Deflation and Bitcoin, the last word on this forum

[JoelKatz]

However, in order for this instrument to be treated as a substitute of Bitcoin, it needs to provide identical, or better utility.

Whatever its utility, there will be some price point at which many people are ambivalent to whether they have X Bitcoins or the instrument.

Let's say that there is a high demand for iPhones. This demand probably cannot be satisfied by clay bricks cut into same dimensions as an iPhone. Merely because they have the same size, it does not follow that from economic point of view, they act as substitutes. Similarly, merely because a financial instrument is denominated in Bitcoins, it does not follow that its supply satisfies the demand for Bitcoins, because they are not necessarily substitutes from economic point of view.

Right, but in principle, anything with the same value can act as a substitute for that currency. There isn't, however, universal substitution because there's friction. If you have a currency where demand exceeds supply, you will quickly overcome that friction.

The most likely scenario, based on the technologically publicly known to exist to day, is alternate crypto-currencies. Perhaps including partially-centralized ones backed by something (including things like gold or fiat currency). Say you owe me 100 Bitcoins, but Bitcoins are hard to get due to scarcity. There will be some number of DollarCoins, GoldCoins, or whatever, that I'll be willing to take in exchange for those 100 Bitcoins and some number of those coins you'd be willing to pay me. The more scarce a currency is, the more incentive we have to make a deal in some other currency, and that reduces the effective demand (or alternatively, increases the effective supply).

[1714915089]

1714915089

[1714915089]

1714915089

[1714915089]

1714915089

[lonelyminer (Peter Šurda)]

Whatever its utility, there will be some price point at which many people are ambivalent to whether they have X Bitcoins or the instrument.

First of all, I humbly submit that this is a contradiction. If one is indifferent to possession of two goods, they by definition have the same utility for him. If you're trying to say that if a good gets more expensive, it becomes more profitable to manufacture substitutes, this is true, but again, this does not necessarily mean that it this makes it more likely for any good to be treated as a substitute. In particular if the good is well divisible it does not sound likely. Money has an interesting feature, that within reasonable limits (e.g. divisibility, mass, volume), any amount is optimal.

Right, but in principle, anything with the same value can act as a substitute for that currency.

Hold on, you're skipping over the finer points. It isn't true that anything with the same value can act as a substitute for that currency. First of all, there must be a certain level of trust that the exchange ratio of the wannabe substitute will remain stable for the foreseeable future. Otherwise, the wannabe substitute will only be accepted in exchange below par, or not at all. Second of all, the wannabe substitute must have reasonable transaction costs. It cannot have the form, say, of sacks of grain. If you wanted to use a sack of grain as a substitute for one Bitcoin, noone would accept it. These requirements can be, to a certain extent, suppressed by the use of force (= government), but it only works as far. If the confidence fails or the transaction costs spike, the system collapses.

As I tried to explain, with form-specific monetary base (e.g. current fiat currencies or gold), both of these are easy to achieve. With Bitcoin, it's much more difficult. Not only because it's difficult to match the transaction costs, but also because there are other highly liquid instruments which are nevertheless not media of exchange. Let's say that I have some BTC-denominated shares, and am short of BTC when I want to pay. Hypothetically, I can offer the shares in exchange. However, it's more likely that I'll just sell the shares at market price and transfer the bitcoins gained from this transaction when paying. In other words, there would be no shortage even if the amount of media of exchange does not inflate.

There isn't, however, universal substitution because there's friction. If you have a currency where demand exceeds supply, you will quickly overcome that friction.

I humbly disagree. This neglects the existence of highly liquid instruments which are not a medium of exchange. The demand for liquidity can be satisfied by these, it is not necessary that the amount of media of exchange increases.

The most likely scenario, based on the technologically publicly known to exist to day, is alternate crypto-currencies. Perhaps including partially-centralized ones backed by something (including things like gold or fiat currency). Say you owe me 100 Bitcoins, but Bitcoins are hard to get due to scarcity. There will be some number of DollarCoins, GoldCoins, or whatever, that I'll be willing to take in exchange for those 100 Bitcoins and some number of those coins you'd be willing to pay me. The more scarce a currency is, the more incentive we have to make a deal in some other currency, and that reduces the effective demand (or alternatively, increases the effective supply).

