By Ria Bhutoria and Wilson Withiam
Know & go
- An inside look – Difficult adjustment: Bitcoin difficulty 101: Bitcoin mining difficulty continues to rise to new highs, so we provide an inside look at what difficulty is in terms of Bitcoin, its relation to other network components and its impact on mining operations.
- Three things to know: (1) Walmart won a U.S. patent for a stablecoin that may be backed by U.S. dollars (2) BlockFi secured $18.3 million in a Series A led by fintech investor, Valar Ventures (3) A bill aiming to change the IRS’s treatment of cryptocurrency transactions was reintroduced to the House of Representatives.
- Market snapshot: Total crypto market capitalization is around $285.4 billion (+5.2% w/w) at press time. BTC is trading at $11,800 (up 12.6% w/w), ETH is at $208 (down 4.7% w/w). BVOL (the rolling 30-day annualized Bitcoin volatility as calculated by BitMEX) is 96.5%, down from 97.3% last week. (8/9 2:30PM ET)
An inside look: Difficult adjustment: Bitcoin difficulty 101
According to BTC.com, the average block interval in the last 2016 block period ended August 5th was ~9 minutes, lower than the 10 minute target block interval. As a result, the target difficulty was adjusted upwards to 9.9 trillion from 9 trillion (a 10.8% increase) to make it harder to find a valid hash, thus, increase the time needed to mine a block. Simultaneously, the hashrate of the Bitcoin network reached an all-time high of 79.9 EH/s as of August 6th (1 EH is 1 quintillion hashes per second). As we explain in this piece, hashrate and difficulty are interconnected – as the hashrate, or processing power, of the Bitcoin network rises, the difficulty is automatically adjusted to maintain the 10 minute block interval, on average. Hashrate and, as a result, difficulty are also impacted by the price of BTC and the cost of electricity denominated in BTC, as these factors determine the profitability of miners.
What is Bitcoin difficulty?
Bitcoin is designed to generate a new block every 10 minutes on average. Among other reasons, this gives network nodes sufficient time to communicate the last valid block and reduce work wasted on valid blocks that end up being discarded — only one block can be accepted. Bitcoin’s target block interval also keeps the issuance rate of bitcoin relatively constant (at least between halving cycles), since miners are rewarded with newly minted bitcoin for discovering the next valid block.
Our cryptographer, Mira Belenkiy, notes many factors affect the actual block time. Sometimes, a new block appears in less than 10 minutes because a miner is lucky and guesses a good nonce. The appearance of more advanced mining equipment (i.e. ASICs) that can calculate new hashes at a superior rate, as well as a spike in new machines coming online, may also cause a sudden decrease in block generation time. Alternatively, block time can exceed 10 minutes, for instance, if a significant percentage of ASICs miners, or miners in general, drop off of the network.
Theoretically, the speed at which new Bitcoin blocks are generated is proportional to the total computational power of the network, otherwise known as the hashrate. As more miners and advanced processing equipment join the network, the increase in hashrate may reduce the time required to create a new block. An unchecked decrease in block time could disrupt the process of transaction verification and lead to a comparable increase in the frequency of bitcoin subsidy payouts. Therefore, the Bitcoin system incorporates a dynamic difficulty adjustment to ensure relatively consistent block intervals and new coin issuance.
Bitcoin difficulty determines how challenging it is, in terms of processing power expended, to compete for the next block and the associated bitcoin reward. Its value is adjusted every 2016 blocks, which at 10 minutes a block, occurs approximately every two weeks. The direction and magnitude of the adjustment is based on the average block generation time of the previous 2016 block cycle. If the average block time from the last 2016 blocks is less than the desired 10 minutes, the difficulty level is automatically increased. As a result, it requires miners to spend more time and processing power relative to the previous cycle to discover the next valid block. On the other hand, difficulty is reduced if the average block time over the same interval exceeds 10 minutes, likely lessening the time and effort needed to produce a new block of transactions.
