Guo Kai: During the Jackson Hole speech, Chairman Powell refused to comment on the exact level of R*. But when you try to think about how restrictive the policy is, you probably need to have some idea where R* is and where it could be. Are we living in a different world now compared to pre-pandemic that R* has moved around a little bit, and where is it now?

Donald KOHN: There's a huge amount of uncertainty around where R* is. I think it's a useful concept, but maybe less useful because it's not precise, and we don't really know ahead of time. It's not as useful in policymaking as it would be if we really knew.

Whereas in the 2010s, it looked like in real terms R* was a half, very low. But now we have a situation with the growth in the federal government debt, with the investments in green technology and AI pushing out investment, that the investment-saving balance has shifted.

Although the demographics are the same, and so far AI hasn't really shown up in productivity—that would keep it down—I still think the savings-investment balance suggests that it's probably somewhat higher than it was in the 2010s, but we don't know.

Policymakers need to keep that in mind and not pretend they knew with precision.

1. I think most FOMC policymakers were comfortable cutting rates in September, particularly after the July employment report. That wasn't the problem, or that wasn't an issue. The issue is how much and how far to continue and how much focus (to put) on the labor market.

So, my sense from my discussions with some of the other policymakers was, he cut / caught this…just how intent he was on shifting the labor market and not pay attention… not being confident about inflation. They were not quite there yet.

Many of his colleagues on the FOMC in their public statements have talked about gradual and measured pace of easing. He didn't do that at all. So, I think they were with him 90% of the way he went, and maybe he went a little bit further than they expected.

2. The best practice of central banking is the target of forecast, because you know there are lags between what the central bank does and its effect on output and inflation, so you should be targeting a forecast.

But I think what happened during Covid given the unique circumstances globally, with the disruption of supply chain, shutdowns and all that kind of things, the forecasts of inflation—which were made that inflation would come back down very quickly as soon the supply chains opened up—were wrong. And inflation turned out to be much more persistent than the central banks and most private forecasters expected.

So, I think they lost a lot of confidence in the forecast, for good reason. And that meant that to figure out what to do, you had to pay more attention to the incoming data. Incoming data is always important because it would tell, even if you're operating on forecast, you might modify your forecast based on the incoming data.

But when you don't have much confidence in the forecast, it gives more emphasis on the incoming data, and especially the incoming data on inflation, as central banks are putting emphasis on getting inflation down. They've been wrong about it before. And that meant inflation is at the end of a long chain of causation between monetary policy and inflation. So, you were looking at the end of the chain, and it raises the risk that you would be late to act.

So, I think the central banks are migrating back to a more forecast-based approach to policy. When Chair Powell talks about the data, he says incoming data and “its effect on the forecast”, and that's important. I hope that this migration happens and it gets more forecast-based.

3. One suggestion I would have is to not publish the median and not use it in the press conference. Now, Chair Powell is always very careful to say, this is not a plan, this is what will happen if the forecast happens, and we can change our mind, we make decisions meeting by meeting.But I think figuring out a better way of communicating the uncertainty and dispersion.

So, I just wouldn't publish the median. Other people will calculate it, that's fine, but it's different when the Federal Reserve calculates and then uses it in their communication. I would try to find some way of looking at the uncertainty.

And then a number of people have suggested that the forecast be link so that a particular dot cone has a particular set of dots and a particular GDP and a particular…
And you don't even have to identify the people. At least as outsiders, you and I could observe, in general, how the reaction functions, how the members of the committee saw the reaction function. I think that would be constructive.

4. The 2020 framework was put into place at the end of the 2010s when inflation was low, interest rates were low, often at zero, there was a risk that inflation would be persistently below the 2% target and inflation expectations would drop down, so the nominal rates would be even lower, which would make the zero lower bound even more of a constraint.

All the emphasis in 2020 (framework) was how do we cope with the low inflation, low R* world?
There were two inflationary aspects to that. One was of an inflation averaging so that if inflation ran low, they would let it run above the target for a while. They didn't say how much and how long, but above the target. But they never said if inflation runs high, we're going to run it low. They just said you will make up for low, but not for high.

And the other asymmetry was with respect to labor markets.They said we think that employment is below its maximum level, and the unemployment rate is too high. We will pay attention to that. That will cause us to ease policy. But if we think the unemployment rate is too low, if employment is too high, that's not going to cause us to tighten policy. As I said, I had been looking at some of the policies of Volcker and Greenspan in the 80s and 90s. Particularly Greenspan's policies, but Volcker’s as well, were preemptive, which means when seeing tightening in the labor markets, let's raise rates before it shows up in inflation, and by having the asymmetry in the labor market, the 2020 framework kind of ruled out the preemption.

But they saw that as offsetting the other problem of the zero lower bound, but they had two ways of offsetting, and I don't know that they need it both ways. So, I would take a careful look at that. I would make sure that the new framework was robust to a variety of different circumstances, not just… It has to remain robust to the potential for a low inflation, lower R* world, but it should also be robust to a higher inflation and adverse supply shocks kind of world. And I think the 2020 (framework) wasn't, so they need to stress test their framework for that.

5. Another issue, which is relevant to China today and it was certainly relevant to the US in 2008 and 2009, is, when one of your sectors is cratering like that and subtracting from domestic demand, you've got to stimulate the other ones. It takes fiscal policy and monetary policy going all in, not saving…

We used to have discussions in the beginning of 2008 when it looked like things were shaky but hadn't really collapsed yet. Some people were arguing that we should keep some of our powder dry. We shouldn't use or we shouldn't be aggressive with easing. I think that was a mistake.

I thought we were pretty aggressive. I certainly argued in meetings for being aggressive, particularly close to zero lower bound. It's better to get ahead of that problem than letting it come. So, I think being aggressive with the easing to stabilize the economy even as the sector is adjusting is really important.

6. I wish I knew a way of leveling out that cycle, taking actions, perhaps building resilience and raising capital requirements when you see macroprudential policy, when you see a sector that looks very ebullient and growing very rapidly, that's when risks are building up, and that's when you need to build up the capital.

To some extent, the regulators and the banking system may have overreacted to the problems of 2006 and 2007, or at least they should have found another way. So, bank mortgage lending went only to the very safest borrowers. That, I think, has helped to contribute to the shortage.

Now there are lots of other things contributing to the shortage and the supply side of the housing market, including regulations about building and where you can build, the restrictions, etc. But I do think it's not so much a central bank problem, but perhaps a government problem, thinking about stopping the up part of the cycle and then not letting it overshoot on the downside so that 10 or 15 years later you're looking at the opposite problem.