February 9, 2015
Production Quotas
As you may have heard, Bitcoin usage is growing. In every way, that graph represents the kind of healthy exponential adoption that we all want to see.
Unfortunately there's a problem on the horizon which threatens to stop or even reverse Bitcoin adoption.
Several years ago, Satoshi added a protocol limit to the maximum size of a Bitcoin block. Prior to this change, there was no explicit limit, just an implicit 32 megabyte maximum message size. This limit was explained as a temporary anti-spam measure, and Satoshi said at the time that it could be raised when the network needed the additional capability.
The economic effect of having a maximum block size is that of a production quota. Production quotas are tools of economic central planning that either mandate or limit the amount of production of a good or service, as opposed to allowing the production rates to be governed by supply and demand.
Production quotas are inherently harmful to an economy, as shown in the 1920 essay, Economic Calculation in the Socialist Commonwealth, and so do not represent a sustainable long term strategy for allocating the supply of Bitcoin transactions.
From the time it was implemented until now, the harm caused by the block size limit has been hypothetical rather than real since there has not yet been enough demand for Bitcoin transactions to be hampered by the limit. For the last few years, the block size limit has been like a minimum wage law that forbids salaries lower than $0.01 per year. There is no market demand for salaries that low, and so such a minimum wage law might as well not exist it has no effect on the economy. Likewise, the 1 MB block size limit has not yet had any effect on the Bitcoin economy since there has not yet been a market demand for more than 1 MB of transactions every 10 minutes.
Since its inception, Bitcoin has been operating as if there was no block size limit at all. If this limit is kept in place when the market demand for transactions rises above 1 MB/10 minutes, then suddenly Bitcoin will be in uncharted economic territory.
People will want to use Bitcoin, but they will be forbidden by protocol from doing so. No matter how much they are willing to pay, no matter how willing miners are to include their transactions in a block, no matter how willing the full node operators are to forward their transactions, they simply wont be allowed to transact.
A block size limit that is low enough to have a real effect on actual block sizes is the ultimate blacklist.
The Alternative to Central Planning
The best alternative to a production quota on Bitcoin transactions is, like in any other situation of central planning, to allow the market to decide the optimum block size.
Nobody gives McDonalds a maximum number of Big Macs they are allowed to produce each day their customers tell them how many they want to buy and McDonalds responds appropriately. At any given time, there exists a price at which the willingness of McDonalds to produce Big Macs is exactly equal to the willingness of their customers to buy them, and that determines the number that will be produced. The process of price discovery is an emergent property of the actions of millions of independent actors expressing their preferences in a competitive open market.
This is exactly how we want Bitcoin to behave. The Bitcoin network, like any other product or service in the economy, should change its production capability to respond to supply and demand.
It is not currently designed to do this, however, and one of the barriers preventing Bitcoin from being improved in this way is a series of economic fallacies or misconceptions that cause otherwise skilled people to distrust market-driven price discovery over central planning, or to assume that resource allocation in computer networks operates in a manner fundamentally different than resource allocation in any other part of the economy.
Objections to Market-Determined Block Sizes
What if the market-set transaction fee doesn't pay enough for a hash rate that protects the network for well-funded adversaries?
If there is not enough market demand for Bitcoin transactions such to pay for sufficient hashing power to protect the network, then Bitcoin will fail.
This will happen with or without a block size limit.
Since its inception, Bitcoin has been on a collision course with extremely well funded, entitled, and politically powerful interests. Its only hope of surviving this collision is by attracting a very broad base of support.
Bitcoin needs millions, and then billions, of users who demand better money. The demand for Bitcoin must be strong enough that they will break the law if that's what it takes to obtain it.
The above strategy isn't particularly novel or extreme this strategy has been employed in the conflict between peer-to-peer filesharing networks vs the copyright mafia, and more recently by ridesharing companies vs the taxi licensing cartels.
When a new technology has to compete against entrenched interests on a non-level political playing field, civil disobedience is a proven effective tactic.
We don't yet know whether or not Bitcoin will gain enough of the right kind of support it needs to survive.
We can say, however, that arbitrarily rationing the transaction rate is counterproductive toward achieving that end.
What if competition results in the profitability of mining being so low that it drives out smaller pools and Bitcoin mining converges to a monopoly?
This objection is a restatement of the natural monopoly argument, and is in no way specific to Bitcoin.
Natural monopoly as an economic theory has been conclusively debunked, and the same principles that explain why natural monopolies do not emerge from free market forces in classically-cited industries apply equally well to Bitcoin mining.
Rather than repeat those arguments here, anyone who is concerned about natural monopolies should read The Myth of Natural Monopoly by Thomas J. DiLorenzo.
If the market should set the block size, why shouldn't it also set the block reward?
This objection is based either on a fundamental misunderstanding on the nature of money, or else on the misconception that Bitcoins value is not derived from its monetary properties.
Exploring these misconceptions fully will be the subject of a future article.
