A Big-Picture Look at the Economy - February 21, 2018
Charles Davidson: Welcome to the Federal Reserve Bank of Atlanta's ECONversations. I'm Charles Davidson, of the Atlanta Fed's public affairs department. Thanks for joining us today. I'm here with Paula Tkac. Paula is associate director of research here at the Atlanta Fed and a financial economist with deep experience studying the macroeconomy and financial markets in particular. Paula, thanks for your time today.
Paula Tkac: Thanks, Charles.
Davidson: All right. Please feel free to ask questions. You'll see a button on your screen through which you can submit queries, so do that as we go through our chat today. What we're going to do is talk a little bit about how we here at the Atlanta Fed prepare our president, Raphael Bostic, to take part in the Federal Open Market Committee [FOMC] meetings in Washington, which is the body of the Federal Reserve that sets monetary policy. So, Paula, can you get us started by talking a little bit about how we help to prepare Raphael to go and participate in the FOMC?
Tkac: Sure. Let's start with a picture: the model we use here at the Atlanta Fed is one that brings together two really disparate groups of folks that oftentimes don't get to talk. The picture you're seeing here is actually a network built on data of friendships and relationships within a high school, but it actually applies pretty well to the policymaking process in the economy. Think of the yellow dots as all the business folks out there—households, all of us in our individual lives, transacting, investing, building things, working—and all the blue dots, think of those as all of the economists who are trying to figure out what all the yellow dots are doing. And you notice here that there are not that many points of connection. Oftentimes economists use statistical methodologies and models to try to understand the world. We here at the Atlanta Fed like to build a bridge between these two worlds and place our president, Raphael Bostic, right in one of those nodes that connects businesses to the rest of our profession. We believe that by bridging both of these perspectives, we can come up with a much better understanding of the way the economy works, and thus help him prepare to have the best discussions possible in terms of setting monetary policy.
Davidson: Right. So plenty of data there, but we also want to hear from people out there who are actually making things happen.
Tkac: Exactly—that's what we're going to do. So today we're going to talk a little bit about data, and a little bit about what we're hearing from businesses, and give you a flavor for the ways that we try to pull that information together.
Davidson: All right, excellent.
Tkac: Awesome. Let's start with GDP. GDP stands for gross domestic product. It's a measure of...think of it as the sum total of economic activity that occurs in our economy over a certain amount of time. Here we're looking at a graph of quarterly gross domestic product, in particular the growth rate of that domestic product—think of how fast that pie of economic activity is growing over time. Here, we're looking at a historical chart that covers most of the recovery from the Great Recession. If I were to average all of these quarterly growth rates, we would see that on average the U.S. economy grew at about 2 percent per year in the recovery.
Davidson: Now, Paula, is this showing growth from the previous quarter, or the same quarter of the year before?
Tkac: This is an annualized growth rate based on the growth in that quarter relative to the quarter before. So we're taking the growth and then we're saying, "If it occurred in an annual basis, what would that look like?"
Davidson: Got it.
Tkac: So what you see here are the first quarter of each year highlighted in red, and quarters two, three, and four in blue. And one thing that we've noticed here at the Atlanta Fed, as have all Fed watchers, is that—at least over this period—the first quarter of each year, January through March, tends to be a bit weaker than remaining quarters in the year. If I zero in here on the last year, 2017, we see that in the second, third, and fourth quarters the U.S. economy was growing at a rate of just over 3 percent (2.5 percent in the fourth quarter). Our question here at the Atlanta Fed is, is this growth rate—which is somewhat higher than our average over the last seven or eight years—is that likely to continue? So now we're in the midst of the first quarter of 2018, and we want to know: are we going to see a weak quarter, as history might lead us to expect? Or are we going to see some continuation of that slightly higher growth? Up at the top, underneath the GDP title, I have the most recent print for our current tracker. We have a tool called GDPNow on our website that pulls in data as it's coming in over the course of the first quarter and tries to forecast what the total growth for GDP will be for the first quarter. So in this case, as of most recent print, we're expecting a 3.2 percent growth rate in the first quarter—that would suggest the momentum is continuing.
Davidson: Right. Now, GDP Now—I know it tends to swing a little bit. It will have a reading of maybe four or five, and then it's three. Can you talk just very briefly, Paula, about why we'll see a fairly substantial move in that?