This concept of difficulty was derived from the Hashcash proof-of-work system first proposed by Adam Back in 1997. Hashcash was initially used by email systems to deter spam and denial-of-service attacks as it required senders to spend a select amount of CPU time producing a verifiable hash to prove the email came from a legitimate source. The difficulty of these hashing calculations was programmed to increase over time to match the advances made in computational capacity according to Moore’s law. Bitcoin employs a similar proof-of-work system as part of its mining algorithm — though with some notable changes, including the choice of SHA256 instead of SHA1 as the hashing algorithm and a more dynamic difficulty adjustment mechanism to better accommodate fluctuations in hashpower and maintain consistent block times.
How difficulty operates
Difficulty is determined by multiplying the previous difficulty by the ratio of the actual time it took to generate the last 2016 blocks to the expected time to mine 2016 blocks (at 10 minutes a block, this would equal 20,160 minutes, or exactly two weeks). The amount of this product is then used to calculate the “target value,” the number that determines whether a block header, the hexadecimal value that represents each mined block, is considered valid or not. Miners will continue to generate prospective block headers (i.e. hashing block data components with the previous block header) until one finds a header that is numerically lower than the target value. [Note: This is where you may have heard block headers must contain a certain number of leading zeros to be considered valid. While this is a great high level explanation, it is more accurate to compare the actual decimal values.]
Difficulty adjustments thus impact both the target value and the range of acceptable block headers. The difficulty increases if actual time < expected time of the previous 2016 block period (i.e. if actual time < 20,160 minutes). If so, the target value and range of valid numbers decreases. Thus, the time and effort dedicated to generating new hashes should rise accordingly. In comparison, a reduction in difficulty due to actual time > expected time raises the target and widens the target range – the two are inversely related. The intended consequence is to make it easier and quicker for miners to find a valid number and hash.
The equation to determine the new target can be summarized as follows:
New target = Old target * (Actual time of last 2016 blocks / 20,160 minutes)
Andreas Antonopoulos points out that the difficulty adjustment is capped at a factor of four in a given 2016 block cycle to avoid large changes in difficulty. If the necessary adjustment is greater than four, it takes place in the next cycle, which means that significant differences between hashrate and difficulty can take multiple cycles to balance out.
Factors impacting difficulty (and vice versa)
Price impact on difficulty
According to Christopher Bendiksen, “Bitcoin is structured such that the hashrate follows price.” This trailing effect occurs due to the delay between the time of investment in new mining equipment and “when the gear is actually switched on.” Hashrate will therefore trail Bitcoin price increases, in some cases on the order of months. For reference, the previous peak in the 7-day rolling average hashrate was recorded in early October 2018, 10 months after price surged to its near $20k all time high. Alternatively, hashrate responds more quickly to substantial decreases in price since shutting down mining operations or redirecting hashpower to mine on a different crypto network that uses the same algorithm is more simple — the notable exception is mining operations that purchase fixed supply electricity contracts.
Difficulty is by design a lagging indicator of hashrate. It takes approximately two weeks (or more if the adjustment exceeds a factor of four) to adjust to fluctuations in hashpower. Therefore, difficulty trails bitcoin price movements to an even greater extent. But despite not appearing to have a significant (immediate) correlation to price, difficulty adjustments can impact miner behavior and, as a result, might provide some insight into bitcoin market cycles.
Difficulty impact on miner behavior
Bitcoin mining operations are primarily profit driven, and their breakeven cost is dependent on bitcoin price and operating expenses. While difficulty may not directly impact price, difficulty adjustments do affect the time and computational power required to win block rewards. Therefore, a stagnant or declining bitcoin price combined with a significant increase in difficulty may cut profit margins to the point where some miners (especially those that do not prepay for electricity) sell all of their bitcoin holdings, depressing prices further, or decide to drop off of the network.
The same remains true in the opposite direction. As bitcoin prices decline and profit margins compress, the least efficient miners are forced to unplug or shift crypto networks, dragging down the hashrate. Bitcoin then intuitively resets the difficulty to lower levels, until “the all-in cost of mining falls to a level where it is again right below the price of bitcoin” (that said, the general trend regarding difficulty has been upwards, with the Bitcoin network only experiencing difficulty declines a handful of times during its 10-year lifetime). Miners are incentivized to rejoin the network under the new, more profitable conditions. Difficulty is therefore one of the features of Bitcoin that prevents some theoretical mass capitulation events, such as the mining death spiral, since a return to profitability encourages participation.
