Modeling Bitcoin Value with Scarcity

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
EfficientInvestor
Posts: 261
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Modeling Bitcoin Value with Scarcity

Post by EfficientInvestor » Mon Nov 04, 2019 3:03 pm

Like many Bogleheads, I generally have not looked at Bitcoin as an investment. I have just viewed it as a fun lottery ticket to own in the event that it continues to go up. For instance, I bought an amount equal to about 0.5% of my net worth in mid-2017. I sold a little after the big run up in late 2017 and haven't touched it since. Entering today, my total balance was still about 0.5% of my net worth.

I read an article today (see "source" link below) by an anonymous Bitcoin expert that goes by the name PlanB. The article presents data regarding the correlation between the price of Bitcoin and the stock-to-flow ratio. The stock-to-flow ratio is the current amount of a product in circulation divided by the annual amount of product added into circulation. Since inception of Bitcoin, this ratio has been a solid predictor of market cap. As you can see in the image, the R2 value is 95%. The article also points out that due to the way Bitcoin works, the stock-to-flow ratio is due to double in May 2020. This is because the amount of bitcoin awarded per block to the miners will cut in half around that time. If this correlation still holds, this could take the market cap of Bitcoin (currently $167 Billion) up to $1 Trillion.

While I'm still cautious about the future of Bitcoin, I must say this is the most compelling argument I have read for owning it in my portfolio. Based on this data, I decided to increase my allocation to 1% and may go up to as much as 2%. I will still continue to look at it like a lottery ticket. However, this is a lottery ticket that has some pretty convincing probabilities behind it. It will be very interesting to see if the correlation continues to hold.

What do others think about this data and whether the trend can continue? My thoughts are that even if Bitcoin becomes the world currency, it has to come to equilibrium at some point and that the stock-to-flow (and market cap) cannot continue to rise at a similar rate forever.

Image

source: https://medium.com/@100trillionUSD/mode ... fa0fc03e25

For those interested in more info, We Study Billionaires (The Investor's Podcast) did a very interesting podcast interview with PlanB - https://www.theinvestorspodcast.com/epi ... ue-plan-b/

david
Posts: 190
Joined: Thu Jan 03, 2013 12:39 am

Re: Modeling Bitcoin Value with Scarcity

Post by david » Mon Nov 04, 2019 5:41 pm

There are a lot of rare things. That doesn't make it valuable.

If people stop using Bitcoin because: there's a better system, some flaw identified, not enough currency in circulation due to lost private keys etc. It doesn't matter if you have all the Bitcoin if there is no demand.

For example, Google just launched a proof of concept quantum computer. That device may be able to calculate hash values in seconds what takes decides years or millennia. This might be a great to mine Bitcoin, but it will be a huge flaw in Bitcoin as it all works on public key cryptography which can be broken by quantum devices. Any hardcoded system could be vulnerable. Thus could make Bitcoin value drop like a stone.

Additionally, I have not seen large increases in people accepting Bitcoin as a currency.

Unless there is demand, I don't know whether value has to go up because of scarcity.

NotTooDeepLearning
Posts: 95
Joined: Wed Oct 24, 2018 2:04 pm

Re: Modeling Bitcoin Value with Scarcity

Post by NotTooDeepLearning » Mon Nov 04, 2019 6:42 pm

I skimmed the article. It seems the author just discovered that if you take the log of two things their relationship becomes surprisingly linear. Then when you transform the variables back to their original scale (the scale the variables actually exist in) the relationship disappears. Just shoddy data manipulations. PlanB's misinterpretation of the results of the hypothesis test doesn't help their case either.

Edit: It seems from the comments that they also ignored the critical assumption that the regression residuals be independent and normally distributed. This alone invalidates the results of the F-test.

Northern Flicker
Posts: 5055
Joined: Fri Apr 10, 2015 12:29 am

Re: Modeling Bitcoin Value with Scarcity

Post by Northern Flicker » Mon Nov 04, 2019 7:22 pm

There may be a limit to how many bitcoins there can be in existence, but there is no limit to how many different crypto “currencies” can be in existence. Scarcity is an illusion.
Index fund investor since 1987.

User avatar
JoMoney
Posts: 8155
Joined: Tue Jul 23, 2013 5:31 am

Re: Modeling Bitcoin Value with Scarcity

Post by JoMoney » Mon Nov 04, 2019 7:50 pm

I have no idea how big it will get before it collapses... but regardless, less money will come out of Bitcoin then people exchanged into it.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

columbia
Posts: 2141
Joined: Tue Aug 27, 2013 5:30 am

Re: Modeling Bitcoin Value with Scarcity

Post by columbia » Mon Nov 04, 2019 9:20 pm

At least one can turn gold into jewelry (although I don’t place any intrinsic value on that either) and industrial parts. And dental crowns too, if you go to an out of date dentist. :mrgreen:

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Modeling Bitcoin Value with Scarcity

Post by Thesaints » Tue Nov 05, 2019 2:31 am

There are countless scarce items that are nonetheless worthless.

User avatar
nisiprius
Advisory Board
Posts: 39737
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Modeling Bitcoin Value with Scarcity

Post by nisiprius » Tue Nov 05, 2019 6:07 am

And there are countless perfectly sound speculations--speculations for which there is some obvious value other than simply scarcity itself.

You could convince me that X is not worthless, that X is a sounder than a lottery ticket, and that people who buy X as a speculative investment are not necessarily out of their minds, without it adding up to "a compelling argument for owning X in my portfolio."

