The deflationary problem

[Quantumplation]

Swift,

Here's another aspect:

Lets imagine that there are no transaction fees in 2040.   At that time, there should be strong Bitcoin banks and other financial institutions.  These institution will have every reason to run miner to keep the integrity of their own transactions and to make sure that deposits are safe from attack.  If I'm not mistaken, miners can pick and choose which transactions to include in a block.  I believe Mt Gox runs miners to speed verification of transaction.  Mining helps bitcoin financial institutions even without transaction fees or rewards from new blocks; The incentive already exists. 

Yes!  Finally someone who gets it!  I've been trying to highlight this for months.  Many of the intelligent forum members don't equate mining to a strong bank vault or other security measures.

I pointed this out in another thread (or maybe this one, can't remember), and it doesn't just apply to bankers.  It also applies to financial investment firms, because they can choose to accept their OWN transactions into the block without a fee, thus significantly reducing their cost overhead for investing in stocks and other such things that I don't have a head for.

[rebuilder]

I read through this whole thread just to make sure it really was just on the 5th page that someone brought up the point that if Bitcoin is still around in 2040 in any form that it makes sense to attack, there will be so many parties with vested interests in keeping the system stable that they'll mine their asses off even if it isn't directly profitable. (OK, someone on the third page kind of said that too, but not so directly).

Seriously. Poor show, guys.

Now, to digress a little bit... If you assume direct rewards from mining eventually decrease to the point where it's no longer seen as a profit vector, but rather a necessary cost of doing business in Bitcoin - is that a form of inflation built into the system itself? If holding Bitcoin means having to spend money to secure the system, what does that mean for spending incentives? I'll admit it's a pretty contrived comparison, but there it is.

[MoonShadow]

Swift,

Here's another aspect:

Lets imagine that there are no transaction fees in 2040.   At that time, there should be strong Bitcoin banks and other financial institutions.  These institution will have every reason to run miner to keep the integrity of their own transactions and to make sure that deposits are safe from attack.  If I'm not mistaken, miners can pick and choose which transactions to include in a block.  I believe Mt Gox runs miners to speed verification of transaction.  Mining helps bitcoin financial institutions even without transaction fees or rewards from new blocks; The incentive already exists. 

Yes!  Finally someone who gets it!  I've been trying to highlight this for months.  Many of the intelligent forum members don't equate mining to a strong bank vault or other security measures.

I pointed this out in another thread (or maybe this one, can't remember), and it doesn't just apply to bankers.  It also applies to financial investment firms, because they can choose to accept their OWN transactions into the block without a fee, thus significantly reducing their cost overhead for investing in stocks and other such things that I don't have a head for.

I have pointed this theme out in other threads as the "Wal-Mart" perspective.  When the day comes that either Wal-MArt or one of it's major competitors starts accepting bitcoins, Wal-mart (and competitors) will have an outsized interest in timely processing of free transactions (for themselves not their competitors) so Wal-Mart might fund a mining cluster (they already have partially funded a datacenter) while smaller competitors such as Target would likely contract out to a major mining company to process their customers' transactions for free, so that they don't have to wait for free transactions to process on the open network nor ask their customers to pay a transaction fee.  Wal-Mart doesn't need a 'financial institution', they are the financial institution.

[Sweft]

The ONLY way transaction fees will increase is through the velocity of money.  Bitcoin is designed to favor a deflationary model.  The problem with depending on transaction fees for hash growth is that velocity of money will decrease, which decreases transaction fees.  The value of bitcoins is dependent on hash.  If miners are not profiting, there's no point to increase hash, the network is less stable and secure.

If mining is unprofitable and we begin to rely on institutions to exert complete control over the system, without any individual miners, the hash becomes centralized to a few authorities - you also decrease the security of the system.  The most healthy and vibrant system is where miners are small and abundant.  When centralized authorities control large percentages of the mining pool (say if banks or walmart invested quantities of money), the network will be LESS safe.

In effect, removing reward of mining still destroys the system.

