← All episodes · Forward Guidance
How To Trade The New Warsh Fed | Bob Sheehan
Overview
Bob Sheehan of Lighthouse Macro joins Felix on Forward Guidance to lay out his framework for trading the new Kevin Warsh-led Fed. Sheehan's central thesis is that the "Fed put" — the long-standing market assumption that the Fed will always step in to backstop falling risk assets — is dead under Warsh, who has deliberately stripped away forward guidance (Sheehan cites the post-meeting statement word count getting cut from roughly 340 words to around 170) and is no longer signaling a reflexive path to rate cuts during sell-offs. That shift, he argues, forces investors to rely much more on raw economic data and much less on Fed rhetoric, and it opens the door to significantly more volatility because market participants no longer share a common script for what the Fed will do next.
Sheehan also walks through his background — from a dual-mandate macro/equity portfolio role at Bank of America, through stints at Tran Macro, a data-science program at BrainStation, a hedge fund (Stratton/Stro Capital) doing discretionary global macro, and Ihor Dusaniwsky's short-data shop EquityLend — before launching Lighthouse Macro earlier this year. That background underpins his core investing philosophy: back every view with data and probabilities, and be explicit about what would prove the thesis wrong, rather than trading on narrative or headlines.
The bulk of the conversation covers how Sheehan is actually positioning around the new regime: a near-term "bear flattener" as the short end repriced higher on Fed uncertainty while the long end lags, followed by a longer-term steepening driven by Treasury supply, declining foreign buyer demand, and the government's fiscal "doom loop" of rising interest expense outpacing tax receipts. He also discusses rotating into defensive, shorter-duration equity sectors (healthcare, staples) versus longer-duration tech, and notes unusual simultaneous sell-offs in bitcoin and gold as evidence that historical correlations are breaking down in this new, less-guided macro regime.
Topics
Bob Sheehan's path to Lighthouse Macro
Sheehan describes a non-traditional route into macro research. He started at Bank of America on a dual-mandate portfolio team: one side ran a large-cap macro equity strategy (roughly $1.1-1.2 billion in assets) offered as an SMA to internal portfolio managers, where he became the team's de facto macro specialist setting sector weightings and top-down views; the other side managed institutional multi-asset portfolios spanning fixed income, commodities, and private investments for high-net-worth and institutional clients including pension funds. He frames this as unusual because most macro professionals come from a research-first background and only later attach a real book to their views, whereas he was forced from day one to back his thinking with an actual investment thesis and defend it to a team that gave him real trust early in his career.
After Bank of America, he moved to Tran Macro working alongside Frank Watrahan, then stepped away to attend BrainStation to formally study data science, wanting to combine his self-described 'data geek' tendencies with macro analysis. From there he joined macro hedge fund Stro Capital Management (found via Twitter/X by PM Mike Kbell) doing discretionary global macro with a quantitative bent, then worked at EquityLend, the premier short-sale data provider, applying a macro lens to short-interest data used by clients ranging from BlackRock to state pension funds like California's teachers' retirement system.
He founded Lighthouse Macro while still in school, initially just to share his views publicly; a well-received piece on Treasury auction data brought him early attention and podcast invitations. The firm was formally established in January 2026 (about five to six months before this recording), with the stated goal of bringing institutional-quality, data-backed research to a broader audience in more accessible language.
Data-driven macro framework: probabilities, not absolutes
Sheehan's core philosophy is to put numbers behind every macro claim rather than relying on narrative or 'reading the headlines and feeling a certain way,' which he considers a disservice to readers who want to genuinely understand the economics affecting their lives. His approach is explicitly probabilistic: he tries to show where the historical data suggests probabilities lie, while simultaneously being clear about what evidence would falsify his view.
He stresses humility about being wrong — even running real institutional money at Bank of America with skilled colleagues, the team was never right 100% of the time. He argues that most macro commentators who don't quantify their views can dodge accountability when wrong, whereas being explicit about the data-driven thresholds that would invalidate a thesis lets him be transparent with readers when he misses. This 'here's history, here's where it could break' framing is presented as the defining discipline of Lighthouse Macro's research.
The Fed put is dead under Warsh
Sheehan's central thesis, previewed at the very start of the episode, is that new Fed Chair Kevin Warsh has effectively killed the 'Fed put' — the market's long-held assumption that sufficiently large drops in risk assets will trigger a rescue via rate cuts and liquidity. That assumption held broadly true through the prior chair's tenure (and even under Yellen to a degree): every major sell-off over roughly the past decade was met with 'don't worry, the Fed's coming.'