Difficulty as an evaluation metric
The effect of difficulty on miner behavior has led some explore how difficulty adjustments can impact bitcoin price and valuation. Willy Woo introduced a new metric, dubbed the Bitcoin Difficulty Ribbon, earlier this month. The difficulty ribbon consists of multiple simple moving averages on mining difficulty (specifically 9D, 14D, 25D, 40D, 60D, 90D, 128D and 200D), creating a visualization of the rate of change in difficulty and how it relates to bitcoin price. Willy notes that when the moving averages compress or invert (shorter term moving averages drop below longer term moving averages), it has historically suggested bitcoin may be undervalued. A possible explanation is the least efficient miners have capitulated at this point, relieving the downward pressure on price and, as Willy states, “leaving more room for more bullish price action.”
The Difficulty Ribbon is still in development and will require more data to be considered reliable. However, it highlights the indirect impact adjustments in mining difficult may have on bitcoin price and offers a new and unique method for valuing bitcoin.
Difficulty is one factor that sets bitcoin apart from other assets. For commodities like gold, issuance is modulated by price. Increases in gold’s price will encourage gold miners to ramp up production, extracting more and more of the asset until operating costs equal the market price. This dynamic occurs similarly in the opposite direction. But “there is no such effect in Bitcoin.” Issuance is algorithmically determined — difficulty adjustments ensure relatively consistent block intervals, and therefore the issuance of new coins is not subject to changes in price. As a result, automated difficulty adjustments, among other components, contribute to Bitcoin’s sound money mechanics.
Additionally, while it is a lagging indicator, difficulty is an important metric to track as it provides insight around how effectively the network is adjusting to changes in price and hashrate.
Note: In this piece, we covered the Bitcoin network’s global mining difficulty. Mining pools may have a pool specific “share difficulty” that miners must consider when dedicating resources to a network.
Thank you to Mira Belenkiy for her contributions and corrections.
Weekly market snapshot
In other news
- The U.S. Patent and Trademark Office published a patent filed by Walmart to launch a stablecoin that may be backed by U.S. dollars. According to the filing, Walmart could structure the stablecoin such that it may only be used at select partners and retailers. It could also provide unbanked, low income households a place to “handle wealth at an institution that can supply the majority of their day-to-day financial and product needs” and potentially even earn interest. Source.
- Litecoin underwent its second halving on Monday, August 5th, reducing the reward from 25 LTC to 12.5 LTC per block. Leading up to the halving event, Litecoin’s price has appreciated from $30 in January to $120 in June. It is now trading at $85, down 12% w/w. The idea is that market participants bid up the price of a proof-of-work coin ahead of a halving event due to an eventual reduction in selling pressure from miners offloading their rewards on the market. Source.
- This week, Binance revealed that a hacker had contacted the exchange, claiming to have thousands of photos and pieces of information that resemble photos and data collected during Binance’s KYC process. The hacker was requesting 300 BTC in return for withholding the information, though Binance said in a statement that it did not give in to the extortion. Recently, the hacker began circulating certain pieces of data online, to media outlets and within a Telegram group. While the information bears similarity to Binance’s KYC data, the exchange said that the images do not have its digital watermark, so it is still unclear where and how the hacker obtained the data. According to this Coindesk piece, the hacker says the information was leaked with the help of an insider. Binance is offering a 25 BTC reward for information on the hacker, claiming to be “white hat”. Source.
- Blockstream revealed information around its mining facilities in Quebec, Canada and Adel, Georgia, catering to enterprise level mining operations and its own operations, with plans to eventually open up to smaller miners. Current customers include the Fidelity Center for Applied Technology and LinkedIn founder Reid Hoffman. At full capacity with the latest hardware, the facilities can contribute up to 6 EH/s of processing power to the Bitcoin network. Another notable feature of Blockstream’s facilities is that they will use the BetterHash protocol to improve mining decentralization. Source.
