Bitcoin : Where can it fail ?

[Iv]

I would like to start a conversation about bitcoin, a virtual currency that is very different from its failed predecessors. It has been recently discussed on slashdot where it received a lot of criticism that I think it does not deserve.

In a nutshell, bitcoin is a distributed system with no single point of failure that maintains a history of transactions. It is cryptographically protected against counterfeiting or double-spending. It will asymptotically reach 21 millions of monetary units.

People have focused a lot on the fact that it seems to base its value on processing power, where it is not really the case apart from the bootstrapping period. Ultimately it will be based on trust of acceptance of this currency, like many modern currencies are. Bitcoins however has a strength in this regard : no one can create inflation or debts by issuing money that was never generated. Hence, it should be more trustworthy in the long term. Bitcoins (BTC) currently exchange at 0.8$ and 0.5€.

I am not a crypto specialist so I cannot judge the technical merits of the implementation, but it seems that at least in theory it could work.

The first time I saw it, I dismissed it a bit too quickly. Now looking at it again and wondering how we can have a monetary system without relying on unreliable and opaque banking organizations, it looks like a plausible way.

What are your thoughts ? What are its weaknesses ? How can it fail ? How would you attack it if you were a criminal organization, a banking institution or a government ?

[iop]

One problem may be that it was not created by an economist, i.e. it was created by someone who doesn't really understand what a central bank does, or that a loan from a (non-government) bank is actually something useful, even though it means that money is being created.

[Iv]

One problem may be that it was not created by an economist, i.e. it was created by someone who doesn't really understand what a central bank does, or that a loan from a (non-government) bank is actually something useful, even though it means that money is being created.

So how would it fail, missing such a feature ? Loans are still possible, but money-creation through loans is not (you can not lend more money than you own). This has obviously some drawbacks but also improves the trust in the value of the currency. Can it be a show-stopper ?

[Bubbles McCoy]

From an economics perspective, I can think of two main problems with this system -

1 - Modern currency is still propped up in a bizarre sense by taxation. A government issues currency, and demands that certain transactions must be accounted for by a payment of currency to the state- government taxation creates an implicit demand for government currency, essentially.

2 - More importantly, the way modern currencies work is much more complicated, and ultimately beneficial, from just being traded around a fixed supply. It's actually reasonably important to business decisions that the rate a currency trades for is fairly stable, and a currency with a fixed stock tends to see rapid periods of deflation during recessions, exacerbating the problem (see the monetary and economic history of the US during the 19th century for examples of this, that deflation spiked during downturns is considered in part what turned recessions into depressions). Central banks nowadays try to expand/contract the money supply so inflation expectations are relatively stable or at least converge to a more stable point (using concepts like the Taylor rule). Central banks may be occasionally unstable, but their ability to modify money supply actually makes them more dependable and predictable then a currency which is has value solely dependent on current demand.

The problem with a currency only being dependent only on demand is that it creates positive feedback loops, whereas a central bank can create negative feedback loops. Say bitcoins suddenly become quite popular. Then their value will rise with demand for them, and the appearance of the rise in value will make them more appealing to buy, and demand will rise even further. The reverse can also happen, as a decline in demand can cause a panic and everyone fleeing from the currency. When there's a flexible amount of currency, the controllers of the currency can modify the amount in circulation as to keep pace with demand, which allows for more stable price regimes.

[firechicago]

More generally, there's already a tried and true system for establishing a currency which is relatively immune to inflation and manipulation: the gold standard.

Of course the gold standard is about as popular among economists as unlimited carbon emissions are among climatologists, but in order to understand why that is would require actually having a working understanding of monetary policy and central banking, rather than just coming up with a neat implementation of a cryptography system that could be used like a currency.

In short, this is a technical solution in search of a problem.

[iChef]

I fail to see how the gold standard is any better than a fiat currency. It was only really in use from the very late 1800's until the early 1970's. Most monetary systems before that were based on silver from classical antiquity until the 1800's when silver crashed because of large deposits being found in the Americas. This could happen to gold as well. Bits of metal don't have any more value than slips of paper aside from the value we agree to give to them. Aside from any value the materials may have in a manufacturing sense. Then you have to choose between money, jewelry or high grade electronics the gold can't be all three at once.

The value of the Bitcoin comes from the same place as all other currencies, the people's willingness to take it. If tomorrow everyone decides they don't want US dollars anymore there isn't anything the central bank can do about it.

[Dark567]

More generally, there's already a tried and true system for establishing a currency which is relatively immune to inflation and manipulation: the gold standard.

What about deflation??? That tends to be much worse than inflation, and gold isn't immune.

[KnightExemplar]

More generally, there's already a tried and true system for establishing a currency which is relatively immune to inflation and manipulation: the gold standard.

What about deflation??? That tends to be much worse than inflation, and gold isn't immune.

BTW: I've always wondered about this...

I've heard that slight inflation is apparently the ideal. Any reasoning for it? I'd assume that if we had deflation, then no one would ever make an investment. (ie: its better to just horde your money for the future, when it would be worth more). In inflation, the general population is encouraged to invest into infrastructure... (ie: 401K plans provide investment dollars for companies, and a few years down the road everyone benefits... at least in theory)

At least, thats always how I've always looked at it. But if someone who actually understood economics explained it to me, that would be great :-p

[Silknor]

I've heard that slight inflation is apparently the ideal. Any reasoning for it?

Without getting into the discussion about Bitcoin, I can try to answer this question. We'll assume that inflation is expected and low (this means we'll ignore things like inflation higher than expected transfering wealth from creditors to debtors, because the interest rate on a loan takes into account expected inflation). Why would that be better than a currency without any inflation?

One reason is that people in general are bad at understanding inflation and translating things into real terms. So if inflation is 2%, and you get a raise of 2%, you'll feel richer, even though your actual income hasn't changed. Stagnating wages can breed discontentment.

Another is to allow more effective monetary policy in a recession. One of the main ways for the Federal Reserve to stimulate the economy is to lower interest rates. This encourages borrowing by companies so they can invest, thus creating demand (same for consumers who borrow to spend). While nominal (not inflation-adjusted) interest rates can't really go below 0, it's the real rate that matters. If inflation is at 2%, then a loan at 2% is just like borrowing money interest free if there was no inflation. But if the interest rate is less than inflation, then the Fed can effectively make loans cheaper than free, encouraging a greater level of borrowing and investment, helping to create the aggregate demand needed to get out of a recession.

And in general, inflation encourages people to invest their money in physical assets. This sort of investment is necessary for long-term economic growth.

[Zamfir]

Another is to allow more effective monetary policy in a recession. One of the main ways for the Federal Reserve to stimulate the economy is to lower interest rates. This encourages borrowing by companies so they can invest, thus creating demand (same for consumers who borrow to spend). While nominal (not inflation-adjusted) interest rates can't really go below 0, it's the real rate that matters. If inflation is at 2%, then a loan at 2% is just like borrowing money interest free if there was no inflation. But if the interest rate is less than inflation, then the Fed can effectively make loans cheaper than free, encouraging a greater level of borrowing and investment, helping to create the aggregate demand needed to get out of a recession.

