Would Bitcoin fall off a cliff if it dropped to $100 or $150?

Topic: 
Tags: 

Bitcoin's been on a long decline over the past year, and today is around $220 per coin. The value has always been based on speculation about Bitcoin's future value, not its present value, so it's been very hard to predict and investment in the coins has been risky.

Some thinking led me to a scary conclusion. Recent news has revealed that a number of "cloud mining" companies have shut down after the price drop. Let me explain why.

Over time, all bitcoin mining has been done using specialized ASIC hardware. The hardware is priced so that you can make a decent but not ridiculous profit with it. All the bitcoins mined go mostly into paying for mining hardware and electricity -- much less goes into profit for the miners. In the past, the electricity was the big cost, but mining hardware got fast enough and expensive enough that most of the cost of mining has been paying off your mining hardware, with electricity dropping to being 20% or less of the cost.

In other words, most of the 3600 btc/day mining revenues of the bitcoin system have been going into the people making mining chips and rigs, but that's another story.

With the drop in price, electricity is back up to being half your cost. That puts a squeeze on the cost of mining equipment. With cloud mining, as with Amazon Web Services, you rented mining equipment and power by the hour. People who bought their mining equipment will still run it as long as the revenue is more than the operating cost. For cloud mining, you need the revenue to exceed the operating and capital cost, because the capital costs are amortized into the operating cost. While cloud mining companies could cut their fees to cut their losses, some have instead just left the business. As noted, those who bought mining equipment are running it now at less profit, but as long as the mining brings in more than the electricity cost, it's still worth running -- the mining gear is all paid for, and even though you will never make back your money, it's worse if you shut it off.

You can get a good analysis of the cost and profitability of mining rigs at this mining calculator.

What if a panic dropped a bitcoin under $100?

It's not out of the question that a sudden panic might drop Bitcoin quickly down to $100. It probably won't happen, but it certainly could. At this point, with current generation mining equipment, most miners then see their revenue drop below the cost of electricity. If they are rational and strictly profit-oriented, they cry into their beer and turn off the mining rig. And the cloud miners have already done that, and some other miners have done the same sooner than they expected, and the network hashrate (the measure of how much mining power there is) has had minor sustained drops for the first time in years.

(It's worst than this. Even at $150, all but the most recent mining rigs become unprofitable to keep turned on, and so a major drop would happen with much less of a drop needed. New mining equipment expected to ship in the next few months is profitable at even lower prices, though.)

The way Bitcoin works, when they turn off the rig, it doesn't mean more coins for the other miners. Bitcoin sets the reward rate with a "difficulty" number that makes the Bitcoin lottery problem harder the more mining capacity is out there. Your reward rate is a strict function of the difficulty and the power of your miners.

Every 2016 blocks, the difficulty adjusts based on how much capacity seems to be mining. Under normal operations, 2016 blocks is two weeks, as long as people are mining at the rate seen in the 2 weeks prior to setting the current difficulty. If large volumes of miners shut off their rigs as non-productive, the mining rate would crash. The wait for a new difficulty could be not just two weeks if this happened at the wrong time, but 4 weeks if half the miners shut down, or 8 weeks if 3/4 of them left. In terms of the Bitcoin world, it's effectively forever, and long before that, confidence in the coin price would probably drop further, causing more miners to shut off their rigs. Only dedicated fans willing to lose money to preserve the system would keep mining.

In such a panic, the Bitcoin Foundation and others might propose an emergency modification of the Bitcoin software base which is able to do an emergency reduction of the difficulty number. Alternately they could propose bumping the mining reward back to 50 coins instead of 25. This would still take days, which I think is too long. But if they did, it's a sticky issue. As soon as you drop the difficulty enough, all those miners come back online, and now the difficulty is too low. To do it right, an estimate would have to be made of how much mining capacity is cost effective and set the difficulty so that only some of the miners come back online, a number tied to that difficulty. For example, one might look at the various mining rigs out there, and set the difficulty such that they are (barely) profitable while others are not. Problem is, the profitability depends on the price of a bitcoin, which will be wildly fluctuating. It's not clear how to solve this.

If the electricity cost exceeds the reward, but you still want bitcoins for future investment, the rational thing is not to mine, but to just buy bitcoins on the exchanges and keep the price up.

What would happen after such a collapse? Could it be stopped?

The collapse would probably spread to altcoins, but some might survive and become successors to Bitcoin. In addition, there are many people devoted to Bitcoin who would continue to mine, even at a loss, to get it back on its feet. After all, the early years of Bitcoin, all mining was at a loss, though it turned into a huge bonanza later and was a wise idea in hindsight. With the large number of well funded companies in the space, we could see companies willing to maintain unprofitable mining for some time if the alternative is the destruction of the thing they've based their business on. They might even buy up the rigs of failed miners, or pay them to mine. Perhaps, if they are ready, they could heed the warning in this message and make contracts with enough miners to say, "we'll pay you to keep mining if a collapse happens."

Alternately, Bitcoin users and boosters could just start deliberately leaving large transaction fees in their transactions to make the cost of mining worthwhile again. While hard to sustain long term, it is in their interest to spend their bitcoins to keep the mining system going, since those coins probably drop immensely if it falls down. It also keeps faith in the mining system since if the coin owners ran the miners, they might corrupt the network with that much power. It should be noted that it's always been part of the plan for Bitcoin that higher transaction fees would arise as the coinbase rewards dropped, but not this early, and because the reward dropped in btc, not dollars.

