Skip to Main Content
Speaking of the Economy
Kartik Athreya, the Richmond Fed's research director, answers listeners' questions
Speaking of the Economy
Dec. 20, 2023

Ask the Richmond Fed

Audiences: General Public, Economists, Workforce Sector Leaders

Kartik Athreya, the Richmond Fed's research director, answers questions submitted by listeners about housing, the Fed's bond purchases, interest rate targeting, and artificial intelligence.


Tim Sablik: Hello, I'm Tim Sablik, a senior economics writer at the Richmond Fed. My guest today is Kartik Athreya, executive vice president and director of research at the Richmond Fed.

Kartik, welcome back to the show.

Karthik Athreya: Great to be here, Tim.

Sablik: We've got a special episode for listeners today. We've asked you and other members of the public to submit questions for Kartik to answer on the show. We've picked a few on a variety of topics to tackle today. So, without further ado, Kartik, are you ready?

Athreya: Yes, I am.

Sablik: The first question we have is about housing. Housing has become increasingly unaffordable for many Americans over the last two years. What is the Fed doing on this issue?

Athreya: Housing has become quite expensive relative to other things that we all buy and need. That is obviously a problem. It's a problem for the vitality of cities, of towns where workforce housing has been cited in our own exchanges with people in [the Fifth] District. We hear this all the time.

As far as what we are doing, the Federal Reserve by itself doesn't have a direct responsibility or the ability, for that matter, to really govern house prices and their movements as we would wish. We instead have an inflation mandate and an employment mandate.

That said, we as a regional Fed, of course care about what is happening in our District. And one of the things that we've tried to focus on for the last several years is trying to understand the roots of what drives housing prices. Our president, Tom Barkin, has spoken about this in speeches that he's given recently.

A long list of suspects, of course, can be determined. They raise questions about the way in which societies agree to use land, the way societies agree to have policies that ultimately attract people to live in close proximity to each other. So, we are learning things about housing. But directly we are not an entity that is charged with or capable of moving house prices.

There are times when the economy is booming where we might think that monetary policy needs adjustment. And those are times when you might get housing prices being high at the same time that you have other things that the Fed is on the hook for, like inflation, also being high. So, we are connected in that sense to housing prices. But we do not directly control them.

Sablik: We had another question asking whether you could explain the Fed's involvement in buying bonds, how that's changed in recent years, and what effect that has on long-term interest rates and mortgage rates.

Athreya: This is a really important question. I'm glad it was asked.

Let's for a second cast our minds back to the period before the Great Financial Crisis, where the asset holdings of the Fed involved having a relatively small amount of U.S. Treasury bonds. What we would do, roughly, was buy and sell those small quantities in order to make short-term interest rates close to what we would target to have for the economy as a whole.

In the Great Financial Crisis, however, this changed. We went from using relatively small amounts of government bonds and relatively small transactions in those markets to target interest rates to generally holding a very large quantity of government debt and other kinds of debt, including mortgage-backed securities. We had to find ways to provide accommodative policy at a time when we faced other constraints on our ability to change the short-term interest rate more directly.

Now, the second part of that question was about how that, by itself, may have impacted long-term interest rates and mortgage rates. Well, in the Great Financial Crisis, part of the goal was to allow us to have a little more influence on longer term interest rates in a situation where we couldn't change the short-term interest rate very directly. This is something that people call the effective lower bound or the zero lower bound constraint on our ability to set directly the short-term interest rate.

So, our response was to try to move longer run interest rates down more directly by purchasing more long-term assets than we were before. That's another change. That, in turn, of course, affected mortgage rates and so on.

It's crucial to not think of that change in the way we implemented monetary policy as representing a change in what our goals for policy were. We did not move to that system in order to affect mortgage rates in a way we were not trying to do before. In the end, we're always trying to alter interest rates in ways that are consistent with our mandate. That never changed in this process, but the implementation changed very substantially pre- and post-Great-Financial-Crisis.

Sablik: Keeping on the theme and topic of interest rates, interest rates are higher now than they've been in over a decade. But inflation remains still a bit above where the Fed would like it. We had a question from one listener about whether the Fed might be reconsidering its 2 percent inflation target.

Athreya: Another good question, Tim. I'm not speaking for the Federal Open Market Committee, I'm an advisor to one of its members. I would say I've seen nothing to suggest that the Fed is actually reconsidering its 2 percent inflation target.

What's important here, Tim, is to think for a second about why 2 percent as opposed to other numbers that one might think about. The answer that we've come up with as economists and policymakers is 2 percent trades things off in a fairly nice way. There are pros and cons for higher and lower inflation, and this a resting place that is deemed to be pretty good.

A couple of reasons why. First, there is reason to think that strictly no inflation — or price stability, if you will — is very useful in any world where it's clunky to change the prices of the goods and services that we all consume. Doing that, however, requires us to really be able to measure prices and inflation more precisely than we actually can. One of the problems with many indexes of inflation — essentially all of them — is that they have an upward bias that is difficult to completely purge. So, there's a sense in which really low and positive rates like 2 percent represent something much closer to actual price stability without hitting something that is deflationary.

A second more substantive force, aside from the measurement issue that I just described, is that settings in which the inflation rate is really low are also settings in which interest rates normally are really low. For those of you old enough to remember, there were periods in the '80s where inflation was high and interest rates were high — mortgage interest rates were in the double digits, for instance. Turning to the more recent data, we see that with relatively low inflation until the COVID era, interest rates were awfully low.

I mentioned earlier the effective lower bound constraint that policymakers worry about, where they cannot reduce interest rates to negative numbers very easily. Well, in a low-inflation environment or a zero-inflation environment, that problem recurs more frequently. While economists still don't have a clear understanding of how the zero lower bound interferes with the behavior of the economy, there is sufficient concern to warrant — in the minds of many, including myself — that you want to try to keep an inflation target a little higher just to reduce the likelihood of the economy being constrained by that bound.

