Welcome to Heller Hurwicz Office Hours, the podcast where we feature University of Minnesota economists and the research that helps us make sense of a complex world. In this fourth season, we are continuing our deep dive into how rigorous economic research informs our understanding of the rapidly shifting global landscape. Geopolitical conflict, policy uncertainty, and challenges to the Fed's independence are all converging to reshape the nation's economic outlook. Today, we're joined by Neel Kashkari, President of the Federal Reserve Bank of Minneapolis, to discuss this pivotal moment for the US economy. So, welcome.
Thank you. Great to be with you.
So, the most recent data on the consumer price index is consumer prices rose 3.8% in April from a year ago, and that's up from 3.3% the previous month and 2.4% in February. So, how concerned are you? It seems that the progress against inflation not only has stalled, but we're getting some erosion here.
Well, before the Iran conflict started, I was feeling pretty good about where inflation was headed. We made a lot of progress in the prior several years, bringing inflation down. It was still too high, but I had confidence it was heading slowly back to the right direction, back to our 2% target. The Iran war has really upended that, and now we've got this energy shockwave ricocheting around the world. Energy—we strip it out and we talk about core inflation, but energy goes into everything. It goes into the food that we eat, it goes into the goods that we buy. Anything that has plastics ultimately comes from oil and petroleum. So energy really matters a lot to the global economy and to what Americans buy, and so right now there's just a great deal of uncertainty about where inflation is headed, because there's so much uncertainty about what's happening in the Middle East right now.
And so you explained in your letter about your dissent—you agreed with the decision that was made by monetary policy to hold rates steady, but you wanted more flexibility about well, maybe we're going to increase, maybe stay the same or decrease. So, what motivated your dissent?
Well, what you're describing is something that we call—it's a technical term—forward guidance, which is us guiding the financial markets of where monetary policy is likely headed in the future. The reason that that matters is if markets and people believe that rates are going to be lower in the future, that brings forward into some decision making today, and that has the effect of actually easing financial conditions today a little bit. And for the past several months, there's been what we call an easing bias. The language that we use has indicated, hey, the next rate move is likely a decrease. Well, we adopted that language when inflation seemed like it was heading back towards our 2% target. Now that the Iran war has really thrown a curveball at that, and I don't know where inflation is heading. I'm simply saying I don't want to have that easing bias anymore. I think we should be neutral. The next move might be down if the Strait of Hormuz reopens quickly, but it might also be up if it ends up being closed for an extended period of time, and if oil prices keep climbing from here.
So one of the things I find fascinating with everything that's been going on—when you look at the Treasury Inflation-Protected Securities, looking out several years, it still seems that most investors are optimistic about the inflation front.
Well, it's funny you say that. You're right. If you look at the data in financial markets, it suggests that inflation will be short-lived and go right back down to our 2% target. So, on one hand, that gives us at the Fed some comfort that markets have confidence, but what we don't know—what's embedded into that—is financial markets' beliefs about what we're going to do. So, if markets are convinced that the Fed is going to be really aggressive and bring inflation back down to target, then you would see those prices suggest, yes, inflation is going to come back down to target. But then we have to walk the walk. We have to follow through and validate those expectations. And so yes, it's good that financial markets are indicating inflation will return back to target, but that doesn't get us off the hook. We might still have to deliver the medicine that's necessary in order to achieve that positive inflation path.
And part of this is also as an institution, the Fed is independent, and there's been obviously President Trump—he'd like to reduce the Fed's independence. How important, when we're thinking about inflation expectations being anchored, is Fed independence? Because you've gone through a period of time where you have to sort of think this through.
Yeah, it's foundational. I mean, if you look around the world where countries have either a central bank that's not independent of short-term political decisions, or that independence gets compromised—where that shows up is in higher inflation and higher inflation expectations. I'll give you just a perfect example. Argentina has this long history of very high inflation. Eventually, there are reforms to bring inflation back down, but the reforms, the institutional reforms, don't persist. And then another government comes in, and they say, 'No, no, we want to balloon the budget again,' or 'We want to cut rates again,' and then those institutional structures are broken, and inflation takes off again, and it's very hard to put it back together once you violated that trust. And so think about NATO as another example, very different, a foreign policy example. NATO is so powerful because we have this mutual defense pact that if one NATO country is attacked, it's an attack on all. Well, imagine that a NATO country was attacked, and then the other countries didn't come to its defense. Are you going to trust it next time? Does the pact have any meaning? And so it's a little bit like the Fed has to walk the walk and get inflation back down, and our political system needs to support us doing so, so that we have that kind of strong credibility in the future.
So, speaking of credibility, what causes inflation? I know that's a naive question, but when you look back at the recent bout of inflation, there's a lot of talk. Well, it's money—we learned from Milton Friedman, it's about the money supply, but then we had the post-COVID world with all the supply chain issues, and now we're talking about trust and credibility. How do you now look at inflation? What is it that you pay attention to? So, speaking of the labor market, how are you looking at the labor market? It's been at least, what, a year now that we've had this low hire, low fire market. And so on the one hand, the unemployment number is good—I mean, it's low, and yet when you interview workers, everyone seems nervous.