The subsidy would have to be enough to overcome losses and provide a modest or even very small profit. The network cost pays 3600 bitcoins/day in mining fees (or $360K at $100/bitcoin.) The subsidy might be more in the range of $50K or $100K per day -- affordable to keep the network alive for up to 14 days to survival.

Another idea would be to develop a way to make the difficulty more dynamic, or provide some mechanism for an emergency reduction. (An emergency increase would mean something was really wrong and would probably also mean somebody had more than half the mining capacity, another must-not-happen.)

What sort of events could cause such a huge drop, to 45% of the current value? That's not been seen in a short time, but a big political event, such as a suggestion the USA or EU might forbid or impede Bitcoin could do it. But there are many other things that can cause panic. A shutdown of exchanges (a common technique in stock market panics) would probably do little, as there are exchanges all over the world and all will not shut down. A call to miners to sacrifice might work, at least for a while, to allow time to fix the problem.

Latent mining capacity

Mining rigs are shut down all the time as non-profitable, but in the past that's always been because newer, better rigs were out there dominating the mining space and pushing up the difficulty. It would be a new idea to have rigs shut down because the dollar price dropped. When such rigs shut down, they would not be permanently useless, and unless torn down, they would be able to restart at any time. For example, if the difficulty dropped (because they all shut down) they would all start running again, and blocks would come out faster than intended. Then, 2016 blocks later, the difficulty would be recalculated up again -- and they would stop again. Miners would also start and stop based on the day's price as well, and the price might even swing around the expected rises and drops in difficulty. This seems like it would be chaos.

Once the electricity cost dominates, the important metric in mining equipment is not gigahashes/second, but gigahashes per joule. At 10 cents/kwh, you need around 2 gigahashes/joule to beat the electricity cost with $100 bitcoins and today's difficulty number. At today's $220 bitcoins, 0.9 gigahash/joule will do. Most miners are under 2, but there are some that do close to 3, and there is the promise of 5. If the trends in the rest of computing are an indicator, operations per joule will eventually level off, even as transistor counts continue to increase. If that happens we will stop seeing big increases in mining power and the upward spiral would end.

Comments

Looks to me like the bubble has almost burst. Ironically, some people saw Bitcoin as an alternative currency free from dangers which real currencies suffer from, but it seems that the concept has eaten itself.

This danger is hardly one that other currencies suffer, and perhaps it can be planned for. Most of the alternate coins vary in their mining plan, and the "proof of stake" mining systems in theory should not suffer from the risk of mining becoming more expensive than the reward. Likewise the solar randomness system I proposed earlier.

Is there a danger that as profit-seeking miners exit, there's an opening for cost-insensitive players (such as national governments or billionaire investors) to essentially take over the Bitcoin network, by becoming the majority of players remaining - at least temporarily? What would be the implications: price manipulation, destroying credibility through double transactions, etc?

If the Bitcoin Foundation changes the rules in reaction to fluctuations in price, does that make them a de facto Federal Reserve of the currency? Isn't the point of Bitcoin to be decentralized?

It does seem that the system is set up to drive old hardware offline or force upgrade spending. Dips or crashes in price have the effect of batching these upgrades into clusters. Is this issue likely to be mitigated if the network grows large or widely used enough?

Yes, if most miners abandon things then miners willing to take a loss could take over -- but if they did, it would not gain them much, it would instead destroy confidence in Bitcoin. You can't alter the past by taking over the mining network, what you can do is double spend.

That's my my proposal that the holders of large amounts of bitcoin not become the miners, but instead just put in transaction fees (which get awarded to the next winning miner.) That way nobody doubts the integrity of the mining system, though they know it vanishes if the subsidy stops.

Or maybe change the rules. The Bitcoin Foundation has no power to change the rules, but they are the place through which most proposals to change the code (and thus the rules) happen. As such, they have the best chance to convince all the miners to switch to some new code base. In an emergency, I can't see who else could do it.

Yes, all the miners. Or rather, so many, that the other miners, sticking to the old code base, see themselves in a minority fork that is a waste of their time, and so they switch to the new one.

The system has always driven old hardware offline as it has grown. The problem I am discussing is what happens if it shrinks too fast.

Brad,

Silly Question: Are you a miner?

I do not mine, though I have in the past joined some pools. Not for some time. As I have written elsewhere, I am in the camp that is bothered by proof of work (ie. proof of waste) as the foundation on which the real thing (proof of luck) is based.

This loop mechanism has been identified before, but it's not a problem.

> The wait for a new difficulty could be not just two weeks if this happened at the wrong time, but 4 weeks if half the miners shut down, or 8 weeks if 3/4 of them left.

Yes, but consider the other blade: if blocks aren't being mined, that also means that mining rewards aren't happening to dilute the monetary base further, there's less risk of mining a stale block, and if the drop is that catastrophic, continuing to mine means you get to reap the reward of being an active miner when the adjustment happens.

In practice, we've seen this sort of adjustment freeze affect some of the smallest altcoins. What happens is that the altcoin just works through it (the remaining miners keep at it, the adjustment happens, and all is well) and it's a nuisance but not fatal - unless the drop in mining is due to an exogenous factor like the altcoin being discredited or revealed to be a scam, in which case the chain stops updating but the adjustment freeze was, ultimately, only a proximate cause of death.

