Coin Metrics' State of the Network: Issue 38

Tuesday, February 18th, 2020

Weekly Feature

Analyzing Crypto Supply Distribution Patterns

By Nate Maddrey and the Coin Metrics Team

Those who control the wealth often control the power. But up until now, wealth distribution has been relatively hard to track. People often hide their wealth or obfuscate the true amount of assets that they hold. Cryptoassets take a big step towards making wealth distribution more transparent. 

Cryptoassets are the first asset class where it’s possible to track the full supply distribution throughout its history. Since every cryptoasset transaction is public and auditable, on-chain data can be used to calculate the balances held by every address at any given block. We can then look at the distribution of the size of the balances held by individual addresses to gain insights about the supply. 

However, supply distribution is not a perfect representation of wealth distribution. People often create multiple addresses, and it is difficult to figure out which addresses belong to a specific individual. Additionally, one address could be owned by many individuals, like an exchange cold wallet. To get an accurate cryptoasset wealth distribution you would need to know who controls each address. But transparent, auditable supply distribution gives a fascinating estimation of wealth distribution, and can also tell a lot about the usage patterns of a cryptoasset. 

For example, supply getting consistently more distributed could be a sign that the asset is getting real usage as a medium of exchange. Furthermore, analyzing sudden changes in the amount of supply held by addresses with large balances may lead to insights about selling and trading patterns. 

In this piece we explore the supply distributions of eight cryptoassets, and analyze what the changes in distribution tell us about each asset’s usage. 

Methodology

The charts throughout this piece show the percentage of supply held by addresses holding certain fractions of the total supply. 

We first looked at the balances held by each individual address. We then created groups of addresses holding different sized balances, ranging from relatively small to relatively large. To remain consistent across different cryptoassets, we grouped address balances by fractions of total supply, starting with addresses that hold at least one ten-billionth (1/10B) of total supply (0.0000000001%) and going up to addresses holding at least one one-thousandth (1/1K) of total supply (0.001%). For context, at time of writing, the total Bitcoin (BTC) supply is 18,214,117 so one ten-billionth of total BTC supply is 0.0018214117 BTC, equivalent to about $19. 

We then grouped these addresses into different discrete ranges based on balance size. We started with addresses that hold at least 1/10B but not more than 1/1B, then at least 1/1B but not more than 1/100M, etc., going up to addresses that hold 1/1K of total supply or greater (1/1K+). 

Finally, we calculated the sum of the supply held by all the addresses in each range, to get a percent of total supply held by each group of addresses. We include the cryptoasset’s price on the second y-axis axis (using log scale) to provide context about price changes during supply distribution movements.

It’s also important to note that there are some meaningful differences between the protocol design of different blockchains. For example, the supply of UTXO-based blockchains like Bitcoin becomes slightly more distributed over time as the UTXO set becomes more dispersed due to natural usage (new addresses are often created for each transaction on Bitcoin). This does not happen, however, in account-based chains like Ethereum where addresses are frequently re-used.

All of the data used in this piece is available as part of our Network Data Pro product. You can find more information about Coin Metrics Network Data Pro here

Supply Distributions

Bitcoin

BTC supply was initially held by a few individuals, but over time it has gradually been distributed to millions of different addresses.

The percentage of BTC supply held by large addresses (with a balance of at least 1/1K of total supply) peaked at about 33% in February 2011. As of February 2020, those addresses hold about 11% of total supply. Conversely, the percentage of supply held by smaller addresses with balances of 1/10M and lower has been steadily increasing since 2011. 

There was a relatively large decrease in percentage of supply held large addresses near the end of 2011 through early 2013, before large price increases. Additionally, there was a decrease in December 2018 that was likely caused by Coinbase redistributing its cold wallets.

Source: Coin Metrics Network Data Pro

Ethereum

Unlike BTC, Ethereum had a crowdsale to initially distribute Ether (ETH). The supply of ETH started off highly concentrated but has gradually become more distributed over time. 

The percentage of supply held by addresses with the largest balances (at least 1/1K of total supply) peaked at about 60% in July 2016. The amount held by these large addresses saw a significant decline as the ICO bubble deflated throughout the end of 2017 and into 2018. As of February 2020, these addresses hold about 40% of total ETH supply.

The percentage of supply held by relatively small addresses (with 1/100K of total supply and lower) has been steadily increasing since 2016. 

Source: Coin Metrics Network Data Pro

Litecoin

Litecoin (LTC) had several large dips in the amount held by large addresses (at least 1/1K of total supply) throughout 2013 just prior to the December 2013 price spike, and throughout 2017 before the January 2018 price peak. Interestingly, nearly 46% of supply is still held in large LTC account compared to 11% held in large Bitcoin accounts. 

Source: Coin Metrics Network Data Pro

Bitcoin Forks

Bitcoin forks inherit BTC’s supply distribution (at the time of forking), so may appear distributed simply because BTC itself is relatively distributed. But unlike BTC, Bitcoin Cash (BCH) supply held by large addresses has gotten more concentrated over time.

In August 2017, when it forked from BTC, about 14% of BCH supply was held by large addresses with balances of at least 1/1K of total supply. As of February 2020, large addresses hold about 29% of BCH, compared to about 11% for BTC.

Source: Coin Metrics Network Data Pro

Bitcoin SV (BSV) percentage of supply held by addresses with balance of at least 1/1K has remained relatively flat, outside of a significant dip in February 2019, and a sudden increase in June 2019. In August 2018, when BSV forked from BTC, these large addresses held about 26% of BSV supply. As of February 2020, they hold about 24%.

Source: Coin Metrics Network Data Pro

Ripple and Stellar

Ripple (XRP) and Stellar (XLM) are both account-based chains, and both have official foundations that hold a large percentage of supply. About 85% of total XRP supply is held by addresses with balance of at least 1/1K. 

About  95% of total XLM supply is held by addresses with a balance of at least 1/1K of total supply. This is largely because the Stellar Development Foundation (SDF) holds over half of XLM supply. According to the SDF’s mandate, it currently holds 29.4B XLM. Additionally, the SDF recently burned 50% of total XLM, bringing the supply down to 50B. These burned XLM still appear on-chain since they were sent to a burn address, and therefore get counted as part of the supply distribution. 

Source: Coin Metrics Network Data Pro

Tether

Tether, which is the largest stablecoin by most measures, has released tokens on multiple blockchains. For this analysis, we looked at the Omni (USDT-Omni), Ethereum (USDT-ETH), and Tron (USDT-TRX) versions of Tether separately.

All three versions of Tether started out 100% concentrated. But USDT-Omni and USDT-ETH have gotten increasingly distributed over time. This could be a signal that they are being used as a medium of exchange, which would explain why supply is flowing from addresses holding large balances to addresses holding smaller balances. The Tron version of Tether (USDT-TRX), however, has stayed almost 100% concentrated, which signals that it is likely not getting much usage as a medium of exchange (however, Tether was only introduced on Tron in May of 2019, so is still relatively new).

Also of note, the USDT-Omni distribution trend reversed and started becoming more concentrated in January 2018, near the peak of the market wide price bubble.

Source: Coin Metrics Network Data Pro

Conclusion

Cryptoasset supply distribution gives a clearer window into wealth distribution than any prior asset class, and also provides some interesting insights into trading patterns. The increasing distribution of assets like BTC and Tether is a positive sign that these assets may be getting real usage, and are ending up in the hands of more individual users. We will continue to analyze supply distribution and report on this in the future.

Network Data Insights

Summary Metrics

It was another positive week for the major cryptoassets. ETH continues its strong run, leading the pack in most metrics. Notably, ETH’s realized cap, which can be thought of as the average cost basis of all holders of the asset, increased by 3.6%, while BTC’s increased by 1.3%.

IOTA has been in the news recently after the network was shut down following a hack. We analyze the price implications of this incident in this week’s Market Data Insights section.

Network Highlights

The median transaction fee for both BTC and ETH has increased at least 60% over the last 30 days, outpacing all other major cryptoassets. Median block fees typically rise due to an increased demand for block space, potentially because of increased usage.

Source: Coin Metrics Network Data Pro

Dai (DAI), Paxos (PAX), USD Coin (USDC), and True USD (TUSD)  transfer counts have all been growing faster than Omni-based Tether (USDT), Ethereum-based Tether (USDT_ETH), and Tron-based Tether (USDT_TRX) over the last 30 days. Although Tether is still by far the largest stablecoin in terms of market cap, this may be an early sign that other stablecoins could start closing the gap in 2020.

Source: Coin Metrics Network Data Pro

Market Data Insights

Many assets were relatively flat for the week with a few important exceptions: ETH (+14%) and Tezos (XTZ) (+21%). 

The BTC options market has been pricing in increased volatility over the next several months as reflected in the spread between realized and implied volatility, in part because of elevated open interest in BTC futures markets. But it has yet to materialize. In fact, BTC has traded in a narrow range over the past week compared to other assets. The spread between BTC’s realized volatility and other assets has widened, most significantly in ETH. 

Source: Coin Metrics Reference Rates

Investigating Recent IOTA Price Action

Market efficiency and maturation is a recurring theme for The State of the Network. The recent IOTA incident is another valuable data point to benchmark the industry’s progress. 

At 12:00 PM (17:00 UTC) on Tuesday, February 12, 2020, the IOTA Foundation sent a tweet stating that they were investigating suspicious behavior with the Trinity wallet.

Less than 30 minutes later, IOTA announced on their Status Page that they were shutting down the Coordinator, effectively shutting down the network.

