Title: Fed Walking Into 'Mistake': Economist Warns What Happens Next | Anna Wong Show: David Lin (YouTube; host David Lin) Guest: Anna Wong — Chief US Economist, Bloomberg Economics (Chicago PhD; a decade at the US Treasury, Federal Reserve and White House CEA — chief international economist 2019–20) Date: 2026-09-11 URL: https://youtu.be/vttNrjl8tpY Length: 39:05 Note: YouTube auto-transcript. Fillers (uh/um/you know) and stutters removed; garbled names corrected (Kalshi, Kevin Warsh, Scott Bessent, Kevin Hassett, IEEPA, ECO <GO>, Bill Ackman, Jackson Hole, FX intervention); wording otherwise verbatim. Kalshi sponsor segments (00:56–01:36, 38:36–end) kept but not analysed.
00:00 It's Friday, September 11th. This morning, August headline CPI rose 0.4% on the month and 3.4% over the year. It's more or less in line with market expectations, while core prices rose 0.3% hotter than market expectations. Gasoline jumped 3.9% in August and accounted for more than one-third of the monthly CPI gain.
00:20 The 10-year Treasury yield is near 4.93%, the highest since 2023. And this comes after Thursday. Treasury Secretary Scott Bessent announced that the US Treasury bond buybacks will increase to $6 billion per operation. Meanwhile, Brent crude is just now at 107 a barrel and WTI is just under 100 as of recording at 12:30 p.m. Eastern time.
00:40 Oil dropped 3% after Iran and the Gulf States announced their meeting to discuss a deal. So, is a Fed rate hike next week pretty much a certainty? And how much longer can the consumer last on the double whammy of higher interest rates and higher oil? Well, right now on Kalshi, there's a trade going on.
00:56 Traders are predicting that there's an 80% chance of a 25 basis point hike next week. This surged dramatically from 60% just yesterday, now to 80%. So, if you agree with markets, a $50 trade can yield a payout of $60 if you're right, and they do hike next week. This video is sponsored by Kalshi, the largest prediction market in the US.
01:15 Unlike a sports book, you're trading peer-to-peer on real world events, from economic data to political outcomes, and the price moves based on public opinion, not a house. Go to the link in the description down below or scan the QR code here and use my code lin and new users can get $25 when they trade $25. Kalshi is CFTC approved and available in all 50 states including California and Texas.
01:36 Bloomberg Economics chief US economist Anna Wong joins us now. Prior to joining Bloomberg, Anna earned her PhD in economics from the University of Chicago and then spent a decade at the US Treasury, Federal Reserve, and the White House where she served as chief international economist from 2019 until 2020. Welcome back, Anna.
01:52 Good to see you again. >> Hi, David. Nice to be here. >> Let's start by talking about today's CPI report. August CPI rose 0.4% monthly, 3.4% annually. Is this a temporary setback or the start of renewed inflation? Like I stated in the introduction, a lot of this was accounted for by the rise in gasoline and oil prices over the last month.
02:14 >> Yeah. So the headline and core surprised on the high side, but when you dig into the details it is really due to a couple outliers. One of them is the wireless phone services category. So in Bloomberg we have created a massive price mapping project called Bloomberg Price Project where we collect over a million prices across 300,000 categories each month, and what we notice is that what's contributing to this jump in wireless phone plans is due to Verizon. And also there's another
03:01 discrete jump in hotel prices. And these two categories are sort of irregular, one-off, temporary, volatile type of categories, right? On the other hand, we still see a lot of disinflationary pulses in the report. For example, housing rents, primary rents and owners' equivalent rents, are low and they have moderated, continue on a moderating trajectory.
03:32 We are seeing several food categories also coming down. So there had been some supply shocks several months ago in lettuce, tomatoes; we are seeing a mean reversion in those prices. And at the same time we are also seeing that medical commodities, so both prescription and non-prescription drugs, are seeing declines in prices.
04:02 So all in all, if you drill into the details of the report, it is one that's more mixed and nuanced than the hotter-than-expected headline and core readings suggest. Unfortunately for the Fed, there's now not much room to maneuver to consider these nuances, and that's because of two things.