> In such a panic, the Bitcoin Foundation and others might propose an emergency modification of the Bitcoin software base which is able to do an emergency reduction of the difficulty number.

That sort of hardfork would itself risk killing Bitcoin: if you change the fundamental rules about mining rewards, you could also change the rules about the mining reward diminishing and the supply of Bitcoins ultimately being fixed at 21m, and such a move would demonstrate that Bitcoin had become de facto centralized.

A major revision to the bitcoin rules is indeed super difficult. But I could see it happening if it's abundantly clear that the system is collapsing if nothing is done. Many of the miners own stocks of bitcoin which they can't afford to see become devalued, and of course they have mining equipment which is useless if there is a total collapse. (This is, or was, one of the arguments in favour of scrypt mining, that it was done on hardware that still has another purpose.)

Another thought was that the fork would be to return to awarding 50 coins per block instead of 25. This would instantly make mining highly profitable again for more recently purchased rigs, and is a less radical tweak than having an ability to alter the difficulty.

The altcoins have survived, and bitcoin of course had all its miners doing it unprofitably for the first few years, but today it's different. Today bitcoin mining is a business, and strictly business for a lot of the miners. It is possible they might mine at a loss because the collapse of Bitcoin would mean a loss of their remaining mining investment, but it is the holders of bitcoins who have the strongest motive (and thus might offer fat transaction fees.)

I don't think it's true that continuing to mine lets you reap the reward of mining when the adjustment happens. You can turn your miner back on any time and start mining, including just when the adjustment block arrives.

But this is also a hard to understand situation. If people shut off mining rigs en masse, and hashrate drops, it's not like the hardware goes away. If an adjustment block suddenly reports a much lower difficulty, then all that turned off gear turns on again, and blocks start appearing fast and furious, and then another adjustment and bam, it's not economical again. That sounds chaotic.

Also, can you point me to some prior threads about this. The phrase "loop mechanism" does not seem to find anything.

> Another thought was that the fork would be to return to awarding 50 coins per block instead of 25. This would instantly make mining highly profitable again for more recently purchased rigs,

Only if you assume the price would not react at all and you can inflate the monetary supply for free. Which it would, and so you cannot. This move would increase the total supply from 21m to... not sure, maybe another 10m? Or no, it started at 50 coins so it would essentially be a reset, so it'd double the total supply to 42m. That would bake in a price decrease of 50% to offset the increased reward - helping no one. (Now you have to sell 2 coins rather than 1, and any problems with mining remain...)

> and is a less radical tweak than having an ability to alter the difficulty.

No, changing the block reward would *way* more radical than, say, tweaking the difficulty to reset every few days rather than weeks! It is an attack on one of the key properties of Bitcoin: a decentralized fixed-supply currency. At that point you might as well implement a blacklist mechanism too and declare the Bitcoin dream dead, and everyone will move on to an altcoin.

> I don’t think it’s true that continuing to mine lets you reap the reward of mining when the adjustment happens. You can turn your miner back on any time and start mining, including just when the adjustment block arrives.

You can mine a private chain. Think of one of the big miners with say 25%; they can mine on a private fork, they'll catch up in a few weeks, the difficulty will ratchet down, and they can mine easier blocks for a while, then at some point go public with weeks of mining-rewards locked in; normally this would be idiotic because the network would ignore low-difficulty blocks in favor of longer/harder chains, but in your hypothetical everyone's stopped mining, so there is no competing chain with more work in it and the fait accompli will work. And others can't jump in before a lot of rewards have been claimed because they don't have access to the private fork.

(If this sounds strange to you and like it couldn't happen - good. You're posing a strange hypothetical which I don't think could happen.)

> That sounds chaotic.

It's not chaotic. If a freeze somehow happens due to some sort of exogenous shock and a private fork is made to fix the adjustment, then when it goes public, mining will be profitable again, hashpower will turn on, new blocks will happen, transactions can resume, and a new equilibrium has been reached.

> Also, can you point me to some prior threads about this.

Sorry; I read about the altcoin instances on Bitcointalk but I didn't save the specific threads or discussions because it was obviously only an issue for a coin which was already dead and of no real-world relevance. (The one relevant attack was big mining pools mining a small altcoin for a few adjustment periods, driving up the difficulty skyhigh and forcing either a freeze of possibly months before the altcoin miners could mine enough blocks to adjust difficulty back down or a blockchain reset/fork back before the attack. But by its nature, the damage is limited to small altcoins.)

No, I would not propose a permanent reset to 50. The idea would be a "get past the crisis" reset. Though it is hard to figure out how to do it.

The problem is that while Bitcoin's plan is for the mining reward to be in bitcoins, miners (thanks to the move to professional ASIC mining) look at the reward in terms of fiat currency. Miners sell their coins quickly for fiat currency to pay mining expenses. Mining companies price their products based on what the market will pay, which is based on the fiat currency reward. (Some miners can be bought with bitcoins, but typically converted immediately to fiat like most transactions.)

What I am worrying about is whether the bitcoin mining system can handle well the chaos that comes when its incentives are tied to the volatile exchange rates with fiat.

I don't understand your private chain strategy. With miners shut off, remaining miners mine up to 2015 blocks at the unprofitable difficulty, taking longer than 2 weeks to do so. You don't know the final new difficulty number until the last block is mined, so how can you build your private chain? You don't know the hash of the final block until it arrives.

