Normally, economists expect a somewhat stable relationship between job openings and the unemployment rate. More job openings = more people employed. Lately, however, the shape of this relationship has changed. Job openings are absolutely soaring. And yet total employment in the economy is well below pre-pandemic levels. On this episode, we speak with Thomas Lubik, a senior advisor in the Research Department at the Richmond Fed, who has been researching and writing about this unusual state of the labor market.
Tracy Alloway:
Hello and welcome to another episode of the Odd Lots podcast. I'm Tracy Alloway.
Joe Weisenthal:
And I'm Joe Weisenthal.
Tracy:
So Joe, I know we've been spending a lot of time on the question of inflation and whether or not it's transitory, but I feel like a lot of that discourse is sort of happening at the expense of a greater focus on the labor market. And I know that sounds like a weird thing to say, but if you think that what the Fed is saying right now is that they're going to keep rates very, very low until the labor market fully recovers or recovers even more, then really we should be digging into the labor market and what full employment actually looks like.
Joe:
Right. And I think there's two questions — or there's a million questions — but a) as you say, what does ‘full employment’ or maximum employment, which the Fed has established as a precondition for rates lift off, what does that look like? That's one thing. And then 2), why have we not seen faster job growth? And that seems like a funny question to ask because the labor market growth has been incredibly fast since last year. Nonetheless, we do seem to be in this weird mismatch where there's lots of job openings and lots of people who are not employed and what are the reasons that people who left the labor force a year ago haven't come back yet.
Tracy:
Right. This is the mystery of the labor market at the moment. So on the one hand, unemployment is higher than it was before the pandemic. I think we're still something like 4 million jobs short of where we were back in February 2020. And if you look at, you know, the line of where we would have been had the pandemic never actually happened, I think we're about 7 million jobs short. And yet at the same time, you have a lot of companies — and, you know, we've spoken to at least one of them on the show — talking about the idea of a labor shortage, that they can't get the right workers in the jobs that they have open at the moment. And it turns out there's really a perfect economic principle to capture this sort of tension between job openings and the unemployment rate. And that is something called the Beveridge Curve.
Joe:
Yeah. You had a great post on this recently.
Tracy:
Thank you. So the Beveridge Curve is basically the relationship… That's all I wanted you to say, Joe,
Joe:
Yeah I know, I just wanted to kick it back to you.
Tracy:
Thanks. It's the relationship between the unemployment rate and the job opening rate. And, you know, normally if you look at the Beveridge Curve in a usual business cycle, it would be expected to move in a sort of counterclockwise loop. So as the unemployment rate initially jumps, you would expect the rate of job openings to stay very small and then sort of gradually recover and start moving to the left as the economy healed. Spoiler alert, that is not what has happened in this particular business cycle. Instead, we've seen something that is normally a curve — you know, the clue is in the name, the Beveridge Curve — it's basically morphed into an up and down line, meaning that unemployment basically isn't moving even as the number of job openings is going higher and higher and higher. So something really appears to have changed in the labor market here. And I am pleased to say, we have the perfect person to discuss all of this. He's actually the author of a recent bulletin on the Beveridge Curve on this exact topic. We're going to be speaking with Thomas Lubick. He's a senior advisor in the research department at the Richmond Fed. So Tom, thanks so much for coming on!
Thomas Lubik:
Okay. Thanks so much for having me and thanks for the kind introduction.
Tracy:
Yeah, I was really interested in this paper. Maybe just to begin with, you could lay out, you know, what is the Beveridge Curve and how would you expect it to act in normal times?
Thomas:
So I think you've already introduced the concept of the Beveridge Curve perfectly. So just to restate what you discussed, so the Beveridge Curve is the relationship between the unemployment rate and the job openings rate. So job openings are open positions that businesses want to fill and are looking to hire for. And this relationship is — it's a negative relationship. So when the unemployment rate is high during a downturn, job openings are low because well, the economy is not working very well. And firms are reluctant to hire new workers because of the uncertainty of how the economy might improve. But as the economy improves and the downturn turns into an upturn, the unemployment rate falls because the job openings are being filled. And as the economy improves firms posts more open positions. So the job openings rate rises as the unemployment rate falls. And this is what we see in every recession.