However, it was another 24 hours before IOTA sent a second tweet about pausing the Coordinator, saying they were working with law enforcement and cybersecurity experts to investigate a coordinated attackand paused the Coordinator in order to protect users.

As the below chart indicates, the price was not particularly responsive to this news. In fact, the price did not even drop to levels seen on February 10th. The only noticeable change in price occurred around 08:00UTC on February 13th, roughly 15 hours after the first tweet and 8 hours before the second tweet. 

This lack of substantial price action is somewhat surprising given that the net effect was the shut down of IOTA.

Source: Coin Metrics Reference Rates

To our knowledge, none of the constituent exchanges for the CM Reference Rates halted trading in IOTA. This is notable for two reasons: 

  1. There was no way for users to deposit or withdraw IOTA once the Coordinator had been shut down. This means that only IOTA already on exchanges could be traded.

  2. More importantly, IOTA claims that the attacker was using exchanges to liquidate their stolen holdings, after some obfuscation, and that exchanges have flagged the applicable transactions.

One might expect these developments to contribute to reduced trade volume following an initial spike. As can be noted from the charts below, while there does appear to be a spike in trade volume, it primarily occurs only after the second IOTA tweet-- the response to the first tweet was limited. Additionally, rather than dry up completely after the spike, trade volume appears to actually increase in some markets, even hours after the second tweet. 

A final interesting note regarding the trading activity: at various intervals, there appears to be large volume spikes across some markets prior to the first tweet from IOTA. The earliest social media activity we could find about the IOTA issue was on the IOTA Discord channel, starting at around 10:35AM EST on Wednesday, February 12th. This roughly co-incidents with the timing of a volume spike in the bitfinex-miota-btc-spot market, as can be seen in the chart below. 

Source: CM Market Data Feed

Note that Coin Metrics uses the ticker ‘miota’ to refer to IOTA.

CM Bletchley Indexes (CMBI) Insights

After a strong start to the week, most Indexes gave up their returns over the weekend to close the week out relatively flat. The CMBI Ethereum Index was the best performer of the week, reaching returns of 25% intra-week before finishing the week 14% up. The Bletchley 20 (mid-caps) had their first negative week for the year, finishing as this week’s worst performer, down 3%.

Source: Coin Metrics CMBI

Coin Metrics Updates

This week’s updates from the Coin Metrics team:

  • Coin Metrics is hiring! We recently opened up 5 new roles, including Blockchain Data Engineer and Data Quality and Operations Lead. Please check out our Careers page to view the openings.

As always, if you have any feedback or requests, don’t hesitate to reach out at info@coinmetrics.io.

Subscribe and Past Issues

Coin Metrics’ State of the Network, is an unbiased, weekly view of the crypto market informed by our own network (on-chain) and market data.

If you'd like to get State of the Network in your inbox, please subscribe here. You can see previous issues of State of the Network here.

Check out the Coin Metrics Blog for more in depth research and analysis.

Coin Metrics' State of the Network: Issue 37

Tuesday, February 11th, 2020

Weekly Feature 

Cryptoasset Valuation Research Primer, Part 1

by Kevin Lu and the Coin Metrics Team

Cryptoassets represent a significant innovation in the evolution of money and the modern financial system. Valuing cryptoassets remains very much an open question. Foundational concepts on which a formal discipline of crypto valuation can be built are only beginning to be established. Without a firm anchor to existing methods of asset valuation, we have seen intense experimentation over the past 10 years. 

We conducted a comprehensive literature review to identify all major facets of cryptoasset valuation research that has been conducted so far. All methods were considered, from theoretical valuation frameworks, to empirical valuation models, to novel indicators that have application to valuation. 

In short, we are interested in all research that can be used to understand the current value of cryptoassets, estimate the value of cryptoassets, or predict future values of cryptoassets. All publication mediums are considered, regardless of pedigree, from forum postings to academic journals. The most salient articles from both academic and industry researchers are included. 

In this piece, we explore six main categories of crypto valuation research: equation of exchange, discounted future utilities model. Metcalfe’s law, price regression models, cost of production models, and asset bubble identification. In an upcoming Weekly Feature, we will review factor investing, transfer value-based ratios, realized capitalization, and the emerging field of UTXO analysis as Part 2 of our cryptoasset valuation research primer.

Equation of Exchange 

Fisher’s equation of exchange embodies a strand in the literature that has been influential in the field of token design. Originally designed to explore the relationship between money supply and price level in the field of monetary economics, it has been applied to the field of cryptoasset valuation and currently stands as one of the most widely explored theoretical frameworks. 

The core idea is simple and intuitive: the equation of exchange is the relation MV = PQ, where M is the nominal amount of money, V is the velocity of money, P is the price level, and Q is the index of real expenditures. This is succinctly explained in Wang (2014).

The equation is adapted to take into account the unique aspects of cryptoassets such as Bitcoin (BTC). First, all quantities are defined in units of fiat currency, setting P equal to 1. M is defined as the number of cryptoasset in existence multiplied by the price of a single unit of the cryptoasset (i.e. the market capitalization). Q is the amount of value transferred across the network. And the interpretation of V remains unchanged. 

The implication of this model is that the value of a cryptoasset has an inverse relationship with velocity -- that is, high levels of velocity lead to lower cryptoasset valuations. Wang (2014) finds that the price of BTC is determined solely by the likelihood that BTC will be saved. Ciaian, Rajcaniova, and Kancs (2015) empirically test the impact of supply and demand factors on BTC price, including velocity (which they proxy by using BTC days destroyed), and find that they have a significant impact. 

BTC velocity has decreased in recent years as price has increased, which fits with the equation of exchange model. The following chart shows adjusted velocity (i.e. velocity computed using adjusted transfer value which filters out self-sends and spam transactions) of the one-year active supply (i.e. supply that has been transacted at least once within the last year) of BTC.

The literature’s fixation on velocity continues in Buterin (2017) and Pfeffer (2017), two seminal articles that apply the equation of exchange to the valuation of utility tokens. Both researchers posit that utility tokens are susceptible to extremely high levels of token velocity because users only procure small amounts of tokens to use a service provided by the network and providers of the service immediately sell any tokens they receive. Buterin (2017) concludes that the value of a cryptoasset “depends crucially on the holding time” of the token and argues for token sinks -- mechanisms which reduce the token supply or token velocity (for example, transaction fee burning). Pfeffer (2017) concludes that token velocity could be high at equilibrium, and the value of a utility token will converge to a low level that is a fraction of the actual cost of the computing resources needed to maintain the network. 

ETH adjusted velocity of one-year active supply increased leading up to the 2018 price peak, and has since leveled off. ETH’s 2017 rise in velocity also coincides with the ICO boom and the rise of utility tokens, which has since subsided. ETH’s velocity appears to be relatively in line with ETH price.

Samani (2017) and Samani (2018) present ideas on token economic models which address the “velocity problem” to allow a utility token to accrue value. Burning and staking models are considered. Collectively, the articles published by industry researchers under this topic in 2017 and 2018 were strongly influential in the field of token design. 

The high velocity thesis has garnered its fair share of criticism, however. Evans (2018) presents one of the most salient criticisms, suggesting that previous frameworks view token velocity as an exogenous variable (a value determined outside the model) that can be tuned to a desired level through token mechanism design. Instead, Evans (2018) models velocity as endogenous to the model and a function of PQ. Koralewski (2018) makes similar criticisms regarding the exogeneity of velocity. Weber (2018) challenges the entire foundation of the velocity thesis by arguing that the equation of exchange has been mathematically misapplied and presents two correct applications. Similarly, Locklin (2019) presents a devastating criticism of the equation of exchange and argues that commonly held conclusions regarding token velocity and optimal token design are false. 

Despite its criticisms, the equation of exchange remains one of the most widely adopted frameworks for token valuation. In the coming years, as winners and losers emerge in the utility token space, we expect more empirical research to be done in this area to test the token velocity thesis. 

Discounted Future Utility Models

The discounted future utility family of models takes inspiration from valuation using a total addressable market approach, discounted cash flow analysis, and the equation of exchange. It represents a significant extension of the theory presented in equation of exchange research by advancing an empirical model to estimate the value of utility tokens. Token velocity, among other assumptions, is one of the key parameters in the model. This family of models most closely resembles a discounted cash flow model found in traditional finance, but adapted to the unique characteristics of cryptoassets. 

Burniske (2017a) represents the first significant contribution to the literature by introducing a model to value a hypothetical utility token. Key determinants of the model are the token’s supply characteristics such as the number of tokens in circulation, the total size of the market that the utility token is used to purchase services from, the adoption curve of the network, token velocity, and the discount rate. Winton (2017) introduces a similar model with many of the same model parameters but also allows for different return expectations from different cohorts of investors. 

Burniske (2017b) introduces the terms “current utility value” and “discounted expected utility value”, and presents a theoretical framework for how value derived from the two sources fluctuates throughout the lifecycle of an utility token. 

The application of discounted future utility models remains an area where considerable empirical research can be conducted. Functional networks upon which utility tokens are built are still under active development, and a few significant projects are seeing meaningful amounts of activity. Modeling the value of utility tokens with novel token designs such as burning, discounting, and staking also remain an active area of experimentation and research.

Metcalfe’s Law 

Metcalfe’s law states that the value of a communications network is proportional to the square of the number of connected users of the system. The foundation of this law is in the mathematical relationship that each user of a communication network can make (n - 1) connections with other users. If each connection is considered equally valuable, the total value of the network is proportional to n (n - 1) / 2, which asymptotically approaches n2. Metcalfe’s law has been successfully applied to the valuation of social networks. As an example, Zhang, Liu, and Xu (2015) empirically test Metcalfe’s law and several other network effect laws by using data from Tencent and Facebook and find that Metcalfe’s law fits better than competing laws.