Tkac: Sure. Unlike some other GDP forecasting tools that you might see on TV and in the press, GDPNow is a purely statistical model. There's no judgment imposed, we're not sitting upstairs saying, "Well, what do we think that number means?" We feed the data as it's coming in into sophisticated statistical models. And based on the data that comes in, we get the best estimate of what the total growth will look like when it's finally done at the end of March and reported by the BEA [U.S. Bureau of Economic Analysis] at the end of April. So if, in particular, a really good manufacturing report comes out—higher than it has been in previous quarters—it will cause the model to expect more good data as the quarter goes on, and so the current read will look very high. If that data doesn't come in, the model will tend to kind of equilibrate back into something that's more standard.
Davidson: Sure, sure, okay.
Tkac: Great. So now, if that's the data: what are we hearing from businesses?
Davidson: These are those other—the giant clump of dots we saw—what are they telling us?
Tkac: These are the yellow guys. The ones we just saw were the blue guys, the economists. So the yellow guys come from our business contacts. We have a national Business Inflation Expectations survey, and we asked them in late 2017—looking ahead, what were their biggest areas of concern?
Davidson: Now, this is businesses from across the country, not just the Southeast?
Tkac: So we asked them, "What are you concerned about?" We want to understand: what risks or headwinds might there be for the economy? So the first thing I want to draw your attention to is about halfway down the chart: regulation. Only about 7 percent of the businesses that we talked to included this in their responses. That's different from what we have seen over, I would say, the past two or three years, where regulation was a much greater concern. You see that firms are not, also, that worried, really, about demand and sales. Their optimism about demand for their products and their ability to produce productively and sell their goods is pretty good. They're also not that concerned right now with geopolitical risk. What they are concerned about, at the very top, is labor quality and availability, and the cost of hiring that labor—thinking about the people that they're going to need to sell into that demand that they seem to be slightly more confident of.
Davidson: What do we think is behind that, Paula?
Tkac: Well, I think it's, again, best to go back to some data as we try to answer that question, so let's dig inside what firms are thinking first by looking again at the macroeconomic data. The labor market spider chart you see here, again, is a tool that we have on our website. We've got 15 labor market indicators arrayed along the axes of this chart. Here at the Atlanta Fed, we believe that you can't look at the U.S. labor market with one statistic—the unemployment rate, for example. That's not really giving you a very good picture of what's going on. So we've assembled these 15 indicators, and the way to think about this chart is the following: the green sort of polygon shape there in the middle represents the depth of the last recession—so about as bad as the labor market got, along all these indicators.
Davidson: So the smaller the shape formed by the lines, the worse it is? Got it.
Tkac: The smaller the shape, the worse it is—right. The dark blue is actually where the labor market was at the peak of the last expansion before the recession. So think of that as: that was good, then we shrunk down to the middle. Now the question is, where are we now?
Davidson: So we're outside that...
Tkac: Right. The orange shows us where we are now. Along almost every indicator, including initial claims, payroll employment, private job opening rates, quit rates—which, of course, people are quitting when they think they can get another job, so quit rates getting higher are actually kind of a good thing—along all those indicators we're actually out beyond where we were at the peak of the last expansion.
Davidson: Right. Over on this side, we're not quite there.
Tkac: Yes, the two places we're not, really, are in the area of wages and the utilization of our entire workforce.
Davidson: Okay, that's "slack." We hear that term.
Tkac: You can think of that as slack, right. I think of it as—is the economy fully utilizing the skills and talents and human capital in the available workforce, and are people who have skills actually in the labor market looking for employment?
Davidson: Right. We don't want to have electrical engineers working at Starbucks...that's an extreme example, but...
Tkac: Right. Or, thinking that maybe they can't find a job and so they're not in the labor market looking for a job, yet they have productive skills that that they could be contributing.
Davidson: Right, sure.
Tkac: So if you look on the bottom left, there are two good measures of this utilization—one is these marginally attached workers. These are folks who have looked for work in the past year but have not looked for work in the past month. So they're technically not in the labor force, they won't be in an unemployment rate, but they do represent sort of a shadow labor force that you believe are productive, and you would like them to be utilizing those skills. The other set of folks right next to them are those that are working part-time for economic reasons.
Davidson: So they'd prefer a full-time job, but they can't find one, or...
Tkac: They would. And according to them—it's self-reported—their business can't sustain the demand or the activity that would allow them to be then working full-time.
Davidson: Got it.
