Monetary Policy and Its Unintended Consequences

>> John Taylor: Welcome, Raghu.

>> Raghuram Rajan: Thanks, John. Shall I hop into it?

>> John Taylor: At least, let me say the title, Monetary Policy and its Unintended Consequences. Raghu Rajan from Chicago speaking all the way. Are you in Chicago?

>> Raghuram Rajan: Yep, I am in my office. Well, thanks, John, for having me.

Actually, I think you gave a previous talk at the Swiss National Bank and then you wrote a book in 2017. I think I was the guy who followed you in giving a talk. And they typically ask you to write a book. But I've been defaulting for a number of years, so this is why it's a 2023 book, rather than.

>> John Taylor: I wondered, okay.

>> Raghuram Rajan: So I finally put pen to paper and I did the easy thing, which is put together a bunch of speeches I've given over the years. The nice thing is the speeches have been pretty consistent. Monetary policy has its limits and downsides come from trying too hard.

So I think we're on a similar page, John. Alan Sanderson sent me this cartoon, which is kind of a summary of what I'm going to say. Which is, look, we don't need to go to the extremes on monetary policy to get reasonable outcomes that perhaps we can stop at some point.

So what's the details? Well, clearly we are in a period where we have high inflation coming down, of course, but worries going forward about potential fiscal and financial dominance. And the question that I ask is, do central banks have any responsibility for how we got here? And we also had some changes, most prominently in the Fed, of central bank frameworks.

Well, how's that played out? Partly on the inflation financial stability angle, but also partly on what happens next. Now that we are way above any norm on average inflation, are we gonna see a substantial contraction induced by the Fed to make up the sort of the backlog on inflation?

So I'm gonna use the Fed as an example, not because I think the Fed is particularly egregious in these matters. But it's the leading central bank in the world, and I think it offers a lot of data to see what's happening. So why are we in the current situation?

And of course, there's a lot that has to do with the supply side. Ben Bernanke emphasizes this. The pandemic, the consequences of the pandemic to goods supply, the extraordinary fiscal and monetary response which obviously affected demand. Basically a lot of stuff happened, but I wanna argue that much had changed before then.

The post GFC lowflation was an important factor, with USPC inflation averaging about 1.4% in 2012 and 2020, way below the 2% target. And I would argue this created a lot of pressure on central banks even before the pandemic to do something. From the political side, you could argue it was the commentary who bailed out Wall street and neglected main street.

What are you going to do about us? And if you're underperforming your inflation target, isn't there some stimulus that you're not delivering, can't you do better? And of course, central banks did try doing some things, including quantitative easing. And to some extent, this also then created pressure on other central banks.

I remember in Basel meetings, one European central banker who shall remain unnamed kept telling me the US, with its quantitative easing, is putting tremendous pressure on us to join in. Because otherwise we're seeing the euro appreciate and we really have to do something about that. Otherwise our growth is also being held back.

So even as some central banks expanded in accommodation, in which this built over. And this is a point that John makes in his book also. And central bankers didn't reject the responsibility of doing something. Remember the mantra, monetary policy is the only game in town? Well, essentially it put pressure on them to do something.

And given that they were already at the zero lower bound, the natural thing to do was to embark on unconventional monetary policy. I would argue that the first elements of that, which was repairing markets, which was consistent with what happened in QE one, was probably sensible and did have fairly positive effects.

It took the guise of Draghi and his attempt to prevent fragmentation of the European market with his ordinary monetary transactions. And his famous speech in which he ended by saying, believe me, could be enough. That created confidence that the ECB would support sovereign debt and elevated the prices of sovereign debt, at least for its.

This was, I think, what many would see as not necessarily a bad thing, and may have actually also helped restore bank health. But it went further with QE two, QE three, and then the pandemic QE in the Fed. But also with QE in the ECB asset purchases, including of government debt, in an attempt to somehow further activity through mechanisms that we can debate.

Forward guidance was also an important element. Trust me, we won't elevate interest rates until dozens such happens. And sometimes the two came together, QE essentially giving a time frame over which the Fed would remain on hold, with the kind of accompanying statement saying. So long as we're doing QE, trust us, we're not gonna raise interest rates.

And to some extent, I think the Fed actually respected this kind of message by waiting until it terminated QE before it started raising interest rates in March of 2022, and similarly with the European Central bank. And of course, some banks also engaged in directed credit programs. What the Fed did, which went further than other central banks, was it also changed frameworks.