Application of Metcalfe’s law to valuing cryptoassets is straightforward and was first conducted in gbianchi (2014), where the key insight was made to define a user of the Bitcoin network as the number of addresses with zero balance, a proxy chosen after backtesting and considering other alternatives. A formula to predict the price of BTC based on the square of the number of addresses with zero balance is presented. The article also made a significant contribution to another thread in the literature by introducing the idea of tracking the number of addresses between certain balance thresholds. 

Alabi (2017) tests Metcalfe’s law on BTC, Ethereum (ETH), and Dash (DASH), and illustrates how deviations from predicted values can be used to identify asset bubbles. Peterson (2018) applies Metcalfe’s law using an alternative representation for users: the number of wallets from blockchain.info. The model is used to identify a period of suspected market manipulation in 2013. Franek (2018) tests Metcalfe’s law and competing laws on BTC and ETH, and introduces a price-to-Metcalfe ratio to identify periods of over or undervaluation. Kalichkin (2018) combines predictions from Metcalfe’s law with predictions from Odlyzko’s law and similarly introduces a price-to-Metcalfe ratio. 

The application of Metcalfe’s law to cryptoassets touches upon another emergent element in the literature: the study of the number of users of a particular cryptoasset. While the transparency afforded by blockchain ledgers allows for many candidate proxies to represent the number of users, we still lack the clarity of a concept similar to daily active users, a common metric used to track the usage of internet applications. The mapping of on-chain activity to real world entities and individuals is still unclear and an active area of research. As more work is done in this adjacent area, a more precise application of Metcalfe’s law to valuing cryptoassets is possible. 

Price Regression Models

Price regression models refer to an approach to cryptoasset valuation where price is regressed on another variable, typically time (or a variable that is a function of time). The defining characteristic of this approach is that predicted price values can be generated far into the future, with some models predicting prices that are unimaginable today. While some practitioners may dismiss this family of models because of its simple approach, we believe it is a mistake to ignore them entirely -- early models have had remarkably accurate out-of-sample results, have reliably identified historical periods of over and undervaluation, and still receive considerable attention from market participants.

Trololo (2014), building on early prior research in the field of cryptoasset valuation, represents the genesis of the price regression family of models. Using a model that regresses price on the natural log of time, Trololo (2014) was able to predict the date that BTC would reach $10,000 with an error of only a few days at a time when the current price was $275. Residuals from the model are used to identify periods of over and undervaluation. 

Awe & Wonder (2018) uses a similar approach with updated data and provides a prediction for the low of the market cycle with very good out-of-sample accuracy. Burger (2019) presents various price regression models using subsets of the data to test for robustness of fit.  

PlanB (2019) was the next significant advancement in the field and represents one of the most impactful articles in cryptoasset valuation research. Taking inspiration from Ammous (2018), PlanB (2019) posits that there is a relationship between BTC’s value and its stock-to-flow ratio. Stock-to-flow ratio is defined as the inverse of annualized supply issuance, and represents BTC’s scarcity and suitability as a store of value. A test using empirical data finds co-integration between market value and stock-to-flow ratio with high goodness of fit. The model predicts a BTC price of $55,000 after the next halving in May 2020. 

Price regression models have had remarkable success in prediction and adoption. But their legitimacy rests on the assumption that BTC, in particular, will continue on a path to a long-term equilibrium where it becomes a global store-of-value asset, comparable to gold. The next few years will generate additional data which can be used to test the primary assumptions of these models, such as the co-integration between BTC’s value and its stock-to-flow ratio. 

Cost of Production Models 

Cost of production models touch on a strand in the literature which quantify the costs of mining in order to value cryptoassets. Such an approach is intuitively straightforward and is rooted in classical economics where, for example, Adam Smith introduced the concept of a natural price and market price for commodities. The natural price is the price level that is equal to the cost of the various factors of production necessary in producing a commodity. The market price is the actual price that the commodity is sold. Smith argues that the natural price is the central price to which all commodities are continually gravitating. 

Satoshi Nakamoto succinctly explains the foundational logic behind this approach: “The price of any commodity tends to gravitate toward the production cost.  If the price is below cost, then production slows down. If the price is above cost, profit can be made by generating and selling more.  At the same time, the increased production would increase the difficulty, pushing the cost of generating towards the price.”

The application of a cost of production model for valuing BTC occurred out of necessity as early as 2009, the first year of BTC’s existence. New Liberty Standard, the first website to offer a BTC exchange, was also the first to establish an exchange rate for BTC -- the first published exchange rate in October 2009 was 1,309.03 BTC to one U.S. Dollar. In the absence of an established market, the administrator of the website calculated the exchange rate using a simple model that approximated the cost of electricity needed to mine BTC. 

Hayes (2015) represents the first serious treatment of the subject and offers a model based on the cost of electricity, the efficiency of miner technology, the market price of BTC, and the difficulty of mining. Hayes (2016) conducts a cross-sectional analysis on 66 cryptocurrencies and finds that a cryptoassets value can be explained by mining difficulty, the rate of supply issuance, and the type of mining algorithm used. 

The Cambridge Bitcoin Electricity Consumption Index, using miner hardware performance in Bevand (2017) provides a lower bound, upper bound, and best guess estimate for the electricity consumption of BTC. Edwards (2019) presents a model where BTC’s value (referred to as its energy value) is a function of energy input, supply issuance, and the fiat cost of energy input. Predicted values from the model using empirical data find good fit. 

Understanding BTC’s cost of production has an important impact on miners who have a role in the formation of asset bubbles-and-crashes due to their procyclical behavior. The economics of mining and conceptual model for estimating the cost of production of cryptoassets is now well understood. Further advancements in this field are likely to be in gathering more accurate empirical data upon which these models are based on. 

Asset Bubble Identification 

The tendency for speculators to create bubbles in financial assets is deeply rooted in human psychology. Cryptoassets, without a firm anchor to traditional methods of asset valuation, are particularly susceptible to bubbles and bubble-and-crash cycles have happened several times in BTC’s short history. The size and frequency of bubbles in cryptoassets invites the application of bubble detection techniques, first developed in traditional financial assets. 

Cheah and Fry (2015) is among the earliest articles to apply established bubble detection techniques to BTC. Using a variety of bubble detection models, it finds empirical evidence that BTC is prone to substantial speculative bubbles. Using a recently developed detection technique, Cheung and Su (2015) finds many short-lived bubbles and three large bubbles in the period between 2011 to 2013. Wheatley, Sornette, Huber, Reppen, and Gantner (2018) presents a generalized version of Metcalfe’s law which does not require network value to grow proportionally to the square of the number of users to model the fundamental value of BTC. Deviations from predicted values from this model are considered bubbles and they are formally tested using a textbook bubble detection technique. Four bubbles are detected and an ex ante prediction is provided that performs well out-of-sample. 

Conclusion

The Dutch East India Company, founded in 1602, was the first corporate entity to issue bonds and shares to the public, and in doing so became the world’s first formally listed public company. It then took a period of over 300 years for the necessary foundational concepts to be developed until the formal discipline of equity valuation was established in the 1930s. With cryptoassets, we stand on the shoulders of giants, and substantial progress has been made over the past 10 years in the emergent field of cryptoasset valuation research. Existing valuation methods across many disciplines are being adapted to suit cryptoassets. At the same time, unique cryptoasset-specific methods are being actively developed. Foundational concepts are only beginning to be established and many concepts likely remain undiscovered. 

In an upcoming issue of State of the Network, we continue with Part 2 of our comprehensive literature review and cover other significant advances in cryptoasset valuation research. Factor investing, transfer value-based ratios, realized capitalization, and the emerging field of UTXO analysis are among the topics covered. We also share our outlook on the most promising future directions of valuation research. 

Network Data Insights

Summary Metrics

The major cryptoassets had another strong week, continuing a hot start to 2020. BTC was up across most metrics, but ETH was up even more in almost every category. Although BTC led the way for much of 2019, we have now seen ETH and many smaller-cap assets outperform BTC at the start of 2020. ETH posted five days of consecutive positive returns from February 5th through February 9th, which has only happened 21 times in ETH’s history.

ETH also outpaced BTC this past week in usage growth. ETH active addresses grew by 21.5% and transaction count grew by 13.2% week-over-week, while BTC active addresses grew by 4.2% and transaction count grew by 3.3%.

Network Highlights

Velocity of one-year active supply (i.e. the supply that has been transacted at least once within the last year) of stablecoins is near all-time highs. The following chart shows the average velocity of one-year active supply for the following stablecoins: Tether (Omni, Ethereum, and Tron), Paxos, USD COIN, DAI, TrueUSD, and Gemini Dollar. Increasing velocity suggests that stablecoins are changing hands more often, which suggests they are potentially increasingly being used as a medium of exchange.

Tezos (XTZ) realized cap has been growing faster than most other assets over the start of 2020. Realized cap can serve as a proxy for investor cost basis. Although BTC realized cap growth outpaced XTZ over most of 2019, XTZ has surged past BTC in 2020.

Market Data Insights

A large move widely expected by market participants has not yet materialized. Open interest on BitMEX’s XBTUSD contract remains high and the difference between realized and implied volatility in the options market remains elevated. A sharp drop in BTC’s price from $10,100 to $9,800 over the weekend caused some liquidations in the XBTUSD contract but failed to trigger a cascade of further liquidations. 

