Blocksize Economics

[jonny1000]

Dear Gavin

Thanks for providing the update on the Scalability Roadmap piece and for focusing more on some of the economic issues.  See https://bitcoinfoundation.org/2014/10/blocksize-economics/ I think it’s very good that you have addressed and discussed some of these crucial economic and incentive type issues.

First let me address your comment that “economic theory says that in a competitive market, supply, demand, and price will find an equilibrium where the price is equal to the marginal cost to suppliers plus some net income (because suppliers can always choose to do something more profitable with their time or money). In this case, price is transaction fees, supply is the willingness or ability of miners to confirm transactions, and demand is the number of transactions people want to have confirmed.”  It is important to realise that Bitcoin is in some respects unique and unlike any system that came before it, it is not always possible to apply the normal rules of “economic theory” to this system.  This is not a normal market with normal supply and demand.  There are several key differences:

1.      Normal markets do not have a difficulty adjustment every two weeks
2.      In normal markets there is no reason why having a large number of suppliers competing is important
3.      In normal markets suppliers don’t compete in a bingo like random competition to be awarded the contract

Contrasting the market for space in blocks with other markets is a useful exercise, however one must realise the dynamics are potentially very different and requires careful analysis.

Neither I, nor the other people saying that having no arbitrary limit in the block size would mess up mining economics are saying the limit does not need to be increased.  I think there is a strong consensus that an increase is necessary.  I merely maintain that some limit is necessary for economic reasons.  When the block reward falls there is no guarantee that any system will produce enough revenue to miners to secure the network.  An arbitrary level of scarcity in the block size limit (at a much higher level than 1MB), is the best method I can think of for providing this security, even though it is far from perfect.

Many thanks

[piotr_n]

The cruel thing about blockchain economics is that it's not for Gain, nor for any of his friends to decide about sizes of the blocks.
It's the mining,  that he doesn't care of.

So I just wonder what is in this guy's head when he assumes that miners actually care about him and his opinion on a size of their blocks? Who is he actually speaking to with poems like this and what is a point of this?


People who understand the problem  know that the key to make bitcoin really scalable and keeping transactions cheap  is adding a layer of off-chain  transactions on top of the existing protocol.
Even Gavin knows that - but apparently he prefers to have been spending the past two years of his life on developing an 'intelligent' fee discovery system and writing papers on why the blocks must be bigger.

And don't worry - he won't mind me saying it,  I'm on his ignore list.

[Cubic Earth]

The cruel thing about blockchain economics is that it's not for Gain, nor for any of his friends to decide about sizes of the blocks.
It's the mining,  that he doesn't care of.

So I just wonder what is in this guy's head when he assumes that miners actually care about him and his opinion on a size of their blocks? Who is he actually speaking to with poems like this and what is a point of this?


People who understand the problem  know that the key to make bitcoin really scalable and keeping transactions cheap  is adding a layer of off-chain  transactions on top of the existing protocol.
Even Gavin knows that - but apparently he prefers to have been spending the past two years of his life on developing an 'intelligent' fee discovery system and writing papers on why the blocks must be bigger.

And don't worry - he won't mind me saying it,  I'm on his ignore list.

I can see why you might be on Gavin's ignore list.... your post has a very grumpy tinge to it.  Getting onto your concerns, bitcoin is more than just mining.  The users and people who build services are important too.  Bitcoin is strong not because of some computers, but because of how, as a community, we use those computers.  So it is in important to keep everyone together as much as possible.

pitor_n, it sounds like you are speaking as if Gavin was proposing to create a MaxBlockSize when there was none before.  If some miners started creating 3MB blocks right now, they would just be ignored.  Why is that?


Gavin's proposal is basically encouraging the network to give miners more freedom.  The blocksizelimit was actually Satoshi's idea, to protect the network from attack.  Satoshi's concept limited what miners could do, and Gavin is proposing to lift that limit, thereby giving miners more freedom.  And you think Gavin is trying to control the miners?  Do you think miners care what Satoshi's opinion was on the size of their blocks?  Do you think miners care what other miners think about the size of their blocks?  And what if some miners looked to Gavin as a leader in the community?  I'm a miner, and for the moment, it matters to me what Gavin thinks.  I read lots of opinions.  I don't follow blindly.

