The Real Cost of Money
Last month we named two variables that could shape the next two quarters: the price of oil and the cost of long-term borrowing. This month the second one is the story. The 10-year Treasury yield rose from 3.97% in late February to 5.31% on October 5, passing its 2007 peak at the end of September to reach a level last seen in 2002. A move that size in the world’s benchmark rate reprices everything financed against it, from mortgages and corporate debt to the value investors place on future earnings. The oil shock looks like the obvious explanation. The data point elsewhere.
A long-term interest rate can rise for three reasons: investors expect more inflation over the life of the bond, they expect short-term rates net of inflation to run higher, or they demand more compensation for committing money for a decade. Strong investment demand, heavy government borrowing and supply shocks can press on all three at once. The headline yield says only that money got more expensive. Its composition shows which of the three moved.
Inflation-protected Treasuries separate the first reason from the other two. Between February 27 and October 5 the 10-year real yield rose from 1.72% to 2.95%, accounting for more than nine-tenths of the 1.34-percentage-point rise in the ordinary yield. Inflation compensation, the gap between ordinary and inflation-protected yields, added barely a tenth of a point, to 2.36%. The more recent stretch since the end of August looks the same, with almost all of a further half-point rise coming through real yields.
The consequences are already visible beyond the bond market. The average 30-year mortgage rate reached 7.3% at the end of September, its highest since November 2023, and every long-lived investment now has to clear a higher real hurdle. Composition tells us what moved. It does not settle what caused it.
Oil gave every reason to expect a different result. Brent traded near $71 a barrel before the conflict with Iran began in late February and closed March near $127. It fell back during a summer ceasefire, then climbed above $100 again in September with the Strait of Hormuz still largely shut. Headline inflation has felt it. CPI held at 3.4% from a year earlier in August, with energy doing most of the lifting, while core inflation stood at 2.4%.
The bond market has treated the shock as a problem for the next few years, not the next decade. Through March, the inflation investors expected over the next five years rose slightly, while expectations for the five years after that actually fell. By October 5 those longer-run expectations stood about a quarter of a point above their pre-war level: a real move, but far short of a broad repricing of inflation.
The shock has bitten through policy instead. On September 16 the Federal Reserve raised its target range to 3.75–4.00%, its first increase since 2023 and the first under Chair Kevin Warsh. Sixteen of eighteen policymakers expect at least one more hike this year, and the median projection for the end of 2027 rose to 4.1% from 3.6% in June. In market terms, a central bank prepared to hold rates higher for longer to keep an energy shock from becoming lasting inflation is a higher real rate. That channel has the most concrete support in the data.
With inflation expectations explaining so little, the rest of the rise has to come from the expected path of short-term rates or from the term premium, the extra return investors demand for holding a long-term bond instead of a series of short-term ones. The Federal Reserve publishes two widely used models that perform this split. Both decompose the ordinary yield rather than the inflation-protected one, so they offer a complementary view of the move rather than a direct explanation of the rise in real yields. From late February through October 2, both attribute roughly 1.2 percentage points to those two sources combined, and then part company. One, from the New York Fed, assigns about three quarters to expected policy. The other, from the Fed’s Board, splits it almost evenly, with the term premium slightly ahead.
The case for a rising term premium is easy to make. New York Fed President John Williams has described robust AI investment running into bottlenecks in the goods it needs, which raises the economy’s demand for capital. Treasury raised its third-quarter borrowing estimate by $68 billion, to $739 billion, although it kept the size of its regular bond sales unchanged. Both are plausible contributors, and neither can be measured precisely. When two of the Fed’s own models disagree this much, any confident attribution should be treated with caution.
Either way, the real cost of financing a long-lived project rose by more than a percentage point in seven months. Last month we argued that high expected returns on AI investment can absorb financing costs that would stop other projects. The hurdle those returns have to clear just moved up.
The kind of shock also decides what bonds do for a portfolio. In a demand slump, earnings and interest rates fall together, so Treasuries rally while stocks decline. In a supply shock, earnings weaken while inflation pressure builds, and both can lose at once. The energy shock of 2026 is the second kind.
Over the past five years, stock and long-term Treasury returns have tended to move together, with a correlation of +0.56. That looks strange only against recent memory. The negative correlation that made Treasuries the natural hedge for equities held from late 2001 to late 2022, and across the full record since 1925 the correlation has been below zero in only about 30% of months.
The protection numbers put a figure on it. In months when the S&P 500 fell between 2002 and 2019, long-term Treasuries gained an average of 2.1%. Since 2020 they have lost an average of 2.0% in those same months. Before 2002 the figure was close to zero: across eight decades and nearly 400 down months, Treasuries rose in about half of them and averaged a small loss. Bonds still lower a portfolio’s overall volatility, but the cushion investors came to rely on has not been there in this cycle.
Weaker protection is a plausible reason investors would demand more term premium to hold long bonds, though it is a backdrop to this selloff and not a measured cause. The portfolio lesson does not depend on settling that. When the dominant risk is a supply shock, diversification has to come partly from assets whose returns do not rely on falling interest rates, such as real assets and strategies that can follow sustained trends in either direction.
