This issue’s syllabus · Issue No. 16 · August 16, 2026
Extra Credit Series · Fixed Income · Part 1 of 2Extra Credit Series: Fixed Income.
Assuming you already understand core relationship between price and yield learnt from the Foundation, we are going to dive a bit deeper.
We are going to see what goes into what each end of the curve is actually pricing, how much a bond’s price moves for a given change in yield, and what a real historical inversion looked like once you run the numbers.
There are three questions worth asking before you react to any yield headline:
A given change in yield doesn’t affect every bond equally. The longer the maturity, the harder a bond gets hit, or helped, by the same shift in rates. A bond maturing in two years barely moves. One maturing in twenty can move significantly with the same rate change. This is known as the duration. The intuition is that longer maturities mean more of the bond’s fixed payments sit far out in the future, locked in at a rate that stays stale for longer if rates move.
So before reacting to a headline that says bond prices fell, the useful question is which bonds, because a two year and a twenty year did not just have the same day.
The 2 year Treasury yield reflects the market’s expectation for the average path of the central bank’s policy rate over roughly the next two years, not the policy rate itself. Because it is pricing in expectations, it often moves well ahead of an actual rate decision, sometimes weeks or months before a central bank even meets.
The 10 year is different. It reflects the market’s longer run view of growth and inflation over the next decade, and it barely reacts to any single policy meeting. So when a headline says yields moved, the useful question is whether the market is repricing what a central bank will likely do soon, or its long run growth and inflation outlook is actually changing.
What’s historically preceded recessions is sustained inversion, often for months, with a lag anywhere from roughly six months to two years between inversion and any resulting downturn. Before treating one inverted reading as a signal it’s important to see if it is persistent.
For example, let’s imagine the 2Y Treasury yields 4.0%, and the 10Y yields 4.5%. The spread is +0.5 percentage points. That’s the ordinary shape, longer term debt paying a bit more to compensate investors for locking up money for longer.
Now suppose inflation picks up and the central bank signals it plans to raise its policy rate aggressively over the coming year. The market moves fast to price that in. The 2Y yield jumps to 5.0% but because the market now expects the average policy rate over the next two years to be close to that. Meanwhile the 10Y barely budges, maybe it will move to 4.3%, because the market believes those hikes will eventually slow growth and inflation back down, which keeps long run expectations comparatively anchored.
The spread is now -0.7 percentage points. The curve has inverted.
Short term yields rose sharply on repriced near term expectations. Long term yields stayed roughly where they were, or even eased slightly, on unchanged long run expectations. The two ends of the curve provide different expectations about the future at the same time, and the short end’s story briefly drowned out the long end’s.
| 2Y Treasury | 10Y Treasury | Spread | |
|---|---|---|---|
| Before | 4.0% | 4.5% | +0.5pp |
| After | 5.0% | 4.3% | −0.7pp (inverted) |
Starting in mid-2022, the 2s10s spread turned negative as the Fed raised its policy rate aggressively to fight inflation, while longer term yields rose far more slowly by comparison. The spread first inverted in July 2022 and stayed inverted for more than 700 straight days, reaching a peak closing inversion of -108 basis points in July 2023, the deepest since the early 1980s. That measure is the longest sustained inversion in the modern data.
Every prior sustained inversion in that dataset had eventually been followed by a recession, typically six months to two years later. However, this episode is genuinely unusual because that expected recession didn’t clearly show up within the normal window. Real GDP growth in the US came in at 2.9% in 2023 and topped 3% on an annualized basis in Q2 and Q3 2024, according to the Bureau of Economic Analysis, going against what the historical pattern would have suggested.
An inverted curve has historically been a strong signal, and the data behind that claim is real. But it’s not a guarantee. The lag between signal and outcome can run long and inconsistent, and this episode is a good example of why treating it as a countdown clock is a mistake. A few things likely contributed to the divergence. Many households and businesses had locked in low borrowing costs before the tightening cycle began, and government spending stayed unusually large well into the period.
| Detail | |
|---|---|
| First inverted | July 2022 |
| Duration of inversion | 700+ straight days |
| Peak closing inversion | −108 bps (July 2023) |
| US real GDP growth, 2023 | 2.9% |
| US annualized growth, Q2–Q3 2024 | Topped 3% |
A bond’s price and its yield will always sit on opposite ends of the same seesaw. The yield curve is that same relationship playing out across every maturity at once. The 2s10s spread is just the version of its people quote the most. When that spread inverts, the market is telling you it expects something different in the near term than it does further out. That gap has historically been worth paying attention to.