Tkac: So again, these are kind of underutilization numbers. The other thing that hasn't recovered, at least in terms of growth rates, are wages, quite frankly. So again, you see here the places that are inside the blue polygon are the average hourly earnings growth and the employment cost index growth. Even as the labor market has improved—we've added hundreds of thousands of jobs every month—wage growth is not as high as it was prerecession. And that creates a bit of a subtle and more complicated picture. Typically, in a tight labor market, you might expect to see high wage growth as firms bid for workers and there are less and less workers out there on the fringes to attract in.
Davidson: Supply and demand, right?
Davidson: If supply is low, demand goes up—price, typically...but not so here.
Tkac: You would think. So again, to kind of understand how this is occurring—and again, what businesses are thinking about with their labor availability—we talk to businesses. This chart is a bit of a tour of the labor market and how businesses have responded since the Great Recession. If we start at the smallest circle, the first thing they did when they had less sales and less activity was lay folks off and increase the job duties of people that were there, and as demand increased and the economy began to recover, they just tried to do more with the same amount of workers.
Davidson: More with less...or, more with the same.
Tkac: Exactly. Then they began substituting some technology for labor, right? So figuring out ways to become more efficient, using technology—that helped save on labor, and they were able again to grow with the economy without necessarily hiring a lot of extra workers. Then we began to see skill-based tightness. We heard about the difficulty in getting IT workers, or cybersecurity workers, or even truck drivers. There were pockets of occupations where labor was particularly tight, and those folks were being paid a premium because they were in high demand. At some point, firms began saying, "Well, you know, sometimes we can't find the purple squirrel that we might actually like, that perfect person—but we can train folks." And so they began training people to get the skills that they needed.
Davidson: We could buy some purple paint, maybe.
Tkac: We could, right, exactly. Or put a tail on some other animal, and turn them into the worker that is optimal and productive for our firm.
Tkac: Then we heard about firms—again, in an attempt to keep the workers they have, now that they've got them trained up and they are productive—that began thinking about nonwage benefits. Rather than increasing wages, they might provide more flexibility to work remotely, or trade off vacation time for wages, or provide a company car, or bring your dog to work—all kinds of things that might, again, be attractive and add value for a worker but aren't just wage payments. Now, most recently, we've been hearing that the wage premiums—these extra wages that are paid for those in high demand—have spread out into industries and geographic areas.
Davidson: Paula, can I ask you a quick question?
Davidson: This notion of sort of fringe benefits, let's call them—bringing the dog to the office, ping-pong table, and so forth—is there any kind of statistic to capture that kind of thing?
Tkac: Not that we've been able to find. Most of our work in this area has been talking to businesses, because it's been very difficult to quantify—especially because much of the activity that we're talking about here are things that firms say that they've not typically done before. So one of the ways that we've found out about these things is not just individual firms, but convening round tables of HR professionals and others, folks that run search firms who are matching workers with firms, and saying, "What kinds of trends are you seeing?" So I think there certainly is more evidence, again in the anecdotal sense, but there's no real way to look at nonwage benefits, other than things like health insurance...
Davidson: Which shows up in some form...
Tkac: Which comes up in the employment cost index.
Davidson: You don't have to register the dogs who come to work every day. [laughs]
Tkac: Right. [laughs] So in the employment cost index, there are some nonwage benefits. But most of the stuff we're talking about is an innovation that folks are doing.
Davidson: Right, makes sense.
Tkac: So now we're seeing that there are some geographic areas—like especially south Florida, with the travel and tourism industry—experiencing big pockets of tightness and difficulty getting workers. The construction industry, increasingly...
Davidson: It's kind of booming almost everywhere, I guess, isn't it?
Tkac: Indeed. And we're also finding that employers are trying to reduce the barriers that might stand between a worker being able to successfully perform in a job. So, it could be something like transportation to work—right? And so you might find, now, firms who will get van pools together to help people with the commute or the transportation, because they realize that's a barrier to getting the productive workers in on time, all the time. Again, this is a way, outside of paying them more, that you might solve the problem.
Davidson: And it's worth the money they're going to spend doing that to have the return there. All right—makes sense.
Tkac: Exactly. So as we look at low wage growth, these are some of the ways that firms have told us that they're attracting and retaining workers without just, very simplistically, raising their wages at an increasing rate.
Davidson: Right, right. Is it the case, Paula, that firms in general prefer not to raise wages? Not to cast any sort of aspersions or anything like that, but just purely economics, I mean...
