Newsfilter article unlocker

What's driving inflation and how sticky do we think that those drivers of inflation are going to be? It's a really interesting time to talk to Stephen Myron, who was previously the chair of the Council of Economic Advisors and on the Federal Reserve Board of Governors. And if it were him, he says he would have done it differently when it comes to the interest rate strategy in the U.S. right now. You know, it's pretty clear that hiking to keep the long end down, you know, I don't I don' t view that as a successful enterprise. So we'll find out what he thinks about the underlying economy. thank you so much for joining us. You've returned to Hudson Bay after a tour in Washington, CEA, Federal Reserve, as we were just talking about. You know, it's hard to not start with the current interest rate environment. Before we go to what the Federal Reserve looks like and is doing right now, I do want to get your opinion on what is going on in the long end of the curve, because arguably this is the much more important benchmark to be looking at. Sure. So first of all, thank you for having me and thank you all for being here. So I think it's useful to divide the recent history in the long end into two periods. And the first one would be the period up to the most recent Federal Reserve policy decision where they raised interest rates. And if you do this division, two segments of time, really up to mid-September, the increase in bond yields had been driven entirely, almost entirely by an increase in growth expectations. So if you divide long yields into three components, the growth component, the inflation component, and the risk premium component, of course, in treasuries, we call this term premium. The three components. Inflation had been very well behaved. In fact, it was extraordinarily well behaved, there was no doubt whatsoever in the inflation market's mind about the path of inflation returning to the Fed's target in coming months and years. The risk premium or term premium component had also been very well behaved. And in fact, it was at slightly lower levels than it was the start of the year. And all of the increase had been in the real growth component. Now, to me, that's actually a really good growth story, right? That's a good increase in bond yields and not an increase in bond yields that I think that needs to be fought either by the monetary or the fiscal authority in terms of the level. Because if bond yields are going up because growth expectations are going up. That's a great fiscal story. When the economy is doing well, revenues grow faster than that at least, and they grow faster in the increase in interest payments, and that brings the deficit down over time. Now, so I think it was actually a really good story, the increase in bond yields. And we'll double-click. To be clear, we'll double-click on inflation and term premium. But growth was good. What's happening now? Well, so, I think if you look at what's happened since the Fed, I think since the FED meeting in September, I think we've got a new reaction function from the Federal Reserve. Inflation data had been coming down relative to where they were in the previous month. The inflation data the last few months have been actually quite good. We got further downward revisions to inflation, which were also well known to be right around the corner at the time of the September meeting. And despite the inflation data coming down and the inflation date looking better, normally a central bank would become more dovish into that. But this central bank actually became more hawkish into the inflation debt coming down. And so what we've got is a completely new reaction function at the Federal Reserve, where the, you know, the old reaction function from the Powell Fed, I think, is no more. And we've got a new reaction function, uh, from the new Federal Reserve. But we don't know what it is yet. So, we don' t know, um, what the new reaction function is. Chairman Warsh is, uh... famously opposed to providing forward guidance. Famously opposed to telling the market what's going to happen next. I su... You know, I support that withdrawal of forward guidance. But the flip side of that is the increased uncertainty over exactly how monetary policy gets made and exactly how to map economic data into a policy outcome means the additional uncertainty over how monetary policy gets made gets reflected in higher uncertainty, higher risk premium in the yield curve. And so I think what we've seen since the September Federal Reserve meeting has been an increase in term premium related to this uncertainty. That's kind of a very large thing for you to say, which is why I am double-clicking on this. For you to saying that the term premium is rising and that is what's creating more of a sell-off in the long end of the yield curve. What stops that? Do you think that that means that the long of the curve is A, beyond the Fed's control, beyond the Treasury's control and has kind of a very meaningful unknown upper bound on yields? Well, so I think the expression beyond the control of this is kind of a loaded one. It's not how I'd really look at it. The bond yields will respond to actions taken by the Treasury and the Fed. And as I said before, up until the middle of last month, I think the increase in bond yields was entirely benign and not one that needed to be aggressively fought. I think as we learn more about the new Fed's reaction function, I Think that uncertainty premium will come down. However, it'll take some time. Now, of course, as I say, I support the withdrawal of forward guidance and the reduction in the Fed telling the market exactly what's going to happen in each meeting, because I think that uncertainty is helpful to the Fed to have flexibility to change policy when the economy changes. And I think it's helpful for the market to have to do its own risk management as opposed to just outsourcing that to the fed, because if the Fed gets it wrong, as they did in 21-22, you get situations like Silicon Valley Bank, where the Fed told them, hey, rates will be zero forever. And as a result, they basically loaded up on all that bad interest rate risk. So I think forcing that risk management back into markets is a helpful thing as well as giving the Fed flexibility to operate. Now, the flip side of additional uncertainty is the higher risk premium. But to me, in my mind, if you're going to have tighter financial conditions as a result of more uncertainty over how monetary policy gets made, in equilibrium, that's just going to lower the neutral interest rate. Because if the Fed is targeting inflation and unemployment at a given level, and then you have a tightening of financial conditions due to something like uncertainty changing, then the Fed's ability to achieve inflation and unemployment at the same way is going to be altered. And so the Fed would be offsetting that with a lower policy rate. So to me, there's two sides to this coin. One is higher term premium. The other side of it is lower neutral rate. And I think that's a trade-off that I support taking. But there has been a change since the September FOMC meeting. When you look at the data, you look to the New York Fed's benchmark term premium estimates, it was pretty flat through the year until the middle of last month. I'm going to try to avoid making you make a long-term interest rate call. But in the short term, it's interesting because before the most recent Federal Reserve meeting, you had said that you believe that the Fed shouldn't cut rates. And you're saying now that the neutral rate is lower, but actually the neutral rate was projected to be higher in the most recent summary of economic projections. What's going on? Bring me into the divide. Yeah. So just to clarify, I thought they shouldn't hike rates, and I still think they shouldn't hike rates. I think the inflation data have been doing well in the last few months. They're on track to converge 2% over time. The key thing here is that monetary policy works with lags. If you change interest rates today, it doesn't hit the economy for 12 to 18 months from now. This is Milton Friedman's famous fool in the shower story, where somebody goes into the shower, turns the water on, and it stays cold for a few minutes. It takes some time for the water to heat up, stays cold for a few seconds, takes some time for the water heat up. But because the water stays cold, the person turns it all the way to the right. And then when the water finally does come out warm, it comes out scalding hot. And so the person turn it to the left, takes a few seconds to cool down, still scalding, hot, turns it always left, and then it comes out ice cold again. And, so, the point is that if there's a lag between when you turn the shower dial and when the water gets hot, you need to sort of just wait and wait for the water to change temperature. It's the same thing with monetary policy. If you move interest rates, it takes a year to 18 months to actually hit the economy. And so that's why you historically would look through the type of supply shock from energy that would boost inflation, because it boosts inflation immediately. If oil prices go up, gas prices move up very quickly. They don't move up a year from now. Whereas the labor market, if the labor market tightens up, wages are a bit of a lagging actor. And so when the labor markets tighten up, you get inflation a year from now, right? And so that's why you'd be very responsive to the labor market as a monetary policymaker, but less so to energy markets. So back to your question about neutral rates. I think neutral rates actually have been moving up a little bit over the course of this year, which is why I've been calling for fewer cuts than I was earlier this year. And I think a lot of that is related to the AI build-out. There's a huge amount of borrowing related to building AI due to the investment, the reduction in free cash flow at the tech companies. And this increased need for investment funds works in the models to boost the neutral rate. I think that's very real. Now, the flip side of that is that this term premium has been going up lately, and we'll see if it stays elevated. We'll see if the Fed provides a little bit more clarity over how they're mapping economic data into monetary policy. But from my perspective, if the inflation data get better and inflation comes down, which is what happened over the summer, and the central bank becomes more hawkish, those things are moving in the opposite direction, and usually they move together. So the question is, what are the data that the Fed needs to see in order to hold, in order to hike? I think we need a little bit more clarity over that for the term premium to settle down. So to recap, in September, you believe that they should still cut. They went off and did hike. And through the dot plot, you could see that there are more hikes really baked in. You also believe that there's a really meaningful transition mechanism that takes a long period of