I don't think that everybody has stopped mining, but I could easily see half the people stopping, which is enough for a giant problem. (The difficulty won't drop more than by 75% in any event.)

Brad:

You're pointing out some things that seem true, that I haven't thought through before, and I don't see any flaw in your reasoning, but I also side with Gwern that it wouldn't be a doomsday scenario for Bitcoin. I think that where you and I differ is on the "doomsday" question is just that I think Bitcoin has a great deal of intrinsic usefulness and a great deal of goodwill, and so even though the mechanism you describe seems real, I don't imagine the Bitcoin community caving in, even if it were to happen.

Anyway, let me muse out loud about the mechanisms you've mentioned. If the price of Bitcoin drops, then some fraction of the current miners stop mining. The graph of "how many miners will mine vs. the price of ?" should be monotonic, but other than that it could be full of surprises and non-linearities — there are probably several thresholds of the price which, if crossed, would each trigger large swathes of the current miners to turn off their machines.

Now, you're imagining a positive feedback loop in which miners turning off machines stimulates a further drop in the price. That part I'm not so sure about — like I say, I think of the Bitcoin community as more patient and optimistic than that.

One thing that you didn't mention is the effect on transaction confirmation times. If mining dropped to 1/10 of its current power then, until the next difficulty adjustment, blocks would take on average 100 minutes instead of 10 minutes. People waiting for six confirmations would wait on average 10 hours instead of 1 hour (with a long tail of even longer waits). I'm pretty curious how much of a difficulty that would be in practice.

I think a lot of people would keep mining to support Bitcoin. But the problem is that I get the sense that most mining is now "professional" and driven by fiat rewards.

I think even half the miners turning off and blocks taking 20 minutes, the the difficulty adjustment now up to a month away is a very unstable state. I think it would push a lot of speculators to sell their bitcoins. (Though miners would not be selling bitcoins in the same volume, and some attribute miner sales as the cause of the price drops.)

If it does, it's a death spiral. We might not fall down it but we don't want to be poised on the edge of it. Perhaps the believers rally and snatch up cheap bitcoins and push the price back up. That part's hard to predict.

What's not hard to predict is that mining which can't pay for the cost of electricity is not sustainable in today's world of professional ASIC miners. Something has to change -- the reward has to go up (in dollars,) or the difficulty has to be very precisely adjusted to bring back enough profitable miners to drive the system while not bringing back all of them (which just bumps up the difficulty to unprofitability again.) Of course, the other factor is the arrival of new mining hardware. Vendors will see new mining hardware (and have it in the pipeline) that does more GH/joule and is profitable on an electricity basis, but a real risk to buy in terms of being profitable on an amortization basis. This is very hard to model.

My comments about the Bitcoin community's resilience were not suggesting that a substantial amount of mining power would continue to run at a loss. They were suggesting that maybe the *price* wouldn't fall. There's a proposed positive feedback loop: price decline ? mining power reduction; mining power reduction ? price decline. It is the second part that I'm questioning.

I've been chewing on your idea more, ever since I read it. One disturbing fact is that Bitcoin mining probably always drives itself to this "precipice" state, over time. Over time, the fraction of the aggregate hash power which is perched on the razor's edge of marginal profitability grows.

It's been a while since I interacted with software that doesn't accept non-ASCII chars. Thanks to your blog for taking me back. Here's the ascii-art version of that part of my comment in which my unicode "rightward arrow" char (U+2192) got translated into "?":

price decline -> mining power reduction; mining power reduction -> price decline

I don't say a feedback loop like that is assured. Rather, I don't see a pressure against it. My fear is that a serious departure of miners, one so large that the blocks started coming noticeably slower could cause a panic among the speculators.

It is an uncomfortable truth that bitcoin's fiat value is almost entirely due to speculators. This is not an unusual thing. Tesla's value is also due to speculators. People are speculating on what bitcoin could be worth in the future, not saying it has already attained 3 billion today. But as I said, this is normal. But it also means if the speculators smell trouble in the wind, they can bolt, because nobody wants to be the last to get out. Investments even fall below their true current values when investors don't think they will realize them. (That's usually because of lack of faith in management, not an issue with Bitcoin.)

Either way, though, there's a price at which none of the miners online are viable at the current difficulty, and even a price where some of the best pre-ordered mining gear is not viable, if it ever arrives (mining gear is notorious for not arriving.) And with a bit of work we could build a sheet to figure out the price at which ordering even theoretical gear is not viable.

There are some places where it might always be viable to mine. For example, in Quebec, electricity is not just cheap, but many people heat their homes with it. In the winter you could effectively mine for free, but what would you do in summer? In Iceland you could mine longer. But this is hardly what we want for a robust mining system.

So my fear is that mining power reduction -> panic -> price decline. And I am not sure why you could not enter unicode arrows.

(The software seems to be deliberately filtering out the unicode arrows. When I type them, they show up when I say "preview comment" so it knows how to show them, but when I publish them it filters them out. Security thing in drupal I guess.)

There are other places with intermittently marginally-free electricity, whenever output isn't perfectly matched (or rapidly adjustable) to current demands. Think solar panels at mid-day, utility generators with step outputs always exceeding grid demand, etc.

You suggest this isn't what's wanted for "a robust mining system", but are you sure?