[teukon]

Thanks for the post Gavin.  There was nothing of substance I disagreed with.  I didn't care for the opening of the "Transaction Fee Death Spiral" section though.

What is written:

The argument for not allowing arbitrarily large blocks: a maximum block size is necessary to create artificial scarcity so transaction fees do not drop to zero, leaving miners with no income, leading to no mining and the death of the network.

How I read this:

The argument for not allowing arbitrarily large blocks: a maximum block size is necessary because miners are willing to process transactions free of charge, but miners are not willing to process transactions free of charge, therefore death of the network.

I see no need to appeal to economic theory here.  Perhaps I misinterpreted the argument.

[piotr_n]

The cruel thing about blockchain economics is that it's not for Gain, nor for any of his friends to decide about sizes of the blocks.
It's the mining,  that he doesn't care of.

So I just wonder what is in this guy's head when he assumes that miners actually care about him and his opinion on a size of their blocks? Who is he actually speaking to with poems like this and what is a point of this?


People who understand the problem  know that the key to make bitcoin really scalable and keeping transactions cheap  is adding a layer of off-chain  transactions on top of the existing protocol.
Even Gavin knows that - but apparently he prefers to have been spending the past two years of his life on developing an 'intelligent' fee discovery system and writing papers on why the blocks must be bigger.

And don't worry - he won't mind me saying it,  I'm on his ignore list.

I can see why you might be on Gavin's ignore list.... your post has a very grumpy tinge to it.  Getting onto your concerns, bitcoin is more than just mining.  The users and people who build services are important too.  Bitcoin is strong not because of some computers, but because of how, as a community, we use those computers.  So it is in important to keep everyone together as much as possible.

pitor_n, it sounds like you are speaking as if Gavin was proposing to create a MaxBlockSize when there was none before.  If some miners started creating 3MB blocks right now, they would just be ignored.  Why is that?


Gavin's proposal is basically encouraging the network to give miners more freedom.  The blocksizelimit was actually Satoshi's idea, to protect the network from attack.  Satoshi's concept limited what miners could do, and Gavin is proposing to lift that limit, thereby giving miners more freedom.  And you think Gavin is trying to control the miners?  Do you think miners care what Satoshi's opinion was on the size of their blocks?  Do you think miners care what other miners think about the size of their blocks?  And what if some miners looked to Gavin as a leader in the community?  I'm a miner, and for the moment, it matters to me what Gavin thinks.  I read lots of opinions.  I don't follow blindly.

There is a negative incentive in mining too big blocks, as they propagate longer thus giving a higher chance of getting orphaned.
You can try to lift the block size limit in the protocol, but even if the miners had allowed you to do it,  they would still be limiting this very size,  because of personal incentives - and that's all a developer needs to know about blocksize economics.

[Crowex]

 This is not a normal market with normal supply and demand.  There are several key differences:

1.      Normal markets do not have a difficulty adjustment every two weeks

A producer of goods and services in any market might have many factors that  would affect the difficulty in producing his goods or services and these factors might change at regular intervals or not. For a farmer it might be weather, disease, interest rates, import/export duties....

2.      In normal markets there is no reason why having a large number of suppliers competing is important

In most free market situations this is quite important to keep costs down, provide choice, drive innovation and
because if one player gets too powerful it can provide opportunities for market manipulation and fraud, which is why there are often laws that don't allow monopolies and cartels.

3.      In normal markets suppliers don’t compete in a bingo like random competition to be awarded the contract

Every time a trawler goes out fishing it might catch nothing or might catch a huge haul of fish. The trawler skipper has to rely on the fact that, on average he will catch enough to cover his costs and make a profit. All business is a calculated gamble, investing time and money betting on the outcome of unpredictable events.