Market breadth is quite positive with nearly all assets in our coverage universe holding onto strong weekly gains. The recent pattern of other assets outperforming BTC has continued with BTC only gaining +9%. ETH (+21%) is a notable performer, along with XTZ (+41%) reaching an all-time high, and Binance Coin (+31%). 

Ethereum Classic (+3%), Dash (+12%), and ZCash (+7%) have been significant outperformers over the past several weeks but only saw relatively modest gains this week. NEM (+35%) is now a project to watch -- it had one of the strongest weekly gains and is up +98% over the past month. 

CM Bletchley Indexes (CMBI) Insights

This week, cryptoassets and indexes continued their impressive start to the year. Even-weight indexes performed strongest again, demonstrating the market wide nature of this week's performance and the strength of some of the lower value assets in each index. 

Of the market cap weighted indexes, the Bletchley 20 (mid-cap) and Bletchley 40 (small-cap) continued to perform best, providing weekly returns of 19% and 18% respectively. Both these indexes are yet to have a negative performing week in 2020. 

All indexes outperformed the CMBI Bitcoin Index for the week, with the Bletchley 20 Even and the CMBI Ethereum Index performing the strongest, returning 24% and 21% respectively.

Coin Metrics Updates

This week’s updates from the Coin Metrics team:

  • Coin Metrics is hiring! We recently opened up 4 new roles, including Blockchain Data Engineer and Data Quality and Operations Lead. Please check out our Careers page to view the openings.

As always, if you have any feedback or requests, don’t hesitate to reach out at info@coinmetrics.io.

Subscribe and Past Issues

Coin Metrics’ State of the Network, is an unbiased, weekly view of the crypto market informed by our own network (on-chain) and market data.

If you'd like to get State of the Network in your inbox, please subscribe here. You can see previous issues of State of the Network here.

Check out the Coin Metrics Blog for more in depth research and analysis.

Coin Metrics' State of the Network: Issue 36

Tuesday, February 4th, 2020

Setting the Stage for A Total Return Crypto Index 

by Huyette Spring, Ben Celermajer, and the Coin Metrics Team

A Background on Corporate Actions

In traditional capital markets, corporate actions are events that could bring a change to the securities (equity or debt) of a public company. Such events can include mergers and acquisitions, stock splits, dividends and rights issues. Effective handling of such events is standardized and routine, and most investors experience them seamlessly and automatically (the dividend just “shows up” in your brokerage account).

While unglamorous, the significance of accounting for corporate actions is hard to overstate. Imagine that you bought the S&P 500 on the first day of 1988. If you only received the return generated from changes in the prices of the index, you would have earned 1,183%. But if you accounted for the corporate actions of all the companies within the S&P 500, in a so-called Total Return Index, your return was more than double, at over 2,500%.

The key insight here is that financial product creation cannot simply abstract away the economic reality of holding the underlying assets -- asset holders are entitled to those full returns.  

Cryptoassets aren’t just an incremental asset class innovation -- they are a first principles re-evaluation of ‘money’. As such, there has yet to be a defined or transparently articulated set of industry standards that dictate how to treat crypto corporate actions or define what an investor will receive under a set of circumstances. The result is that even if cryptoasset holders are credited with the results of any crypto corporate actions there is inconsistency in application and no clarity or certainty around the process. 

Such inconsistencies are a gating item for broader adoption. The crypto industry’s nascency has masked the criticality of effective corporate actions handling (along with other market infrastructure components), but this is not a situation that can persist. As the industry grows, both retail and institutional holders of cryptoassets will expect to receive the economic returns of the assets that they hold. 

This feature will discuss the methodology and objective eligibility criteria that Coin Metrics has designed to manage corporate actions in crypto, specifically hard forks.  

CMBI Corporate Actions Policy

As part of the recent launch of the CMBI Single Asset Index Series, Coin Metrics has developed the CMBI Fork Legitimacy Policy in an attempt to promote standardization, improve transparency and apply institutional rigor to such crypto corporate actions. 

Generally speaking, there are currently three main types of crypto corporate actions:

  1. Forks

  2. Airdrops

  3. Staking Yields

The CMBI Fork Legitimacy Policy initially addresses forks, which we consider to be the most significant of the three crypto corporate actions, while the other two (airdrops and staking yields) will be the focus of later iterations.

As noted above, crypto financial products must provide holders with the economic return of underlying assets. With the circumstances under which asset holders should experience a return defined, Coin Metrics can develop two series for every index product: 

  • A Price index, which simply tracks the price

  • A Total Return index, which tracks to the return investors would experience by holding the underlying assets (i.e. a cryptoasset and all its associated ‘legitimate’ forks)

Deep Dive on CMBI Hard Fork Legitimacy Policy

Unknown to many, there are over 73 forks of Bitcoin alone. But only a handful of the new cryptoassets created by these forks are sufficiently large enough or adopted enough to have an economic impact. 

Coin Metrics deems a hard fork to have occurred if:

  1. Two or more distinct blockchains with their own clients are in existence post-fork.

  2. Each blockchain shares the same pre-fork blockchain history.

  3. Native tokens on each chain are distinct assets and are not interchangeable.

Once a hard fork has been identified, the CMBI Fork Legitimacy Policy provides a framework utilizing both market data and network data to answer key questions as to the legitimacy of forked assets. 

For CMBI Indexes, the criteria outlined below are observed for up to 12 months after the fork event to determine legitimacy. 

It is worth noting that individually, each of the criteria defined below has limitations. To mitigate this, Coin Metrics have taken the following approach:

  1. Criteria are evaluated over a 30 day time period. Price, volume, and on-chain transaction activity can be manipulated, but often at a cost. By requiring criteria to be met over 30 consecutive days, it becomes increasingly unlikely that the economic benefits of this manipulation exceed the costs.

  2. Creating a set of market and network data criteria, which together represent a comprehensive and manipulation-resistant test. As such, a fork is deemed legitimate only once it meets all criteria outlined below.

In the section below, we will demonstrate how two forks were analyzed against the the legitimacy criteria by reviewing one fork which passed, Bitcoin Cash, and one fork which failed, Bitcoin Gold. The results are summarized in the table below. 

Market Data Criteria

Exchange Support

The adoption of a fork by exchanges plays a critical role in its investability. Only with support from multiple large exchanges can investors have liquidity to buy and sell the forked token and process large transactions. Liquidity and presence on multiple exchanges also indicates that there is enough trading taking place to determine a fair price for the forked asset. 

As such, newly forked tokens will pass the Coin Metrics exchange support criteria if there is support from:

  1. At least one market with a quote currency in U.S. dollars, Bitcoin, or Ethereum on three different exchanges in Coin Metrics’ exchange coverage universe.

  2. At least one exchange that is headquartered and incorporated in the United States and is registered as a Money Services Business with FinCEN or a New York BitLicense. 

Bitcoin Cash and Bitcoin Gold both passed this test.

Price

Along with the actions of exchanges, Coin Metrics deems it important to gauge the perception of a forked token by investors and trading market participants. Under the assumption that markets are at least semi-efficient, the price and market capitalization of forked tokens are proxies for investor/trader acceptance. Since forks can happen during various market regimes and the size of cryptocurrency assets continues to grow, the price of the forked asset as a percent of the price of the parent chain asset is examined. 

Considering all of this, newly forked tokens will pass the Coin Metrics price criteria if the token trades on whitelisted exchanges with the following characteristics:

  1. A new native token will only be considered eligible once its 7-day price volatility (where the price is quoted in units of the parent chain) remains less than 7.5% for 30 consecutive days. Low volatility might indicate that price discovery has occurred and the manipulative price practices that sometimes occur around fork time have subdued. Since cryptoassets are inherently volatile, the test measures volatility against the parent asset (and not against USD) so that volatile movements that are in line with the market are acceptable.

  2. The price of the new native token must remain at least 10% of the price of the native token on the parent chain for 30 consecutive days. This assumes that the fork resulted in a 1:1 issuance. If another ratio is observed, the criteria is adjusted accordingly (e.g. 1:2 would require a 5% ratio between new token price and previous chain token price).

Bitcoin Cash passed both of these tests. Bitcoin Gold passed the Volatility test but failed the Price test, thus failing the test overall. 

Volume

In order for financial institutions and large asset managers to liquidate forked tokens, there must be the presence of significant volume in the market. Volume acts as a further measure to ensure that the exchange, investor and trading community adopt the newly forked token. As such, newly forked tokens will pass the Coin Metrics volume criteria if the native token trades on an exchange in Coin Metrics’ exchange coverage universe, with the following characteristics:

  1. The volume of the new native token must remain at least 10% of the volume of the native token from the parent chain for 30 consecutive days.

Bitcoin Cash passed this test while Bitcoin Gold failed.

Network Data Criteria

Fork Uptake

Fork uptake is a measure of the number of native units that appear active on the newly forked blockchain from the time of the fork. Here, activated is defined as being sent to an address post a fork event. 

This measure provides an indication of how many owners of the parent chain are choosing to “activate” their newly forked native units to either unlock their utility or to sell them (i.e. how much of the supply gets transacted at least once, as opposed to staying indefinitely dormant). This measure is relatively resistant to manipulation as long as the parent asset’s supply is reasonably distributed. If the parent asset is decentralized, it would take coordination amongst many different holders to fake fork uptake.  

Newly forked tokens will pass the Coin Metrics fork update criteria if it meets the following definition:

  1. The fork uptake of the new native token must exceed 10% of the supply of the native token from the parent chain at the time of the fork.

