At Statecraft, we’ve talked to several economists about working within the federal government. Those conversations have largely been about how their work is structured. But I also wanted answers to some other concrete questions I had: about stimulus packages, inflation, labor force participation, and demographics.
So I spoke to Ernie Tedeschi, former Chief Economist at the Council of Economic Advisers (These days he’s Director of Economics at The Budget Lab).
We discussed:
“Fighting the last war” in stimulus packages
What’s driving inflation
The tools CEA economists have, and the tools they wish they had
Are high interest rates driving low consumer sentiment?
The demographic trend toward lower employment rates
Why men are dropping out of the labor force
[NB: This interview was recorded in late April.]
Let me present you with a recent quote from Ezra Klein. He says:
“My sense is economics has weakened quite a bit as a public player in the past couple of years. Not just, by the way, in the Biden administration, but in the Trump administration too.”
He goes on to suggest that economics doesn't speak as a discipline with as clear or unanimous a voice as it once did and that political circumstances have changed.
Is that true?
In some respects, yes, in some respects, no. I think in certain areas economist voices have gotten through more, are louder, and are having an effect. For example, when you look at the response of both fiscal policy and monetary policy to the pandemic: that response would not have been nearly as swift or decisive if the economics discipline had no voice. Coming out of the Great Recession, there had been a lot of circumspection about the response in 2008/2009 and the subsequent very long recovery. That was a sluggish recovery for the first five years of that cycle, and really didn't start accelerating until 2014 or so.
There were lessons coming out of that. You saw, mostly for better, sometimes for worse, the pandemic response was really fighting the last war. In some ways, it was the sort of Great Recession response that a lot of economists wish we had done.
The way in which Klein might be right is, I do think that there was this infatuation with economists maybe for the first decade of the 21st century. It reminds me a lot of defense experts in the Vietnam War. There was this sort of technocratic fascination with them, that they could solve a lot of these problems.
I think economists obviously have a lot to say about policy. But we are not the only people, we are not the only lens, we are not the only way of thinking in the room. That's something that working at the White House really brought home for me.
I worked at the Council of Economic Advisors. I was surrounded by economists. But when I was collaborating within the administration, it was lawyers and engineers and climate scientists and different people who often spoke a very different technical language and different but extremely valid ways of looking at a problem.
And yes, there was a political perspective too. Politics has become a bit of a pejorative term, the way it's commonly used nowadays. But at its best, politics is a responsiveness to the way that people want to order society. And in a democracy, politics in its purest form is the expression of the will of the people. So yes, the perspective of people on the ground should absolutely hold sway over economic debate.
One last thing I'll say on that, just to make it real: in macroeconomics, one way that we measure trend inflation, which is where we think inflation might go over the next 6 to 12 months, is core inflation. And core inflation is inflation, taking out food and energy. The reason we take out food and energy is that historically food and energy prices have been volatile. If gas prices spike, they're probably going to come back down relatively quickly, so that spike doesn't really signal much about where other prices are going.
[For more on the nature of oil markets, see our recent interview with Skanda Amarnath and Arnab Datta, below]
It's a very standard metric in macroeconomics. But if we get too caught up in that, we forget that food and energy are extremely important to real people. We can't get too hung up on idiosyncratic things that make sense to economists, but that don't speak to the needs of real people.
Two follow-on questions: How much of the “fighting the last war,” as you described on the economic policy response to COVID, is attributable to economists?
And are we getting better at responding in a more calibrated way to financial crises since the Great Recession and COVID?
Let me tackle the second one first. I think the answer is yes. The economic toolkit that we have for fighting financial crises is undeniably broader than it was 14 years ago. Maybe not intellectually broader: there were creative thinkers 20, 30 years ago thinking about ways to do this. But we've actually tried many of these things now and we've gotten comfortable doing them. And look: that is thanks in large part to economists. Ben Bernanke in 2008, thinking, “Okay, in this financial crisis, cutting the overnight rate to zero is not enough. What unconventional policy can we introduce that will ease conditions more?” Quantitative easing (QE), for example, or forward guidance. It took the Federal Reserve a little while in that crisis to get comfortable with them, but they eventually introduced them. Fast forward to the pandemic and the Fed cut to zero, started QE and had issued forward guidance in March of 2020, before we had a single bad piece of unemployment data. That's how rapid the response was, because the Overton window had been widened already.