Anyway the point I'm trying to make is that the market of providing the service of getting transactions into a block isn't so different from other markets and it's perfectly valid to analyse it this way.

 I'm not sure I agree with the centralization death spiral argument as the economies of scale still apply whether there is a block size limit or not.
  I agree with piotr_n, there are negative incentives to mining too large blocks. Maybe it would be better to just remove the block limit altogether.

[coindate]

There is a negative incentive in mining too big blocks, as they propagate longer thus giving a higher chance of getting orphaned.
You can try to lift the block size limit in the protocol, but even if the miners had allowed you to do it,  they would still be limiting this very size,  because of personal incentives - and that's all a developer needs to know about blocksize economics.

That's because the system award makes up most of their incomings, the situation will be different when the tx fee go dominant.
So it makes sense to enlarge the block size before that happens.

[teukon]

Anyway the point I'm trying to make is that the market of providing the service of getting transactions into a block isn't so different from other markets and it's perfectly valid to analyse it this way.

It isn't so different but, as jonny1000 correctly observes, the competition mechanics between miners are atypical.  We really shouldn't, for example, make assertions about how mining companies will operate and how mining power will be distributed by appealing to economic theory which assumes that these miners are simply competing for business with people that want transactions confirmed.

However, in the simple case of debunking the Transaction Fee Death Spiral argument, we do not need to consider competition between miners.  Here, we care only about the distinction between some mining and "no mining = network death".  Basic supply and demand certainly does apply: if people desire transaction confirmations, then there will be at least one miner.

I merely maintain that some limit is necessary for economic reasons.  When the block reward falls there is no guarantee that any system will produce enough revenue to miners to secure the network.  An arbitrary level of scarcity in the block size limit (at a much higher level than 1MB), is the best method I can think of for providing this security, even though it is far from perfect.

Just to clarify: You prefer block-space scarcity to assurance contracts as a method of ensuring network security?  Even knowing the contents of Gavin's post?  Could you please link me to the best argument you know of for this?

[DumbFruit]

However, in the simple case of debunking the Transaction Fee Death Spiral argument, we do not need to consider competition between miners.  Here, we care only about the distinction between some mining and "no mining = network death".  Basic supply and demand certainly does apply: if people desire transaction confirmations, then there will be at least one miner.

One miner is equivalent to Bitcoin's death, therefore your "simple case of debunking" does no such thing.

I've been thinking about this problem alot lately and as far as I can see there is no way to solve this problem with a perfectly efficient algorithm because the solution demands A Priori knowledge that can not be known.

In order to accurately expand the Block Size you need to know two things;

1.) Current and future market demand for transactions. (The amount people are willing to pay for a given amount of transaction bytes.)
2.) The current and future hardware costs of supplying those transactions.

Traditionally, this problem has been solved through private ownership of the capital used to provide the service, and allowing consumers to negotiate the market clearing price.

The equivalent of applying this solution to Bitcoin is to remove the block size cap. This would lower the transaction fee to as close to cost as competition can muster, and would necessarily drive out bad competitors.

In most cases, this is fantastic, but Bitcoin's utility is partially derived from artificial inefficiency in the form of redundant competition throughout the world.

Therefore, two things are true as far as I can tell;

1.) Bitcoin is necessarily arbitrarily inefficient.
2.) This inefficiency is a direct trade off between low transaction fees and high centralization and high transaction fees and high decentralization.

A good way to look at it is for every competitor each transaction fee needs to pay for a copy of the outputs on every single competitor. 1000x competitors (nodes) necessarily means at least 1000x higher transaction costs.

If what I have said is true, then Bitcoin users need to come to the sober realization that any solution is not going to be perfect.

See the Economic Calculation Problem, and the Tragedy of the Commons.

[amaclin]

http://en.wikipedia.org/wiki/Metcalfe's_law

[Crowex]

It isn't so different but, as jonny1000 correctly observes, the competition mechanics between miners are atypical.  We really shouldn't, for example, make assertions about how mining companies will operate and how mining power will be distributed by appealing to economic theory which assumes that these miners are simply competing for business with people that want transactions confirmed.

