The Federal Reserve raised its benchmark rate a quarter point to 3.75%–4% on 16 September, unanimously, and sixteen members of the committee put a second increase in their own forecasts for this year. The projections also push the return to 2% inflation out to 2029.
The committee's own statement calls that combination a timelier return to target. Nobody on the Bloomberg Surveillance desk could work out how both things are true at once.
"So I'm not sure what timelier means in this case."
Three outside economists took the decision apart in the half hour before Kevin Warsh's news conference: Rich Clarida, who sat in the Fed's vice chair's seat from 2018 to 2022; Diane Swonk, who has been forecasting the US economy for more than three decades; and Matt Luzzetti, who had already published a call for 75 basis points of tightening in total.
The full segment is covered here so you can skip it. 31 minutes of audio, 16 minutes of reading.
Here are the 11 arguments that matter.
👤 Guests: Rich Clarida, former Vice Chair of the Federal Reserve; Diane Swonk, Chief Economist at KPMG US; and Matt Luzzetti, Chief US Economist at Deutsche Bank
🎙️ Hosts: Jonathan Ferro, Lisa Abramowicz and Tom Keene, who anchor Bloomberg Surveillance
👥 Also on: Mike McKee, Bloomberg's economics and policy correspondent, reporting from the Fed in Washington
📰 Published: 16 September 2026 on the Bloomberg Surveillance feed
🟣 Apple Podcasts | ⏱️ 31 min | ✅ Time saved: 15 min
Key Takeaways
The Fed raised rates unanimously and pushed its own 2% target out two more years in the same document
Sixteen of the committee's members penciled in a second increase this year; only two would hold
Swonk thinks the rate the Fed just hiked to is neutral, which would make policy not restrictive at all
The committee's own long-run estimate moved only to 3.2% from 3.1%, and she says that is still too low
Clarida argues the PCE price index is overstating underlying inflation, so the Fed has less ground to cover than it looks
He also called the committee's 3.4% core PCE forecast for this year a little lofty
The anchor's objection to the whole afternoon is that the Fed cannot fix a supply shock
"This Federal Reserve can't print molecules"
Luzzetti's base case is that the Fed takes back all 75 basis points of last year's cuts
He reads eight dots showing the funds rate 75 basis points higher through the end of next year
A 25 basis point hike does nothing to the parts of the economy holding prices up
Luzzetti conceded AI investment will not slow because of it, and pointed to the dollar and equities instead
Swonk sees two labor markets running at once: entry-level shortages and an AI-driven slowdown in wage gains
Nobody on the desk thinks the Fed will accept a recession to reach 2%, and equities agree
1. The Decision in Numbers
Mike McKee had the statement and the projections in Washington and read them straight off the page before anyone could react to them.
"It is a rate increase, the first of a cycle, a unanimous decision to raise their benchmark rate a quarter point to three and three quarters to four percent." That is the first increase after the cuts the Fed delivered last year.
The dot plot is where the second hike lives. Sixteen members of the committee anticipate another increase this year and only two would hold, McKee said. Kevin Warsh, the chairman, does not submit a dot at all.
Eight members — almost half — see another increase next year, six call for no change, and four see rate cuts in 2027. One of those four has rates falling to 3.25%.
"Almost as significant, they raise their long-run view, essentially the neutral rate, to 3.2% from 3.1% in June." That is the rate the committee thinks neither speeds the economy up nor slows it down.
Growth is marked up for this year and next, to 2.3% and 2.4%, a tenth higher each than the June forecasts. Unemployment is forecast at 4.1% this year and held there through 2029.
Inflation is where the projections get awkward. Headline PCE is now seen at 3.7% this year against 3.6% in June, falling to 2.3% next year — and the 2% target gets pushed out two more years, to 2029. Core PCE is 3.4% at the end of 2026, up a tenth, and 2.5% next year, unchanged.
McKee finished on the language the rest of the half hour argued about: "The statement concludes, today's policy action will support a timelier return to the committee's 2% goal."
2. Timelier, and 2029
The desk fixed immediately on the one new word in the statement, and on the fact that the forecasts underneath it say the opposite.
Abramowicz's read: "The word that I think is going to be analyzed and overanalyzed will be timelier." The median projection, she noted, does not include Warsh himself and still carries an additional 25 basis point hike this year.