You are conflating scarcity and liquidity. The concept of a currency being "more scarce" or "less scarce" makes no economic sense. What is sought is liquidity. This can be achieved by selling assets (more liquid ones will be more likely to be sold) or by taking loans (presumably, against interest). If Bitcoin is a dominant currency and for some reason people decide to hoard all their bitcoins, the price of goods will fall and the interest rate will rise, making it more attractive to spend or invest, so the prices and interest rate will move towards equilibrium.

[steelhouse]

Consider a society of bitcoin users, say 1000. If bitcoin was the only money, you would need to use some for cost of living, entertainment, and travel.  If the price of bitcoin rises with deflation, the value of your bitcoin stash increases.  As result you will have more money to spend.  Nobody hoards in safe investments or currency for the fun of it.  They do so to buy something in the future or for emergencies.  Spending will increase.  Eventually all the spending of society equals all the savings or hoarding of society.  When that happens, the deflation and inflation stop. 

Suppose somebody dumps bitcoin and the price drops.  People panic and they sell their bitcoin.  Maybe to buy factories or gold.  Eventually the price will drop to say $1.  However, the history of the bitcoin will give the coin value.  It will reach a point where the panic stops.  Then rebuild.  It is time and number of users that will prevent the complete collapse of bitcoin.

As the number of users increases and use increases, the value will increase.  Next year the number of users might be 2000.  This should basically double the value of bitcoin since they are limited.

[jtimon]

Do we agree that deflation can make it harder for capital to be profitable?

Yes, but you have the cause and effect precisely backwards.

Not sure how we agree then.

You're assuming you can hold the monetary system constant as you imagine various changes in economic productivity. But the monetary system responds to the economy.

AND the other way around. The economy responds to the monetary system. What we're discussing is how price deflation (an indicator in the monetary system) affects investments and therefore the economy. And most people here are saying "Deflation doesn't affect real investments negatively at all, if anything, deflation is good for the economy". If I try to make the examples with "two parallel and identical economies but only changing the deflation rate" is to make things simple. Not that I actually think that is possible. I know that the economy is a complex system with no isolated parts. I know that if the two economies were actually identical, they would present an identical deflation rate. The question still is, does deflation discourage the production of new real capital (production goods)?
If so, is it only as the expense of a shift to consumption?
If so, how is that shift to consumption justified?

For example, Bitcoins can become scarce because of their built-in scarcity. But they can also become scarce because demand goes up due to economic productivity. And either way, the result will be in increased effective supply of Bitcoins in the form of things like demand notes, mortgages, or bonds circulating as currency.

Are you saying that deflation will be naturally postponed while the system accepts more credit/debt that can act as a substitute? I didn't had the courage to say something similar myself, but actually I think that the credit cycles are caused by the basic interest making the credit substitutes grow exponentially. Back to the claim I think you're making...
If deflation is postponed while the system accepts more credit/debt, eventually the credit won't be able to grow more and there you have the cycle bust. With the "not anymore postpone-able" deflation accelerating the unavoidable credit destruction process. Until the level of debt is "sustainable" again.

What will happen, unless you imagine a coercive government mucking with the process (say by printing currency, borrowing endlessly, or strong arming interest rates), is that the economy will drive the currency to encourage sensible investment, or non-investment, as makes sense.

You seem to be saying that if resources are not going to investment, they have to be going to consumption for a good reason.
What concerns me is not that "resources are spent on consumption" or "resources are spent on investment". Is "resources aren't spent on consumption nor on investment". Because there are resources like food and labor that happen to irreversibly disappear if not used at a certain time.

Take into account that I'm not attacking here the free market. I attribute these problems to the gold standard, for example. Which is not an intrinsic part of the free market, a free monetary market is. Bitcoin as an only currency would suffer the same problems, but I don't fear a "bitcoin standard" to be enforced. I also think that the Keynesian solution of government spending is just a bad patch that creates more problems without really solving anything, only postponing things. But yet they acknowledge there's a problem to be solved.

The point is that if the price of the money in that community is rising and is expected to continue to do so (probably you need collapsing credit/debt for that), people will prefer to keep the money over buying the closed factories and put them to work again. But, yes, at some point (when the deflation has destroyed enough credit/debt) that will stop, the financial market will get restored, real investments will resume and the factories will be reopened.

I don't think deflation is the problem per se, it's just one of the symptoms and the natural (but destructive) solution and unavoidable  conclusion (well, there's also hyperinflation to the debt jubilee by destruction of the currency, but that doesn't deserve to be called a solution) of credit/debt cycles.

