Yields at a 20-Year High, Stocks at a Record High — Notes on MacroMicro After Meeting EP. 217

Contents

1. 2004 to 2006: 425 basis points of hikes, and the long end stayed put
2. Early 2026 to October: one hike, and the long end went from 4% to 5.3%
3. Early October in France: a budget that cannot pass, and a yield through 5%
4. September to October in the US: payrolls of 29,000, read as resilience
5. Where this stands: the term premium is the line to watch
6. What to watch next: 94%, and next year’s base effect
7. “The bad news keeps coming — why does the market keep sidestepping it”
8. ”How do I tell which kind of rise I’m looking at”
9. The One Thing to Take Away

Hold fast to the mechanism of the illness; assign each sign to what governs it. Where something is present, seek it out; where nothing is present, seek it out still.

—— Huangdi Neijing, Suwen, “Zhizhen Yao Da Lun” (compiled before the Western Han; translation by the author)

On MacroMicro’s After Meeting EP. 217, published 11 October 2026, host Roger and US researcher Ralice took on an apparent contradiction: the US 10-year Treasury yield reached 5.3%, a 20-year high, while the S&P 500 and Nasdaq set record highs in the same week. Ralice reads the move as driven by real rates, which in turn reflect upgraded growth expectations — second-quarter GDP was revised from 1.5% to 2.2% in the final estimate, and consumption’s contribution to growth returned to 2.5 percentage points from 0.5 the prior quarter — with inflation expectations the steadiest of the three components. She was explicit about the condition attached: the term premium must not be the next leg up, because that component prices fiscal strain and doubts about a government’s ability to pay, and it has no ceiling.

2004 to 2006: 425 basis points of hikes, and the long end stayed put

Ralice mentioned she only noticed this chart after a colleague, Jason, shared it in a weekly meeting. Stretch the axis back to 2000: under Greenspan, the Fed hiked from 2004 through 2006 for a total of 425 basis points, and the 10-year yield barely moved. The front end rose, the long end sat still, and the textbook conclusion of that cycle was that long rates price long-run growth and inflation rather than the policy rate.

That comparison matters because it makes this year’s shape stand out.

Early 2026 to October: one hike, and the long end went from 4% to 5.3%

Through the recording date there has been a single hike this year, yet the 10-year moved from roughly 4% to above 5%, with the 30-year closing in on 5.7%. Same direction of policy, opposite behaviour at the long end. Ralice attributes it to structural shifts in bond supply and demand, in three layers.

On the demand-for-capital side, the large platforms are raising money for AI infrastructure, and the spending extends into energy, mining and defence. On the savings side, ageing lowers saving — retirees have no salary and spend down what they saved — and the return gap in favour of US equities means some money that would once have been allocated to Treasuries simply stays in stocks.

The third layer is the one I found most interesting, because it comes up least often: the holders have changed. The chart shown in the episode has net inflows into every US asset class positive since 2023, with total flows rising — the “end of US exceptionalism, money leaves America” story does not appear in the flow data. What did change is composition: private-sector holdings of foreign assets now exceed official reserves. A central bank buying Treasuries holds reserves, and the price is beside the point. A mutual fund or pension fund buying Treasuries wants a return, and is far more sensitive to the level of rates. Once the marginal holder is return-seeking, price swings get larger.

Early October in France: a budget that cannot pass, and a yield through 5%

The yield that rose harder than America’s that week was France’s. A little above 3% at the start of the year, it traded through 5% intraday on 2 October, the highest since 2002. The France-Germany spread cleared 150 basis points, the euro fell to its lowest since May 2025, and European equities were dragged along.

The cause is fiscal. In early October the government presented its 2027 budget, aiming to use roughly €54 billion of consolidation to cut the deficit from about 5.4% of GDP to 5%. The direction is sound; passage is the problem. No party or bloc holds a majority, the Socialists who helped the budget through last year have said they will not do so again, the right-wing National Rally opposes freezing pensions, and students are protesting for a larger education budget. After the 2024 snap election, the Barnier government fell to a no-confidence vote, the Bayrou government that followed fell the same way, and the current prime minister bought Socialist support by suspending pension reform — the 2026 budget still slipped past New Year before passing. The rating has walked down from AAA to a single A.

The phrase that comes to mind is “euro debt crisis,” and Ralice thinks that is far off — further off, in her view, than even the UK’s mini-budget episode of 2022. Her reasoning is specific. In 2022 the Bank of England announced balance-sheet reduction and the Treasury announced the largest fiscal stimulus in fifty years the following day: one foot on the brake, one on the accelerator, and gilts and sterling were sold hard enough that the central bank had to step in with emergency purchases. That conflict is absent now. Euro-area HICP came in at 3.8% year on year for September, which looks high, but the rising component is energy; core inflation is mild, the deposit facility rate already sits near core, and the European Central Bank has little need for a run of large hikes.

September to October in the US: payrolls of 29,000, read as resilience

The headline data looked soft. The latest payroll print was a gain of 29,000, below expectations, and unemployment ticked from 4.1% to 4.2%. The episode’s approach is to look one layer down: more than half of industries are still adding jobs, including education and health, leisure and hospitality, and manufacturing. The rise in unemployment came alongside participation moving from 61.6% to 61.8% with total employment higher than the prior month — the kind of increase that comes from workers returning.