The RQA Economic Forecast Model rose to 0.44 in September, its fourth consecutive monthly gain and its highest reading since November 2024. Services activity, real incomes and consumer spending all improved, while stock market gains and manufacturing cooled from strong August readings.
More than half of September’s gain came from the yield curve, as long-term rates rose faster than short-term ones and steepened both curve measures the model tracks. A steeper curve has historically pointed to stronger growth, and the model reads it that way. This steepening accompanied a sharp rise in long-term real yields, so we would put less weight on it than on the improvement in services and incomes.
The September column carries August CPI, which held at 3.3% from a year earlier on our series (3.4% as published by the BLS; the difference reflects RQA’s seasonal adjustment) while core eased to 2.4%, so energy is doing most of the work in headline inflation. Core PCE reads 3.0% in the September column, down from 3.5% a month earlier, but most of that drop is revision. The September 30 release carried the government’s annual revision of GDP and income data, which lowered July’s figure to 3.0% as well, so the underlying rate held steady instead of falling half a point.
Output is where the economy looks strongest. The ISM services index posted its best reading since March, and manufacturing, though it eased from a strong August, remains comfortably in expansion. Residential permits held positive. Consumers kept spending, with retail sales and the consumer composite both firming, and real incomes jumped, though that jump partly reflects the annual revision and should be read with care.
Labor is the soft spot, and the latest news is weaker than the September column shows. The heatmap carries August payrolls, still slightly negative over twelve months. The September report, released October 2, added just 29,000 jobs, revised July to a loss and lifted unemployment to 4.2%, while wage growth slowed to its weakest since 2021. Part of the rise in unemployment came from people joining the labor force, which softens the signal, but hiring has lost momentum. Layoffs remain contained, with initial claims still well below year-ago levels.
Financial conditions reflect the spotlight. The gap between 10-year and 3-month yields is now the widest in the table, the direct imprint of long-term yields rising faster than short-term ones. The extra yield on corporate bonds remains lower than a year ago, and stocks are still up strongly over twelve months, though the S&P 500’s annual gain slowed as yields rose. Consumer sentiment remains deeply negative.
| Indicator | Sep-26 | Aug-26 | Jul-26 | Jun-26 | May-26 | Apr-26 | Mar-26 | Feb-26 | Jan-26 | Dec-25 | Nov-25 | Oct-25 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Labor | ||||||||||||
| Non-Farm Payrolls (YoY%) | -0.3 | -0.4 | -0.5 | -0.4 | -0.5 | -0.5 | -0.5 | -0.3 | 0.0 | 0.2 | 0.4 | 0.3 |
| Initial Unemployment Claims (Inverse YoY%) | 9.6 | 11.4 | 9.6 | 8.9 | 10.4 | 14.9 | 6.2 | 12.4 | -1.0 | 9.1 | -1.4 | -0.9 |
| Employment-to-Population Ratio (YoY%) | -0.8 | -1.2 | -1.2 | -0.8 | -1.5 | -1.2 | -1.0 | -0.5 | -0.5 | -0.3 | -0.5 | -1.0 |
| Average Weekly Hours Worked (YoY%) | 0.3 | 0.3 | 0.6 | 0.3 | 0.0 | 0.0 | 0.6 | 0.9 | 0.0 | 0.0 | 0.0 | 0.0 |
| RQA Labor Composite (YoY%) | 2.2 | 2.5 | 2.1 | 2.0 | 2.1 | 3.3 | 1.3 | 3.1 | -0.4 | 2.2 | -0.4 | -0.4 |
| Commercial Output | ||||||||||||
| ISM Manufacturing PMI (% over Base) | 9.2 | 11.2 | 6.6 | 8.0 | 5.4 | 5.4 | 4.8 | 5.2 | -4.2 | -4.0 | -2.4 | -2.2 |
| ISM Services PMI (% over Base) | 10.8 | 8.2 | 8.0 | 9.0 | 7.2 | 8.0 | 12.2 | 7.6 | 8.8 | 5.2 | 4.8 | 0.0 |
| Industrial Production Index (YoY%) | -0.8 | -1.0 | -1.3 | -0.9 | -1.3 | -2.0 | -1.6 | -1.1 | -0.8 | -0.2 | 1.6 | 1.2 |
| Residential Real Estate Permits (YoY%) | 6.2 | 6.6 | -2.1 | 1.4 | 2.1 | -7.4 | -5.5 | -2.4 | -4.8 | -12.8 | -7.3 | -8.1 |
| Income & Consumption | ||||||||||||
| Real Personal Incomes (ex. Transfer Receipts) (YoY%) | 1.9 | 0.4 | 0.7 | 0.2 | -0.3 | 1.1 | 2.0 | 1.1 | 1.3 | 1.9 | 2.1 | 1.8 |
| Retail Sales (YoY%) | 2.3 | 1.8 | 3.2 | 2.5 | 0.7 | -1.0 | -0.6 | -0.6 | -1.4 | -1.3 | -0.8 | -0.3 |
| Real Personal Consumption Expenditures (YoY%) | — | 1.8 | 2.1 | 1.8 | 1.2 | 1.2 | 1.7 | 1.7 | 1.1 | 1.2 | 1.5 | 1.6 |
