How to Value Cryptoassets (Part II)

Iris
WhaleAcademy
Published in
8 min readOct 18, 2018

Prophesying dips and even lower dips with the power of valuation

“crescent moon digital wallpaper” by Nino Yang on Unsplash

In the previous article, we discussed the oft-used NVT Ratio and NVT Signal as valuation models for large cap cryptoassets. In this article, we will look at a valuation technique based on Metcalfe’s Law of Network.

Although cryptoassets lack proven valuation models, it is crucial to have models in place as a standardized measure of the value projects or coins bring to the ecosystem. Valuation models are consistently proposed to the community, acting as a guide to investors or traders to tell if a coin is overvalued or undervalued, even without ‘earnings’ or ‘cashflow’ data available. NVT Ratio and Signal are very widely used valuation techniques as covered in our Part 1 article.

So, if we already have NVT…

Why look at other valuation techniques?

There are 3 main reasons why:

1. NVT only accounts for on-chain transactions. Hence, the trading activity that happens on exchanges daily are not accounted for in the equation. This presents itself as a problem because off-chain transactions account for a majority of the transactions done through the use of Bitcoin or Ethereum. Thus, NVT might or might not give an accurate picture in providing an estimate of a cryptoasset’s value. To visualize this:

Graph of transaction volumes (on- and off-chains) for BTC, ETH and XRP, with data from Coinmetrics, Coinmarketcap, Binance, blockchain.info and etherscan.io. Artwork: Author’s own.

Compare the off-chain transaction volumes of Bitcoin and Ethereum and the respective on-chain transaction volumes. You will easily be able to see that off-chain transaction volumes may match or even exceed that of transaction volumes on-chain. Reasons why the community might prefer to transact offline and trust an off-chain transaction provider may be due to the feasibility of active trading, its lower transaction costs and shorter time it takes. Bitcartel explores some of the cons of transacting off-chain.

Despite the debate ongoing about on- and off-chain transactions, the fact is that not accounting for them may severely compromise the accuracy of the value estimation of the said cryptoasset.

2. On top of discounting off-chain transactions, the data on daily transaction volume is not easily obtainable. Although the Coin Metrics aggregates daily cryptocurrency data, they openly state that the “transaction volume in USD terms is highly unreliable and may be overstated by a factor of 5–10 or more”. Understand why they say this in their article here.

Unsurprisingly, when comparing the different sources for ‘on-chain transaction volume’ across the internet, we see huge discrepancies. For instance, on 16 October 2018, USD$ 280.5 million reported by blockchain.info as BTC’s on-chain transaction volume was only a fraction of the USD$ 2729 million Coinmetrics reported. Irregularities in compilation and amount of data available cumulate into these discrepancies. Such poses as an issue in the quest to value cryptoassets.

3. A famous critique of NVT attacks it fundamental usage of transaction volume to determine worth of currency. Thomas Clarke of William Blair’s macro fund argues that more transactions should not increase the worth of said currency. Meaning, increasing transactions do not form a positive linear relationship with value. For instance, the Swiss franc has held value better than USD though USD usage is far greater than the Swiss franc.

Hence, valuations would benefit from having other measures to more accurately determine the ‘true’ value of an asset. Although NVT accounts cryptoasset value based on network value and on-chain transactions (albeit imperfectly due to data available), we still need other methods to estimate the value of the network. That other way, suggested by Clearblocks, is inspired by the Metcalfe’s Law.

What is Metcalfe’s Law of Network?

Metcalfe’s Law states that

“The effect of a telecommunications network is proportional to the square of the number of connected users of the system (n²).”

The math behind is as such (and simple to understand, I promise):

If there are n people within a network, the value of the network to each user is proportional to the number of other users (i.e. Instagram is only worth using if your friends are on it). Hence, the total value of the network to the user community is calculated as

n *(n-1)= n²-n

To illustrate the growth in value proportional to its network growth:

If the value of a network of 10 users to an individual user is $1, then the total value of the network to all users is:

10 * (10–1) = $90

If the network grows to 100 users, then the total value of the network to its community is:

100 * (100–1) = $9900

An increase in network size in the tens results in an increase in network value by a hunderedfold.

This law has been influential in quantifying and studying network effects and in valuing online networks. Metcalfe’s formula (M=n²) was proven to fit growths of both Facebook’s and Tencent’s (China’s largest social media network) the best compared to other similar laws like (Sarnoff’s, Zipf’s and Reed’s).

When applied to digital blockchain networks, Ken Alabi shows that Metcalfe’s Law works in this space too. His study, focused on Bitcoin, Ethereum and Dash networks, showed that networks were fairly well modeled by Metcalfe’s, wherein the value of the network is proportional to the number of nodes/end users, squared.

Clearblocks has thus proposed a formula which builds on the fact that price correlates closely with transactions (or active addresses/users).

Price to Metcalfe’s Ratio (PMR)

PMR is developed to be a ratio analysis, which explores the relationship between price/value of a cryptoasset and its network usage (i.e. active addressers or users in this case).

Price to Metcalfe’s Ratio, conceptualized by Clearblocks

In this equation, Clearblocks took the experimental route over formal testing: visually checking for high correlations and tweaking the equation by trial and error. If you prefer to look at their own explanation and figures (of Pearson’s r) you can do so here.