Bitcoin Cash and Bitcoin Gold both passed this test.

Hash Rate

Miners are another important stakeholder in the cryptoasset ecosystem. As such, hash rate is an important metric to examine because it reflects the consensus of miners. Hash rate is also relatively resistant to manipulation because mining equipment is a scarce asset that incurs high variable costs in the form of electricity.

Newly forked tokens will pass the Coin Metrics hash rate criteria if it meets the following definition: 

  1. If the forked asset shares the same consensus algorithm as the parent chain, the hash rate of the forked chain must exceed 10% of the hash rate of the parent chain for 30 consecutive days. If the forked asset uses a different consensus algorithm, this criterion cannot be applied and is ignored.

Bitcoin Cash passed this test. Bitcoin Gold has a different consensus algorithm than Bitcoin, thus this criteria is ignored.

Active Addresses

Active addresses are the number of unique addresses that were either the recipient or originator of a ledger change and can reflect the estimated amount of activity on a blockchain. 

Newly forked tokens will pass the Coin Metrics active addresses criteria if it meets the following definition:

  1. The active addresses of the forked asset must exceed 3% of the active addresses of the parent chain for 30 consecutive days. 

Bitcoin Cash passed this test while Bitcoin Gold failed.

Conclusion

Broader adoption of cryptoassets requires clarity, transparency and consistency. Nowhere is this more important than in determining the economic returns entitled to cryptoasset holders. 

While it will undoubtedly change over time, the CMBI Fork Legitimacy Policy aims to bring standardization, transparency and institutional rigor to crypto corporate actions.  Hopefully the handling of corporate actions for cryptoassets will become as routine and automatic as traditional assets. Standardization around events like forks sets the stage for a crypto Total Return index, which is an important step for the continuing maturation of the crypto industry.  

Network Data Insights

Summary Metrics

Crypto markets rallied this past week, as Bitcoin (BTC) passed $9,000. There is growing evidence that BTC is beginning to predictably react to geopolitical events, and this past week’s cryptoasset rally may have (at least partially) been a reaction to the recent drop in the Chinese stock market. We explore this more in today’s Market Data Insights section.

Adjusted transfer value increased by at least 20% for all five cryptoassets in our sample, outpacing the increases in market cap. Bitcoin Cash’s (BCH) adjusted transfer value is relatively even with Ethereum’s (ETH) -- over the past week, BCH had a daily average of $217M adjusted transfer value while ETH had $234M. BTC still dwarfs them both, with a daily average of $1.9B. 

Network Highlights

Tether continues to gain market share versus all other non-Tether stablecoins. 

Coin Metrics now tracks the amount of Tether that has been issued on Tron, in addition to Ethereum and Omni. The below chart shows how the combined market cap of Tether issued on Tron, Ethereum, and Omni compares to the combined market cap for all of the other stablecoins we track (DAI, USDC, GUSD, PAX, and TUSD). Tether currently accounts for about 85% of the total stablecoin market cap. Comparatively, Tether made up about 77% of the market cap on January 1st, 2019.

Bitcoin SV (BSV)  OP_RETURN transaction count has been increasing over the past week, after falling over the past two months. OP_RETURN transactions are often used to write arbitrary data onto a blockchain and are therefore often used for on-chain data storage. 

BSV OP_RETURN transaction count passed BTC and Bitcoin Cash (BCH) OP_RETURN transaction count in mid-2019, and has been mostly trending upwards since. Check out State of the Network Issue 8 for more of our coverage on how Bitcoin SV is being used for data storage.

Market Data Insights

Recent events such as the U.S.-Iran military conflict demonstrate that under certain circumstances, BTC reacts to geopolitical events. If this cause-and-effect relationship continues to strengthen, the narrative that BTC is uncorrelated to financial assets may need to be re-examined. BTC’s reaction to China’s equity market reopening after the Lunar New Year holiday adds to the growing body of evidence that BTC reacts to global events. As China’s equity markets re-opened, most shares fell by the daily limit within minutes, and a small but significant increase in BTC price was observed. 

Volatility exhibits mean-reverting behavior partially because low levels of volatility encourage higher levels of leverage and risk taking by market participants. BTC realized volatility, measured on a three month rolling basis, is now at 52% and approaching levels that it has historically bounced off of. 

In addition to this, BitMEX’s XBTUSD perpetual swap contract’s open interest has recently exceeded $1 billion dollars, a level that also has historical significance. During the summer of last year, market sell-offs were closely associated with open interest breaching this level. 

Nearly all major assets saw strong gains this week with high dispersion in returns. While BTC outperformed most assets for the majority of 2019, a trend of certain assets significantly outperforming BTC has been recently established. Strong performers this week include Litecoin (+26%), Tezos (+24%), and Cardano (+26%). 

CM Bletchley Indexes (CMBI) Insights

All Bletchley Indexes performed strongly throughout the week, finishing the week between 9% and 19% up. Despite a strong week for the CMBI Bitcoin Index, returning 9%, it was the weakest performer of all indexes. It was the small-cap assets that had the best week, with the Bletchley 40 increasing a staggering 16% for the week, after constituents MonaCoin, ZCoin and BitShares all returned over 50% and Siacoin, Ziliqa and Nano all returned over 20%.

The weekly returns above added what was already a very positive month for cryptoassets.  Over the month it was the Bletchley 20 (mid-cap assets) that performed the best, returning 70% in just 31 days. Large-cap and small-cap assets both performed in line with each other, returning ~35% for the month. 

A trend that has been discussed a bit through the January is the performance of the even indexes. This has been no truer than for the Bletchley 10, where the performance of the even index was almost double the performance of the market cap weighted index. Index design and construction in traditional capital markets can result in very interesting return profiles during different market conditions, something that Coin Metrics hopes to emulate within the cryptoasset market.

Coin Metrics Updates

This week’s updates from the Coin Metrics team:

  • Coin Metrics is hiring! We recently opened up 5 new roles, including Blockchain Data Engineer and Data Quality and Operations Lead. Please check out our Careers page to view the openings.

As always, if you have any feedback or requests, don’t hesitate to reach out at info@coinmetrics.io.

Subscribe and Past Issues

Coin Metrics’ State of the Network, is an unbiased, weekly view of the crypto market informed by our own network (on-chain) and market data.

If you'd like to get State of the Network in your inbox, please subscribe here. You can see previous issues of State of the Network here.

Check out the Coin Metrics Blog for more in depth research and analysis.

Coin Metrics' State of the Network: Issue 35

Tuesday, January 28th, 2020

Weekly Feature

Re-examining Four of the Largest Bitcoin Hacks

By Antoine Le Calvez and the Coin Metrics Team

When cryptoasset exchanges get hacked and large monetary amounts get stolen, news tends to spread fairly quickly. However, articles tend to focus largely on the monetary amount stolen. Rarely do they explore the deeper consequences and fallout resulting from these shocks.

In this feature, we use both on-chain and market data to analyze four of the largest Bitcoin exchange hacks and look at the deep consequences of each, both positive and negative.

Bitcoinica

Often unknown by newcomers to the industry, the Bitcoinica hack was one of the most influential hacks of all-time. Bitcoinica launched in September 2011 and was a Bitcoin trading platform created by Zhou Tong, a teenager at the time. It quickly gained traction and attracted deposits from many prominent community members. In late 2011, Zhou Tong sold Bitcoinica to Intersango (a UK-based exchange) but stayed involved as CEO and lead developer.

From March to July 2012, Bitcoinica suffered a series of catastrophic incidents:

Linode compromise

In March 2012, Bitcoinica’s servers were hosted by Linode. A Linode web portal was compromised by someone that explicitly looked for customers showing any signs of Bitcoin activity. Bitcoinica’s server was therefore targeted and its wallet emptied out.

Very quickly, Zhou Tong made the theft public, even publishing the hacker’s transactions. The theft was made possible by the use of an un-encrypted wallet.

Hot wallet theft

A few weeks after the Linode compromise, another 18.5k BTC was stolen. Zhou Tong promptly disclosed the theft, along with the transaction’s hash. The publicly known cause was the exploit of an email server that escalated into an exploit of the exchange’s hot wallet. Zhou Tong took control quickly enough to avoid the theft of Bitcoinica’s Mt. Gox API key, which could have led to another 15k BTC being stolen (Bitcoinica held BTC on Mt. Gox in order to fill orders). 

Mt. Gox API key exploit

Following a leak of Bitcoinica’s source code, its old Mt. Gox API key was revealed. Unfortunately, it was used as a password to a LastPass account which contained the new Mt. Gox API key. Someone took advantage of this and proceeded to steal 40k BTC + $40k out of Bitcoinica’s Mt. Gox account (the maximum daily withdrawal possible).

Source: Coin Metrics Reference Rates

Consequences 

Overall, 102,101 BTC and $40k of user funds were stolen from Bitcoinica. Roger Ver was probably one of the largest creditors, having held 24,841 BTC on Bitcoinica prior to July 2012. Bitcoin’s price was largely unaffected by all of the hacks and even rallied following the Mt. Gox API key exploit. 

On the positive side of things, the publication of Bitcoinica’s source code inspired many new entrepreneurs. Most notably, Bitfinex’s early codebase was directly issued from Bitcoinica’s. The disappearance of a very successful exchange also left room for competitors to grow.

Unfortunately for its creditors, the downfall of Mt. Gox tied up 64,673 of Bitcoinica’s BTC in bankruptcy proceedings that are still ongoing to this day.

Mt. Gox

Of course, when speaking about Bitcoin exchange hacks, one has to mention Mt. Gox. It was one of the first fiat on-ramps and quickly gained the majority of fiat inflows into Bitcoin from 2010 to 2013. Originally created by Jeb McCaleb (who then went on to help create Ripple and Stellar), it was later purchased and operated by Mark Karpelès. From its inception to its death, Mt. Gox went through a series of hacks that went largely unidentified eventually culminating in its collapse in 2014. 

Following its catastrophic collapse in early 2014, the public finally learned the scale of its mismanagement. An excellent analysis by Kim Nilsson, using Mt. Gox proprietary data, shed more light into how BTC was siphoned off Mt. Gox.

The following excerpt is from the transitional period between Jeb McCaleb and Mark Karpelès, when the first major Mt. Gox hack occurred in March of 2011.

Excerpt from chat log between Jeb McCaleb (Mt. Gox) and Mark Karpelès

79,956 BTC (worth around $70k) were taken out of Mt. Gox’s wallet in March 2011 after the server hosting the wallet was hacked. As mentioned in a previous State of the Network feature, none of this BTC has ever moved since, so it is unknown whether the thief still has the address’ private key.

Later, in September 2011, someone got access to Mt. Gox’s hot wallet file. It contained keys that held BTC at the time, and also unused keys that would end up as deposit addresses afterwards. Over time, the thief slowly withdrew money from the wallet, undetected by inexistent wallet monitoring.

As the thief’s wallet was a copy of Mt. Gox’s, some of the thief’s spending was interpreted as deposits by the Mt. Gox system further muddying the traces of the thefts.

By 2013 practically no BTC was left to be stolen, and Mt. Gox was fully insolvent (apart from 200k BTC held in cold storage, now at the center of bankruptcy proceedings). It wasn’t until February of 2014 that the public became fully aware of the hacks when Mt. Gox halted withdrawals. The price of Bitcoin subsequently crashed. 

Source: Coin Metrics Reference Rates

Consequences

Mt. Gox’s insolvency had a major impact on Bitcoin. Suspicious trading behavior attributable to Mt. Gox occurred during the late 2013 price run-up leading some to think the incredible rise of Bitcoin’s price at the time was not entirely natural.

Its collapse durably depressed Bitcoin’s price following the 2013 run-up. It took slightly more than 3 years for Bitcoin to reach another all-time high.

Mt. Gox was also the introduction to Bitcoin for many in the mainstream crowd. The stigma associated with Bitcoin due to Mt. Gox is still very strong to this day. Had Mt. Gox not happened, one can only imagine what Bitcoin’s current public image would be.

As Mt. Gox concentrated most of Bitcoin’s trading for years, its disappearance left the field open for many competitors. Since then, no other exchange has dominated Bitcoin exchange market share as much as Mt. Gox at its peak. It also highlighted the need for exchanges to monitor their Bitcoin holdings on a constant basis, something even Bitcoinica managed to do.

At Coin Metrics we track on-chain exchange activity as part of our CM Network Data Pro offering.  Specifically, we track the supply of BTC and ETH held by most major exchanges, as well as the amount flowing into and out of each exchange. We believe that tracking exchange on-chain activity is more important now than ever, and it’s crucial to keep exchanges accountable and hold them to a high standard. We have also vouched for the concept of “Proof of Reserves” as a means of exchanges publicly verifying their holdings. Check out State of the Network Issue 34 for more information on how we track exchange health using on-chain data.

Bitfinex

Born from Bitcoinica’s ashes, Bitfinex grew over time as it added more currencies and features to become one of the largest and more influential exchanges of today. From our estimates, Bitfinex had at least 225k BTC under custody on August 1st 2016, just prior to its largest hack.

On August 2nd 2016, 119,756 of these BTC were stolen. They were jointly custodied by BitGo and Bitfinex in 2 out of 3 multisig addresses (meaning 2 out of 3 keys have to sign a withdrawal transactions) where BitGo held one key and Bitfinex the others.

While the details are still unclear, Bitfinex’s BitGo API key was compromised. Due to the lack of checks on how much BTC could be withdrawn in a given time window, very large amounts of BTC were stolen.

At the time news of a breach was made public, there was uncertainty about the amount involved. However, it was possible to get a very accurate estimate using on-chain analysis.

BitGo uses special addresses, known as P2SH (pay-to-script-hash), which enable complex multisignature setups and are well-suited to custody large amounts of BTC. The thief elected to withdraw the heist money to non-P2SH addresses. The specialized website p2sh.info (now txstats.com, a joint BitMEX and Coin Metrics property) tracked the number of BTC stored in P2SH addresses and reflected this large movement a few blocks after they happened, which according to the timeline above, made it the first source of the existence and size of the breach.

Reconstitution of what p2sh.info’s (now txstats.com) homepage would have displayed just after the hack

At the time of the hack, the price of Bitcoin dropped more than $200 but it recovered quickly, in just under 3 months. 

Source: Coin Metrics Reference Rates

Consequences

Bitfinex’s hack is unique inasmuch the exchange survived despite losing 36% of its reserves (on a USD basis). Bitfinex even managed to thrive afterwards, generating $730M in profit over 2017-2018.

Instead of electing to go into a very long and complex bankruptcy procedure (as highlighted by the Bitcoinica and Mt. Gox cases), Bitfinex management decided to use financial engineering to get out of the hole created by the breach.

Each account received a 36.067% reduction in all balances (even though only BTC was stolen) and was credited with an amount of BFX tokens. Bitfinex would either buy back BFX at a ratio of 1 BFX per dollar lost or convert for shares in iFinex Inc, the BVI registered company behind Bitfinex.

Creditors electing to convert their BFX for iFinex Inc shares would also receive Recovery Right Tokens (RRT) allowing them to get exposure to any recovered heist funds once all BFX had been bought back or converted for shares. Any RRT held would give rights to $1 in heist funds recovered.

Furthermore, an open market allowing trading of BFX and RRT tokens was created on Bitfinex allowing creditors to sell their BFX and therefore enable a market-based pricing of each token.

At first BFX traded at 38 cents on the dollar and RRT at 20 cents on the dollar. BFX trading ended in April 2017 close to par when all tokens were either redempted or converted to iFinex Inc shares. RRT still trades to this day at 2.9 cents on the dollar.

All BFX tokens were redeemed or converted to iFinex Inc shares. The use of the BFX token allowed Bitfinex to survive this otherwise critical event. It even turned out to be a profitable trade for creditors that converted their BFX to iFinex Inc shares, as the entity distributed over $500M in dividends in the 2 years that followed.

Bitfinex also used a similar idea to get past the seizure of $850M deposited at payment processor Crypto Capital to raise $1B by selling 1B Unus Sed Leo (LEO) tokens. Each LEO token gives exposure to any recovery of funds from the heist. Any money left after redemption of RRT tokens, legal and other fees, will go towards buying LEO on the open market.

To this day, only 28 BTC have been recovered from this heist. In June 2019, two Israeli brothers were arrested in relation to the Bitfinex hack, but no more funds have been recovered yet.

Binance

The last hack covered by this feature happened on Binance, an exchange that went on to dominate altcoin trading from late 2017 onward. Binance attracted many retail traders and amassed considerable Bitcoin and altcoin reserves.

On May 8th 2019, a 7,000 BTC withdrawal from its hot wallet was triggered. Hackers supposedly broke into many retail accounts via various methods and managed to fool Binance’s hot wallet system into processing such a large withdrawal.

While the details of how the hackers managed to pull-off this heist are sparse, one theory has emerged over time as to how hackers managed to withdraw large amounts of BTC.

Binance lists many exchange pairs (601 active pairs as of writing), the majority of which are illiquid, and therefore cannot support large trades. Hackers can exploit these pairs to concentrate funds from many hacked accounts into fewer ones.

Over time and through various methods, hackers acquired two types of Binance accounts:

  • trade-only API keys that can only be used to send trades from unsuspecting accounts

  • full accounts, authorized to withdraw large amounts of BTC

The hackers placed buy orders at very high prices on illiquid pairs from accounts authorized to withdraw large amounts of BTC and used the many hacked API keys to exhaust the order book of that pair all at once, filling all the buy orders, reaching the orders the hackers placed on the withdrawal accounts. Once all this is done, a large percentage of hacked funds were funneled into the right accounts for withdrawals.

News of the hack had no impact on Bitcoin’s price, in fact, the price rallied shortly thereafter.

Source: Coin Metrics Reference Rates

Consequences

Thankfully for Binance, a prior initiative called SAFU (Secure Asset Fund for Users) was launched in August 2018 and allowed them to avoid insolvency following the theft. They were saving 10% of trading fees in a separate, cold, wallet to handle exactly this kind of situation.

Conclusion

From the genesis of Bitcoin exchange hacks with the Bitcoinica hacking of a single server to the highly complex and well-orchestrated Binance hack, the constant duel between exchanges and whoever wants to steal their reserves has intensified.

Despite the millions in lost funds and the many victims of these hacks, each stands as an important milestone in the maturation of an asset and asset class, providing many lessons for future market participants:

  • Bitcoinica traumatized many, but at the same time allowed new exchanges to be born via its open codebase

  • Mt. Gox’s implosion pushed Bitcoin into the mainstream, resulting in a more fragmented but industrious spot market and granted long-term enthusiasts low Bitcoin prices for many years

  • Bitfinex’s hack and subsequent recovery via financial engineering might have been the genesis idea behind many exchange tokens

  • Binance’s recent hack showed the usefulness of self-insurance as well as the increased sophistication of hackers

Network Data Insights

Summary Metrics

Source: Coin Metrics Network Data Pro

BTC had a relatively stable week, with market cap, realized cap, active addresses, and transaction count all fluctuating by less than 3% week-over-week. ETH, however, saw more of a usage drop over the week, with active addresses decreasing by 17.8% and transactions falling by 21.2%. XRP also had a particularly bad week, as active addresses dropped by over 67%. 

