# When the Rubber Meets the Road: Jurrien Timmer on Rates, Earnings, and 60/20/20 Source: Realpha Blog (blog.getrealpha.com) Original article and charts: https://blog.getrealpha.com/en/blog/compound-2026-10-09-this-is-where-the-rubber-meets-the-road-with-jurri/ > Listening notes on the October 9, 2026 episode of The Compound and Friends (#263). Fidelity's director of global macro Jurrien Timmer on how a 5.3% 10-year yield and 2.9% real rates compress valuations, why a falling P/E doesn't rule out risk, and how to think about allocation once the 60/40 premise breaks. Educational notes only — not investment advice, no stock picks, no price targets. Published: 2026-10-09 Locale: en Tags: interest rates, valuation, asset allocation, AI capex, treasuries ![A long hall inside an archive room, both walls covered floor to ceiling with oversized hand-drawn long-term market charts under museum lighting, one person standing deep down the hall pointing up at a chart](/covers/compound-2026-10-09-this-is-where-the-rubber-meets-the-road-with-jurri-cover.png) > The ruler is the boat; the common people are the water. The water carries the boat, and the water capsizes the boat. > > —— Xunzi, "Duke Ai" chapter (Warring States period; translation mine) On the October 9, 2026 episode of The Compound and Friends (#263), Fidelity's director of global macro Jurrien Timmer came back to make one argument: what is pressing on this market is the cost of capital. The numbers he put on the table were a 5.3% nominal yield on the 10-year Treasury and real rates near 2.9%, against a tech-sector P/E that has gone from 21 at the April 2025 launch point of this rally down to 20 — earnings running faster than price. His case rests on one condition: that earnings growth holds. If earnings stall, the same numbers tell a different story. ## Why won't he give a year-end index target? Because he doesn't think price is the thing worth forecasting. Timmer called year-end targets one of the silliest things Wall Street does. They make headlines and put people on record, but price is a byproduct of earnings and valuation. His own order of operations runs backwards from that: solve for earnings, then solve for what people will pay for those earnings — which depends on how long the earnings cycle lasts, the cost of capital, the risk premium, and how much of the earnings come back to shareholders through dividends and buybacks. Put those into a discounted cash flow model and price falls out as the residual, not the starting point. He was also candid that compliance doesn't let him make predictions, so his stock answer on television is that he has been in the business 41 years and that is the reason he doesn't answer questions like that. He grants his peers their constraint, though: the market rises about 67% of the time, so standing on the bullish side means being wrong less often, and being a bear is a hard living. What struck me is that the number we all want is exactly what he refuses to hand over. What he hands over instead is the arithmetic to compute it yourself. ## Valuations are falling — so why does this still feel like a bubble to people? Because the suspicious part of this cycle sits in earnings, not in the multiple. Timmer doesn't think this is a bubble, and his reason is that earnings have exploded enough that valuations are coming down: the multiple went from 21 to 20 over a year while prices rose. Semiconductor earnings are up 178% year over year, equipment suppliers can't keep up with demand, and some companies are talking about five-year backlogs. On a multiple basis this is nothing like the 70x tech sector of 2000. ![Two rising lines diverge from the same origin on the left, with earnings climbing more steeply than the share price; on the right, two bars of nearly equal height labelled 21x and 20x carry a downward arrow, showing that the denominator is growing faster than price, so the multiple falls.](/figures/earnings-outrun-price-en.svg) But the bears changed their argument, and he concedes this version has something to it: the charge is now over-earning. The semiconductor index has a clear 40-month earnings cycle, and 178% growth was never going to persist; capex also pulls demand forward. The real risk scenario is that one day these companies start missing, analysts scramble to ratchet estimates down, and 21x becomes 14x while the denominator is shrinking too. His own framing: he isn't worried about the earnings story so much as about too much capital chasing unknown returns on investment. He raised one angle I hadn't weighed. These cash cows used to return roughly 90 cents on the dollar of earnings through buybacks and dividends; now they spend that cash flow on capex. A purist would applaud — a company is supposed to invest. But a buyback is a known return and the ROI on a data center is an unknown one, and markets pay up for certainty. Same company, same cash flow, different use, and the multiple deserves to come down. As for whether earnings can be right and investors still lose, he reached for 1973–74 and the Nifty Fifty: those high-quality leaders didn't skip a beat on earnings through a 48% bear market and got annihilated anyway, because inflation kills valuations. At that extreme, the top 50 stocks traded at twice the P/E of the bottom 450. ![On the left, ten solid blocks stand for cash flow, and nine of them move by arrow to dashed blocks on the right; the left side is labelled a known return and the right side an unknown rate of return, with a downward arrow at the bottom marking the multiple the market pays.](/figures/buyback-to-capex-en.svg) ## The index looks fine — why does my portfolio feel awful? Because the damage is ordered by rate sensitivity, and the index is held up by the least sensitive names in it. The week of the recording, only 25% of index constituents were above their 50-day average and 46% above their 200-day. Timmer calls this a stealth correction and points to 1994 for the template: Greenspan doubled the cost of capital, the index went nowhere for a year with two 10% drawdowns, and underneath, more than two-thirds of constituents fell over 20%. Earnings were strong then and the market was cap-heavy, so the index hid it. The ending he remembers from that year is that once Greenspan took his heavy hand off the rate market, stocks exploded higher. ![A horizontal waterline runs across the chart with a nearly flat index line above it; below the line, two proportion bars are filled only to a quarter and to under a half, the rest left empty, showing that the index is fine while most of its members sit below their moving averages.](/figures/stealth-correction-waterline-en.svg) His description of the mechanism is the part I'll keep. It looks as though every portfolio manager independently decided to hold less rate-sensitive exposure and more long-duration tech, and the aggregate of all those separate