Transcript
Dave Altig, Executive Vice President and Director of Research
John Robertson, Senior Policy Adviser and Economist
Research Department
Federal Reserve Bank of Atlanta
Charles Davidson: Welcome to the Federal Reserve Bank of Atlanta's ECONversations. I'm Charles Davidson, from the Bank's Public Affairs Department. Please feel free to ask questions during our session today. We will be answering those, so ask via the Ask a Question button on your screen. It's pretty simple.
In today's webcast, we're going to discuss the Bank's 2016 annual report. To do that, I'm joined by John Robertson, a senior policy adviser here at the Bank, and Dave Altig, our chief economist and research director.
Thanks for being here today, guys.
David Altig: Glad to be here.
Davidson: All right. Well, let's get to it. First question, guys: the annual report looked at the economy over the last decade, and we drew distinctions between secular and cyclical changes that have occurred over the last decade. First off, what do we mean by "secular"? What is that?
John Robertson: One way to think about that, Charles, is, if you thought of the economy consisting of supply—the supply of goods and services—and the demand for those goods and services, and the secular factors are kind of on the supply side. One way I like to think about it is if the economy was one big factory, producing output—right?—the rate of output growth that that factory would have is kind of determined by how fast is the pool of labor resources in the factory growing, and then how you use those labor resources, in terms of the types of capital that you give those resources, the education of those resources.
And all those things go together to produce stuff, so it's kind of like you've got a labor supply piece and you've got a productivity piece, and those can change over time. If you think about the labor supply, then a key determinant is just the growth rate in the population, for example.
Davidson: Okay. Well, what makes something cyclical?
Altig: Well, in the simplest sense, what we call cyclical movements are temporary movements away from the secular trends that John was describing. It's a little bit complicated sometimes to understand how much is a secular trend and how much is a cyclical trend. And there can be supply-side movements that are temporary, that move you away from your long-run path. I think maybe a good example might be a substantial change in regulation, which would require some period of adjustment.
But usually when we're talking about cyclical things, we're thinking about—to use the contrast with John's supply-side kind of thinking about secular trends—the things that are created by demand, by demand shocks, and as a consequence of it being about demand, those are the things that we think things like monetary policy can actually address.
If we go back to the crisis, there were very substantial changes, which fortunately turned out to be temporary. We would think of those as big, cyclical movements—during the recession, for sure, but even afterwards.
Davidson: So we're in the monetary policy business here at the Fed, obviously. How did monetary policy respond to—and let's hope influence—these cyclical dynamics?
Altig: Well, I think everyone is pretty familiar with the fact that there were very substantial monetary policy responses. This last recession was an interesting one, compared to the recessions that most of us have lived through—[during] a good part of the post-World War II period, if we're old enough—and what makes it interesting is it was associated with a financial crisis. So some of the tools that monetary policy and the central bank—the Fed—brought into play were very familiar. You cut interest rates. The idea is you cut interest rates and that makes it cheaper to lend, cheaper to consume, cheaper to invest—and that helps stimulate the economy. We don't usually find that we cut interest rates all the way to zero, and get to the point where we really can't cut them anymore. That, of course, happened this time, and that necessitated some creative thinking about, how do you support the cyclical recovery? How do you support the economy, when you don't have the interest rate tool to adjust anymore?
So, into play comes the things that are popularly known as quantitative easing—we call them large-scale asset purchase programs within the Federal Reserve. They were particularly interesting because, in many cases, they were aimed at two things. In some instances, those policies were aimed at replacing the stimulus that you might get out of cutting interest rates. But particularly with respect to the first quantitative easing program, it was also about addressing certain dislocations and problems in credit markets in particular—mortgage-backed security purchases, for example, directly supporting the housing market—and they were more than your traditional monetary policy tool. They were about the traditional monetary policy tool, but they were also about—not just making credit cheaper, but also addressing some real dislocations within the markets for borrowing and lending.
Davidson: What about the secular forces? Does monetary policy have much influence on those?
Robertson: Probably the most important influence is trying to prevent the demand factors [from] spilling over and becoming supply factors. So think of, for example, the labor force participation rate—so the fraction of the population that's engaged in the labor market, either looking for work or in a job. When a recession comes, unemployment rises, temporarily, and then when the recession ends, the unemployment rate starts to fall.