[rebuilder]

Sweft: So, assuming centralization really becomes that extreme, Wal-Mart could abuse their hashpower to double-spend their BTC. They could print their own USD right now and "profit", but the risk is too high. They're in business, and screwing over their customers like that is usually a bad idea for such a big target.

[Sweft]

Sweft: So, assuming centralization really becomes that extreme, Wal-Mart could abuse their hashpower to double-spend their BTC. They could print their own USD right now and "profit", but the risk is too high. They're in business, and screwing over their customers like that is usually a bad idea for such a big target.

Well avoid the 2040 analogy for a second.  What i'm trying to say is that it's a threat for adoption.  Bitcoins will become a target if they keep growing.  There is no question about it, no one should really assume otherwise.  Everything should be created with the assumption that people will attempt to destroy the currency whenever possible.  If you base every design decision on that assumption, you create the soundest currency possible.  My conclusion is that a deflating currency limits the security of the currency and opens it up to attacks.

Technology advances ever year, if there's no investment in the infrastructure, the system will be prone to attack.  If mining is less profitable, there's little reason to invest in infrastructure.  If hash remains constant, ie. no new infrastructure is created to support a growing hash.  In 2 years, hardware efficiency doubles, in 4 years, it quadruples, in 6 years it's 8x more efficient.  That means hash has to grow in response to technological advancements.  If it doesn't, the barrier to control the market becomes very small.

You can make arguments, come up with theoretical possibilities - how to defend a deflating currency.  People may mine at very large loses to protect their coins.  Sure.  If there's no profit motive, what is the point of increasing hash?  Only to protect your own money.

So, in any way - my original argument is that deflation is bad for the system.

Inflation creates a reason to mine, which reinforces the system.  Once you remove inflation the argument trends into waters of possibility.  It may be possible to defend the system, but it's much more difficult and the arguments revolve around unclear means.  

There's absolutely no reason to favor deflation, there is reason to favor inflation.

[rebuilder]

Inflation creates a reason to mine, which reinforces the system.  Once you remove inflation the argument trends into waters of possibility.  It may be possible to defend the system, but it's much more difficult and the arguments revolve around unclear means.  There's absolutely no reason to favor deflation, there is reason to favor inflation.

There is a lot of reason to favor deflation. Let's disregard the wild exchange rate fluctuations for the moment, since we're assuming deflation exists. If I offer you payment in BTC and you expect it will deflate, and I also offer to pay you in some inflationary currency, which would you rather take, absent anyone forcing you to use one currency over the other?

I'm not at all convinced an inflationary currency can be competitive without some sort of (coercive probably) immediate incentive for people to use it.

[MoonShadow]

I'm not at all convinced an inflationary currency can be competitive without some sort of (coercive probably) immediate incentive for people to use it.

Well, it doesn't really have to be coercive.  Bad money will chase good money out of the market, because while the vendor would prefer to receive his funds in the good money, if the customer is only offering the bad money, then the vendor will take that with a risk premium added on.  This is why dimes made of real silver still exist, but never circulate.  That and inflation has made the silver content worth over $3 apiece.

In a truly free marketplace, Bitcoin would never have been more than a transfer mechanism with websites like eGold operating unmolested.  Of course, if eGold was still around, Bitcoin would likely never have been created.  It's the lack of a free market in money that drives the demand for Bitcoin to begin with.  If the governments of the world really want to destroy Bitcoin, all they have to do is (credibily) return to the Gold standard.  I don't forsee that happening, however.

[Sweft]

Inflation creates a reason to mine, which reinforces the system.  Once you remove inflation the argument trends into waters of possibility.  It may be possible to defend the system, but it's much more difficult and the arguments revolve around unclear means.  There's absolutely no reason to favor deflation, there is reason to favor inflation.

There is a lot of reason to favor deflation. Let's disregard the wild exchange rate fluctuations for the moment, since we're assuming deflation exists. If I offer you payment in BTC and you expect it will deflate, and I also offer to pay you in some inflationary currency, which would you rather take, absent anyone forcing you to use one currency over the other?

I'm not at all convinced an inflationary currency can be competitive without some sort of (coercive probably) immediate incentive for people to use it.