- BlockFi secured $18.3 million in a Series A led by fintech investor, Valar Ventures, including participation from previous investors such as Morgan Creek, Fidelity and Galaxy Digital. BlockFi will use the funding to grow headcount from 40 to 60 by the end of the year. Eventually, the company wants to expand into complementary products such as credit cards offering bitcoin rewards and new regions. Source.
- E-sports platform, Matcherino, raised $1.5 million in a Series A-1 including Galaxy EOS VC Fund and Wells Fargo. The company is exploring the development of tools using EOSIO software. Matcherino provides infrastructure for e-sports events and companies. Source.
- FTX has raised an $8 million round by Proof of Capital, Consensus Lab, FBG and Galois Capital for its derivatives and trading platform. It was incubated by Alameda Research which trades up to $1 billion per day and manages $100 million in assets. FTX is an exchange that offers futures trading, leveraged tokens, and OTC trading. Source.
- Mobile gaming firm Animoca paid approximately $8 million to acquire Quidd, a startup creating a marketplace for blockchain-based digital collectibles. Sequoia-backed Quidd previously raised $13 million in its Series A and used the funding to purchase the rights to ~325 brands from Disney, HBO, and Marvel, among others. Quidd found success in attracting users and generating revenue, but an exceedingly high burn rate led to the buyout. Animoca aims to help Quidd expand beyond the North American markets, especially into Asia. Blockchain startup Harmony was also strategic partner involved in the acquisition. Source.
Global regulatory roundup
- The Seoul Metropolitan Government is aiming to complete at least three blockchain-based initiatives by the end of the year. The three priorities include: (1) a blockchain points system in which residents will be rewarded with S-coins for using public services; (2) a blockchain service enabling citizens to submit qualifications without needing paper documents; and (3) a blockchain-based system that will allow part-time and temporary employees to enter into contractual agreements with employers. Source.
- Kik filed an answer to the U.S. SEC’s complaint against the company. The government agency announced a lawsuit against Kik in June, alleging Kik sold unregistered securities as part of its $100 million Kin ICO in 2017. In its recent response, Kik claims some of the original statements made by the SEC were taken out of context and appears to be willing to continue the litigation process. For a full breakdown, we recommend Katherine Wu’s annotated version of Kik’s response to the SEC.
- The Virtual Value Tax Fix Act, a bill aiming to change the IRS’s treatment of cryptocurrency transactions, reached the U.S. House of Representatives for a vote. Rep. Ted Budd, the man who reintroduced the bill, said “tax concerns and transaction record-keeping act as a deterrent to adoption” with regard to crypto. The tax bill would amend 1986’s Internal Revenue Code to eliminate the double taxation and record-keeping on cryptocurrency transactions. Source.
What we’re reading
- Becoming Decentralized Enough: The Case For DAOs by Mohamed Fouda (Token Daily)
- Cryptocurrency in China: Over the Counter, Under the Table by Dovey Wan
- Contextualized Analysis of Bitcoin Drawdowns by Luka Jankovic & William Nuelle (Galaxy Digital Research)
- Proof of Life: Why Bitcoin is a Living Organism by Gigi
- How Ray Dalio-And A Politicized Fed-Could Catalyze A New Generation Of Bitcoin Investors by Matt Hougan (Bitwise)
- The future of decentralized finance by Linda Xie
What we’re listening to
- What Bitcoin Did: Down the Bitcoin Rabbit Hole with Matt Odell & Marty Bent
- Unchained: All Things Crypto Regulation With Jake Chervinsky
- Unconfirmed: The IRS Is Cracking Down on Crypto Taxes: What You Need to Know
- Chain Reaction: Ryan Zurrer: Response To Tom’s Polkadot Report, Web3 and Lessons Learned In China
- Base Layer: Riccardo Spagni (Monero)
- Epicenter: Enabling Scaling Through Trust in Public Notaries
- Blockchain Insider: Who’s Been Using Bitcoin?
Circle in the news
- Jeremy Allaire, Circle CEO, appeared on Squawk Box to discuss bitcoin’s recent price action.
Where we’ll be in August
- TPI Aspen Forum, Aspen, CO, 8/18-20
- Blockchain Summit Singapore, 8/20