A few years ago, Willem Buiter had a good column on ways to get negative nominal interest rates, at least at central bank level: http://blogs.ft.com/maverecon/2009/05/negative-interest-rates-when-are-they-coming-to-a-central-bank-near-you/ It's a really interesting piece, because the question forces you to look sharply at the different functions and charcteristics of things like securities and currency, and to what extent those are separable.

[Iulus Cofield]

Jane Jacobs proposed an interesting idea in Cities and the Wealth of Nations. She said inflation and deflation were good, as long as a currency was tied to a single economic region, i.e. a city. When in/deflation occur on the scale of a single economy, they have corrective factors on un/employment, wages, interregional trade prices, purchasing power, etc. Furthermore, she claimed that national currencies can be bad, because a market crash in New York can affect markets in San Francisco that would otherwise be unscathed. Bitcoin would seem to exacerbate this problem by further removing currency from the actual economies its used in.

But she could be crazy and I'd never know, because economics makes my head hurt.

[firechicago]

I fail to see how the gold standard is any better than a fiat currency. It was only really in use from the very late 1800's until the early 1970's. Most monetary systems before that were based on silver from classical antiquity until the 1800's when silver crashed because of large deposits being found in the Americas. This could happen to gold as well. Bits of metal don't have any more value than slips of paper aside from the value we agree to give to them. Aside from any value the materials may have in a manufacturing sense. Then you have to choose between money, jewelry or high grade electronics the gold can't be all three at once.

The value of the Bitcoin comes from the same place as all other currencies, the people's willingness to take it. If tomorrow everyone decides they don't want US dollars anymore there isn't anything the central bank can do about it.

The gold standard is "better" in the same ways that it is claimed that bitcoin is better, it serves as an independent store of value, not dependent on any single central bank. And the fact is that people have been accepting gold as a store of value for at least a couple thousand times longer than they've been accepting bitcoin, which suggests to me that gold would be considerably more stable.

(It's also worth noting that your history of gold and silver is grossly oversimplified. The big New World silver and gold finds mostly occurred in the 16th and 17th centuries. The thing that made the value silver collapse in the latter half of the 19th century was the fact that several big economies moved from a dual standard to a gold standard, thus dramatically reducing the demand for silver. There's no possible currency that could survive a move like that unscathed.)

More generally, there's already a tried and true system for establishing a currency which is relatively immune to inflation and manipulation: the gold standard.

What about deflation??? That tends to be much worse than inflation, and gold isn't immune.

Of course, that's why I said that the gold standard was wildly unpopular among economists. (In fact, the gold standard can be worse in an economic downturn than just using gold as a currency, because central banks need to protect their gold stocks, so they raise interest rates to make cash more valuable than gold. But raising interest rates only deepens the liquidity trap that causes the deflationary spiral.)

[Iv]

1 - Modern currency is still propped up in a bizarre sense by taxation. A government issues currency, and demands that certain transactions must be accounted for by a payment of currency to the state- government taxation creates an implicit demand for government currency, essentially.

That is an interesting point. *It does give an edge to the official legal tender but I do not think that it is a show stopper for an alternative currency.

2 - (On the regulating power of central banks)

Well central banks can play with some levers to regulate some economic fluxes. I have the opinion that is really akin to cheating by using a cognitive bias : Americans say "a dollar is a dollar" and they have the perception that a dollar's value is constant. Or rather, they used to have this perception. When a bank used to say "Give me 1000 dollars and by the end of the year you will have 1020 dollars" the reaction is not "cool! 20 free dollars !" anymore but rather "Does this beat inflation ?" Most people (including economical actors) know today that currencies values are changing and that money creation actually dilutes the value of money. Instead, an entity that would want to regulate bitcoins economy would "just" have to store a lot of it. If an entity owns 10% of the circulating BTCs, it can regulate many things by offering loans or saving accounts. Is that impossible ? Move fast if you want that kind of power, but you can buy it for ~400,000$ right now. Well probably as much as twice that because BTCs price will go up when you begin to buy them.

A BTC regulating bank is possible. It takes a lot more funds but it would have a lot more credibility.

More generally, there's already a tried and true system for establishing a currency which is relatively immune to inflation and manipulation: the gold standard.

It still is dependent on the random process of gold mining, is subject to variation if technological needs for gold explode (or if fashion suddenly become more crazy about it), is hard to transmit, requires skill to check its authenticity, does not divide exactly, can not be transmitted by wires, cannot be own anonymously, requires safe and expensive storage.

In short, this is a technical solution in search of a problem.

The irresponsibility of central authorities is a very serious problem to solve. If I am not confident about the way the currency of my country is managed, about how debts are so carelessly contracted. I want a way to use an alternative system. Buying some gold is a solution, but it has the aforementioned problems.

[Silknor]

A few years ago, Willem Buiter had a good column on ways to get negative nominal interest rates, at least at central bank level: http://blogs.ft.com/maverecon/2009/05/negative-interest-rates-when-are-they-coming-to-a-central-bank-near-you/ It's a really interesting piece, because the question forces you to look sharply at the different functions and charcteristics of things like securities and currency, and to what extent those are separable.

Quite an interesting article. But I can't imagine ever seeing any of those things implemented.

[Algrokoz]

I've heard that slight inflation is apparently the ideal. Any reasoning for it? I'd assume that if we had deflation, then no one would ever make an investment. (ie: its better to just horde your money for the future, when it would be worth more). In inflation, the general population is encouraged to invest into infrastructure... (ie: 401K plans provide investment dollars for companies, and a few years down the road everyone benefits... at least in theory)

The reason that inflation is required is because the Fed loans money to the U.S. Treasury department in exchange for Treasury bonds. Treasury bonds carry with them a payment of interest. But the nominal value of the dollars given to the Tresuary Dept. and the nominal value of the bonds given to the Fed are the same. That means that the Treasury Dept. needs to pay back the Fed with money that doesn't (and can't even theoretically) exist. Therefore the only solution to the problem is to create new money by issuing more Treasury bonds. This of course dilutes the purchasing power of all the previous money and causes inflation. The only reason that we do not have a fiat money system in the U.S. is because BANKERS wrote the laws creating and governing the Fed. It makes them money. And as other people have pointed out, inflation allows for banks to get cheaper than free loans from the federal government. It's all a house of cards, and I would challenge anyone who disagrees to TRY and prove me wrong.

[Zamfir]

That is an interesting point. *It does give an edge to the official legal tender but I do not think that it is a show stopper for an alternative currency.

I think you might be underestimating how important taxation is as backstop of the system. It's not an "edge", it's absolutely critical to the way the system works at the moment.

