Why you should consider buying Bitcoin right now (Jan 2015) if you have high risk tolerance

LessWrong is where I learned about Bit­coin, sev­eral years ago, and my great­est re­gret is that I did not in­ves­ti­gate it more as soon as pos­si­ble, that peo­ple here did not yell at me louder that it was im­por­tant, and to go take a look at it. In that spirit, I will do so now.

First of all, sev­eral caveats:

* You should not go blindly buy­ing any­thing that you do not un­der­stand. If you don’t know about Bit­coin, you should start by read­ing about its his­tory, read Satoshi’s whitepa­per, etc. I will as­sume that hte rest of the read­ers who con­tinue read­ing this have a de­cent idea of what Bit­coin is.

* Un­der ab­solutely no cir­cum­stances should you in­vest money into Bit­coin that you can­not af­ford to lose. “Risk money” only! That means that if you were to lose 100% of you money, it would not par­tic­u­larly dam­age your life. Do not spend money that you will need within the next sev­eral years, or ever. You might in fact want to men­tally write off the en­tire thing as a 100% loss from the start, if that helps.

* Even more strongly, un­der ab­solutely no cir­cum­stances what­so­ever will you bor­row money in or­der to buy Bit­coins, such as us­ing mar­gin, credit card loans, us­ing your stu­dent loan, etc. This is very much similar to tak­ing out a loan, go­ing to a cas­ino and bet­ting it all on black on the roulette wheel. You would ei­ther get very lucky or po­ten­tially ruin your life. Its not worth it, this is re­al­ity, and there are no laws of the uni­verse pre­vent­ing you from los­ing.

* This post is not “in­vest­ment ad­vice”.

* I own Bit­coins, which makes me bi­ased. You should up­date to re­flect that I am go­ing to pre­sent a pro-Bit­coin case.

So why is this po­ten­tially a time to buy Bit­coins? One could think of mar­kets like a pen­du­lum, where price swings from one ex­treme to an­other over time, with a very high price cor­re­spond­ing to over-en­thu­si­asm, and a very low price cor­re­spond­ing to de­spair. As War­ren Buffett said, Mr. Mar­ket is like a manic de­pres­sive. One day he walks into your office and is ex­u­ber­ant, and offers to buy your stocks at a high price. Another day he is de­pressed and will sell them for a frac­tion of that.

The root cause of this phe­nomenon is con­fir­ma­tion bias. When things are go­ing well, and the fun­da­men­tals of a stock or com­mod­ity are strong, the price is driven up, and this re­sults in a pos­i­tive feed­back loop. In­vestors re­ceive con­fir­ma­tion of their be­lief that things are go­ing good from the price in­crease, con­firm­ing their bias. The pro­cess re­peats and builds upon it­self dur­ing a bull mar­ket, un­til it reaches a point of eu­pho­ria, in which bad news is com­pletely ig­nored or dis­be­lieved in.

The same pro­cess hap­pens in re­verse dur­ing a price de­cline, or bear mar­ket. In­vestors re­ceive the feed­back that the price is go­ing down ⇒ things are bad, and good news is ig­nored and dis­be­lieved. Both of these pro­cesses run away for a while un­til they reach enough of an ex­treme that the “smart money” (most well in­formed and in­tel­li­gent agents in the sys­tem) re­al­izes that the pro­cess has gone too far and switches sides.

Bit­coin at this point is cer­tainly some­where in the de­spair side of the pen­du­lum. I don’t want to im­ply in any way that it is not pos­si­ble for it to go lower. Pick­ing a bot­tom is prob­a­bly the most difficult thing to do in mar­kets, es­pe­cially be­fore it hap­pens, and ev­ery­one who has claimed that Bit­coin was at a bot­tom for the past year has been re­peat­edly proven wrong. (In fact, I feel a tremen­dous amount of fear in stick­ing my neck out to cre­ate this post, well aware that I could look like a com­plete idiot weeks or months or years from now and ut­terly de­stroy my rep­u­ta­tion, yet I will con­tinue any­way).

First of all, lets look at the fun­da­men­tals of Bit­coin. On one hand, things are go­ing well.