[JoelKatz]

Do we agree that deflation can make it harder for capital to be profitable?

Yes, but you have the cause and effect precisely backwards.

Not sure how we agree then.

Maybe, maybe not. It's hard to say. In a free market, deflation is a sign that it is hard for capital to be profitable. It's not that deflation makes it hard for capital to be profitable, it's that these two things go together. If some external force compelled deflation, such as Bitcoin's inherent monetary deflation, a free market will respond by creating alternatives to Bitcoins that serve the same purpose, assuming it is free to do so. The net result will be that the *effective* deflation will accurately reflect how hard it is for capital to be profitable.

You seem to be saying that if resources are not going to investment, they have to be going to consumption for a good reason.
What concerns me is not that "resources are spent on consumption" or "resources are spent on investment". Is "resources aren't spent on consumption nor on investment". Because there are resources like food and labor that happen to irreversibly disappear if not used at a certain time.

Sure, but that will only happen if someone forces the market to do that or if the market cannot avoid it. One force, like the existence of a scarce currency even if it's the primary currency simply can't do that on its own. If Wendy's can't sell enough hamburgers, the result won't be $100 hamburgers -- it'll be more Burger King's or more chicken sandwiches.

Take into account that I'm not attacking here the free market. I attribute these problems to the gold standard, for example. Which is not an intrinsic part of the free market, a free monetary market is. Bitcoin as an only currency would suffer the same problems, but I don't fear a "bitcoin standard" to be enforced. I also think that the Keynesian solution of government spending is just a bad patch that creates more problems without really solving anything, only postponing things. But yet they acknowledge there's a problem to be solved.

I agree. If a coercive force compelled a gold standard, it would be about as bad as compelling a fiat standard. You wouldn't have the evils of unchecked spending causing inflation, but you'd still have the problem that you'd need centralized control over monetary policy with all the mistakes central planners inevitably make.

If you imagine a Bitcoin-based world where other currencies are outlawed or heavily regulated, you would have the same problem. The Bitcoin built-in deflation would act as a retarded central planner, forcing a monetary policy that is very unlikely to make sense. But so long as that doesn't happen, there's no problem with Bitcoin's built-in deflation. If the currency becomes too scarce, the market will introduce alternatives to the currency that serve the same role unless some force prevents it from doing so.

The point is that if the price of the money in that community is rising and is expected to continue to do so (probably you need collapsing credit/debt for that), people will prefer to keep the money over buying the closed factories and put them to work again. But, yes, at some point (when the deflation has destroyed enough credit/debt) that will stop, the financial market will get restored, real investments will resume and the factories will be reopened.

Right, so the question is -- why would a free market allow the price of money to rise when it's trivial to introduce new currencies or use debt instruments as currencies? And the answer is -- if the price *should* rise and thus there's no efficiencies to be gained (and thus no profit) in stopping it from rising.

[lonelyminer (Peter Šurda)]

Maybe, maybe not. It's hard to say. In a free market, deflation is a sign that it is hard for capital to be profitable. It's not that deflation makes it hard for capital to be profitable, it's that these two things go together. If some external force compelled deflation, such as Bitcoin's inherent monetary deflation, a free market will respond by creating alternatives to Bitcoins that serve the same purpose, assuming it is free to do so. The net result will be that the *effective* deflation will accurately reflect how hard it is for capital to be profitable.

I still don't see why this should be the consequence. I find it more likely that the result will be adaptation of the production structure and the change in the nominal rate of interest. The best description of how this works I found in deSoto's Money, Bank Credit and Economic Cycles. If I interpret it correctly, as long as the money supply does not shrink, there is always an equilibrium which fits the time preference of economic actors. Selgin argues that however that this alone is not sufficient for people to hold the optimum amount of medium of exchange, but that's a different variable and the Mises/Rothbardian branch probably frowns upon arguments like this.

You basically argue that if a medium of exchange has an appreciating value, this would make it more likely for people to choose something else as a medium of exchange. Regrettably, I cannot see why it should be so. If anything, it should be exactly the opposite: the more the value appreciates, the more will people demand it, and the more difficult it would be to get someone to accept something else in exchange (ignoring transaction costs and network effect). I can imagine that a fixed exchange ratio (aka Gresham's law) would have the effect you describe, but this can only be done by force (=state). On a free market, people would adjust their behaviour based on the expectation of exchange rate changes, for example, they might request a higher amount of the alternative medium of exchange to compensate for the price difference between the time they accept it and the time they expect to spend it, or as the economists say, the other medium of exchange would trade at a discount.