The obvious problem with a currency (fiat but also non-fiat) is that is worth far more as long as it stays the currency. That creates an instrinsic instability: if other people stop trusting a currency, you have to stop trusting it too. And the more people don't think the currency will stay a currency, the more likely it is to stop being a currency.

The only way to deal with such feedback loops is to guarantee that they can be stopped, no matter how large they are. For small deviations, the market will do this itself: Suppose there is someone who wants to get rid of a million dollars (and is willing to accept below-market prices) because he believes the dollar will stop being a currency. Then there is always someone else who will buy those dollars in exchange for euros, or oil, or whatever assets they seller would like. But if the numbers get big enough, the market won't always have that much liquidity. So you get short-term price fluctuations. And those could conceivably explode in a kind of bank run, where everyone is selling the currency because they think that other people are going to sell. Unlike a house or a stock, there is no reason why such a run should stop before it reaches zero.

At such a moment, there has to be backer who can buy all the currency that is offered, no matter how much, in order to stabilize the market. This is in principle a profitable business: you buy low, and you can sell slowly when the price is stable again. As long as you are big enough to buy everything in the market, and if you are sure that the run will not exceed your reserves.

Now, who is big enough? The government. Because they can tax people, so they can quickly loan enormous amounts of money and still guarantee payback. Suppose that tomorrow afternoon some group of people have a 100 billion dollars together (5% of m1, apparently). And they do not trust the dollar anymore, so they will sell them for other currencies or other stuff at any rate people are willing to offer. Without the government, people would have to scratch their heads, and wonder if the dollar will survive. Because if you buy one billion of those dollars and no one else does, you have just exchanged something valuable for a billion worthless dollars.

But for the US government, a 100 billion is not much. Just a few weeks of taxation. The Fed has that much foreign currency lying around as spare change. And if they don't have it, they can easily borrow a hundred billion elsewhere, because they can obviously pay it back out of future taxation. In most countries, the central bank's foreign currency reserves plus the short term lending power of the government is enough to buy most, probably even all of the outstanding currency quickly, if needed.

In other words, the amount of pure currency that an economy needs is large compared to most actors in that economy. But it is not particularly large for a government of that same economy. A government can dampen swings in the supply of their currency even if those swings are a noticable part of all the outstanding currency. In practice, a national currency can only collapse to zero if the government itself is failing, for example by printing money instead of taxing people.

[Iv]

I think you might be underestimating how important taxation is as backstop of the system. It's not an "edge", it's absolutely critical to the way the system works at the moment.

I am not arguing that the current dollar system could work without central authority, but that an alternative system could without one. Actually I am not even arguing it, I am looking for flaws in what is proposed in bitcoins.

Unless I am mistaken, what you propose is that there is a need for a big player that has an interest in the stability of the currency. Imagine Google using a datacenter during 3 years to mint bitcoins and to buy all that are available on exchange markets. They could reach 10% of the monetary mass. What wouldn't they be able to do that the government does ? By selling or buying at a chosen price, different from the market price, they can influence the whole currency.

From what I understand about the current crisis (and analysis differ in frightingly large ways) the problem is that many small European countries have right now done exactly what you propose : they used their weight to restore confidence in their economy and to do that they invested large sums of money, digging their debt a bit deeper. Now we are talking about states going bankrupt. There is really nothing magical about a state : it is just an actor with deep pockets and huge loan capacities. I doubt that it would be less sane if their only lever to correct fluctuation was the value they really owned instead of a bigger value, largely imaginary, based on the trust that a given state won't go bankrupt.

Bitcoins sees itself as a currency for "Ze Internetz". Imagine your ISP, registrar and web hosting service requiring bitcoins for payement (some webhosting services already accept bitcoins). Wouldn't this be akin to a tax ? The federation of actors using bitcoins have a common interest in not letting it fail : it allows them to short-circuit banking institutions that make international transactions costly and cumbersome. Sure we have no example of such a system working in the past, but I still think it is worth a try...

[Iulus Cofield]

[quote="Iv"]Imagine Google using a datacenter during 3 years to mint bitcoins and to buy all that are available on exchange markets. They could reach 10% of the monetary mass./quote]

This is probably the most troubling thing to me. Why not indeed? Why did they choose to distribute in a semi-random but dependent on processing power model? I don't understand the benefit to that, other than the incentive for people to participate so that they can get some free money.

[Iv]

Imagine Google using a datacenter during 3 years to mint bitcoins and to buy all that are available on exchange markets. They could reach 10% of the monetary mass./quote]This is probably the most troubling thing to me. Why not indeed? Why did they choose to distribute in a semi-random but dependent on processing power model? I don't understand the benefit to that, other than the incentive for people to participate so that they can get some free money.

That is the only benefit : to allow people to be rewarded for participation and to have a pseudo-fair distribution of the initial coins.

What would be the problem with Google doing the former ?

[Iulus Cofield]

But it's not fair. It gives people with a lot of servers, people who are probably already rich, a bunch of free money. Suppose Bitcoins do become a widely accepted internet currency, a bunch of corporations with large server farms can all work together to farm a substantial percentage (say 40%) of all Bitcoins produced. Since no more Bitcoins are produced after 21 million, these corporations can hold onto their Bitcoins to drive up the value relative to national currencies, then once the high value is achieved they can start lending out their Bitcoins in credit lines, credit lines that could also be paid for in real cash. Plus, companies with large server farms are also likely to companies that can easily start accepting Bitcoins as payment in their regular business, giving them even more potential to sweep a huge market share.

Hmmm, I think I need to talk to some CEO's. I smell easy money with some investment.

[Zamfir]

Unless I am mistaken, what you propose is that there is a need for a big player that has an interest in the stability of the currency. Imagine Google using a datacenter during 3 years to mint bitcoins and to buy all that are available on exchange markets. They could reach 10% of the monetary mass. What wouldn't they be able to do that the government does ? By selling or buying at a chosen price, different from the market price, they can influence the whole currency.

That solves the selling part, but that was never the problem anyway. The problem is buying: if you buy BTCs, you have to offer something else in return. And that has to be something the seller of BTCs will accept, so it has to be something liquid: something the seller can easily sell on again.

Google could hold a reserve of dollars, or other currencies, or more likely currency-denominated bank accounts. But in that case, Google would just be a bank and BTCs would be bank accounts. In practice, BTCs would be dollars. If Google held a reserve smaller than the number of outstanding BTCs, they would be dollar-creating bank, and the Fed would control them and guarantee them like one. So in order to create an independent currency, Google would have to be able to offer something liquid to potential BTC-sellers, but that liquid thing cannot be an existing currency.

Google has such a thing: they could give out IOUs that promise the bearer X amount of advertising at Google, whenever they want in the future. But that would not be very liquid, since most people do not want advertisement on Google. So if you have such an IOU, but you want bread, you'll have to find a baker who wants to advertise some more, or a baker who knows someone else who wants to advertise, etc.


