On Treasury Volatility, Zoom Out—Not In

Here are four sets of numbers:

First: 0.08, 0.07, 0.05.

Second: 0.38, 0.35, 0.11.

Third: 0.45, 0.37, 0.12.

And, fourth: 0.51, 0.41, 0.11.^[i]

These, friends, are the cumulative change in 30-year, 20-year and 10-year Treasury yields, respectively, in percentage points (ppt) since Tuesday’s close, three months ago, six months ago and one year ago. What do you see in them? Do they seem large … or small? Do they indicate panic or calm? Nothingness or crisis?

We ask these questions because it appears to us pundits have decided a 0.08 ppt or 0.07 ppt move in very long-dated Treasury yields since Tuesday’s close are a sign something in bond markets is awry. They point to an auction of 20-year bonds Wednesday, which allegedly saw “weak” demand, causing “sharp declines in [bond prices],” which would spike yields (since the two move inversely).^[ii] Many pinned Wednesday’s US stock market selloff on this exact point. Zoomed in, on a Y-axis that goes all the way from 4.3% to 5.2%, maybe the upturn in 30-year yields looks big.^[iii] To some, the spike this chart seems to show even indicates the often-feared, rarely seen “bond vigilantes” of yore are back to wreak havoc on the overindebted US.

But zooming in to less than a percentage point is a ridiculous scale to use, too small to give proper perspective on the subject. Hence, our series of numbers. In our view, stripping out the emotion in this fashion reveals the reality: Yes, yields are up over the last two days, three months, six months and year. But the magnitude isn’t alarming—these moves are far from large by historical standards. Nor are these rates’ levels particularly high when viewed longer term.

You can say the same of the allegedly “weak” demand at Wednesday’s auction. In all the coverage we read, few mentioned the bid-to-cover ratio. This metric compares the dollar amount of bids to the dollar amount of debt sold. For the 20-year debt the Treasury sold this week, it was 2.46.^[iv] This means there were nearly 2.5 times more bids than bonds sold. Is that bad? Well, here is another set of numbers. In February, the bid-to-cover on a 20-year auction was 2.43—nearly identical.^[v] In November? 2.34.^[vi] That is lower! And a year ago, it was 2.51.^[vii] Yes, Wednesday’s bid-to-cover was down from April’s 2.63 and March’s 2.78. But, like yields, there is little sign of a sea change here and more to suggest demand is bouncing in a fairly typical range. That auction also occurred the same day overseas bond auctions were disrupted by a Bloomberg terminal outage.^[viii] The outage of this widely used system by bond investors doesn’t seem to have directly affected the US auction, but it could easily have swayed sentiment.

Exhibit 1: 20-Year Treasury Auction Bid-to-Cover Ratios

Source: US Treasury, as of 5/21/2025.

As for the bond vigilantes, take all the claims of their return now with many grains of salt. There were warnings aplenty in late 2023 of their return.^[ix] In February 2018, pundits warned, “[the bond vigilantes] have saddled up, and they’re riding high. The posse is getting ready. They’re getting the message out.”^[x] In May 2009, headlines warned the vigilantes were back with knives out over the federal budget deficit.^[xi] A generation of low rates followed, lasting so long that many people psychologically reframed. They now see those historically low rates as “normal,” stoking fear of anything actually normal longer term.

When fear is running hot, a crucial tool you have in your toolkit is zooming out. Volatility can tempt you to do the opposite. Don’t let it.