Ninety Days From Today

Count it out on a calendar. Ninety days from this morning is December 17, a Thursday. Between here and there sit two Federal Reserve meetings, on October 27 and 28 and again on December 8 and 9, a midterm election on November 3, a quarterly refunding on November 4, and a Thanksgiving on November 26. That is the window this article is about: one quarter of a year in which the world’s largest bond market has to price a war that shows no sign of ending and a Federal Reserve that just proved it will act anyway.

Two weeks ago this series told you the Fed would hike on September 16 and that the ten-year would test 5% around that meeting. Both happened. The Fed raised its policy rate a quarter point to a range of 3.75% to 4% in a unanimous 12-0 vote, its first increase since July 2023, and the ten-year finished that session at 5.01%, its highest close since 2007 (CNBC; FRED). Being right about that is worth less than being honest about what came next, and what came next went the other way from what this column expected. We will get to that.

The war with Iran entered its seventh month this week and the news is genuinely mixed for the first time. President Trump said Wednesday he is “hopefully” near the end of it, and that Tehran wants to make a deal. He told reporters he had spoken with the Iranians directly. He plans to meet the six Gulf Cooperation Council states on the sidelines of the United Nations General Assembly, possibly next Tuesday, to work on a day-after plan that his own team does not expect to finalize until after the midterms (CNBC). At the same time the fighting widened: the Houthis spent the week consolidating the Red Sea coastal plain and building berms above the Bab al-Mandeb, Saudi Arabia intercepted a Houthi drone over Taif, and a United Nations fact-finding mission said there are “reasonable grounds” to believe American forces committed war crimes in Iran, a finding Washington rejects (ISW; CNBC).

So the premise stands, with a new wrinkle. Assume the war does not improve on its own timetable, and assume the market keeps half-believing it might end. What does that do to the bond market by December 17?

Where the Market Stands Today

The numbers as of this week:

  • The 10-year Treasury yield closed at 5.01% on September 16 and reached 5.041% intraday on September 15, its highest level since July 2007 (FRED; CNBC).
  • The 30-year yield is at 5.35%, after touching 5.401% intraday, its highest since June 2007 (FRED; CNBC).
  • The 2-year yield is at 4.74%, up seven basis points on Fed day and the sharpest mover of the week (FRED).
  • Brent crude settled at $104.82 on September 17 and traded near $104.64 on Friday; WTI closed at $101.91 and touched $103.05. U.S. crude is up more than 18% on the month (CNBC; CNBC).
  • Diesel hit an all-time high of $6.31 a gallon on September 16, roughly 70% higher than a year ago, and gasoline averaged $4.32, up 36% year over year (CNBC; CNBC).
  • The average 30-year mortgage is 6.95%, up from 6.76% the week before and 6.26% a year ago, per Freddie Mac; Mortgage News Daily puts the prevailing rate at 7.19% (Freddie Mac; CNBC).
  • The Fed’s own dot plot shows 16 of 18 participants expecting at least one more hike this year, with four of those seeing two. It marks up 2026 headline PCE to 3.7% and core to 3.4%, and it does not expect to reach 2% until 2029 (CNBC).

The committee’s entire post-meeting statement ran 130 words. “Inflation remains elevated,” it said. “The Committee will deliver price stability.” Kevin Warsh, at his news conference, was blunter: inflation has been “too high and has been for too long,” and the standard for standing still “has not been satisfied” (CNBC).

One more figure belongs here, because it is the one that reaches into your kitchen. Moody’s Analytics puts the cumulative cost of the war at $1,760 per household as of September 11: $930 of that from energy, $425 from higher interest rates, and $405 from military spending that will be paid for in debt or taxes (CNBC). That is the bill the bond market is arguing about.

Why the Flight to Quality Has Not Come

Most people’s instinct about wartime bonds was formed in 1990, and it is worth restating why it does not apply, because the instinct is about to be tested again by the peace talk.

When Iraq invaded Kuwait, long Treasuries initially sold off with oil and then rallied hard through the winter as capital fled to quality. The Gulf War was a demand shock: a slowing economy pulls yields down, and the bond market does its job (New York Times, August 23, 1990).

This war is a supply shock, and the distinction has held for seven months. Oil is blocked at a chokepoint, shipping insurance has repriced, and diesel at $6.31 a gallon feeds into everything that moves by truck. A supply shock raises prices while it slows growth, and a central bank facing that pair cannot ride to the bond market’s rescue, because cutting rates into rising inflation is pouring fuel on the fire. That was the 1970s lesson, and after the 1979 shock it took a federal funds rate near 20% and a deep recession to undo it (Federal Reserve History).

