One Fed Hike Isn't The Mistake, Five Could Break The Economy

00:00 The setup: what changed in two months

• The 10-year breached 5% intraday for the first time since 2023; oil over $100, diesel over $6, August core CPI +0.3%, the 30-year at 5.31%, and better than 90% odds of a Wednesday hike. Two months ago Rosenberg said the next move was a cut.

01:21 162,000 jobs: "a lot of hair in that number"

• A figure with a "100% chance of getting revised" after a year of downward revisions, flattered by back-to-school seasonals in state and local education and World Cup effects in leisure/hospitality. "If you're trading on faulty data, my heart goes out to you" — and a Fed responding to it is "on the precipice of a policy misstep."

02:31 Mapping industry data into the CPI: core "close to being flat"

• Hotel/motel industry rates were negative for early August, yet CPI lodging jumped; CPI telecom services rose the most on record while PPI telecom was negative; the Manheim index had used cars down while CPI had them +0.4.
• "That's data mining. No, I actually refer to it as data analysis." Not a controversy — just "a little spurious… not something as a central banker that I would be raising interest rates on."

03:32 Esther George, a historic hawk, would sit it out

• The former Kansas City Fed president "said if she was going to vote… she would probably just stay on the sidelines."

03:58 Why he flipped: Warsh changed his modus operandi

• Warsh arrived "more dovish than Jay Powell," which is "how he got the job," then sounded hawkish at his press conference and Jackson Hole, with three dissents for a hike already on the record.
• The cut call assumed a fragile Iran truce and well-behaved oil; the truce failed. "I had no choice because it looked like the Fed chairman changed his mind at the same time." The hawks want to fix "five years with core inflation above target" — "I don't know what raising rates today is going to do to change the rearview mirror."

05:56 Core at ~2.5% with no labor-market pass-through

• "Core inflation has a two handle," the trend flat; oil is range-trading at the top of the range, not a new high.
• 2022–23 had an 18-month wage-price spiral, not the 1970s' 10-year one (unionized workforce, COLA clauses). Today's inflation is oil and its "first cousins" (airfares, delivery) with "no real sign of any broad-based pass through."

07:44 Warsh's breadth test, applied to August CPI

• Warsh introduced a breadth measure on the PCE deflator at Jackson Hole; run on August CPI, 45% of hundreds of subcomponents were flat or negative vs a historical norm just over 40% — "the diffusion is actually improving."
• Shelter's strength defied falling home prices in the industry data.

09:10 The one variable: nominal wages are decelerating

• If labor were tight and second-round effects real, nominal wage growth would be accelerating (as in 2021–23). It is decelerating. "Labor is the biggest cost… that's what's gone missing in this particular inflation narrative."
• Price shock + cooling labor market = negative real wages (six months running), negative real consumer spending and a profit-margin squeeze: "it is a tax hike on the private sector. This is not a source of inflation."
• A Wednesday hike is "an exercise in flexing its anti-inflation muscles"; markets already price a 4.5% funds rate next year, above the most hawkish June dot (under 4%).

12:36 No monetary inflation; full employment may be 3.5%

• Money supply and velocity show no monetary inflation; growth is running 1.5–2%, below potential.
• Don't read tightness off the 4.1% U3 rate — read the price of labor. "For all we know the full employment unemployment in the United States is close to 3 and a half%." Warsh "gave no lip service to nominal wage growth… no lip service to the housing market."

14:28 History: the Fed used to look through oil

• Greenspan cut five times when oil doubled in Gulf War I; Bernanke did not hike as oil went $50 → $150 in 2008. "What is the interest rate peak that's going to combat oil prices?" Only a destabilizing recession.

15:36 Global knee-jerk: "like eating potato chips"

• Canada's underlying inflation is ~2% with decelerating wages, yet as much or more tightening is priced after a hawkish-sounding Tiff Macklem; ECB, BoE and RBA hiking too — "a global monetary tightening cycle."
• Markets front-ran cuts and now front-run hikes: two cuts priced in late February (10-year under 4%) → five hikes now. "I think it's way overdone."

17:13 The ECB's July-2008 mistake — and a divided Fed

• Asked if Warsh is repeating the ECB's 2008 hike into the crisis: "Yes. Yes, I do."
• Communication has been "bungled"; at least nine FOMC members are comfortable standing pat. What if the hike comes with two dissents, "and what if by the way Powell is one of those dissents?"
• "I don't think one rate hike will be a policy mistake, but if they raise their dots and they ratify what the market's priced in… four or five, I don't think the economy can withstand it."

19:31 AI build-out needs cheap money and energy

• The leveraged AI boom needs low financing costs; with push-back on the build-out, receding capex means the debt binge recedes, "and that might take some of the pressure off the general level of interest rates too."

20:24 5% on the 10-year: block your nose and buy

• Last time was October 2023, when "the economy was a lot stronger… and inflation was higher." "Really juicy yields… and nobody seems to want to own it."

21:07 Catalysts: Nov-3 midterms, then Nov-4 refunding

• Data should slow "precipitously." Democrats likely take the House, the Senate a toss-up → fiscal gridlock after six years of 5%-plus-of-GDP deficits: it "is going to cut into aggregate demand no matter what."
• The next day, November 4, is the Treasury refunding announcement.