time. So say they hike again, right? what is the risk there in your view in terms of how long that takes to feed through the economy and what really is impacted right we're still pretty close to record highs on the S&P 500 I mean it's not like we've seen anything break yet so so to be clear I don't think I would have voted to cut at the center meeting I probably would have owed it to hike in the reason why I would really Sorry. I would have voted to hold. I'm all tongue-tied. I would have voted to hold at the SEMDR meeting. The reason is we're talking about bond deals and there's been a lot of volatility there. This is my view about what's driving the bond deals. Nevertheless, in the interest of caution, I would have voted to hold, not wanting to rock the boat anymore because things could get a little bit unhinged. Financial markets are prone to what a economists call excess volatility. And so I wouldn't have seen a reason why you couldn't have waited a little bit for markets to settle down a bit. But I do think it's pretty clear that hiking to keep the long end down, I don't view that as a successful enterprise. It's clear that it hasn't happened. Part of the reason why it hasn'T happened is, of course, if you are hiking enough to induce a recession, you will get the long-end under control. That's not what's happening. If you're just increasing the short end of interest rates a little bit, and your marginal buyer of treasuries is now somebody who is sensitive to those short rates, it just gets passed through to the long end. You've raised the funding cost for the guys that are going to be buying bonds. So I don't view that as a successful enterprise. So I wouldn't have voted to hike. I would have voted for hold if I'd been there. But, you know, of course I wasn't. So the inflation backdrop, there are a lot of investors and strategists out there who would say, well, wait a minute, you know, even if you believe that inflation is coming down, that there is an element of stickiness to the core prints that are impossibly hard to ignore at this point in time. You know, what do you say to those people? Right. Because you have plenty of people who don't agree with you right now on that. Yeah, I think it's important to to sort of ask what's driving inflation and how sticky do we think that those that those drivers of inflation are going to be? If you look at core PCE, there's no question that it's elevated, that it'S materially above the Fed's the Fed'S target. If you Look at core CPI, it'S the lowest level since March of 2021. right it's undone almost the entirety of the boost to inflation that's occurred as a result of the pandemic and everything that's happened afterwards uh now you know the way that cpi is calculated there's an upward bias in cpi of about four tenths of a percent due to the way the weighting methodologies are so on a historic average if cpi were at that level core pc would be at 2%, right? The deviation is due to the inversion of that relationship is due to a small number of components that are extraordinarily out of whack historically. Now one of those has been the portfolio management services which until very recently just tracked the stock market higher, so the stock market moves higher and that mechanically gets written down as inflation, which to me is nonsense. Another one has been AI, right? AI has driven up some components of prices like memory prices. It's memory prices, it's driven up, some software prices as well. To me there's also a very large element of measurement error in that component as well, it gets into some, you know, I'm happy to go into as much detail as you want on this, I've made a very large study of these things, but basically the way that the surveys work is that things get up weighted by 35x in PCE versus CPI because of the way that, sorry, the software component, software and accessories, including the flash drives gets up-weighted by 35x in PC relative to CPI, just because of different data collection methods. And if you take the data at face value, they imply that a typical family is spending about $500 a year on flash drives, which I don't think is really the case. I think that's a statistical anomaly that's the result of the data collection method that doesn't reflect what people actually buy. And so these two categories alone, going into the September meeting, we're adding 70 basis points to core PC, right? So of that 1% inversion plus to the CPI-PC relationship, about seven-tenths of a percent was due to these two categories alone. Now, BEA, the Bureau of Economic Analysis, recognized that there's very significant measurement error here, and they recently took steps to try and address it. And those two categories, they made some revisions, and those categories ended up reducing their contribution to core BC from about 70 basis points to 35 basis points, right? So they cut that in half. To me, that's a good start. It's not enough. So the question is, what's driving inflation higher? If I look at the portfolio management bias, I don't see that necessarily as recurring. If I looked at the software and accessories contribution from AI, look, the AI investment surge is really big, and it seems very durable. That I could see continuing to persist. And so the question is given monetary policy lags, given that monetary policy affects the economy a year to a year and a half from now, where do you see these components that are contributing excess inflation a year to a half a year from now? Do I see oil and oil pass