On a positive note, security metrics were mostly up across the board. BTC and BCH led the way, with BTC estimated hash rate growing by 5.7%, and BCH growing by 7.4%.

Network Highlights

The rise of Tether on Ethereum was one of the big stories of 2019. Over the course of the year, Ethereum-based Tether (USDT-ETH) rapidly overtook the Bitcoin-based, Omni protocol version of Tether (USDT) in terms of market cap and usage.

Towards the end of 2019, Tether started gaining traction on a new platform: Tron. There is now close to a billion dollars worth of Tether issued on Tron (USDT-TRX), in addition to the $2.29B and and $1.55B issued on Ethereum and Omni, respectively. 

We recently added USDT-TRX data to our Network Data Pro offering. Find out more about Coin Metrics Network Data Pro here

Source: Coin Metrics Network Data Pro

Tron-based Tether is starting to gain some of the share of Tether transfer value, but the Ethereum version still dominates. The following chart shows the percent share of the total adjusted transfer value of USDT, USDT-ETH, and USDT-TRX. As of January 26th, they have 11%, 74%, and 15% share, respectively.

Source: Coin Metrics Network Data Pro

Market Data Insights

Markets were largely unchanged this week. Tezos (+6%) and Cardano (+6%) were the only notable gainers among major assets. The start of the Chinese New Year celebration which lasts  for one week may have contributed to the muted market activity. 

Developments surrounding the 2019-nCoV novel coronavirus and its uncertain future impact have already had significant impacts on financial assets. Equity markets around the world, particularly Chinese markets, have sold off sharply and we have observed safe haven capital flows to gold, U.S. treasuries, and the typical reserve currencies. Interestingly, Bitcoin sold off slightly in concert with Chinese equities. This risk-off behavior seen in Bitcoin complicates the Bitcoin safe haven theory that has been advocated by many market participants.

While the evidence suggests that Bitcoin may react in a risk-off manner to the 2019-nCoV novel coronavirus, another event that occurred over the weekend provided evidence that further bolsters the safe haven theory. On January 26, 2020 at 16:38 UTC, reports on Twitter surfaced that the U.S. embassy in Baghdad was under attack by several rockets

In a previous issue of the State of the Network, we found that there may be limitations to the degree of Bitcoin’s market efficiency in our study of Bitcoin’s price response to the recent U.S.-Iran military conflict. The most recent developments show, however, that Bitcoin was able to respond instantaneously to an unexpected increase in geopolitical tensions.

Returning to market performance, Ethereum Classic (+7%), Dash (+6%), and ZCash (+4%) continue to outperform the broader market. Maker (-5%) and NEO (-4%) saw slight losses this week. 

Source: Coin Metrics Reference Rates

CM Bletchley Indexes (CMBI) Insights

This week Coin Metrics announced the launch of the CMBI Bitcoin Index and CMBI Ethereum Index. These indexes mark the first to be launched under the CMBI branding, with more single asset indexes, market cap weighted indexes and smart beta indexes to be designed and launched in the future. These initial single asset index products complement the Bletchley Indexes which currently provide a broader market perspective.

To read and learn more about our indexes and how they fit into our broader strategy please check out our website or the recent CMBI launch announcement.

This week, the CMBI Bitcoin Index fell 1% against the USD, whilst the CMBI Ethereum Index finished slightly up. The best performing market cap weighted index was the Bletchley 20 (mid caps), which ended the week 1.5% up, outperforming the CMBI Bitcoin Index by 2.5%. In what has been a recurring theme for the start of 2020, all even weighted indexes continued to outperform their market cap weighted counterparts.

Source: Coin Metrics Bletchley Indexes

Coin Metrics Updates

This week’s updates from the Coin Metrics team:

  • Coin Metrics is hiring! We recently opened up 5 new roles, including Blockchain Data Engineer and Data Quality and Operations Lead. Please check out our Careers page to view the openings.

As always, if you have any feedback or requests, don’t hesitate to reach out at info@coinmetrics.io.

Subscribe and Past Issues

Coin Metrics’ State of the Network, is an unbiased, weekly view of the crypto market informed by our own network (on-chain) and market data.

If you'd like to get State of the Network in your inbox, please subscribe here. You can see previous issues of State of the Network here.

Check out the Coin Metrics Blog for more in depth research and analysis.

Coin Metrics' State of the Network: Issue 34

Tuesday, January 21, 2020

Weekly Feature

Analyzing Exchange Health Using On-chain Supply

by Nate Maddrey and the Coin Metrics Team

On February 7th, 2014, Japan based exchange Mt. Gox halted all Bitcoin withdrawals, claiming that they were trying to resolve an issue that was caused by "a bug in the bitcoin software.” Unbeknownst to the general public at the time, Mt. Gox had suffered number hacks and security breaches from various attackers over the previous few years. In all, about 850,000 BTC was lost, which was close to 6% of the total supply. 

On February 6th, the day before the hack became public, BTC price was $763, and had topped $1,000 for the first time ever just two months earlier. 

On February 24th, the Mt. Gox website went offline for good. That same day, BTC price dipped to $542, and would not top $700 again for over two years.

Source: Coin Metrics Network Data Pro

At the time of the hack, Mt. Gox reportedly accounted for over 70% of BTC trades. Exchanges can be a big potential systematic risk factor for the industry as a whole. If one particular exchange has a huge portion of the market share and that exchange is compromised, the entire market is adversely affected. Therefore, it’s crucial to track exchange activity to know if an exchange is becoming too dominant, and also in order to keep tabs on exchanges and look out for any suspicious activity. 

In retrospect, the Mt. Gox hack was an inflection point for the cryptocurrency industry. It kickstarted a new push for accountability and regulation of exchanges that is still underway today. “Proof of reserves” is the idea that exchanges should publicly verify the reserves that they claim to hold. Although Kraken released a proof of reserves audit in 2014, the industry at large has still done little to move towards implementing proof of reserves. This is partially because proof of reserves poses operational challenges, and also presents potential security risks due to exchange addresses being publicly exposed. Exchanges allowing third party assessments of their balances could be an intermediate step. We welcome the opportunity to collaborate with exchanges on this domain, as they try and build user trust.

In this issue of SOTN, we analyze the amount of BTC and ETH held by exchanges over time using on-chain data. Analyzing supply held by exchanges gives a picture of which exchanges are getting the most usage, and if any exchange (or the market as a whole) is at risk.

Exchange Data Overview

At Coin Metrics we track exchange activity as part of our CM Network Data Pro offering.  Specifically, we track the supply of BTC and ETH held by most major exchanges, as well as the amount flowing into and out of each exchange. We do this by finding and tagging all of the addresses operated by each exchange and tracking the aggregate activity for those addresses. This requires a mix of automated and manual processes with daily oversight. 

For this report, we cover the following exchanges: Bitfinex, BitMEX, Binance, Bitstamp, Bittrex, Gemini, Huobi, Kraken, and Poloniex. Although these exchanges cover a significant portion of the overall volume, it’s important to note that there are hundreds of other exchanges that we did not include in this analysis. Furthermore, our metrics are estimates of the real number of BTC/ETH held by exchanges - there is no way to know the true number unless exchanges prove their holdings through an independent audit. Notably, Coinbase is not included in this report, but we are actively working towards tracking it in the future.

The amount of supply held by exchanges can be thought of as an estimate of the exchange’s usage. Of course, it’s not a perfect measure. Exchange supply increases could mean there are more traders using the exchange, or it could be because more traders are effectively using the exchange as a bank to store their assets. But there’s some evidence that usage correlates with the amount held on exchange. 

The following chart shows Poloniex’s BTC reserves compared to the amount of users connecting to the exchange’s websocket API (that is, the number of users that are online at a given time - not the total number of users with accounts). As BTC reserves declined, so did the number of active API users.

BTC and ETH Supply Held by Exchanges

Over the last five years, the total amount of BTC held on the nine exchanges in our sample (Bitfinex, BitMEX, Binance, Bitstamp, Bittrex, Gemini, Huobi, Kraken, and Poloniex). has generally trended upwards, regardless of market conditions. The following chart shows the aggregate amount of BTC held on the exchanges plotted against BTC’s USD price. 

Source: Coin Metrics Network Data Pro

Individual exchanges, however, have more volatility. While some exchanges, like Binance, have continued to grow after the 2018 price bubble, others, like Gemini and Bittrex, saw a large increase in early 2018 and have since leveled off. One other thing to note is that Bitstamp previously moved their cold storage back and forth between Xapo, which caused a sudden dip in their supply.

Source: Coin Metrics Network Data Pro

This is reflected in the following chart, which shows each exchange’s percentage share (on a monthly basis, averaged over the month) of the total BTC held on the nine exchanges in our sample. Over the last five years, exchange supply distribution for the exchanges in our sample has become significantly more distributed, which is a positive sign for the health of the overall market. 

