There seems to be quite a bit of a Bitcoin interest around here, with several articles about it already: [1 2 3 4 5 6 7]

I propose that links and generic Bitcoin comments should be posted here, instead of making a new discussion thread for each interesting article about the subject.

New to LessWrong?

New Comment
50 comments, sorted by Click to highlight new comments since: Today at 5:48 AM

"Financial experts on Quora all agree that Bitcoin is either a horrible or an excellent idea." Several long, thoughtful comments on BC.

Has anybody seen any reply to Tyler Cowen's argument that Bitcoin's monetary velocity is unstable?

The arguments I have encountered focus on Bitcoins' strengths as a medium of exchange, but not (as Cowen points out) as a store of value, as one currency among a monetary universe composed of many money-like substitutes. Why hold non-negligible amounts of Bitcoin (as opposed to cash for its state-enforced liquidity, or any less-liquid but higher-return financial instrument of choice - recall Fisher's equation here)?

The mainstream Keynesians like to talk up liquidity preference. The post-Keynesians and modern monetary theory types talk about fiat money demand as driven by the state (to pay taxes, etc.). Well, Bitcoin cannot do that, either - it is fiat money without a fiat. We know all about Bitcoin's stable money supply. What determines its money demand? Demand for anonymous money? But the supply of anonymous-money-substitutes is not limited to Bitcoin.

(An alternative way to phrase the problem, via a standard result from international macro - flexible exchange rates and high rates of currency substitution imply unstable exchange rates; the more perfect and costless the substitution, the more unstable the exchange rate, with the rate becoming indeterminate as cost goes to zero. This is the point made by the paper Cowen mentions. Currencies in use in the world tend to correlate remarkably well with geopolitical boundaries, regardless of whatever optimal currency areas that may exist, for reasons that have less to do with their intrinsic economic properties than with state measures; states have central banks and central banks can defend currencies, so the problem presented by rampant currency substitution is really only limited to weak states in military turmoil. Bitcoin is another matter, though - it doesn't suffer any of the vagaries of bad central banking, but it can't claim any of the strengths either.)

edit: "unstable exchange rates" is not quite clear. "Exchange rates highly elastic to changes in money demand and supply" might be better.

I would say that demand for anonymous money coupled with first mover advantage seems to prop up bitcoin.

One long run scenario is this -

Almost all of the current developed nations that are not resource rich will have issues with their currency. In the absence of strong democratic will (a commodity in great shortage), growing budget deficits due to an aging population will result in the invisible tax, inflation, being used for effective relief. Nobody knows how long the resource exporter nations will continue the dance of accepting currency that isn't backed by something solid or effectively limited (like bitcoin).

What will really change bitcoin's fate is some small nations using it as a currency (post-revolution, maybe) Then, a money laundering crackdown in the US and EU would not be able to halt it.

What will really change bitcoin's fate is some small nations using it as a currency

I am unpersuaded by your argument that that any nation would do that (because nations that cannot make a credible commitment to good monetary policy can just peg their currency to the dollar). Much more likely IMHO that some wealthy actor like Tehran or the Chavez administration will increase the appeal of accepting and holding bitcoins by offering to swap bitcoins for, e.g., heating oil as a way of sticking it to the US government.

When I think of your scenario, I find it much more plausible than mine. Thanks, Richard. I update my belief.

I haven't seen Cowan's argument, but wouldn't the argument apply just as well to gold and silver, which were used for thousands of years as currencies?

Well, no. Concisely put, the problem is under-determined money demand because of readily available money or money-like substitutes (in the theoretical framework of money demand/money supply). This is an issue limited to the period of readily available new money, of which Bitcoin itself is one, really. For those thousands of years there were few such substitutes, and substitution would have been costly anyway, so the problem does not apply there.


(The problem does apply to gold speculation in this day and age, though, right? I will BTW readily concede that gold speculation in this day and age is high risk.)

It would apply if gold were legally enforced and usable as a currency, but I don't think it is.

It does apply to forex speculation, though.

OK, so investors buying and holding gold do not cause the problem.

Undergound drug market built on Bitcoin:

Thanks, excellent link! I will (not) tell you about the experiences I (don't) have when making use of the resource.

