Stephen Miran says the Federal Reserve kept buying mortgage debt well into the recovery from COVID even after home prices had already risen 20%, and calls it "a very clear mistake."
Hatzius says the shift away from Fed transparency he has spent his career defending is a mistake. Kohn, a Fed policymaker for 40 years, is only half-convinced. Miran, who left the Fed's own Board of Governors himself in May, says the change doesn't go far enough.
"I disagree with that because markets price what they think the Fed will do, not what they think the Fed should do."
Kohn spent 40 years inside the Federal Reserve System, the last four as the Board's vice chairman with an office next to Kevin Warsh's during the 2008 financial crisis. Miran served on the Fed's own Board of Governors himself until resigning in May, while also chairing the White House's Council of Economic Advisers. Hatzius has been Goldman Sachs' chief economist since 2011, and his team has topped Institutional Investor's economist rankings every year since.
I listened to the full episode so you can skip it. 22 minutes of audio, 15 minutes of reading.
Here are the 9 takeaways that matter.
🎙️ Host: Allison Nathan, a senior strategist in Goldman Sachs Global Investment Research who created and edits the firm's Top of Mind report
👥 Also on: Jan Hatzius, Goldman Sachs' Chief Economist and Head of Goldman Sachs Research; Donald Kohn, a 40-year Federal Reserve veteran and its Vice Chairman from 2006 to 2010, now a Senior Fellow at Brookings; Stephen Miran, who resigned from the Fed's Board of Governors in May 2026 after also chairing the White House Council of Economic Advisers
📰 Published: 9 September 2026 · recorded August 2026
🔴 YouTube | 🟢 Spotify | 🟣 Apple Podcasts | ⏱️ 22 min | ✅ Time saved: 10 min
Key Takeaways
Forward guidance forced the Fed to keep buying mortgage debt while home prices were already up 20%, Miran says
He calls it "a very clear mistake" with no good economic justification
Miran ties Silicon Valley Bank's collapse to the same trust in Fed guidance
Its managers extended duration because the Fed had promised rates near zero indefinitely
Kohn says the FOMC's median dot is a weak signal that a single vote can move
Kohn calls the COVID-era pledge to hold rates near zero until full employment a real mistake
It implied accepting a real interest rate as low as minus 2% at full employment
Kohn's real complaint about Warsh is silence on the economy, not silence on rates
He wants a "narrative," the way Alan Greenspan always had one
Miran wants the policy dot gone but would keep the Fed's economic projections
Constructing the inflation data alone involves "10,000 methodological choices," he says
Hatzius rejects Warsh's claim that less transparency gives markets "unfiltered" information
Markets price what the Fed will do, not what it should — obscuring the reaction function just produces worse guesses
Kohn proposes a "golden mean": tell markets the story of what the Fed is watching, without a number attached
Kohn doubts Warsh's current silence can last, and points to the July press conference as evidence it already backfired
Short-term rates fell and long-term rates rose the same day
Miran says the tools built for the zero lower bound shouldn't be spent again until the Fed is back near zero
1. Two Kinds of Clarity
Nathan opened by asking Hatzius whether the shift toward a less transparent Fed under Warsh is a good or a bad development. He drew a line between two things people call transparency and answered them differently.
Being clear about the Fed's reaction function is good, full stop. "I do think that the transparency revolution in central banking over the last 40 years or so has been a good thing," he said, arguing that when markets understand what data will trigger what response, "you'll see faster monetary policy transmission" because financial conditions adjust before the next meeting even happens. "So, I think unwinding transparency about the reaction function, in my opinion, would be bad."
Forward guidance about the actual path of rates is a separate and more debatable thing. He distinguished Odyssean guidance — a hard commitment, named for Odysseus tying himself to the mast — from Delphic guidance, where the policy is simply conditional on the data. Odyssean guidance "doesn't make any sense under normal circumstances," in his view, though it can be defensible at the zero lower bound
He defended the dot plot as a mild version of the second, better kind. "The dot plot is not, in my opinion, strong forward guidance, not even strong Delphic forward guidance," he said, describing it as FOMC participants' own economic expectations rather than a promise — information about the reaction function, not a commitment to a path
2. The COVID Guidance Mistake
Kohn agreed the Fed needs to be clear about its reaction function, but said he has some sympathy for Warsh on forward guidance specifically, starting with how the Fed reads its own median dot.
The median forecast in the Fed's Summary of Economic Projections moves on one vote. "That is a very weak indication of where the committee is when just one person shifting can shift that median," he said, adding that he has watched it happen and the market react anyway. "So, it's crazy to focus on that median."