The Fed confirmed the shape of the trap on Wednesday. It hiked into an oil shock not because it wants to slow the economy but because it is afraid the shock becomes an expectation. Warsh cited the breadth of the problem, the labor market’s stability, and the Middle East, and said all three “lend themselves to a firm unanimous decision today.” Notably, three officials who dissented in favor of a hike in July got their hike, and the three who had argued for patience, including Governor Christopher Waller, fell in line rather than dissent (CNBC; CNBC).

There is a second force, and it has nothing to do with the Fed: supply. The war has now cost the United States tens of billions, every week adds more, and all of it lands on deficits that were already demanding heavy issuance from a Treasury market of roughly $30 trillion. The marginal buyer of long-dated paper has changed from foreign central banks to domestic investors who want to be paid for the risk. When the people who lend you money for thirty years start demanding a premium, you do not get to argue with them.

What the 10s30s Spread Is Saying Now

Here is where this installment has to hold itself accountable, because last month’s reading was wrong in an instructive way.

In the September 3 piece, I wrote that the spread between the 10-year and the 30-year was the quietest and most honest signal in the market, sitting at 48 basis points, and I named the two ways it could break. Either it re-widens toward historical norms as the long end wakes up to the supply story, which I called the vigilante scenario and the one “consistent with everything else in this article,” or it keeps compressing, which I called the complacent reading and gave a short shelf life.

It compressed. On the eve of the war in late February, the 10-year yielded 3.97% and the 30-year 4.64%, a gap of 67 basis points. On August 19 it was 54. Entering September it was 45. On September 16, the day of the hike, it closed at 34 basis points, the tightest of 2026 (Treasury constant-maturity data via FRED). Both ends of the curve rose hard this month. The 10-year went from 4.83% on September 9 to 5.01%, while the 30-year went from 5.28% to 5.35%. The belly did the work and the long end barely moved.

So the vigilante showed up late and left early. What does a 34 basis point gap actually mean, given that this spread averaged about 55 basis points from 1990 through 2019 and about 40 since 2020? It means the market has decided the next two years of Fed policy are a bigger deal than the next three decades of Treasury supply. It means investors are being paid almost nothing to extend from ten years to thirty. And it means the term premium, the extra yield that is supposed to compensate for holding duration, is being squeezed by the one actor with unlimited balance sheet and no profit motive: the Fed, whose hike removed the last argument for a long-end bid based on imminent cuts.

Two readings follow, and they point in opposite directions for the window:

  • The compression is a signal of confidence. If the long end is calm while the front end rises, the market is saying inflation expectations are anchored, the Fed will win, and the fiscal problem is a slow-burn issue rather than a crisis. Warsh’s own communication, terse and consistent, may have helped: RBC’s Andrzej Skiba argued after the meeting that clear messaging “could actually help support Treasury prices further out the curve” (CNBC).
  • The compression is a signal of fragility. A flat long end with a hot front end is what you get when the marginal buyer has been induced to show up rather than convinced to. There is a specific inducement in the market right now, and it has a maturity date.

The Buyback Scorecard, Now Actually Tested

Two weeks ago this section had an asterisk: the Treasury had doubled its long-end buyback operations to at least $4 billion each, effective September 9, but the program itself had not yet run, so only the announcement had been tested. That asterisk is gone.

The doubled operations have now been live for a full week and a half, and the ten-year has risen from 4.83% on September 9, the first day of the larger operations, to 5.01%. The thirty-year went from 5.28% to 5.35% (Treasury; FRED). Higher yields on bigger buybacks is not a rounding error. It is the program failing on its own terms.

Set aside for a moment whether the program is good policy and look only at the arithmetic, which has not changed. A $4 billion operation, run a few times a week, against a market that turns over hundreds of billions in a single day is a fraction of one percent of daily volume. Evercore’s Krishna Guha called the plan “a weak form of Operation Twist” that “in itself will have little enduring impact and could backfire if it is seen as signaling concern about the ability to fund longer-term at acceptable cost.” The backfire is now the observable outcome. The visible signal sent was not “we have this under control.” It was “we are worried too.”