I would argue that the way the Fed changed its framework was consistent with Paul Krugman's famous exhortation that. Central banks should commit to being more rationally irresponsible in an attempt to elevate inflation. And for the non-economists here, the idea is basically, if I am a central bank, that is preemptive words of inflation at the first sign in the future, perhaps that constrains inflation today and keeps it excessively low.

If, on the other hand, I say, look, I'm going to stare inflation, stare at it for a while, see if it's sustainable, not take any action, be more relaxed about it, then perhaps the prospect of future inflation will elevate current inflation. Whatever the rationale, it did turn out the Fed put in place aspects that would imply that it would no longer be preemptive.

The whole mantra that if you see inflation in the eyeballs, it's too late was thrown out of the window. Instead, you were going to watch it, measure it, look to see if it was sustainable before acting. So targeting average inflation over an undefined period helped in that, because it said once you saw inflation, you could afford to wait for a while until inflation caught up with the average that you were targeting because you had underperformed in the past.

Other measures made you more relaxed, included an employment goal which was broad-based and inclusive. Unfortunately, minorities are hired at the end of the job cycle, and so that made you, again, more likely to be patient to wait till minority hiring started and went on. And of course, rather than minimizing deviations from maximum employment, you made it one sided.

You emphasize shortfalls from maximum employment, as the Fed is concerned. So a variety of reasons how the Fed's framework changed and made it a little more relaxed about inflation. What were the micro-effects of these actions? Asset prices rose, leverage built up, and we have some evidence for the effects, sometimes in fairly positive directions.

The efforts to repair specific markets through explicit or implicit guarantees. Implicit guarantees of the kind that Draghi gave, seem to work in recapitalizing banks and driving more lending. And directed lending, which some central banks did, also expanded the flow of credit to targeted segments. There are papers on all these.

The macro effects, as some of you in the room have pointed out, are harder to discern. Perhaps the effect that I believe most strongly is what was determined by Arvind Krishnamurthy and Annette Vissing-Jorgensen, who suggested that QE seem to signal that monetary policy will remain low for long.

Again, the point that I made earlier, so long as the central bank is buying assets, it won't raise rates. And there seems to be some evidence of that. Harder to find evidence more generally, including on real activity. And there's this tongue-in-cheek paper by Farbo Lubos pastor, my colleague and others, which examines 54 studies on the effects of QE on output and inflation.

And basically finds there a mixed bag central bankers that did QE typically report statistically significant QE effects on output, but only half the academic papers do. What I thought was the most interesting finding was the Bundesbank, which has historically been against QE, is the one central bank which finds an even lower effect than the academics.

So the bottom line is this is an area where it's really hard to find effects. And I think if it's hard to find effects, generally, you have to worry about whether it's an effective tool and should be used in the future, because it also has downsides. One of the downsides didn't seem to matter early on, which is you tend to anesthetize government bond markets because there's a big buyer.

How much of a role that plays, I think we will have to wait to find out. But certainly it seemed that this might play a part if there was pressure on the government to raise financing quickly, as turned out to be the case during the pandemic. Now, the other problems that emerge once you start thinking about the way central banks operate, is that when central banks buy long-term government debt during the process of quantitative easing and issue reserves to finance it.

You're essentially financing a long-term asset with a liability which has infinite maturity but zero duration. It reprices every day. That's what a reserve is. And what that effectively does is shorten the duration of the consolidated bank government balance sheet. Another way of saying this is the central bank does a kind of asset swap with the public by taking their long term paper and giving them short term paper in return.

But when you consolidate the balance sheet of the central bank and the government, essentially the long term paper the government has put out has been converted in some measure to short term paper. So in a period where interest rates on long-term debt were really low, you were adding to the process of shortening the maturity of government debt.

One example worked out in the UK is when you look at the UK's average debt maturity, it's 15 years. When you looked at the median maturity, it was 11 years. But if you looked at the median maturity after accounting for QE, it went down to 4 years. Why does that matter?

It matters because when interest rates move up, it very quickly leads to higher debt service, higher fiscal deficits, and potential fiscal dominance of the kind that sometimes John talks about, I think more in terms of perhaps becoming a factor in central bank thinking as it raises interest rates.

Because this will impact government budget deficits in a serious way also, and therefore limit central bank policy. Other effects, I've talked before in this forum about the consequences of commercial bank balance sheet expansion to keep pace with central bank balance sheet expansion. Let me just show you some graphs, especially since we've had Silicon Valley Bank earlier in the year when central bank balance sheets expand.