If they did not, this would create an opportunity for arbitrage. In particular with media of exchange with low transaction costs and highly liquid markets, such as Bitcoin and similar, the arbitrage is swift and uncompromising. And if it's possible to trade derivatives (e.g. options), I see practically zero chance the disequilibrium would stay for long. This would compensate for the appreciation effect. So even if you were correct in the assumption of people's choice, it still would work differently due to arbitrage. Even if people tended to make a particular choice contrary to my assessment, it still does not follow that an arbitrageur would forego a profit opportunity.

[JoelKatz]

You basically argue that if a medium of exchange has an appreciating value, this would make it more likely for people to choose something else as a medium of exchange. Regrettably, I cannot see why it should be so. If anything, it should be exactly the opposite: the more the value appreciates, the more will people demand it, and the more difficult it would be to get someone to accept something else in exchange (ignoring transaction costs and network effect).

I believe your argument is essentially: The more desirable the currency is to hold, the more people will want to acquire it. Thus the more they will tend to negotiate to get the currency in transactions. Thus increasing the demand for that currency relative to other possible ways value could be exchanged.

But if that was correct, so would this argument be: The more the currency appreciates, the less those who have it wish to part with it, so the more likely they'll negotiate to use something other than the currency in their transactions. This will reduce the demand for the currency because fewer transactions will use it.

Since they lead to opposite conclusions and are both equally justified, the justification for both must be inadequate.

Whatever currency might do in the future, people will tend to agree that X units of currency have, as currency, value Y (taking into account the future value, risk, fungibility, and so on). They will, in general, be indifferent to receiving Y value in that particular currency or in any other form.

If it costs me $20 to sell a goat, all things considered, and I can sell that goat for $60. Then I don't care whether you give me a goat or $40. Normally, we'll use the money because it's just easier. But if money is super ridiculously scarce for some reason, there is no rational reason for us to use money in preference to a goat. Naturally, this goat would have a very low efficiency because of your costs to transfer it to me and my costs to sell it. So more likely we'd use something more efficient, like gold, Bitcoins, or whatever.

[lonelyminer (Peter Šurda)]

But if that was correct, so would this argument be: The more the currency appreciates, the less those who have it wish to part with it, so the more likely they'll negotiate to use something other than the currency in their transactions. This will reduce the demand for the currency because fewer transactions will use it.

I would say that it's the person accepting payment that plays a more significant role in the decision making process. The person paying must already have particular media of exchange in his possession (or at least have a way of purchasing them), which limits his choices, but the person accepting the payment is not restricted this way. And if forex markets are highly liquid and have low fees, then it's possible that this effect is eliminated altogether, because it's possible for the payer and payee to use a different medium of exchange (just like you can use an euro-denominated debit card to pay someone who wants dollars, or how Bit-Pay allows merchants to receive something else than Bitcoin). However, I am not entirely sure this hybrid solution affects the money supply.

Since they lead to opposite conclusions and are both equally justified, the justification for both must be inadequate.

Of course it's inadequate, it neglects transaction costs and network effect.

Whatever currency might do in the future, people will tend to agree that X units of currency have, as currency, value Y (taking into account the future value, risk, fungibility, and so on). They will, in general, be indifferent to receiving Y value in that particular currency or in any other form.

Well, it turns out you're neglecting the transaction costs and the network effect (also the discounting procedure I described in a former post). If we for argument's sake assume that people are indifferent to receiving a particular value in any form or denomination, the transaction costs associated with each of them are not equal and influence the decision. The network effect favours incumbents. And the discounting procedure means that people might expect a different value of a different medium of exchange depending on the expected exchange ratio development and the duration they expect to hold it.

If it costs me $20 to sell a goat, all things considered, and I can sell that goat for $60. Then I don't care whether you give me a goat or $40. Normally, we'll use the money because it's just easier. But if money is super ridiculously scarce for some reason, there is no rational reason for us to use money in preference to a goat. Naturally, this goat would have a very low efficiency because of your costs to transfer it to me and my costs to sell it. So more likely we'd use something more efficient, like gold, Bitcoins, or whatever.