I think that there has been a conclusion among economists that the response to the Great Recession was too tepid and that we should not make that mistake again. The size of the CARES Act and the American Rescue Plan was, in large part, driven by that perspective from economists.
I think it was also driven by the fact that this was an exogenous shock. Unlike the financial crisis, where you could at least make a moral hazard argument, nobody thought that COVID was anyone's fault. Especially with the CARES Act, that led to bipartisan support that is unusual for this sort of fiscal response.
A great deal of the sorts of things that were done, the unemployment insurance, the PPP loans for businesses, were ideas that were in many ways informed by economists.
CPI and PCE are two different metrics of inflation. I've got a tweet from you here from yesterday [April 10th, 2024]. “It's fairly clear now the CPI is signaling some recent stubbornness and disinflation.” Walk me through that as a layman.
So we have two different inflation measures. I don't know what the exact ratio is, but they share a great deal of data. They generally follow one another very well. The Consumer Price Index is much more well-known: It comes out earlier in the month, it gets a lot of press. But in some ways, the Personal Consumption Expenditures price index, PCE, is more important, because that's what the Fed follows for its two percent target.
Yesterday we got CPI for March. When you look broadly over the last two years, CPI inflation has come down. It peaked at just north of 9% in the middle of 2022. Now it's at about 3.25% core, which I mentioned earlier, the metric that economists like to look at for an idea of what trends are going on.
But that core measure ticked up faster than expected just for March individually. Now, normally you would not want to pay much attention to any one month and any sort of outlier in a month. The Bureau of Labor Statistics that collects these data, they are economic heroes. They do the best job they possibly can. They think about this all day. But at the end of the day, they're issuing a survey and collecting prices, and there can be noise in that survey. So you never want to focus too much on one month.
However, February and January were a little bit hotter than expected as well. So once you have three months in a row where you're having these upside surprises, then you need to rethink your priors and your assessment of the economy. Before January, it looked broadly CPI and PCE inflation were coming down. We were getting closer and closer to the Fed's 2 percent target. We expected that, basically by the end of the year, we were going to get there, and that it was safe for the Fed to begin cutting rates this year. Now, at least in CPI, it seems like we're stalling out on core, and that's raising some worries that the last mile of disinflation might be harder than we thought. I am worried about that.
I'm a little bit less worried than some folks, though, because the primary reason inflation is still high is very different than when inflation was hot in 2022. In 2022, all prices were hot and all prices were accelerating. Now, it's really just one or two big major components. It's almost entirely housing and then a little bit of auto insurance, especially in CPI. And those are both real but esoteric categories. In particular, with the way we measure housing, there's a lag between what we see on the ground with market data and then incorporating that market data into these official surveys, both CPI and PCE.
We know from market data, put out by places like Zillow, that rents are growing, but they're now growing at a normal, pre-pandemic pace. It takes about a year for that normalization to work its way into the data. And we haven't had a year of that yet. There's good reason to think that's in the pipeline.
That said, the Fed is not in the hopium business. They are in the “We need to actually see this in the data” business. I get that, they are nervous.
To be clear, I'm still a little bit nervous about it too. The dynamics of housing could very well be different than they were pre-pandemic. We may have more housing churn in equilibrium than we did before, for example. Maybe with working from home, the demand for space is just greater now in equilibrium than before. Maybe monetary policy is less effective than we used to think it was.
As you look at this month-over-month data at CEA, some decisions shouldn't be made based on individual months.
How does CEA think about recommendations in a month-to-month sense?
So our job at CEA is to make sure that robust economic data analysis and research are part of the policy conversation. The goal is not to make sure that economics is the paramount criterion in any decision that's being made. We just want to make sure that it's part of the considerations that are made; that when policymakers get together in the administration, they at least understand the economics. Then they integrate other considerations and, if they go against the economics, that happens. We just have to make sure that they understand the economics.
When it comes to monthly data, our basic job is to make sure that the president knows any new data that comes out month-to-month. And so we literally have to tell him, look, we added this many jobs, inflation came in here, expectations were here. There’s a sort of mechanical month-to-month update of the data in there.
