Demystifying Digital Asset Data: A Look at Bitcoin Transaction Value

Introduction

In its raw and unfiltered form, digital asset data can be misleading. When taken at face value, some metrics overstate certain activity, while others may undercount usage. A commonly-cited measure that is often overstated, at least in economic terms, is transaction value on the Bitcoin network, which uses an unspent transaction output (UTXO) system. While the ledger does not distinguish between economic and non-economic transactions, we believe it is crucial to strip out non-economic transactions when feasible.

Transaction Value, a Commonly-Cited Measure

Data providers, such as Coin Metrics, have made strides in separating economic value from non-economic value. A couple filters they apply to the raw data are the exclusion of:

1. Known change outputs, or the portion of a transaction input sent back to the originating address, and
2. Transactions that may be counted multiple times due to successive spends and transfers (resulting from exchange processes, mixers, etc.)¹.

As a result, the difference between adjusted and unadjusted transaction value figures is significant. Bitcoin’s total adjusted transaction value from inception to date is approximately $2.2 trillion compared to the unadjusted figure of approximately $7.5 trillion. In other words, the latest cumulative raw figure overstates economic transaction value by a factor of 3.3x.

Chart 1 demonstrates the difference between cumulative raw transaction value and the adjusted value, which adjusts for change outputs among other factors.

Chart 2 demonstrates the ratio between the raw transaction value and the adjusted value on bitcoin.

Using the unfiltered measure of transaction value also has a notable impact on ratios that use the figure as an input, such as the network value to daily transactions value (NVT) ratio and NVT Signal (NVTS) ratio (network value to 90-day moving average daily transaction value).

Chart 3 demonstrates the difference between using adjusted and unadjusted transaction value figures in the denominator of the NVT ratio.

Given transaction value is the denominator, the raw, overstated figure results in an understated ratio all else held equal. This could be especially misleading when comparing bitcoin to other digital assets based on the NVT/NVTS ratios.

NVT Ratio Introduced as a Type of Price-to-Earnings Ratio

The concept of NVT ratio was developed more than two years ago by Willy Woo, as a type of price-to-earnings ratio for bitcoin, and was later named by Chris Burniske. NVTS is a successor to NVT introduced by Dmitry Kalichkin as a more descriptive measure of bitcoin cycles.

While some factors require making downward adjustments to gain clarity on economic transaction value, certain other factors, such as transactions taking place on Layer 2 scaling solutions, like the Lightning Network, and volume on centralized exchanges, result in an understatement in economic value transacted (and overstatement in ratios that use the input as the denominator).

While digital asset networks provide an abundance of real-time, granular data, it should not be taken at face value and requires deeper analysis to filter out noise and form well-informed conclusions to the extent possible.

Thank you to the teams at Coin Metrics and Avon Ventures for their feedback and suggestions on this analysis.