"All of this speaks to the idea that this is part of a cycle." She said the unanimity mattered more than the quarter point, and was a big tell given how many people got on board, the Fed chair included.
McKee's own verdict on the language was blunt: "So I'm not sure what timelier means in this case." He pointed out the committee pushed the 2% target out another two years in the same set of projections, as it almost always does.
The higher long-run neutral rate is the part he thinks matters. Officials have been discussing the possibility that spending on artificial intelligence lifts it, he said, and real rates have already moved that way.
Tom Keene's objection was to the labor forecast, which assumes nothing happens: "I'm absolutely fascinated by the labor call, which I know is off the radar right now, but they don't look for any change in the labor economy." His summary of the implied path was that "This is all going to get solved gloriously by inflation coming down, and output will be fine."
Jonathan Ferro ran the price action. The 10-year Treasury yield had breached 5% earlier in the week, a level he said had not been seen since 2007, and the S&P 500 was up 0.2% on the afternoon, off its session highs.
The first outside reaction to land in Ferro's inbox was Neil Dutta's at Renaissance Macro, and he read it out: "The Fed is not done. The odds are they are underestimating just how much work they need to do." Ferro added that this is why the two-year yield sits materially above the policy rate; Abramowicz put the gap at 100 basis points and asked which end closes it.
3. Not One and Done
Rich Clarida, who was the Fed's vice chair through the pandemic, was the first guest on and treated the dot plot as the story.
"You had 17 folks indicating, I think, another rate hike later this year, which had been our call that this would not be one and done." He counted Warsh among them even though the chairman files no dot.
Asked what changed between July and September to produce unanimity, he reached for precedent: "Well, the history is, Lisa, is that we typically don't see one and done." If the Fed is more than 100 basis points from its target, he said, a single quarter point does very little on its own.
He thinks the committee's own inflation forecast is too high: "I actually think the 3.4 percent on core PCE, which is what I think I heard Mike say, I think is a little lofty." He expects a lower print, partly because of methodology revisions.
The mechanism he offered for the July-to-September shift is that officials set a bar in public and the data missed it. John Williams had put out a 0.2% per month standard on core inflation, Clarida said, and the months since have not delivered it.
"I think Chris Waller said, give disinflation a chance, and they hadn't seen it." Ratifying that reaction function after Jackson Hole, he said, is what made the vote unanimous.
Keene pushed him on shocks — his own academic work is on dynamic stochastic general equilibrium models, and the stochastic is the shock term. Clarida said the committee is not split on whether shocks exist: "And so I think it's not so much shocks or no shocks. It's how persistent are they going to be?" His examples were memory chip prices a year out, where tariffs settle, and the shape of the oil futures curve.
4. Naive, Not Nirvana
Diane Swonk came on next and was the most hawkish voice of the three.
"Well, one of the things that I think is really important is this is the beginning of a rate hiking cycle." Two hikes are already signaled, she said, and the Fed does not want to get ahead of itself because it manages to economic aggregates.
"At the end of the day, though, their job is to derail inflation." If stickiness in services prices is still there in 2027, she said, the committee will have to go further and choose between derailing growth and hoping for a softer landing.
The retail sales figures released the same morning are her evidence that demand is not cooling: "At the end of the day, we had the retail sales today, and they were stunningly strong and broad-based."
Part of that strength is people buying ahead of tariffs, which she says is the problem rather than a mitigating detail. As she put it, "Some of those retail sales reflect actually buying ahead of feared increases in tariffs in the vehicle sector, but they weren't all that. And that is, of course, the very behavior that the Fed is tasked to avert."
Keene asked whether inflation can come down without real GDP coming down with it — the outcome he called nirvana. Swonk's answer was one line: "I think it's more naive than nirvana. And I am worried."
Her evidence is other economists' pipelines, not her own model. The industry economists she talks to are worried, she said, about cost pressures in the pipeline, and about how much more sustained those pressures will be.
Abramowicz noted the statement's passing reference to geopolitical developments and asked whether the Fed is content to look through an oil shock as long as the rest of the data is strong. Swonk's answer was that this is now a combination of demand and supply shocks, which is why more than one hike sits in the forecast.
5. Two Labor Markets at Once
Ferro asked Swonk whether the labor market is tight enough to generate the wage growth that would support higher prices, and whether consumers will tolerate them.
"In terms of the labor market right now, I'm seeing two labor markets at the same time."