Yes, that is true. At the moment we are not in a situation where the transaction fees have great economic relevance.

 However I could imagine a future when miners might offer reduced fees if they have the exclusive opportunity to mine a customer's transactions. Or online wallets partnering with miners who would exclusively mine their transactions.
 Transactions don't have to be broadcast to everybody.

 If transactions fees become more economically relevant then I would not be surprised if competition will occur in the same way that it does in other markets.

[jonny1000]

Let take a step back and consider why we even have the option for transaction fees in the first place?  In this post I attempted to outline four possible reasons:

https://bitcointalk.org/index.php?topic=669243.0

If there is no scarcity in block space and it is not to incentivise miners then the primary reasons for the transaction fee seem to disappear.  In reality this is not the case, transaction fees are an integral, necessary and crucial part of Bitcoin.

The market for transaction fees can be considered in some ways which make it appear to look fundamentally different from a “normal” market from certain perspectives, however I understand many viewpoints expressed which imply it is a “normal” type of market.  I personally struggle with both of these concepts a lot.

Lets consider again the two primary “reasons” for the transaction fee: (Once the block reward becomes low)



All users of Bitcoin benefit from a secure network and a tragedy of the commons type problem could occur, in the sense that “Why should I pay a transaction fee for my transaction, when others could pay the higher fee to incentivise miners?”  This problem clearly occurs for the second "reason" and not the first.

A market based mechanism could work for the first “reason” for the transaction fee, but not the second.  This is why I propose that an artificial scarcity is required, this would manipulate the otherwise “normal” market for transaction fees, such that mining incentivisation occurs as an indirect benefit of users wanting their transactions included in blocks.  The arrow in the above image represents the transactions fees being used to incentivise an activity/"reason" which the users are not directly paying for.

I understand that in many people’s minds, the distinction between “reasons” for the transaction fee is not necessary or relevant and mining is just mining, whatever the "reason".  However, I think the above illustrates the economic dynamics of the transaction fee is not as simple as some people think and the network can be analysed in a number of valid ways.  When the block reward does fall to an economically low level I do not know what will happen, but it is important to try to look at the system from many different angles now to prepare for this.

[coindate]

http://image.bayimg.com/4e6822913d1c31c0a756bca4ca4188e35d0443c8.jpg

All users of Bitcoin benefit from a secure network and a tragedy of the commons type problem could occur, in the sense that “Why should I pay a transaction fee for my transaction, when others could pay the higher fee to incentivise miners?”  This problem clearly occurs for the second "reason" and not the first.

From the miners' point of view, they want to profit from mining, so they won't include txs without enough fee, and the fees makes the network safer, so the reasons for transaction fee are actually coinside of each other.

[gglon]

From miner perspective, in the short run, it is most profitable to include tx with fee that is grater than penalty from broadcasting this transaction. As I understand, as soon as block O(1) propagation is included this penalty is greatly reduced and amounts to almost 0. Therefore if miner is greedy it is best to include almost all transactions with non-zero fee as soon as possible. This however will destroy fee market -> fees will collapse to 0. This is not free market. Good well known analogy is situation when there is an unexpected surplus of some goods eg. food in the wake of some international sanctions. It is then not possible to sell all goods in the domestic market as the demand is not enough even if the wholesale price is 0. Therefore artificial limitation (destroying crops) is crucial for securing farmer revenues.  

From miners perspective the best limitation is the one that brings miners the biggest revenue. That is probably when the fee is competitive with the other means of payment for small to medium types of transactions as those types of transactions are the most common ones. Therefore the ideal tx fee should probably be somewhere between $.1-$1.

If miners cooperate they have the power to limit soft maxblocksize by simply not mining larger blocks. But is this kind of cooperation possible in the view of the fact that greedy miner would mine all non-zero txs and have higher revenue? I think Gavin thinks it is possible so the hard maxblocksize role should be different. It should be just an upper limit of soft maxblocksize that miners can  impose and that would guarantee network decentralization.

