Change in difficulty after block award drops from 50 to 25 BTC?

[chickenado]

Curious to know what the community expects to happen. I predict that it will drop more than 80%, because the competitive nature of the mining market will mean that the majority of miners are just barely profitable, and a drop in net income by 50% will make most of them unprofitable.

[dannyjpw]

doesn't that rather depend on the USDBTC rate given that eleccy and rigs are priced in USD?

[theGECK]

What's the timeline for this poll? 1 month afterwards? 1 week afterwards?

[FreeMoney]

Curious to know what the community expects to happen. I predict that it will drop more than 80%, because the competitive nature of the mining market will mean that the majority of miners are just barely profitable, and a drop in net income by 50% will make most of them unprofitable.

I agree that miners will tend to be only marginally profitable considering all costs, but once they get set up only marginal cost considerations matter. On top of that when some drop out it becomes more profitable for the rest. 50% is a hard ceiling for the drop imo, I choose 5% to 30% as most likely.

One more thing to consider is that if coins are harder to get the price could pop a little around the same time, or before in anticipation.

[freeman]

[..] the competitive nature of the mining market will mean that the majority of miners are just barely profitable, and a drop in net income by 50% will make most of them unprofitable.

That's what I originally thought too and was going to argue that the reward should have dropped off gradually from 50.  However, now I wonder if the free market forces will handle this sharp jump without difficulty.  Couldn't people with marginal profits start charging small transaction fees?  I guess the transaction rate would have to be high enough to ensure that bitcoin users could support those fees.

Some quick math (please correct if I'm wrong).  To make the extra 25 per block, assume the miners charge a fee based on the amount transfered.  For 2500 in transactions, the fee would be 1%.  If 25,000 happened in 10 minutes then only 0.1% would have to be charged. Obviously we don't know how popular the system will be at that time but it seems possible that transaction fees could make up the difference. 

[FreeMoney]

[..] the competitive nature of the mining market will mean that the majority of miners are just barely profitable, and a drop in net income by 50% will make most of them unprofitable.

That's what I originally thought too and was going to argue that the reward should have dropped off gradually from 50.  However, now I wonder if the free market forces will handle this sharp jump without difficulty.  Couldn't people with marginal profits start charging small transaction fees?  I guess the transaction rate would have to be high enough to ensure that bitcoin users could support those fees.

Some quick math (please correct if I'm wrong).  To make the extra 25 per block, assume the miners charge a fee based on the amount transfered.  For 2500 in transactions, the fee would be 1%.  If 25,000 happened in 10 minutes then only 0.1% would have to be charged. Obviously we don't know how popular the system will be at that time but it seems possible that transaction fees could make up the difference. 

I don't know if it changes your analysis, but miners can't charge fees, they can only decide not to include a transaction if they don't think the fee is high enough. I just want to make sure no one gets the wrong idea about fees.

[freeman]

I don't know if it changes your analysis, but miners can't charge fees, they can only decide not to include a transaction if they don't think the fee is high enough.

Yes, that was my understanding.  Hopefully I'm not talking complete nonsense but I don't think that actually changes my analysis.  Basically  a large group of miners could get together and decide among themselves "we will not process blocks unless we are paid at least <some fee scheme> for transactions".  Supply and demand would determine what those fees are going to be.  It's a little messy since the clients have to figure out what fees are necessary.

My naive understanding is that the halving of the mining bonuses is Satoshi's way of weaning the miners off the inflation gravy train and transitioning to a transaction fee based system.  It very well could be that 25 coins is still sufficient payment for running the processing machines without requiring transaction fees.  It all depends on the demand for bitcoins at that point.

[chickenado]

What's the timeline for this poll? 1 month afterwards? 1 week afterwards?

The reward will drop from 50 to 25 @ block 210,000

The first difficulty change after the drop will take place @ block 211,680

That's 1680 blocks, or approx. 11 days afterwards.

[MoonShadow]

My naive understanding is that the halving of the mining bonuses is Satoshi's way of weaning the miners off the inflation gravy train and transitioning to a transaction fee based system. 

That's correct, but it's also how the system limits the total bitcoin that will ever be, as the block reward halves every 210,000 blocks, the total number of bitcoins issued approach a  mathmatical limit, never quite reaching 21,000,000 bitcoins.  At the inflection points, the difficulty will almost certainly drop, but by that time the value of those 25 bitcoins will likely be much higher than 50 bitcoins is right now.

[FreeMoney]

I don't know if it changes your analysis, but miners can't charge fees, they can only decide not to include a transaction if they don't think the fee is high enough.

Yes, that was my understanding.  Hopefully I'm not talking complete nonsense but I don't think that actually changes my analysis.  Basically  a large group of miners could get together and decide among themselves "we will not process blocks unless we are paid at least <some fee scheme> for transactions".  Supply and demand would determine what those fees are going to be.  It's a little messy since the clients have to figure out what fees are necessary.

My naive understanding is that the halving of the mining bonuses is Satoshi's way of weaning the miners off the inflation gravy train and transitioning to a transaction fee based system.  It very well could be that 25 coins is still sufficient payment for running the processing machines without requiring transaction fees.  It all depends on the demand for bitcoins at that point.

This is essentially the most difficult to enforce cartel in all of history. Even if the miners hook up all the generation they have now to an organizing pool server or something if it is more profitable for an individual to include transaction below whatever the cartel chooses then that will be set up on the side either by 'cheating' cartel members or by other's enticed by the artificial profits that the cartel induces. There simply is no cartel equilibrium here.

I don't see any way that difficulty falls by more than 50%. So what if you used to get 9 blocks of 50BTC/mo and now you get 18 of 25BTC? It will be exactly as profitable for you as before.

[freeman]

There simply is no cartel equilibrium here.

I'm not an economist but I don't think a cartel is required.  Assume more then 50% of the miners are not profitable with no transaction fees after the drop to 25 BTC.  They could drop out of the mining business completely (and try to liquidate their GPUs, etc) or they could require a certain fee to include each transaction in their next block.  Each miner could determine for themselves, "I won't include a transaction unless I get X fee schedule". I'm a little sketchy on the details but I think the market would naturally determine a fee schedule.  The client software would obviously have to be adapted to handle the emergent fee schedules before the halving.

After more thinking, it seems to me that the drastic drop in the block reward is required to bootstrap the transaction fee market.  If the reward gradually decreased from 50 BTC to 0 then I believe the miners would gradually stop participating.  The block difficulty would adapt but we would be left with a vulnerable system due to lack of total processing power.

[mimarob]

Fascinating discussion!

Either miners would liquidate a number of gpu's but an other option (or the equivalent behaviour) would be to launch a service where gpu's are offered as general computing power for anyone who needs a lot of computation, the natural choice of payments would of course be bitcoins!

Another possibility is of course that someone starts up a bitcoin fork and many gpu's goes there instead...

I just read a blog statement that the bitcoin economy is now consuming 6 MW of power...

I think the intention of the author was to first generate a critical mass number of coins for an economy and then gradually transitioning this economy into something useful, like selling storage, c/gpu power, network bandwidth and also physical goods online.

[theGECK]

Necro'ing this topic because I found the results of the poll from over two years ago interesting. Did you remember that 80% of us thought the difficulty would drop after the reward dropped from 50 to 25 Bitcoins? We were very, very wrong.