← Analysis page  ·  Jeffrey Gundlach hub  ·  Research hub

Actionable insights — Gundlach Unlocked: The Fed's Next Move

Not what he allocates to, but how he reads the bond market — the rate models, hedge tests, credit tells and valuation screens behind the calls, written so each can be re-run on next month's data.
2026-SEP-10 · DoubleLine — Gundlach Unlocked (episode 3) · Jeffrey Gundlach (DoubleLine Capital CEO) · ▶ Watch · full analysis · transcript
How to read this page: each insight is a reusable method — the indicator, the test, and the signal to monitor when you re-run it. The boxed line shows where it points right now. Timestamps deep-link into the video. (The webcast ended before its recommendations segment, so these are the diagnostics only.)

3:10 1. A big move that won't retrace is a continuation signal

The repeatable method
  1. Measure the size of the last major trend move in a yield (or price) series — here the 30-year from its 2020 low.
  2. Watch the consolidation that follows: a healthy trend normally gives back a meaningful fraction. A long sideways period with almost no retracement means the counter-trend side lacks buyers.
  3. Treat the failed retrace as evidence the next leg continues the prior trend, and position accordingly (avoid the duration it would hurt).
Now: the 30-year rose from 27bp to ~5.24% and "didn't retrace hardly at all" — so he expects the upward trend in long yields to continue (TLT negative).
Watch for

5:28 2. Map each Fed outcome to its long-end consequence before the meeting

The repeatable method
  1. Read the market-implied odds (Bloomberg WIRP from the short-end curve) and the 2-year-implied fair funds rate (2y minus funds gap).
  2. Form your own view of the chair — not the pricing — and assign low/high conviction.
  3. Pre-map both outcomes to the long end: if the Fed under-delivers versus what the 2-year demands, long yields rise (inflation credibility lost); if it delivers, the long end holds. Size duration exposure for the worse branch.
Now: WIRP ~60% hike; 2-year says funds ~50bp too low; Gundlach leans no hike (distrusts Warsh) → "long-term interest rates to rise fairly significantly."
Watch for

12:13 3. Test a "hedge" by charting its spread to the thing it hedges

The repeatable method
  1. Before buying an instrument as protection against another, plot both yields and — crucially — the difference between them.
  2. If the spread has been flat for years through the very move you fear, the "hedge" carries the same risk; it only protects against the component that actually varies (here, inflation expectations, not real rates).
  3. Split by maturity: where the implied component is mispriced (short-end breakevens pricing 2% immediately), the instrument is attractive; where it isn't, skip it.
Now: 30-year TIPS vs nominal spread "absolutely stable for the past 5 years" → Long-term TIPS are no hedge; Short-term TIPS are "too cheap" because breakevens assume 2% now.
Watch for

9:07 4. Split credit spreads by the hot sector to find supply indigestion

The repeatable method
  1. Chart IG and HY spreads for the market ex the crowded borrowing theme versus the theme only.
  2. A widening that is confined to the theme while the rest sits at its tights is issuance pressure, not a macro credit cycle — the market is charging more to absorb supply.
  3. Ask who the marginal buyer is (captive insurers, private-credit vehicles); opaque or forced buyers make the eventual repricing worse.
Now: AI-sector IG 50 → 125bp, HY 180 → 325bp; ex-AI flat/tight → AI corporate bonds negative.
Watch for

14:54 5. Average the unadjusted import and export price indices

The repeatable method
  1. Pull the year-over-year import and export price indices — no seasonal or hedonic adjustment, "just prices."
  2. Average the two for a clean read of underlying inflation and compare with PCE/CPI (and whether the 6-month annualized rate is above the 12-month).
  3. Cross-check with household-level pressures (electricity prices, oil and SPR/inventory levels) that explain weak sentiment.
Now: exports 8.25%, imports 5.95% → ~7%, vs PCE 3.7% and a Fed promising 2%.
Watch for

22:50 6. Translate CAPE into an expected 10-year real return

The repeatable method
  1. Plot starting Shiller CAPE against the subsequent 10-year real return (1965–2015) and fit the regression line.
  2. Read today's CAPE off the x-axis: note the range of outcomes historically observed at that level, not just the fitted point.
  3. Compare that expected real return with what bonds/loans now yield; when fixed income pays ~6–7% and equities imply negative real returns, cut cap-weighted equity exposure. Add a concentration check (largest sector weight vs 1999).
Now: CAPE 42 → never a positive 10-year real return, typically −5% to −9% a year; tech 38% of the index → he recommends none of the cap-weighted SPY.
Watch for

30:21 7. Overlay relative-performance lines on the dollar to pick the regional tilt

The repeatable method
  1. Chart S&P vs MSCI EM, and US corporates vs JPM EM local-currency debt, against the (inverted) trade-weighted broad dollar.
  2. If the shapes match, the regional bet is really a dollar bet — form the dollar view first (trend since the peak, stability, policy).
  3. Anchor with valuation (MSCI US vs ex-US price/book) and temper timing with seasonals (September–October weakness for risk assets).
Now: DXY 110 → below 100; P/B 5.72 vs 2.49; S&P −20% vs EM since end-2024 → EEM, EMLC, Non-US equities positive — but he's "coming close to home" near term.
Watch for

Methods distilled from the public YouTube video (transcript in transcript.txt) for personal study. Not investment advice. © DoubleLine Capital for source material.