Watch November 4. That is the date the doubled program expires and the date of the next quarterly refunding, when Treasury will tell the market what the buyback sizes will be for the next quarter. It is also the day after the midterm election. A quiet rollover at the same size would say the department is comfortable. A bigger operation would say the opposite, and the market will read it in the long end within minutes.

The Governance Question

This series treats the corruption story as a standing part of the bond market, not a political aside, because the term premium is where governance is priced. Three threads ran this month, and each one is a small addition to the premium.

The White House is now openly describing the Fed’s votes as political. Trump said after the hike that he had told Warsh, “You might as well vote with the board. It’s not going to matter,” and called the Fed board “very hostile,” “very political,” and said its members are “raising the rates to make Trump do as bad as they can possibly can do.” He repeated his demand that rates be cut to “1%, or less,” and threatened to halt trade with countries that run surpluses with the United States unless the Fed cuts (CNBC). Warsh’s answer was a refusal to engage: “Part of the independence of the Federal Reserve is we stay in our lane. Independence is a two-way street” (CNBC).

Read that pairing carefully, because it cuts both ways. A unanimous 12-0 hike against the president’s public demand is the strongest possible evidence of Fed independence, and it should be credited. But a president who describes the central bank’s decisions as a personal attack, and who has restarted the process of firing Governor Lisa Cook while keeping a Department of Justice inquiry into the former chair on the shelf, has told the market what he thinks the institution is for (CNBC). Investors price the risk that he eventually gets his way. That is a basis point cost that no hawkish statement can refund.

Tariff authority is being handed out with an exemption channel attached. The House passed a Russia sanctions bill that lets the president impose tariffs of up to 100% on countries buying Russian oil. China took half of Russia’s crude exports as of August, India 37%, Turkey and the European Union 5% each (CNBC). The mechanism is the story for our purposes. The bill “contains exceptions in certain circumstances,” and one trade lawyer told CNBC that the Indian government will “quietly seek confirmation” from the White House whether it qualifies. That is the architecture of favoritism: a broad punitive power, a discretionary gate, and a queue of governments asking privately which side of the gate they are on. A bond market that watches the rule of law watch the rules get discretionary will charge for it.

The trade war is now a bans regime. Washington will ban imports of Canadian motorcycles, whey, molasses, nonalcoholic beer, wine, cider, whisky, and vodka starting September 29, replacing 50% tariffs on those goods. Canada has retaliated with tariffs on C$27.6 billion of American imports (CNBC). Tariffs that a court can review are one thing. Flat prohibitions on named consumer goods are a different kind of instrument, and each escalation gives the long end one more reason to demand compensation for holding a claim on a government that keeps changing the terms of trade.

The Ninety-Day Map

Take the premise seriously: no improvement, no breakthrough, no reopening of the strait. Here is the path from September 18 to December 17.

Late September. Trump meets the Gulf states around the UN General Assembly, and a UN report accusing American forces of war crimes sits on the table. Watch oil first and bonds second: if a Gulf meeting produces anything that looks like a framework, crude falls and the long end rallies before the Fed says a word. The August PCE report lands September 30 and the September jobs report October 2, the two data points that will decide how isolated the December hike is.

October 14. September CPI. If diesel at $6.31 and gasoline at $4.32 have flowed through, headline inflation stays above 3% and the December hike becomes consensus rather than probable. This is also the month the Fed is likely to skip.

October 27-28. The Fed meets with no projection materials and no expectation of a move. Goldman Sachs Asset Management expects a skip, describing a committee that “does not at this stage envisage an aggressive tightening cycle” (CNBC). The interesting question is the statement’s length, not its content. At 130 words, every addition is a signal.

November 3 and 4. The midterms, then the refunding. A contested or slow-counted election during a war adds a risk premium rather than removing one. The refunding the next morning tells you whether the buyback experiment continues at size, and whether Treasury’s issuance mix leans further on bills, which is how a government avoids admitting that the long end is expensive.

November 26. Thanksgiving. This is the date the diesel bill becomes a household item rather than a statistic. The National Energy Assistance Directors Association expects home heating oil customers to pay as much as 31% more this winter if prices hold, and the federal heating assistance program is unlikely to get more money with Congress in recess until November (CNBC).

December 8-9. The last FOMC meeting of the year, with a fresh dot plot. Sixteen of eighteen participants already see another hike; if the data holds, this is where it lands. The realistic landing zone: the 10-year in the 5% to 5.4% range, the 30-year above its 2007 high on a closing basis, mortgages at 7% or above, and a 10s30s spread that has either finally begun to re-widen or has gone flat enough to stop being a signal at all.