And the Beveridge Curve basically describes this relationship. So you can think of it as a scatterplot of monthly data on the unemployment rate and the job openings rate. And they line up nicely around this downwards sloping Beveridge Curve. Now, what is absolutely striking in the Covid recession and Covid recovery is that the Beveridge Curve that we've seen in the last 20 years pre-Covid, has changed dramatically. And in two ways. It has shifted outward, and it also has become much more steep. So the outward shift of the Beveridge Curve, that occurred right after the Covid shock hit. So essentially starting in March, 2020, April, when the unemployment rate shot up, and then declined very, very rapidly once the economy worked itself through the initial Covid shock. But this is connected with an outward shift in the Beveridge Curve in the following sense. So any given unemployment rate in normal times, you can derive a job openings rate that is required to maintain a certain employment level, or unemployment level. Now during the first few months of the Covid shock, the unemployment rate stayed high, but declined, but so did the job openings rate, but the job openings rate was at a level that actually we hadn't seen before for this given unemployment rate. And this has been quite striking.
Joe:
I want to obviously dive into why this relationship is the way it is right now, but actually before we get to that, I actually have a more straightforward question, which is how is the job openings rate derived? Where's that number come from? Because with the unemployment rate, you know, you imagine you ask a thousand people, what percentage of you want a job and don't have one now, etc., and you can come up with one. How do they determine the job openings rate?
Thomas:
So it comes from the so-called Jolts data. So the job openings and labor turnover survey that the Bureau of Labor Statistics conducts and they ask businesses a set of questions, where they really dig deep into what a job opening is. So a job opening is defined as a position that is immediately ready to be filled for which there's funding and for which firms actively engage in recruiting efforts. And the survey goes out to businesses and they report their job openings to the Bureau of Labor Statistics. And this then ends up as the job openings rate. So both total level, but also relative to the labor force, for instance.
Tracy:
So I'm gonna jump to the next obvious question that Joe just kind of foreshadowed, but what do you think is happening here? So, you know, you mentioned this is sort of an unprecedented move in a relationship that has historically held across many different types of recessions in many different countries. What's your theory about what's happening?
Thomas:
So as you know, we've seen historically high job openings rates. So we hit 7% in July, it’s come down a little bit from this and the job openings, the level, is about 10 and a half million when total unemployment is around 7 million. So there's a huge gap and this is the highest gap ever measured. So at this point, I think it is fair to say that we don't have one clear explanation of why the job openings rate is so high. I think it's a combination of a variety of things. We have seen — even before Covid — relatively high job openings rates. So even before Covid, we heard stories from, or reports from the business sector, about mismatch, if you will. That firms find it difficult to hire the right workers because they don't have the right skills, even if they interview, they don't show up for the beginning of a job and all kinds of things that we've heard anecdotally from firms. Now, in a sense, the Covid shock scaled this problem up.
So we know that the economy shut down, the positions were still there. Funding for these positions was there through various programs, the payment protection program, PPP first and foremost. And aggregate demand held up incredibly well during this recession. So all of these signs point to what's a very strong demand for labor that shows up in the job openings rate. Now, the question is, why is this so outsized? And I think one of the answers to this question is that there is a sense amongst businesses that the labor market is changing, that the requirements on workers is changing, and that, in a way, businesses post open positions preemptively. So one aspect of this is the high quit rate. So we've seen historically high quit rates. So in the quit rates, they also come from the job openings and labor turnover survey, and they represent voluntary quits of workers presumably into other positions. So there's a very high turnover in the labor market. A lot of churn. We can think of it this way. So a quit is connected with two job openings. So if a worker quits to take another position, well that fills one job opening, but it creates immediately a new job opening for the firm that was left behind, so to speak. And firms, businesses may respond to this preemptively and post open positions because they are concerned that they may not find the right workers or replace them.
Joe:
So this actually sort of gets back to my previous question, which is if that's the case, as you described, okay, firms are anticipating churn, they're anticipating quits. Is the nature of posting job openings, has it changed such that at least part the mismatch or part of the changing shape of the Beveridge Curve is not simply a relationship between employers and employees changing, but a reflection, at least in part, of the nature of how and when businesses say they're looking for an employee?
Thomas:
It’s a very good point. And this is certainly part of the story. So a lot of job search, I would say the majority by now has no now moved online, which also has reduced the cost of finding workers. And I think this is also one of the drivers, and this is actually what our economic models would tell us, that if the cost of searching for workers of job search is lower than more vacancies, more open positions are being posted because it simply is less costly.