[teukon]

From the miners' point of view, they want to profit from mining, so they won't include txs without enough fee, and the fees makes the network safer, so the reasons for transaction fee are actually coinside of each other.

Yes, this is understood.  jonny1000 is intentionally breaking the roles and incentives down for analysis.

It is true that the fees make the network safer, but the big question is whether or not they provide the correct level of safety?  What if the fees are low and the network is vulnerable?  What if the fees are forced to be very high through some selection of magic numbers baked into Bitcoin but this generates too much safety (is wasteful of resources).

The eventual realisation of this study is that rules to do with fees and block scarcity do affect safety but not in a robust way.  There is no price mechanism and it's easy to undercook or overcook the safety level.  Given that market solutions exist "assurance contracts", it seems that the burden of proof rests with those that would jerry-rig the fee system to provide an appropriate level of safety.

[teukon]

From miner perspective, in the short run, it is most profitable to include tx with fee that is grater than penalty from broadcasting this transaction. As I understand, as soon as block O(1) propagation is included this penalty is greatly reduced and amounts to almost 0. Therefore if miner is greedy it is best to include almost all transactions with non-zero fee as soon as possible. This however will destroy fee market -> fees will collapse to 0. This is not free market.

No, this is free market.  Here, an innovation has increased the efficiency of mining and this is communicated to the customer via the price system.  All exchanges involved are voluntary.

[gglon]

No, this is free market.  Here, an innovation has increased the efficiency of mining and this is communicated to the customer via the price system.  All exchanges involved are voluntary.

My bad. Of course it is free market. It is not preferable type of market for miners though, since their profit from tx fee tends to 0. This in the future, as previously noted by @jonny1000, won't allow to sustain large amount of mining power, decrease Bitcoin network value and create death spiral.

[NewLiberty]

Let take a step back and consider why we even have the option for transaction fees in the first place?  In this post I attempted to outline four possible reasons:

https://bitcointalk.org/index.php?topic=669243.0

If there is no scarcity in block space and it is not to incentivise miners then the primary reasons for the transaction fee seem to disappear.  In reality this is not the case, transaction fees are an integral, necessary and crucial part of Bitcoin.

The market for transaction fees can be considered in some ways which make it appear to look fundamentally different from a “normal” market from certain perspectives, however I understand many viewpoints expressed which imply it is a “normal” type of market.  I personally struggle with both of these concepts a lot.

Lets consider again the two primary “reasons” for the transaction fee: (Once the block reward becomes low)



All users of Bitcoin benefit from a secure network and a tragedy of the commons type problem could occur, in the sense that “Why should I pay a transaction fee for my transaction, when others could pay the higher fee to incentivise miners?”  This problem clearly occurs for the second "reason" and not the first.

A market based mechanism could work for the first “reason” for the transaction fee, but not the second.  This is why I propose that an artificial scarcity is required, this would manipulate the otherwise “normal” market for transaction fees, such that mining incentivisation occurs as an indirect benefit of users wanting their transactions included in blocks.  The arrow in the above image represents the transactions fees being used to incentivise an activity/"reason" which the users are not directly paying for.

I understand that in many people’s minds, the distinction between “reasons” for the transaction fee is not necessary or relevant and mining is just mining, whatever the "reason".  However, I think the above illustrates the economic dynamics of the transaction fee is not as simple as some people think and the network can be analysed in a number of valid ways.  When the block reward does fall to an economically low level I do not know what will happen, but it is important to try to look at the system from many different angles now to prepare for this.

Transaction fees also serve an anti-spam (by non-miner) function.  Miners can of course insert as many transactions into their own blocks and only pay themselves the fees, or no fees, limited to the MAX_BLOCK_SIZE.

Gavin's paper does a very good job of describing some of the motivations and incentives in the blocksize economics.  It does not address any of the economics outside of the bitcoin economy which may have incentives to do things against the interests of the success of Bitcoin network.  It may make sense to mention some of these for purpose of example as a 3rd.