The other, more important problem is that Google has only so much (valuable) advertisements to sell. If they gave out IOUs for their total advertisement space for years, then they will have less operating income in the future, and odds are they won't be around untile ll those IOUs are redeemed. So a Google-IOU backed currency can only function if the total outstanding value of that currency is not too big compared to Google itself. Perhaps Google could back a few billion dollar-equivalent worth of alternative currency, but that's it. Observation suggests that an economy needs a highly liquid currency worth a few month of the entire prodution of that economy, so Google alone can never do more than offer a little extra currency.

That's why governments,and taxation, are important: governments can give out "Pay-your-future-taxes" IOUs. Most people pay taxes, so most people can use such gov-IOUs. And government taxes add up to several months of the entire output of the economy every year. Most governments already have outstanding IOUs worth a 9 to 12 months of their entire economy's prodution. They can add a few months more if need be, and still be credible that they will redeem the IOUs when needed.


Of course, you are right that governments of small countries can guarantee so much that they could never redeem those guarantees if they all come in at the same time. In effect, they are backing liquidity for more than just their own economy. But that's a choice of some individual countries, not a necessity and not an important feature of the current system. If all countries backed a conservative, safe amount of liquidity, there would be still enough liquidity to keep the economy function.

But withou governments, that becomes tricky. You'd have to collect private organizations as backers with a total size (or more exactly, credibility) comparable to all governments in the world combined, or you would have to move to systsem where most liquidity is backed by much less conservative margins than is now the case. Or perhaps a mix, with some gvernment backed currencies and some private currencies. But you still need to find organizations with some other valuable business that are also willing to back liquidity.

[Iv]

But it's not fair.

Which never stopped a monetary system from working
Money is supposed to be exchanged. If all what people do with bitcoins is to mine them and sell them for dollars, it will have failed. But if some people get rich in BTCs and other people want to get rich in BTCs, it doesn't matter much who gets them in the start. And also, by providing huge servers to the BTC network, such an actor would have helped it start, so it is not really that unfair.

Hmmm, I think I need to talk to some CEO's. I smell easy money with some investment.

Please do, any BTC node is welcomed to join

That solves the selling part, but that was never the problem anyway. The problem is buying: if you buy BTCs, you have to offer something else in return. And that has to be something the seller of BTCs will accept, so it has to be something liquid: something the seller can easily sell on again.

Ok, I understand your point. If people start distrusting bitcoins, it can spread quickly as a self-fulfilling prophecy and can only be stopped by someone guaranteeing that it will continue to accept BTCs as payment for something worthwhile. I am tempted to say that this is not a problem. If people sell BTCs in huge numbers, it means they have another currency that has a better trust-value in their opinion, so let it be. Anyway it is not like BTCs will disappear, their localization and transactions are published in cleartext by every node (another unique characteristic of this system). It is really more like an alternative system usable only when regular currencies and central banks do not do their job correctly.

[Zamfir]

I am tempted to say that this is not a problem. If people sell BTCs in huge numbers, it means they have another currency that has a better trust-value in their opinion, so let it be.

I can't really think of an example of hyperinflation that was not extremely harmful to its economy, and to the lives of the people in it. Of course, hyperinflation tends to be caused by more or less problematic circumstances, so it is hard to say how much of the damage is cause of the problem and how much is caused by the hyperinflation. But the consensus seems to be that in historic examples, the collapse of a currency made the situation significantly worse.

Perhaps there is some way to have gradual, painless collapses. But at least up to now, economies appear to switch from a collapsing currency to another in an unexpected, uncoordinated process that disables large parts of the economy for at least months on end. That means there is a enormous amount of potential production that just doesn't happen. That's a loss of real wealth, not just nominal wealth. And in the aftermath, the (justified!) fear for another collapse distorts the economy for years on end, with people either consuming directly, or investing in highly illiquid assets. It is highly inflexible if companies and people don't have much liquid cash around for unplanned events.

And sometimes the economy doesn't manage to define a new currency, but adopts an existing one. That process is a direct transfer of wealth from the old currency holders to the holders of the new currency. Parts of South America dollarized around 1990, which basically meant they had to perform real services and trade real stuff to buy dollars abroad, without selling those dollars afterwards to import stuff in return.

Perhaps there is some way to have an economy where currencies are often collapsing and new ones come into existence, without the traditional levels of pain. Perhaps if everyone is always using a mix of currencies. Or perhaps there could be organizations that monitor for signs of collapse and pre-emptively step in to make the process more smooth. A bit like bankruptcy proceedings. Buut I don't think the BTC people are hoping to give large powers to a mix of a central bank, the IMF and a bankruptcy court.

Truth is, we have no idea how such a system might look like, let alone whether it could work at all. That's the big problem with BTC: the cryptographic trick is neat, but it is just a tiny, insignificant part of the real changes required to make an different currency system work. They have done the equivalent of designing a new car key, and then claim it could be a revolution in transportation.

[iop]

Imagine Google using a datacenter during 3 years to mint bitcoins and to buy all that are available on exchange markets. They could reach 10% of the monetary mass. What wouldn't they be able to do that the government does ? By selling or buying at a chosen price, different from the market price, they can influence the whole currency.

Why would you trust Google more with governing money that a democratically legitimized government?

From what I understand about the current crisis (and analysis differ in frightingly large ways) the problem is that many small European countries have right now done exactly what you propose : they used their weight to restore confidence in their economy and to do that they invested large sums of money, digging their debt a bit deeper. Now we are talking about states going bankrupt. There is really nothing magical about a state : it is just an actor with deep pockets and huge loan capacities. I doubt that it would be less sane if their only lever to correct fluctuation was the value they really owned instead of a bigger value, largely imaginary, based on the trust that a given state won't go bankrupt.

One big problem in the currency crisis is the Euro (in particular the lax enforcement of the economic rules that only allow healthy economies to be members). If e.g. Greece still had the Drachme, they could just devalue their currency, which would reduce both assets and debts held in the currency relative to the rest of the world. This would be a harsh move, but it is an out for a failed national economy (see Argentina, which is, btw., recovering comparatively nicely). With the bitcoin as global currency, you tie all economies together, exacerbating the problem Europe has.

The other problem with a currency that cannot grow (or shrink) according to the economy (note that the amount of gold kept increasing as more was dug up, and the amount of gold-backed money grew as percentage of backing was reduced) is deflation. Assume that lunch costs 5 bitcoin. If 1bn people go for lunch, we need 5bn bitcoins to exist. For the sake of the example, let this be the total amount of bitcoin out there. Now one additional person wants to have lunch (i.e. the economy grows). Either this person cannot eat, because there aren't enough bitcoins, or the price of food has to decrease. Thus, as the economy grows, and prices decrease (because of the limited amount of bitcoins), it becomes extremely valuable to keep the money for as long as you can afford, because by doing nothing you get richer. This may be good for the environment (less investment will lead to less consumption), but not good for those who like gadgets, and progress.