That is the base case. Two branches sit on either side of it:

The escalation branch. A full closure of the Strait of Hormuz, or strikes that draw Iran’s neighbors in directly, takes roughly a fifth of the world’s oil off the market in a stroke. That is the 1979 scenario, and the 1979 precedent says the Fed’s choice becomes Volcker’s choice: break the inflation or bequeath it. Bonds sell off hard at the long end no matter what the Fed does, the 10s30s finally re-widens, and the question stops being whether yields hit 5% and becomes whether the Fed will do what it takes to make 5% temporary. For the slower-burning equity analog, 1990’s Gulf War drawdown was 19.9%; the 1973-74 bear market fell 48.2% from its January 1973 peak to its October 1974 trough, and it is the precedent that matters when a supply shock meets a central bank that will not look through it (PortfolioCalc).

The de-escalation branch. This is the new one, and it deserves more than a footnote because the market has started to price it. Trump says the Iranians want a deal and that he has spoken with them directly. Warsh says nothing about peace, because he cannot. If a framework emerges from the Gulf meetings, oil falls, the Fed’s inflation problem shrinks without the Fed doing anything, and long bonds rally hard, precisely because the long end has spent seven months being told that the inflation force wins. In that world the ten-year could give back 50 basis points in a fortnight, and the 10s30s spread would likely steepen on the way down rather than up. The prior installment’s paradox branch was a recession. This installment’s is a peace, and it is the single largest risk to the bearish case sketched above.

What a Careful Investor Does

None of this is a prediction that yields rise in a straight line. It is a statement of asymmetry: if the war does not improve, the risks to long-term yields are skewed upward, and the price of being wrong about duration is much higher than the reward for being right. The de-escalation branch is the reason the asymmetry is no longer as one-sided as it looked two weeks ago, and it is why the playbook is about structure rather than direction.

The playbook is not new, and it is not clever:

  • Keep duration short at the front end. Bills and short notes let you reinvest as the curve moves instead of locking a long-end loss into your statement. The 2-year at 4.74% is a defensible home for new money that needs to be liquid within a few years (FRED). The 30-year at 5.35%, with the 10-year only 34 basis points below it, is paying you very little for thirty years of inflation and issuance risk.
  • Do not confuse the compressed spread for a bargain. Thirty-four basis points is not compensation for the two risks that live exclusively at the long end. If you want duration, own it deliberately and size it small, not because the gap between the ten and the thirty looks tidy.
  • Own inflation directly. The August CPI print was 3.4% headline and 2.4% core, and the Fed’s own forecast has headline PCE at 3.7% for the year (CNBC). In a market where the nominal coupon may understate what your money will buy, TIPS do work the coupon cannot.
  • Ladder, do not time. Nobody knows whether December brings a second hike or a Gulf framework. A ladder converts that ignorance into a schedule, and it is the only structure that survives both branches intact.
  • Keep liquidity for whichever branch shows up first. The de-escalation rally and the flight-to-quality rally look identical in the first hour and are opposite trades by the second week. Cash is what lets you tell them apart before you commit.

The Long Way of Saying It

The bond market is not a casino and it is not a democracy. It is an ongoing referendum on whether a government’s promises are worth holding, and over the next ninety days that referendum has four questions on the ballot: an oil chokepoint that stays half-shut, a central bank that just chose inflation credibility over a president’s demand, a treasury department buying its own long bonds while the long end climbs anyway, and an election that will decide how much of this anyone in Washington is willing to fix.

There is also, for the first time since February, a fifth question underneath the other four: what does the price of a thirty-year bond look like in a world where the war ends? Nobody knows, which is why the long end stopped moving this month while the front end did all the work. That is not calm. That is a market holding its breath.

You do not need to know the answers to invest well through them. You need to respect what the market is already saying. The ten-year went from below 4% before the war to a 2007 high this week. The Fed hiked unanimously against the president’s wishes and told the market to expect more. Diesel is at an all-time high, and Treasury is buying long bonds while yields rise. Storms do not send invitations, but they do leave clues, and right now the clues are written in basis points.

Ninety days is not a long time. It is long enough to have your defenses built before the lightning strikes, and long enough to remember that the opportunity hidden in uncertainty belongs to the ones who kept their powder dry.

PRH | huffmanwrites.org | © Philip Huffman

Sources