| RQA Consumer Spending Composite (YoY%) | 2.3 | 1.8 | 2.6 | 2.1 | 1.0 | 0.1 | 0.5 | 0.5 | -0.1 | 0.0 | 0.4 | 0.7 |
| Financials & Sentiment | ||||||||||||
| Treasury Yield Curve Spread - 10-Yr Less 3-Month | 1.1 | 0.8 | 0.9 | 0.6 | 0.8 | 0.7 | 0.6 | 0.3 | 0.6 | 0.5 | 0.1 | 0.2 |
| Treasury Yield Curve Spread - 10-Yr Less 2-Yr | 0.4 | 0.4 | 0.5 | 0.3 | 0.5 | 0.5 | 0.5 | 0.6 | 0.7 | 0.7 | 0.6 | 0.5 |
| Corporate Bond Spreads (Inverse YoY%) | 32.3 | 33.8 | 30.4 | 28.0 | 16.4 | 12.5 | 15.0 | 12.9 | 1.7 | 6.2 | 10.3 | 16.2 |
| U.S. Monetary Base (YoY%) | -4.8 | -3.8 | -4.5 | -1.9 | -4.6 | -5.5 | -4.0 | -3.8 | -4.1 | -5.6 | -3.7 | -2.0 |
| S&P 500 Return (YoY%) | 14.4 | 19.0 | 18.1 | 20.9 | 28.2 | 29.4 | 16.3 | 15.5 | 14.9 | 16.4 | 13.5 | 19.9 |
| Consumer Sentiment (YoY%) | -12.7 | -11.2 | -10.5 | -18.5 | -14.2 | -4.6 | -6.5 | -12.5 | -21.3 | -28.5 | -29.0 | -24.0 |
| Inflation & Money Supply | ||||||||||||
| CPI (YoY%) | 3.3 | 3.3 | 3.4 | 4.2 | 3.8 | 3.3 | 2.4 | 2.4 | 2.6 | 2.7 | 2.8 | 3.1 |
| Core PCE (YoY%) | 3.0 | 3.5 | 3.4 | 3.6 | 3.5 | 3.4 | 2.9 | 2.9 | 2.8 | 2.6 | 2.5 | 2.8 |
| M2 Money Supply (YoY%) | 5.2 | 5.0 | 5.2 | 5.1 | 4.3 | 4.2 | 4.6 | 4.1 | 4.1 | 4.1 | 4.6 | 4.7 |
On levels, the economy remains where it has been all year: Quadrant II, the inflationary boom, with the growth model at 0.39 and CPI at 3.4% on the three-month basis the map uses. The direction-of-travel ribbon turned to Quadrant I for the first time since April, as growth kept rising and the three-month inflation average fell. Read literally, the data are drifting toward Goldilocks.
Two cautions keep us from leaning on that. Much of the drop in the inflation average is the June column’s 4.2% reading rolling out of the window, and oil climbed back above $100 during September, after the latest CPI reading in the exhibit. The bond market is pricing a different economy: a Fed that has started tightening, real yields at multi-decade highs, and Treasuries that cushion equity losses less than investors are used to. When the data drift toward Goldilocks while borrowing costs move the other way, we give more weight to the bond market.
For positioning, the regime argues for the same discipline as last month, with more weight on the rate side. Higher real yields raise the bar every investment has to clear, which favors businesses that fund themselves from cash flow and earn returns above a now-higher cost of capital. Long-term bonds are no longer the reliable offset to stock losses they were for two decades, so portfolios built on that assumption are worth revisiting. Energy remains the swing factor for inflation, and the Fed’s October 27–28 meeting is the next test of how far policy will lean against it.
The economy’s own data improved in September as growth broadened, while energy kept inflation elevated. Bond yields point the other way, with real yields behind nearly all of a 1.34-point rise in the 10-year, a Fed that has begun tightening, and Treasuries that no longer reliably offset equity losses. Quadrant II on the data, with the real cost of money now the variable to watch.
Disclosures. These materials have been prepared solely for informational and educational purposes and do not constitute investment advice or a recommendation to make or dispose of any investment or to engage in any particular investment strategy. Information and data shown were obtained from sources believed to be reliable, but accuracy is not guaranteed. All investments involve risk, including the potential loss of principal. Past performance is not indicative of future results. References to specific securities, funds and issuers are for illustrative purposes only and are not intended as recommendations. Third-party data and research, including material published by the Federal Reserve Board, the Federal Reserve Bank of New York, the U.S. Treasury, the Bureau of Labor Statistics, the Bureau of Economic Analysis, and the Mortgage Bankers Association, are cited for context and do not represent the views of RQA. Term-structure model decompositions are estimates and are presented as published. Richmond Quantitative Advisors, LLC is an SEC-registered investment adviser; registration does not imply a certain level of skill or training.