Tkac: Well, I think what we've heard from employers is that there's a lot of complexity around a wage structure. Let's say you're going to bring some new people in at the lower end of your skill set in your firm, and you now need to pay them a premium in order to get them. Well, if you don't raise everyone else's wages, you've got compression, right? You might even get some inversion, where certain folks are making more than others. Those have negative consequences for the productivity, and the culture of the firm, and morale. But if you think about increasing everyone's wages, that becomes a much more costly proposition than just the wage that you might pay the new person. So it's not that they don't "want to." The question is, how do you manage this very complex work force? And of course, they're dealing with people who are leaving for a higher wage or a better set of benefits, and they have to figure out how much productivity is lost in that turnover and how they might want to, again, compete for those workers or try to work to retain folks over time. And so it's not an easy problem.
Davidson: Right. There's no "one size fits all" solution, clearly.
Tkac: No, no, there's not. That's one of the things we've learned by connecting the yellow dots with the blue dots.
Davidson: [laughs] The yellow dots may look the same on the screen—but they are not the same in real life. Nor are the blue dots, for that matter.
Tkac: They are not the same—no, they're not. So this is the picture of labor markets, which is half of the FOMC's dual mandate: maximum employment. We can look to the labor market and these kinds of numbers and anecdotes to understand whether or not we're at full employment, whether or not labor markets are tight.
Davidson: Right. And there's no set definition for "full employment," correct?
Tkac: No, there isn't. It most certainly is not "every person in the economy working a full-time job," right?
Davidson: Right—not realistic, I guess.
Tkac: Younger folks—obviously, we're not interested in six-year-olds working full-time jobs—but folks are in school, they're retraining, they might be taking care of family members. Folks—many of my friends—think about retirement, and other things that they want to do. So it's not the case where everyone will be employed in full-time jobs, so it makes it a little bit fuzzy, and there are many folks who have slightly different definitions that they might find to be their favorite.
Davidson: Yes, yes. Full-time work may not be for everybody. If you can get away without it...
Tkac: That's true as well. Many students do not work full-time, but that for them is optimal. So, the other half of the FOMC's mandate, of course, is price stability—that means we need to turn to inflation numbers. And in this case, if we think of labor markets as seemingly looking very healthy and robust, and an unemployment rate at 4.1 percent, the puzzle on the inflation side is, with the economy growing at 2 percent for, again, on average the last seven or eight years, the momentum there in the last, say, year or so on GDP, the labor market continuing to expand...we really aren't seeing the inflation that many folks predicted when the FOMC brought rates down to zero and purchased a lot of assets for our balance sheet, and we're actually seeing inflation—you can see here in the chart—that is below the FOMC's desired target, which is 2 percent. The blue line here is the personal consumption expenditure index. That includes food and energy, so the big swings that you see in that blue line are predominately related, in this case, mostly to energy prices.
Davidson: Right, oil prices.
Tkac: Exactly. But the red line takes out food and energy, not because they're not important—they are important, we all purchase them. Even the "blue dot" economists are putting gas in their cars and food on the table—but because they are noisy and they are related to lots of supply-and-demand factors that are outside of what the economy is doing, oftentimes weather and geopolitical events and those things. So if we strip those out, that core red line gives us a better handle for what all the other prices in the economy are doing. And even here, we see that the growth rate over, let's say the past 12 months, is at 1.5 percent in that core number—again, relative to what the FOMC would like it to be, which is 2 [percent]. So one of the puzzles, for us, is, just looking at the data, why isn't inflation somewhat higher, and do we have to worry about it breaking out of the box quickly? Maybe it's low, but is there a risk that it could become significantly higher in a very fast way? That's what a monetary policy maker would want to be thinking about.
Davidson: Yes. So, Paula, when we talk about "core," I think maybe that gets to the question of... I think a lot of people confuse inflation in these terms with just purely their cost of living, and they're not the same thing.
Tkac: Indeed no, they're not. In fact, we have a tool on our website, myCPI, where you can go in and put your profile in and it will compute the basket of goods and services that you buy, and you can see how fast your cost of living is estimated to go up. These indexes that we use here actually think about market baskets for the economy as a whole, right, so they may not be reflective of any particular person's experience with the particular things that you buy.
Davidson: Right. And also, I guess, a related point—with monetary policy, we hear economists often refer to it as a blunt instrument. It sounds a little unseemly [laughs], but what does that mean?