Joe:
So that makes me wonder, I mean, I think when people look at these curves, when they look at these mismatches and my sense is that there's a lot of reasons, but we'll get into those, when we look at these mismatches and we say, oh, maybe it's people have all these savings. They're not rushing back to work. Maybe it's a labor and skills mismatch, maybe it's unemployment insurance, maybe it's people lack childcare. Maybe it's people don't want to go back into an office for fear of risking Covid. At some element, there is also just this change as you point out, being able to post a job opening online, at least some of it might just be that, all things equal, firms may be posting more openings today than they would've had similar economic conditions to say occurred in the 1980s.
Thomas:
Yes, I think that's a very fair point. But to be clear. So in a way, what is nice about this Beveridge Curve idea that both the job openings rate and the unemployment rate are normalized by the labor force? So essentially you can make this discussion about labor force participation, which we know has declined and is much below where it was pre-Covid, but the Beveridge Curve essentially nets out labor force participation effects.
Tracy:
As we observe this relationship sort of breaking down — and maybe it's because of this job creation condition that you just described, you know, the idea that employers are expecting churn and so they're posting more and more potential jobs on various websites to make up for that — but we still come back to this problem or this issue of employers and employees struggling to match with each other. And I'm wondering, do you see that more as a problem of some sort of match inefficiency, the way we're assigning jobs or hiring people for jobs I should say, is just somehow more inefficient than it used to be? Or is it that we're seeing more competition for workers and people are having to, or employers are having to compete for a sort of smaller potential pool of employees?
Thomas:
That's a very good point. I think it's both of these factors. So when we dial back the clock to the financial crisis, and this is when I became first interested in the Beveridge Curve. So during the financial crisis, the unemployment rate hit 10% stayed around between 9% and 10% for a while, and would not come down for a substantial period of time. And then it declined rapidly as the economy was working through this unemployment overhang. But at that time, we were talking a lot about precisely this mismatch phenomenon. So the mismatch phenomenon describes the idea that, well, simply speaking, that workers and firms don't find each other on the matching market. So this could have geographic reasons during the financial crisis. Some regions in the United States were hit harder than others, which has an implication for job openings and open positions.
There was a sectoral change away from construction, manufacturing to healthcare, other services. So all of these issues combine into mismatch. The idea that workers and firms don't find each other. Now we see the same phenomenon, if you will, in the Beveridge Curve in the sense that this Beveridge Curve relationship has shifted outward in the scatterplot sense. So in other words, for a given unemployment rate, now many more job openings need to be posted to reduce the unemployment rate. So this is, I would say prima facie evidence that yes, there is mismatch going on. Now, the question is, is this the same type of mismatch that we've seen in the great recession? And I would argue likely not because Covid hit most of the U.S. in a similar way. Now we don't see many regional differences. And we know that the service sector was hit much harder by Covid than the financial services sector, for instance.
So there may be some aspect to this, but I think overall speaking, I think there is an aspect of mismatch going on that we don't quite fully understand yet, which is connected to this nexus, as Tracy mentioned, of workers being reluctant to return to work, childcare issues, Covid is still around and also potentially skills — changing skills requirement. So but in that sense, I think I would argue that what we've seen in the second half of the last decade, so between 2015 to 2020, that this may have been more of an outlier than was acknowledged at the time. So we saw the labor force participation going up despite all these demographic changes. So it brought in a lot of people into the labor force, and this has disappeared through Covid. So in terms of labor force participation, I would argue that we are now almost back to the pre-2015 trend based on demographics. So long story short, I think there's a lot going on in the labor market that we don't quite fully understand yet, simply because we are lacking the micro data as of this point. And the Beveridge Curve just gives this overall snapshot that something's going on in the labor market, a structural change.
Joe:
As you mentioned. I mean, if you look, and I'm looking right now at bls.gov, their Beveridge Curve, they have a nice chart of it over different time periods. And as you mentioned, we did see a shift outwards in the Beveridge Curve after the great financial crisis. So that was already a bit of a change. And then of course, this current shape of the Beveridge Curve, it's blown out way past that. So it's clearly to a very high degree. And one of the possibilities is the so-called mismatch and maybe that's geographical, some people think, this is a very popular theory, after the GFC that it had to do with skills or the lack of skills or mismatch, but, you know, you mentioned construction and so forth. But on the other hand, you know, that led to people arguably prematurely thinking that we had hit maximum employment, or, you know, at this point, there's nothing more that monetary policy can do or fiscal policy, because at this point we just need to up-skill workers, etc. Instead, what we saw is that the unemployment rate just kept going down. It was slow, but I think we got down to like 3.4% prior to Covid. So what are, in your view, some of the policy upshots from this idea of mismatch? What does mismatch, however you wanna define it, tell policymakers?