When we think that there is a broader story going on, we try very hard to pivot away from the trees and toward the forest, and show three-month moving averages or year-on-year growth rates or whatever we feel is a better metric of the broader contours of the economy. I found that one of the most fun and challenging parts of the job. You are both a spreadsheet person and a storyteller at the same time. You have to be good at both of those skills.
In my circles, you're known as notoriously careful with data, what it shows and what it doesn't.
Oh, thank you.
I asked around. You also spend time in the private sector. Tell me about the data that you get access to at CEA, vs. the data tools you use in the private sector. Are they different?
Very similar, actually. So for example, in both places, we use a tool called Haver Analytics, which is if you've ever used the St. Louis Fed’s FRED website [Federal Reserve Economic Data], it's a hyped-up version of that with much more data. The big difference with CEA is that if you work on data memos for the president, you get the data the night before, on a confidential basis. So that's one difference. But no, it's a lot of the same tools.
CEA traditionally is a place where academics go. Generally, senior economists at CEA are tenured academics who will take a year or two of leave and do a rotation at the White House to advise.
Academics are brilliant, and what's great about CEA is it's one of the few places in the government that can bring that skill set into the policymaking process. One of the challenges is that academia has a very different cadence than the speed of government. I know that people like to stereotype government as being this lumbering mastodon that moves very slowly. But I guarantee you, the White House is a mile a minute all the time.
That can be challenging coming from academia to that sort of environment, where you don't have a year to ruminate on a question and to do umpteen iterations of a paper to get the graphs right. You have two days to give it your best shot. That's very much like the private sector, One of the reasons I think I was hired at CEA is because of my private sector background, and I think that made the transition for me a little bit easier.
You shouted out the Bureau of Labor Statistics a moment ago. Martha Gimbel said something similar in our interview. If it was up to you, what additional data would the federal government collect, whether through BLS or elsewhere?
The two things that were the most challenging for me at the CEA were consumption data and savings data. We do have some consumption data. There's the Consumer Expenditure Survey that the Census Bureau releases. It's a survey, people fill out diaries of what they buy. It doesn't come out with nearly as high a frequency as the consumption questions that we have and that we get.
I found that I really wanted scanner data on what people were buying at grocery stores, to get at questions about what prices people were actually seeing. How are they responding to those prices? Are they buying smaller things? Are they going from name brand to store brand? Are they buying less? What is going on here? Just, in general, beyond inflation, you get a check from the child tax credit. What are you spending that on? That would be extremely helpful, whether it were a government version or us working with a private provider.
The other thing I mentioned was savings data. The best savings data that we get comes from a high-quality survey that the Federal Reserve does, called the Survey of Consumer Finances, but it only comes out once every three years.
Particularly in this cycle, there were questions all the time about, “How much more savings do people have? Are they spending down their savings?” There were several months where we had very strong aggregate consumer spending data, but we didn't have great real-time data to answer that question. Some private vendors, like the JPMorganChase Institute, ended up running some numbers using JPMorganChase accounts to show checking account balances over time on a higher frequency basis, and that was very helpful to us. But then, that raised questions, like, “What about the unbanked? Is this a self-selected sample?” Things like that, I found myself begging for.
[For our interview with former CEA Senior Adviser Martha Gimbel, see below]
We spoke to Ronnie Chatterji, who was at the National Economic Council. And he talked about wanting a very different pile of data about supply chain dynamics, a totally different world of economic data. Talk to me about the difference between the NEC and CEA.
[For our interview with Ronnie Chatterji, see below]
Yeah, sure. That’s the number one question I get. There's no reason people should need to know that on the outside. But think of CEA as the think tank of the White House. And then think of NEC as the traffic cops who manage the policy process.
There are legitimate experts and extremely smart people at NEC. NEC is led by an economist right now, Lael Brainard. But NEC is generally not full of economists per se. It's mostly staffed by lawyers who focus on economic policy. Their job is really to manage the interagency. What I mean by that is a question comes up: let's say it's forgiving student loans or something. Lots of different people within the federal government, different agencies, different stakeholders have expertise, have thoughts, have a stake in that question. NEC’s job is to convene all of these people.
So in the case of student loans, that could involve CEA because we have education economists. It could mean the Department of Education, it could mean the Treasury Department, because we're talking about debt that's being managed by the federal government.