At the bottom: "There's pockets of labor shortages where wages are starting to pick up." She said a bank chief executive told her the day before that his clients cannot find workers at entry level — not new graduates, but positions she said would once have been filled by foreign-born workers and now go unfilled.
At the top, the effect runs the other way and it is about pay, not headcount: "We're not seeing job losses per se from AI, but we are seeing a slowdown in wage gains in those sectors most exposed." She attributed that to economic research she had just seen.
Her point about what a central bank can and cannot do closes the section. As she put it, "And the bottom line is, again, they can only deal with the economic aggregates. They can't deal with the inequalities and the unevenness of this expansion, which has gotten extremely concentrated in the AI build-out as well."
6. Neutral Is Still Too Low
Abramowicz read out a message from Steve Chiavarone, Chief Investment Officer at Federated Hermes, and Swonk used it to disagree with the Fed's own estimate.
Chiavarone's line, as read on air: "If Warsh characterizes this as a recalibration to higher neutral rate, reflecting higher nominal growth, he can land the plane."
Swonk's response was that the committee's upgraded neutral rate is still wrong. She expected the move up, she said, but "It's still too low."
"I actually think this is neutral, what we're at right now. And that means we're way too accommodative at the moment." On her reading, a rate of 3.75% to 4% is not restrictive, which is a very different starting point from the one in the summary of economic projections.
Keene asked whether she would miss the dot plot if the Fed scrapped it. Swonk said the statement had already given the game away: "They give us an inclination that they're in a rate hiking cycle." The verbiage in the statement itself, she said, established the cycle without the dots.
On how far this goes: "As I agree with Rich, we actually have two more additional rate hikes and it could be more than that."
Ferro's counterpoint after she left is that the dots were doing real work today. Without them there would be no implied path at all, he said, and the chairman is not expected to supply one verbally at the press conference.
7. The Fed Can't Print Crude
The strongest argument of the half hour came from the host, not a guest, and it is about what a rate rise can physically do.
Ferro's objection is that the inflation the Fed is responding to is a supply problem: "This Federal Reserve can't print molecules. It can't print barrels of crude. It can't build refineries. It can't sign peace accords."
He was explicit that the target has been missed for years regardless, but said the energy shock is what turned the conversation into one about hikes. Without the war and the move in energy prices, he said, neither the European Central Bank nor the Fed would be discussing additional increases.
"The evidence of the last five years, though, Tom, is it's tremendously difficult to get inflation back to target with nominal GDP this high and with the crude story in the mix, too."
Abramowicz's read of the market's answer to whether the Fed will accept a recession to finish the job: the equity market says no. "Based on the fact that equities are up, I would say the answer to that right now is perceived to be no, that ultimately they will stop short of engineering a full-blown recession because of the reluctance to do so." That, she said, is why nobody is penciling in four, five or six hikes.
Keene cited a research note that separates Waller and Warsh on how much weight to give shocks, and complained that an institution like the Fed will not hold that discussion in public on decision day.
8. PCE Overstates Inflation
Clarida came back for a final word, and Ferro put the central question to him directly: can inflation get back to target without destroying demand?
His answer was yes, and the precedent he cited is recent: "We saw an example of disinflation without a recession between 2022 and 2024 and 5."
His reason is a measurement argument rather than a demand one: "I think, to be blunt, I think the PCE price index is overstating underlying inflation in the economy." Look at the labor market and at the consumer price index, he said, and inflation is above target but not as far above it as PCE shows.
"I think they have less ground to cover than maybe some of the indicators suggest." He said some of the gap may be revised away, as it has been before.
Asked whether the chairman could make that argument without damaging the Fed's credibility, Clarida said he would be interested to see whether Warsh tries. He chose not to at Jackson Hole.
Abramowicz's own counter-evidence was core CPI, which strips out energy and food: the monthly increase was 0% in June, 0.2% in July and 0.3% in August. That pattern is broad-based, she said, and it is why people read the Fed as acting.
9. Deutsche Bank's 75bps
Matt Luzzetti had written before the meeting that it was not clear the Fed is sufficiently restrictive, and that Deutsche Bank now expects 75 basis points of tightening in total. Ferro asked whether the other 50 had been validated.
His read of the dots was more hawkish than he expected: "16 out of the 18 dots expect at least two rate hikes this year."
The committee is close to evenly split on whether next year brings a second or a third, he said: "I think that there's eight dots that show the Fed funds rate 75 basis points above yesterday's levels through the end of next year."