As for changing the target, which is what you asked me at the beginning, there is an immediate tactical issue for any central bank thinking about changing its target, and that's of communication. It's not easy to get the public to coalesce around a set of inflation expectations. Arguably, we've been incredibly successful for about 30 years now at having inflation expectations being low and stable at around 2 percent. Our president, Tom Barkin, has spoken of how this has really allowed the private sector as a whole to make decisions in a much more simplified manner than if they had to guess what our inflation level was going to be in the near future, or in the distant future for that matter.

The other thing that is true is that if you're going to reconsider a target, it's presumably a lot less challenging to communicate why you're doing it if it's at a time when inflation is right at target or really close to it. As you pointed out, Tim, inflation is not yet to target. There is a sense in which I think [changing the target] would potentially risk muddling the Fed's commitment to price stability more generally.

Sablik: Moving away from some of the more Fed-centric topics, we have a question here on your thoughts about how the development of AI might impact employment and the economy.

Athreya: This is a huge question. I think the honest, short answer has to be I don't yet know, and I think most people don't know yet.

What is true is we can learn something from the way in which important technological changes in our own past [and] in the past of other major economies in the world have ultimately led to changes in the lives of people. That historical perspective leads me generally to be optimistic.

I look at the enormous change in the way we produce goods and services compared to, let's say, 100 or 150 years ago or 200 years ago. Those changes — especially ones that involved industrialization and the move from the farm to the factory — were also huge. I think it's important just to have a sense of scale for how the United States between 1870 and 1940 went from being a heavily rural economy to being a heavily urbanized one, concentrated in cities [and] producing in factories.

Those changes also created huge ripple effects in terms of what else people needed. Most notable, in my mind, are the changes for the way in which people insured each other, the way in which communities protected their own members. We went from having that being done informally in agriculture, with large families and communities looking out for each other, to one in which people were relatively anonymized in very big cities.

These are very large changes, not just technologically but culturally and socially. But in the end, those were managed remarkably well. One summary statistic that you can look at is the unemployment rate. Quite remarkably, over the course of 100-plus years, tectonic changes didn't ultimately lead to vast pools of dystopian unemployed outcomes and so on. These things didn't happen. Instead, what happened was that the unemployment rate stayed flat. We got vastly richer — on the order of six, maybe eight times as rich per person as we were 100, 150 years ago. And the news has been very good.

Sablik: That future sounds pretty good. But what would you say to people that are maybe a little more pessimistic?

Athreya: I think the case for pessimism rests in the fact that this does look a little different in some important ways than before. People in our own District, including Anton Korinek, have given this a lot of thought and I would urge you to kind of take a look at what he's written. He's a professor of economics at UVA and a friend of our department.

Those who are more pessimistic are worried about the fact that previous changes in technology that have come along have largely been able to augment the productivity of workers, especially helping us deal with things that are routine and often noncognitive in their nature. AI almost looks like a giant group of superhuman beings that are going to be present. There's no reason to think that, as a competitor-type of labor force to something that can do everything better than you and I can do, we ought to come up roses in that scenario.

Now, let me give you one last reason to be positive. In a world in which AI can do everything better than you or me or everybody else can, what's clear is that society as a whole can have more of everything. The pie almost necessarily has potentially gotten way bigger. Then it becomes about being creative and building the societal consensus necessary to have everybody come out of that with a bigger slice. That's a politics problem, that's a political economy problem. The way that's solved will ultimately hold the key to whether, even under the pessimistic view, we all come out ahead or not.

Sablik: Well, Kartik, thanks so much for joining me today and sharing your optimism.

Sadly, this might be one of the last times I get to have you on as a guest of the show, as you will be transitioning to a new role as director of research and head of the Research and Statistics Group for the New York Fed. Reflecting on your time as research director here at Richmond, what are some of the things that you're proudest of?

Athreya: Well, Tim, it's bittersweet for me, too. I think it'll be very hard to leave behind a place that I've called home for more than two decades now. It's been an amazing experience, and I thank everybody in the Bank [and] in the department for helping me survive and then thrive as a colleague.

As for the things that I'm proudest of, a couple of things come to mind. First and foremost, we're a department that really operates as one cohesive unit, that sees the mission of the Research department to really matter in economics as a profession [and] to matter in the District and in our region. These are things that I feel have permeated all of us. We do not speak frequently as only economists or only economics writers or only community development specialists. I think that's been a very real cultural achievement.

The second thing that I would point to as being really proud of is the way in which we have very smartly used hierarchy, if you will, as an aid to make decisions and not as something that is important in the deliberations that we're engaged in. I think hierarchy has a place to facilitate efficient decision making — somebody has to make a call on something [and] it's better that that person understand that they have to, it's better that everybody else understand that they're responsible for making the call. But that's about process for the final decision. It is not necessarily, at all, about how you get to the right decision. That's about deliberation and deliberation requires all hands on deck.

Beyond those two cultural attributes, I think that I'm very proud to be a colleague of the people here. I've seen people take their jobs really seriously [and] try to do good things all the time. To me, that's been the kind of thing that's put a good kind of pressure on me when I get up in the morning to say, "Well, everybody else is trying hard. What are you doing?"

Sablik: I'm sure I speak for all the researchers here at the Richmond Fed when I say thank you for all that you've done. We'll certainly miss you but wish you the best of luck in New York.

Athreya: Thanks, Tim.

Sablik: Well, that's going do it for today. If you enjoyed this episode, please consider leaving us a rating and review on your favorite podcast app.

Phone Icon Contact Us

Research Department (804) 697-8000