Yeah, this is the big problem with the papers or blog posts pointing out how many coins are not moving around and then insinuating that Bitcoin is a scam or that it will soon be destroyed as an old miner dumps a few hundred thousand coins: for most of them, the parsimonious explanation for them not moving is simply that they've been lost. But there's no way to prove this! Did Satoshi wipe his hard drive when he moved on to other 'projects'? We have no way of knowing.
(There apparently are ways to verifiably destroy bitcoins - send them to impossible addresses or something like that - but I haven't heard that anyone has bothered doing that and this wouldn't apply to people losing coins accidentally or apathetically.)
This chart gives a somewhat measure of if the hoarding has gone up or not. It gives us a good idea about bitcoin speculation.
http://blockchain.info/charts/tx-trade-ratio
The lower the number on this chart, the less the speculation. Recently, the number has almost always been under 20. This number was at 225 during the July 2011 bitcoin bubble.
(More explanation about this chart: http://codinginmysleep.com/measuring-bitcoin-speculation/)
So while we would think that more people would be hoarding bitcoins, the data shows thats not necessarily true.
Combining this chart with the number of bitcoin transactions per day chart gives us a good indication that the price is increasing because more people are hearing about bitcoins and using it. (Number of transactions per day have gone up from 6,000 a year ago to 70,000 today.)
My bet is that some people will hoarde money, but it doesn't really matter. There will always be people who will use bitcoins for real economic activity, and 'speculators' will keep the value inline with other currencies. Prices will be adjusted for deflation just like they're adjusted for inflation.
Saying that the rational action is to hoarde the money until it rises in value is like saying that the rational action when buying computers is to wait an infinite amount of time, since you're always going to get a better computer by waiting 6 months.
When a really rich person doesn't have to do anything at all to keep getting richer, what will he do? If he hoards until he dies, how will that affect the economy?
I too want to see empirical data on this.
Money is not a store of value. It is currency, for current exchange.
I think the argument is a sufficiently deflationary currency creates an investor's dream: once you get enough of it, you can spend just what you need and what you have left gains value just sitting there. You don't have to work anymore or invest in anything besides the currency itself. If that becomes the most profitable investment strategy, it's in each individual's best interest to hoard money, but if enough people do that the economy collapses.
If one person "cheats" the system by hoarding money and gets to spend less and less of it every day because it's actively gaining value, that person gets something out of the deflation. If everyone tries it, the system's in deep trouble.
That seems to be the argument, anyway.
What would actually happen? I don't know… I'm a programmer, not an economist.
A crypto currency with managed inflation wouldn't make early adopters as wealthy, but it would be a better medium of exchange.
With central banks, it just became accepted that when dealing with unpredictable economies, the tendency to inflate was less dangerous than the tendency to deflate and thus why most banks will err on that side when making policy decisions. There are some great historical examples where banks organized mass deflation (I can't find one right now, but the one in my mind involved the entire country dropping zeroes from the end of their notes) and by making the change predictable, they were able to avoid many of the harmful effects.
If it were legitimately used in widespread transactions, the currencies market wouldn't impact the valuation as much, and without that turbulance, since BTC's can't "disappear" under your couch, and you have a mathematically fixed growth rate up through 2140, the currency shouldn't deflate. Just as a frequently cited example, the US had a crisis in the 70's not because pegging the dollar to gold was bad, but because the real gold reserves were covering vastly less of the actual dollars in the market drastically due to money printing.
BTC doesn't have those issues - they are fixed in circulation, have a predictable rate of disbursement that will slow to nothing in 2140, and at that point the exchange rates for goods with BTC would only fluctuate its fixed value against other currencies or against goods scarcity. I wouldn't imagine, if it actually became a well grounded currency, significant swathes of the market making a run on it and depreciating its value as badly by that point. It lets you keep currency speculation, though, without the rampant inflationary effects of quantitative easing and money printing at whims.
The problem with a coin like this is initial adoption (which is required to create the network effect that makes the coin useful). Why would anyone put serious money into a currency that is guaranteed to lose value forever through inflation and with no useful network?
The only reason inflation works for a currency like USD is because there are already billions of users locked in by the extremely large network effect, and similar inflation on any viable alternatives like the EUR.
Look at inflating currencies with smaller networks (south america) and how hard it is to keep people from getting rid of that money, usually it requires government (violent) intervention.
If the real rate return of investments over any period of time is on average less than simply hoarding the currency (an inevitable consequence of a permanently deflationary economy) your economy is screwed up.
If I'm misunderstanding, please clarify.