[Iv]

@Zamfir : Do you have an example of hyperinflation that was caused by the lack of action of a central authority ? I have this feeling from the (very) little economy news I read that hyperinflation usually denotes a real loss of value of the currency ant that trying to hide that fact would only displace problems. Sure a central authority can iron out some irrational variations but when there is a true reason to the currency's variation, there is little it can do. I realize I am probably very biased by recent events but the main role that a central bank seems to have is political : it allows a government to spend a ridiculously huge amount of money and delay the consequences until the faulty politicians are retired.

Why would you trust Google more with governing money that a democratically legitimized government?

Indeed, why would you ? I was not addressing the moral/political issue but merely pointing out that the bitcoins system is not mutually exclusive with a big entity offering some kind of regulation or offering loans. I am still uncertain as to whether this is necessary, but this is certainly a possibility. Actually, a single person seems to be in possession of 400k BTCs, which gives it quite an amount of power on the system.

One big problem in the currency crisis is the Euro (in particular the lax enforcement of the economic rules that only allow healthy economies to be members)

Exactly. With a modern currency, it is easy to have a negative budget and a big deficit in the economy as a whole. It is possible to have badly manage debt that create money in an uncontrolled way. I have the feeling that this kind of scheme is impossible in bitcoins. You can not lend money you don't have. Sure you can imagine people trading IOU notes on top of the bitcoin economy but as long as you only accept bitcoins rather than "IOU bitcoins" notes, you should be safe.

[MrConor]

Jane Jacobs proposed an interesting idea in Cities and the Wealth of Nations. She said inflation and deflation were good, as long as a currency was tied to a single economic region, i.e. a city. When in/deflation occur on the scale of a single economy, they have corrective factors on un/employment, wages, interregional trade prices, purchasing power, etc. Furthermore, she claimed that national currencies can be bad, because a market crash in New York can affect markets in San Francisco that would otherwise be unscathed. Bitcoin would seem to exacerbate this problem by further removing currency from the actual economies its used in.

But she could be crazy and I'd never know, because economics makes my head hurt.

It's not crazy. A scaled up version of this argument applies in Europe: whether individual countries should join the Eurozone or not, and use a single European currency rather than their own national currencies. It's a matter of weighing the costs and benefits against one another: for example, in the European case, adopting the currency can increase inter-European trade but also gives national governments less control over their monetary policy.

The same arguments would arise for adopting a global currency: it's a matter of which system favours your local economy more.

[iop]

One big problem in the currency crisis is the Euro (in particular the lax enforcement of the economic rules that only allow healthy economies to be members)

Exactly. With a modern currency, it is easy to have a negative budget and a big deficit in the economy as a whole. It is possible to have badly manage debt that create money in an uncontrolled way. I have the feeling that this kind of scheme is impossible in bitcoins. You can not lend money you don't have. Sure you can imagine people trading IOU notes on top of the bitcoin economy but as long as you only accept bitcoins rather than "IOU bitcoins" notes, you should be safe.

Ok, I didn't formulate that well. The reason the Euro made the crisis worse is that the regional currency took away the possibility of devaluation from Greece. The stability rules had been designed to avoid having a badly-managed local economy impacting everybody else, but they weren't enforced - Greece should have been kicked out of the Euro long before.

In a global bitcoin economy, you can still create and badly manage debt (note that Greece wasn't able to print Euros!) - anytime you're don't immediately pay for a service or goods you create debt, and if you can't pay what you promised, you're screwed - but economic shocks created in one economy are going to be felt everywhere, and there is the issue of deflation that, to me, is the really big show-stopper.

[Iv]

In a global bitcoin economy, you can still create and badly manage debt (note that Greece wasn't able to print Euros!)

Actually you can't and it was. First, Greece : While it was not able to actuallt print euro bills, the fact that it was allowed to have a deficit budget and contract debt is an effective way of creating money. The main one used today actually.

In a bitcoin economy you can not do that. Right now a bank can lend about 5 times the total worth of what it actually owns. A bank owning 1 billion of assets can lend a total of 5 billions. In a bitcoin economy you can only make transactions with money that you own. Such scams become impossible.

and there is the issue of deflation that, to me, is the really big show-stopper.

I can see how a deflation economy is a catastrophe for the banking sector, because they become unnecessary for gaining profit from one's savings, but I challenge the idea that deflation causes problems if non-addressed. If that is so, why not forbid banks to offer rates on their saving accounts that beat inflation ? It should be a catastrophe to the economy according to this theory. In fact I can't find examples of catastrophic deflations that caused big problems, instead all of the wikipedia examples seem to be examples of big problems causing deflation.

[iop]

In a global bitcoin economy, you can still create and badly manage debt (note that Greece wasn't able to print Euros!)

Actually you can't and it was. First, Greece : While it was not able to actuallt print euro bills, the fact that it was allowed to have a deficit budget and contract debt is an effective way of creating money. The main one used today actually.

So in a bitcoin economy, I cannot promise to pay you tomorrow? And I cannot get a loan from you? Because that's how you get into debt. Incidentially, that's basically what Greece did. They just accrued too much debt given their meager tax income and their bleak growth prospects.

In a bitcoin economy you can not do that. Right now a bank can lend about 5 times the total worth of what it actually owns. A bank owning 1 billion of assets can lend a total of 5 billions. In a bitcoin economy you can only make transactions with money that you own. Such scams become impossible.

and there is the issue of deflation that, to me, is the really big show-stopper.

I can see how a deflation economy is a catastrophe for the banking sector, because they become unnecessary for gaining profit from one's savings, but I challenge the idea that deflation causes problems if non-addressed.

Do you agree that deflation curbs investment? Basically, if your money will be worth more in the future, why invest now? And since fractional reserve banking ("the scam") is not allowed, you have to be even more careful about investing, because once your bitcoins are invested, they're gone for a while.

So, who needs investment? Businesses and individuals who are temporarily short on cash (for example a farmer whose tractor broke down and who will only be able to pay back after the harvest which he can't get if he has no tractor). People who want to start a company. Countries that want to build something big, like a new high-speed railway system. Basically lack of investment is a catastrophe for anything expanding, and assuming that work keeps getting more efficient, this also means that there will be fewer and fewer jobs. Good times!

[Iv]

So in a bitcoin economy, I cannot promise to pay you tomorrow? And I cannot get a loan from you? Because that's how you get into debt. Incidentially, that's basically what Greece did. They just accrued too much debt given their meager tax income and their bleak growth prospects.

Ok, that's right, you can get a loan and you can get some debts but what you can't have is loans in the way banks, central banks and governments practice them : by giving you a drawing right at the ATM that can exceed what they actually own. That is what we are talking when we talk about badly managed national debt.

Do you agree that deflation curbs investment?