Source: Coin Metrics Network Data Pro

Unlike BTC, the total amount of ETH held on exchanges in our sample has not consistently trended upward over the last five years. It is difficult to draw global conclusions from this chart since our exchange sample for ETH supply only includes eight exchanges and does not include Coinbase, among others. But one interesting observation is that the the amount of ETH dropped significantly in 2017 and throughout 2018 during the price bubble peak and burst, and has since leveled out. 

Source: Coin Metrics Network Data Pro

Kraken was the first major fiat exchange where ETH was tradeable, so accumulated a relatively large amount of ETH early on. But Kraken was unable to hold onto this early lead. Kraken held close to 9M ETH in January 2017 and had less than 3M by January 2018. Gemini and Poloniex also had their peak ETH supplies by early 2017 and have since declined. 

Source: Coin Metrics Network Data Pro

ETH exchange supply has gotten more distributed over the last five years. As of January 2020, Huobi, Bitfinex, and Binance had 23%, 20%, and 15% of ETH supply, respectively. 

Source: Coin Metrics Network Data Pro

Exchange Case Studies

On-chain exchange supply is also useful for tracking individual exchange activity. Below, we take a look at how key events affected the BTC and ETH supplies for three exchanges: Poloniex, BitMEX, and Huobi.

Poloniex

Poloniex launched in January 2014 and quickly became one of the largest exchanges in the world. During its early years, Poloniex thrived in a largely unregulated market. During 2016 and 2017, Poloniex handled a lot of altcoin trading. But by late 2017, Poloniex was plagued by support issues and was having trouble scaling quickly enough to cope with its new users.

Poloniex was acquired by Circle in February, 2018, with plans to work with regulators and improve Poloniex’s infrastructure. Those plans had a dramatic effect, as Poloniex’s BTC and ETH holdings began to plummet immediately after Circle’s acquisition. By October 2019, Circle announced that they were spinning out Poloniex as a separate entity. Poloniex officially ended support for US customers in November 2019 (and subsequently dropped KYC requirements for accounts with balances of less than $10k), sending their on-chain supplies to their lowest levels since January, 2016. 

Source: Coin Metrics Network Data Pro

BitMEX

BitMEX is a Seychelles-registered derivatives trading platform that offers high leverage trading (up to 100x) on futures and perpetual contracts. BitMEX only allows for BTC deposits (and not ETH), but allows BTC to be traded against many other currencies. 

Similar to Poloniex, BitMEX was also founded in 2014. But unlike Poloniex, BitMEX started off slowly and has been steadily growing over the last few years. Notably, BitMEX has increased their BTC supply during the 2018 bear market. However, it’s important to note that BitMEX has a relatively large “insurance fund” of BTC that reportedly grew by over 63% in 2019. 

BitMEX has also suffered some recent setbacks due to regulatory issues. In July 2019, Bloomberg reported that BitMEX was officially under investigation by the U.S. Commodity Futures Trading Commission (CFTC) about whether BitMEX broke the law by allowing Americans to trade on their platform. This caused BitMEX BTC supply to temporarily dip, but it has since recovered. In November 2019, BitMEX revealed that over 20,000 client emails had been leaked as part of a security breach. However, this did not appear to affect BitMEX BTC holdings. 

Source: Coin Metrics Network Data Pro

Huobi

Huobi was founded in 2013. Huobi had moderate success over its first five years of operations, but is most notable for its large increase in both BTC and ETH supply during the second half of 2019. 

The PlusToken scam was a China based ponzi scheme which was similar in structure to Bitconnect. PlusToken reportedly stole billions of dollars worth of cryptocurrency over 2018 and 2019. It reportedly first started in early July, 2018 then abruptly stopped accepting payments on June 30th, 2019 after publicly posting the message “sorry we have to run.”

In August, 2019, reports first started coming out that BTC and ETH from the PlusToken scam were being sold in large quantities on Huobi. In December 2019, Chainalysis published research claiming that PlusToken coins can be tracked to a few OTC desks that were operating on Huobi. There has also been speculation that the PlusToken selloff may have been directly linked to the BTC rally and subsequent crash during the summer of 2019.

Source: Coin Metrics Network Data Pro

Conclusion

Tracking exchange on-chain activity is more important now than ever. As the industry continues to evolve and regulatory pressure increases, it is important to keep exchanges accountable and hold them to a high standard. Overall, exchange supply appears to be getting more distributed, and the public is starting to become more aware when exchanges are involved with suspicious activity, which is a positive sign for the health of the industry. Up to this point, exchange balances have largely been characterized by third parties like us. Moving forward, we welcome the opportunity to work with exchanges to attest to these findings, and hope to continue to see the industry moving towards transparency and accountability.

Network Data Insights

Summary Metrics

Source: Coin Metrics Network Data Pro

The major cryptoassets continued to trend upwards this past week. ETH and XRP transfers were both up, growing 10.5% and 20.7%, respectively. XRP also led the way in active address growth, growing 34.1% week over week. However, it’s important to note that XRP still has significantly fewer active addresses and transfers than BTC, ETH, LTC, and BCH. Notably, BTC usage and security numbers increased less than the other four assets in our sample, which was not the case for most of 2019.

Network Highlights

BSV thirty-day active supply increased 43% over the past week (from Jan. 13th - 19th), more than any other asset in the below sample of 17 major cryptoassets. This was likely related to BSV’s upward price action as it outpaced all major assets with a 67% price increase over the last week, as noted in the below Market Data Insights section.

Source: Coin Metrics Network Data Pro

BSV active address growth, however, was negative on the week. BSV active addresses decreased by 7% over the week, which is less than most of the other assets in our sample. 

Source: Coin Metrics Network Data Pro

Market Data Insights

Nearly all cryptoassets are continuing to see moderate gains this week. Among the major assets, Bitcoin Cash SV (+67%) performed the best with some market participants attributing the price action to legal developments surrounding Craig Wright. 

Recent market action demonstrates that cryptoasset markets are still susceptible to short-lived bouts of violent price swings. Bitcoin Cash SV was particularly susceptible because a significant number of its holders have not yet claimed their coins since the fork occurred (narrowly defined here as moving their coins from one address to another) and many reputable exchanges have delisted the asset. This creates a situation where liquidity is low, order book depth is shallow, and price discovery occurs on second-tier exchanges that are more amenable to price manipulation. Furthermore, gaining short exposure to assets like Bitcoin Cash SV is either extremely difficult or impossible. 

Source: Coin Metrics Reference Rates

Ethereum Classic (+53%), Dash (+58%), ZCash (+44%), and IOTA (+30%) have similarly seen outsized gains over the past week. The fact that these types of movements happen with regular frequency suggests a momentum-based strategy related to these assets may be possible. 

Source: Coin Metrics Reference Rates

We previously examined Bitcoin’s historical price cycles in State of the Network Issue 27. We found that the previous cycles indicate a pattern of lengthening where each cycle takes longer to complete than the previous cycle -- an expected result if cycles are driven by a new wave of adoption and awareness from a certain group of users, each bigger than the last. 

Source: Coin Metrics Reference Rates

Bitcoin’s price correction following the summer of last year has put the current cycle inline with the previous cycle that started in early 2015. If historical price cycles are reliable guide, we should expect further periods of only moderate price growth interspersed with brief periods of rapid growth and corrections. 

Source: Coin Metrics Reference Rates

Still, market sentiment seems to have significantly improved over the past month. Looking at the distribution of estimated cost basis (an extension of the realized capitalization concept that we discussed in The Psychology of Bitcoin Bubbles as Measured by Investor Cost Basis) reveals that about 72% of all Bitcoin now has unrealized gains. This is a significant improvement from one month ago where this metric stood at 50%. The current distribution shows that the vast majority of Bitcoin has an estimated cost basis less than $12,000 and that at these levels, only a small increase in price is needed to dramatically improve investor sentiment. 

Source: Coin Metrics Network Data Pro

CM Bletchley Indexes (CMBI) Insights

All CM Bletchley Indexes continued their impressive start to the year, performing extremely strongly again last week. Over the last year, it was infrequent to witness Bitcoin underperform the entirety of the Bletchley indexes by as much as it did this week, with the CMBI Bletchley Index only returning 6.5% over the week. Comparatively, the Bletchley 20 (mid cap assets) performed the best during the week with the Bletchley 40 (small cap assets) not too far behind, returning 20.5% and 17.5% respectively.

Despite Bitcoin being one of the weaker market performers over the week, the Bletchley 10 Even was the strongest weekly performer. This was largely due to the staggering performances of Bitcoin SV (67%), Bitcoin Cash (26%) and Stellar (23%).

Source: Coin Metrics Bletchley Indexes

As mentioned above, this week’s performance continued the exciting start to the year for cryptoassets. From the first of Jan, all CM Bletchley Indexes have returned between 25% and 48%. The breadth of the performance across the market can be witnessed in the below chart, where it can be observed that the annual returns of the even weighted indexes has exceeded or been relatively similar to its related market cap weighted index.

Source: Coin Metrics Bletchley Indexes

Coin Metrics Updates

This week’s updates from the Coin Metrics team:

  • Coin Metrics is hiring! We recently opened up 5 new roles, including Blockchain Data Engineer and Data Quality and Operations Lead. Please check out our Careers page to view the openings.

As always, if you have any feedback or requests, don’t hesitate to reach out at info@coinmetrics.io.

Subscribe and Past Issues

Coin Metrics’ State of the Network, is an unbiased, weekly view of the crypto market informed by our own network (on-chain) and market data.

If you'd like to get State of the Network in your inbox, please subscribe here. You can see previous issues of State of the Network here.

Check out the Coin Metrics Blog for more in depth research and analysis.

Loading more posts…