premise that there will always be price-agnostic bidders defies the laws of economics, and there are too many competing cryptos that fulfill the same function as bitcoin for this to be true indefinitely. What you have is a tremendously small sample size (relative to the whole bitcoin market) that you are projecting into an absolute truth.

Litecoin has been "competing" with bitcoin for years and no one is using it for any non-speculative purposes. I'm not saying that the existence of bitcoin's price floor is an absolute truth but it will be true for the next 1-2 years, which is what I'm interested in.

I wouldn't call litecoin competing to bitcoin. I would rather look at it as living next to it or being in symbiosis.

Litecoin is great for its cheaper price and faster blocks, while still being secure.

Can you buy gold NOW with Litecoin?  If so, well then LTC might become a viable contender to BTC.  If not..., then my guess is that BTC's advantages as the first useful crypto are too hard to break for the medium-term.  

Retail acceptance is an important part of the BTC Equation...

[pitham1]

Demand:
1) My experience as a seller on localbitcoins.com tells me that there is a baseline daily demand of bitcoin coming from people who are price-agnostic such as Joe Schmoe who is buying $2000 worth of bitcoin per day for.. something. Maybe remittances. I never ask. But I will tell you that every day he needs $2000 worth of bitcoin no matter what the price is. He doesn't care what the price is. This component of total demand, measured in USD, is inelastic.

Should you be measuring demand in terms of USD? That might not give the true picture. If Joe Schmoe was buying 10 bitcoins a day, no matter what the price is, I would agree that this component of total demand is inelastic. But if he is buying $2000 worth of bitcoins everyday, I am not sure you should be considering this demand inelastic.