I am out for opinions on this. From what I understand, a scenario with a 0.1% deflation is the same as 3% inflation where there are 3.1% interest rates on saving accounts. I agree that the theoretical homo economicus will probably slow its investments, and professional investors probably have a behavior close to that, but many people who invest money are expecting more than than just a return on investment. Some want to become their own boss, some have unrealistic expectations of gain, some want to make a company that does activity X. I don't know if all these people account for a marginal part of the investment economy or if it is a bigger part. Do we have numbers on that ? Also, economical conditions where a 0% return is a successful investment has its advantages.

And since fractional reserve banking ("the scam") is not allowed, you have to be even more careful about investing, because once your bitcoins are invested, they're gone for a while.

Yep. Less investment, more cautious ones, less imaginary value. I agree that it makes it hard to compete with an economy where reserve banking is allowed and is practiced responsibly, but it is a fallback economy for when irresponsibility happens.

So, who needs investment? Businesses and individuals who are temporarily short on cash (for example a farmer whose tractor broke down and who will only be able to pay back after the harvest which he can't get if he has no tractor). People who want to start a company. Countries that want to build something big, like a new high-speed railway system. Basically lack of investment is a catastrophe for anything expanding, and assuming that work keeps getting more efficient, this also means that there will be fewer and fewer jobs. Good times!

It is still possible to get loans and investments, but these are a real transfer of money. How fair is the current system where individuals can immobilize 1$ to invest 1$ but banks can immobilize 1$ to invest 5$ ? About countries getting in debt with money-making investments, I don't agree anymore that this is a good thing. Usually, one administration will contract the debt but will not have to undergo the consequences. An elected body is simply not responsible enough to be able to get into debts (even the possible bitcoins debts of contracting an obligation toward a big owner). They either have the money or they don't. They raise taxes if they need more, they make a reserve, they don't allow negative budgets when the reserve is too low.

[Zamfir]

Iv, Perhaps I am misreading you, but your previous post suggests you think that fractional reserve banking (or similar money-creatng activities) are made possible by governments and central banks. Butbthat's about the opposite of the situation: central banks restrict money creation.

If an organization engages in an activity that increases the supply of the official currency, they are required to register with the central bank, open their books, and in general become part of a tightly controlled system that enables the government to limit money creation. Angd in most countries, the creation of an alternative currency outside of that control is limited or even forbidden.

All those measures are in place reduce (or at least control) fractional banking, and therefore to create a relatively stable and trustworthy currency.

In a system with private currencies and no central control, you would have far more fractional banking. That's why central banks were started in the first place, because uncontrolled banks created instabilities.

Imagine a bank that offers to pay you a small interest on BTCs. The terms of the loan are the same as when you loan money to a standard bank payment account: you can withdraw the BTCs whenever you want, the give you a debit card so you can pay with BTCs from your loan in shops.

In such a situation, you would treat that BTC account just the way you treat your normal bank account: you would act if you are holding the BTCs yourself, because you can spend them anytime you want. But at the same time, the bank has loaned most of those BTCs to other people, who also keep them in an a account and also act as if they hold all of them.

That's how money is created, and it works just as well for BTCs as for government money.

[Iv]

Here is what I understand about how it works :
When bank A wants to loan money, it has to buy some rights at the central bank, proving that it owns enough assets to be robust enough if many people withdraw money at the same time. Buying one billion at the central bank allows them to lend 5 billions. That's the figure I had in mind. If all the people at a bank were to withdraw their money, only ~20% of them would succeed. I am not sure about governments, but I had the feeling that they could bypass many rules to finance their deficits. Maybe that is a wrong assumption.

Of course, you can give your bitcoins to a banker you chose to trust because you think he is a wise investor and that he will pay you back. But the interesting idea of bitcoins is the ability to have a digital money account that you can use to make international payment without having to trust a bank. But I suspect that people who will be willing to accept IOU tickets instead of hard bitcoin cash won't be that numerous.

[Bubbles McCoy]

Yeah, that's not quite how fractional reserve banking works.

I believe banks sometimes do get money from central banks without paying for it immediately, but these actions always take the form of loans and are generally only performed in crisis for a higher then market interest rate (see lender of last resort); they have to give it all back someday though. Common currency, on the other hand, is exchanged for government bonds in a one-to-one ratio.

Fractional reserve banking is where a bank is essentially allowed to loan out money it has been given as a deposit. The thing is money doesn't spend a whole lot of time in the transaction phase; generally speaking people take small increments out of a bank account, buy something, and then the seller deposits the money again in his own account. So, if I deposit $100,000, if there's a reserve ratio of 20% then the bank will then lend out $80,000 to someone. If this someone then spends the money, the seller will then have $80,000 which they will in all likelihood deposit in a bank for the time being, where again there's a 20% reserve, and $64,000 is lent out, etc. However, at this point in time I believe I have $100,000 in hard cash, and the seller believes they have $80,000 is hard cash, yet only $100,000 is backing this directly - the other $80,000 comes from the promise the lender will pay this back someday.

The amount of money that a dollar can come to represent under this policy is then geometrically restrained by the reserve requirement. Without one, in theory this could continue on forever, and a single dollar can represent any number of apparent dollars after it's been loaned and spent enough. This is quite unstable in the event people start to withdraw money though, as very few actual dollars are in possession of the bank at any one point in time (a deleveraging crisis). So, with the above reserver requirement of 20%, the geometric limits make it so that at no point will my $100,000 ever come to represent more then $500,000. Thus the "additional" dollars created by fractional reserve banking is entirely a symptom of the nature of investment itself, it has nothing to do with the rights to loan money sold by a central bank.

Going back to deflation - in the case where there's .1% inflation, there's a low chance of a problem manifesting. Where issues arise with deflation is more when there's deflation of something like 3%, but the real interest rate is at 1% - this would imply that the nominal interest rate should be -2%. Yet this is impossible, as holding money then is a better bet than investment. This creates a cycle of deflation, as when people start to withdraw money from investments to hold cash. As people are demanding currency, more and more dollars are being held just as savings, causing further deflation, and more withdrawal from investment. As mentioned earlier, this is well documented by several depressions in the nineteenth and early twentieth century.

[Zamfir]

. But I suspect that people who will be willing to accept IOU tickets instead of hard bitcoin cash won't be that numerous.

But there is nothing 'hard' about bitcoins. A bitcoin is like a bank account with a bank that doesn't own anything at all. A bank account (even in the Cayman islands without government guarantees) is at least backed by assets, namely the loans the bank made.

Not all of those assets are liquid, but that's exactly the function or a bank: a bank takes in short loans, like your bank account or a 2-year fixed deposit. Then they loan most of that on as long and often illiquid loans, like mortgages or company loans that won't be repaid for decades. And they only have to keep a smaller amount of liquid assets around, which is still enough in most circumstances.

You, as a regular guy, cannot afford to have all your savings tied up in 30 years mortgages. You need part of your savings around as liquid money, to buy stuff immediately if needed. But by pooling your savings witter those of others, a bank can offer you the liquidity of money, while still investing the majority of your savings in long term investments.