Listen please explain to me how mining is benefited in an deflationary currency.

Gold mining is inflationary, we mine much more gold now then we did before, but the price keeps increasing.

Why? Because it requires significant input energy to mine it.  The same follows for bitcoins.

Why would you not accept bitcoins that were mined later since they used significant energy to mine it.  It makes absolutely no sense.  Bitcoin is not fiat, and inflation does not affect it in the same way.

Lets please recognize this point.  Otherwise people cry inflation.  You people who do do not understand that it requires a lot of money to inflat bitcoin unlike fiat.  Same as gold, which requires abundent energy.

[mcdeeda]

hah, this all implies that bitcoins will last as a currency for more than a few months before collapsing spectacularly

[MoonShadow]

Inflation creates a reason to mine, which reinforces the system.  Once you remove inflation the argument trends into waters of possibility.  It may be possible to defend the system, but it's much more difficult and the arguments revolve around unclear means.  There's absolutely no reason to favor deflation, there is reason to favor inflation.

There is a lot of reason to favor deflation. Let's disregard the wild exchange rate fluctuations for the moment, since we're assuming deflation exists. If I offer you payment in BTC and you expect it will deflate, and I also offer to pay you in some inflationary currency, which would you rather take, absent anyone forcing you to use one currency over the other?

I'm not at all convinced an inflationary currency can be competitive without some sort of (coercive probably) immediate incentive for people to use it.

Listen please explain to me how mining is benefited in an deflationary currency.

Bitcoin is not deflating.  Bitcoin is inflating at roughly 40% APR, and will not actually be deflating until at least 2129.  This is a non-issue in my lifetime.

[ryepdx]

hah, this all implies that bitcoins will last as a currency for more than a few months before collapsing spectacularly

Well... it's already lasted for 2 years... so I don't think it's going to just collapse out of the blue. Price might drop, but every market does that from time to time.

[Klestin]

On the off chance that OP isn't trolling, I'll give a go at removing the main concern of the post.

Although as others have pointed out, 2040 is not a special date, we can at least agree that the coin creation reward for block solving will have diminished by then to a much smaller reward than exists today.

In fact, in 2041, the coin creation reward for solving a block will drop from 0.39BTC to 0.19 BTC.  Sounds drastic, right?  Who would mine for 0.19 BTC???  But that reward is only a small part of the picture.  We need to consider transaction fees and exchange rates too.

Let's consider a pretty conservative (but of course not guaranteed) prediction of a future in which BTC is commonly used and has been adopted by a good portion of the populace, and is used for a portion of Internet commerce.  There might, in this world, be 10,000 transactions in each 10 minute block, and let's say they pay on average 0.002 BTC per (a small fraction of the suggested transaction fee for today).  That works out to 20 BTC in transaction fees, or 20.39 including the creation reward.

Now, 2040 rolls into 2041 and (gasp) the creation reward drops to 0.19.  Suddenly and without warning, the total block creation reward drops to 20.19 (from 20.39)!  Oh noes!  

Who can say what the exchange rate in USD will be at that time?  $100? $1000?

As time goes on, and adoption increases, the transaction fees increase.  In even a very conservative prediction, transaction fees outweigh the creation reward well before 2040.

[Sweft]

On the off chance that OP isn't trolling, I'll give a go at removing the main concern of the post.

Although as others have pointed out, 2040 is not a special date, we can at least agree that the coin creation reward for block solving will have diminished by then to a much smaller reward than exists today.

In fact, in 2041, the coin creation reward for solving a block will drop from 0.39BTC to 0.19 BTC.  Sounds drastic, right?  Who would mine for 0.19 BTC???  But that reward is only a small part of the picture.  We need to consider transaction fees and exchange rates too.

Let's consider a pretty conservative (but of course not guaranteed) prediction of a future in which BTC is commonly used and has been adopted by a good portion of the populace, and is used for a portion of Internet commerce.  There might, in this world, be 10,000 transactions in each 10 minute block, and let's say they pay on average 0.002 BTC per (a small fraction of the suggested transaction fee for today).  That works out to 20 BTC in transaction fees, or 20.39 including the creation reward.