22:32 The Treasury's lever is supply, not demand

• The Fed controls demand (QE, balance sheet); the Treasury controls supply and where on the curve it issues. At the October 2023 refunding it flooded the system with bills and cut long issuance — the 10-year fell 100bp in the last three months of 2023: "If you were long duration… you made your year."
• Bessent's buybacks were mainly a signal — criticized by Druckenmiller as too small; "you don't have to go back to Operation Twist in the early 60s."

25:18 The Fed doesn't control the 10-year

• Policy spillover reaches perhaps 2–3 year Treasuries; beyond that, issuance mix shapes the curve. 2023's "huge bull flattener" came with deficits still huge — purely from where the Treasury financed them. He expects Bessent to pivot the same way.

26:36 AI capex cuts also cut corporate bond supply

• AI executives signalling slower capex hits demand growth and corporate debt issuance — a setup for "certainly a partial reversal of this runup in yields," oil permitting.
• Inflation expectations aren't out of control; what changed is the Fed's reaction function. Bonds lead stocks: rising real yields are now compressing P/E multiples.

28:20 The fiscal juice ends; the yield cushion is 500bp

• "People will wake up on November the 4th also realizing that the fiscal goodies are over" — fiscal juice has supported the economy and profits.
• In 2021 the cushion was 60bp; "today you're 500 basis points." No sure thing, don't put all eggs in one basket.

29:05 Bonds beat stocks next; the short-covering rout

• ROSY's bond exposure "has not been a very good trade the past few months," but he expects bonds to outperform stocks, then stocks to follow with a lag.
• The net spec short at the Board of Trade is "almost at a record high": a 20bp move to 4.80% "is going to freak out all the shorts… Next thing you know, we'll be at four and a half" within 1–3 months.

30:46 Gold is an investment, not a trade

• Gold broke below $4,300 (low $4,253) on the dollar's best day since June. Rosenberg: bullish since $1,000 in 2010 at Gluskin Sheff; "12 or 15 corrections" in 25 years.
• A triple bottom at $4,000 despite the strong dollar and real-rate boom — "why didn't it go even lower?" Central banks keep buying; his view changes only when they stop.

32:21 Real yields at a TIPS-era record, likelier to fall

• The ~3% real yield on the long bond has "never been this high… in the lifetime of the TIPS market." Risk-managed, "they're more likely to fall than they are to go up."

32:59 Fundamentally bearish the dollar

• Policy "an absolute mess," gridlock, impeachment proceedings if Democrats take both houses, "$5,000 dividend checks." "The start of the sunset of the US economic hegemony"; a long-term bear market regardless of 2028.
• ROSY is "70% inversely correlated to the US dollar" — "if you're bullish on the US dollar… don't be interested in my portfolio."

35:04 Inside ROSY: a barbell of cheap bonds and hard assets

• Equities mostly emerging markets and Japan; local-currency EM debt; Treasuries (2s and 10s).
• Disinflation in services and housing, inflation in goods → "very long basic materials": base metals, gold bullion (no miners), rare earths, power and energy infrastructure, pipelines, aerospace/defense, and oil as a geopolitical hedge.
• Commodity supply deficits make this less cyclical than it looks; no hotels, restaurants, airlines or real-estate services — cyclical services "bear the brunt."

37:38 If he's wrong

• If the Fed "goes bonkers raising rates," the Treasury leg won't do well, the dollar rises and basic materials suffer. "Your assumptions drive your conclusions… if you don't have a view, you don't have a portfolio. Just move to cash."

39:18 From model portfolio to listed ETF

• A "Eureka moment" at the end of 2022: a fund of ETFs expressing Rosenberg Research's themes for 2,300 clients in 40 countries across currencies, commodities, fixed income and equities.
• Since Feb 2023: low-to-moderate risk, "a passive strategy that's been run actively," ~0.4 beta to the S&P, ~0.7 to a 60/40, up 60% in 3.5 years — with one unit holder (him) until clients mirrored it and pushed for a listing. ROSY listed in Toronto last week with Corton Capital as manager.

42:11 Why Rosenberg Research, why Canada

• After 12 years at Gluskin Sheff, "everything sounded the same" on the sell side; he started the firm at 60 in early 2020 with 1,000 clients.
• Canada lacks a research-based, truly diversified, global vehicle ("60/40 is not diversified"); "home bias… is the most difficult bias to break"; ~45–50% of global equity sits outside the S&P 500, away from "AI concentration risk." "I don't tell people what to do. I just tell them what I'm doing."

45:02 Not a trading vehicle — thesis shelf life

• Built for "a market cycle view," rebalanced; "the only way you make money, by the way, is booking profits."
• When a theme is fully priced but the thesis unchanged, exit and rotate to a "plan B" top-conviction idea — e.g. "uranium… India… Canadian bonds."

47:33 Jarvis: trading fund vs investment fund

• A PM must choose: short-term dislocations or theses that play out over time; the best focus on theses. His job is to "pare it down" to the core themes that translate into an investable ETF.

49:24 Horizon: 6–12 months minimum; the dollar thesis is 5–10 years

• A thesis realized in three months gets its profits taken; others run for years. The core one — dollar devaluation — he sees over 5–10 years, as governments and pension plans decouple from US exposure.
• Judge the fund over at least 6–12 months; a day's dollar pop "to me that's noise."