04:32 Number one, Kevin Warsh had given a very hawkish speech at Jackson Hole a couple weeks ago and that speech has pretty much signaled to the market that, well, if the CPI report does not show declining year-over-year inflation, then market should go ahead and price a hike in. And so that's what the market has internalized from Warsh's speech, and subsequently Governor Chris Waller also gave a speech of similar tone, which is that he said we need to see the CPI report; if the CPI report shows
05:15 inflation is hot, we move; if it doesn't, we hold. Well, the second factor that really has boxed the Fed in is that Kevin Warsh has said he's reducing forward guidance; rather, he wants to lean on the market signals, the unfiltered market signal, to shape economic reality. Well, the market signal right after the CPI report, which is unfiltered and not contaminated by Fed speak, is that they want the Fed to hike.
05:49 They expect the Fed to hike. When market pricing is almost at 90% for a hike and you have a Fed chairman who has pretty much said that the market will guide what we have to do, well, then as a chairman you have to be true to your words. He has to be true to all his words, and the words that he has given is that number one, markets is the king.
06:15 We have to listen to the market. Number two, that he thinks inflation is too high and the CPI report will govern what they would do. So that's where we are, David. I think that the lack of room to discuss the disinflationary details in this report is a result of >> this attempt at reducing forward guidance and as a result limiting the flexibility and maneuverability around the details of a report. >> The ECB raised rates earlier this week; the explanation was of course higher
06:54 gasoline prices and higher oil prices. Here's a headline from NBC News. Diesel hits an all-time high today, $6 per gallon. And there's a quote here from a farmer: We are endangering one of the strongest economic pillars that we have in this country. Can you just walk us through from an economist perspective how much weight we should put on diesel's impact on core CPI and consumer prices?
07:22 In other words, how likely is it that diesel will actually make other things more expensive very quickly? >> I'm glad you show that headline because it encapsulates all that matters. Number one, diesel is hitting the headline CPI, and likely in September, we're 11 days into September,
07:49 that is going to translate to more than 5% increase in gasoline prices, which will add roughly 0.2 percentage point on headline CPI. So in September the headline CPI is going to be probably at least 0.4%, pretty high. But this headline also shows that there is an economic feedback loop when you have diesel and gasoline prices that high, and that is that it will be restraining economic activity.
08:21 Here in this headline you show that farmers are going to be hit because farmers use a lot of diesel, and you know what else uses a lot of diesel is freight companies, trucks, and whoever has to buy from trucks. So what kind of goods have to rely on transportation costs? Well, food services, food, groceries, appliances.
08:46 So we are going to see some pass-through and higher pressure on some grocery prices and also on some heavy bulk appliance goods. And actually we saw a glimpse of it in the CPI report. So heavy appliances, kitchen appliances saw a pop in inflation. That's not due to tariffs, 'cause the tariff pass-through is over.
09:09 It is due to this transportation cost increase. >> Is it fair to say that these things that you mentioned, groceries, bulk appliances, these things are more sticky upwards than downwards? In other words, once the prices are raised, it'll take some time for prices to come down, if that's even possible. >> Not food.
09:31 Food is very, very flexibly priced, but food producers also have less margin to absorb the hits, so that's why they tend to pass through the prices very quickly but also cut back very quickly, just like eggs, if you remember. >> The flu. >> So is it bottom line then, is it fair to say that if we have some sort of de-escalation in Iran, which may bring down the oil price, then this inflation scare that we're seeing right now is going to go away for the most part? >> In short, yes.
10:08 In fact, it was going to come down very quickly in the first quarter of next year. So headline CPI could possibly fall below even 2% if gasoline prices or crude oil were to come down to a more regular 70 or 60. And headline CPI falls really fast as it rises very fast. Yes.
10:37 >> The faster it rises, the faster it falls, and this is why typically the Fed should not and does not respond to headline gasoline price increases. >> Okay, let me show you the probability of a Fed rate hike next week. On the CME FedWatch tool, we have 88, 89% of 25 basis points.
11:00 On Kalshi prediction market, it's pretty similar, 81%. Importantly, this chart shows it was at only 60% yesterday. CME FedWatch had 66% yesterday. The CPI report that we quoted earlier was more or less in line with expectations. So why did markets jump up all of a sudden on the probability of a Fed hike, significantly, in just one day?