decisions is the sector leaderboard we're staring at. The sectors filed under defensive — utilities, real estate, staples — are the year's worst, because they have been used as bond substitutes inside equity portfolios for fifteen years. When people get nervous now they buy Nvidia and Apple. Whether those are defensive is a separate argument; they are large, liquid, and their earnings are going to show up. One more chart stopped me: the top five companies now equal the combined value of the bottom 434 in the index. Run the same math in 2018 and it was the bottom 282. In eight years, most of the back half of the list got absorbed into five lines. ## Why won't yields come down? Because an entire class of buyers has left the bid while corporations compete for the same money. Timmer's first chart was the roughly $17 trillion increase in US federal debt since the pandemic, against an absence of non-economic buyers — the ones who buy regardless of the rate: reserve-currency central banks, large European banks whose risk budgets are policed by regulators. Central banks aren't buying, Japan and China are selling, and commercial banks aren't stepping in either. ![A balance tips to the left, its left pan piled with supply blocks of government and corporate issuance while the right pan looks empty because all three buyer entries are crossed out, and an upward arrow on the right is labelled yields.](/figures/supply-demand-tilt-en.svg) Short a buyer class, the supply side also added a competitor, which he calls reverse crowding out. The textbook version has governments crowding out corporates; this round has the hyperscalers borrowing about half a trillion dollars this year, with his team estimating another half trillion next year. Add corporate bonds and net equity issuance and roughly $3.6 trillion of new capital has been raised in twelve months. His read is that it's being absorbed so far, with nobody knowing when the market chokes on the fire hose. Two more sources get overlooked. One is sticky inflation out of genuine bottlenecks — he mentioned a colleague tracking oil tanker rates, up tenfold. The other is deliberate policy uncertainty: he reads Kevin Warsh as wanting Greenspan's opacity, which Timmer thinks is honorable, since a market led by the hand loses price discovery. The cost is that investors demand to be paid for uncertainty, and the term premium sits at 90 basis points. Higher yields are a byproduct of that choice. As for how close the danger is, his test is clean: compare the funding rate to nominal growth. France grows 2% nominally with yields at 4.5%, interest expense eats the budget, and that is debt unsustainability. US potential nominal growth runs around 6% against a 5.3% 10-year — not there, close enough that people have started doing the subtraction. ![Two small panels each show two horizontal levels: in France the funding cost sits above nominal growth and the lines have already crossed, while in the United States nominal growth is still slightly above the funding cost but the gap is very narrow.](/figures/growth-versus-funding-cost-en.svg) ## So what can I actually do with this? Start with something I got wrong for years: I treated diversification as a question of names being different. Timmer lays out the failure of 60/40 plainly. That mix delivered about a 9% annualized return from the late 1990s to the pandemic on the premise that the 40 was negatively correlated to the 60. Raise the cost of capital until the risk-free asset is competitive and the correlation flips positive — so bonds now cause the wobble in stocks instead of cushioning it. The alternative he has been pounding the table on for four years is 60/20/20, where the last 20 holds things that don't move with either of the first two. He named short-to-intermediate TIPS (duration around 2.5, real yield around 2.5), leveraged loans with no duration risk, gold, bitcoin, commodities, managed futures — while saying out loud that this is back-of-the-envelope work and nobody's advice. His risk/reward arithmetic on the 10-year is worth copying as a template. Yields fall 100 basis points: you collect a 5% coupon plus about 6% from duration. Yields rise 100: 5% coupon minus 6% in price, down 1%. Eleven to minus one. That isn't an opinion; it's a calculation you can rerun yourself on any bond. ![A zero line runs across the chart with a tall upward bar on the left labelled a 100 basis point fall in yields for a gain of 11, and a very short downward bar on the right labelled a 100 basis point rise for a loss of 1, the two sides wildly unequal in length.](/figures/asymmetric-bond-payoff-en.svg) The same episode supplies the counterweight. Gold's anchor has already changed once — before 2022 you could flip the real TIPS yield upside down and get the gold price, that broke, and he now explains it with global money supply. And when the Iran situation hit, gold and Treasuries sold off together, because to a Gulf state both are reserve assets you can liquidate when you need cash. Bitcoin has the same vulnerability; it can revert to a Nasdaq proxy in two seconds. That's the real question diversification asks: will these buckets be sold on the same day by the same people for the same reason. So the use I'd make of this episode is dumb and concrete. Lay out your holdings and write one line next to each: what happens to this if rates rise another 100 basis points. I did it once and five lines gave me the same answer. Five names, one position. ## Worth a look - The Compound and Friends #263 (October 9, 2026), with Jurrien Timmer - Timmer publishes the research he compiles each Sunday on LinkedIn, which links through to X; he also does a half-hour chart-driven show every Monday - Third-party research cited in the episode: 22V Research (hyperscaler bonds as a share of net new Treasury borrowing) and Duality Research (how tech and ex-tech indices interact) ## The One Thing to Take Away **One idea**: diversification is about whether your positions get sold on the same day for the same reason. Different names, different sectors, even different asset classes don't settle it. Gold and Treasuries are separated by a few thousand years of history and still fell together in one week because the same group of holders needed cash. The thread running through this whole episode — the cost of capital — is exactly the kind of reason that cuts across buckets you assumed were unrelated. **One thing you can do today**: write down your three fallbacks. Maybe it's this job, a side project, a friend who forwards you opportunities. Then ask each one the same question: if the thing that collapses is the same thing, do these still stand? When I did it, all three of mine hung on the health of one industry — the week that industry cools, three slots go to zero together. Noticing that doesn't require changing jobs. It requires knowing you hold three names and one position.