Use of Bit­coin (net­work effect):

One mea­sure­ment of Bit­coin’s value is the strenght of its net­work effect. By Met­calfe’s law, the value of a net­work is pro­pori­tonal to the square of the num­ber of nodes in the net­work.

http://​​en.wikipe­dia.org/​​wiki/​​Met­calfe%27s_law

Over the long term, Bit­coin’s price has gen­er­ally fol­lowed this law (though with wild swings to both the up­side and down­side as the pen­du­lum swings).

In terms of net­work effect, Bit­coin is do­ing well.

Bit­coin trans­ac­tions are hit­ting all time highs: (28 day av­er­age of num­ber of trans­ac­tions).

https://​​blockchain.info/​​charts/​​n-trans­ac­tions-ex­clud­ing-pop­u­lar?times­pan=2year&showDataPoints=false&daysAver­ageString=28&show_header=true&scale=0&ad­dress=

Num­ber of Bit­coin ad­dresses are hit­ting all time highs:

https://​​blockchain.info/​​charts/​​n-unique-ad­dresses?times­pan=2year&showDataPoints=false&daysAver­ageString=28&show_header=true&scale=0&ad­dress=

Mer­chant adop­tion con­tinues to hit new highs:

BitPay/​Coin­base con­tinue to re­port 10% monthly growth in the num­ber of mer­chants that ac­cept Bit­coin.

Promi­nent com­pa­nies that be­gan ac­cept­ing Bit­coin in the past year in­clude: Dell, Over­stock, Pay­pal, Microsoft, etc.

On the other hand, due to the sus­tained price de­cline, many Bt­coin busi­nesses that started up in the past two years with ven­ture cap­i­tal fund­ing have shut down. This is more of an effect of the price de­cline than a cause how­ever. In the past month es­pe­cially there has been a num­ber of bear­ish news sto­ries, such as Bit­pay lay­ing off em­ploy­ees, ex­changes Vault of Satoshi and CEX.io de­cid­ing to shut down, ex­change Bit­stamp be­ing hacked and shut down for 3 days, but ul­ti­mately is back up with­out los­ing cus­tomer funds, etc.

The cost to mine a Bit­coin is com­monly seen as one in­di­ca­tor of price. Note that the cost to mine a Bit­coin does not di­rectly de­ter­mine the *value* or use­ful­ness of a Bit­coin. I do not be­lieve in the la­bor the­ory of value: http://​​en.wikipe­dia.org/​​wiki/​​La­bor_the­ory_of_value

How­ever, there is a sta­bi­liz­ing effect in com­modi­ties, in which over time, the price of an item will of­ten con­verge to­wards the cost to pro­duce it due to mar­ket forces.

If a Bit­coin is be­ing priced at a value much greater than the cost (in min­ing equip­ment and elec­tric­ity) to cre­ate it, peo­ple will in­vest in min­ing equip­ment. This re­sults in in­creased ‘difficulty’ of min­ing and drives down the amount of Bit­coin that you can cre­ate with a par­tic­u­lar piece of min­ing equip­ment. (The amount of Bit­coins cre­ated is a fixed amount per unit of time, and thus the more min­ing equip­ment that ex­ists, the less Bit­coin each miner will get).

If Bit­coin is be­ing priced at a value be­low the cost to cre­ate it, peo­ple will stop in­vest­ing in min­ing equip­ment. This may be a sig­nal that the price is get­ting too low, and could rise.

His­tor­i­cally, the one pe­riod of time where Bit­coin was priced sig­nifi­cantly be­low the cost to pro­duce it was in late 2011. It was noted on LessWrong. The price has not cur­rently fallen to quite the same ex­tent as it did back then (which may in­di­cate that it has fur­ther to fall), how­ever the cur­rent price rel­a­tive to the min­ing cost in­di­cates we are very much in the bear­ish side of the pen­du­lum.