Thomas:
So I think my preferred angle in terms of thinking about mismatches, it's really a skills story. So does the workforce have the right skill set set of skills for the requirements that firms in a changing economic environment want to have? And of course there's a lot of research, lots of policy discussions about it, but I think the overall sense is that that the labor force is lagging behind this in terms of STEM skills, if you will. Now to your point, I think it is also fair to say that we were caught by surprise how well the labor market performed in the second half of the last decade, which brought in a lot of people from the sidelines. So, and in that sense, yes, a growing economy, growing at or above trend can bring in additional workers from the sidelines. But much of this I think was driven by an expansion in services, a shift away from manufacturing to some extent, healthcare, the sector expanded. So I think there is a structural reason for this.
Tracy:
Just to play devil's advocate for a second, could you flip that slightly and argue that maybe the labor market or the economy isn't offering up the type of jobs that people want? And, you know, I say this after the experience of the pandemic, when we had a lot of anecdotes about people who simply didn't want to go back to their service jobs and maybe didn't have to because they had higher unemployment payouts, and maybe they were day-trading GameStop stock in their basements or cryptocurrency or whatever, and didn't feel as much pressure to go back to a job that they didn't like or value. Is that something that's potentially on your radar here?
Thomas:
It is, because it is connected to, well, the outside option of the workers, if you will, to use the technical term, or the willingness of workers to work in low wage jobs, if you will. So, and yes, I mean, this is definitely something that we see in the data already. That particularly employment in leisure and hospitality in the service sector is lagging much behind the financial services sector, for instance. And there are good reasons for this. So all the support programs during the Covid shock and the recovery, certainly changed the financial position of workers. There was no need to look for a shop that may not offer the same benefits and the same wage that one could get through the government support programs. So now this is starting to disappear. So to take you up on the devil's advocate question — so I think if we want to wrest out the remainder of the slack in the labor market, if you will, so from the gap between four and a half percent where we are now to below 4%, I think this is probably where it is coming from. But to your point, I think this is certainly an issue that is on the radar, namely in terms of what is maximum employment in this economy?
Joe:
This seems to be the question that the Fed, and by the way, I think it's interesting, people should know we are recording this Monday, November 22nd, and we start our recording at 9:00 AM, right at the moment, literally we started recording right at the moment that it got officially announced that President Biden would be renominating Powell. So assuming he gets confirmed, this is the question that Powell will have to answer. How do you think about this question of maximum employment? And this idea of, as you mentioned in the second half of the 2010s economists were taken by surprise by how strong the labor market could get, how many people it could bring in, what it could do for labor force participation, how low the unemployment rate could get without triggering a meaningful increase in inflation and so forth. So on this question of, you know, obviously there's some head-scratching that's going to happen. The Fed is obviously reluctant to put any sort of number on maximum employment. It kind of feels like a ‘we'll know, it when we see it’ type of thing — inherently subjective — but from your perspective and from your research, how should the Fed be thinking about answering that question of what success on the employment front looks like?
Thomas:
So the Federal Reserve Act stipulates maximum employment as the first mandate and the Fed has always interpreted this well, first, a little bit more narrower, and then a broader goal in terms of an unemployment rate that is sustainable without generating additional inflation and where the labor market and the trend in the labor market, very loosely speaking, are in equilibrium. So when I went to college, then graduate school. So we were thinking of maximum employment as an unemployment rate of 5%, which we dropped to 4.5% in the 2000s. So this unemployment rate tied to maximum employment can change, of course. And in a sense, this is where we were surprised that the unemployment rate could sustain low inflation rates below 4% for a very, very long time. And I think it's fair to say that this experience of below 4% unemployment really informs the current discussion right now, but I think with the experience of the 2010s and the recovery from the global financial crisis also taught us that the unemployment rate — the headline unemployment rate — may not be the single best gauge for what maximum employment means.