Something just occurred to me: is Bitcoin unnecessarily causing itself grief by being promoted as "money"? Once you get it classified as money, all sorts of political (not to mention emotional) baggage gets attached: the government wants to regulate it and such.

What if you kept everything the same, but promoted it as a "social status codification mechanism" or something like that? That is, make Bitcoin ownership a way of accumulating social status (among some sub-community -- Bitcoiners, if nothing else). The more coins you have relative to others, the higher your rank. Of course, nobody knows how much any person has, just how much each address has. But as long as there's the option to reveal your ownership, you have the potential for status.

If it were promoted this way, and had always been regarded this way, it would be harder for its opponents to conceive of it as competing with national currencies and therefore be harder to justify banning or taxing. After all, you don't have to report "accumulation of social status" as income on any tax form, right? You would be able to "trade" the status to buy things -- but since Bitcoin is, at root, a public database ("money is information"), any such transaction can be equivalently represented as someone gaining and someone losing social recognition.

One could no more justify taxing "status payments" than they could justify taxing "promotion to manager" or "popularity".

Just a thought...


That would also apply to people who would otherwise sell real things for bitcoins.

What if you kept everything the same, but promoted it as a "social status codification mechanism" or something like that? That is, make Bitcoin ownership a way of accumulating social status (among some sub-community -- Bitcoiners, if nothing else). The more coins you have relative to others, the higher your rank.

Interesting, and such a thing might be valuable, but that would seem to imply non-existence of currency exchanges and prevent it from serving as both social status and money simultaneously.

Btw, I want to compare Bitcoin mining rigs to supercomputers. I figured a reasonable way to do this would be to find how many floating point operation equivalents per second they are calculating. Since the hashrate is known, all that's left to make this calculation is the number of flop-equivalents it takes to perform a SHA256 hash.

I can't find this information anywhere. Do you know how to derive it or where the information would be?

(hashes/sec) * (flop-equivalents/hash) = flop-equivalents/sec

(Replying to you so someone actually sees this comment.)

I don't think there's a well-defined conversion rate. The main issue is that flops are a measure of floating-point arithmetic performance, but SHA256 hashing is mostly bitwise operations that aren't captured in that metric.

However, you can still figure out how much hashing a supercomputer can do, if you can find out how many CPUs it has and what type they are, and how many GPUs it has and what type they are. The same parts are typically used in both supercomputers and desktops, so you should be able to find benchmarks, and the way they're arranged doesn't matter much. (This is a big difference between mining and the tasks supercomputers normally perform; most of the expense of a supercomputer is the I/O backplane, which will go mostly unused.) I'm pretty sure supercomputers will end up losing badly in hashes per dollar.

All true, but I was thinking about a measure that abstracts away from the parallelism/serialness tradeoff. Obviously, supercomputers aren't going to be optimized for ultra-paralellizable tasks like mining rigs are, and I want a measure that doesn't penalize them for this.

And you don't have to guess about supercomputers being less cost-efficient in hashing -- that's the whole reason that amateurs like me, without any experience building one, can put to gether a cluster that's hugely ROR-competitive with existing rentable computing services (a theme often noted on the Bitcoin forums).

Still, there are a number of necessary operations at the assembly/machine level to perform a flop, and presumably much of the same operations are used when computing a hash. At the very least, you have to move around memory, add values, etc. There should be level of commensurably in that respect, right?

Still, there are a number of necessary operations at the assembly/machine level to perform a flop, and presumably much of the same operations are used when computing a hash. At the very least, you have to move around memory, add values, etc. There should be level of commensurably in that respect, right?

Unfortunately, there isn't; in most architectures, the integer and bitwise operations that SHA256 uses and the floating-point operations that FLOPs measure aren't even using the same silicon, except for some common parts that set up the operations but don't limit the rate at which they're done. A typical CPU will do both types of operations, just not with the same transistors, and not with any predictable ratio between the two performance numbers. A GPU will typically be specialized towards one or the other, and this is why AMD does so much better than nVidia. An FPGA or ASIC won't do floating point at all.

But certainly all of these components can do floating point arithmetic, even if it requires special programming. People could use computers to add decimals before floating-point specialized subsystems existed. And you wouldn't say that an abacus can't handle floating point arithmetic "because it has no mechanism to split the beads".