through into core raising prices a year form now? Are airline fees going to be going up at the start of 2028? I don't see that likely to be the case. Portfolio management fees? No. Software and accessories? Yeah, probably. So we'll talk more about AI in a second and we'll more about one of the largest components here that real Americans feel. It's housing. Before I get there, though, I do want to make sure that I'm properly understanding your view on this, because on oil, right, on one hand, you can completely understand the hesitation on hiking into an oil shock, right? On the other hand, you have diesel above $6, right. That feeds through. It is hard to see that not feeding through the economy. And so even if you don't think 2028 is going to be starting with that type of a pass-through. I mean, isn't it the case that it could get ugly before it gets better? Yeah. So there absolutely can be some pass-thru from diesel into the rest of the core. That's something that I think very much can happen. However, I think it's going to be limited. And part of the reason it's gonna be limited is because the labor market is just not very tight. So if you look at what happened with claims of tariff pass-though, right? I think that there's no question that what's been going on in international economics and the reconfiguration of international trade has had some effect on inflation. However, the predictions that it was going to lead to very severe inflation didn't pan out. And so, for example, you had some very prominent economists sort of say that we'd have an inflation spike worse than we had with COVID, which, of course, nothing like that remotely happened. One of the reasons why tariff inflation has been not super high, has been because tariffs are actually a very small part of the economy. They're only 11, 12%. Another reason has been because the labor market was not overheated in a way that allowed those prices to be passed through and reverberated throughout the economy, right? And so what you had is decreasing margins at the importer level. I think you have a sort of similar situation today where you've got wage growth that is just not strong enough for increased diesel prices to find their way into large-scale pass-through into the economy. If you had a situation like 2021, 2022, where the central bank had rates at zero and was doing $120 billion a month of QE, and the government was tossing trillions of fiscal dollars into the economy, into an economy that was quite healthily opening up and growing on its own, you have situation in which supply shocks can really reverberate through the economy. And so a supply shock in 2021-22 can ricochet throughout because it's interacting with monetary policy, it's interacting with fiscal policy. You don't have that now. Monetary policy is probably, let's call it roughly neutral, maybe it's a hair restrictive. And fiscal policy is also similar, let'S call it roughly neutral. It's nowhere near what they were in 21-22. So it's interesting that you mentioned fiscal versus monetary, because I was going to ask you how you felt about the ways that the long end could be controlled. What do you think about what the Treasury is doing in terms of using bills to finance buybacks of the long end? Yeah, so as I said before, you know, really until the Fed meeting last month, I think the increase in interest rates was driven by improving growth expectations, right? And that's a movement to a new equilibrium that's a result of actually, you know, good things in the economy. If the market is moving to a New Equilibrium, it would take a heck of a lot of firepower to prevent it from getting there, right. And what the Treasury is doing is a few billion dollars a month of extra purchases, right, for comparison the Fed did a hundred, as I said before, the Fed did a 120 billion dollars a month of QE every month for many months in a row. This is such that it added up to trillions, right? But this is a little philosophical, isn't it? No, no, no. It's not. Because if you think that Treasury is trying to actually change the new market equilibrium, it needs trillions of dollars more than it's engaged. So what I'm saying is that that's not what Treasury is trying to do. What Treasury is doing is minor interventions that add liquidity to the most illiquid part of the yield curve. And when they started this in late August, it was the most illiquid time of year. So what Treasury is doing is not getting to some new equilibrium level. What Treasury isdoing is it's altering the path by which you get to that new equilibrium level, in my opinion. So it's reducing volatility in doing so. While it does that, though, you do have a Fed chair that not only is he against the idea of forward guidance, he's against the idea of expanding the balance sheet of the Federal Reserve. So, it's a little strange to see the Treasury buying at the long end when, you know, you're thinking that QE might be a thing of the past. Well, I mean, I don't see those as inconsistent because it's not the Fed's balance sheet that's being used. It's the Treasury's balance sheets. And it's not even an expansion in the size of the balance sheets, it's just swapping one, you know, maturities of different obligations for each other. Like fiscal interventionists versus monetary interventionists? No, look, the maturity distribution of public debt, the distribution of maturity dates of debt held by the public is the core of fiscal policy, right? Like that is the core of the Treasury's policy