His sharper example is the pledge, made coming out of COVID, to hold rates at zero until the economy reached full employment almost regardless of inflation. "And I think just looking at the structure of that forward guidance, in what world is it okay to have at the highest a real rate of minus 2% at full employment?" he said. "So, I think that forward guidance was not well-conceived."
He said that guidance constrained the committee with real costs, and that forward guidance in general becomes less useful the further the Fed is from the zero lower bound — "I would be a reluctant user" away from it, though some guidance can still be necessary at the bound itself
3. Warsh Needs a Narrative
Kohn said Warsh's decision not to give forward guidance is fine on its own. His objection is to a different silence — Warsh's refusal to describe how the committee sees the economy at all.
"But he refuses to talk about how the committee sees the economy evolving. And I think that's not helpful." Kohn called explaining the committee's thinking part of what keeps markets stabilizing rather than destabilizing, and part of the Fed's own accountability
He invoked Warsh's own stated model. "Alan Greenspan, whom Chairman Warsh says he wants to imitate in many respects, always had a story," Kohn said — a narrative Greenspan could check against incoming data and revise when the data stopped lining up
"So, I think the story, the narrative is really, really important." Kohn said he hopes Warsh's own communications task force delivers that message
4. Guidance Breeds Blowups
Nathan turned to Miran, who took the opposite side from Hatzius and Kohn: less transparency is a clear positive, because forward guidance trades calm markets today for larger accidents later.
His framing is a straight trade. Forward guidance "dampened volatility in the short run," he said, "but the flip side of that is that it has increased it in the long run" — because telling markets what the Fed will do makes the Fed slower to change course when conditions do
His example is the Fed's own post-COVID mortgage buying. "Why was the Fed buying mortgages when home prices were up 20%? I don't think there's a really good economic reason for doing so," he said, attributing it to a calendar-based commitment the Fed had already made. "So, that was a very clear mistake that in my mind was a result of forward guidance."
He extended the same argument to Silicon Valley Bank. "Part of the reason they were terrible risk managers is because they listened to the Fed," he said. "The Fed said rates are going to be zero pretty much indefinitely." In his account, that promise is what let SVB extend the duration of its portfolio and take on more interest-rate risk than it otherwise would have
5. Clarity Becomes Guidance
Nathan pushed Miran on whether he was conflating two different things: a commitment to a rate path, and transparency about the reaction function itself. Miran said the two blur in practice, because inflation data is not a fact the way an oil price is.
"I can tell you the price of a barrel of oil because that exists. It's concrete." Inflation, by contrast, is built by statisticians. "And in constructing the general price level, there are 10,000 methodological choices that get made along the way," many with no objectively correct answer — which means too much clarity about the reaction function can make an ordinary data revision look like the Fed moving the goalposts
He does draw a line between the two kinds of guidance Nathan asked about, and cares much more about one of them. Telling markets what the Fed will do worries him far more than telling markets what it cares about
On the dots specifically: the policy dot should go, the economic dots can stay. "My view is the policy dot absolutely needs to go," he said. "Markets and private sector agents should not be taking the Fed's indication of where policy is going when creating their own expectations for the future." The economic dots, in his view, carry less downside and can remain, though the whole presentation could be narrower
6. Markets Price the Fed's Move
Nathan put Warsh's own argument to Hatzius directly: that less transparency lets markets give the Fed more "direct and unfiltered" information about the economy. Hatzius rejected the premise outright.
"I disagree with that because markets price what they think the Fed will do, not what they think the Fed should do. And that's not going to change." Obscuring the reaction function, in his account, just produces worse guesses about Fed behavior — not cleaner information about the economy
He distinguished productive volatility from wasted volatility. A genuine change in the outlook should move markets, he said, but "if you just have less clarity about the reaction function, then market pricing is going to jump around more for not necessarily very good reasons"
Obscuring the reaction function can also slow monetary transmission, he said, because the moves that should have happened immediately end up happening later instead
7. Kohn's Golden Mean
Asked directly whether more short-term volatility from less guidance is a good or bad trade, Kohn proposed a middle path, and Miran argued there is no version of a market signal that comes without volatility attached.