Tkac: Well, essentially the lever that the FOMC has to achieve its dual mandate of price stability and maximum employment really is the setting of a short-term—an overnight—interest rate. That interest rate translates into longer-term interest rates, which then affect economic activity and business activity, but there is no precise tool that, let's say, will create more jobs in a certain sector, right? That's sort of the fiscal side of the government policy house. The monetary policy side really is about the amount of money circulating in the system, and so when we do some things like purchase assets we can affect the amount of reserves in the banking system, which will play into the inflation picture. And by promoting economic activity, then, there is a connection to labor markets, but there's no real direct path to influencing labor markets.
Davidson: Right. If we see a lot of unemployment suddenly in the software industry, we don't have a way to target that.
Tkac: No, exactly. So that's the data. What have we heard from firms? Well, we've actually heard some interesting things about increased competition, especially firms who talk to us about their suppliers. They say they've actually been able to lower their costs by increasing the competition among all their suppliers. They can run competitions on platforms where suppliers now can bid competitively, so the contract isn't necessarily going to your long-time supplier. You can look for cheaper prices. There's also a fair amount of consolidation in a lot of industries: gaining economies of scale, firms becoming more efficient—which is good, but they still report to us that they don't see their ability to charge higher prices to consumers, or to really increase prices very quickly. And finally, we're hearing a lot about new pricing strategies, the ability to charge different customers different prices or change your price based on time of day or location. Think about airline tickets, right?
Davidson: Yes, sure, or surge pricing with Uber or Lyft, et cetera.
Tkac: Exactly. Prices can move all over the place now, and firms—because of technology—are increasingly able to have much more robust and complex pricing schemes. So one question that I personally have is, how does that and the other things we're hearing from business ultimately translate into the measures we have for inflation and could maybe there be something in that translation that would help us square what we hear happening with what we see in the data?
Davidson: Yes, another puzzle, I guess. Seems like with economics there are few hard and fast answers, are there?
Tkac: Yes, none, probably. [laughter] Which is part of the fun, actually.
Davidson: Yes, sure. It keeps everybody here busy.
Tkac: That's right, it sure does. The last thing I wanted to do is show you the view from the FOMC. If you put yourself in a policymaker's chair around the FOMC table, you're going to need to balance the labor market data and things you're hearing from businesses with the inflation picture and things you're hearing from businesses, and come up with some assessment of what you believe as a policymaker would be an appropriate path for this short-term interest rate, this federal funds rate. So what you see here is something called the dot plot. It's released by the Committee four times a year—this one's from December of last year. Each dot is a policymaker who sits around the table, and they're asked to describe their version of appropriate policy for the next several years based on what they see happening in the economy and their forecast.
Davidson: Right. So each of these is a regional Federal Reserve Bank president, or a member of the Board of Governors of the Federal Reserve?
Tkac: Indeed, yes. And so one thing you can see here: at the end of 2017, they were all pretty sure, or pretty much in agreement, about where they thought the appropriate rate was, but as we move to the end of 2018, 2019, 2020, you see the dots are pretty spread out. So there's a lot of diversity of opinion. Some of it diversity of opinion about how the economy will evolve, some of it might be diversity of opinion based on what the appropriate way to manage monetary policy is, even conditional on how the economy evolves. But this difference of opinion is what creates a really diverse discussion in the system at the policy table, as they come up with—eight times a year—their consensus view and take a policy action, or keep policy in the same place, at each one of their meetings.
Davidson: Right, okay. Well, Paula—thanks so much for that. We've got time to grab a couple of questions. The stock market—financial markets in general—bounced around a little bit lately, and I know that's not a primary focus for monetary policy, but do we factor that into the monetary policy deliberations?
Tkac: As a financial economist, I think one way to think about what's going on with financial markets is, they're constantly looking ahead, the way all of us are, assessing all of the information that's coming in. But they're processing it in real time, and as you see the result of that can be a lot of volatility. You can have a piece of news that might come out that might look good, and markets may go up a lot. And then another piece of news comes out which may change your interpretation of the first piece of news, and they'll go down a lot. So in the sense of sitting, as I do, as an advisor—things that are happening at a high-frequency level with financial markets are most of the time not of concern, because that's what markets do.
Davidson: Day-to-day swings, basically, you're talking about.
Tkac: Right. At some point, like in the financial crisis, there are things going on in financial markets that are more fundamental. One of the challenges is always to figure out, when does it turn from noise, if you will, into something more fundamental?
Davidson: Right, right. All right—well, that's just about all of our time today. Thank you for joining us. Thank you for your time, Paula.
Tkac: Thanks, Charles.
Davidson: And please come back in June for our next Atlanta Fed ECONversation. Thanks for watching.