Now, 2040 rolls into 2041 and (gasp) the creation reward drops to 0.19.  Suddenly and without warning, the total block creation reward drops to 20.19 (from 20.39)!  Oh noes!  

Who can say what the exchange rate in USD will be at that time?  $100? $1000?

As time goes on, and adoption increases, the transaction fees increase.  In even a very conservative prediction, transaction fees outweigh the creation reward well before 2040.

Yeah, unfortunately with transaction fees you have to deal with the velocity of money.  In a deflating currency, the velocity of money declines, less transactions, less fees.  The numbers you have created are pure speculation. 

The other problem is that for hash to increase, there must be either an increase in velocity or block inflation.  We reach a point where block creation is negligible.  Thus, since we agree that the deflationary aspect causes hording (which i'm not arguing is bad in commodities like gold, etc) but the proof of gold does not depend on mining.  It's the proof itself.  A bitcoin requires future proof of work.  Gold requires past proof of work.

Deflation poses numerous problems to the future of bitcoin.  They should be easily solved by a competing cryptocurrency.

[nathanrees19]

The value of bitcoins is dependent on hash.

No.

[nathanrees19]

Yeah, unfortunately with transaction fees you have to deal with the velocity of money.  In a deflating currency, the velocity of money declines, less transactions

No, you get smaller transactions, not less transactions.

less fees.

I don't think you know how the fees work.

[marcus_of_augustus]

Yeah, unfortunately with transaction fees you have to deal with the velocity of money.  In a deflating currency, the velocity of money declines, less transactions, less fees.  The numbers you have created are pure speculation. 

You stopped at the final hurdle. (remember this is in far, far future when btc is actually deflationary)

Velocity of money declines, less transactions, less fees ....

less fees, less mining, difficulty eases ...

difficulty eases, valuations decline slightly (inflationary expectations creep in), people start spending ...

velocity of money increases, more fees, more mining .... until velocity of money declines ... begin loop again, it will track the demand for money.

It appears there is a self-regulating mechanism built in even in the far, far future when your grandkids might be debating these same things. (Be careful, they maybe reading what you write today).

[Sweft]

Yeah, unfortunately with transaction fees you have to deal with the velocity of money.  In a deflating currency, the velocity of money declines, less transactions, less fees.  The numbers you have created are pure speculation. 

You stopped at the final hurdle. (remember this is in far, far future when btc is actually deflationary)

Velocity of money declines, less transactions, less fees ....

less fees, less mining, difficulty eases ...

difficulty eases, valuations decline slightly (inflationary expectations creep in), people start spending ...

velocity of money increases, more fees, more mining .... until velocity of money declines ... begin loop again, it will track the demand for money.

It appears there is a self-regulating mechanism built in even in the far, far future when your grandkids might be debating these same things. (Be careful, they maybe reading what you write today).

Sure, there is self regulation.  That's fine.

You don't understand one point.  Without increasing hash, the network is not secure.  If hash begins to decrease or remain at equilibrium the network will be vulnerable to compromise.  This is inescapable.

Without an increase in hash, the network is vulnerable.  If miners don't profit, the network is vulnerable. 

Inflation expectations cannot creep in to a currency that is mathematically devoid of inflation.

[Sweft]

The value of bitcoins is dependent on hash.

No.

You don't understand bitcoins, sorry.  Read the thread, hopefully you will understand.  If not, again, sorry.

[Sweft]

Yeah, unfortunately with transaction fees you have to deal with the velocity of money.  In a deflating currency, the velocity of money declines, less transactions

No, you get smaller transactions, not less transactions.

less fees.

I don't think you know how the fees work.

Velocity of money means a decrease in the exchange of money.  That's what it means.  A currency that will be worth more in the future will decline in transactions because there is little reason to spend it, when you earn money by holding it.  A decrease in transaction fees leads to a decrease in mining profitability, which leads to a decrease in hash.

A decrease in hash will compromise the network.