You would be more likely to accept something that you can use again in payment, and something that reasonably holds value. It's easier to pay for a random thing using dollars than goats. Also the goat needs feeding and will eventually die (or can breed). This creates an order of preference among the media of exchange even if people would really be indifferent to the current value. Also, the concept of money being more or less scarce makes no economic sense to me. It's just a metaphor.

[JoelKatz]

You would be more likely to accept something that you can use again in payment, and something that reasonably holds value. It's easier to pay for a random thing using dollars than goats. Also the goat needs feeding and will eventually die (or can breed). This creates an order of preference among the media of exchange even if people would really be indifferent to the current value. Also, the concept of money being more or less scarce makes no economic sense to me. It's just a metaphor.

I don't think it makes sense to consider the preference independent of the value. Not preferring something is the same as valuing it less.

If a currency has an unusually high or unusually low value to just one party in a transaction, that makes it a lousy medium of exchange. If it has abnormally high value to the seller, the buyer won't be willing to part with enough of it to make it worth the seller's while. If it has abnormally high value to the buyer, the buyer won't be willing to part with enough of it to make it worth the buyer's while. Either way, the buyer and seller will be more likely to make the deal in another currency. So, if you think that inflation or deflation change the relative value of a currency to the seller and the buyer, then a currency that is either inflating or deflating for reasons internal to that currency (and not due to market conditions generally) will tend to become less efficient as a medium of exchange and will lose out to more stable currencies.

If we had a Bitcoin-based currency and the built in deflation made the currency scarce, the solution would simply be to supplement the currency with other currencies with comparable characteristics or to use credit instruments as currency. So long as these instruments had roughly equal value to both the buyer and the seller, rational buyers and sellers would be indifferent to the form in which they were paid. All that matters is that I can acquire something you value at X for not much more than X and you get not much less than X back for it. The closer these three values are to each other, the more useful the currency.

[lonelyminer (Peter Šurda)]

[I don't think it makes sense to consider the preference independent of the value. Not preferring something is the same as valuing it less.

My apologies for the imprecise formulation. I was attempting to point out that you omitted important factors in your model.

If a currency has an unusually high or unusually low value to just one party in a transaction, that makes it a lousy medium of exchange. If it has abnormally high value to the seller, the buyer won't be willing to part with enough of it to make it worth the seller's while. If it has abnormally high value to the buyer, the buyer won't be willing to part with enough of it to make it worth the buyer's while. Either way, the buyer and seller will be more likely to make the deal in another currency. So, if you think that inflation or deflation change the relative value of a currency to the seller and the buyer, then a currency that is either inflating or deflating for reasons internal to that currency (and not due to market conditions generally) will tend to become less efficient as a medium of exchange and will lose out to more stable currencies.

If the expected exchange rate change was the reason for the decision to use a particular medium of exchange, discounting would compensate against it. And on a free market this is the expected behaviour.

If we had a Bitcoin-based currency and the built in deflation made the currency scarce, the solution would simply be to supplement the currency with other currencies with comparable characteristics or to use credit instruments as currency.

First of all, let me repeat, the concept of money being more scarce or less scarce makes no sense. There can, in principle, be specific deficiencies, such as lack of coins or notes with a particular denomination, or a shrinking of the money supply, or regulation could inhibit the flow of capital or changes in prices. Hopefully we can agree that this is not what we are talking about.

If you want to pay with Bitcoins and do not have any, you have following options:

• attempt to use some other liquid asset in payment
• sell another liquid asset (to a third party) for Bitcoins and use Bitcoins to pay
• attempt to use an IOU

Which one of these is more likely depends on several factors. For example, network effect would favour 2. However, transaction costs might outweigh the network effect. Or, you might not have a sufficient amount liquid assets, which also precludes 1. Or, if IOUs already act as substitutes, that favours 3. The alleged factor of "scarcity" has no influence on the decision (omitting the situations excluded above).

Now, it could happen that people decide to consume less as such (irrespective of the medium of exchange). This is a decrease of the time preference, and not an increase in the scarcity of money. To attempt to compensate this with an increase of the money supply would go counter to the preferences of the users of the medium of exchange and cause disruptions in the coordination of plans between producers and consumers.

So long as these instruments had roughly equal value to both the buyer and the seller, rational buyers and sellers would be indifferent to the form in which they were paid.

