In conversation: US inflation and policy outlook with Joseph Gagnon
Beat Siegenthaler: First of all, let me ask what you expect for the FOMC [Federal Open Market Committee] decision this week.
Joseph E. Gagnon: I don't expect any surprises. The Fed has said it needs to see significantly more inflation before starting to change its mind. We've had some uptick of inflation, but it's mostly base effects, a rebound from last year's unusual drop. The Fed is expecting that and not going to be alarmed by it. So there will be no change in their communication.
BS: Where do you stand in the debate on whether the inflation spike this year will be transitory or become more persistent?
JG: I do not expect persistent inflation of the type that we had in the late 1960s and 1970s; that's not in the cards because the Fed wouldn’t allow it. But I do think we will get more than just a brief burst of activity. We have two years in a row of the biggest budget deficits the United States has seen in our lifetimes since World War II. That's likely to have a big effect on the economy, especially if the vaccines are successful, the pandemic becomes much less of a factor, and the economy fully reopens. There are tremendous savings that people have piled up, sitting unused in their bank accounts. They might want to spend that later this [year] and into next year. And of course the government is funding all kinds of programs. I expect this to run for some two years. You can call that a temporary boom, but it's significant, and I believe it will push inflation up more than people generally expect.
BS: What do you say to those who argue that the global deflationary forces will continue to work against higher inflation, as they have for the last few decades?
JG: Larry Summers had a good analogy, that in the last few recessions the bathtub got filled up only a quarter or so. We didn't really pour a lot of stimulus into the economy, it wasn't enough to overflow the bathtub, not even close. But now we're pouring two or three times its size into the bathtub. It would be bizarre if it didn't overflow. It's just a matter of arithmetic. I do get that people haven't seen inflation in their lifetimes, they don't know what it looks like or what causes it. And what you've never seen you don't expect. But government stimulus is many times larger than even the largest estimates of the output gap in the economy. So it's bound to overheat the economy, the only question being by how much.
It's possible that a lot of the stimulus money will be saved, that saving rates will be permanently higher. In this case the bathtub won't overflow that much. But if the savings do get spent, it will overflow by a lot. There will be more spending than the economy can produce, and the mathematical result of that will be inflation. Basically, businesses can't keep up with demand, so they will have to raise prices, and they will have to raise wages to get more workers. The economy overheats and then you get inflation. The last time that happened was the 1970s.
BS: What about beyond the two-year time frame? I recently had Charles Goodhart and Manoj Pradhan here and they argued that higher inflation will be more persistent in the medium [term] for structural reasons as major demographic trends of the last few decades are reversing.
JG: I think there are two parts to that question. There is the US fiscal aspect, which is what I've focused on so far. And that aspect can overwhelm the global forces for a while. It's not like you can't have inflation in an environment of global deflationary forces. It is true that most countries have chosen policies that in this global environment gave them low inflation, but we're about to see that change in the United States. On the Goodhart argument, he's focusing on demographic forces as the baby boomers get older, pushing more people into retirement, saving less and spending more, which will reduce excess savings and push up aggregate demand and inflation. But one piece he is missing is that people are living longer, and that means they also need to save more. It looks like the baby boomers who are retiring now might indeed not spend as much as Goodhart thinks, and younger generations need to save more than before. So the two effects of an aging population and longer lifespans may kind of balance each other out. And so my guess is that in terms of global forces we'll probably stay where we are for a while with only a very gradual rise towards higher inflation and higher interest rates thereafter.
BS: In a recent interview on the RBA outlook I was struck by how much the new Fed strategy of focusing on outcomes rather than forecasts is taking hold at other central banks around the world. Do economic models and forecasts have any value anymore, as basically nobody knows how low unemployment can go before we get wage pressure?
JG: Well, Chair Powell has said that he doesn't want to trust the models. My own view is that the Phillips curve is still a curve, but that the original Phillips curve was highly nonlinear while the models we use are linear. That makes a huge difference if the real world is not linear. A linear model would say that if you have excess unemployment you will get a lot of downward pressure on inflation. Which is not what we have seen. A nonlinear model with a flat section when unemployment is high would say that no matter how high unemployment is, you get essentially no downward pressure on inflation. The problem is that in the past 30 years, advanced economies around the world have all had excess unemployment and been below potential output. Looking at inflation and basing judgements on linear models, central banks thought that we were close to potential, which turned out to be wrong in a nonlinear world. We all made this mistake, some more than others. Jay Powell, and I guess the RBA, are saying no, we are not making that mistake anymore. They say that in this world where the traditional model doesn’t work we need to search for the point at which inflation starts to rise. We need to search for the maximum employment, the lowest unemployment rate we can get before inflation rises. That's where we should always be aiming policy. That's a big change, and I think a good change.
BS: If the economy does overheat, at which point do you think the Fed will react?
