BIS Study Finds Large Gaps in Bitcoin On-Chain Transaction Data
BIS researchers find Bitcoin metrics vary significantly
A study by the Bank for International Settlements (BIS) has revealed that common methods used to measure Bitcoin transactions may be inaccurate. Researchers found that the value of Bitcoin sent on the blockchain—a digital ledger that records all transactions—can vary by as much as six times depending on the calculation method used.
The discrepancy is primarily caused by how Bitcoin transactions are structured. When a user sends funds, any remaining balance is often returned to the sender as "change." Some measurement tools count this returned change as a new transfer, even though the money never changed owners. This can lead to a massive overestimation of actual economic activity.
Key findings from the BIS report
- Bitcoin on-chain transfer values can differ by 600% depending on how change outputs are handled.
- Traditional market capitalization for Bitcoin has been up to four times higher than its realized capitalization, which tracks the price of coins when they last moved.
- Roughly 54 million out of 67.5 million active smart contracts—automated programs on a blockchain—on Ethereum could not be categorized.
- Stablecoins like USDT, which are digital tokens pegged to the US dollar, serve different purposes depending on the network they use.
Stablecoin activity varies by network
The study, which analyzed 100 billion blockchain records, noted that stablecoin usage is not uniform across different platforms. On the Ethereum network, USDT is mostly used for Decentralized Finance (DeFi), which involves financial services like lending or trading without a central bank. In contrast, USDT on the Tron network is used more for payments and as a way to store value.
The researchers explained that combining these different types of activity into a single number can hide how people actually use stablecoins. In 2022, smart contracts held over 20% of the USDT on Ethereum, but only about 1% of the USDT on Tron.
Industry efforts to refine data
Some financial companies are already working to provide more accurate data. Visa uses an analytics dashboard that filters out non-economic activity like high-frequency trading and automated bots. These filtered results show a significant difference from raw data.
For example, Visa's data showed that while the total volume of stablecoin transactions was $6.4 trillion over a 30-day period, the "adjusted" volume reflecting real economic activity was only $313.1 billion.
Why accurate measurement matters
The BIS researchers concluded that current blockchain indicators should be seen as "noisy approximations" rather than exact measures. Because these numbers are often used to gauge the health and size of the crypto market, inaccurate data can lead to a misunderstanding of how much real money is moving through the ecosystem.
Confirmed facts and uncertainties
The study confirmed that measurement challenges exist across Bitcoin, Ethereum, and Tron. It is also confirmed that current market cap and transaction volume metrics often suggest a level of accuracy that the underlying data does not support. However, it remains unclear how many of the uncategorized smart contracts on Ethereum are being used for legitimate economic purposes versus other activities.