11:20 >> Yeah, exactly. This is the dilemma that the Fed faced that I just described. They have boxed themselves in. When you look at the CPI report, it really only surprised by five bips. So, we had going in 0.24 for core CPI. The actual is 0.29. And that five bips of price is entirely due to wireless telephones and Airbnb shifting their policy, like very minutiae things, right? Not a broad inflationary impulse increase, right? And however, it rounds up on the Bloomberg terminal from a baseline
12:03 expectation of 0.2 to 0.3, and that is considered an upside surprise, and I think market participants just summarize that increase as, oh, this checks the box of what Warsh said, a hot CPI, when they just don't have time even to look at the details, that, oh, it's the wireless phone; it's the headline that matters. And so, yeah, market pricing jumped by 20 percentage points after the report. >> Do you more or less agree with the probability here? I will show you the
12:43 BLS labor situation here. Total non-farm payrolls increased by 162,000 in August. Factoring in the employment situation with the CPI data that we saw today, do you think 80% is reasonable? >> So I would say 80. So as I said before, Kevin Warsh said the market signals from now on will shape economic reality. Markets have spoken.
13:13 They said they expect a hike. So if market says it's almost 90%, 80%, it's 90%, 80%, because the Fed is going to listen to them under Kevin Warsh. However, I think this hike is a mistake and the economy will suffer for it. When you drill down into the details, I already have described how it's a few straggler outlier items in the CPI report that created that five bips of surprise. For us, on the jobs report also, once you look at the details, in fact, it's pretty weak.
13:57 So it is the seasonally adjusted payrolls that surprised on the high side, but when we look at the non-seasonally adjusted payrolls increase, which is really how people feel because that's the reality, that figure is not contaminated by any models or any adjustment, that unseasonally adjusted increase in payroll
14:23 was actually lower than August last year, and in fact even lower than my own expectations and our forecast. And so, well, how is it that an even higher NSA figure from last August actually produced a mere 20,000 increase in payrolls, which was subsequently revised down to 70,000, and a weaker NSA in August this year actually produced over 160 seasonally adjusted figure?
14:56 So I think the strength in the labor market report was a seasonal quirk. There's not much substance in it. But unfortunately, again, it is the headline that drove all this market pricing. And you know what? Even though I am from the Chicago school, I got my economics education in Chicago, I'm all about markets,
15:23 I do think that markets are not always right. I mean, in terms of the substance of things, markets like to also do groupthink, markets herd, market has network effects. So it might not be the right thing to do, and hence it is quite important for policy makers to maintain their independent thinking, and I think the actual closed-door deliberation of the Fed next Tuesday and Wednesday will be actually a more close call than the near certainty that the market has priced in right now.
15:58 This is why I still only expect one hike this year. >> Could this hike happen after September? By the way, we have to factor in the fact that yes, maybe Kevin Warsh wants to follow what the markets are doing, but is he following what Trump wants to do? Trump basically last week explained on Truth Social that the Fed better get wise, and if they don't keep rates the same, he's going to push tariffs on other countries.
16:32 Do you see that threat? >> Yes. So note that Trump is saying that to the Fed, not to Kevin Warsh himself; the threat is not specifically mentioning Kevin Warsh. When you listen to what Kevin Hassett, the NEC director, and Trump say about Warsh, they say he has to do what he has to do. >> Yes. >> So it's almost like they are blaming the rest of the FOMC but not Kevin Warsh.
17:11 And you can see that Kevin Warsh would go to Trump in this situation next week and say, well, he's outvoted, the majority of the FOMC prefers a hike, the market overwhelmingly prefers a hike, if they don't hike the long end of the yield curve would blow out. This is what he could say. It's a masterful job at managing the politics around the situation, if you ask me, because he basically deflected all the blame from himself.
17:48 The fact is he was the one who guided the market to look at the markets. He was the one who said he wants market signals to shape economic reality. Well, and he let the market push the Fed around. So he created this very difficult situation, but yet he will survive under Trump due to the mastery of his way of dealing with the politics of this.
18:17 >> I want to come back to this mistake that you mentioned from the Federal Reserve. But I want to show you a line from today's BLS real wages report. From August 2025 to August 2026, real average hourly earnings decreased 0.3% seasonally adjusted. However, we have here a situation where there's a 0.6% increase in the average work week.