Put that next to the GDP revision to 2.2% and consumption back at 2.5 percentage points, and consensus expectations for next year’s growth have been marked up toward 2.2% since the September meeting, converging on the Fed’s own projections.

The dispersion among officials remains wide, from Bowman seeing no urgent need for further hikes to Logan seeing room for another 50 basis points or more. What Ralice picked out of the remarks and the September statement was a shared word: timelier. Her reading is that the September turn toward tightening was aimed at bringing inflation back to 2% sooner, rather than a response to a sudden jump in upside inflation risk — the economy has been verified as sturdy enough to accommodate a more restrictive stance. She added that this is her own subjective reading, which I appreciated.

Where this stands: the term premium is the line to watch

The chart worth saving from this episode is the decomposition: real rates, inflation expectations, term premium. Roger pointed out they have walked through it not three times but thirty, and told listeners to bookmark it themselves.

Right now inflation expectations are the steadiest of the three, the move is led by real rates, and the rise in real rates traces back to upgraded growth expectations. The episode included a backtest run by their colleague Dylan: correlations between US sector returns, the manufacturing cycle index, and the 10-year yield. Cyclicals — information technology, financials, durables — correlate positively with the manufacturing cycle, which is intuitive. The point is that their correlation with the 10-year yield is also above zero. When expansion pushes yields up, cyclical returns can rise alongside them.

Then the warning. The term premium has turned up again and has edged out of its prior range. Ralice’s distinction: real rates have a ceiling, because once rates reach a certain level the market starts pricing recession and the Fed’s room narrows. The term premium prices geopolitical risk, fiscal strain and distrust of a government’s capacity to service debt, and that kind of fear has no upper bound. A world where the term premium leads, growth momentum fades, and rates stay high is a different environment for equities. She called that a risk to track rather than the base case.

What to watch next: 94%, and next year’s base effect

The inflation worry persists because the Israel-Iran conflict keeps oil elevated — WTI near $90 and Brent around the $100 mark at recording. This is the question Ralice gets most often from corporate audiences: if the fighting runs into next year, is oil stuck up here?

Her answer came with a number I did not expect. Using an energy research firm’s figures and January 2026 — before the war — as the baseline, Middle East crude export volumes recovered to 94% of pre-war levels by September. Roger’s reaction in the episode was “wasn’t the strait blockaded?” and mine was much the same. On reflection it fits: selling oil is the main source of income for producers, so even without a durable ceasefire, the other Gulf producers have the economic incentive to press Iran to let shipments continue, and an understanding gets reached below the surface. Roger’s term for it was a Schrödinger’s strait — I blockade my strait, you ship your oil. Iran’s own exports, as the primary target of sanctions, face something stricter; the other producers do not.

The second item is the base effect. This year’s high oil prices become next year’s high base, so inflation can slow without much of a fall in crude. Ralice puts peak energy-inflation pressure in the second half of this year through the first quarter of next, after which the base does the work.

As for the latest PCE print coming in below expectations, part of it reflects a statistical revision by the Bureau of Economic Analysis to three previously overstated components, which lowered core PCE by roughly 0.4%. Three- and six-month annualised rates and average monthly changes all point to a slowdown, with the monthly average near the 0.2% comfort zone.

“The bad news keeps coming — why does the market keep sidestepping it”

The episode’s title uses the image of a sidestep, and I think it captures a common frustration: long rates are high, oil is high, a war is ongoing, each headline worse than the last, and the market keeps slipping past them.

Treat headlines as input and prices as output and this has no solution. The episode’s method is to insert a layer in between: ask what is driving the number. A rising yield can mean “people expect the economy to do better” or “people are starting to doubt this government can pay.” The first can coexist with rising cyclical earnings; the second compresses valuation and confidence together. One number, two mechanisms, opposite treatments.

The mistake I made for years was using these series as thresholds — trim when the 10-year clears 5%, step back when crude clears $100. A threshold erases the mechanism and keeps only the symptom. Asking about the driver instead left me tracking more series and making fewer decisions.

”How do I tell which kind of rise I’m looking at”

The episode hands you the answer: watch the line nobody usually looks at.

Steady inflation expectations with rising real rates is a growth story. A rising term premium is a confidence story. France is the live example of the second — the same 2002-high headline, driven by a budget that cannot pass, a government that might fall, and a rating stepping down, which is why the euro and European equities fell together instead of stocks and bonds going separate ways as in the US.

What works for me is writing down in advance what I will watch, rather than deciding what a number means after it prints. The three tracking lines from this episode are all publicly available: whether the term premium keeps pushing out of its range, whether Middle East crude exports hold near 90% of pre-war volumes, and whether the average monthly core inflation print stays near 0.2%. The benefit of writing them down is that next month, when yields rise again, I don’t have to guess all over again.

The One Thing to Take Away

One idea: the same symptom can have opposite causes, and the treatment follows the cause rather than the symptom. A 5.3% yield can mean the economy improved or that confidence eroded, and the difference lives in the three components. Watching the symptom leaves you anxious in place; watching the cause tells you what to track next.

One exercise you can do today, which I have tried myself: pick something you run into repeatedly and handle the same way every time — a child’s grades slipping, a team falling behind, waking at 3am. Write down two mutually contradictory possible causes (“he isn’t trying” versus “he never understood the material”; “we’re short-staffed” versus “the target was wrong from the start”), and pair each with one sign that would only appear if that cause were the real one. For the next week, skip the treatment and go looking for the sign.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.