This concludes part 1. Future articles in this series will address the subject of how to build economically-scalable P2P networks, and why the block reward is fundamentally different than the block size limit.
Part 1, Scarcity:
http://web.archive.org/web/20160423041910/https://bitcoinism.liberty.me/economic-fallacies-and-the-block-size-limit-part-1-scarcity/
Part 2, Price Discovery:
http://web.archive.org/web/20160318120539/https://bitcoinism.liberty.me/economic-fallacies-and-the-block-size-limit-part-2-price-discovery/
The average size of a block is approaching the 1 MB protocol limit for the first time in Bitcoins history, and not everybody agrees regarding what to do about it. Many objections to raising or removing the block size limit are based on misunderstandings about the nature of economic scarcity and operation of markets in general.
ECONOMIC FALLACIES AND THE BLOCK SIZE LIMIT
This is the first article in a series about the economics of the Bitcoin protocols block size limit. This article covers scarcity. The second article covers price discovery and the third article will cover why the block reward is fundamentally different than the block size.
http://web.archive.org/web/20160423041910/https://bitcoinism.liberty.me/economic-fallacies-and-the-block-size-limit-part-1-scarcity/
Part 2, Price Discovery:
http://web.archive.org/web/20160318120539/https://bitcoinism.liberty.me/economic-fallacies-and-the-block-size-limit-part-2-price-discovery/
The average size of a block is approaching the 1 MB protocol limit for the first time in Bitcoins history, and not everybody agrees regarding what to do about it. Many objections to raising or removing the block size limit are based on misunderstandings about the nature of economic scarcity and operation of markets in general.
ECONOMIC FALLACIES AND THE BLOCK SIZE LIMIT
This is the first article in a series about the economics of the Bitcoin protocols block size limit. This article covers scarcity. The second article covers price discovery and the third article will cover why the block reward is fundamentally different than the block size.
Production Quotas
As you may have heard, Bitcoin usage is growing. In every way, that graph represents the kind of healthy exponential adoption that we all want to see.
Unfortunately there's a problem on the horizon which threatens to stop or even reverse Bitcoin adoption.
Several years ago, Satoshi added a protocol limit to the maximum size of a Bitcoin block. Prior to this change, there was no explicit limit, just an implicit 32 megabyte maximum message size. This limit was explained as a temporary anti-spam measure, and Satoshi said at the time that it could be raised when the network needed the additional capability.
The economic effect of having a maximum block size is that of a production quota. Production quotas are tools of economic central planning that either mandate or limit the amount of production of a good or service, as opposed to allowing the production rates to be governed by supply and demand.
Production quotas are inherently harmful to an economy, as shown in the 1920 essay, Economic Calculation in the Socialist Commonwealth, and so do not represent a sustainable long term strategy for allocating the supply of Bitcoin transactions.
From the time it was implemented until now, the harm caused by the block size limit has been hypothetical rather than real since there has not yet been enough demand for Bitcoin transactions to be hampered by the limit. For the last few years, the block size limit has been like a minimum wage law that forbids salaries lower than $0.01 per year. There is no market demand for salaries that low, and so such a minimum wage law might as well not exist it has no effect on the economy. Likewise, the 1 MB block size limit has not yet had any effect on the Bitcoin economy since there has not yet been a market demand for more than 1 MB of transactions every 10 minutes.
Since its inception, Bitcoin has been operating as if there was no block size limit at all. If this limit is kept in place when the market demand for transactions rises above 1 MB/10 minutes, then suddenly Bitcoin will be in uncharted economic territory.
People will want to use Bitcoin, but they will be forbidden by protocol from doing so. No matter how much they are willing to pay, no matter how willing miners are to include their transactions in a block, no matter how willing the full node operators are to forward their transactions, they simply wont be allowed to transact.
A block size limit that is low enough to have a real effect on actual block sizes is the ultimate blacklist.
The Alternative to Central Planning
The best alternative to a production quota on Bitcoin transactions is, like in any other situation of central planning, to allow the market to decide the optimum block size.
Nobody gives McDonalds a maximum number of Big Macs they are allowed to produce each day their customers tell them how many they want to buy and McDonalds responds appropriately. At any given time, there exists a price at which the willingness of McDonalds to produce Big Macs is exactly equal to the willingness of their customers to buy them, and that determines the number that will be produced. The process of price discovery is an emergent property of the actions of millions of independent actors expressing their preferences in a competitive open market.
This is exactly how we want Bitcoin to behave. The Bitcoin network, like any other product or service in the economy, should change its production capability to respond to supply and demand.
It is not currently designed to do this, however, and one of the barriers preventing Bitcoin from being improved in this way is a series of economic fallacies or misconceptions that cause otherwise skilled people to distrust market-driven price discovery over central planning, or to assume that resource allocation in computer networks operates in a manner fundamentally different than resource allocation in any other part of the economy.