But there can also be labor supply responses at the margin of whether people are even looking for work or not. We saw in the last recession a large buildup in the number of people who were not officially counted as unemployed, yet they were kind of on the sideline because they were still saying they wanted a job, but for whatever reason—sometimes because of discouragement about the job-finding prospects—they left the labor market. Economists have studied this a lot, and found that once people leave—detach themselves from the labor market—unless it's for reasons like going to school, where you're going to be building your human capital, probably so that you can come back in and get a better job in the future—if you leave, there's a good chance you're not going to come back again.
The aggressive response of the Fed, if you like, was in a way to try to prevent that pool of people [from] leaving the labor force, from growing too rapidly and getting too large. And then part of the cyclical recovery was also bringing some of those people from the margins back in again as the economy strengthened and there were more job opportunities for people.
Altig: Let me follow up on what John said, with a couple of additional points—one very much related to what John was saying. We know, for example, that the longer someone is unemployed, the harder it can be to become reemployed. So the longer a downturn persists—the longer a recession lasts—the longer people may stay unemployed, and then it makes all that much harder to realize the rebound.
Davidson: It sort of feeds on itself.
Altig: Right. Another example would be a long recession tends to be associated with low investment by businesses, low rates of capital accumulation. One of the things that makes workers productive—and contributes to higher wages and better incomes—is, in fact, growing the capital stock through business investment. Again, the longer the downturn takes hold, the farther you get behind, in some sense, from building capital to where it really ought to be, and the harder and longer it will be to dig out of the hole.
Davidson: We had a deep hole, and it just continually got deeper, so it was tough to get out of.
Altig: Yes, it was very deep. Fortunately, in many respects, we have dug out of that hole. GDP [gross domestic product] fell by $600- some billion over the course of the recession. We're now approaching almost $2 trillion of GDP, adjusted for inflation—that's higher than we were when the recession started. The unemployment rate, which doubled—the official unemployment rate doubled—is now back below where it was before the recession started. Housing prices fell by 22 percent, or something like that, from the start of the recession to its lowest point. Now, housing prices, at least by some measures, have on average fully recovered.
So there was a really substantial recovery that took place. It just took a really long time.
Davidson: Right. And some of the reasons for that were, as you mentioned, extraordinary: coming out of a recession triggered by a financial crisis tends to be much worse.
Altig: There are some elements that haven't completely bounced back, germane to some of the secular issues that John was referring to—for example, productivity. Labor productivity is still quite far behind where it was in terms of the annual growth rate prior to the crisis, and that's a problem for these secular trends that John discussed.
Davidson: I want to switch gears just a little bit, guys, to a topic that has received a lot of attention lately—and seems to continually receive more—and that is inequality in the distribution of wealth in the economy. How do the cyclical and secular forces play into this question of inequality?
Altig: I think there is a general consensus, or the starting point of our thinking about policies that deal with issues of distribution of wealth and income inequality and the like. It starts with the presumption that monetary policy is fairly limited—in fact, quite limited—in what it can do; and another big...
Davidson: The old blunt instrument....
Altig: And not only a blunt instrument, but the notion that monetary policy is quite good at dealing with these cyclical developments we're talking about, and not so powerful, if at all, in dealing with the secular trends that will persist far beyond when the effects of a recession are over. So we turn to things like fiscal policies, workforce development policies, all sorts of policies outside of monetary policy that add to the skill level of workers, for example—which monetary policy doesn't address very well at all.
Now having said that, I think it's also fair to say that that's not a position we take for granted. We've locked ourselves into a situation where we're going to say, "Forever and always, we know monetary policy can't do anything about any of these challenges." And it is an ongoing debate, and it's an ongoing object of research within the Federal Reserve System and the broader economics community to try to understand whether or not monetary policy has a role to play—maybe for the reasons that John described. If we don't get the cyclical stuff out of the way, it affects the secular stuff down the road. And if it does have such effects, what is the bounce that we can expect from monetary policy?
Davidson: Dave, I want to follow up on that a little bit here. Is something like that as simple as saying, "Well, just keep pushing. Keep making monetary policy more accommodative in order to draw in even more people into the labor force"? I mean, is that the kind of avenue that that could go toward?