[Iv]

@Bubbles: I agree that without central banks, the situation would be far worse with banks able to create the amount of money they want. But still, the sum of all the money that people believe they own is far less than the assets that the bank actually owns. And the value of the currency can be easily influenced by politicians who decide or not to have a big debts, to issue governmental loans, etc...

But there is nothing 'hard' about bitcoins.

There is something hard about bitcoin that the dollar doesn't have : the certitude that there are 21 million BTCs and that it won't ever go higher than that number. Your bitcoins do not disappear if the bank goes bankrupt, if the country goes bankrupt, if someone decides to print money because God said to do it.

The regular currencies don't manage to choose between assets-backed and trust-based system. They hybrid both by having some assets and assuming some trust and trying to balance both. Bitcoins chooses the 100% trust one. Acutally this conversation took a turn I did not expect. I was not really wondering about the outcomes of a global economy that would be based on bitcoins but rather I was wondering how one would attack the bitcoin system if he wanted it to fail and had consequent power (like a government). The current discussion is still interesting, but I was actually fearing that bitcoins would fail before these questions became relevant.

A bitcoin is like a bank account with a bank that doesn't own anything at all. A bank account (even in the Cayman islands without government guarantees) is at least backed by assets, namely the loans the bank made.

Isn't that a bit circular ? You can trust it because so many people trust it already ? It looks like 100% trust to me. The problem being that a central authority (the Cayman bank itself) can decide to blow up this trust for whatever reason and completely destroy the system.

You, as a regular guy, cannot afford to have all your savings tied up in 30 years mortgages. You need part of your savings around as liquid money, to buy stuff immediately if needed. But by pooling your savings witter those of others, a bank can offer you the liquidity of money, while still investing the majority of your savings in long term investments.

I'll be happy to have all my savings on a flat bitcoin account. And as most people think that deflation is looming to this system, it means that I don't need to make complicated maneuvers to get an interest rate.

[Zamfir]

Banks do have, more or less, assets for every dollar they owe their creditors (including you as account holder). What they don't have are enough liquid assets. If all creditors wanted their money back right now, the bank would have to hand over illiquid assets, because it does not have enough liquid assets for everyone.

So instead of money that you can directly use to buy stuff, you would find that someone now owes their mortgage payments to you, or that you own part of an office block in Tokyo, or 30 years bonds given out by the municipality of Karstheim-am-Rhein. That's the sort of stuff banks hold, instead of only money.

That's a problem if all those creditors then try to sell their share in that office block at the same time because they don't want to own an office block but want to buy iPads and food. The market for illiquid assets is by definition small. So the price for such office blocks would temporarily crash, and the creditors would find that they haves much less cash to buy iPads. Someone else will have gotten an office block at a bargain price.

In regulated countries, this would not happen. Instead, the government will take over those assets (in practice they just take over the bank as whole), and they will pay cash money to the creditors which they can borrow because of their size and tax powers. Although in theory they will only pay up to a maximum only. The illiquidity then becomes a problem of the government, who is much more able to slowly sell stuff over a long time.

But in return for giving this guarantee, the government wants control over the asset-buying choices of those banks, requiring them to have enough liquid assets at hand to make the scenario implausible. That doesn't have be just money, but for example also short-term bonds of national governments. Those can readily be sold on the market without much loss.

For an unregulated bank, like one in the Cayman islands, the account holders don'have that special protection. So in a bank run they will end up with illiquid assets, just like the other creditors of a defaulting company. But the losses only come because creditors are all selling those assets at the same time, not because the bank didn't have enough assets in the first place.

There is something hard about bitcoin that the dollar doesn't have : the certitude that there are 21 million BTCs and that it won't ever go higher than that number. Your bitcoins do not disappear if the bank goes bankrupt, if the country goes bankrupt, if someone decides to print money because God said to do it.

People care about the value of currency units, not their number. If people stop using bitcoins, their values drops to nothing. The fact that there are only 21 million has no influence on that at all. For national currencies, the number of people using the currency is relatively constant, namely everybody in thtr country, and the use they make of it is fairly constant too, since the same currency use it for almost all of their monetary affairs. Under those circumstances, changes in the amount of currency in circulation can be one of the main drivers of changes in value. You can't extrapolate that situation to a private currency at all

[creighto]

What are your thoughts ? What are its weaknesses ? How can it fail ? How would you attack it if you were a criminal organization, a banking institution or a government?

Disclaimer: I've a personal interest in the success of bitcoins, now that I have quite a few. However, I spent months observing and learning about the system before jumping into it. I am not a programmer or a cryptologist myself; but I am a praxeologist, and this thing is (IMHO) very elegant.

In the early stages, Bitcoin could have been easily destroyed by a malicious party with much computational power attacking the blockchain. As time goes on, and the difficulty level continues to increase as a direct result of new players adding compuational power to the collective, such an attack becomes less and less viable. Not impossible, mind you, but astronomicly unlikely. The present collective power of the Bitcoin P2P network is already competitive in the world supercomputer class.

The currency itself cannot fail in the manner that others on this list have proposed. It would be beyond a forum post to explain why, but the short answer is that Bitcoin is not a form of currency that be compared to current fiat currency regimes that people are familiar with. It is different from fiat, as well as any commodity standard currency such as gold or silver. It is designed so that both the inflationary process that creates new currency, and it's predictable future monetary base, mimics the best characteristics of a gold standard. That said, it's unlike a gold standard as well. The closest example of a monetary system that Bitcoin could be compared to would be a LETS system (http://en.wikipedia.org/wiki/Local_Exchange_Trading_Systems) wherein the 'locality' it was designed for is the whole of the Internet. However, even that comparison fails, because a LETS is based on mutual credit, and all credits in the system balance out to zero, and do not have a natural market value. As such, all LETS systems are 'pegged' to some other value, such as hours of unskilled labor (Ithica Hours) or to the national fiat currency. Bitcoin, however, maintains a known monetary base of a positive value that is not pegged to anything whatever, and therefore floats in value. Bitcoin is an altogether new monetary model, so comparisons to any past or present monetary systems, or their flaws, is fraught with error.

The only weakness for outright destruction of the system by malicious activities that I have encoutered and found credible, is the complete shutdown of the Internet. And I do mean *complete*, for if sections are simple cut off from one another for a period of time, this will only result in a blockchain 'split'; which is a self-correcting event for Bitcoin.

As for the issue that a bitcoin can drop in value. Everything fluctuates in value. This isn't really an issue in the long run. In the short run, don't buy bitcoins on a rally.

BTW, it's not true that Bitcoins are 'deflationary', or even of a fixed number. At least not yet, as the 21 million mark (which is actually 2,100,000,000,000,000 as the client hides the real number from the user by adding a decimal point at the eighth digit for readibility) is a limit that can never be reached, and won't even come close for 120 years. Presently, bitcoins are *very* inflationary, running just under 50% APR. This will drop off quickly in the next few years, dropping to 6.5% APR roughly around Jan 2013 and then below 2% APR by 2020; but calling it 'deflationary' is wildly inaccurate.