And NEC itself will have thoughts. The president's chief of staff will have thoughts. The Domestic Policy Council might have thoughts. Others will want to weigh in as well. NEC's job involves very much an element of art, of people management, so that we can reach consensus and something final can be brought to the president's attention for approval.
Has CEA become more political over time?
Some CEAs have been more political than others. I think both Cecilia Rouse and the current chair, Jared Bernstein, have been very good about emphasizing the rigor of economics, because that's really what CEAs value add is: to be absolutely clear and transparent. That said, our advice is independent, but we are part of the administration. We work for the president. Our whole job is to make sure that the president understands and sees economic data, research considerations. CEA is not like, say, the Congressional Budget Office, or an independent regulatory body.
But the advice CEA gives is meant to be reflective of the state of economic research. In that sense, it is independent. Some CEAs have been more political than others. Alan Blinder had a really interesting book on the history of fiscal and monetary policy in the United States in the post-WWII world. What is striking is actually how less in control of the policy process CEA has become over time. Remember, the NEC was a Clinton administration creation. So many pieces of economic policy before NEC — the one that comes to mind is the Kennedy tax cut — for example, were the brainchildren of CEA. That role of CEA has faded over time.
We still give advice and do analysis, but fewer of the actual grand policy ideas come out of CEA. More and more come out of NEC. So in that sense of political, CEA has become less political.
CEA reports, in my experience, are written very clearly. They're readable. Do people read them?
I don't know. If people read The Economic Report of the President cover to cover, I don't know how I would feel about those people.
I am above the median American in nerdiness and wackiness, I think it's fair to say. Before I joined CEA, I would get The Economic Report of the President. During both the Obama and the Trump years, I'd flip through, and if there were interesting charts or interesting sections, I would dig into those charts and read those sections. But I would not read it like a novel.
There have been some interesting reports that CEA produced that I thought were well done. So I remember under the Jason Furman CEA, they produced a report on long-term interest rates. It was a 30-page report, but it was extremely clear for a topic that is often muddled and vaporous and hard to get your head around. That report also got praise from people from across the political aisle. John Cochrane, who was not an Obama supporter by any stretch of the imagination, praised that report and said they really nailed the economics.
There's obviously a cluster of intellectual interest, whether you call it “Econ Twitter” or something else, that pays attention to what CEA does.
Do things like Jobs Day discourse shape how you think about creating products from the inside?
So yes and no. Take our Jobs Day tweets or our inflation tweets, for example. In crafting the template for those tweets, we did think about what metrics seem to resonate with people. That was definitely a factor that we took into account when we were composing them.
But we also took pride in the fact that we generally would follow the same template for that tweet thread, regardless of whether the data were good or bad. We saw that as an opportunity to communicate the seriousness of CEA and to teach the public a little bit about the data.
We had a couple of bad jobs reports and a healthy mix of good and bad inflation reports, and our Twitter thread was the same format, same charts, telling it like it was. I took pride in that, that CEA did it that way.
To go back to your point about the relationship between inflation and housing, how does CEA think about something like that, where supply dynamics are outside of the Fed's control?
Yeah, it's a really good question. Broadening this a little bit: It has definitely not been the case that the Federal Reserve has always been independent of political influence, or that the administration has refrained from talking about monetary policy publicly.
I think this CEA did a solid job from Day One to lay out the expectation that part of our inflation-fighting strategy in the administration was to give the Fed room without political pressure to do their job as they saw fit. We reiterated that constantly.
Just to cut you off there, is that an institutionalist mindset, that both institutions work better when they keep to their lanes, or is it a pragmatic decision?
It's a bit of both. There are quite a few Fed detailees who come to CEA for a small amount of time, so we know what the Fed is thinking about, or at least how they think about things.
And I guarantee you that anything that CEA thinks on inflation, the Fed has 10 times as many economists who have already thought about it. This is the Fed's job, right? And they are not clowns. They are serious professionals who think about inflation constantly.
So both the institutionalists and the practical mindset was we need to give them the flexibility to be able to do what they see fit.
And then, sorry, where was that question leading?
We were… I cut you off and both of us forgot.
Oh, housing. How do you think about it at CEA?
Housing is funny, right? Because on the one hand, housing is the textbook example of an interest rate-sensitive sector. You think about where monetary policy would have an effect, and housing should be top of your list, because virtually everybody takes out a mortgage to be able to buy a house. And the interest rate is an important consideration.