For example, in one or two of his racing novels, Dick Francis has his characters present a case that investing in a race horse (as an owner) can be a really good investment. Great. A Google search suggests that the typical price of a race horse might be in five figures, which means that, yes, I could swing it. (Ignoring likely continuing costs, vet bills, trainers, stables, transport to races, etc.) I'm not buying any racehorses.

Nor bull semen, another investment in the same category.

A relative of mine was a farmer who owned a woodlot, and it made good money for him. I'm not buying any woodlots myself.

I know someone who inherited an M. C. Escher print that his dad had bought in the 1950s for $100. He sold it at auction around the year 2000 for $10,000. $7,500 after Sotheby's commission. Tax on capital gains. Shipping costs, insurance costs, conservator's costs, yadda yadda. Include 6X inflation between purchase and sale. It still works out to a real return of better than 10X over forty years = 6% per year. I'm not buying any Escher prints myself.

If bitcoin sings to you, the way racehorses or trees or fine art prints sing to people, if you just want to be involved somehow, and appreciate bitcoin for its own sake, buy it because you love it.
Last edited by nisiprius on Wed Nov 06, 2019 7:52 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

columbia
Posts: 2141
Joined: Tue Aug 27, 2013 5:30 am

Re: Modeling Bitcoin Value with Scarcity

Post by columbia » Tue Nov 05, 2019 6:33 am

This topic reminds me of the guitar market. Last year, I paid $400 for an off brand one made in 1965. At the time, my best research indicated that it originally sold for about $50 new.

It’s scarce, no doubt, but...

PVW
Posts: 612
Joined: Mon Dec 29, 2014 11:01 am

Re: Modeling Bitcoin Value with Scarcity

Post by PVW » Tue Nov 05, 2019 8:57 am

The article and discussion have the feel of a bunch of undergrads that just learned regression analysis. PlanB didn't discover a new correlation in bitcoin price, s/he rediscovered that many things show a correlation when plotted on a log-log plot. Take a look at the graph and data points - some points on the graph are off of the trend line by a factor of 10. What is a reasonable confidence interval on the predicted value of bitcoin when the scarcity doubles?

Digital currency, gold and silver are all objects of desire and their price is a function of this desire. However, there is a lot of inertia in the desire for gold and silver. Not so with bitcoin. Desire for bitcoin can change in the blink of an eye.

As with all financial predictions, you should be skeptical. Believing that this pattern was correct in the past costs you nothing. Believing that it will hold in the future can be very expensive.

User avatar
JoMoney
Posts: 8155
Joined: Tue Jul 23, 2013 5:31 am

Re: Modeling Bitcoin Value with Scarcity

Post by JoMoney » Tue Nov 05, 2019 9:35 am

I know dollars are required to pay taxes, and will be needed to repay ever larger social debts created by borrowing dollars.
At least with gold and silver I would get a shiny rock to look at, maybe even fashioned into some wearable jewelry or a coin with numismatic collectible qualities.
I still haven't seen a legitimate need for bitcoin that doesn't have a better solution that already exists.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

NewMoneyMustBeSmart
Posts: 144
Joined: Sat Jun 01, 2019 10:28 pm

Re: Modeling Bitcoin Value with Scarcity

Post by NewMoneyMustBeSmart » Wed Nov 06, 2019 7:31 am

NotTooDeepLearning wrote:
Mon Nov 04, 2019 6:42 pm
I skimmed the article. It seems the author just discovered that if you take the log of two things their relationship becomes surprisingly linear. Then when you transform the variables back to their original scale (the scale the variables actually exist in) the relationship disappears. Just shoddy data manipulations. PlanB's misinterpretation of the results of the hypothesis test doesn't help their case either.

Edit: It seems from the comments that they also ignored the critical assumption that the regression residuals be independent and normally distributed. This alone invalidates the results of the F-test.
^- this

User avatar
Blueskies123
Posts: 605
Joined: Sat Nov 15, 2014 7:18 pm
Location: South Florida

Re: Modeling Bitcoin Value with Scarcity

Post by Blueskies123 » Wed Nov 06, 2019 7:47 am

They may not be that rare. Like the Hunt brothers with silver 30 or 40 years ago I suspect the shortage is being created:
https://www.bloomberg.com/news/articles ... the-market
If you find yourself in a hole, stop digging

sean.mcgrath
Posts: 165
Joined: Thu Dec 29, 2016 6:15 am
Location: US in NL

Re: Modeling Bitcoin Value with Scarcity

Post by sean.mcgrath » Wed Nov 06, 2019 9:53 am

It surprises me that people seem to think you can't price Bitcoin.

If you assume that Bitcoin has a purpose, you should be able to price it using the marginal cost of coin generation. The next coin is fairly easy to calculate, based on your own electricity cost. The nvidia chips have a world market price, but electricity costs are wildly variable -- variance per country, plus people who have creatively managed to source for free. You'd then need to do some calculus to create a price curve over a medium period of time.

Understood that there are technology, security, usability risks -- but for a one to two year period, it should give a reasonable idea of what a "correct" price should look like. There are tons of people running these calculations, and looking for cheap/free computational resources and electricity. The wild price fluctuations come from market manipulators and sheep investors, but with some effort you could calculate an expected marginal cost two years out, and what an efficient price would be. Then, taking the mentioned risks into account, you could decide whether it's an interesting investment.

[full disclosure: zero interest here. Not a fan of investing in illiquid, intransparent, poorly-regulated markets, even though historically there is lots of money to be made in them]

Post Reply