Warsh, by contrast, has been vocal historically about hawkish views and is deliberately taking monetary policy in a different direction — not just on the Fed funds rate (the short end that most people watch) but also via balance sheet management, which Sheehan says turns what used to be a single unified 'Fed story' into two separate stories (a rates story and a balance-sheet/supply story). Sheehan concedes he could be wrong that the put is fully gone, but argues the degree to which Warsh has removed it is meaningful enough that it should change how investors think about Fed meetings and rate-cut expectations going forward — a genuinely new regime that younger traders in particular have never experienced.
Less forward guidance means more volatility and harder-to-parse language
Sheehan notes that Warsh has drastically shortened the language used in Fed communications — he cites doing research showing the words in Fed statements/guidance dropped from roughly 340 to about 170 (he caveats he may not be exactly precise), a roughly 50% cut. He frames this using a bowling-alley bumper metaphor: less guidance means fewer 'bumpers' steering market expectations down a predictable path, which mechanically opens the door to a much wider range of possible outcomes as more competing opinions and interpretations fill the vacuum.
He also points to voting-committee dynamics as evidence of rising uncertainty: Warsh himself abstained from submitting a dot-plot projection, and Sheehan notes a jump to 9 of 18 Fed officials shifting their projected number of hikes for September and December — a sign that the committee itself is far less unified than before. He argues this should make traders more cautious about placing large, one-sided bets, since without Fed guidance walking the market to a known outcome, being on the wrong side of consensus carries more risk (though being right against consensus could also pay off bigger). The overall effect, he says, is a much more dynamic macro landscape where parsing the Fed's now much shorter statements becomes both more difficult and more important.
Rethinking how to use economic data (NFP, CPI, PCE) without Fed guidance
Asked how the reduced guidance changes his use of hard economic data releases like NFP, CPI, and JOLTS, Sheehan explains that with the Fed no longer telegraphing its reaction function, he now puts less weight on trying to model what the Fed specifically thinks about any given data point, and instead evaluates data releases purely against long-run historical market reactions. He is clear he is not a short-term data-release trader (not positioning the day before a print to flip it the next day) but also not so long-term that data releases are irrelevant.
He frames this as a temporary state of affairs — since Warsh is only weeks into the role, the market simply hasn't yet built up a long enough track record of how he reacts to specific data to model his reaction function the way it could with prior chairs. For now, his approach is to treat each release on its own historical statistical merits (e.g., how markets have historically responded to a given core CPI or PCE surprise) rather than layering on assumptions about the Fed's likely response, since Warsh has signaled that data-in-the-moment, not forward guidance, should now drive expectations.
The two-trade curve framework: bear flattener now, steepener later
Sheehan lays out what he calls the most important market-reaction takeaway from the new regime: there are now two distinct trades operating on two different time horizons. In the near term (over the next month or so), he sees a 'bear flattener' playing out — the short end of the curve moving higher faster than the long end, driven by repricing of Fed-related uncertainty and recent inflation prints. This is the trade he says has already been visible over the preceding one to two weeks.
Separately, and over a longer horizon (months to a couple of quarters), he expects the long end to be driven primarily by a supply and term-premium story rather than direct inflation expectations: balance-sheet reduction, waning demand from foreign buyers (he pushes back on analyses claiming foreign buyer levels are steady, arguing the straightforward read of the data shows genuine foreign buyer drop-off), and fiscal deficits all point toward upward pressure on long-end yields as term premium reasserts itself. He describes the sequencing as: short end moves first (bear flattener), then the long end responds later (steepener) as the supply/term-premium dynamics take over — meaning the curve dynamic itself unfolds in stages rather than as a single simultaneous move.
Equity positioning: rotating defensive on duration grounds
Extending the curve framework into equities, Sheehan explains Lighthouse Macro's current book is positioned defensively, favoring healthcare and consumer staples over longer-duration growth/tech names. His logic treats equity sectors as having implicit 'duration': value and defensive stocks behave like shorter-duration assets, while high-growth tech names (further out on expected cash flows) behave like longer-duration assets that are more sensitive to the still-uncertain long end of the curve.
Given his view that the market is in a transitional, 'muddied water' period where the short end is volatile and the long end has yet to fully reprice to the supply story, he argues for favoring shorter-duration equity exposure until the curve dynamics settle. He notes this is separate from (though can compound with) the well-known AI/tech-driven rotation narrative already underway.
Breakdown of historical correlations: bitcoin and gold selling off together
Sheehan flags an unusual market signal as evidence that the new regime is scrambling historically reliable relationships: bitcoin and gold selling off simultaneously, with gold down roughly 3% on one recent day. Historically these assets have not moved in tandem in this way, and he says this kind of dislocation is striking even to market veterans, not just younger traders unfamiliar with pre-Fed-put regimes. He frames it as a live example of how removing Fed forward guidance is producing market behavior that doesn't fit prior playbooks, reinforcing his broader thesis that this is a genuinely new and harder-to-model macro environment.