On the other hand, there is a longstanding supply issue in the United States that predates Biden and predates Trump. There’s a structural shortage of housing in the United States. The simple answer to your question: at CEA we were Fed watchers like everyone else.
We had opinions, but we didn't have any insider information about what the Fed was going to do. We were watching the press conferences and reading the private sector notes and talking amongst ourselves about what we thought the Fed might do, and then seeing if we were right or wrong. On the substance of housing, we tried to understand what the data showed, and then if there was anything outside the realm of monetary policy where we can move the dial. Of course, that's very difficult without legislation, but sometimes there are tweaks to policy that an administration can make to ease that burden.
In 2022, the Biden administration announced a whole package of legislation it was going to push for, and then tweaks to housing policy on the margin to try to encourage more supply. That was very much informed by economic analysis, looking at the importance of housing and inflation and convincing policymakers in the administration that housing is an inflationary issue.
You've been outside the CEA, inside, and outside again. Are there things structurally that CEA could change to be more effective in its role?
That's a great question. One thing that this CEA did well that I would encourage in future CEAs is to bring in many different types of economists. There are academic economists, there are government economists at places like Census or BLS, there are private sector economists, and there are think tank economists. They all do economics, but they have different skill sets and are used to thinking about problems at different cadences, different levels of depth versus breadth. This CEA had a good mix of all of those types of economists.
We had the perspectives of academic economists who would normally think very deeply about a problem for a very long amount of time, and knew exactly how to structure an analysis to robustly measure something hard to measure, and what controls to throw in.
We had government economists who knew official data extremely well, had contacts at other agencies, and could get questions answered for us. We had private sector economists like me, who knew how markets were going to approach a question and what metrics they were looking at, and think tank economists who often had the skill to be able to communicate with the broader public, which is not always a skill that economists have.
Can I push you one more time on the question, which is about changes to CEA? That's a great answer, but I'm wondering what the CEA is not doing that it should be doing.
Oh, I see. So I think CEA should do the same thing that I think a lot of economists could do more of. CEA needs to get out of its comfort zone sometimes and find forums or avenues or mechanisms for talking with folks who are not normally the types of people that economists talk to.
We get every private sector note on Wall Street, right? We see every think tank report that is published. Those perspectives are very well represented in CEA. What is not as well represented are the perspectives from community groups, churches, local nonprofits: people on the ground who maybe don't think about national politics often, but who are deeply involved in their communities and who might have a better perspective.
When consumer sentiment was low, even though inflation was falling, the question arose in the administration: what is going on? I think that if we had talked to everyday people, we would have arrived at what I think is the right answer more quickly. We eventually got there, but we would have arrived at the right answer more quickly. When you put inflation in a model and price changes come down, your model gets confused when sentiment is still low, but what everybody is complaining about is high price levels, right? Especially when they go to the grocery store or the gas pump. And we need to find a way to integrate that into our models.
Is your view that low consumer sentiment is about levels primarily, and not, say, high interest rates?
Yeah, a little bit. I should emphasize that it could very well be an “and,” not an either/or, in this case. There's not a consensus. There are lots of different ways of modeling this. I will tell you that I find the interest rate hypothesis very plausible. It just makes a lot of sense.
However, I've dug into the micro data on sentiment, not just the aggregate numbers, but the individual responses to sentiment. And I’ve got to tell you, it's really hard to parse out the effect of interest rates. And I think the reason why it might just be a data reason is that the Fed raises interest rates in part when inflation is hot, right?
So once you control for inflation in your model of sentiment, you've gotten a lot of that effect. When I was modeling sentiment, I put in inflation, I put in price levels, put in the unemployment rate, the stock market, housing prices, whatever. And what I found was there was no variation left over to be explained by interest rates. Every which way that I put it in, it wasn't significant.
That doesn't decisively mean that interest rates are not important. Again, personally, that doesn't make sense to me. There must be an effect on the cost of capital. The point is that it's really hard to tease that out in an econometric setting.
One very specific version of that argument you often hear is, “Oh, people are down on the economy because it's so hard to buy a house. It's not just that interest rates are high, but that housing prices are high.”
And that's really a double-edged sword that is much clearer to see in the data. I have no doubt that if you are young and don't have a house yet, you probably hate the housing market right now. It's hard to get in.