Commercial bank balance sheets also have to expand to finance that central bank balance sheet expansion. It's not given, but it does happen, and I'll show you in just a second. Basically, commercial banks increase their holdings of reserves financed with uninsured, demandable deposits. They also reduce time deposits, at least that was the case in the US, and increased credit lines effectively.

You've got a commercial Bank balance sheet, which on the one hand has many more liquid reserve assets, but also has many more runnable liabilities. Here's what the graphs look like. What you have in blue here is reserves. And the vertical dashed lines represent QE1, QE2, QE3. Post QE is the period the Fed did relatively little.

QT1 is when it actively shrank its balance sheet. Turns out, in terms of bank reserves, both contributed to a decline in bank reserves. But then you had in September 2019, all hell break loose in the financial markets. The Fed went back into quantitative easing to supply reserves to the banking sector.

Tha'ts what happens in 2019. And of course, in March 2020, the Fed goes back full steam with quantitative easing as the pandemic hits and increases its reserves once again to a very high level. So that's the reserve buildup after 2022. It started coming down sharply until Silicon Valley Bank happened.

After which, bank reserves went up once again, because the Fed supplies reserves to the banking system through special facilities. If you look at demand deposits, if you look at deposits more generally, you see them also going up as reserves go up, but in a much more secular fashion.

When the reserves go down, the deposits don't go down, they keep going up. And what is particularly interesting is credit lines. These are undrawn credit lines. These are also claims on bank liquidity issued by the commercial banks. They also go up steadily with the various periods of QE and also with the pandemic QE.

In other words, not only are banks issuing deposits, they're also issuing other claims on liquidity directly to firms, which they can call down in times of trouble. And they do call them down in times of trouble, as you can see. As you saw during the initial days of the pandemic, we break down deposits into time and demand.

What you find is time deposits are coming down while demand deposits are going up. But within demand deposits, it's most prominently the uninsured demand deposits. That's the dark line, which is going up tremendously over time, and that's been a source of concern. Now, if you look within banks, the uninsured demand deposits have been going up.

Any question?

>> Male Speaker 1: The uninsured, insured, some of that, given you have this over 15 years or something. Some of that's just the fact that the insured limit is unindexed. So it'd be a natural upward drift anyway, as people got richer and held more.

>> Raghuram Rajan: Well, it went up from 100 to 250.

 

>> Male Speaker 1: But before this period.

>> Raghuram Rajan: Yeah, yeah, yeah. But it went up from 100 to 250 and stayed at 250 for much of this period. So that's not a big factor. But you're right, we should point out when that changed. But what is interesting is that the uninsured is going up particularly fast for the small and medium sized banks.

It's going up for all banks, but it's going up for the small and medium sized banks most. That's the green line, the below 50 billion bank size. I mean, 50 billion is pretty big size, but their share of the uninsured is going up more. What is particularly worrisome is if you look at their share, their uninsured demand deposits.

And it doesn't matter whether you add credit lines to this or not, you get a similar picture. And divide by their access to liquidity, that is, their reserves, as well as the eligible assets that they can take to the Fed and get liquidity directly. You see that for the small banks, the below 50 billion, it's been ratcheting up considerably fast, and in fact is now, just before the Silicon Valley Bank happened, was at a level consistent with the large banks.

The large banks have been getting safer in the sense of their ratio of claims on liquidity to available liquidity was much higher, has been coming down substantially. Some of this is the effect of LCR regulations, etc, on them. Medium sized banks have been about flat. But the big concern is small banks have become much more exposed to liquidity, at least by this measure, which is the share of claims on liquidity divided by available liquidity.

What this means is, when there is a shortage of liquidity, a dash for cash, many of these banks are out on a limb. I mean, the most egregious example was Silicon Valley Bank, with 97% uninsured deposits amongst its deposits. But there are a lot of banks which, though not so much on limb, are pretty exposed.

And when everybody is running for cash, for reserves, you are the person who finds it hard to get a seat, especially if at those times, you find that banks also start hoarding. So this is again an unintended consequence of quantitative easing. You are changing risks in the commercial banks.

And importantly, this is not taking into account any kind of search for yield which may entail these banks also investing in long term, high duration securities in order to eke out an additional return. Which again leaves them exposed to losses when interest rates go up. We've seen that too happen.

