The U.S. government now pays more to borrow for ten years than it has at almost any point this century. On Wednesday 7 October 2026 the 10-year Treasury yield reached about 5.35% during the session — the highest since 2002, according to TheStreet — before settling at 5.29% in the afternoon, a 52-week high, while the 30-year yield hit 5.671% (TheStreet, 7 Oct). The Treasury’s own end-of-day par yield for the 10-year was 5.28%, one basis point above Tuesday and a hair under the 2026 closing high of 5.31% set on 5 October (U.S. Treasury daily par yield curve).
For a crypto audience the question is simple: when government bonds pay 5.3%, why is anyone holding an asset that pays nothing? The answer in this year’s data is stranger than either bulls or bears usually admit. Since 30 June, bitcoin has risen 42.3% ($58,524 to $83,275) while the 10-year yield rose 84 basis points (4.44% to 5.28%). Yet on the individual days when yields jumped, bitcoin tended to fall. Both things are true, and the gap between them is the whole story.
Two ways to look at the same relationship
There are two separate questions. The first is about levels: does a high yield stop bitcoin from rising over months? The second is about days: when yields move sharply, does bitcoin move the opposite way that day? This desk matched every Treasury trading day in 2026 (192 sessions, 2 January to 7 October) with the corresponding Coinbase daily close, and measured both.
| Window | Bitcoin change | 10-year yield change | Reading |
|---|---|---|---|
| 30 Jun to 7 Oct 2026 | +42.3% ($58,524 to $83,275) | +84bp (4.44% to 5.28%) | Bitcoin rose while yields rose |
| 2 Jan to 7 Oct 2026 | −7.4% ($89,958 to $83,275) | +109bp (4.19% to 5.28%) | Bitcoin fell while yields rose |
| 16 Sep (Fed hike) to 7 Oct | +9.4% ($76,145 to $83,275) | +27bp (5.01% to 5.28%) | Bitcoin rose after the hike |
Over three different windows, the sign flips. That is exactly what a weak relationship looks like: the level of bitcoin is being set by something other than the 10-year yield, whether that is positioning after the June low, ETF demand, or the market’s view on regulation. The yield is not the driver of the multi-month trend, in this sample.
The daily test: yields up, bitcoin down
The daily picture is different. Counting only days when the 10-year yield moved by at least 5 basis points in either direction, this is what bitcoin did on the same date:
| Days (from 1 Jul 2026) | Count | Average bitcoin daily return | Share of days bitcoin rose |
|---|---|---|---|
| 10-year up 5bp or more | 12 | −0.92% | 25% |
| 10-year down 5bp or more | 4 | +1.94% | 75% |
| 10-year within ±5bp | 52 | +0.73% | 58% |
| Days (from 2 Jan 2026) | Count | Average bitcoin daily return | Share of days bitcoin rose |
|---|---|---|---|
| 10-year up 5bp or more | 27 | −1.11% | 30% |
| 10-year down 5bp or more | 15 | −0.15% | 67% |
| 10-year within ±5bp | 150 | +0.21% | 50% |
When yields jumped, bitcoin fell more often than not — on 9 of the 12 jump days since July, it fell. Those are the days that make the narrative “bitcoin hates rising yields” feel true. But three caveats keep it honest. First, 12 days is a very small sample: one or two outliers can move the mean by a full percentage point. Second, the down-yield group has only four days. Third, the standard measure of how tightly two series move together is the correlation coefficient: for daily bitcoin returns against daily changes in the 10-year yield, it is −0.10 across all of 2026 (192 sessions) and −0.29 since 1 July (68 changes). Squaring the second figure gives roughly 8%: yield changes explain about one-twelfth of bitcoin’s daily variation in the recent window, and under 1% over the full year. The sign is negative, which fits the textbook story, but the grip is loose.
Video: WhiteBoard Finance, published 22 September 2026, discussing the 10-year at 5%. The presenter’s views are his own; the figures in this article come from the Treasury and Coinbase.
Why Wednesday does not fit the pattern
The day that prompted this analysis is, if anything, a counter-example. Bitcoin fell 2.65% on 7 October (Coinbase: $85,540 to $83,275), yet the Treasury 10-year moved only one basis point (5.27% to 5.28%) and the 2-year only two (4.79% to 4.77%, a decline). The Fed’s minutes, which said most participants see another hike as likely appropriate by year-end (Federal Reserve), came at 18:00 UTC — and, as this desk showed this morning in its hourly breakdown, bitcoin had already fallen about 2.9% by then. So even in the week of a 5.35% intraday yield, the relationship that matters most to a one-day move was not the one on the bond screen.
There is also a timing mismatch in any such study. The Treasury’s par yield curve is a roughly 3:30 p.m. ET snapshot of the New York bond market, while Coinbase’s daily close is 00:00 UTC (8 p.m. ET). Roughly four and a half hours of bitcoin price action therefore fall outside the matching bond window, which would blur a real relationship rather than create a false one — so the true daily link could be somewhat stronger than −0.29, or it could be an artefact of 68 observations. This desk cannot tell which from this data alone.
Three readings, and what each would require
- “Bitcoin is a rate-sensitive risk asset.” This reading predicts bitcoin underperforms on hawkish days, which the jump-day table supports, but it struggles to explain a 42% gain over a period when the 10-year rose 84bp.
- “Bitcoin is decoupled from rates.” The multi-month sign flips support this, but the −0.29 correlation since July argues against calling it independent on a daily basis.
- “Both: rates set the daily noise, other forces set the trend.” This is the reading most consistent with the numbers on the page, and also the least testable, because “other forces” is doing the explaining.
What would change this analysis
If the 10-year closes decisively above the 5.31% closing high while bitcoin keeps rising, the decoupling reading gets stronger. If the next several yield spikes (the next U.S. inflation report is on 14 October and the next Fed decision on 29 October) are each followed by a falling bitcoin close, the 12-day sample grows toward something more meaningful. Neither outcome would prove cause and effect. As this desk showed in its peak-anniversary analysis, bitcoin is 33% below its 6 October 2025 closing high of $124,720 while the Nasdaq trades near records, and the 10-year yield is more than a full percentage point higher. A market that has repriced the cost of money by that much is entitled to a lower price for the most speculative asset; the surprise in the data is how little of the bitcoin move the daily yield series can explain.
Context from Wednesday: oil rose on Middle East tensions and equities fell modestly (S&P 500 −0.22%, Nasdaq −0.22%), per TheStreet. This desk does not include oil in the statistics above, because the daily series it could access did not reconcile with futures quotes, and it will not publish a correlation it cannot stand behind.
Method: bitcoin prices, highs, lows and closes are Coinbase Exchange BTC-USD candles (UTC day or UTC hour) pulled by Bitcoin Mastery at about 06:20 UTC on Thursday 8 October 2026; the 8 October candle is still open at that time, so any closing figure quoted is the 7 October close unless stated. Treasury yields are the U.S. Treasury’s own daily par yield curve (constant-maturity) series. Third-party figures are attributed to their source and date. Statistics use 192 matched days from 2 January to 7 October 2026; daily returns are close-to-close on Coinbase UTC candles and yield changes are in basis points on the Treasury 10-year constant-maturity series. Small-sample results (12 and 4 days) are descriptive only.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. Bitcoin and other cryptocurrencies are volatile and you can lose some or all of the money you put in. Nothing here is a recommendation to buy, sell or hold bitcoin, any exchange-traded fund, any listed security or any other asset, and the technical levels, probabilities and scenarios discussed are descriptions of published data, not forecasts. Do your own research and consult a licensed financial advisor before making investment decisions.