[Iv]

Zamfir, please don't get me wrong, but what is the point of your explanation ? I agree with most of what you said, but I don't see how this is related to bitcoin. i know that banks do have some assets, and that it is a fraction of what they owe their creditors. The figure I had in mind is 20% of assets, so that means that 80% of the value comes from trust. I am of the opinion that this is a dangerous situation because most people believe there is a higher part in real assets and are very unwilling to trust bankers, but this political opinion is largely irrelevant to the discussion (If you wish we can start another one, more politically oriented).

I still fail to see why the example of the current system makes it impossible to have a 100% trust-based currency. By "trust", I mean "trust that this currency will be accepted as a payment method". Regular electronic money systems require an additional trust toward the bank, that is, the trust that the bank will not create money to accounts it likes, that it will keep an accurate count of the money I have and that it will do its best to maintain the currency at a level that is generally beneficial to the economy. Replacing the trust-toward-the-bank by a neat cryptographic algorithm seems completely feasible and the 100% trust based currency already exists today if I understand your Cayman islands example.

So I fail to see where this explains a flaw of bitcoin.

[creighto]

One problem may be that it was not created by an economist, i.e. it was created by someone who doesn't really understand what a central bank does, or that a loan from a (non-government) bank is actually something useful, even though it means that money is being created.

Interesting statement. Possiblely false. The creator of Bitcoin, Satoshi Nakamoto, is a very private person. It is a common, perhaps a majority, opinion on the official Bitcoin forum that that name is not of a real person; but if it is the name of the true creator of Bitcoin, then he is very good at hiding his identity online. More likely, it is either a false name chosen to hide the creator's real identity or it is a false name created to hide the real creators' identities. Considering the varied expertise required to have created Bitcoin in it's present form, largely without input from the open source community, I consider the latter to be more likely. In that condition, there is most likely at least one professionally trained economist involved in it's development. I personally know that Bitcoin is not unknown among economists, but none are yet willing to place their professional reputations upon supporting a beta level attempt at a cryptocurrency. I suspect that several semi-famous Austrian economists are watching developments closely; but since Bitcoin does permit the careful to hide their identities pretty well, there is no way to confirm this.

However, if one individual came up with Bitcoin, then s/he is a true polymath.

[iop]

BTW, it's not true that Bitcoins are 'deflationary', or even of a fixed number. At least not yet, as the 21 million mark (which is actually 2,100,000,000,000,000 as the client hides the real number from the user by adding a decimal point at the eighth digit for readibility) is a limit that can never be reached, and won't even come close for 120 years. Presently, bitcoins are *very* inflationary, running just under 50% APR. This will drop off quickly in the next few years, dropping to 6.5% APR roughly around Jan 2013 and then below 2% APR by 2020; but calling it 'deflationary' is wildly inaccurate.

Somehow I misunderstood - I had thought that the limit will almost be reached in 20 years. However, this doesn't change anything about the main problems.

Whenever the economy shrinks relative to the bitcoin there is inflation (e.g. because the pool of bitcoins grows faster than the economy, or because the economy contracts, for example after a natural disaster), and whenever the economy grows, there will be deflation. In other words, there is very little price stability with the bitcoin, unlike what we have right now. Keeping prices stable by adapting the pool of money to the size of the economy is one of the most important roles of a modern central bank. Why is price stability important? Because it lets you plan ahead, which is useful if you want to run an efficient economy.

I went into this discussion assuming that an efficient, and growing, economy is something that we'd like to keep. Why should we? For one, technological advance has made us a lot more productive. This means that producing the goods that people used to use now takes a lot less manpower. In other words, if we need fewer people to produce the same (or more!) amount of goods, and there are more people, we need to have a growing economy in order to give jobs to everyone. I like to have a job. However, not everybody needs a job as much as I do, so to them, a stagnating, or shrinking economy is not so bad, and would not be considered a possible failure of bitcoin economy.

Do you agree that deflation curbs investment?

I am out for opinions on this. From what I understand, a scenario with a 0.1% deflation is the same as 3% inflation where there are 3.1% interest rates on saving accounts.

There is one important difference: In the inflation scenario, your money needs to be in the bank, which allows the bank to give out loans. In the deflation scenario, your money is under your mattress (or on your hard drive), which does not allow to create loans.

The big investors, by the way, are organisms like retirement funds, or insurance companies - entities that need to make sure that they make the most out of the money they're storing. They will invest in whatever gives the best returns with an acceptable risk. In a deflationary economy, they will, of course, not invest anything. But even with a bit of inflation the amount of bitcoin is severely limited - thus there will not be many investments, and certainly not too many long-term ones, and they need to promise very good returns. Forget about a new hydro-power plant, for example. A startup company is unlikely to give good returns given the risk, of course (so whoever wants to start their company have to be more patient so that they and their friends have the time to save up all that money), and a person who needs a loan, well, sucks to be them unless they pay outrageous interest. Currently, many loans go to existing companies: As long as the expected return on investment is above the interest rate, it's worth putting money into a company, and if the interest on the loan is smaller than the expected return, then a company will benefit from taking a loan in order to be able to get that new plant up faster. If they can't get these loans (because there is just fewer bitcoin to go around), they'll grow more slowly, and that is what I won't like. YMMV, of course.

So in a bitcoin economy, I cannot promise to pay you tomorrow? And I cannot get a loan from you? Because that's how you get into debt. Incidentially, that's basically what Greece did. They just accrued too much debt given their meager tax income and their bleak growth prospects.

Ok, that's right, you can get a loan and you can get some debts but what you can't have is loans in the way banks, central banks and governments practice them : by giving you a drawing right at the ATM that can exceed what they actually own. That is what we are talking when we talk about badly managed national debt.

Huh? Which country spends money they don't have? If countries need money, they get a loan (by issuing bonds). The interest rate they have to pay on this depends on their credit. Basically, a country can get more loans at a decent interest rate if the market believes that their net income (tax minus spending) will grow, either through a growing economy (see a trend here?) or through cutting expenses (which is what Greece and Portugal have to do right now, because no one believes in their economy much). If they lie about their finances (like Greece), or if they let spending go out of hand while cutting taxes, then they will find themselves in a situation where they have been beyond their means and then they fail.
The only thing bitcoin can possibly change here is that if there is much less investment overall, countries have a much harder time to get loans, and thus they cannot run as much debt as they can now. Everything else that pushed e.g. Greece over the edge has nothing to do with fiat currencies - except that it would have been much less bad if Greece hadn't been in a quasi-bitcoin situation in that they weren't able to devalue their currency.


To summarize: If you don't like a growing economy, support bitcoin. If you don't care about price stability, support bitcoin. Don't hope countries, or companies, or individuals, can no longer go bankrupt, because bad financial management is possible with any monetary system that allows loans.