These points of temporarily-arbitrarily-cheap electricity may be individually transient – but globally reliable on a smooth daily cycle. Bitcoin mining could be a complementary activity bolted onto the side of every generation process, to recapture value from otherwise unsellable surplus electricity, for exactly those periods when the surplus exists.

With adjusted difficulty, Bitcoin can adapt to whatever level of hashpower such 'recapture mining' supports. This might even result in an ideal decentralization scenario: mining that's most economical when paired with highly-geographically-distributed and uncoordinated edge/off-grid electrical generation.

The problem with mining from an intermittently usable source is that you have to buy custom mining hardware, and it's depreciating every day. If you run it when the electricity is cheap, great, but if that's only a fraction of time you may not be able to get back your hardware investment before the hardware is obsoleted by newer chips. Presuming there is a market for new hardware, of course, since if the price of btc is not high enough, it's hard to justify new mining rigs. In any case, very intermittent electricity like grid adjustments is not going to cut it. Solar panels mid-day still feed the grid.

But no, I don't think it's a robust mining system if miners all have to be in places that heat their buildings with cheap electricity or have electricity to spare. In theory one could imagine physically shipping hardware from Iceland in northern winter to somewhere south in their winter -- but it seems a bit crazy to me.

It's vital -- the whole purpose of mining -- to have a very diverse set of miners out there at any time, widely distributing the mining load. Mining exists so we know the people writing the blocks are selected fairly, at random. You can't have all the miners in a few geographic zones.

Right now some people are mining with subsidized electricity in countries that do that, or buildings that foolishly give a flat rate electricity to tenants. A nice hack, but not a sustainable basis for the system. Pro miners are moving to places with the cheapest electricity.

Of course, this all fixes if bitcoins sail up to high prices again. As long as electricity is only a minor part of the cost, people can mine anywhere, and they (or more likely mining chip companies) will make money. But the difficulty is up and the price of btc is down and the hashes/joule number moves in infrequent jumps and so we're close to the danger zone, where electricity is a large part of the cost of mining even with the newest rig.

Also worth considering is how fast the hashrate grew in the last couple of years. Way faster than Moore's Law. That's because the price got so high, but with a price decline it's a different story. Bitcoin's design many not be robust in a world of declining prices and declining difficulty.

Hardware becomes obsolete with regard to a certain electricity cost. At marginally-free electricity, lots of otherwise 'obsolete' rigs are practical again. So, power-generators buy used equipment cheap from people who had pay for electricity. That is, these intermittent-miners are not facing quite the same obsolescence-threats you're alleging.

And such 'peaker' mining would be more geographically distributed than anything else – it has to turn on and off with the spinning of a generator, or the local cloud cover, or gusts of wind – so it must occur right next-to various kinds of generation. Generation is itself distributed due to transmission losses, and plausibly becoming more distributed with new edge and off-grid technologies.

Search for [difficulty retargeting algorithms] for a sampling of discussion around related altcoin crises.

I haven't noticed Bitcoin miners stopping-on-a-dime based on the relation of current prices to their internal breakeven costs. Either out of bullishness, foolishness, or enlightened self-interest, their behavior seems a bit laggy – and that's helpful.

Only big discontinuous mining-power withdrawals, outracing the usual adjustment, are a problem, and if one of those became evident in block rates, I'd expect a lot of panic and argument over a few days, and maybe even some rushed changes from the reference-core team... but then survival at an adjusted price/difficulty. The stall would be viewed as a bug, and for a bug, "anything that works" could win a broad consensus. (OTOH, the Bitcoin creation-rate and final-total are central doctrine, and any changes there would cause a religious schism.)

If cycling hysteresis between "difficulty so low everyone wants to mine" and "difficulty so high no one wants to mine" were likely, perhaps because of tight cost uniformity among mining operations, we would have already found the doom loop, price volatility or not. (Price volatility might even be helpful for smoothing miner behavior, by generating a diversity of expectations rather than complete synchronization.) So far, the usual behavior has been incrementalism, up-and-down, as-designed.

Thus I can't see why one perfect-storm price-and-hashpower stall, followed by some gut-wrenching hackish jump-restart, would necessarily make people expect repeats as mining power reenters. People are more likely to think, "maybe the crisis happens once every 6+ years but now fixes are in place", than to think "we've entered a permanent new phase of even wilder price & hashpower swings".

Finally, and possibly under the category of "my extreme fringe opinions": I don't expect that even a total incentives/network collapse of the current blockchain would totally destroy the value of Bitcoin balances (txouts). It might even provide an evolutionary kick-in-the-pants that, over a longer term, makes the value of those balances larger.

Why?

As long as there's not a crypto-collapse, the whole historical blockchain will still be there. Multiple teams, with competing theories about Bitcoin's flaws, would be free to bootstrap their own next-generation systems as 'SpinCoins', reusing the Bitcoin endowments to immediately seed their system to all cryptocoin early-adopters. Every Bitcoin balance at the 'collapse' could thus yield usable balances in multiple competing successor 'BabyBitcoins', a bit like a series of corporate spin-outs. You only have to believe that one (or more) of these will eventually figure things out for ur-Bitcoin balances to retain positive, and perhaps significant, value.

See the bitcointalk thread on [spin-off altcoins] for more on this approach, and especially the diabolically parasitic idea of launching an 'æthereum' coin that's exactly Ethereum but with Bitcoin starting balances, if Ethereum proves to have legs.