Kohn's answer was that a specific number, not information itself, is what damps the market's reaction to new data. "So, I think there's a golden mean here in which you give them some information," he said, describing it as the same narrative he had already argued for: enough for markets to work out the likely path of rates on their own, without the Fed naming a number
Asked whether the resulting volatility improves or confuses the market signal, Kohn said it improves it, within limits. "I think it could improve the signal. So, if it is damped and we take off the damper but give markets enough information to react intelligently to whatever is happening, then a little volatility is fine," he said — so long as it doesn't tip into open-ended uncertainty about where the economy or the policy is headed, which would damp investment instead
Miran said the choice isn't between volatility and no volatility — it's between volatility with a signal and no volatility with no signal. With forward guidance, "there's very little volatility and there's no signal in the front end. It just reflects the forward guidance." "The first best world of market signal without volatility doesn't exist," he said. "So, you can't let perfection be your standard."
8. The Quiet Fed's Limits
Nathan asked how durable a less transparent Fed regime can actually be. Hatzius and Kohn each doubted a different part of it.
Hatzius said the chairman's own quietness is durable, because it is entirely within his control. "He controls what he says at the press conference, for example. That's something that nobody can take away from him," Hatzius said. What he doubts is the rest of the institution staying quiet: "That is not going likely to be durable, because I think the Reserve Bank presidents, you're not going to be able to shut them down." Regional bank boards want their presidents talking, he said, so a more centralized Fed voice could be replaced by a louder, less coordinated one
Kohn doubts the chairman's own silence holds, and pointed to a specific press conference as the tell. "I'm guessing that he knows that the July press conference was not a good look for a Fed chair," he said, describing short-term rates falling and long-term rates rising the same day — a sign, in Kohn's reading, that markets were pricing in more inflation, uncertainty or risk premium than Warsh would have wanted
Kohn's confidence that Warsh will adjust rests partly on personally knowing him. "This is a very smart man. I worked very closely with him for nearly four years. Our offices were next to each other at the Fed. We worked hand-in-hand through the financial crisis," Kohn said, adding that he expects Warsh's own communications task force to help move him toward more explanation
9. Squandering the GFC Tools
Nathan asked Miran, the strongest advocate of the trade-off, whether there's a level of volatility so disruptive it would force a rethink. Miran said yes, but argued that point is nowhere close today.
He grouped forward guidance with quantitative easing and the formal inflation target as one set of tools, all built for one specific circumstance. "These are extraordinary tools for use at the zero lower bound when you are afraid of deflation risk that you can't head off," he said, and he would not have opposed using them in 2009 or 2010
By the middle of the last decade, in his view, they had outlived their purpose. "By the time you get to 2014/2015, are they needed anymore? Absolutely not. They all should have been repealed at that point completely," he said, making the same point about the Fed's continued mortgage-bond buying discussed earlier
His comparison was to antibiotics. "You take antibiotics when you're sick. If you take antibiotics when you're not sick, you just create super bugs," he said. "And their preciousness shouldn't be squandered."
His bar for bringing the tools back is a return to the zero lower bound itself. If volatility ever got bad enough to push the Fed funds rate back to zero, forward guidance might be appropriate again, he said. "But we're very, very far from that type of outcome."
Bonus Insights
Nathan's own framing of what changed under Warsh: shorter post-meeting statements, less forward guidance, and a smaller role for the Fed's economic projections — the specific changes all three guests were asked to weigh in on
This episode is drawn from Goldman Sachs Research's monthly Top of Mind report, in which Nathan interviews investors, policymakers and academics on one pressing market question each month
Kohn's view on forward guidance is conditional on where rates sit: some guidance can still be necessary at the zero lower bound itself, and only becomes unhelpful once the Fed has room to move rates normally
None of the three would call today's setup durable: Hatzius says obscuring the Fed's reaction function only makes markets guess worse, Kohn wants the committee to keep telling a story even while dropping the numeric promise attached to it, and Miran would drop both forms of guidance and accept the resulting volatility, on the view that the tools built for the zero lower bound shouldn't be spent again until the economy is back near it.
Products, Companies & Tools Mentioned
Federal Reserve (The institution at the center of the conversation — Chairman Kevin Warsh's shift toward shorter statements, less forward guidance and a smaller role for its economic projections is what all three guests were asked to weigh in on)
Summary of Economic Projections, the "dot plot" (Kohn wants the median forecast retired as a market signal; Miran wants the policy dot gone specifically while keeping the economic dots)
Silicon Valley Bank (Miran's example of "terrible risk managers" who extended duration because they trusted the Fed's promise that rates would stay near zero)
Books & Resources Mentioned
Assessing a Less Transparent Fed – Goldman Sachs Research's Top of Mind (The report this episode was built around)
If this was worth your time, send it to someone who has to have a view on this.
Get the latest market chatter as it happens: