BIS paper finds major gap in Bitcoin onchain transfer estimates

Researchers at the Bank for International Settlements say standard onchain metrics can substantially overstate or misrepresent underlying economic activity. For Bitcoin, the BIS found that total onchain transfer-value estimates can differ by as much as sixfold depending on how transactions are counted — a gap driven largely by how measurement methods treat change outputs and other transfers back to the sender. The discrepancy reflects Bitcoin’s UTXO transaction model, where unspent funds commonly return to the sender as change and can be recorded as separate outputs even though they do not represent value moving to a different party.

The measurement issues extend beyond transfer totals. The BIS report notes that the conventional way of calculating Bitcoin market capitalization has at times been up to four times greater than a realized-capitalization measure, which values each coin at the price observed when it last moved. The researchers based their conclusions on an analysis of about 100 billion blockchain records spanning Bitcoin, Ethereum and Tron, and say similar data-structure and classification problems affect other parts of the crypto ecosystem.

Ethereum poses its own challenges because of the sheer number of smart contracts. Of roughly 67.5 million active contracts the study examined, about 54 million could not be classified using the study’s categories. Stablecoins also behave differently across chains: the BIS found that USDT activity on Ethereum was more closely linked to decentralized finance uses, while USDT on Tron appeared more associated with payment-like and store-of-value functions. The share of USDT held by smart contracts in 2022 exceeded 20% on Ethereum but was near 1% on Tron, a divergence that the authors say makes cross-chain aggregation misleading.

Because of these limitations, the BIS recommends treating onchain indicators as noisy approximations rather than direct measures of economic activity. Some analytics firms already try to separate raw blockchain traffic from adjusted metrics that aim to better reflect economic use. As an example, Visa’s Onchain Analytics dashboard (using data from Allium Labs) reports both total and adjusted stablecoin transaction volumes, saying the adjusted figures remove distortions such as high-frequency trading, bots, bridge routing and internal exchange operations. Over the past 30 days, the dashboard shows $6.4 trillion in total stablecoin transaction volume across tracked networks versus $313.1 billion after adjustments.