Objections to Market-Determined Block Sizes
"Transaction fees are too low and won't rise unless space in a block is scarce. We need a block size limit to ensure block space scarcity and thus price transactions."
This objection is based on a common misunderstanding of the word scarcity as it applies to economics.
In economic terms, something is scarce if people cant have an infinite amount of it at a price of zero.
On the surface of the Earth, air is not scarce. Its not scarce because every human can breathe as much as they are capable of breathing, without paying for the air, and theres still enough to go around. Because everybody can consume as much as they are capable of without reducing anyone elses ability to do the same, air does not require allocation. In practical terms, the amount of available air at a price of zero is infinite, therefore air is not scarce.
Almost everything else is scarce, certainly any services that require time and/or energy to produce.
The space in a block will always be scarce as long as our computers are still made of matter and still occupy space. Constructing a block isn't free, storing a block isn't free, and the bandwidth needed to transmit a block is not free.
There will always exist some cost to a miner to add a transaction to a block. That cost may be very small, but it will never be zero.
If transaction fees emerge from the operation of a competitive open market, then we would expect them to approach the marginal cost of production, plus a small profit margin.
This objection is based on a common misunderstanding of the word scarcity as it applies to economics.
In economic terms, something is scarce if people cant have an infinite amount of it at a price of zero.
On the surface of the Earth, air is not scarce. Its not scarce because every human can breathe as much as they are capable of breathing, without paying for the air, and theres still enough to go around. Because everybody can consume as much as they are capable of without reducing anyone elses ability to do the same, air does not require allocation. In practical terms, the amount of available air at a price of zero is infinite, therefore air is not scarce.
Almost everything else is scarce, certainly any services that require time and/or energy to produce.
The space in a block will always be scarce as long as our computers are still made of matter and still occupy space. Constructing a block isn't free, storing a block isn't free, and the bandwidth needed to transmit a block is not free.
There will always exist some cost to a miner to add a transaction to a block. That cost may be very small, but it will never be zero.
If transaction fees emerge from the operation of a competitive open market, then we would expect them to approach the marginal cost of production, plus a small profit margin.
What if the the market-set block size is so big that only Google can afford to run full nodes?
This is a real problem that could emerge, and is actually the reason that the block size limit was enacted in the first place.
The reason this could happen is because of poor P2P network design: miners do not need to pay the cost of relaying blocks throughout the network, therefore this cost becomes an externality which is not reflected in their marginal cost of production.
The solution to this objection is a better P2P network design, not a production quota that limits the maximum transaction rate.
A description of how to build a better P2P network will be the subject of a future article.
This is a real problem that could emerge, and is actually the reason that the block size limit was enacted in the first place.
The reason this could happen is because of poor P2P network design: miners do not need to pay the cost of relaying blocks throughout the network, therefore this cost becomes an externality which is not reflected in their marginal cost of production.
The solution to this objection is a better P2P network design, not a production quota that limits the maximum transaction rate.
A description of how to build a better P2P network will be the subject of a future article.
What if the market-set transaction fee doesn't pay enough for a hash rate that protects the network for well-funded adversaries?
If there is not enough market demand for Bitcoin transactions such to pay for sufficient hashing power to protect the network, then Bitcoin will fail.
This will happen with or without a block size limit.
Since its inception, Bitcoin has been on a collision course with extremely well funded, entitled, and politically powerful interests. Its only hope of surviving this collision is by attracting a very broad base of support.
Bitcoin needs millions, and then billions, of users who demand better money. The demand for Bitcoin must be strong enough that they will break the law if that's what it takes to obtain it.
The above strategy isn't particularly novel or extreme this strategy has been employed in the conflict between peer-to-peer filesharing networks vs the copyright mafia, and more recently by ridesharing companies vs the taxi licensing cartels.
When a new technology has to compete against entrenched interests on a non-level political playing field, civil disobedience is a proven effective tactic.
We don't yet know whether or not Bitcoin will gain enough of the right kind of support it needs to survive.
We can say, however, that arbitrarily rationing the transaction rate is counterproductive toward achieving that end.
What if competition results in the profitability of mining being so low that it drives out smaller pools and Bitcoin mining converges to a monopoly?
This objection is a restatement of the natural monopoly argument, and is in no way specific to Bitcoin.
Natural monopoly as an economic theory has been conclusively debunked, and the same principles that explain why natural monopolies do not emerge from free market forces in classically-cited industries apply equally well to Bitcoin mining.
Rather than repeat those arguments here, anyone who is concerned about natural monopolies should read The Myth of Natural Monopoly by Thomas J. DiLorenzo.
If the market should set the block size, why shouldn't it also set the block reward?
This objection is based either on a fundamental misunderstanding on the nature of money, or else on the misconception that Bitcoins value is not derived from its monetary properties.
Exploring these misconceptions fully will be the subject of a future article.
This concludes part 1. Future articles in this series will address the subject of how to build economically-scalable P2P networks, and why the block reward is fundamentally different than the block size limit.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.