A different difficulty algorithm probably is needed by bitcoin, though such a major change is not an easy thing to do, people would really need to see the crisis. For example, if you just got a new Neptune miner with more gh/s than everybody else, you prefer the difficulty stay high, precisely because it pushes everybody else out of the game. You need everybody, especially the fastest miners, to buy into any major revision to the algorithms.

I continue to feel, though, that while the economics governing mining are primarily based on the fiat exchange rate, and miners are professionals who will not sustain money-losing mining for long that there is an issue, and the issue is particularly strong when the fiat price approaches the point which makes the bulk of currently operating miners unprofitable.

Worse, I am talking about being unprofitable in the "not covering electricity" zone. Right now even the fanciest new miners become unprofitable when you include the cost of the mining rig if the price drops in half. I suppose if nobody buys new mining gear it does stabilize things a bit, and Cointerra's bankruptcy may be a sign of what's to come.

KnCMiner claims 13 gh/joule for their new Solar miners. Perhaps that will take us out of the danger zone. But right now I see the following fundamental disconnect:

  • Electricity (and miners) must be bought with fiat currency, so the economics of those depend on the dollar/BTC rate
  • Bitcoin's internal systems of difficulty adjustment, hashrate and reward are not tied to fiat prices.

The exchange rate to dollars is volatile and based on what speculators think the future value of bitcoins will be. The only connection is that when the exchange price goes high, people are willing to pay more for mining rigs. The price of electricity is the same, no matter what the other factors are. When the price of electricity becomes dominant in the cost of mining, the exchange rate becomes everything.

This disconnect is dangerous, and creates a danger zone in the price region where the cost of electricity heavily dominates. We are close to that danger zone.

I can't stop thinking about this now, so I'm coming back here, and I'm happy to read the new comments since I last looked.

So, let me sketch this out one more time to help me think:

1. (fact?) Proof-of-Work mining, because it offers no barrier to entry, drives the marginal revenue of a miner down. Miners exit when the marginal revenue ? their marginal costs (which we're mostly assuming is electricity, although cooling might be a big component of marginal cost, too). Therefore, eventually the surviving miners are those which have a low marginal cost (let's think in terms of marginal cost per hash instead of per time or per unit of equipment or whatever), but even they, the survivors, also have low marginal revenue, because the way they drove their more cost-sensitive competitors out of business was to increase the system-wide difficulty, harming their own marginal revenue as well as everyone else's.

In short: we can expect that in a mature Proof-of-Work system there are huge swathes — maybe even a large majority — of the mining power perched on the edge of marginal profitability.

2. (fact) The market price of the coin can swing up or down.

3. (fact) Brad's first observation: if the price drops below some threshold, some fraction of those marginal miners will shut down to avoid marginal costs. (This is a fact because almost all of these miners are not hobbyists/enthusiasts — they are pros with their eyes on their bottom line. For a nice video of what professional Bitcoin mining was like a year ago: https://www.youtube.com/watch?v=K8kua5B5K3I .)

4. (fact) If a substantial fraction of miners shut down, the average block confirmation time increases proportionally, until the next difficulty adjustment. E.g. a 90th percentile of 35 minutes instead of 23 minutes, if 2% of the mining capacity is exiting every day; I got these numbers from Dave Hudson's good blog post on the topic: http://hashingit.com/analysis/41-waiting-for-blocks .

5. (speculation) If the block time increases, proportionally fewer coinbases are given out to miners, and perhaps miners are currently selling all their fresh coinbases, which puts downward pressure on the price, so if they do that less then there will be a reduction in downward pressure on the price. Thanks to Dave Hudson's blog for mentioning this speculation, and also Gwern above alludes to a similar consideration when he points out that longer block times mean a slower rate of dilution of the monetary base.) I don't have an opinion on whether this would be a significant factor.

6. (fact) If the average block confirmation time increases, the time to know that a payment to you is confirmed increases proportionally. E.g. a 90th percentile for 6-confirmations of 143 minutes, instead of 93 minutes (ibid).

7. (fact) If the average block confirmation time increases, the transaction bandwidth limit could be reached, so that transactions that come with no or low fees take even longer, or stop being serviced entirely and average transaction fees go up. (According to Dave Hudson, Bitcoin typically currently has 30% of its txn capacity in use: http://hashingit.com/analysis/39-the-myth-of-the-megabyte-bitcoin-block .)

8. (speculation) Because of point 7, average transaction fees could rise enough to make mining more profitable again. I'm skeptical of this.

9. (speculation) Because of point 7, people could switch to alternatives away from Bitcoin, reducing demand for it and further depressing its price. I'm skeptical of this, too. :-)

10. (speculation) Brad's second observation: due to some combination of 3, 4, 6, and 9, if the mining power drops, this could cause a further drop in the price of the coin. I'm not sure what I think of the likelihood of this.

So, here's why this is potentially important: if 10 happens, then the combination of 3 and 10 would make a strong positive feedback loop that simultaneously crashes the price of the coin and the performance of the transaction processing system.

I think this is worth worrying about.

Does anyone (Brad, Gwern, Gordon) see a mistake in the above?

Most of this is correct. I have often heard it said that the reason for Bitcoin's 2014 price slide was that miners were all selling rather than holding and speculating. I don't know if this is true, or more to the point, I don't know if the selling of 3600 fresh btc per day has been enough to cause the price drop from $1200 to $200. I mean, that's a $14 billion dollar drop in the bitcoin market cap, so if it is true, it just says that bitcoin's support was very weak, so that extra coins on the market could devalue it this much. If so, we don't know where the bottom is, where the support for the price is strong enough to withstand 3600 btc per day extra.