But even without that, you have a lot of liquidity exposure, which creates potential problems. So this is all sort of set of arguments to say much had changed even before COVID. Remember, even the change in the Fed framework was well underway at that time. And of course, COVID created a whole set of new problems, real disruptions and massive fiscal expansion and the accompanying monetary easing, including balance sheet expansion, to deal with the early days of COVID.

What this meant was post recovery, and this is familiar ground to many of you, given your changed framework, given that you decided to wait till QE was done, you couldn't react to inflation early. And so the Fed delayed and had to catch up. Risks were higher in the banking system.

We've already talked about the liability side changes. But there were also substantial asset side changes, some of which Amit Seru and his co-authors have documented. Carry trade, motivated by a search for yield, essentially caused them to invest in long term assets, which essentially became much lower value when interest rates went up.

So as the Fed sort of moved, when the economy recovered, started raising interest rates and shifting QE to QT, liquidity risk and solvency risk came to the fore. Deposit inflows turn to outflows, and we saw the problems in Silicon Valley Bank. Now, the failure of these four banks may seem like small potatoes, right?

After all, we've solved the problem, banking system risk is no longer a headline concern. But remember, there was unprecedented Fed and Treasury intervention to make this happen. Effectively, all uninsured demand deposits were insured by the statements of the Fed and the Treasury. There was a new lending facility from the Fed and over $1 trillion in lending by the federal home loan banks, which effectively are a way for the Fed to intermediate funds to the banking system.

What this did was reserves were coming down at the banks, down to 3 trillion. They've gone up to 3.7 trillion. And incrementally, as the Fed engages in QT, where the reserves are coming off is really of the money market funds and not of the banks. What happens when the money market funds no longer have reserves at the Fed and it starts cutting into bank reserves once again?

We'll have to wait and see what happens then. But what Fed intervention has done is converted a possible panic into a slower burning problem. If long term interest rates come down significantly, maybe there's no solvency problem anymore. Because all the losses that Amit and his co authors have documented may in fact be absorbed reasonably easily.

If, however, interest rates go up and stay up higher, what you have is a much bigger problem of solvency coupled with liquidity in the banking system and potential financial dominance, as the Fed's ability to raise interest rates becomes more constrained by the effects on the financial sector. Now, what bottom lines do we take away from this, I think financial stability and monetary policy have become deeply intertwined.

Many central bankers would not agree, they keep saying there's a separation principle. The central bank worries about monetary policy, focus on economic activity, and inflation. That's all it cares about. And the supervisors focus on financial stability and make sure that doesn't become a problem. Well, we've seen that not working out particularly well.

When the central bank has its foot firmly placed on the accelerator, it's very hard for the supervisors and the regulators to somehow reduce risk taking within the system. And time and again, we've not seen this work out. One paper recently that caught my eye on this, which I want to just show you the principle graphs there.

It's a paper by Jose-Luis Peydro and co-authors from the Bank of Spain and elsewhere. And the left-hand graph, the vertical axis is interest rates, and the vertical line is when a banking crisis starts. And what they show is there's typically no pattern in interest rates when banking crises don't happen.

I don't know what you wanna take away from that, but before every banking crisis, they find a u-shape in interest rates. First, interest rates are cut. That's when policy becomes really accommodative, that's when the risks start being taken. And then, perhaps because of inflation, policy tightens considerably. And that's when the crisis gets set off, and then things go downward for some time.

That's true of every crisis, they look at a large number of crisis over the 21st and 20th century. They have different crisis definitions, that's the second graph. And they look at post world war crisis. The picture is always the same, easy money followed by tight money typically leads to a crisis.

Now, I don't know if there's a theorem or if there's a rule that comes from this. But it does say that it's very hard to effect separation of monetary policy and financial stability. Of course, we haven't even talked about how these unusual monetary policies have perhaps created some kind of fiscal space and even fiscal dominance.

Consequently, I mean, what was the role of QE in limiting fiscal concerns as Congress and parliament spend? I don't know, I think it's worth finding out. And of course, we've already talked about the consequence of QE and raising interest sensitivity of the consolidated debt. And whether there's some unpleasant fiscal arithmetic to come, again, we have to wait and see.

So, last point, and I'll end there. What should central banks do on frameworks going forward? That's the last question I want to tackle. Augustin Carstens, from the BIS points to a way of categorizing inflation regimes. There's a low inflation regime when price shocks do not feed on each other.

