Finite Supply vs Steadily Increasing Supply

[johnyj]

The problem is that you can never keep an economy grow at a steady increasing pace, it is like wave, always have ups and downs. There are already too many variables in economy, the money supply should be a constant to reduce uncertainty, and its value will fluctuate depends on market demand

[godislove]

Nowhere in your wall of text did you realize that contracts could specify valuation in terms of a basket of goods without putting the currency itself in the hands of madmen?

If you know of a financial instrument that allows people to keep constant value in terms of a basket of commodities, do tell.  I'm talking about spot value, not a 9% annual loss from ETFs that require a monthly rolling over of futures contracts. 

[teukon]

But that's just it.  The deflationary effect of lost Bitcoins is already offset by the risk involved with holding them.  There's already balance.

I think I am starting to see what you are saying here.  Is it something along the lines of "people know about the risk of losing Bitcoins and account for it accordingly"?  Sorry, I'm probably being dense on this one...

Don't worry, I was oversimplifying my point and I guess that led you astray.  Some people have claimed that general coin loss causes people to hoard bitcoins but they fail to account the risk of personal coin loss faced by the hoarders.

Your latest post has given me a much better understanding of what you're thinking about and it seems I was slightly off-topic; my apologies.

I feel that we are aiming for the stability of subtly different things. 

I am looking to make economic calculation as easy as possible.  The more stable prices are (i.e. the degree to which they are likely to change over time) the more easily someone can plan for the future.

I'm certainly with you here.  Planning should be made as fruitful as possible and this implies eliminating all unnecessary uncertainty in the currency itself.  However, I don't see how predictable monetary inflation (which is essentially predictable wealth redistribution) can fundamentally assist planning.  In particular, how would having more money that expected impact negatively on a person's plans?

Coming at this from another angle: if price stability is the goal then why not price everything, including bitcoins, in terms of a more stable commodity (perhaps a basket of goods and services).  Instead of having a fixed balance and falling prices, people will see their balances rise (like receiving interest) and enjoy stable prices.

Please tell me what you think about the following scenario (you may assume no variation in the size of the underlying economy, no coin loss, no block rewards, and no divisibility limit):

I design an altcoin called Deflatacoin.  This is identical to Bitcoin but where on each day, the balance of each wallet is reduced by 0.1% (0.1% of all deflatacoins are destroyed).  This causes serious price deflation (about 44% per year).

Is this price deflation economically problematic?  Should it be countered by an equal amount of monetary inflation?  Is this currency more stable with or without the extra inflation?

I would say it is not problematic as it is a static amount each day, which is something people can account for.  If the amount were variable, as in the case of lost Bitcoins, there is no way people could account for such loses.  Giving them a know offsetting amount allows them to plan better, although still not perfectly.

Thanks, I see.  But if the unpredictability of coin loss is the problem, how would predictable monetary inflation help?  Surely the sum of these two effects will be practically as unpredictable as the former.

[temp1029]

Don't worry, I was oversimplifying my point and I guess that led you astray.  Some people have claimed that general coin loss causes people to hoard bitcoins but they fail to account the risk of personal coin loss faced by the hoarders.

I appreciate you clarifying your points from earlier, I now understand what you were aiming at and agree with you that this type of "problem" really isn't.

However, I don't see how predictable monetary inflation (which is essentially predictable wealth redistribution) can fundamentally assist planning.  In particular, how would having more money that expected impact negatively on a person's plans?

Maybe that is where my theory is wrong.  I view the re-distributive power of monetary inflation as only occurring when it is not predictable (i.e. in the way the fed does it).  If we were to know 100% for sure what the fed would do next week, there wouldn't be a problem being ready for it.  In fact, we can see this type of preparation whenever the fed discusses tapering (or the converse, more "QE").  This type of knowledge would actually take the teeth out of monetary inflation, as prices would adjust before the new money could be spent.  I could certainly have this wrong; any thoughts?

I'm not really sure it would have a negative impact, but it would certainly make planning harder (by at least a little bit).

Coming at this from another angle: if price stability is the goal then why not price everything, including bitcoins, in terms of a more stable commodity (perhaps a basket of goods and services).  Instead of having a fixed balance and falling prices, people will see their balances rise (like receiving interest) and enjoy stable prices.

Unfortunately, even a basket of goods is subject to changes in supply.  As examples, natural disasters can decrease supply and technological innovation can increase supply.  On the other side, demand can also change; people want bananas instead of oranges this month, next month they decide apples are the hot fruit (my wife is eating an apple right now, hence the fruit example ).

Thanks, I see.  But if the unpredictability of coin loss is the problem, how would predictable monetary inflation help?  Surely the sum of these two effects will be practically as unpredictable as the former.

I believe it would help because static inflation would offset, at least partially, the dynamic deflationary effects of coin loss.  Doing so would give people a certain offsetting amount that they could plan against.  Basically, given uncertainty in one direction certainty in the other direction helps balance things out, although admittedly not entirely.