I already explained that you omit other factors that influence the decision.

All that matters is that I can acquire something you value at X for not much more than X and you get not much less than X back for it. The closer these three values are to each other, the more useful the currency.

You also omit the possibility that this "something" is traded at specialised markets instead of being used as a medium of exchange. For example, instead of paying with Namecoins, I can sell Namecoins against Bitcoins and use Bitcoins for payment. If I have Namecoins but no Bitcoins, it does not necessarily make it more likely that Namecoins are used as a medium of exchange as a substitute for Bitcoin. This raises further questions with respect to the assumption that there can be a "scarcity" of a currency.

[johnyj]

Let's say that I have some BTC-denominated shares, and am short of BTC when I want to pay.

Oh my god, I see a brilliant idea here: BTC-denominated shares!

With enough amount of BTC reserve, a BTC Bank could issue their own BTC notes, thus create multiple supply of exchange medium...

On second thought... isn't this back to the old time that gold/silver were so scare that bankers produced their notes based on the amount of gold they have in their reserve?

[lonelyminer (Peter Šurda)]

With enough amount of BTC reserve, a BTC Bank could issue their own BTC notes, thus create multiple supply of exchange medium...

I already addressed this in the past, e.g in the Bitcoin wiki article on FRB. Bitcoin is form-invariant, any medium that can store 64 bytes of data (=ECDSA keypair) can be used as Bitcoin. Since "BTC denominated notes" do not provide an advantage on transaction costs (unlike with systems based on gold, for example, because gold is not form-invariant and cannot exist in paper or electronic form), it would not be generally accepted as a medium of exchange. Similarly as sacks of grain in the weight of 5.5 tons would not be generally accepted as suitable alternative payment method instead of an ounce of gold, even if their market price might be the same (I roughly calculated it based on the current prices).

On second thought... isn't this back to the old time that gold/silver were so scare that bankers produced their notes based on the amount of gold they have in their reserve?

Bankers didn't produce notes because gold/silver was scarce, but because the notes provided an increased benefit for the depositor (e.g. safer storage, smaller weight, etc), in other words they reduced their transaction costs. FRB came after the notes were already accepted as a medium of exchange and I find it doubtful that it can occur the other way around.

EDIT:
Based on what I read, historically, deposit accounts predated bank notes, but the situation is analogous to notes. Deposit accounts offer a ledger and other services, and with gold, this decreases transaction costs. With Bitcoin, the ledger is an integrated component and the other services are either already present or obsolete.

[JoelKatz]

I already addressed this in the past, e.g in the Bitcoin wiki article on FRB. Bitcoin is form-invariant, any medium that can store 64 bytes of data (=ECDSA keypair) can be used as Bitcoin. Since "BTC denominated notes" do not provide an advantage on transaction costs (unlike with systems based on gold, for example, because gold is not form-invariant and cannot exist in paper or electronic form), it would not be generally accepted as a medium of exchange. Similarly as sacks of grain in the weight of 5.5 tons would not be generally accepted as suitable alternative payment method instead of an ounce of gold, even if their market price might be the same (I roughly calculated it based on the current prices).

So long as they don't have a disadvantage, why do they need to provide an advantage to be competitive? One can imagine any number of advantages they could provide including providing interest by means of fractional reserve, insurance against theft, or anything else the free market can dream up. Sure, if Bitcoins work perfectly, there will be no reason to try to compete with them. But if they don't, there will be.

[lonelyminer (Peter Šurda)]

So long as they don't have a disadvantage, why do they need to provide an advantage to be competitive?

For example due to the network effect, the costs of having them redeemed, counterparty risk (with debt instruments) / unforeseen exchange rate fluctuations (with alternative currencies).

One can imagine any number of advantages they could provide including providing interest by means of fractional reserve, insurance against theft, or anything else the free market can dream up. Sure, if Bitcoins work perfectly, there will be no reason to try to compete with them. But if they don't, there will be.

Since Bitcoin is form-invariant, all of the needs that come up are hypothetically possible natively, without having to create a new instrument or another currency. As for practical examples, Bitcoinica provides interest on deposits and on demand redemption without FRB (well, technically they do overissue, but the derivatives they create are not usable as a method of payment, and there is no maturity mismatch unlike FRB). Theft protection can be implemented with split-keys. Also, none of these particular needs you mentioned create a demand for a new method of payment (unlike, for example, transaction costs, which do create a demand for a new payment method).