And inflation may be low, even too low and below targets. And there's a high inflation regime where price shocks become more correlated, and you get generalized inflation quickly. In which case, central banks need to react quickly and preemptively, almost to head off generalized inflation. Certainly, our recent experience has been of this kind.

Whether you can write down a model to capture this effectively, I will leave as a question. But if this is true, if you have the low inflation regime, remember 2014 oil prices went through the roof, but inflation did budge. It somehow seemed to absorb all that without inflation.

Maybe expectations were entrenched, but more recently that has not been as strong. It's true that we have had sharper inflation rises. And of course, one of the worries right now is if we have another oil shock, will that upset this process of declining inflation, even though it is merely a supply shock?

So maybe, if this is all true, you need a different framework for each regime. But of course, how do you keep changing frameworks? You have the very relaxed regime when you want to pump up inflation. That's the regime we're in right now. Average inflation targeting, at least in the period when past inflation was really low, gave you the ability to accommodate inflation.

But now maybe we need a stronger regime, which is more anti inflation, the preemptive regime, which we are in the past. How do you shift? And really it's impossible to shift. You can't keep shifting frameworks. You need one framework for all regimes. So do you adopt the relaxed regime that we have currently, or do you adopt a preemptive regime that we had in the past?

And I would argue maybe the answer has to do with transitions. The transition between a low inflation to a high inflation regime under a relaxed framework is really problematic because you wait too long. Financial risks build up when the policy is accommodative, and financial instability can result as you try and catch up by raising rates quickly.

Maybe we have been lucky this time. But if we wanna take the lessons from all this, I would say perhaps we should fix the framework that contains high inflation and not worry too much about low inflation unless it becomes galloping deflation. We don't have too many cures for low inflation, but whatever cures we have may be worse than the disease.

Maybe we learn to live with low inflation while not making too much of a fetish about it. Bottom line, I think, is central banks perhaps can achieve more by doing less. And I'll stop there. Happy to take questions.

>> John Taylor: So, the first question is from Bob Hall.

>> Bob Hall: So you didn't, say anything about real interest rates, but in principle, they have a central role over this period.

The short term real interest rate, the one year interest rate in the US at the beginning of 2022 was minus 6%, which just an intensely expansionary value. And since the economy was already at full employment, that's the source of the inflation. So it's self building. This instability is well known, but people just don't talk about it.

That's reversing now, of course. The policy rate was in real terms. And for some reason, the Fed never talks about the policy rate in real terms, even though in the Taylor rule, it's written out in real terms plainly. So there's just a huge error in thinking, and there's been a big swing of rising real interest rates.

And I think we're beginning to see the consequences of that in real activity. But of course, we're also seeing the disinflation. And the disinflation is traceable directly to the rapid increase in real interest rates that has occurred.

>> Raghuram Rajan: I don't disagree with anything you've just said. I guess the only thing I would add to what you said is, why did they not act when real interest rates went seriously negative, and they saw that this would have consequences?

And I think they were held back in part by their framework, in part by their attempt to validate quantitative easing as an effective tool. Wait till you end it, we said we would end it by this time. Or we have to say that we'll end it before we start raising interest rates.

So I think they were prisoners of a bunch of things, including perhaps the hope that somehow all this would go away and they wouldn't go back to low inflation without them doing too much.

>> John Taylor: Question here, and then Bob.

>> Male Speaker 2: On the bank sector, you've laid out how the rising interest rates are causing bank distress with the Silicon Valley episode.

But also, we were worried about the health of the banking sector when interest rates are very low in the previous regime. Profitability was severely compressed in banking sectors. So is it not fair to say that the policy rate, consistent with a healthy banking sector, seems to be somehow shrinking, or at least much more constrained than, say, in the 90s?

When we had bond sell offs in 1994, which did not cause a banking crisis. And I just had one very short point. I was wondering if you could elaborate on the point that the average maturity was so severely shortened by QE, and what in your view, the underlying policy rationale was to compress the maturity so drastically.

Thanks.

>> Raghuram Rajan: So on the first point, what is the golden range of policy rates in which banks can thrive? I think that's an important question. It's hard to pick a number there. But it is true that certainly in the United States, partly as a result of a lot of being imposed on the banks, banks are becoming less and less of balance sheet creatures and more and more off balance sheet creatures.