18:39 This resulted in a 0.3% increase in real average weekly earnings over this period. So people's average weekly earnings have gotten better, inflation adjusted, due to the fact that they're working longer hours on average. When you see a stat like this, what do you think? >> Well, working longer hours suggests the underlying economic momentum has improved, and that's what we have been writing all along from last year.
19:08 So, we said last fall that we think the economic cycle has bottomed, that we are in early cycle now. That's actually exactly the time when we wrote that, and that coming into this year the labor market will improve. So we were quite early in pointing out these early business cycle dynamics, but in this year I also have noted that this is a pretty fragile recovery and it still needs the support of the Fed, and you can't kill
19:45 this recovery too quickly by getting back into a hiking cycle. I think where we are in the business cycle is we are in the early to middle business cycle, and the Fed tends to hike late cycle, not in the early to middle cycle. So this is rather early, and I think this is going to squeeze and reduce the runway of this labor market recovery.
20:15 >> I'd like to play for you a clip of Treasury Secretary Bessent yesterday, and we'll talk about the Treasury's buyback program, but take a listen first. This is what he had to say. >> And look, if some of the Bloomberg terminal bros are unhappy with what I'm doing, well, that's too bad. And I will tell you that we have the best performing bond market in the world, and if you look, bond yields have never been more correlated to the energy price, and that's my point here, that is, we have a supply shock and
20:51 we will get to the other side of this. >> Okay, before I comment, I'll let you comment. What's your comment on hearing that? >> Well, I don't know what he meant by Bloomberg terminal bros. There are Bloomberg terminal sis, too. >> Yeah. As a self-identified Bloomberg terminal sis,
21:18 what is your response to the market fighting the Treasury, which is kind of what he's alluding to here? >> David, he's pointing his finger at the wrong person, because I am actually one of the more sympathetic figures to what he's trying to achieve here. After all, I had spent a decade in public policy and seeing things from the perspective of policy makers.
21:43 So from his eyes, he's seeing this long end of the yield curve blowing out, and you want to do whatever you could to contain the situation. And one solution is to increase the buyback at the long end and issue in the short end. That's like a natural thing to do. Many reasonable people will choose to do this.
22:09 And second of all, when he announced this a couple weeks ago, the one-day and two-day effects were very huge; immediately the long end, 10 and 30-year, fell substantially. And the most comparable policy to this is the Fed's Operation Twist back in 2011, >> which is 400 billion of buyback.
22:35 So that is like 100 times higher than the 4 billion buyback that he announced a couple weeks ago. And yet the one-day effect of Bessent's announcement is substantially more than one-hundredth, proportionate to the size of the announcement. I think what that suggests to us is that the long-end market is very illiquid right now and there's a lot of room for price discovery, and that, from a policy maker's perspective, is when you want to intervene.
23:13 This is also a reason why many policy makers, not just US but in other countries, like to intervene in August, when generally the market is not as liquid. And on top of that, in Bessent's case, he could see the immediate price discovery impact his announcement had. Now, as to the 6 billion the other day, >> it was not as large as what markets had hoped for.
23:45 So that's why yields went back up. So I think the Treasury still has a lot of firepower, and from when I was a student or researcher studying FX intervention, which is where my background was, more on the international side: the way you think about intervention, when is it effective, is when the market is disorderly.
24:13 You have to have a surprise element to it, and you have to prove to the market that you have abundant resources and you're ready to go all in. This kind of rule as to how you should intervene applies not just to FX, like to the yen, how he is dealing with the yen as well, but also to how the Fed in 2020 and during 2008, when you're dealing with what you think is the crisis moment, you want to show all the firepower, and you're like, open the hose, and you're going to do it.
24:48 And so before you even do it, you immediately exert an impact on the market, and that's how you want to do it. So I think the problem with the 6 billion is it's not large enough. Bottom line, Treasury still has a lot of firepower, and I'm more optimistic that he's going to have an impact ultimately.
25:12 >> Ultimately, okay. He did mention that Treasury yields and the price of oil have never moved together like this before. I'm paraphrasing here, but that's kind of what he's alluding to. He has a point here. I have here the 10-year yield and the WTI price. Is this a spurious correlation here or are we looking at causation? He mentioned supply shock.