(In the unlikely event that Bitcoins ever became widely popular with speculators, then governments will tax or restrict domestic purchasers or exchanges for Bitcoins (they can't and won't be especially efficient at doing this; the US/EU don't need to be to severely damage the exchange value of a Bitcoin.).
Arcane hypothetical edge cases are unhelpful for exposition. It's easier to look at the conventional alternative: everybody in the world uses a means of payment which diminishes in value at a slow but stable rate. Now the average person that invests in production gains, and the average person holding cash is incentivised to convert that cash into something they want or funding the production of something others want. Isn't this better?
(1)strictly speaking we're better off looking at money as a flow, in which case the velocity of circulation comes into place, but a fixed money supply certainly wouldn't induce that to increase....
This statement is wrong, because you measure profits in money rather than value. With bitcoin the amount of money in the economy stays the same, true. However the value of each unit grows as soon as you invest and produce something useful that people want. This means, you may receive back even less than you invested in absolute units of money, but more in value.
Deflation is bad because it makes loan repayment more difficult during both recessions and periods of growth. In the extreme, deflation will worsen recessions and hinder expansions of the economy.
It is easy to think that lending is a bad thing, but let's put it this way: it is because we are able to take out loans (of various kinds) to start businesses that non-wealthy non-aristocrats are able to participate in the market. Too much credit is a bad thing, but too little credit denies capital to anyone who is not already wealthy.
The bigger issue we should be focusing on is not deflation or inflation but rather who controls it. If it is controlled by one player (government) then it ultimately decides who benefits and who loses from it. It is a much more honest situation when market controls it. Once there are too many people taking loans and going bankrupt, then investors stop investing until a point is reached at which it is more profitable to invest than to hoard again.
In any case, the extent to which people are denied access to capital is almost exactly the same as the extent to which they benefit from hoarding.The distribution of capital changes, rewarding less opportunistic and more hardworking people during deflation. Ironically, it's exactly the opposite of what governments in almost any country tell people.
Finally, I would like to add that even though capital is important, it is not the only component responsible for creating value and wealth. Thus, economy would reward not only people capable of obtaining capital, but also those who have great ideas and skills to implement those ideas.
In my view, Bitcoin will ultimately be little more than another way to make electronic payments. Merchants will accept Bitcoin only to immediately trade the Bitcoin units for some fiat currency on a Bitcoin exchange. Bitcoin exchanges will basically be payment processors. This, of course, assumes that technical problems do not destabilize everyone's trust in Bitcoin as a secure payment system (and given that it fails to meet the formal definition of security used by cryptographers, I suspect that this is bound to happen).
By the way I back my ripple credit entirely with bitcoin -- if you can make a payment go through the network to me, you are entitled to it.
having a secure system with the properties of bitcoin is worth some amount of money compared to money wasted on frictional costs in the current financial system.
If you avoid drinking the kool-aid you don't regard bitcoin as a replacement for current methods of exchange. It only really serves one of the three functions of money very well.
I'm devising a "Friedman Coin" or "Free Coin" to fix these issues. If you're interested, drop me a line.
This could easily be an instance of http://en.wikipedia.org/wiki/Gresham%27s_law in action, whereby BTC is considered "good money" by its holders and USD (etc.) considered "bad". i.e. When you take your sack of coins to the market, you spend the shaved and nicked coins first, reserving the finest coins for some more important transaction.
> thus leading some (like Nobel Prize laureate Paul Krugman) to argue that it's not a currency at all, and any value is purely speculative
This doesn't necessarily follow, if you understand Gresham's law. i.e. it's not a question of what is or isn't currency, but rather which form of currency is held more dear. This is, of course, a form of "speculation", in the broadest sense of the term.
Otherwise quite the opposite is observed in real world. There's even a word for it. Dollarization.
> They also focused mainly on the interaction between different metallic monies, comparing the relative "goodness" of silver to that of gold, which is not what Gresham was speaking of.
(from the wikipedia article)
People take risks according to their preferences and perception of risks and chances. So, if they are willing to invest and hold bitcoins, they will be holding some amount which balances chances and risks accordings to their perception - this can be as little as 10 $ or tens of thousands of dollars.
Behaving rationally, they do not take efforts and go lengths to buy bitcoins, only to spend them for something they can buy otherwise more easily and with less hassle.
But, Bitcoins also have advantages in costs, speed, safety and privacy. For example, due to creditcard costs for backcharges etc., goods paid in bitcoins could be 5 % cheaper. Because margins in online trade are small and costs matter, this is a very significant advantage. Also, time matters.
Once these savings are larger than the cost and effort needed for refilling their bitcoin wallet, people will use them for sure.