In this case, the emulation would be going the other way - using floating point to emulate integer arithmetic. This can probably be done, but it'd be dramatically less efficient than regular integer arithmetic. (Note that "arithmetic" in this case means mainly bitwise rotation, AND, OR, and XOR).

I can see why it would eliminate the possibility for currency exchanges, but why couldn't Bitcoin still function as a currency itself: you're just trading your "Bitcoin social status" for goods that someone else has, who in turn will trade the status they gained for other goods.

And I think a currency exchange could still exist, so long as it classified itself as "social status consulting". It would still have to report profits from the sale of status, but at least it wouldn't be a target of laws and regulations for competing with the national currency.

Just to toot my own horn: I finally got my liquid-cooled mining rig set up. Picture and details here.

I'm curious, what are the economics on making bitcoins? How much did it cost to build your rigs and how much do you (expect) to make?

Pre-liquid cooling, This rig was $1800 (though it could only use 2 of the GPUs so if I had only bought 2 it would be $1300). With the liquid cooling, it comes to about $2600. Right now (given current exchange rate and difficulty) it will generate about $40/day, and I've already generated $650 worth of BTC as valued at the current exchangge rate (which pays for more than two of the cards after deducting for electricity).

The difficutly will increase, but the exchange rate tends to go up with it, canceling this effect. After about 50 more days I'll be profitable (though if you count the value of the hardware, my profits have already outpaced depreciation, so I'm profitable in that sense).

If I had gone with just two cards, and kept it aircooled and running during my down-time, it would probably have already paid for itself.

Do you know why the ATI boards perform better? Nvidia has devoted more transistors and R&D to their GPGPU functions. Has no one coded directly to CUDA?

I don't know. It's always been a mystery to me. People have been surprised that I prefer ATI for parallel computing since, "Nvidia is so much better at it". (You probably looked, but the mining hardware comparison ATI consistently better as well.)

Rick Falkvinge, founder of the Pirate Party movement, argues that Bitcoin will make income taxation impossible as monetary transactions become invisible to the government. He proposes to replace income taxes with a value added tax, and a basic income scheme to take the lowest earners into account.

The VAT is actually a sensible choice here, because it is relatively easy to enforce: much of the enforcement happens "for free" as a result of transactions among suppliers, and between them and retailers. (For instance, a retailer using e-currency has some incentive to report her transactions to the government in order to get a refund for the VAT she pays to her suppliers.) There are other possibilities, though: the land tax is especially attractive because it has zero excess burden, and all governments keep track of land ownership rights as part of their basic functions.

However, I'm not sure that it makes sense to scrap income tax as a response to e-currency, because the income tax is mostly a tax on labor income, and it is also quite easy to enforce such taxes by auditing employers.

It is true that some transactions that are legally taxable will go unreported, but most of these are the kind we would want to leave untaxed in the first place, such as transactions involving second-hand goods or informal exchanges of services.

I disagreed with falkvinge's dismissal of land tax.

He's chosen VAT. I think that wherever it's easy to hide income, it should be easy to hide sales as well. Implementation of his ideas can become really difficult.

he's just poorly recreating Milton Friedman's arguments about the negative income tax.

I'm not familiar with Milton Friedman's arguments, but were they really based on the premise of an anonymous currency?

the arguments for basic income aren't directly related.

This reply is ambiguous. Are your referring to Falkvinge's or Milton's arguments?

Anyway, I find replies of the kind "someone else has already made this point better, duh" pointless and annoying. If they'd include a pointer to the earlier discussion, that'd be useful. But otherwise it's just saying "I'm better than you because I happened to hear about this subject earlier than you did, hah!".

I don't find it pointless and annoying at all if the person in question actually made the point in a much more thorough way. playing the non-strongest version of arguments off against each other is a waste of everyone's time.

The problem is not including a link to said arguments.

Google negative income tax and read the article...

Naz I think you are a little off though. the negative income tax is an implementation of a few possible implementations of a basic income system. Friedman liked it because it was better then normal welfair or the progressive tax we have. He wanted a flat tax. He did not particuly want the NIT, he wanted less welfair overhead and a flat tax.

If you do not have an income tax you can not use a negative income tax to implement a basic income.

