Bitcoin Forum
December 09, 2016, 06:04:25 AM *
News: To be able to use the next phase of the beta forum software, please ensure that your email address is correct/functional.
 
   Home   Help Search Donate Login Register  
Pages: « 1 [2]  All
  Print  
Author Topic: Capital set in motion by BFL ASIC pre-order  (Read 2616 times)
Spekulatius
Legendary
*
Offline Offline

Activity: 1022



View Profile
June 26, 2012, 11:28:47 PM
 #21

Sry, how does mining influence bitcoin prices again? Please be as precise as you can in counting and determining causes and effects!
Thank you

Many economic types tend to focus on a one-size-fits-all theory for how an item gains value. Some, including myself, see it as going through phases in which several schools of thought hold credibility (labor theory of value, subjective theory of value, intrinsic theory of value, etc). I'll discuss (kind of haphazardly) functional items and especially money, leaving things like music and toys for another time. If you already have an understanding of this, consider the first part to be confirming common ground to work with. All values will be in USD.

Origins of value

For small markets, or those with very tight supply/demand dynamics, the marginal cost of production can have a very strong effect. Introduction of an item is almost exclusively dependent upon production cost, or energy input in the case of Bitcoin mining. If a new tool could make my job 100 times more effective, I would certainly consider using it. However, it if will take a decade to construct, I would have to possess sufficient vision to see how it would change things for the better, as well as the drive/passion to see it through to completion.

As market size grows, demand develops based mainly on utility value. If it's useful, it stands a good chance of being adopted, especially if it improves on what currently exists. When its existence becomes known more widely, adoption may accelerate or meet resistance, depending heavily on conceptual familiarity and functionality. If demand accelerates relative to supply, obviously the production cost will be negligible. If the demand remains constant, production cost equalizes at a level that provides enough supply to meet demand in addition to returning enough profit to keep the production operating. In the final case of demand declining, production cost sets the floor in value because it dictates supply.

When the market for an item matures, supply and demand fluctuate within a familiar and stable range. At that point, efficiency improvements related to production cost are rare and demand might have little space to expand, but plenty of room to contract.

Relating to Bitcoin mining

If demand for Bitcoin remains consistent or rises, mining efficiency won't matter much thanks to difficulty adjustments. However, rising demand is dependent upon multiple factors and cannot be predicted with absolute certainty. On the downside in 2011, GPU mining costs were a primary factor in putting a floor under the exchange rate price. The mining operations were not as willing to sell so close to production cost which was largely in the $1-3 range at the time (now mostly in the $3-5 range).

With FPGA mining, production cost could reasonably be projected to have decreased to about $0.10 or an order of magnitude better than GPUs. This means that a mining operation has more competitive pricing leeway to work with, and will retain that flexibility until the majority of the market falls in line with comparable efficiency. Until then, FPGA miners would enjoy a greater profit margin. Input costs still need to be covered as well; because of that, any substantial rise in demand would be met with an equivalent rise in selling so long as sufficient supply can be produced.

The same will occur with ASIC mining, although without any really major advances possible in efficiency, it will be harder to maintain large margins. If it costs $0.01 or less to produce a Bitcoin, there has to be enough demand to justify $6+ as the rate. If bids increment in $0.01 intervals, that means 600+ buyers must be present for every Bitcoin. That's obviously a rather absurd scenario, and an oversimplification, but you can see how the supply leaves ample room for lower prices.

One major factor that would keep the production cost elevated is an expansion of the mining market itself, as reflected in the difficulty level. If 10x the participants started mining with ASICs, driving the difficulty up by the same factor, ASIC mining would cost $0.10+ per Bitcoin. FPGA would be at $1.00+, and GPU at $10.00+. That's where the ratio of different classes matters, with GPUs potentially falling off due to lack of competitive efficiency. FPGAs might put a floor under the exchange price until they're paid off or no longer profitable. Then ASICs would remain as the production cost floor.

Pricing dynamic

Value floors can be either hard, as in production cost, or soft in relation to demand. As you can see from above, the production cost floor would be where miners start shutting down operations. The soft demand floor is where the perceived utility value picks up, and since that can be highly subjective, it can be anywhere above the hard floor. Since several dynamics are in play, with at least three strata on the production side, demand constitutes whatever buffer exists between the production cost and market rates.

I like to use gold for comparison - today, it costs about $1,200 to bring one ounce to market. The market currently values gold at about $1,600 per ounce. That $400 amount is exclusively demand side. If demand evaporates and the price declines below $1,200 there will be some producing operations that slow down or halt entirely. For this example, the price is also like difficulty in Bitcoin. Since gold mining is a mature industry at the equivalent point of Bitcoin ASIC mining, factors that affect the production cost are limited.

On the upside, competition for a share of the Bitcoin economy is likely to bring ASIC mining costs up rather quickly - maybe as fast as units become available, until the network is saturated. Depending on how long that transition takes, with GPU and FPGA mining all but gone, demand may not vary much and won't have an opportunity to test even the GPU hard floor. In the table below, each will bump up to the next level of estimated production cost.

ClassProd. Cost
CPU100.00
GPU3.00
FPGA0.20
ASIC0.02

If we assume 75% of the network is mining using GPUs and 25% are using FPGAs, then the average production cost would be $2.30 - the remaining $4.10 for the ~$6.40 rate today is from demand. Block halving with constant demand would be an effective doubling of demand, which should be reflected in the prices. Mining efficiency could absorb some of that, so the price rise might not double to $12 instantly.