I think "cost" is the only place Bitcoin is actually a winner. Banks can clear transactions far more rapidly than Bitcoin will. Bitcoin does not meet the formal definition of security and there is a known polynomial time attack on Bitcoin (this is another way of saying "broken crypto"). Bitcoin has been repeatedly shown to not offer the anonymity or privacy that people think it offers.
I am a big fan of digital cash. The problem is that Bitcoin is not a secure digital cash system, and it should not be advertised as such.
Waiting a few seconds and confirming that most nodes on the network have received the bitcoin transaction is safe enough for most transactions (those involving values less than $100) to be considered cleared.
Waiting 1 hour makes transactions with values in the tens of thousands of dollars safe to consider cleared.
These transactions can be done 24 hours a day, 7 days a week.
In contrast, bank transactions can only be cleared during bank operating hours. For international wire transfers, it's usually necessary to physically visit a bank branch too.
When the average wait time of waiting for banks to open and the time it takes to physically arrive at a bank teller's counter is added up, it's much longer than the maximum 1 hour one would wait for a very large bitcoin transaction to clear.
The actual transfer of funds sent in an international wire transfer usually takes several hours or days too, so it can be reversed without the cooperation of the receiving bank long after the funds have been credited to the receiving party's account, within that window of time.
The time it takes for a transfer of funds from the sending bank to the receiving bank to cleared is analogous to the one hour wait for six confirmations that one would wait for extremely large bitcoin transactions, so really, even under the worst conditions (when banks are open and the person is already physically at the bank branch and when six confirmations are needed for a very large bitcoin transaction), Bitcoin is much faster than traditional banking.
" Bitcoin does not meet the formal definition of security and there is a known polynomial time attack on Bitcoin"
When you discover a way to create a provably safe decentralized digital currency, please let us know. Bitcoin is the best that is possible given the features it has.
https://en.bitcoin.it/wiki/Double-spending#Brute_force_attac...
The description dances around the issue, but what they are basically saying is this: the work needed to successfully attack Bitcoin is proportional to the total work done by all Bitcoin participants. That is an amount of work that is polynomial in the parameters of the system.
As a workaround, merchants will just demand more confirmations for higher-valued transactions, like waiting for a check to clear. Checks and physical cash also wouldn't meet your definition of formal security, but seem to work in practice.
Is there some part that does not run in polynomial time in the parameters of the system?
"I would argue incorrect to describe the result as "broken crypto", as this has little to do with the cryptographic components of Bitcoin, and is more a fundamental design limitation."
Cryptography is not limited to hash functions and digital signatures (the two primary cryptographic primitives used in Bitcoin). The "broken crypto" refers to the fact that Bitcoin fails to meet the security definition for digital cash, a type of cryptosystem. Moreover, Bitcoin is not secure as a multiparty computation protocol (also a kind of crypto) against malicious parties, as the attacker's work is polynomial in the system's parameters.
One should not make the mistake of trying to separate the security of cryptographic building blocks from the security of a system as a whole. My "pet" example of this fallacy is "robust" encryption. In a nutshell, an encryption system is "robust" if only the intended recipient can receive a valid message. Using PGP to sign and then encrypt a message is not robust, even though the signature system, hash functions, and ciphers in PGP are secure as signature systems, hash functions, and ciphers. The reason is pretty clear: if you send a signed+encrypted message to me, I can decrypt it, encrypt the signed message for someone else, and then send them a valid message.
For most PGP users, robustness is not an important security property; usually, the recipient's name or some other identifying detail will be clearly written in the message. On the other hand, for Bitcoin, the protection against double spending is absolutely necessary; a polynomial time double spending attack on Bitcoin is very bad.
"As a workaround, merchants will just demand more confirmations for higher-valued transactions"
Which only forces the attacker to make a polynomial increase in their work. An attacker only needs to do as much work as the sum of the work done by all other participants in the system to remove any workaround. Right now, that is well within the reach of governments, banks, and perhaps the larger criminal enterprises of this world.
"Checks and physical cash also wouldn't meet your definition of formal security, but seem to work in practice."
It would not be the first time someone has proposed that digital cash go beyond what is possible with physical cash (see Page 2, Property (e)):
http://pdf.aminer.org/000/120/358/universal_electronic_cash....
Quick background for those unfamiliar with the issue - Bitcoin was built to be a currency, to be used to trade for goods and services. However, a significant portion of BTC is being hoarded and not traded, thus leading some (like Nobel Prize laureate Paul Krugman) to argue that it's not a currency at all, and any value is purely speculative. For the record, I don't agree with Krugman's argument.