Falkvinge is coming from the other direction. He is saying we will be forced to have a flat tax (VAT) because of bitcoin and that in order to still have wellfair we will need to implement a basic income, his citizens income.

It is all crazy talk though.

I've been thinking of allocating some of my bitcoins to bounties/prizes for various things of LessWrong interest. For example, at the beginning of this month I started a prize for promotional videos for cryonics. Anyone have suggestions as to what would be good ideas for the future?

I bought a hd5850 because mining is still profitable. it might not remain so for much longer though because the difficulty has been ramping up severely. obviously over time in should reach an equilibrium based on electricity/component costs.

I bought four hd5870s but was only able to run two because of heat issues. I took it down to install liquid cooling, which is having a bunch of snags, and the downtime has cost me dearly :-( Fortunately, I made a decent stash while the two were running.

since I plan on leaving my profits unrealized I wasnt willing to spend that much on such a risky venture.

I didn't spend anything to get BitCoin equipment, but I'm still slightly miffed that its recent increase in popularity has brought such a quick ramp in the difficulty factor that my 5770 won't see returns in months.

I've actually quit mining because the excitement of a chance of getting a winning block is now offset by the worry of having the video card 'burn out' before it pays for itself. (Also, the winter's over and the extra hot air from the computer's exhaust are no longer welcome.)

Unfortunately, once one looks behind the curtain of the system, one conclusion seems inevitable: it was deliberately designed as a pyramid scheme to enrich early adopters, not as a currency to trade goods and services. This Quora discussion explains it at some length, but basically, because a) the total currency supply is limited, b) it gets harder and harder to mine bitcoins over time, there’s now a small elite of bitcoin owners. The wealthiest own hundreds of thousands of bitcoins. Nobody will match them in wealth at this point without a prohibitive investment of money and time, and the gap between rich and poor will continue to widen indefinitely as a new bitcoin owner must spend dollars to purchase mere fractions of a coin (or contribute to its growing ecological footprint by becoming a “miner” who wastes computing cycles on the generation of coins). - blog of Erik Moeller of Wikimedia (personal blog, not official).

I would not, personally, attribute malice to the founders and say that it was deliberately a pyramid scheme. Though functionally being one may not be good.

The side effect of making early adopters rich like a pyramid scheme (or e.g. a startup) doesn't prove that it isn't a currency. They are logically separate matters.

Competition for mining power could a good thing, as it would drive the market for more energy-efficient computing equipment.

In a pyramid scheme, the early adopters are immediately given wealth as the new adopters adopt it. In bitcoin, this wealth is notional unless the early adopters sell their bitcoins.

The idea is that the early adopters shepherded the system by holding and mining bitcoin when it was very risky to do so. How risky is it now? With more merchants and more knowledge in the wealth preservation community, it is a smaller risk and its price is reflecting the same.

Bitcoin's present status is similar to that of an asset bubble, and as Mencius Moldbug said, Money is the only bubble that does not pop.

If bitcoin pops, it pops, if it doesn't, then well, anyone who sold at an earlier stage will end up feeling regret.


If I understand correctly, a pyramid scheme is something like when you try to convince me that 2 people will send me a dollar if I agree to send you a dollar. Is bitcoin really like that? In what sense do I have to send someone "already enrolled" in the scheme a dollar to get my two dollars?

It's not a pyramid scheme in the exact sense, but Bitcoin's more valuable with a wide subscriber base, and early adopters have a large advantage in terms of acquiring portions of the Bitcoin space cheaply. The changes in the investment-to-returns ratio over time that these pressures produce do end up looking a lot like a pyramid scheme's in the case where Bitcoin fails as a currency, even though you can make that investment in GPU cycles rather than dollars.

If it succeeds as a currency, of course, that curve ends up looking more like what you'd see in a gold rush or another emerging market. So some of this looks like dueling prognostications to me, or even just different levels of cynicism regarding the concept.

(By way of disclaimer, I don't currently own any Bitcoins.)

Looks like one of those instances where pyramid scheme is used as a catch all term for scams which aim to look like a franchise or investment. Could be that they are thinking of the miners as franchisees but that would only make sense if the founders were selling graphics cards.

I believe the technical term for what the bitcoin founders are making the right moves to be doing is a pump-and-dump.