Your logic suggests that the support might be at a level where it can handle 3,600 btc per day, but it could handle 2,000. That level exists somewhere, but I have nothing to tell me it's at $100 or any other price.

My concern is that, as we all know, the price of btc is based on speculator's view of the future true value of Bitcoin as a system. That coins are valuable because there is a consensus to value them, plus their inherent utility, which today is low but may become high in the future. Of course, as is often debated, fiat is also valuable because of the consensus to value it, plus its inherent utility (you can pay taxes to the US treasury with dollars, for example.)

Anyway, my concern is that with the value containing this large speculation component, that a serious event like the reduction in the block rate to 15 minutes could cause a serious drop in confidence. Bitcoin has a lot of support because it has shown itself to be quite robust. If it shows cracks in the system I think people sell holdings. Not all of them, but some.

Unfortunately, as you note, this issue creates a feedback loop, and we move out of stable equilibrium.

Right now transaction fees are a tiny fraction of reward. It's all in the coinbase. Tx fees would have to go way, way up to make the difference, but in order to break the feedback loop, they would be needed. They would convince people that there is a patch for the crack and the system can survive. However, the large holders of btc, who have the most to lose, are not a unified group, and will not sustain paying large tx fees.

Right now blockchain.info suggests we are paying $7.68 per transaction to the miners, almost all of it in coinbase rewards. If you had to switch a lot of it to fees, that's a very hefty transaction fee, making bitcoin inferior for many transactions until it goes down.

On the plus side, the fiat price is improving today, out of the danger zone. If it stays there until a new generation of mining equipment arrives with much higher megahash/joule numbers, the cliff is escaped for a while.

I was just discussing this with Tara Vancil, and she pointed out that inasmuch as Bitcoin has value as a medium-of-exchang (not just as a store-of-value) then this should exert upward pressure on the market price, even in the face of declining mining power. I think this is a really good point. The part about Brad's Doomsday Scenario positive feedback loop in which ?mining-power causes further ?price assumes (explicitly) that the price is determined solely or primarily by "speculators" (i.e., by people valuing Bitcoin as a store-of-value).

Now, Tara's observation doesn't totally assuage my concerns about Brad's Doomsday Scenario. It could still be a Doomsday, although perhaps not as complete or permanent of a Doom as it would be without value as a medium-of-exchange. Perhaps I dimly intuited this earlier, and that's why I included point 9 in http://ideas.4brad.com/would-bitcoin-fall-cliff-if-it-dropped-100#comment-15801 , which suggests a way that this Doomsday Scenario could cause Bitcoin to lose value as-a-medium-of-exchange, too.

Right now, my impression is that the vast majority of use of bitcoin as a medium of exchange has it being converted to fiat the moment it is spent. That's what all the various payment processing platforms offer as their main method of operation, and it's what most merchants want, because the merchants are not in this to speculate on the price of bitcoins, but just to get paid.

As far as that is concerned, it almost doesn't matter what the price of bitcoin is -- the processors even promise the merchants they will get the full value in fiat they actually demand. It matters to the buyer, but I have heard that buying with bitcoin does not offer all that great an exchange rate (due to that promise to the merchant) but has mainly been popular with people who mined/acquired a lot of bitcoins in the early days and want to spend them. Ie. my impression is there is not much of "I will buy btc with dollars, I will then pay with btc at a merchant who then turns them right into euros."

Where I could be wrong is in the underground markets, where people certainly don't want to use official payment processors or convert into or out of fiat anywhere it might get tracked. Those coins might go to a laundry right away. Underground sellers will tolerate the volatility more than mainstream merchants.

I don't know what fraction of bitcoin use is underground these days. It was more in the past, but there is still plenty of it.

The fiat price of bitcoin is not kept up by people doing buy-btc, spend-btc, merchant-redeems-btc. The only thing that will put upward pressure on bitcoin is people buying bitcoin to hold, or to spend with a merchant who holds it. Likewise the thing that puts pressure downward is people selling bitcoins that were not bought recently, ie. they were held or mined.

On the other hand, every long-holder of bitcoins (such as the lucky early miners) who buys a product online with bitcoins from a mainstream merchant who immediately sells them is in effect selling bitcoins, and pushing down the price.

Now, there are many people who have long term faith in bitcoin as a store of value. So when it drops, they buy, and they support the price. But support has eroded eventually all the way down from $1200 to $220.

Hm, that makes sense, but I also think that what Tara said makes sense. Let me see if I can reconcile this.

There are three functions of money: medium-of-exchange, store-of-value, and unit-of-account. I tentatively think that each of these uses can (but will not necessarily) support the price of a currency, in different ways.

First of all, consider store-of-value. It is easy to see that if in aggregate there are people who are willing to bet on BTC as a store of value, then that supports the price (by increasing demand for BTC). On the other hand, as Brad points out, if those people change their minds, then the demand — and hence price support — can disappear at the speed of the changing of a mind.

Empirically, as Brad says, there is a lot of aggregate "betting on Bitcoin" going on right now, so this is currently probably a real effect.

