Key points

  • Bitcoin transfer-value estimates differed by up to six times across measurement approaches examined by BIS researchers.
  • The paper says technical artefacts, smart-contract proliferation and cross-chain differences can obscure economically meaningful activity.
  • Researchers recommend explicit assumptions, classification and disaggregation instead of treating onchain indicators as direct measurements.

A Bank for International Settlements working paper published on September 15 found that estimates of Bitcoin transfer value can differ by as much as sixfold depending on how analysts interpret blockchain records. The result does not mean Bitcoin's ledger is inaccurate. It shows that a transparent record of technical transactions does not automatically reveal the economic activity behind them. The paper extends that warning to Ethereum smart contracts and stablecoin use across Ethereum and Tron, where identical-looking token movements can reflect different purposes.

Bitcoin outputs create an aggregation problem

Bitcoin transactions can combine several inputs and create several outputs. One output may pay another party, while another returns unspent value to the sender as change. A simple total can therefore count value that did not represent a transfer between economic actors. The BIS authors compared approaches that make different assumptions about change and other technical outputs. Their estimates varied by a factor of up to six, illustrating why a single onchain-volume figure can imply more precision than the underlying method supports.

Related reporting: Joe Lubin predicts Ether could see a 100x surge as Wall Street shifts onto decentralized rails.

Market-cap measures answer different questions

Independent coverage of the paper also highlighted the difference between conventional market capitalisation and realised capitalisation. Conventional market cap multiplies current price by circulating supply, while realised cap values units according to the price when they last moved. The two measures have at times differed by as much as fourfold. That gap is not evidence that one calculation is necessarily wrong; each answers a different question. The practical risk appears when researchers compare or combine such figures without explaining their definitions.

Smart contracts add a classification challenge

The study drew on extensive transaction data from Mercurius, a platform covering Bitcoin, Ethereum and Tron. BIS said its toolkit classified 13 million active contracts, including about 1.4 million tokens. Yet the rapid creation and programmability of contracts make it difficult to separate economically meaningful applications from inactive, duplicated or technically generated records. For regulators and market analysts, raw contract counts may therefore say little about usage unless they are paired with a clear classification system and limits on what the data can establish.

The same stablecoin behaves differently by chain

The researchers found that stablecoin activity on Ethereum was more closely associated with smart-contract interactions. On Tron, balances were more often held outside smart contracts, a pattern the paper said was consistent with transactional and store-of-value motives. Aggregating the same stablecoin across chains can consequently blend DeFi use, payments and savings-like behaviour into one number. The authors also found trading activity to be highly concentrated and centred on stablecoins, making careful cross-chain comparisons especially important.

Disclosure matters more than one definitive number

The paper's conclusion is methodological rather than a verdict on crypto's legitimacy. It recommends granular, data-bounded estimates built from explicit assumptions, technical classification and disaggregation. That approach would let readers see which activity has been included and how sensitive a result is to alternative choices. Data vendors, policymakers and researchers remain free to use different metrics, but the study suggests they should disclose those choices and treat onchain indicators as approximations rather than direct measures of economic activity. The authors also note that their views do not necessarily represent the BIS itself.

Sources

AI-generated editorial image; not a photograph of the reported event. Prepared with AI assistance and source verification.