25:36 What effect does the price of oil have on the long end of the curve directly? >> Yeah, I think this correlation suggests that it is some kind of global risk aversion, a pullback from bonds, that's driving this, because I think what the oil price increase is reflecting is higher risk, higher uncertainty, and in the past when uncertainty increased the Treasury long-end yields would go down because of the safe-haven feature of US Treasury assets, but now we are seeing it increase, which to me is a bit concerning.
26:24 >> Okay, you mentioned that the reasonable thing to do would be to increase the short end, decrease the long end; that's Operation Twist. Why is that the reasonable policy here? From my understanding, Anna, corporates borrow from all ends of the Treasury yield curve, and so would it have more of a positive impact, let's say, to increase the short end of the borrowing curve versus the long end? >> Well, so it is about practical policy making versus following the convention. I think that's the kind of dilemma that Bessent faces,
27:00 because the US Treasury has some longstanding practices on a few things, and all new Treasury secretaries get introduced to it, and many stick with those conventions, but some people may question them, and Bessent is one of those who have questioned them. So some of those longstanding Treasury conventions include, well, the Treasury Secretary always has to reiterate that the US favors a strong dollar.
27:32 I mean, this has started since Bob Rubin during the Clinton years. Nobody can challenge that. As a former Treasury staff, I can tell you everybody knows that you're not supposed to challenge that. >> The second thing is Treasury should design issuance in a regular, predictable manner. The idea is that by a regular, predictable manner across the yield curve,
28:00 the predictability, with lower risk premium, will ultimately lower the cost of financing the fiscal debt. Now I think Bessent is basically challenging this latter convention. He's looking at this in a more practical way. He's like, right now the fiscal deficit is hitting 6% of GDP at the end of this year, and more than half of that would be due to interest payments, and this 6% is higher than the fiscal deficit that CBO projected earlier this year, in large part because of the higher-than-expected Treasury
28:44 yields. So if you continue to issue at this higher long end, which is close to 5%, when you could have issued in the short end, where it's more like 4.2 or something lower, that could make hundreds of billions of difference.
29:08 I think he, as a former hedge fund manager, and also his boss President Trump, as a former businessman, just couldn't pass on this immediate cost-saving incentive. But there's also a lot of value to the longstanding Treasury position on this, to have a predictable manner of issuance that will ultimately lower the risk premium, but that is more of a medium, longer-term consideration, and I think Bessent felt like he's facing a crisis situation right now and he has to utilize crisis-type flexible policymaking.
29:50 >> Well, a lot of people are worried about the absolute level of the Treasury yield right now. Certainly the Treasury is worried about that, otherwise they wouldn't be intervening. Here is a quote I'm just pulling up from the Financial Post here: the 5% level of the 10-year is seen by some as a threshold above which financial markets might go into a meltdown, says John Higgins from Capital Economics.
30:12 While we aren't convinced that 5% is that magic number, higher Treasury yields will certainly pose a risk to the sustainability of the US public finances. What is it about 5%? Which, by the way, is not even higher than where it was prior to 2008. >> I think we hit 5% in October of 2023. I still remember >> that time.
30:36 And in fact, I still very clearly remember, I think it was Bill Gross or Bill Ackman who said he bought the long end at that time; I think when the long end is at five you will have people coming in to buy. At least that's what happened in October 2023.
31:03 I do feel like at 5% 10-year Treasury yields, that is absolutely restraining for the economy. We already have seen the housing sector cooling very rapidly this year as 10-year yields linger above 4.5%, but at 5% you're going to see very, very clear restraining impact on economic activity, as well as on small businesses and eventually on the labor market.
31:35 So after October 2023, we started to see the labor market slowing very, very rapidly. Between then and the spring of 2024 was when payrolls were overstating actual employment growth substantially. And recall in August 2024, there was the flash crash that was triggered by a surprisingly weak labor market report and a jump in the unemployment rate.
32:05 I think all of that is the economic outcome, the lag effect, from the really high long yields back in late 2023. And this is why I think it's a mistake for the Fed to be hiking right now, when the long end has already done much of the job for it, because the long end has risen; 10-year yields have risen by at least
32:32 .5 percentage points since March of this year, at least, and that is equivalent to about 100 basis points of rate hikes. >> Do you buy the argument that if the Fed does not hike then the bond vigilantes will continue pushing up the long end, and if the Fed actually starts hiking then the bond vigilantes go away? >> No, I don't buy that.