So for one, I mean, it, it is a survey-based measure. So it is measured reliably as much as it can reliably be measured, but it does not fully encompass all of the job search that is going on. So, and this is why policymakers in my reading, have shifted to looking at additional metrics. So the employment-population ratio, which may be just as good a measure of full employment, then the labor force participation rate, as well as flow-based measures such as the job openings and labor turnover survey. So I am inclined to agree with you that we know it when we see it, when we have maximum employment. So at this point, I think there's still room to go until we hit 4%. Now, the flip side of this is, of course, our view of maximum employment was informed by the fact that the inflation rate stayed well contained above 2% for a long time in the 2010s. So much of the thinking that goes into what is maximum employment is actually related to the inflation rate, because we did see historical episodes when unemployment was below its natural rate, if you will, inflation would go up. This didn't happen in the 2010s.
Tracy:
This kind of leads into something that I wanted to ask, which is, when would you expect to see this competition for workers — or the sort of inefficiency in matching — when would you expect to see that feed into wage increases? And is it possible that if we've seen a historic break in the Beveridge Curve, you know, the relationship between the unemployment rate and the job opening rate, then is it possible that maybe we're going to see something vastly or wildly different when it comes to wages?
Thomas:
So at this point we haven't seen much of wage inflation at a level where I think policymakers and economists would become uncomfortable. So nominal wages have been going up, but not excessively. So, and actually to the point that real wages have been falling because of the high inflation numbers that we have been seeing, what we do see is some higher wage increases in some sectors. So leisure and hospitality, there we do see higher wage growth, higher nominal wage growth. And we also see the job switches, which brings me back to the quit rate and historically elevated level of quit rate. So job switches have seen high wage increases, which presumably is one of the reasons why they switched jobs. But at the same time, we know that the wage data are lagging behind in the sense that so many wages are determined one year ahead and now many companies are having wage discussions at the end of the year. So what is the compensation picture going to look like for 2022, given the high levels of inflation we have had? So I think going into the next year, I am starting to become a little bit more nervous that we may see much higher nominal wage inflation numbers.
Joe:
Let me ask you, I want to go back to something important and again, this mismatch idea, because ultimately that really is what the chart is showing us on some level, there is some big change. To what extent are these things, you know, you mentioned for example, and, you know, numerous policy makers talk about say skills and STEM skills and tech skills and coding skills, in particular, as always being in short supply. And there's probably not a major business in the entire country that doesn't have numerous job openings for technical skills. But one thing that we did see at the end of the 2010s, in addition to bringing some people back into the labor force, is more anecdotal stories about skills training, where the company itself would pay for skills training or fund skills training, or also where the companies would hire people who were previously incarcerated. And so looking beyond the normal pool of applicants to bring people back into the workforce, should in theory have a very sort of positive long-term effect. To what extent should these mismatches essentially be, or can they essentially be, dealt with by the market? That if, you know, the Fed can recognize the persistence of mismatches, but ultimately if there are — whether it's skills or geography — these are things that employers and employees will ultimately solve if there is a price opportunity there.
Thomas:
Yeah. I mean, I very much agree with you on this point. I mean, in the end, these mismatches will be dealt with by changes in employment relationships and how workers and their employers reorient their relationship. I think the prime example for this is working from home on a hybrid working environment, which decouples the location of the business from the location of the workers. So the geographic mismatch part of the mismatch story, I think would probably largely go because I could, you know, as we're having this interview right now, I mean, I'm sitting here in Richmond, and you're in other locations all over the world. So that part I think can be mitigated. So one thing, with respect to the U.S. labor market, so one thing that struck me — and my background is in Germany and growing up in Germany — it's the lack of apprenticeship programs and on the top training programs, which was very formalized, and very institutionalized in Germany.
And you don't see this to the same extent, actually to a much lower extent in the United States. And I think this is one of those labor market policies that, I think policymakers should try to pay more attention to. On the job training, up-skilling of the workforce, and essentially worker training. But I think the broader issue is can we already discern structural changes in the labor market that would completely decoupled us from, you know, previous working arrangements? And I think the first signs are there. So I think the Covid shock has accelerated this work towards more of the ‘worker as an entrepreneur’ rather than some salaried wage employee.
Tracy:
So Thomas I'm aware that you've been publishing quite a lot at the Richmond Fed. And I mean, you've sort of taken on the very, very big questions of the economic experience of the pandemic, including how to actually model Covid-19 and things like that. But what's been the most interesting work that you've undertaken over the past year or two, or perhaps the most surprising thing that you've seen — other than the Beveridge Curve, of course.