But most Americans own a house. For those 60-odd percent of Americans who own, housing is an asset. When housing prices go up, they love that. That's one of the most robust findings in the sentiment literature: housing is always, no matter how you put it in, a very strong positive on sentiment. It’s unfortunate for the portion who don't own a house, who were probably pissed about the housing market But that's one takeaway from that.
Let me close with some demographic questions. The American working-age native population is shrinking. You've talked about this. What does that mean for economists advising policymakers?
What's going on is that the native-born population in the United States is aging. As time goes on, more and more of the native-born population is retiring. The baby boomers are retiring. It’s one of the least surprising trends in economics: demographers are much better forecasters than economists are.
I think we get a little too obsessed with the idea of a growing population and a growing labor supply in existential terms. The United States would be able to get by with a shrinking population and a shrinking labor supply. I guess it literally can't go on forever, but it's not an existential crisis. It makes things like funding pensions for seniors, providing care for seniors, being able to fund the things that we want in this economy, opening new businesses, expanding existing businesses, all a little bit harder. If we want to make those things easier, we need a growing labor supply.
And that's what we've been getting, through immigration. And this is separate from the question of what is the appropriate border policy? How should we handle asylum seekers? Very important question and very much worthy of debate. But take a step back and just think about the economic context of all of this. If we had not had immigration after 2019, our labor supply would have shrunk, and that would have made all of those things I mentioned before incrementally harder for us. But the fact that it expanded means that we can sustain 300,000 jobs a month in growth, which is a good thing, right? We can open new businesses, we can expand existing businesses and that doesn't just benefit immigrants, that benefits native workers as well, because it means that they can get more of the goods and services that they want and just have an economy that provides for them better.
Let me throw you one more demographic question. I'm looking at your NPOP piece.
Oh, God, that's old.
I've got it here.
Let's go!
It’s a metric that tracks the percentage of the population not in any full-time or part-time employment.
Yeah.
So that trend's been going down. Fewer and fewer folks are employed in some capacity. I'm looking at a couple charts of employment rates of men and women in the U.S., disaggregated. The time series I have runs all the way into this year, and men have not seen a rebound to pre-COVID levels, in the way that other parts of the recovery have rebounded. What should economists make of that stalled-out employment rate for men?
I think aging is an overlay to all of these things. Men and women are getting older overall. So when you're looking at the entire population, we should expect their employment and their labor force participation rates to be structurally falling, because they're all getting older.
There will be variation in the business cycle, but the overall labor force participation rate is falling at a structural rate of about a quarter point a year due to aging. There are a couple of ways to adjust for that, but the simplest way economists do that is to look at just prime-age Americans, which is 25 to 54-year-olds.
It's nice because it excludes the very young people who tend to be in college, and older people who tend to be retired. The recovery for both prime-age men and women in both EPOP and LFPR has been quite strong in this recovery, relative to prior U.S. recoveries. That said, men and women both have challenges, albeit some different challenges, right?
So women have the challenge that the United States still has relatively weak family policy relative to other countries in the world. That gap used to not be that great. And so when you looked at female prime-age labor force participation in the United States, it was quite high relative to the rest of the advanced world around 2000 or so. But that family policy gap has widened over time. And so now we are not just in the middle, but toward the bottom of advanced economies in terms of female labor force participation rate. In fact, some historic laggards like Japan are now outpacing us in female labor force participation because they have introduced policies that are more geared toward bringing women into the labor force.
Men have a separate set of challenges that are very long-term, beginning in the 1950s, which is the decline in manufacturing employment in the United States. There is a strong interaction between education and labor force attachment for all people, but especially for men. For men without a college degree, it's been particularly challenging over the long term.
Now, Americans without a college degree have done relatively well in this pandemic cycle relative to other cycles. Wages for people without a college degree have done extremely well. I think in large part because of the shortages that restaurants had, they ended up having to pay higher wages. People on the low end got a lot of benefit from that.
But in the wider lens, looking over the last 50 to 60 years, there's very much a downward trend in male labor force participation. And this interacts with other social phenomena, including the marriage rates and marriageability of men.
So all of those things have to be taken into consideration. It's a different set of challenges that face women, but they're both challenges that policymakers need to pay attention to.
Share this post