Balancing rates/liquidity mechanics with balance-sheet and fiscal forces
Felix pushes Sheehan to reconcile his real-yield/curve framework with the 'other side' of the picture: reserve management purchases occurring over recent months, deregulation increasing how many Treasuries commercial banks can hold, and Warsh's historical preference for shrinking the balance sheet. Sheehan responds that this is genuinely one of the hardest parts of the current call, requiring investors to parse not just Warsh's public language but his known historical positions and the various 'task forces' he has stood up — which Sheehan characterizes as providing him political cover ('deniability') while he works out what he actually wants to do without committing publicly.
He reiterates his two-trades framework here: the short end is the easier, nearer-term call, while the long end requires separating pure Fed-rate mechanics from actual balance-sheet/money-supply changes, which he calls a 'natural force' that isn't just words — it's an actual change in the money supply and investable universe. He argues that historically Fed guidance was essentially a short-end-only story with the Treasury handling the long end, but under Warsh, balance-sheet actions and reserve management purchases now function as their own independent mechanism that has to be modeled separately rather than assumed to move in lockstep with rate policy.
The fiscal doom loop and why 'dollar to zero' calls are unrealistic
Bringing in the fiscal side, Sheehan references a piece he wrote four to five months earlier describing a 'fiscal doom loop': rising yields force the government to pay more interest, which requires issuing more debt, which pushes yields higher still, in a self-reinforcing cycle. He notes the U.S. is currently running the highest interest expense on its debt in a very long time (he says possibly ever), with government expenditures — driven partly by war-related spending but more broadly by overall outlays — outpacing tax receipts, feeding the loop.
He connects this directly to his long-end thesis: the fiscal doom loop is a real and underappreciated source of upward pressure on long-end yields via the supply channel. However, he explicitly rejects the more extreme bearish-dollar narrative that this dynamic inevitably leads to dollar collapse or debt monetization to zero, calling that an unrealistic, non-relative way to view the world. His view is that macro is fundamentally a relative game — Treasury demand, the dollar, and related assets should be judged relative to alternatives (e.g., China, other currencies) rather than against absolute doomsday benchmarks — and that people will keep buying Treasuries and the system will keep functioning even as pressure builds, up to a point where the drift is measured against everything else going on rather than treated as a standalone collapse signal.
Visuals
- Warsh removes forward guidanceFed statement language cut from ~340 to ~170 words; no dot plot from Warsh; Fed put no longer assumed.
- Short end repricing (near term, ~1 month)Bear flattener: short end yields rise on Fed uncertainty and recent inflation prints, outpacing the long end.
- Long end repricing (months to quarters)Steepener driven by Treasury supply, balance sheet reduction, declining foreign buyer demand, and term premium reasserting itself.
- Equity and cross-asset spilloverRotation toward defensive, shorter-duration equities (healthcare, staples); unusual joint sell-offs in bitcoin and gold as correlations break down.
Takeaways
- The Fed put is dead (or close to it): Warsh has intentionally removed the market's assumption that the Fed will rescue falling risk assets, a meaningful regime change from the prior decade.
- Fed statement language was cut roughly in half (about 340 words to about 170), and Warsh personally abstained from the dot plot — both signal a deliberate retreat from forward guidance.
- Expect a two-stage curve move: a near-term bear flattener (short end up first, over the next ~month) followed by a longer-term steepener (long end rising over months to quarters) driven by Treasury supply and fading foreign demand, not just inflation.
- With less Fed guidance, rely more on raw economic data (CPI, PCE, NFP, JOLTS) evaluated against historical market reactions, and less on trying to anticipate the Fed's reaction function.
- Lighthouse Macro is currently positioned defensively in equities — favoring healthcare and staples (shorter equity 'duration') over longer-duration tech — given curve uncertainty.
- Don't buy the 'dollar to zero' fiscal doom-loop narrative: the fiscal dynamics (high interest expense, rising issuance) are real and pressure long yields higher, but macro is a relative game, not an absolute collapse story.
Notable quotes
The Fed put is essentially the market's assumption that if risk assets fall hard enough, the Fed's going to step in... And I think Warsh has very much tried to signal that that's gone. That's not going to be back.— Bob Sheehan
There are two trades now, and this is the way I'm really trying to frame it in terms of the market reaction.— Bob Sheehan
If the Fed isn't going to guide you ahead, then... I'm going to be a little bit quicker to cut risk, or I'm going to maybe not overly lean on one bet hitting.— Bob Sheehan
We are issuing more to then kind of fix the problem that then we need to kind of issue more, and it's becoming this loop... I've even seen it called the fiscal doom loop.— Bob Sheehan
There's a lot in macro where people want to say absolutes, and it's like that's not the way the world works. Everything's relative.— Bob Sheehan