The bond crisis is not over

How high must the yield on US Treasuries get before someone buys?

US Treasuries will never go unsold. The real issue is: How high must yields climb before buyers step in.

The so-called bond crisis here isn’t about an imminent US default—it’s about the loss of a stable anchor for long-term financing costs.

30-year Treasury hits 5.3%: This isn’t just another rate hike trade

The 30-year US Treasury yield surged to around 5.3%, reaching a new high since 2007. The 10-year yield rose in tandem. This isn’t just an American phenomenon: Long-term rates are rising in the UK, Germany, and Japan as well.

Short-end rates remain relatively steady, while the long end stands out. This isn’t a simple rate hike trade—short-term yields are easing while the long end rises. What the market is repricing are fiscal risk, inflation credibility, and long-term supply.

In July, US retail sales and PPI data were generally soft, leading the market to ease back Fed rate hike expectations. Short-term yields didn’t follow the long end up: The 2-year yield actually fell during the same period, steepening the curve significantly.

But this doesn’t mean long-term inflation concerns have disappeared. The short end coming down only shows the market isn’t still trading for near-term hikes—it doesn’t prove the market is fully comfortable with the future inflation path.

What’s rising is the real yield and term premium. In July, the US federal deficit reached $432.3 billion, the highest since March 2021. Year-to-date, the deficit is nearly $1.8 trillion, and is expected to remain around $1.9 trillion for the full year. Over the next decade, interest expense may soar to $16.2 trillion. Meanwhile, Treasury Secretary Bessent has intervened in FX markets along with Japan, and the new Fed Chair Walsh’s less transparent communication style is amplifying uncertainty in the markets.

There will always be buyers for Treasuries—the question is how much yield it takes

Long-term buyers haven’t disappeared. Pension funds, insurance companies, foreign reserve managers, and asset managers are still around—they just care more about price now.

Banks’ willingness to buy is suppressed by capital constraints, and foreign investors face higher FX hedging costs, which make buying Treasuries less attractive than before.

It’s not a lack of buyers, but a lack of buyers willing to accept low yields. If yields rise, buyers return—at the cost of higher financing costs than before. The era of low rates is likely gone for good.

AI bonds aren’t the origin—but they add to duration supply

Recently, tech giants’ frequent bond issuance is often blamed for pushing up long-term rates. But more accurately: AI bond issuance has coincided with a peak in long-end Treasury supply this Q3—about $42 billion for 20-years, about $69 billion for 30-years. The combined supply pressure is intensified.

It’s a catalyst—not the origin. The true driver is still fiscal deficits and the scale of debt supply.

The Treasury can buy time, not buy out deficits

As 30-year yields approach multi-year highs, Treasury Secretary Bessent stepped in: Doubling buyback operations from $2 billion to at least $4 billion.

Treasury buybacks aren’t QE. The Treasury has to issue new debt to buy back old debt; the Fed’s QE creates reserves out of thin air to buy assets, truly removing duration from markets. The mechanisms differ, and so does the scale—just the combined issuance of new 20-year and 30-year Treasuries this quarter is around $110 billion, while the additional $14 billion in buybacks is a drop in the bucket.

Treasury buybacks can improve liquidity but can’t reduce the government’s borrowing needs. They buy time, not deficits.

There is no sign of deficit convergence and the scale of issuance will only grow. Fiscal tightening can’t be done politically; the Fed can’t restart QE impulsively while inflation remains above target.

So long-term yields will likely continue to grind higher. The bond crisis isn’t over—it’s just on pause for now.

Who gets hurt, who benefits, as long-term yields rise?

5.3% yields may not cause a crisis. The real danger is a sudden 30 basis point surge in a few days—that alone can trigger deleveraging or a liquidity stampede. It’s not high rates per se, but disorderly rates that are dangerous.

More than VIX, the real stress signals for bonds are: the MOVE index, auction tails, the indirect bidding proportion, and repo market funding costs. These are the real thermometers of a bond crisis’s control.

High rates first hit the valuation anchor. In recent years, a lot of asset pricing logic was: “Rates will fall, so we can discount future cash flows at a lower rate.” That’s now undercut. Growth and high-valuation tech stocks are the first to take the hit.

But it’s too early to call a bear market. This wave of pressure looks more like a structural valuation reset— not a full-blown systemic collapse. Assets with stable cash flow and reasonable valuations should outperform. Only if the bond market truly goes out of order does structural stress spill into a full bear market.

Gold and bitcoin face totally different dynamics. For gold, the logic is clear: Higher credit risk premium, ongoing central bank buying, and added geopolitical risk all support gold’s traditional bids. Longer-term, gold outperforms—but this doesn’t mean it’s a short-term momentum play; it works more as portfolio insurance rather than a trading tool.

Bitcoin is often labeled as “digital gold,” but its pricing logic is closer to a high-beta liquidity asset. Its best runs come amid abundant liquidity and falling real yields. Currently, with high real yields and tight liquidity, the environment isn’t friendly. Only if the Fed is forced to turn dovish would bitcoin truly benefit. For now, it’s more a bet on policy pivot—not a safe haven.

The next six months: Policy suppresses volatility, fiscal stress persists

Looking ahead, either the bond market deteriorates further, or tools are used to soothe investor anxiety and get past the midterm elections in the next six months.

The two paths are not mutually exclusive—more likely, volatility is first capped by policy with fiscal pressure returning afterwards.

The next half year is the pre-midterm election window. The baseline: Policy responses will cap disorder, but not push the rate center lower. Treasury and the Fed will likely coordinate more buybacks, regulatory loosening (e.g., SLR adjustments), and more aggressive guidance—the goal, to cap volatility and prevent market disorder ahead of elections.

Key risk triggers include: bond auction failures, a surging MOVE index, or mass deleveraging. Should these signals flash at once, the “periodic easing” playbook breaks.

Conclusion

A few extra basis points on the 30-year yield isn’t the point.

Treasury buybacks can calm markets, AI investment explains part of the funding pressure, and Fed communication shapes short-term swings—but none of these are fundamental. What really determines long-end rates: fiscal deficits, debt supply, long-term buyers, and the dollar’s credibility.

Markets are being forced to revisit a long-ignored problem: The US can keep piling on debt, but not without consequence.

The low rate era gave growth stocks, duration assets, and far-off cash flow stories ample valuation room. That space is now shrinking.

The most likely scenario over the next six months: “Surface calm with underlying tension.” Policymakers will try to keep sentiment steady, but the deficit and debt supply problems won’t simply vanish.

This bond crisis isn’t over. It has just found a quieter way to persist.