It is difficult to calcu­late an ex­act cost to mine a Bit­coin, be­cause this de­pends on the ex­act hard­ware used, your cost of elec­tric­ity, and a pre­dic­tion of the fu­ture difficulty ad­just­ments that will oc­cur. How­ever, we can make es­ti­mates with web­sites such as http://​​www.vn­bit­coin.org/​​bit­co­in­calcu­la­tor.php

Ac­cord­ing to this site, ev­ery available Bit­coin miner will never give you back as much money as it cost, fac­tor­ing in the hard­ware cost and elec­tric­ity cost. Up­com­ing more effi­cient min­ers which have not yet re­leased yet are es­ti­mated to pay off in about a year, if difficulty grows ex­tremely slowly, and that is for up­com­ing tech­nol­ogy which has not yet even been re­leased.

There are two im­por­tant break­points when dis­cussing Bit­coin min­ing prof­ita­bil­ity. The first is the point at which your re­turn is enough that it pays for both the elec­tric­ity and the hard­ware. The sec­ond is the point at which you make more than your elec­tric­ity costs, but can­not re­cover the hard­ware cost.

For ex­am­ple, lets say Alice pays $1000 on Bit­coin min­ing equip­ment. Every day, this min­ing equip­ment can re­turn $10 worth of Bit­coin, but it costs $5 of elec­tric­ity to run. Her gain for the day is $5, and it would take 200 days at this rate be­fore the min­ing equip­ment paid for it­self. Once she has made the de­ci­sion to pur­chase the min­ing equip­ment, the money spent on the miner is a sunk cost. The money spent on elec­tric­ity is not a sunk cost, she con­tinues to have the de­ci­sion ev­ery day of whether or not to run her min­ing equip­ment. The op­ti­mal de­ci­sion is to con­tinue to run the miner as long as it re­turns more than the elec­tric­ity cost.

Over time, the pay­out she will re­ceive from this hard­ware will de­cline, as the difficulty of min­ing Bit­coin in­creases. Even­tu­ally, her pay­out will de­cline be­low the elec­tric­ity cost, and she should shut the miner down. At this point, if her to­tal gain from run­ning the equip­ment was higher than the hard­ware cost, it was a good in­vest­ment. If it did not re­coup its cost, then it was worse than sim­ply spend­ing the money buy­ing Bit­coin on an ex­change in the first place.

This pro­cess cre­ates a feed­back into the mar­ket price of Bit­coins. Imag­ine that Bit­coin in­vestors have two choices, ei­ther they can buy Bit­coins (the com­mod­ity which has already been pro­duced by oth­ers), or they can buy min­ers, and pro­duce Bit­coins for them­self. If the Bit­coin price falls suffi­ciently that min­ing equip­ment will not re­cover its costs over time, in­vest­ment money that would have gone into min­ers in­stead goes into Bit­coin, helping to sup­port the price. As you can see from min­ing cost calcu­la­tors, we have passed this point already. (In fact, we passed it months ago already).

The sec­ond break­point is when the Bit­coin price falls so low that it falls be­low the elec­tric­ity cost of run­ning min­ing equip­ment. We have passed this point for many of the less effi­cient ways to mine. For ex­am­ple, Coin­t­erra re­cently shut down its cloud min­ing pool be­cause it was los­ing money. We have not yet passed this point for more re­cent and effi­cient min­ers, but we are get­ting fairly close to it. Cross­ing this point has oc­curred once in Bit­coin’s his­tory, in late 2011 when the price bot­tomed out near $2, be­fore giv­ing birth to the mas­sive bull run of 2012-2013 in which the price rose by a fac­tor of 500.

Mar­ket Sen­ti­ment:

I was not ac­tive in Bit­coin back in 2011, so I can­not com­pare the pre­sent time to the sen­ti­ment at the Novem­ber 2011 bot­tom. How­ever, sen­ti­ment cur­rently is the worst that I have seen by a sig­nifi­cant mar­gin. Again, this does not mean that things could not get much, much worse be­fore they get bet­ter! After all, sen­ti­ment has been grow­ing worse for months now as the price de­clines, and ev­ery­one who pre­dicted that it was as bad as it could get and the price could not pos­si­bly go be­low $X has been wrong. We are in a feed­back loop which is strongly pump­ing bear­ish­ness into all mar­ket par­ti­ci­pants, and that feed­back loop can con­tinue and has con­tinued for quite a while.