Thomas:
Thanks for your kind words. I mean, I would've answered the Beveridge Curve, because this is truly striking. I mean, I started working on this during the financial crisis and at that time, the shifts in the Beveridge Curves seemed completely out of line with historical record. And then I just plotted late last year the Beveridge Curve, and whoa, suddenly the change in the Beveridge Curve was even much more dramatic, but in terms of other work, so one aspect that occupies my thinking a lot is, well, what we call r*, or the natural real rate of interest, which basically anchors almost all of our policy discussions. And what we can say is that the natural real rate of interest — so when all is said and done, the level of the real interest rate that the economy would naturally gravitate towards — has declined over the last 20, 30, 40 years. The precise level of where r* is, yeah, there's much uncertainty about this. And this fundamentally informs policy questions. So in the long run, where should a natural normal level of policy accommodation or the federal funds rate be? But I think we are far away from having a good answer to this at this point.
Joe:
What about just the, you know, usefulness of these indicators, r*, neutral rate, real rate, you know, this is prior to Covid — and I always forget what year it was, I want to say 2018 — Chair Powell gave a speech sort of questioning whether any of this, any of these numbers, what the neutral rate is, that this is really knowable in real time. And I took it to mean that maybe there are just some questions about like, to what degree should we be using these to build models that inform policy in real time? How do you feel about the confidence that economists should have in being able to derive a number of like r* and actually use it to make a decision from one meeting to the next?
Thomas:
Yeah, thanks for the question. So, I mean, as I said, there's much uncertainty about the levels of these natural rates. Be it u* — the natural rate of unemployment, be it r*. Ultimately I think all of the models that we have in mind, and I'm not just thinking about, you know, the academic models that academic economists build, but also the implicit models that policymakers have in their mind, they ultimately come down to the question, what is the long-run equilibrium outcome in the economy, where we are at maximum employment, where we’ve hit our inflation target and what the long-run interest rate should be. And once you take this issue on board, this immediately leads you to the question, well can we figure out what r* or u* is. So, and to be fair, these are fundamentally unobservable. So we have to use some statistical methods, technical tricks to tease them out from the data, but I think that they are informative nevertheless. So even if it is trust at the level that r* is much lower than it was in the 1980s, which puts us into a different policy environment.
Tracy:
So just on that note, I have a slightly, I guess, strange question, but you know, one of the things that has been repeated in many of our conversations on Odd Lots over the past year or so is the idea that a lot of economic principles that we seem to have taken for granted, actually, no one seems to know how they really work. So something weird is going on with the labor market, something weird is going on with inflation. I think r*, people have been questioning the natural rate of interest idea for a while now. I know you've done some academic work on the output gap not being particularly useful for monetary policy. Is this sort of a moment of doubt for economic researchers? Or does it encourage you to, you know, examine these issues even more and seek out answers? It feels almost like a sort of existential crisis, or at least if you read some of the commentary around higher than expected CPI from a week or two ago, it feels like an existential crisis for the Fed at least.
Thomas:
Yeah. I would not put it in such stark terms.
Tracy:
That's fair. I just, I got to ask.
Thomas:
So in the sense that I think we do have a much better understanding of how the economy works compared to 20, 30, 40 years ago. But I think what we've become more aware of is that there's much more uncertainty about these unobservable variables and again, going back to the Covid shock, we were hit by a completely unprecedented shock that shut down economies worldwide to an unprecedented level. And we've recovered from this incredibly quickly, in terms of the macroeconomic data. And yes, while the Beveridge Curve may look different from what it would have expected had Covid not happened, it still follows the same underlying principles. So when times are good or improving, the unemployment rate falls, job openings go up, but what the Beveridge Curve tells us in that sense, it is, you know, a visualization device of the state of the economy. Yes, but there's something else going on. Now, we don't know exactly what is going on so more research will be poured into it. So when I presented the Beveridge Curve to internal audiences and outside audiences, I introduced it as job security for labor economists. So more research needs to be done. But I think overall our economic models and our economic thinking actually held up pretty well during the Covid shock.
Tracy:
Thomas, it's been great speaking with you and, again, a plug to the Beveridge Curve paper, which you can find on the Richmond Fed website. Thank you so much.