The major factors are, not in any particular order and including ones that weren't explored in detail:
  • Demand
  • Block Reward
  • Mining Efficiency
  • Difficulty (related to #2)
  • Ratio of Miner Classes (CPU/GPU/FPGA/ASIC)
  • Size of the Bitcoin economy (related to #1)
  • External Influence (global markets & banking crises)
8. Hash rate of total Bitcoin network (relating directly to #4)?
also to
1.: I would rather call it "Ratio of Supply and Demand"


Wow, thanks for such a compelling answer! I just got around today to read it.
I see now, that mining does in fact has three effects on prices:

- Setting a price floor determined by the minimum of production cost of new bitcoins, for the mean of all active miners. Therefore this
  floor acts as a strong support level for exchange rates (to speak in terms of chart analysis)
- Pushing exchange rates up stronger, the closer they drop to said floor by miners hoarding coins
- Drawing down on exchange rates the further exchange rates rise from said floors by miners selling hoarded coins on exchanges

I may add a little remark on the side, that does not impede the functioning of the mechanisms described by you, but tries to increase precision in calculating resulting effects. I feel like in your descriptions you have always spared the cost of aquisition, as well as proceeds of sale after use. Those may be -especially with ASICs- substantial for the resulting profitability of mining and should be included in any further calculations IMHO.

As my question only concerned the effects mining can have on exchange rates, you rightfully answered only this aspect and left all other relevant factors that influence exchange prices aside. Thank you for being so specific.
With those factors I think of:
A: Major news on the media
B: Mass psychological effects
, commonly known of panic buying or panic selling or rallys, forming of chart patterns, emotion cycles,..
C: Change of fundamentals
D: Value change of pair currency
(BTC/USD, EUR/BTC,..)
E: Market Manipulation (e.g. Paramount orders in order books commonly known as "Walls", public disinformation)
F: Other external Effects, like Exchange outages, hacks of beacon entities, congestions of the money flow in and out of exchanges

Only for the sake of completeness, I'd like to point out, that mining can only be seen as one of many such factors excerting their influence at different periods of time to a varying degree on the exchange price of bitcoins for other currencies. We may call mining however one of the most stable and predictable influences, as it is reliably measured in many different indicators and metrics, like difficulty, hashrate/time of the entire network, reward,etc .
I would measure the influence mining has on exchange rates by the total number of bitcoins mined every day (~7200 BTC daily at a 50 BTC/block reward). This is of course simplyfied and based on the assumption that in over a longer period of time those times of more frequent hoarding of mined bitcoins and times of frequent immediate spending of mined bitcoins even themselves out. Also that the characters of miners are evenly distributed, with those on one side that rather speculate on higher prices and thus hoard coins in times of low profitability and on the other side those that are more inclined to exchange every bitcoin they mine immediately for other currencies.

Please correct me wherever you feel differently.

Salute,
Spekulatius
1481263465
Hero Member
*
Offline Offline

Posts: 1481263465

View Profile Personal Message (Offline)

Ignore
1481263465
Reply with quote  #2

1481263465
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1481263465
Hero Member
*
Offline Offline

Posts: 1481263465

View Profile Personal Message (Offline)

Ignore
1481263465
Reply with quote  #2

1481263465
Report to moderator
1481263465
Hero Member
*
Offline Offline

Posts: 1481263465

View Profile Personal Message (Offline)

Ignore
1481263465
Reply with quote  #2

1481263465
Report to moderator
miscreanity
Legendary
*
Offline Offline

Activity: 1078


View Profile
June 27, 2012, 04:46:55 AM
 #22

Wow, thanks for such a compelling answer! I just got around today to read it.
I see now, that mining does in fact has three effects on prices:

- Setting a price floor determined by the minimum of production cost of new bitcoins, for the mean of all active miners. Therefore this
  floor acts as a strong support level for exchange rates (to speak in terms of chart analysis)
- Pushing exchange rates up stronger, the closer they drop to said floor by miners hoarding coins
- Drawing down on exchange rates the further exchange rates rise from said floors by miners selling hoarded coins on exchanges

Glad it helped Smiley

Those three aspects of the mining dynamic are about as core as it gets. It may not help with figuring out the demand side, but the supply and resulting price floor can be estimated fairly well.

I may add a little remark on the side, that does not impede the functioning of the mechanisms described by you, but tries to increase precision in calculating resulting effects. I feel like in your descriptions you have always spared the cost of aquisition, as well as proceeds of sale after use. Those may be -especially with ASICs- substantial for the resulting profitability of mining and should be included in any further calculations IMHO.

Yes, I kind of glossed over that. I tend to consider initial outlay to be part of the cost of producing each Bitcoin, so an amortized value included in that cost. Most businesses do this as a normal practice, but I think because Bitcoin mining is so exclusively dependent upon and recoups hardware investment so wildly fast that there's a good deal more emphasis on that than energy input.

From a purely economic view, it's pretty clear in blockchain.info charts how the mining operations that were either unprofitable would've closed up shop in the latter half of 2011. That's not taking psychological factors into consideration, which I'm sure played a hefty part.

I also didn't go into 'free' energy input, since that isn't realistically viable long-term and the larger operations will be forced to source power at some point. Renewable, independent power production is another amortizable cost.

I would measure the influence mining has on exchange rates by the total number of bitcoins mined every day (~7200 BTC daily at a 50 BTC/block reward). This is of course simplyfied and based on the assumption that in over a longer period of time those times of more frequent hoarding of mined bitcoins and times of frequent immediate spending of mined bitcoins even themselves out. Also that the characters of miners are evenly distributed, with those on one side that rather speculate on higher prices and thus hoard coins in times of low profitability and on the other side those that are more inclined to exchange every bitcoin they mine immediately for other currencies.

Please correct me wherever you feel differently.

Salute,
Spekulatius

Spot on. Once the interrelated components are separated out, it becomes easier to see how they affect the whole. Running a mining operation helps Smiley
Pages: « 1 [2]  All
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!