I consider it theoretically possible that some feature cannot be implement efficiently and this would overcome the network effect. This, indeed, might create a demand for an alternative. Whether that would be a new cryptocurrency or a Bitcoin derivative depends on empirical parameters of the inefficiency and the alternative. I just don't think that it's a very likely scenario, more likely in that case Bitcoin would be replaced completely. Also I'm ignoring the effect or state interference and regulations, these could trigger Gresham's law.

However I'd like to stress that I don't see why the concept of "Bitcoin becoming scarce" should play a role in the acceptance of the alternatives as a method of payment.

[JoelKatz]

However I'd like to stress that I don't see why the concept of "Bitcoin becoming scarce" should play a role in the acceptance of the alternatives as a method of payment.

Suppose currency scarcity creates some kind of problem, such as discouraging investment. If you don't think currency scarcity creates any issues, then you won't see any point in trying to "work around it", but neither will you see any need to do so. But if you think quirks of the monetary system will destroy investment opportunities that would otherwise be profitable, then there will be a strong incentive to develop other payment schemes and induce people to accept them as a way of opening up those profit opportunities.

[lonelyminer (Peter Šurda)]

Suppose currency scarcity creates some kind of problem, such as discouraging investment.

As I argued earlier, the concept of "currency scarcity" makes no sense.

If you don't think currency scarcity creates any issues, then you won't see any point in trying to "work around it", but neither will you see any need to do so. But if you think quirks of the monetary system will destroy investment opportunities that would otherwise be profitable, then there will be a strong incentive to develop other payment schemes and induce people to accept them as a way of opening up those profit opportunities.

You imply that "currency scarcity" (which you do not define) can "destroy investment opportunities". As I already explained elsewhere, the changes in the amount of money in the the economy do not influence the amount of "investment opportunities". The only thing they do is that they make some opportunities appear more attractive, and some less. But it's purely an issue of changes in interpretation and evaluation. No investment opportunity is destroyed, it merely appears unattractive. However, this must logically be compensated by a different opportunity becoming more attractive. Furthermore, it can mislead consumers and producers by screwing up with economic calculation, for example the market interest rate and natural rate of interest diverge. So they would think they are investing into a profitable business and will forego an investment into what appears to them to be a non-profitable business, however once the profit does not materialise, the error in calculation will be revealed and the investment would need to be liquidated.

Furthermore, even if you were correct that "currency scarcity destroys investment opportunities", it still does not follow that the market will compensate against this with an increase in the money supply, which is, as I explained, guided by different factors. If the money chosen by the market actors is resistant to increases of the supply, the only way to fix this would be to force them to use a different money.

[johnyj]

With enough amount of BTC reserve, a BTC Bank could issue their own BTC notes, thus create multiple supply of exchange medium...

I already addressed this in the past, e.g in the Bitcoin wiki article on FRB. Bitcoin is form-invariant, any medium that can store 64 bytes of data (=ECDSA keypair) can be used as Bitcoin. Since "BTC denominated notes" do not provide an advantage on transaction costs (unlike with systems based on gold, for example, because gold is not form-invariant and cannot exist in paper or electronic form), it would not be generally accepted as a medium of exchange. Similarly as sacks of grain in the weight of 5.5 tons would not be generally accepted as suitable alternative payment method instead of an ounce of gold, even if their market price might be the same (I roughly calculated it based on the current prices).

On second thought... isn't this back to the old time that gold/silver were so scare that bankers produced their notes based on the amount of gold they have in their reserve?

Bankers didn't produce notes because gold/silver was scarce, but because the notes provided an increased benefit for the depositor (e.g. safer storage, smaller weight, etc), in other words they reduced their transaction costs. FRB came after the notes were already accepted as a medium of exchange and I find it doubtful that it can occur the other way around.

EDIT:
Based on what I read, historically, deposit accounts predated bank notes, but the situation is analogous to notes. Deposit accounts offer a ledger and other services, and with gold, this decreases transaction costs. With Bitcoin, the ledger is an integrated component and the other services are either already present or obsolete.

No matter the form, the scarcity itself will make BTC more possibly to become the target of mortgage, and then many loan based on BTC will be issued, these loans will require increased currency supply, it's the same effect as FRB

[jtimon]

However I'd like to stress that I don't see why the concept of "Bitcoin becoming scarce" should play a role in the acceptance of the alternatives as a method of payment.