mandate. And indeed, this is one reason why the Fed should not get involved in QE, is because it interferes with fiscal decisions, and it blurs the lines between monetary and fiscal policy, and leads ultimately to some significant existential questions about the Fed. So that's a good argument for the Fed not engaging in Qe, but this is something that Treasury should be doing. Do you worry, though, that there's so much issuance that is front-loaded and this uncertainty around the long end of the curve is making it harder for the Treasury to then term out the debt, right? Well, you know, I don't know that that's an active policy goal. I mean, I think you'd have to ask Treasury whether that's something that they want to do or not. But like I said before, my interpretation of what Treasury's been doing with buybacks is to add liquidity, is to reduce volatility to the most illiquid part of the yield curve and make it less likely to be a one-way trade in which excess leverage accumulates. Because as I said beforehand, financial markets are prone to excess volatility. And if you let all of this leverage accumulate and the trade becomes too easy one way, it overshoots and then it snaps back. And that's not good for the real economy. So I think the provision of additional liquidity to slow things down is helpful. So back to the real economy for a second here. I could talk about the Treasury versus the Fed all day long. But the real autonomy, I really want to get your views on the housing market. Sure. I think that they're differentiated. And I tend to agree with a lot of them, actually. The housing market in particular, you're watching mortgage rates start to spike. You and I kind of talked about this a few months ago as you saw people start to sell homes. Is that price lower or is it higher, right? That's the big question that's underpinning this housing market for me. We've had this rapid run-up in asset prices in a zero-rate environment, and now we're figuring out what the cost of debt is really worth for a real household. It's hard to see it doing well with higher interest rates, right, You know, and it's tough to see it doing well. I mean, I think that the housing market has been a persistent source of disinflation as market rents have been growing at a 1% or lower rate for about three years now. And I see that continuing to be the case. It's easy to imagine it, you know, becoming a little bit dicier even with, you now, with higher yields. I mean if you're in a situation in which you've got to sell, I mean you could be in some trouble. What does that mean for kind of the overall inflation story, given everything we know about what shelter means for the cost of goods for an American household? One interesting thing about shelter in the way that inflation is measured is that it is the most persistent part of the inflation index. And that's because it takes a really, really long time for market rents to feed through into the official data. The reason is that the official data track rents that are paid and not market rents that you see on the screen. And that matters because leases get renewed once a year, once every two years. And so if market rentS go up and you don't renew your lease for a year you don' t experience that until a year has passed. So that's why there's a really, really long lag between market rent and the measured data. And this is one reason why inflation in 21-22 was always going to be persistent was because the Fed was still buying mortgages. Home prices were up 20% year over year, and the Fed was still buy mortgage, which is 20% in a year, not enough, more credit into the housing sector. That was always going to be persistent because the way that this passes through into the data takes years. As I said before, market rents have been growing at a 1% rate for three years now. Measured shelter inflation is closer to 3% because it's still reflecting the catch-up that occurred years previously. We're still looking at the housing market of 2023, 2024, even though it's almost the end of 2026 now. So when you think about things now, right, you've sort of got this long period of market rents growing very low that'll continue putting disinflation into the measured data. And I think, you know, maybe rents probably will firm a little bit, but it'll still take a long time for it to feed through. You've still got this disinflATION of the pipeline from that. So it could push inflation higher in, you know, 28, 29, a little bit from here. But I think that given the disinflation we've got in store already, I think it's very, you know, it's, it'S very modest. And then on top of that, if you've got the sales market under pressure, I'm not sure how that interacts. And just very quickly here on the housing market, because it's not like we're calling for like a 2008 style housing crash, right? But, you Know, there, there's a lot of supply that's still needed for this market. But, You know, what are the issues that have perhaps been masked as the transaction volume has been muted. Yeah, I think that there definitely are parts of the country that still need supply, and the higher interest rates are going to actually impede that supply coming online. You know, the thing about housing is, as people say, it's always a local market, right? And there definitely aren't parts ofthe country that are still supply constrained, and they need that supply. Now, I think the best thing to do is actually to ease regulations. Now unfortunately, you know, there are building codes all