Allowing arbitrarily large block size could result in non-economic acts, or rather economic acts within a larger economy than is constrained to the scope of the Bitcoin economy.
For example with an arbitrarily large MAX_BLOCK_SIZE, a mining group hostile to the success of Bitcoin with 1.1% of the network could feasibly solve 1.45 blocks per day on average.  If they did so with very large blocks, such that fully 1/3 of their blocks were orphaned and was sending 1GB blocks, this could grow the block chain by 1GB per day.
Such activity could isolate hobbyists, many of the small nodes, as well as much of the Southern Hemisphere due to bandwidth constraints.  So some nationalist interest that seeks to advance its Bitcoin presence WRT what it may see as its competitors could accomplish this.

Also there do exist "Free" computing and network resources (resources not paid for by those that control them), such that mining activity can be conducted by non-economic actors at below the bandwidth and computing costs.  The deployment of such "Free" resources cuts into the profitability of ALL miners.  Thus making marginal cost = zero for this population.

[solex]

For example with an arbitrarily large MAX_BLOCK_SIZE, a mining group hostile to the success of Bitcoin with 1.1% of the network could feasibly solve 1.45 blocks per day on average.  If they did so with very large blocks, such that fully 1/3 of their blocks were orphaned and was sending 1GB blocks, this could grow the block chain by 1GB per day.

Absolutely, but scaling the block size according to the projected bandwidth growth rate (based upon the recent bandwidth growth rate) means that 1GB blocks are still many years away, mid-2020s. Storage shows phenomenal potential for further growth and durability. http://physicsworld.com/cws/article/news/2013/jul/17/5d-superman-memory-crystal-heralds-unlimited-lifetime-data-storage

In ten years time the block reward will be 3.125 btc, and the fees market much healthier, such that a dynamic "soft" limit could be introduced, if that is considered needed. You mention doing the right thing for future generations of Bitcoin users. We all want to do the right thing for them, but we have to concede that they will have far more information to make design and scaling decisions. What is important now is to introduce *some* scalability, ideally so that the ecosystem can grow, as much as computing technology (in the hands of a hobbyist or semi-professional bitcoiner) will allow, in the meantime.

[NewLiberty]

For example with an arbitrarily large MAX_BLOCK_SIZE, a mining group hostile to the success of Bitcoin with 1.1% of the network could feasibly solve 1.45 blocks per day on average.  If they did so with very large blocks, such that fully 1/3 of their blocks were orphaned and was sending 1GB blocks, this could grow the block chain by 1GB per day.

Absolutely, but scaling the block size according to the projected bandwidth growth rate (based upon the recent bandwidth growth rate) means that 1GB blocks are still many years away, mid-2020s. Storage shows phenomenal potential for further growth and durability. http://physicsworld.com/cws/article/news/2013/jul/17/5d-superman-memory-crystal-heralds-unlimited-lifetime-data-storage

In ten years time the block reward will be 3.125 btc, and the fees market much healthier, such that a dynamic "soft" limit could be introduced, if that is considered needed. You mention doing the right thing for future generations of Bitcoin users. We all want to do the right thing for them, but we have to concede that they will have far more information to make design and scaling decisions. What is important now is to introduce *some* scalability, ideally so that the ecosystem can grow, as much as computing technology (in the hands of a hobbyist or semi-professional bitcoiner) will allow, in the meantime.

We agree on the need for scaling up.  The notion of arbitrarily large sizes was from Gavin's article.  He cited a few problems with it, and dismissed them, this was additional to those he mentioned in the article.

In the ideal case, we could devise an automatic and mechanical way to increase MAX_BLOCK_SIZE, so that we are not creating a central authority to decide what the MAX_BLOCK_SIZE ought be, and won't have to rely on their criteria of the day.  The information that the future generations of Bitcoin users will have, will come from the block chain, which we also have today, telling us what is needed today.  If we are able to solve the problem in a way that accomodates what the block chain is telling us, it may never need to be solved again!  It may be able to adjust to changing environment similarly to the way the difficulty does.