How about unit-of-account? If people use BTC as a unit of account, that is basically the exact same thing as saying that there is price stickiness in prices of goods (when expressed as a price-in-BTC). Price stickiness seems like it supports stability of the price of a currency, resisting changes in either direction. OTOH, it too can change at the speed of the changes of some minds, plus some communication and renegotiation. So, actually that might not be very fast at all, depending on what needs to be communicated and renegotiated.

Empirically, BTC is not used as a unit-of-account at all for anything right now, to my knowledge, so this is solely relevant to a possible future world.

Finally, what about medium-of-exchange? This is what Tara Vancil suggested to me, and I think there is something to it. On the face of it, like Brad argued, pure medium-of-exchange (without any unit-of-account) doesn't affect the price of BTC in terms of $, but like Brad pointed out, buyers who need to acquire BTC in order to buy push up the price, and sellers who want to sell their BTC push down the price. I think you can subtract the aggregate outflow from sellers from the aggregate inflow from buyers to yield the price pressure.

So, I think that means that if the Bitcoin economy in aggregate is currently expanding, even in the absence of any effects from store-of-value or unit-of-account, that this pushes up the price, and if it is currently contracting that will push down the price.

Empirically, Bitcoin is currently used primarily, as far as I know, for underground markets like illegal drugs and Internet crime. Here's a recent article suggesting that there isn't a lot of use of Bitcoin as a medium-of-exchange in above-ground markets: https://twitter.com/tsimonite/status/568138057444163584

All of this might be the same as what Brad already said above, but now that I've worked it out in my own words, I'll remember it better.

A take-away for me is that the three facets of money are curiously interlinked, and that being usable for both store-of-value and medium-of-exchange purposes would make a cryptocurrency much stronger than if it were used for just one facet.

Fuck, the software that runs this site converted all my "(b)" symbols (U+24d1) (http://www.fileformat.info/info/unicode/char/24d1/index.htm), by which I meant "Units of Bitcoin as a currency", in to "?" symbols, which makes it look like I am incoherent and/or up-talking. Argh.

My point is that "I buy bitcoins with $, pay for item from merchant with BTC, merchant immediately cashes out into $" should have minimal effect on the price of Bitcoin to dollars. Legal or illegal.

So how many people are buying from merchants with long-held bitcoins? Since almost all merchants immediately cash out, the more of this, the more the price of a coin goes down -- as it has.

I don't think there is very much "I buy btc with dollars, pay a merchant, and the merchant keeps the btc" going on at all. That would push the price up. Perhaps I am wrong and the people selling weed are keeping the btc, perhaps to pay their own suppliers, but in the end I think somebody cashes out. They will certainly send their btc to a laundry, but why would they hold them after a year of steady declines, unless they also wish to speculate?

To add to this we have miners who now are not very profitable, and must cash out all their coinbases to stay afloat, also pushing the price down.

The only thing pushing it up is speculators -- and I would say this has been true since it first started getting real cash value a few years ago.

> My point is that “I buy bitcoins with $, pay for item from merchant with BTC, merchant immediately cashes out into $” should have minimal effect on the price of Bitcoin to dollars. Legal or illegal.

Hm, it seems to me that if the rate of such transactions today is higher than it was yesterday, then this exerts an upward pressure on the price of BTC in terms of USD. I'm not entirely sure why I think that — maybe because of uncertainty and delay between the time the customer bought the BTC and the time the merchant sold the BTC? If it is all down to friction and delay then some hypothetical tools in the future which eliminate that friction and delay would eliminate my hypothesized upward pressure due to commerce.

I think the pressure of use as a medium of exchange is minor. Also, I am not sure how much of that is going on in the legal market. I mean, every web retailer who takes bitcoin that I have seen also takes dollars, and there is very little reason for me to do a $ -> BTC transaction just to buy from them, other than to show my love of Bitcoin. If I am somebody with a large stash of coins, I have a motive to use them, but that's a downward pressure. For buying drugs, it is a different story. The vendor does not take Visa or Mastercard so people will do it, but it's still mostly a flow-through.

I will also note (though it has probably diminished) that people donate bitcoins. Almost exclusively the donors are miners of old, I think. Most charities convert to cash. I know we do. (And yes, we foolishly gave away our 8,000 coin btc haul to the faucet in the early days.)

And every day, 3600 new coins are mined and, if I read reports right, are probably sold immediately. Some people attribute the 2014 slow drop to this alone.

Coincidentally, TwoBitIdiot just posted the first in a series of posts on what seems to be roughly this same question: does Bitcoin usage as a medium-of-exchange impose a floor on the price of Bitcoin?

https://twitter.com/twobitidiot/status/568280561518817280

The real demise of bitcoin is that it isn't a good currency. To be a good currency stability is required. One of BTCs claims to success was that it would not be subject to inflation. The mere suggestion of changing the amount of coins rewarded in mining wreaks of inflationary strategies. The other promises of BTC which I question are,

1: That it is anonymous. BTC transactions are more traceable than dollar transactions. It's all in the block chain.

2: That it will be the default currency for global peer to peer transactions. The crushing of this myth by Facebook's entry into peer to peer payments through cell phones is the end. I am watching closely this weekend's BTC trading expecting to see a final pump ending late Sunday night with a devastating dump. The other big danger concerning BTC peer to peer over the internet is of coarse security. Apple, for instance, doesn't want BTCs in their phones.

If you think the BTC society are a bunch of do gooders for a cause, well, I doubt that! They are in it to make a fortune! It is they who will be the first to bail. Then they will quietly disperse.

Add new comment