A look at mar­ket in­di­ca­tors tells us that Bit­coin is very, very over­sold, al­most his­tor­i­cally over­sold. Again, this does not mean that it could not get worse be­fore it gets bet­ter.

As I write this, the price of Bit­coin is $230. For per­spec­tive, this is down over 80% from the all time high of $1163 in Novem­ber 2013. It is still higher than the roughly $100 level it spent most of mid 2013 at.

* The av­er­age price of a Bit­coin since the last time it moved is $314.

https://​​www.red­dit.com/​​r/​​Bit­coinMar­kets/​​com­ments/​​2ez90b/​​and_the_av­er­age_bit­coin_cost_ba­sis_is/​​

The cur­rent price is a mul­ti­ple of .73 of this price. This is very low his­tor­i­cally, but not the low­est it has ever ben. THe low­est was about .39 in late 2011.

* Short in­ter­est (the num­ber of Bit­coins that were bor­rowed and sold, and must be re­bought later) hit all time highs this week, ac­cord­ing to data on the ex­change Bit­finex, at more than 25000 Bit­coins sold short:

http://​​www.bfx­data.com/​​swaphis­tory/​​to­tals.php

* Weekly RSI (rel­a­tive strength in­dex), an in­di­ca­tor which tells if a stock or com­mod­ity is ‘over­bought’ or ‘over­sold’ rel­a­tive to its his­tory, just hit its low­est value ever.

Many in­di­ca­tors are tel­ling us that Bit­coin is at or near his­tor­i­cal lev­els in terms of the depth of this bear mar­ket. In per­centage terms, the price de­cline is sur­passed only by the Novem­ber 2011 low. In terms of length, the cur­rent de­cline is more than twice as long as the pre­vi­ous longest bear mar­ket.

To sum­ma­rize: At the pre­sent time, the mar­ket is pric­ing in a sig­nifi­cant prob­a­bil­ity that Bit­coin is dy­ing.

But there are some in­di­ca­tors (such as # of trans­ac­tions) which say it is not dy­ing. Maybe it con­tinues down into oblivion, and the re­main­ing fun­da­men­tals which looked bullish turn down­wards and never re­cover. Re­mem­ber that this is re­al­ity, and any­thing can hap­pen, and noth­ing will save you.

Given all of this, we now have a choice. Peo­ple have of­ten com­pared Bit­coin to mak­ing a bet in which you have a 50% chance of los­ing ev­ery­thing, and a 50% chance of mak­ing mul­ti­ples (far more than 2x) of what you started with.

There are times when the pay­out on that bet is much lower, when ev­ery­one is eu­phoric and has been con­vinced by the pos­i­tive feed­back loop that they will win. And there are times when the pay­out on that bet is much higher, when ev­ery­one else is ex­tremely fear­ful and is con­vinced it will not pay off.

This is a time to be good ra­tio­nal­ists, and in­ves­ti­gate a pos­si­ble op­por­tu­nity, com­par­ing the pre­sent situ­a­tion to his­tor­i­cal ex­am­ples, and mak­ing an in­formed de­ci­sion. Either Bit­coin has be­gun the pro­cess of dy­ing, and this de­cline will con­tinue in stages un­til it hits zero (or some in­cred­ibly low value that is es­sen­tially the same for our pur­poses), or it will live. Based on the new all time high be­ing hit in num­ber of trans­ac­tions, and ways to spend Bit­coin, I think there is at least a rea­son­able chance it will live. Enough of a chance that it is worth tak­ing some money that you can 100% af­ford to lose, and mak­ing a bet. A ra­tio­nal gam­ble that there is a de­cent prob­a­bil­ity that it will sur­vive, at a time when a large num­ber of oth­ers are bet­ting that it will fail.

And then once you do that, try your hard­est to men­tally write it off as a com­plete loss, like you had blown the money on a va­ca­tion or a con­sumer good, and now it is gone, and then wait a long time.