around this country that said at the state and local county, town level that make it more difficult to build. They impose huge regulatory costs on building homes, everything from carbon emission restrictions to zoning rules to everything. And that reduction in supply really has been a primary driver of housing affordability problems. And so the primary solution here is actually not monetary. It's to ease the supply constraints. But that's a state and local policy decision, and it's not a federal government decision. So we only have a few minutes left here. I really do want to talk on a topic that fascinates me the most. And it's not fiscal, it's not monetary, it is totally just artificial intelligence. No shocker. The topic of the year. You were at the Federal Reserve when there really started to be meaningful questions rising about financial stability, the risks that AI poses, you know, increasing cyber incidents and agents gone rogue. How concerned are you about our financial system and its guardrails against all of the above with artificial intelligence? So what concerns me is the cybersecurity element of it, is that we've gone completely from a paper-based banking system and financial system to a digital-based financial system, and that's only as secure as the encryption and passwords that protect it. Now, thus far, we seem to be in a situation in which the frontier models are made available for improving cybersecurity before they're made available to people who would use them for more nefarious purposes. And as long as that continues, it feels pretty good. However, cybersecurity remains, I think, one of the big problems. But this is not a financial-specific risk. This is just a risk to the financial system as it is to everything else. And it's one that I think policymakers will continue investing a lot of effort into. Right. Are we prepared? You know, I think that they do a pretty good job trying to keep on top of the developing cyber threats. But the amount of effort that needs to be invested in this is only going to increase. Right. I mean, agentic solutions are really just getting started in finance. We don't know yet, I feel like, a lot, of the future. Yeah, I wish I did. Before I let you go, it would be a missed opportunity to not ask you about the tariff policy today and the relationships that are forming across North America, what they look like in the future in particular, the U.S. and Canada. How does this all play out in your view as somebody who has worked so closely on tariff policies for so much of your career? Sure. So, you know, it's a really interesting case, and it's one in which the two sides ultimately need to come together and find a solution, because not finding a solution isn't really good. What's an ideal solution in your view? I think you need a solution that isolates North America from predatory foreign supply chains. You think the U.S. and Canada can get it together enough to build that kind of a coalition? I think once there's a little bit of economic pain that sort of convinces everybody to work together, I think that'll happen. Economic pain for who? Well, I Think that it'll probably be more incident on the Canadian side than the U S side. I mean, every single Canadian province trades more with the U s than it does with the rest of Canada. Do you think Americans are going to have to live with some more inflation as we figure out tariff policy globally? So I'm still not of the view that tariffs drove a lot of inflation because I think there were a lot of things going on in international trade. And, you know, the supply chain reconfigurations date back to the pandemic environment in which we had. But it's not just tariffs. It's not just tariffs, it's supply chain risk, it''s political risk, its wars, its export controls. There's so much going on that has driven changing good prices and changing supply chain patterns that, to me, it doesn't make any sense to sort of say it's all tariffs. And I think that if I compare U.S. goods inflation to goods inflation internationally and it were tariffs, I think I would be able to pick the U. S. out of a lineup, and I can't do that. So, you know, I do see inflation on a good path. I think we've gotten good data recently. I think мы're going to continue to get good data on inflation, barring additional supply shocks. You know, you can always have new supply shocks. Stephen, I'm just going to end where we started. We were talking about the long end of the curve and we were talking about the Fed's reaction function, right? Do you think when we put this all together that what it really means at the end of the day is that there is just more volatility to come for investors moving down the road here? You know, I think we're living in a world in which we're trying to figure out how to make things sustainable, right, that means global imbalances. That means fiscal imbalences. That means trade relationships. That means supply chains with choke points. And as the world tries to figure that out, right, and across many different countries and many different economic domains, it has some economic knock-ons and side effects. Has that been something that has been driving inflation data over the last few years? I think that's absolutely an admissible claim. Is it something that's going to continue. Again, I think it is something that is going to continue a little bit. Stephen, thank you so much for being here today. Thank you all for having us and looking forward to spending time with you. Thank You.