JG: I think the Fed will be on hold pretty much all year because we have a long way to go to get people fully hired. We could see very big job growth numbers coming up. Employment will be rising by sometimes over a million jobs for several months, maybe even two million jobs in a month. By the end of the year we will be close to where we were before the pandemic. And in the meantime there may be some temporarily higher inflation, but not too scary, and attributable to temporary factors, which is not going to alarm the Fed. We're going to get strong growth with only slightly higher inflation. It's going to be great. The real question will be next year, say a year from now, March or April 2022, when it will become obvious that the process isn't stopping. That instead the process is continuing and we're blowing past levels of employment we saw two years ago and that price pressures are not subsiding, and that, if anything, price pressures are creeping up further towards 3 percent or more. At that point the Fed is going to have to make a decision. It will either have to change its policy to defend the 2 percent target, or—as I wish they would consider but probably won't—raise the inflation target. I think all around the world we have found that 2 percent is too low a target, that it should be a little higher. Not too high, but 3 percent or at most 4 percent, further away from the zero lower bound where we have been trapped with too low inflation. It's not good for the economy; a little more inflation would be better. But of course raising the target is not the most likely outcome, and most probably they'll stick with 2 percent and then they'll have to raise rates.
BS: I find it remarkable that there has been essentially no hawkish dissent at the Fed recently, is that at least partially because of the political backdrop in the United States with so much focus on the labor market?
JG: It's a good point, we haven't seen hawkish dissent, and there is indeed a strong focus on how the labor market works, that education levels and skill levels matter a lot in terms of how likely someone is to be unemployed, and how cyclical their job is. That is even more true in a pandemic recession, because people who have white collar jobs and make higher incomes have not lost their jobs as much. Almost all the job loss has been among lower paid workers. This has been true in other recessions, but it's much more true this time. And what the Fed is saying is, well, when we previously thought we were at full employment, we were really only achieving full employment for white collar workers, high wage workers. And that's not fair and we think that was wrong. So the Fed wants to push the economy higher this time, with more employment that will mostly reach the lower wage workers. The Fed has realized that they had not paid sufficient attention to the true definition of full employment.
BS: You have argued in a recent blog post that bond markets are not very good at forecasting inflation, so I guess you wouldn't pay much attention to markets painting a benign picture?
JG: Yes, together with Madi Sarsenbayev here at the Peterson Institute we looked at many decades of long-term bond data in various countries. And what we found is that neither the nominal bond yields nor the inflation compensation components ever predicted a significant move in inflation. Never. Those bond yields always moved simultaneously with inflation or even a little bit behind inflation. We couldn't find any example where that was different, really quite amazing.
BS: Since the Fed wants to get behind the curve by design, what’s the risk of a bigger, more disruptive response down the line?
JG: First of all, if inflation does rise more than the Fed thinks it will, I wish they would welcome that and just raise their target. That to me would be the best outcome. But assuming they don't do that and stick to their 2 percent target, it's true that the longer they wait, the more they're going to have to raise rates later. But I don't think they need to cause a recession. That's where I disagree with Larry Summers, who is very worried about inflation. Larry especially is worried about having to fight inflation in a harmful way causing a recession. That would be true if the Fed let inflation expectations become unanchored toward a much higher level, and then had to beat them back down. That was the story of Paul Volcker, of course, and to a lesser extent in a couple of earlier recessions where the Fed had not been anchoring expectations very well.
Right now though, inflation expectations are well-anchored and markets believe that the Fed will not let inflation get out of control. I think that's the correct view. In which case you don't need such a big policy increase to push inflation down. You just need to cool the economy off a bit, without causing a big recession. That was the lesson of the Korean War, which I wrote about in another blog post, where inflation jumped way up with a burst of spending, but came right back down, from zero to 10 percent and back. No one is predicting 10 percent inflation for the United States next year. But it happened in 1951, and by 1952 it had fallen back to 2 percent and then a year later it was down to 1 percent, and [when] there was no recession, people realized that it was a temporary burst of spending. I think something like that is possible now as well. The Fed doesn't need to be too aggressive to make that happen, but it will need to react a bit. The lesson from the 1960s and 1970s is that problems arise when the Fed ignores inflation and pretends that it can't be controlled.
BS: One question from a viewer is, why do you expect employment to keep rising through next year, beyond the pre-Covid levels?
JG: I'll just come back to the bathtub analogy: It's just measuring the amount of water. Look at the size of the fiscal stimulus, between 12-13 percent of GDP that just got dumped on the economy. And on top of that we have 7 percent of GDP in excess saving from last year that people weren't planning to save but did. That's 20 percent of GDP in spending power in an economy where the biggest estimate of the output gap right now is less than 5 percent of GDP. So when you put 20 percent of water into a hole that's 5 percent big, what happens? Well, it doesn't all fit. Maybe not all that 20 percent gets spent, maybe people will save a lot of it and only gradually ramp up spending. But the amount of water to potentially put into that bathtub is huge. It's much bigger than anybody's estimate of how much spare capacity there is. And so we will hire all those workers back and the output gap will be zero. And then there'll be still more spending out there, requiring to hire even more workers and raise prices.