Altig: Well, it could, but I think we need to recognize the likely limits of that, and the unintended consequences of that. There is a position—one that I'm fairly favorably disposed towards—that if we go back to the '70s, for example. The '70s was a period where, much like what we think we're seeing today, there was a downshift in the secular trend of GDP growth that was unrecognized, and so what was sort of this downshift on the supply side, as John described it—policymakers thought it was a cyclical development and therefore attempted to use monetary policy to address it. Well, monetary policy wasn't very good at addressing it, and what we got was inflation that was both high and volatile, and the economy bore a large cost for that, and it was a very painful thing to get rid of.
So, trying to do ambitious things beyond your ability to reach those goals can create more trouble than not.
Davidson: Right. This stuff's tough, right? I mean, it's not easy to look and immediately understand exactly what's going on out there.
Robertson: One of the things I think is a good thing about the structure of the Federal Reserve—the dual mandate that the Federal Reserve is not the just the maximum employment function, it's maximum employment and price stability and moderate interest rates, and all that kind of stuff is its responsibility.
So it has to think about the consequences of its actions, because at some point there's trade-offs here.
Davidson: Everything has a side effect, right?
So, our annual report covered a 10-year period that ran through 2016. Now, not a lot of time has passed in 2017, and instant analysis can be a little dangerous, but in general terms, what have we seen the recovery do since the end of 2016?
Altig: Well, you've only got a quarter in the bag, so there's not a whole lot to say about that. I can tell you that our staff view, to this point, has been, what do we expect out of 2017? We expect a lot of what we saw in 2016 and 2015 and 2014. In other words, this sort of cyclical adjustment process has largely completed itself, and now we're on a growth path—add unemployment rates, in an economic environment that is pretty much steady as you go.
The first quarter, as we know it right now, did not come in particularly favorable to that view: it was a weak quarter, and in particular consumer spending was quite weak. But the early indications—and I want to emphasize very early indications—are that it was one of those "ups and downs" that happens over the course of the year. And the second quarter looks like it would be better than the first, and basically make up in many ways for the shortfall in the first.
So nothing has really changed about the story we told in the annual report.
Robertson: One indication of that is the labor market continued to strengthen. Employment chugged along at a pretty robust pace, the unemployment rate came down some more. Labor force participation—one group that was a particular puzzle, because we know there are some demographic forces that are lowering the rate of labor force participation at the time...
Davidson: Aging, and so forth?
Robertson: The aging of the population, the baby boom generation moving through the age distribution into retirement ages, pulling it down.
But then people who are in their prime working years—we also saw a pretty significant decline in that group's labor force participation during the recession that persisted for quite a while after the end of the recession. And over the last two years, that has kind of turned around. That makes my life complicated, because it complicates determining what part of that is cyclical—kind of a delayed cyclical response—versus just a return to a longer-term trend.
But, it's another indication that firms are hiring. Production, overall, since the end of the recession—the rate of growth in GDP has not been particularly strong. It's been okay, but it's not... If you have strong hiring and just okay GDP growth, that, by definition, means labor productivity growth is pretty low, and that comes back to the question of, why is that?
Because longer term, it's really an important issue, because the demographics are not working in favor of the growth in the labor input. It's going to be hard for the economy to sustain the pace of employment growth we've had over the last few years for the next few years. Just the growth rate of the population is not really enough, unless the unemployment rate continues to fall a lot—but then that's going to start to raise questions about, is that really sustainable? And is the economy...?
So we're going to have to really be careful going forward, as we think about can we have continued declines in the rate of unemployment in the face of so-so output growth. But of course, if so-so output growth turns around because productivity picks up, then that might change again the dynamics of this. I think one of the themes in the annual report is there's this secular stuff and the cyclical stuff, and they're all happening at the same time, and we don't get an independent read on each of those pieces in real time.
We continue to monitor stuff. And I think monetary policy has been pretty explicitly a data-dependent monetary policy for that very reason, that we're in an environment where you need to be looking at the data carefully and trying to think, is this what's fundamentally driving these movements?
Davidson: What would it take for productivity to tick up?
Robertson: Well, one of the things that's been a puzzle since the end of the recession is—we could think of labor productivity growth and breaking it into pieces, and one piece is the technology that you have available to give workers. But it's not necessarily even that. It could be just better management practices. You organize the workplace in a better way so you're more efficient in how you do stuff.