Suppose currency scarcity creates some kind of problem, such as discouraging investment. If you don't think currency scarcity creates any issues, then you won't see any point in trying to "work around it", but neither will you see any need to do so. But if you think quirks of the monetary system will destroy investment opportunities that would otherwise be profitable, then there will be a strong incentive to develop other payment schemes and induce people to accept them as a way of opening up those profit opportunities.

I like your explanation.
These investment and possibly more important payment substitutes must influence the price of pure cash (no one else's liability) since they compete with it as medium of exchange.
Here's another explanation by Silvio Gesell:

As credit transactions have in this way a powerful influence upon the demand for money, we must consider them so"mewhat more closely.

 We said that wares represent a demand for the medium of exchange exactly corresponding to their amount and quality. So, if there were any method of exchanging wares without employing money, the demand for money would be reduced by the amount of the wares so exchanged. This is self-evident when examined with the aid of our conception of the demand for money. Here again we may use a railway-line as an illustration. The demand for rolling stock is exactly equal to the amount of goods awaiting transport. But if a canal is built along the railway, the demand for rolling stock decreases by the amount of the goods transported by canal.

 Credit transactions substituted for money in the exchange of goods have the same effect as such a canal. If A. in Königsberg sends B. in Aix-la-Chapelle a consignment of butter, and B. pays the bill with a consignment of wine, the transaction is completed without a pfennig of money. If B. had no credit with A. or A. had no credit with B. the butter would have been handed over only for money, and the wine could only have been exchanged in the same way. The demand which the wine and butter would have created for money is here eliminated by credit.

http://www.community-exchange.org/docs/Gesell/en/neo/part3/8.htm
http://www.community-exchange.org/docs/Gesell/en/neo/part3/9.htm (the text is taken from this one)
http://www.community-exchange.org/docs/Gesell/en/neo/part3/10.htm
http://www.community-exchange.org/docs/Gesell/en/neo/part3/11.htm (and here he explains the problem with gold and deflation)

"No merchant, speculator or employer will discount a bill at the bank and undertake the obligation of paying interest if he suspects that the product he thinks of buying may fall in price. A fall of price may mean that he does not get back even the amount of his outlay."

But actually he didn't think that credit transactions can prevent price deflation:

If prices rise, that is, if demand exceeds supply, credit comes into play, deprives money of part of the wares to be exchanged and drive prices still higher. If prices fall, credit retires, wares must be exchanged for cash, and prices are still further depressed.

[lonelyminer (Peter Šurda)]

No matter the form, the scarcity itself will make BTC more possibly to become the target of mortgage, and then many loan based on BTC will be issued, these loans will require increased currency supply, it's the same effect as FRB

Let me repeat again:

So far, noone has explained to me how money can be more scarce or less scarce. From economic point of view, this concept makes no sense.

The increase in the money supply through credit expansion is guided by the features of the substitute created, for example, transaction costs. It is not caused by "money being scarce".

[jtimon]

No matter the form, the scarcity itself will make BTC more possibly to become the target of mortgage, and then many loan based on BTC will be issued, these loans will require increased currency supply, it's the same effect as FRB

Let me repeat again:

So far, noone has explained to me how money can be more scarce or less scarce. From economic point of view, this concept makes no sense.

Mutual credit based monies like LETS are not scarce, as opposed to cash-monies. It can be created (and destroyed) when is needed (at transaction time).

The increase in the money supply through credit expansion is guided by the features of the substitute created, for example, transaction costs. It is not caused by "money being scarce".

If I want to buy you a good but I lack the money. But you trust me an accept my IOU, we're expanding the credit. In "credit" I'm including these: http://en.wikipedia.org/wiki/Bill_of_exchange
Are you saying that there cannot be substitutes for bitcoin because it is so easy to transfer and store?
Because there's already bitcoin substitutes. Accounts on markets like mtgox, Open Trasaction's "untraceable cash" backed by bitcoin or ripplepay credit lines denominated in btc.
Bitcoin transactions aren't instant. That could be their feature but I think is more than that. Ripple can enable trade in communities where there's no actual cash to trade with. I don't know, maybe mutual credit is not "real life" enough for you but it seems the best example for "circumventing money scarcity through credit" to me.
Did you hear about Wir money? How do you explain that it gets more used during deflationary/illiquid times?