BS: How likely is it that we'll get yet more fiscal spending via a large infrastructure package?
JG: There's a couple of things to keep in mind: First, the infrastructure spending will not be happening this year. It will only begin in 2022, as the law probably won't even be passed until late this year. The Biden administration's plan would have spending peak in 2024–25. So the infrastructure spending would ramp up several years down the road, when the burst of the stimulus we’ve been talking about would be long gone. So in that sense the timing would be right, not all piled on together. Second, tax increases are being proposed to pay for the infrastructure spending, making it much less expansionary. Third, the Fed will have more time to adjust and plan ahead, and if there was too much spending in the end, 2-3 years down the road, they could raise interest rates and that would be OK.
BS: Moving on to the issue of currency manipulation, you have argued that Treasury Secretary Yellen in the latest semiannual currency [report] “looks the other way.” Were you surprised that the Treasury shied away from putting the manipulator tag on anyone despite three countries meeting the formal criteria?
JG: Well, when Janet Yellen said at her confirmation hearing that she was going to be very aggressive against currency manipulation, I didn't really believe it. I didn't really think her heart was in it. At the time she made those comments, there had already been evidence that currency manipulation had risen a lot recently, but she didn't mention that. So that sort of clued me in that she doesn't really want to deal with it. And I think a lot of economists really don't get terribly excited about the topic. We can debate how harmful currency intervention is, and it can be a long debate. Personally, I do think there is some harm, and more importantly, it goes counter to the agreements that these countries have made. Members of the International Monetary Fund (IMF), which is just about every country in the world, have all agreed not to manipulate their currencies, not to prevent balance of payments adjustment. And then the Treasury came up with these criteria of currency manipulation: Do they have a large current account surplus, is there a big trade surplus with the United States, and are they buying foreign currency? So if all three criteria are met, well, the case seems pretty clear. But instead the Treasury said there was ”insufficient evidence,” which is kind of weird. I can't think of any reason why a country would meet these criteria unless it was trying to maintain a trade surplus and keep its currency from appreciating.
BS: Do you think countries will conclude that the Treasury will remain soft on this topic, that a country like China might get some degrees of freedom back in terms of managing their exchange rate?
JG: I think it's clear that one reason the Treasury didn't take the step of calling Switzerland, Taiwan, and Vietnam currency manipulators is because doing so wouldn't lead to any immediate consequences. It's kind of an empty threat. And so it makes you look bad if you say something that you can't really do anything about. I have argued that Treasury could develop tools with the help of Congress. They could get foreign exchange intervention capacity in order to do counter intervention if they wanted to. But that would require a change of policy and Congress to pass laws. The administration is not ready to go there yet, and it's not high up in their priorities. So if they're not going to do anything about it, then why name someone a currency manipulator? It would be meaningless.
I also think it's unfortunate that the IMF is not really doing its job here, as it has the same evidence and the same rules at work. But the IMF also doesn't have much it can do, as these are not the countries that would need to borrow from the IMF. They're lenders, not borrowers, which means the IMF has no leverage. It's a gap in the international architecture where we have this thing that countries claim they're not supposed to do, but no internationally agreed sanctions. As to whether China sees this as a green light, well, maybe, but I don't think China wants to get back to large trade surpluses. They're happy to have a small surplus, they like stability. They want to manage their exchange rate in a gradual way, to keep the surplus small, maybe above zero, but not by much. Currency manipulation requires a big trade surplus, which China used to have 10 years ago, though by mistake rather than design. It doesn't want that anymore and may also fear an outbreak of protectionism, as the rest of the world wouldn't tolerate it. So I think China is happy not to manipulate even without Treasury penalties.
BS: I got a series of questions on currency manipulation, including how to define it, whether the Treasury's methodology is the right one, and why sovereign wealth funds are not included? And also whether negative interest rates should also be considered currency manipulation?
JG: I'd say negative interest rates are definitely not currency manipulation. Any type of monetary policy, whether it be quantitative easing or negative rates, or just very low rates, may weaken a country's currency, but they don't tend to increase the trade surplus, because they also tend to increase spending inside the country. The key part of the definition according to the IMF is that you buy lots of foreign currency and have a big trade surplus. You're intervening in the market to influence the price mechanism to support a large trade surplus. What the Treasury has done is to put numerical magnitudes on that in order to determine how much is too much on each dimension.
The question of sovereign wealth funds I think is important. Neither the Treasury nor the IMF include sovereign wealth funds, which doesn't make any sense to me. When a country buys foreign assets via a sovereign wealth fund, it has the same effect on exchange rates as when it buys assets in its foreign exchange reserves. So I see no reason to exclude these funds. What you have is countries setting up sovereign wealth funds for the express purpose of avoiding being called currency manipulator. Korea, for example, by my definition would have been a manipulator last year, as I include their sovereign wealth fund. But the Treasury does not include it, which means Korea does not exceed that criterion.