A new Bank for International Settlements working paper has found that Bitcoin on-chain transfer valuation can diverge by up to a factor of six depending on the measurement method, a discrepan
A new Bank for International Settlements working paper has found that Bitcoin on-chain transfer valuation can diverge by up to a factor of six depending on the measurement method, a discrepancy that undercuts the reliability of the network's headline transfer figures used by analysts, policymakers and index providers to gauge economic activity.
The finding sits inside BIS Working Paper No 1377, Hidden by complexity? Measuring stablecoin, crypto and decentralised finance ecosystems, published September 15, 2026 and authored by Timothy Aerts, Ronald Heijmans, Jan Paulick and Violeta Vuletic, according to the BIS research summary. The paper frames the gap as a measurement problem, not a claim about Bitcoin's fair value or exchange trading volume. For related coverage, see Clarity Act Senate Vote to End Debate Fails.
The result lands amid a softer market, with Bitcoin trading near $75,395 and down roughly 4.3% over 24 hours as of this run, against a market capitalization near $1.51 trillion, though the paper's conclusions are analytical rather than price-sensitive. For related coverage, see Kamino Names Yieldstreet Co-Founder Michael Weisz CEO.
The central finding is that estimates of Bitcoin on-chain transaction value differ by up to a sixfold factor depending on how change outputs and self-transfers are handled, meaning the same underlying activity can be reported at wildly different totals. For related coverage, see CLARITY Act Senate Procedural Vote: 60-Vote Threshold.
Bitcoin on-chain transfer estimates
Up to 6×
Bitcoin on-chain transaction-value estimates vary by up to a factor of six across measurement approaches, according to BIS Working Paper No 1377. This measures divergence in transfer estimates, not Bitcoin’s price or exchange trading volume.Source: BIS Working Paper No 1377.The gap is a measurement-uncertainty problem rather than a coverage hole or a single miscalculated figure, since it stems from how outputs are counted rather than from missing data. The authors treat it as evidence that headline transfer totals are approximations shaped by methodological choices.
The paper attributes the divergence to Bitcoin's UTXO model, in which transactions consume whole unspent outputs and return change to the sender; counting those change outputs as transfers inflates totals with movements that never leave the sender's control. That mechanic is native to Bitcoin, not an error in any single dataset.
The study draws on roughly 100 billion records from Mercurius spanning Bitcoin, Ethereum and Tron, though footnote 6 cautions those records include data repeated across processing stages and are not a count of distinct blockchain events or empirical observations.
Section 4.1.1 sets three estimates side by side: an unadjusted total, an adjusted figure that excludes outputs returned to sending addresses, and a conservative lower estimate that strips out the highest output where no self-transfer can be identified. The spread between these approaches is what produces the sixfold range.
The authors illustrate the problem with a 4 BTC input that funds a 1.5 BTC recipient output and returns 2.5 BTC as change, noting that change sent to a fresh address cannot be distinguished from the intended transfer using transaction data alone. That example is illustrative, not an observed transfer.
Critically, footnote 7 describes the lower estimate as a conservative heuristic rather than a mathematically guaranteed bound, given exceptions such as CoinJoin transactions and uncertain classifications, a caveat that separates the paper's floor from a hard minimum. Readers weighing on-chain data should not treat any single adjusted figure as ground truth.
The authors conclude that on-chain indicators should be treated as noisy approximations rather than direct measures of economic activity, a framing that applies to payment volume, settlement value and value-transferred metrics alike.
"Our findings suggest that on-chain indicators should be treated as noisy approximations rather than direct measures of economic activity," the authors wrote in the paper's summary.
The paper addresses how transfer values are measured; it does not assert that Bitcoin is overvalued or undervalued, nor that any specific published volume figure is definitively wrong. Its scope is the interpretation of blockchain data, not an investment signal.
For policy watchers, the implication is procedural: the BIS notes the views are the authors' own and do not necessarily represent official BIS positions, so this is analytical input for how regulators read blockchain metrics rather than a binding ruling. The debate over how authorities treat digital-asset data runs parallel to legislative uncertainty in Washington, where the Senate's failure to advance the CLARITY bill has left market-structure definitions unsettled.
Valuing the amount of BTC moved on-chain is a distinct exercise from pricing the asset itself; the sixfold gap concerns the former, converting transferred coin amounts into value, and says nothing about spot pricing. Broad sentiment sat at a Fear & Greed reading of 69, or "Greed," as of September 15, 2026, a market-wide gauge unrelated to the paper.
The study confines itself to on-chain records across Bitcoin, Ethereum and Tron and does not purport to capture off-chain or exchange-internal activity, which limits any read of it as a complete measure of network throughput. That boundary matters as regulators lean harder on on-chain analytics even while institutional products such as Binance's wealth-management ETF listings push more Bitcoin exposure into regulated, off-chain wrappers.
It is the practice of estimating the total value moved across the Bitcoin blockchain by converting transferred coin amounts into a monetary figure. The BIS paper shows this estimate depends heavily on how change outputs and self-transfers are treated.
It finds that Bitcoin on-chain transaction-value estimates can vary by up to a factor of six across three measurement approaches, an unadjusted total, an adjusted estimate and a conservative lower bound described as a heuristic.
No. The finding concerns divergence in transfer estimates, not Bitcoin's price, its fair value or exchange trading volume, and the authors caution their conclusions represent their own views rather than official BIS positions.
The next concrete trigger to watch is how index providers and central-bank analysts respond to the paper's three-estimate framework, and whether on-chain data vendors begin publishing adjusted transfer figures alongside their unadjusted headline totals in the months ahead.
Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital asset markets carry significant risk. Always do your own research before making decisions.
The post BIS Report Flags Bitcoin On-Chain Transfer Valuation Gap was initially published on Coincu.