Yeah, there's a lot of nervousness in the economy. So, you're right, the headline unemployment rate is 4.3%. That's not a bad unemployment rate. It's been mostly moving sideways, but as you indicated, there are not a lot of people being hired, and thankfully not a lot of people being fired. But workers are nervous. There are a lot of surveys done of households and consumers and businesses. There's an overwhelming economic anxiety that many people are experiencing, even though the actual conditions are still pretty decent right now. So, from a monetary policy perspective, we want to get inflation back down, but we want to keep the labor market at least where it is. We don't really want the labor market to weaken. From here, you'll hear the term stagflation used. A stagflationary shock, like an oil shock, pushes up prices, but it pushes down on economic activity, and on the margin, it weakens the labor market. That's really difficult for a central bank, because we can fix one or the other, but with monetary policy, we can't solve both problems at the same time.
And with that dilemma, my reading of the past couple years is that the Fed will lean toward, if it has a choice, fighting inflation, because the recent experience is that people really hate inflation, and it is hard to bring down. But am I reading this right?
I think you are. It's going to depend on the circumstances, but it's not just that people really hate inflation, which we were reminded of in the last few years, but we've now had five-plus years of elevated inflation. So when the Fed lifted off—we cut the rates to zero during COVID, we finally started raising rates again—inflation had been elevated for about six months, and we raised rates really aggressively. Now, inflation has been elevated for more than five years, and at some point people are going to say, 'Wait a second, this is the new normal for inflation. We just need to get used to it, and I need to go ask my boss for a bigger raise, because my family is not keeping up with where inflation is.' And if that happens, if everybody says, 'I need to go ask for a raise to keep up with inflation,' then it becomes embedded in our economy, and then it becomes a very difficult problem to solve. So, yes, people hate inflation, but the longer inflation stays elevated, the more important it is that we actually finish the job and get it all the way back down to our 2% target.
Okay, so with that 2% target, you have this tsunami of investment going on with artificial intelligence. What goes into all of artificial intelligence? You have improvements in productivity, and some people are attributing—we're going to have improvements in productivity. So how do you factor into this analysis when you're looking ahead? Do we raise rates, do we lower rates, keep them the same? This remarkable technological innovation and all that investment.
I would say there's two parts of that that are related. The first part, most near term, is what does it do to the average rate environment, and what I mean by that is there's some interest rate that neither stimulates the economy nor restrains the economy. It's the neutral rate, we call it. We don't set the neutral rate, we can just try to estimate it. But if there's massive new demand for investment capital going to build data centers and chips and power plants to power those data centers, that's on average going to raise interest rates across the economy, because that is how the economy reallocates capital away from other sectors into this data center, AI sector. So, a concrete example, a stylized example: an apartment building that used to get funded at a 3% mortgage. Well, now mortgage rates are going to be elevated, and now mortgage rates go up to 6%, and now that builder says, well, that apartment building doesn't make economic sense anymore. I can't afford a 6% mortgage rate, so that apartment building doesn't get built, and instead that capital goes and builds data centers. So the first thing to keep in mind is this massive new demand for investment capital is probably pushing up interest rates all across the economy to reallocate capital to data centers. So we need to try to estimate that, because we need to know: are we stimulating the economy, or are we restraining the economy, or are rates at neutral? I think monetary policy is somewhere around neutral today, maybe slightly restricted, but that's at a higher rate level than it was before the pandemic, because I think that neutral interest rate has gone higher. So that's the first important factor. Second factor is long term: what does it do for productivity? If AI really boosts productivity structurally over an extended period of time, then we can grow the pie for the American people without having to worry as much about inflation. The non-inflationary growth rate of the economy will be higher if productivity growth is higher.
What are you watching when it comes to productivity? Because it's a volatile number, it can get revised a lot, but it has been there. It's about doubled from what it was running before this recent period. So how do you judge productivity? What are you watching?
I mean, the simple is that—and you hinted at this in the question—which is, we look at the economic growth, the GDP numbers, and we look at how many people are working, and how many hours they're working. And if businesses aren't hiring very many people, then the same number of people, but they're having big profits and big earnings and big economic growth—the same number of people are producing more, which suggests their productivity is going up. But as you also said, these numbers tend to get revised, and sometimes it takes a long time to really see the underlying trend, because the numbers are volatile. So sometimes it takes a long time to see the underlying trend in productivity. And so, yeah, in the last couple years, productivity has kind of surprised us, but I think it's too soon to say that AI is already leading to a boost in productivity. It could be that coming out of the pandemic, businesses couldn't find workers, and they had to restructure how they did their basic operations to just do more with the few workers that they had. That's productivity enhancing, even before you get to AI.