32:56 And there's evidence for it. After Kevin Warsh's very hawkish speech at Jackson Hole, the long end temporarily went down and the next day went back up again. And I think ultimately what this economy needs is growth. And the rise in the long end is ultimately because of the fiscal situation in the US, and there are many ways to resolve this fiscal situation.
33:24 For example, the Supreme Court ruling against the IEEPA tariffs, which started this refund wave draining cash from the Treasury. Well, that has produced a 1 trillion loss in fiscal revenues over the next 10 years, which is actually one consideration that's driving these fiscal vigilantes out onto the bond market.
33:56 If the Fed were to tank the economy, or the Fed raising rates were to slow the economy substantially, slow the labor market and even puncture the AI boom, then the fiscal dynamics will be out of control, because we are already at 6% fiscal deficit. Under normal economic times, when you have a growth slowdown or recession, there's no fiscal space; you can't get to 10% fiscal; it would just be a downward spiral in the fiscal dynamics. So this is actually where, if I were a policy maker, this is the time to be practical about
34:40 everything, and the number one threat to the US is the fiscal dynamics, and you need fiscal revenues from a policy tool that does not slow growth substantially, and the one that we have in front of us appears to be tariffs, because tariffs have been generating revenues without tanking the economy, and it is a policy tool that both Democrats and Republicans seem to agree on.
35:08 You don't have anything like this on the table. I mean, if there are, then we should go for it, but I don't think there are. And so this is why I think the Fed raising rates is a mistake, because there's already a lot of disinflation impulses, number one. Number two is that, well, if you have a little bit of inflation, it also helps, while you know that inflation is on its way down; you don't want to puncture the growth, which would really then, at that point, what would the Fed do? Lower the rates to zero, or QE, which
35:48 Kevin Warsh doesn't like. And I just think that both the Fed and the US economy are in a deep hole right now because there's just not many options. >> Well, this leads to my final question on growth overall, and we have a few minutes left. Thank you, Anna. 4.4% GDP is the third quarter estimate from the GDPNow cast from the Federal Reserve of Atlanta.
36:13 We have here a situation where, like you said, the economy is restrained by higher interest rates but at the same time also restrained by perhaps higher oil costs and gasoline costs and diesel costs. On the other hand, the AI capex continues to expand, driving output. Which side wins, Anna? >> Well, recall that right before 2008, growth was amazing too, and the Fed was declaring everything a soft landing early in 2008.
36:44 GDP is a contemporaneous indicator. And I do not challenge this. Growth was very good. We were going through an early business cycle recovery from the second half of last year to the first half of this year. However, if the Fed were to come in so quickly to squash this, it could end up pretty ugly, and I also think that the double derivative of AI capex has peaked this year.
37:15 The double derivative meaning the growth of the growth rate. I think the growth rate of capex is the fastest this year. It is the growth rate that contributes to GDP growth. So we estimate that AI capex roughly contributed 1 percentage point to GDP growth in the first half of this year. If capex growth will slow next year, even though it's still positive, it means the GDP contribution would be smaller; we think it could be
37:50 .5 or lower. And so we are at peak AI capex boom contribution to GDP right now; that is going to change. >> Anna, excellent, thank you very much for your time today. Where do we follow you? >> You can follow me on the Bloomberg terminal by going on the terminal and typing ECO <GO>. That is the landing page for my analysis and my colleagues' analysis that covers the economies around the world.
38:16 I also have a Twitter account and you can also follow me there. >> Okay, we'll put the link down below so you can follow Anna's Twitter account, X account, and of course we'll use the ticker on Bloomberg if you want to follow Anna and also her Bloomberg terminal bros and gals colleagues there as well.
38:36 According to Scott Bessent. Welcome back to the show, Anna, it was good to see you and I hope to see you again soon. Take care for now. >> All right, thanks David. >> And thanks for watching. Please do like and subscribe. Follow Anna, links down below, and don't forget to use my code lin when you sign up to Kalshi. Remember, new users who use my code L I N can get $25 when you trade $25.
38:55 Link down below or scan the QR code here.