There two other benefits I see as well and, although these are a bit off-topic for the purposes of this thread, I'll put them in here.
1) Miners would have further incentive to continue providing network security.
2) Fees could be kept low, as miners would will receive rewards from #1.

[teukon]

Maybe that is where my theory is wrong.  I view the re-distributive power of monetary inflation as only occurring when it is not predictable (i.e. in the way the fed does it).  If we were to know 100% for sure what the fed would do next week, there wouldn't be a problem being ready for it.  In fact, we can see this type of preparation whenever the fed discusses tapering (or the converse, more "QE").  This type of knowledge would actually take the teeth out of monetary inflation, as prices would adjust before the new money could be spent.  I could certainly have this wrong; any thoughts?

Inflation implies wealth redistribution in all cases.  I agree that the system you propose lacks the teeth of "Quantitative Easing".  I think it's the ability to print more "when more money is needed" that Keynesians argue can be used to "smooth out the bumps" in economic growth and thereby boost the efficiency of financial planning.  I'll admit that I'm far from an expert on such matters.

Unfortunately, even a basket of goods is subject to changes in supply.  As examples, natural disasters can decrease supply and technological innovation can increase supply.  On the other side, demand can also change; people want bananas instead of oranges this month, next month they decide apples are the hot fruit (my wife is eating an apple right now, hence the fruit example ).

I completely agree (and I've just finished an apple myself).  However, I believe we were concerning ourselves not with prices in the short term, but with the general price level in the long term.  I don't believe that the price fluctuations brought on by supply and demand as you have illustrated are within the purview of monetary inflation.

Perhaps I can make my illustration more direct:  Let us assume that approximately 2% of all bitcoins are truly lost each year.  Let us define a "fixcoin" as (100% - 2%)^n BTC (where n is the number of years since Bitcoin's genesis).  Thus:

• in 2009, a fixcoin was equivalent to 1000 mBTC;
• in 2010, a fixcoin was equivalent to 980 mBTC;
• today (2013), a fixcoin is worth about 922 mBTC.

By pricing everything in fixcoin rather than bitcoin (including Bitcoin wallet balances) we bring the concept of a 2% per year monetary inflation to life.  The only difference is that, instead of going to miners, the extra money appears in Bitcoin wallets as though it were 2% interest.  As I remarked earlier, I don't see how an entity's financial plans can be undermined by receiving such interest.

I believe it would help because static inflation would offset, at least partially, the dynamic deflationary effects of coin loss.  Doing so would give people a certain offsetting amount that they could plan against.  Basically, given uncertainty in one direction certainty in the other direction helps balance things out, although admittedly not entirely.

I agree that the inflation would partially compensate for the deflationary effect of coin loss.  However, you've not convinced me that the balance achieved provides any economic or planning benefits whatsoever.  If you can expand on and supply concrete justification for your intuition here then you'll receive more constructive feedback.

There two other benefits I see as well and, although these are a bit off-topic for the purposes of this thread, I'll put them in here.
1) Miners would have further incentive to continue providing network security.
2) Fees could be kept low, as miners would will receive rewards from #1.

Certainly, such an inflationary scheme would be a boon to long-term network security.  However, there are serious drawbacks to consider and, as you note, this is somewhat off-topic.

[maxxoccupancy]

Simple:
I. Build in a parallel mine and premine that are owned by the Block Chain itself.  Used as a reserve, the Block Chain could buy up Bitcoins/altcoins, selling them for itself in order to take excess altcoins out of circulation.
II. Increase/reduce the reward coins each day, reducing them to proof-of-stake when the altcoin devalues, until the price recovers.
III. Adjust the total number of coins intended to be produced over the life of the altcoin.  If demand goes up, raise the limit.  If demand declines, lower that limit.

[johnyj]

If added money supply can be evenly distributed to the whole society at the same time, then maybe a steady increase in money supply is useful to stabilize the price level, but anyway since GDP will rise slower and slower due to better and better life for everyone, that might not be needed

In today's fiat money system, the added money supply will first go into banker's pocket and flow into the society in form of loans. If the whole society do not want to take a loan, then all those added money supply will stuck at bank's account

[minerva]

If added money supply can be evenly distributed to the whole society at the same time, then maybe a steady increase in money supply is useful to stabilize the price level, but anyway since GDP will rise slower and slower due to better and better life for everyone, that might not be needed

In today's fiat money system, the added money supply will first go into banker's pocket and flow into the society in form of loans. If the whole society do not want to take a loan, then all those added money supply will stuck at bank's account

Not really. QE-2 and QE-Infinity repurchased mortgage-backed securities and treasuries. The money supply does not go into the banker's pocket, but to people who own financial instruments who have sold them to the Federal Reserve.