Here is a breakdown of the elements included in the equation:

Use of ln: Their formulas correlated nicely with prices of cryptoassets on a natural log scale, and thus they decided to go with log-transformed data.

Use of daily USD price of cryptoasset: For PMR sensitivity to market changes in the price of cryptoassets.

Use of 30-day moving average of Metcalfe’s law (volume of daily active addresses squared): Clearblocks explained this succinctly,

“(taking) the 30-day backward-facing moving average of unique active addresses (or transactions for Ethereum) and then (using) those values for (their) daily M2 calculation. 60- or 90-day moving averages could also be used and perform similarly (they are a bit less sensitive but potentially slightly more predictive).”

- panek from Clearblocks, 2018

PMR for Bitcoin

Graph of ln BTC (USD) and ln PMR (with a 30 day MA) from 9 September 2015 to 9 February 2018. Source: Clearblocks

Bitcoin’s 3 major price corrections is accurately predicted by PMR. As PMR (solid orange line) crosses 1.0, we see a correction in bitcoin prices (dark blue line). Unlike NVT Signal, PMR does not tout that it identifies bubbles, only corrections.

One point worthy of highlight is that after the price correction Bitcoin experienced early this year, PMR is still “dangerously close” to 1.0 (even though this data is not plotted in the graph above).

This corroborates with the NVT Signal, where we see a sharp increase in NVT Signal in April 2018 (BTC at USD$ 8225). NVT Signal reached highs at previous price correction levels in Sep 2018.

Graph of Bitcoin and its NVT Signal data, from Apr 2011 to Oct 2018. Souce: Woobull.com

With both signals warning of corrections, the community might continue to witness sustained dips in Bitcoin prices.

PMR for Ethereum

For Ethereum, PMR 1.0 predicted 4 major price corrections and PMR 0.25 seem to be good buying opportunities.

Graph of ln ETH (USD) and ln PMR (with a 30 day MA) from 9 September 2015 to 9 February 2018. Source: Clearblocks

If you look closely, you will realize that PMR, however, does not predict the price correction early this year (2018). In fact, it suggests the opposite: that Etheruem is in the strong buy zone. Clearblocks attributed this discrepancy to Bitcoin’s over-inflated NVT and PMR, then its subsequent correction. Since Bitcoin is the dominant cryptoasset, the market likely moved with it, resulting in the Ethereum price crash.

“green plants on soil” by Francesco Gallarotti on Unsplash

PMR needs to mature

There are a few issues that you should consider before throwing NVT away and using PMR wholesale:

We are not sure when correction will happen even if PMR hits 1.0. For example, for both Bitcoin and Ethereum, there are instances where correction only happened after PMR hits 1.5. If you were an active trader, the time between PMR 1.0 and 1.5 would be a loss in profitable trading opportunity. If you were an long-term investor, an increasing PMR trend says only one thing: steer clear from buying (or take this opportunity to liquidate your position). Hence, in terms of a trading indicator, PMR does not seem to outperform NVT Signal.

Also, even Clearblocks doesn’t understand the reason behind the goodness of fit Metcalfe offers in cryptoasset prices. This is a clear sign to keep PMR as one of your indicators instead of your only one, lest there comes a time where losses are sustained but no one can explain why.

PMR needs to be adjusted to each cryptoasset it values. Arguably, PMR might be more accurate in valuing Bitcoin because Bitcoin transactions directly relate to trading (sending bitcoins from peer to peer). To illustrate this, we borrow the example used by Clearblocks:

“In the case of Ethereum, ERC-20 transactions now represent upwards of one third of all on-chain transactions. Because ERC-20 transactions do not directly relate to trading of ETH, it’s possible that we’ll need to discount transactions in the future by some function to account for the rise in non-ETH transactions. Similarly, as dapps begin to launch and Ethereum sees increased non-speculative usage, further discounting may be needed. On the other hand, perhaps this increased usage will not affect M2’s correlative power.”

Basically, no one knows if PMR will accurately describe and predict price consolidations and potential buying opportunities in the future.

In addition to that, a notable point Clearblocks made is that the cryptomarket moves with Bitcoin. Thus, even though Ethereum’s PMR is in the strong buying zone (well below PMR 1.0, and close to PMR -0.25), its prices crashed from the start of this year. This means that despite PMR working well as an indicator of price VS fundamentals of a coin, it needs to be supplemented by other indicators.

Should I use PMR?

Clearblocks clearly state that,

“some variant of PMR will be (a) useful metric moving forward”

showing PMR still needs to mature before formal use by traders and investors as a more reliable market indicator.

PMR is undeniably a robust way to incorporate the value of daily active addresses (a proxy for daily active users) into valuation. However, be sure to use PMR in conjunction with other indicators, while looking out for its more developed variants.

In the next publication, we will look at yet another Metcalfe’s derivative — Network Value to Metcalfe (NVM) Ratio. Stay tuned!

— Article written by Iris Loh

If you would like to join WhaleBlocks in our quest to build a discerning cryptocommunity, or to know more about us:
Website: https://www.whaleblocks.info/
Telegram Review Channel: https://t.me/whalereviews
Telegram Group: https://t.me/whalecap

--

--