Thomas:
Okay. Thank you so much. It was a pleasure talking to you.
Joe:
Thanks Thomas, that was great.
Tracy:
So, Joe, I thought that was really interesting. And again, I think, you know, all of, well, the two major issues — inflation and the labor market — are of course tied together, but it does feel like the focus of the majority of attention has been on the inflation question. And people are only really talking about the labor market in so far as it might lead to wage increases. So I thought it was quite good to dig into, you know, the Beveridge Curve and a particularly wonky corner of economic research. Yeah.
Joe:
Anyone who just pulls it up and goes to the BLS Beveridge Curve chart, you just instantly see how wild, I mean, look, Covid data is going to leave this imprint on all charts that we look at for decades. Because it was just so weird, whether it's the raw surge in the unemployment rate, the pure speed of the bounce back, the collapse, everything is just so weird, but you look at it and it's obviously very different. And I do think that, like, I don't think there's a totally satisfactory answer yet to why we do seem to have this very robust demand for labor, and yet we continue to have this big hole, you know, at least 5 or 6 million people. It still remains a pretty big puzzle.
Tracy:
Yeah. And I liked Thomas' description of it as a full employment plan for economists. But I do think, I mean on that wider point though, I mean the idea that maybe we know more than we did before about the way the economy works, but we sort of don't have a good way of measuring those unknown economic variables, like maximum employment and r*, I think that's quite, maybe that's the way to think about it. Like it's not just that no one knows how anything works anymore, which would be quite frightening, but it's becoming increasingly hard to gauge those particular ideas maybe.
Joe:
You know what, while we're here. And I imagine maybe some listeners who listen to this episode may work at the Fed or something like that because we had a guest at the Richmond Fed. So I have a request to those listeners. So Tracy, you know, like the Beige Book, it’s like the regional survey. It's like business leaders said ‘oh, you know, we're seeing labor market tightness, having trouble hiring.’ I want a beige book for workers, anecdotes of the labor market, but from the worker perspective. Just asking people questions, okay, we get the actual data. But talking about what was your experience? What was your experience applying for a job that you were ostensibly qualified for? Did companies actually call you back? Was the description actually how it was? I think if employers are going to get to sort of like give their anecdotal takes that aren't really based in data on like, oh, you know, like we can't find the people or the workers aren't showing up after we offered them the job. I think we deserve a Beige Book from the workers’ perspective. And I think that might help us answer some of these questions. So if anyone is listening, maybe we could get some funding for that.
Tracy:
I think that's a great idea. And I mean, I want to know whether or not there is a wealth effect, or greater wealth effect going on from like stock trading, from crypto.
Joe:
Yeah. This would be the kind of thing where instead of just having to speculate, I mean, anecdotes can only go so far. And I think we have to be careful. And one reason we have to be careful about anecdotes is because, you know, employers started calling the labor market tight in 2014 or 2015, and then we had like five more years of unemployment, right? So you have to be careful of anecdotes, but all of these things like ‘oh, are you taking the time off from work because you want to trade crypto,’ or something — like let's ask people in a systematic manner and have it be something that we record rather than just sort of like spectrum.
Tracy:
Totally. And I also want to see how people react to or what they say about the new hiring websites and things like that. Because I gather — look, it's been a while since I've applied to a job on one of those, so thank you Bloomberg — but like my impression of them is that you kind of have to tick all these boxes and it can be really, really frustrating if like one aspect of your resume is slightly out of line with whatever the request is from the potential employer or the Monster.com form or whatever. And I suspect that's also frustrating some of these matching efforts. But I’d love to know.
Joe:
And again that speaks to Thomas's point, which is like okay, these websites have made it a lot easier to post job openings. There are behavioral effects on the employment side where it's like listing in 2021, may be different than listings in, you know, 1986 or something like that. And so again, it feels like there's a lot of ambiguity in all of this and yes, interviewing more people about their experience on these sites, do they actually get callbacks for jobs that they're qualified for that they continue to see posted? I think it would be very useful. So hopefully someone listening, call us up, name it at the Odd Lots survey.
Tracy:
Great idea.
Joe:
The Odd Lots book, the Odd Book. That has a great ring to it.
Tracy:
It does. Alright. Let's leave it there.
Joe:
Let's leave it there.
You can follow the Richmond Fed on Twitter at @RichmondFed.