"So I think it's very much in line with our view at this point, which is that the Fed has started a mild tightening cycle and that it's a pretty strong base case that the Fed is likely to take back the 75 basis points of reductions that they gave us last year."
On what the Fed is trying to achieve, his answer was growth as much as supply shocks. "You have a very strong growth backdrop as well. As we saw with retail sales this morning, the Atlanta Fed GDP tracker is 5.1% annualized. Consumer spending is expected to grow above 4%."
He was candid about the limit: "Now, will they get all the way back to target without significantly reducing demand? I guess I have some skepticism about that, but can they at least get closer?" Getting core inflation from around 3% to nearer 2.5% is the realistic objective, plus buying back credibility so inflation expectations do not move.
Ferro pressed him on why the 2022-to-2024 precedent does not transfer: that disinflation came with a large positive supply shock to labor from immigration, and the US now has negative supply shocks in labor, goods and energy at once. Luzzetti agreed the comparison breaks: "we did see this immaculate disinflation taking place" then, with the Fed tightening aggressively into positive supply-side forces, and "It's very different right now."
"And I think we just now see that the Fed's patience has run out and they think they need to act. I think that's the right move."
10. What a 25bp Hike Buys
Ferro's sharpest challenge was that the parts of the economy holding prices up are the parts a policy rate does not reach.
He put it as a contradiction: the rate-sensitive parts of the economy are already struggling, and the rate-insensitive parts are what is driving prices. So what does this hike actually achieve?
Luzzetti conceded the premise on both counts. Housing is the clear example of a sector where financial conditions are already tight, and "AI-related investment is not going to come down simply because the Fed hiked rates by 25 basis points today."
His answer is that the transmission runs through asset prices rather than through borrowing costs: "It can help to lift the dollar. It can help bring the equity markets down a bit." Wider credit spreads are the third channel he named.
"I think that there's just very strong signals that monetary policy is not restrictive, is not sufficiently restrictive."
Abramowicz was unconvinced that the channel exists at the moment. Ferro's evidence against it: front-end yields are up 100 basis points year to date, equities are higher, and credit spreads have not moved even against heavy new issuance.
"I'm glad you really went hard on that because ultimately a question here is what exactly is 25 basis points going to do, given the fact that we already have had that baked in? What will 75 basis points do? What will 100 basis points do?" Her question is whether the only real transmission mechanism left is the signal itself.
11. 25bps Before an Election
The desk closed on politics and on what a quarter point is for.
Keene's view is that the move is too small to matter economically and large enough to matter politically: "The fact is, 25 basis points is nothing except for the guy at 1600 Pennsylvania Avenue."
He also objected to the language everyone was using. "How do you get inflation down? And you keep saying demand destruction. It's fancy talk, OK? It's just slower economy. Who wants a slower economy into an election?"
Ferro's answer on White House influence was that the Fed has just raised rates by 25 basis points and signaled it is willing to go again.
Ferro also noted that the pain argument is not hypothetical, even with a 4% unemployment rate: "There are people in pain in this economy. I certainly don't want to understate that." But at the aggregate level it does not show up in unemployment or in GDP, which is why the willingness question stays open.
Most people who come on the program, he said, expect the Fed will not go again in October, immediately before the election.
Bonus Insights
One of the anchors read the unanimity as discipline rather than agreement: "Stay out of trouble. Don't dissent." followed by "Keep your head down. Get to the next meeting and then keep your head down for the elections."
Abramowicz read out a note from Kevin Gordon of Charles Schwab that framed the whole problem: the inflation data may be driven by idiosyncrasies at times, but the central bank works in aggregates — "They have a blunt tool. It's their only tool. It is not clean. It potentially will take some casualties."
The Fed has reordered its press room alphabetically. McKee told the desk he is now on the second row and a Wall Street Journal reporter is at the back; Ferro said he had no idea what it means for the order of questions.
When Keene described the Fed's implied path as inflation falling with output unharmed, an unnamed voice on the desk cut in with "Says who? And that's, to me, the huge mystery here."
The bottom line from the desk is that the Fed has started a cycle it may not be able to finish: the hike is real, a second one is in the forecasts, and every economist on the program doubts that a policy rate can reach an inflation problem being driven by energy, tariffs and an AI capital-spending boom.
Listen to the full episode:
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: