New Bitcoin Mining Hedge Indicator: CMBI Bitcoin Hashrate Index and “Observed Work”

Editor's Note: This article comes from Crypto Valley Live (ID: cryptovalley), author: Ben Celermajer & Coin Metrics Team.

Key points:

• To date, Bitcoin miners, as key players in the Bitcoin ecosystem, have not had any hedging tools and are completely dependent on the price of Bitcoin. However, as the mining market continues to mature and traditional market participants continue to join, these companies will seek corresponding mechanisms to hedge their risk exposure and operations, just like other traditional assets.

• The current way of estimating hashrate involves a calculation process that includes a set lookback period (such as 48 hours), which makes it very difficult to design financial products that can help miners hedge their risks.

• Coin Metrics designed the CMBI Bitcoin Index and "Observed Work" as a more responsive, sensitive, and manipulation-resistant way to measure the authenticity of mining activity compared to traditional hashrate estimates.

• Observed Work and Coin Metrics’ CMBI Bitcoin Index can serve as the basis for financial products, providing the market with the tools needed to effectively and efficiently trade or hedge Bitcoin’s hash rate.

introduce

Bitcoin is about to halve, and there has been much speculation about the impact it will have on hashrate. For most of the cryptocurrency community, this is an interesting speculative exercise with relatively low stakes. However, for the mining community, the outcome will determine not only profitability, but also the probability of survival.

This is largely due to the lack of a robust, market-accepted method for hedging the uncertainty of Bitcoin mining operations. In this week’s SOTN feature, we propose two new tools that will enable the financial derivatives market to effectively provide a mechanism for hedging and speculating on Bitcoin hashrate.

1. CMBI Bitcoin Hashrate Index

2. Observed Work

But first, let’s quickly review the importance of Bitcoin’s hashrate to the cryptoasset ecosystem.

The importance of Bitcoin hashrate

Mining is one of the core functions of Bitcoin and one of Satoshi Nakamoto’s important innovative ideas. Simply put, without mining, neither Bitcoin nor cryptocurrencies in general would exist today. Mining helps:

• Ensure the security of the network, prevent corruption, and disincentivize bad actors from tampering with the public ledger.

• Minting new bitcoins into circulation.

• Create and broadcast all transactions that occur on the chain.

• Verify and add new transaction information to the chain, allowing users to conduct transactions in a trustless manner.

Miners can generally understand their costs, which are a function of the hardware/facilities and electricity used to run their mining equipment, and given a fixed hashrate on the network, miners can determine their costs to be profitable under specific Bitcoin price conditions. However, the total number of hashes (computing power) performed on the Bitcoin network is not constant or predictable. Instead, hashrate fluctuates significantly over time and is not always in line with the price of Bitcoin.

Therefore, the ability for miners to hedge their hash rate volatility risk and maintain profitability across a wider range of price and hash rate scenarios is critical.

For example, consider a large institutional mining operation that is deciding whether to enter the market. They have the budget to purchase equipment, and their equipment has enough mining power to earn 1% of the mining hashrate. With this data, they can expect to earn an average of 18 BTC per day. At $8,000 per Bitcoin, if operating costs are more than $144,000/day (at today's reward levels), the miner will not be profitable and should not enter the market. However, if their operating costs are $100,000 per day, then their profit margins are very good and they should consider entering the market.

Three months later, when the equipment arrives at the factory, Bitcoin has risen to $10,000 per coin, but Bitcoin's computing power has doubled. Now, they will only have 0.5% of the computing power, representing an average reward of 9 BTC per day. At $10,000 per BTC, they will have a net income of $90,000 per day, assuming a daily operating cost of $100,000, a net loss of $10,000 per day. This does not bode well for the long-term development of their business.

However, if they are able to hedge their exposure to mining operations by trading hash rate derivatives, they can minimize the risk of macro changes in hash rate.

(Note: This example utilizes illustrative numbers and ignores the impact of the upcoming halving for simplicity.)

Tools needed to design computing power financial products

In a distributed process like mining, it is nearly impossible to get reliable hashrate figures from miners. Therefore, the current best practice for deriving hashrate is to generate an implied value from the rate at which blocks are produced at a given difficulty. This approach might be like how oil futures traders use the observable price at the gas station to calculate the amount of oil being extracted worldwide. In essence, this is deriving the current price (or hashrate) from historical data.

For hashrate, this introduces a lot of undesirable problems when creating financial products. Coin Metrics found that hashrate indexes have three key problems that must be overcome:

1. Predictability at the short-term level.

Since the calculation of hashrate relies on past data, the data used to generate short-term future levels is mostly known. For example, if you use a 48-hour lookback window to calculate hashrate, then out of the 48 hours of data points, you already have 47 hours of data points that will be used to calculate hashrate within an hour. Therefore, this short-term future hashrate is relatively predictable.

2. Given the random block generation process, the implied hashrate tends to follow an oscillating pattern (as shown in the figure below).

This introduces two types of settlement risk to the contract. First, whether the contract settles at the top or bottom of the swing is random, which can have a significant impact on the outcome of the transaction. Second, this rate is highly manipulable, and some large miners control a considerable portion of the computing power.

3. Fixed contract lengths in terms of hashrate do not take into account what happens between the opening and closing of a contract.

Imagine a 3 month contract that opens and closes at the same level. If a miner wanted to hedge their position by going long this contract, they would make $0 at settlement. However, if the hashrate during this period averaged 20% higher than the opening/closing ratio, they would also realize lower returns than expected. This is less considered because theoretically one could trade across the contract to overcome this, but it could also be overcome with some innovative design.

Another way to develop financial derivatives to speculate and hedge hashrate is to use mining difficulty. While mining difficulty offers some benefits over hashrate-derived indices, it also has some issues that need to be overcome.

1. Difficulty is adjusted every 2,016 blocks (≈2 weeks). This means that pricing is very difficult in the early stages of the contract, as estimating future difficulty is essentially impossible and will fluctuate widely.

2. Long-term difficulty contracts do not take into account the difficulty level throughout the contract period. The difficulty can remain at the same level at the opening and closing of the contract, but if the difficulty is higher in the middle of the contract, it will not serve as an effective hedge (unless the contract has liquidity and can be managed in real time).

3. Mining difficulty can be easily manipulated. For example, miners can significantly influence the outcome of the next difficulty adjustment by intentionally turning off their equipment a few days before the contract closes.

Coin Metrics has reduced the impact of many of these issues by designing the following tools, which together can form the basis for effective derivatives market hashrate products.

1: CMBI Bitcoin Hashrate Index

There is currently no definitive way to understand how much hashing power is contributing to the Bitcoin network. Instead, the implied hashing power can be measured by looking at the recent historical time it took miners to produce a block.

The average block time for Bitcoin is 10 minutes. Every two weeks, Bitcoin's difficulty is adjusted to maintain an average block time of 10 minutes. Since solving Bitcoin blocks is a random process that follows a Poisson distribution, the time between blocks can vary greatly.

This can cause fluctuations in hashrate on the network. To date, the industry standard has been to use a 24-hour lookback to view hashrate. However, Coin Metrics believes this is too unpredictable and volatile for a structured financial product. Therefore, we have introduced and will utilize a 48-hour lookback for the CMBI Bitcoin Hashrate Index.

In particular, three blocks took more than 50 minutes to mine in 24 hours in September, causing the industry standard implied 24-hour hashrate to drop by more than 30%. However, this is most likely just a random low-probability event. Although the 48-hour lookback period was also affected, the decline was less than 20%.

More generally, in the chart above, you can observe that the 48 hour lookback produces much less volatility than its 24 hour lookback.

2: Observed Work

As mentioned above, publishing a hashrate index alone is not enough given the short-term predictability, randomness (from oscillations), and tradability issues of traditional hashrate calculations. Therefore, we created Observed Work.

Traditionally, the chain's work calculation assigns a fixed hash value to each block based on the current difficulty (i.e., between difficulty adjustments, whether a block is 1 second or 1 hour apart, the chain's work calculation assigns the same value).

In order to better reflect the workload done by miners and the amount of hash calculations performed in a fixed time period, Coin Metrics' method of observing workload is as follows.

Observed Work = (48-hour implied hashrate) × (time taken to generate the latest block)

By introducing an implied level of hashrate and the time it takes to find the latest block, this representation of work is more consistent with the reality of mining activity than on-chain work calculations.

Coin Metrics’ proposed Observed Work futures contract

Recall from above that miners know the amount of hashrate their equipment can generate, but do not know how much hashrate other miners can and will generate in the future. Based on the implied hashrate, miners understand their current share of the total hashrate, and thus their expected revenue/block reward share.

The Observation workload was developed for financial services institutions to create structured financial products.

• For market participants to calculate the computing power.

• For miners, it is an effective way to manage their computing power risk by hedging the observed computing power over a period of time.

Bitcoin's difficulty level provides insight into the network's expected hashrate at each 2,016-block epoch. Therefore, a financial product that leverages "observed work" allows users to effectively trade expectations of hashrate numbers against the unknown nature of hashrate changes over a fixed period of time.

Here is a theoretical example of observing a 50 minute contract (3,000 seconds) of workload, assuming:

1. At t = 0, the market's expected hashrate is 100 exahash per second

2. Each Bitcoin block, as defined in the Bitcoin white paper, is expected to take 600 seconds

At the contract open, a reasonable expected settlement price would be 300,000 exahashes (3,000 seconds ✖ 100 exahashes/second). However, as shown below, despite the implied hash rate closing at the same level as the open, over time the contract will adjust and close higher than expected at 308,050 exahashes.

Analyze the whole process:

• Block 1 took 600 seconds to be discovered, so the implied hashrate did not change and the observed work of 60,000 exahashes equals the expected work.

• Block 2 is found in 10 seconds, which increases the implied hashrate because the block time is less than the protocol-defined 600 seconds. At t=0, the expected work in 10 seconds is 1,000 exahashes (10 seconds * 100 exahashes/second). However, due to the increase in hashrate, the observed work increases to 1,050.

• Likewise, block 3 is a fast block, so the implied hashrate rises and the observed workload is higher than expected at time 0.

• Block 4 was the first slow block, taking over 600 seconds, so the implied hashrate dropped to 105 exahashes per second. This is still higher than the 100 exahashes per second we would expect at time 0, which causes the observed work for this block to also be higher than expected.

• Block 5 was another slow block, resulting in an implied hashrate of 100 exahashes per second, the same rate as when the hashrate started. Taking this into account, the observed work for this block equals the expected work, producing 119,000 exahashes.

This situation is not uncommon in the Bitcoin protocol. A recent period that demonstrated this result very clearly was mid-January of this year (see the figure below). It can be observed that the implied hashrate opened and closed at approximately the same value over the 2,016 block period. Despite this, the difficulty increased by 5% because the hashrate was above expected levels for most of the time. If miners held a long hashrate position during this contract period, they would not have received the returns they expected from mining at the beginning of the period because the average hashrate was higher than expected. In addition, they would have made little profit on the long hashrate futures contract because the hashrate closed around the level at which it opened.

However, exploring how a two-week work futures contract would perform during this period, the following scenarios show:

• The total amount of work done by miners during the entire period was higher than expected.

• If miners were basing their earnings on a fixed hashrate per second, their returns over the entire period would be lower than they expected.

• If miners are bullish on this contract, they will profit from the increase in observed work over the entire period.

This can be further modeled over longer contract lengths to provide long-term exposure and possible required hashrate hedging (e.g., between ordering and receiving equipment). Below is an example of a 3-month observed work contract after the first difficulty adjustment in 2020. One can visually observe how the expectations of such a contract will change over time as more information becomes available and the contract settlement date approaches.

in conclusion

From the examples above, we can see that such structured financial products will overcome many of the issues discussed earlier that hinder the success of hashrate products.

1. Predictability

The amount of work performed during the contract is always increasing, which makes it less susceptible to the predictability issues that hashrate faces due to its fluctuating patterns. Additionally, while expected work is well understood, observed work is highly dependent on the randomness of block times combined with fluctuations in hashrate. In this case, the difference between expected and observed work is more difficult to predict than short-term hashrate fluctuations.

2. Measuring the performance of the contract over time

Hashrate contracts may close at the top or bottom of an oscillation, which introduces unwanted random risk to traders. Additionally, hashrate levels do not reflect the behavior of the metric over the duration of the contract. Observed Work reflects the history of "work" over the entire contract period and is not subject to the same oscillatory patterns of hashrate.

3. Manipulability

Given that some miners have a large amount of hashrate, they can have a significant and rapid impact on hashrate and difficulty. Observed Work can improve the resistance of hashrate products to manipulation by adding a time-weighted dimension that follows a random Poisson distribution.

Mining is one of Bitcoin's core features and innovations, allowing all of us to benefit from a decentralized, distributed, non-sovereign currency. Therefore, hashrate is a very important on-chain metric that provides the market and network participants with an indication of network strength and security.

To date, the critical role of miners has been unhedged and completely dependent on the price of Bitcoin. However, as the mining market continues to mature and venture-backed businesses and traditional market participants join, these companies will seek mechanisms to hedge their risk exposure and operations, just as they do with other traditional assets.

Together, the CMBI Bitcoin Hashrate Index and Observed Work hope to serve as the basis for financial products that can ultimately provide the market with the tools needed to effectively and efficiently trade or hedge Bitcoin hashrate.