So you get some bang from that. You get some bang from the education of the workforce—you have a smarter workforce who can do more stuff—but a big driver, typically, has been workers get better capital, and they get better machines. To get the better machines, the businesses have to be investing, and the pace of investment during the last six years has actually been pretty weak on a historical basis. Firms are hiring faster than they are growing the capital stock, and so you're not getting the bang from having bigger ovens to bake bread, right? [We have] the same size ovens, the same technology.
And so, we're really looking for some turnaround in that investment piece as a key that's going to drive that piece of productivity higher, and then lift the overall productivity, because, as I said, longer term, we're not going to be able to rely on growth in the labor supply as being the driver of the output growth.
Altig: And I'll say that one of the sources of optimism, coming out of the election—we can pretty much date it to the election—was that businesses and consumers were both expressing greater optimism and, with that, sort of the hope that we would see more expending on plant and equipment, software, by businesses, following from this better sentiment. It hasn't happened yet, so we are still in "wait and see" mode to see if these dynamics that John rightly indicates are so important will eventually take hold.
Davidson: Okay. We've got a few audience questions, so we'll turn to a couple of those, mixed in with a few more we had on hand ourselves here. So an audience member asks: as part of the Fed's monetary policy and unwinding of the balance sheet, is there a possibility the Fed will cease paying interest on excess reserves?
I'm not sure you guys can give a real definitive answer to that, but...
Altig: Well, being the chief economist doesn't make me the chief. That's a policy decision. Excess payments on excess reserves are allowed by law. They are a primary instrument of monetary policy right now. I will say that, technically, as long as the balance sheet is large—and how large large is, is still sort of an open question.
Davidson: Probably large.
Altig: It is large now. [laughter] You have to use something like interest on excess reserves as your policy instrument, so there's a feasibility question, and there's the policy question. And right now, the feasibility question really says the answer to that is likely to be no, this is our instrument for conducting monetary policy, and will be the primary instrument for some time. Even though the federal funds rate is the target, the way to implement that is through excess interest on excess reserves, interest on reserves.
Davidson: So we've touched on some of the elements of this question already, I think, but it sounds like, do we continue to see moderate GDP growth ahead for the next couple of years—2 percentish? And what would it take for GDP growth to accelerate substantially from that?
Altig: I think the answer to that is both very difficult and very hard. Our internal staff forecasts see 2 percent growth going out for as far as the eye can see. What it would take to get something different, I think John has pretty much laid out the arithmetic there: it's going to be productivity growth.
So the bottom line of all of this is that, whatever is the solution to increasing productivity growth is the longer-run solution to a faster pace of growth in the U.S. economy—or any economy in the world right now, for that matter.
Davidson: So is it fair, Dave—maybe I'm oversimplifying—to conclude from that answer that this productivity puzzle is the most critical question we're facing at this point?
Altig: I would put my money there.
Davidson: Okay. Well, turning back a little here to the topic of the annual report—cyclical forces, secular forces—we saw those two diverge through the course of the recession, and even during the recovery. Is that unusual? Are they usually moving more together through history, or is that a tough question to even answer?
Robertson: You could think of the secular things as being the trend, and the cyclical things being the deviations from that trend, so they're always kind of like waving to each other in the night as they pass. [laughter] The goal is, yes. There's no cyclical piece, and it's all trend—I mean "goal" in the sense of...
Davidson: The ideal, maybe?
Robertson: Yes. In the long run, there's no distinction between cyclical and secular—although, as Keynes said: in the long run we're all dead, so....
Davidson: [laughter] The dismal science, right?
Robertson: Exactly. The cyclical is kind of those deviations from the trend, so when a recession comes, that's when you see the big cyclical movements. And then during an expansion, you're catching up to that trend again. We think, if you look at the labor market—based on what we see in the labor market—we think we have pretty much caught up to trend. There may be still a little bit of slack in there, in terms of... Another way you can think of it is: What's the trend in GDP growth? What's the potential rate of growth there, which is from that math of productivity and labor supply growth? And I think we're catching up to that trend as well.
And then the goal, from a policy perspective, is sustain that along the trend.
Davidson: Well, on that positive note, I guess we'll close things out. John, Dave, thanks for your time today.
And thanks for joining us. Please come back for future ECONversations. There's always economic information on our website at frbatlanta.org
Thank you.