Treasury Yields Are Not Just About Iran - The Plumbing Explains Why They're Not Coming Back Down

Not a great week for anyone who owns duration, even if the mainstream narrative gives you a neat reason to blame it on. The 10-year Treasury yield closed July 31 at 4.75%. That's up 27 basis points from 4.48% on July 1. The 2-year sat at 4.28%, leaving a spread of 47 basis points - the curve is no longer inverted, but the move to get here is what matters. The 30-year bond hit its highest yield since 2007. That's the part most commentary misses, because everyone is focused on the headline driver.

Here's the narrative: the Iran war pushed oil higher, inflation expectations came with it, the Fed might have to raise rates, so yields moved up. Yes, that's the mechanism for part of this move. Brent crude surged 7% in a single session to $100 per barrel on July 23, and traders priced in a 36% chance of a Fed rate hike at the next meeting. The FOMC ended up holding the federal funds target at 3.50–3.75% on July 29, but it was a 9–3 vote - Beth Hammack, Neel Kashkari, and Lorie Logan wanted an immediate quarter-point increase. Under Kevin Warsh, who took over as Fed chair in May, the committee's split is part of the new normal.

But the oil-and-inflation story doesn't explain why these yields aren't pulling back, even when the geopolitical headlines calm down. Understanding what I understand about spreads and the plumbing would tell you that the real issue isn't just what's pushing yields higher - it's that the system has lost its shock absorber.

During the last phase of quantitative tightening, when the Treasury General Account was being rebuilt through massive bill issuance, there was a cushion. Money market funds held hundreds of billions of dollars in the Fed's overnight reverse repurchase facility - what traders call ON RRP. When the Treasury needed to issue more bills and pull cash from the system, money market funds could simply move that cash out of the RRP window and into the newly issued bills. Reserves at banks didn't budge because the cash was coming from the RRP bucket, not from bank balance sheets. Same deficit. Same Treasury bills. Reserves stayed stable because the funding source didn't change.

That pool has been drained. As of July 29, total reverse repo standing at the Fed was $337 billion. Of that, $334 billion belongs to foreign official and international accounts. The domestic money-fund buffer - the one that actually cushioned reserve drain - is down to roughly $2.6 billion. It's essentially gone.

Which means a greater share of future TGA rebuilding may pull directly from bank reserves rather than from a reverse repo reservoir that no longer exists. And when reserves tighten, SOFR spikes, repo financing gets expensive, and leveraged Treasury positions start consuming balance sheet capacity at dealers. It's not that the plumbing has already seized - it's that the next unit of stress has nowhere to go.

This is the part most market commentary skips. They'll tell you the yield curve is normalizing, which it is - the spread between the 2-year and 10-year has gone from deeply inverted back to roughly 50 basis points. They'll tell you the Fed is on hold, which it is, barely. What they won't walk you through is why the 30-year is sitting at levels not seen since 2007 while everyone argues about whether the Fed is dotting the i's.

The term premium - the extra yield investors demand for holding long-duration Treasuries over a ladder of shorter bills - has reappeared because supply is large, positioning is crowded, and the plumbing is thinner. Jamie Dimon told a podcast he wouldn't buy 10-year Treasuries at current prices. That's not a contrarian sound bite. It's a signal that the biggest potential buyer of U.S. debt at the long end sees the math as unattractive, not because of the Iran headlines, but because of the deficit trajectory and the inflation regime we're still inside.

Even in March 2000, during the dot-com bubble, the 10-year yield was around 6.0%. So 4.75% isn't apocalyptic. But even in 2007, before the financial crisis broke, the plumbing held - the Fed could deploy liquidity because there was room to expand. The comparison isn't in the yield level; it's in the fact that the 30-year hitting 2007 highs while the Fed's balance sheet is shrinking, not growing, is a structural difference that tells you the market is doing the work the central bank used to do.

Here's the conditional chain going forward. If oil pulls back and Iran de-escalates, short-end yields can retrace because rate-hike expectations will ease. The 2-year, which is tethered to Fed funds, has the most room to move back. But the 10-year and 30-year - those are driven by supply, term premium, and the plumbing, not by the Fed's next meeting. Unless reserve balances stabilize or the Fed signals a pause in quantitative tightening, the long end doesn't come back down by much, and any rally is going to be shallow.

If oil doesn't come down, or if the TGA keeps growing through the August and September issuance cycle, you get the double whammy: inflation expectations lock in, and reserves get pulled tighter simultaneously. That's the scenario where yields don't just hold at 4.75% - they push toward 5%.

What to watch. The weekly H.4.1 release from the Fed shows reserve balances and ON RRP usage. If reserves drop below the $3 trillion threshold, you'll see repo rates spike and dealer positioning tighten. That's the plumbing flashing red. Watch the SOFR-to-IORB spread, the difference between what the market charges for overnight secured funding and what the Fed pays on reserves. When that spread widens, cash is getting expensive even if the Fed funds rate hasn't moved. And watch the 30-year, not the 10-year. The long end is where the market's real conviction shows up, and right now it's voting with its feet.

The Iran war is the spark. The exhausted RRP buffer is the fuel. Most of the commentary is still talking about the spark.

Views expressed here are personal and do not constitute investment advice. All data as of July 31, 2026.