So for example, a lot of the loans that banks make are sort of put into collateral loan obligations. Those become the balance sheet vehicles to hold the loans, while banks then make fees from the origination, but also make additional fees from offering lines of credit to these collateral loan obligation vehicles.

So that they become emergency funders rather than funders all the time. And why are they emergency funders? It may have something to do with their access to central bank liquidity, which makes them more effective emergency funders. So it may be that these consequences are less important for US banks than for European banks.

But certainly, there is an issue of what the appropriate range is within which banks can flourish. The problem with too low for too long is banks tend to then fund a lot more risk in an attempt to eke out interest rates. I mean, you saw the specter of Silicon Valley Bank taking duration risk because there's nothing to earn on short term reserves.

Why not take ten year bonds? After all, interest rates have been low for long, and they'll stay low for long, and we'll be fine. And obviously, that's a huge mistake. The point about QE is, let me try and make it as transparent as I can. The central bank buys government paper from the hands of the public.

So let's say it buys ten year paper. And then it, for the sake of simplicity, it issues reserves to the entities it buys the paper from. So, from the perspective of the public, they've exchanged ten year paper for overnight reserves on the central bank. From the perspective of the government and the central bank combined, the liabilities now consist of the ten year paper that is still outstanding with the public, plus a whole bunch of reserves that the central bank has issued.

So the average duration of the debt of the government is the weighted average maturity of those which has come down because the central bank has done this financing. Now, the point, of course, that maybe John Cochrane would raise immediately, is why doesn't the treasury extend the maturity of its debt?

Well, part of the rationale for why the central bank wanted to buy this long term paper out of the hands of the public was some sort of thought that there would be a portfolio rebalance, that they would go towards the long end in other investments, maybe in private sector lending or in other ways.

So therefore, it wouldn't particularly be effective QE if the Fed took the long term paper out of the public's hands and then the treasury issued it back into their hands. You wouldn't get the portfolio rebalancing. But some of our work suggests you don't get the portfolio rebalancing for another reason also.

Which is the banks themselves shorten the maturity of their liabilities and become less positioned to fund long term assets when QE takes place. So that's another reason in addition to what I just said.

>> John Taylor: Dan, sorry, then John. Dan?

>> Dan Kessler: Thanks, Raghu. I just wanted to get your reaction to a recent book and ask if you've taken a look at it and what you think, called The Lords of Easy Money by this guy, Christopher Leonard.

You familiar with that? And what do you think if you are?

>> Raghuram Rajan: I am not. If you tell me what the gist is, I could react.

>> Dan Kessler: Okay.

>> Raghuram Rajan: But I'd prefer to read it and then react more effectively to you. So let me look for it.

>> Dan Kessler: Okay, the gist is that it's sort of a journalist's anecdotal accounting of the real effects of the zero interest rate policy for so long.

And I was just wondering if you thought this guy was a crack pot or not, but.

>> Raghuram Rajan: I can't say without reading. But what I'll say is we came out with a report from the group of 30 maybe a week back. The report is authored by Marcus Brunemeier, but a whole bunch of us have signed on to it, and that basically makes this point.

Look, let's not try to do too much as central banks. And some of the instruments we're trying to create low for long, low forever, actually have side effects. It's not one directional and we have to worry about some of those.

>> Dan Kessler: Thanks.

>> John Taylor: Okay, John Cochrane.

>> John Cochrane: I want to make a couple comments.

Raghu, this is great. One, I'm glad to infer you're joining the team that sees the long period of the zero bound is not a particular problem. There was back in the time tremendous worry about here comes the deflation spiral. Here comes the deflation spiral. Tremendous, we need to provide more stimulus.

And in retrospect, I think that was all mistake. A wonderful paper by Bob Hall and Marianna Kudlyak, which you saw on Monday, essentially saying we were at the natural unemployment rate the whole time, which I loved. Second point, it is interesting, why did they not move faster? Now, some of it was they had pledged average inflation targeting, whatever.

But some of it was surely failures of perception. They kept saying, it's just supply shocks, it's just particular prices, it's just things that'll all go away. Their forecasts never showed the inflation. So the long, it's not just the working out of the promise for average inflation targeting, there was also a tremendous lack of perception that somehow needs to be fixed.

I'm curious at your view of the end of inflation. Bob Hall said, which now I'll disagree with Bob, of course they raised real interest rates, inflation went down. But the mechanism there is supposed to be raise real rates. That hurts the economy and the magic of the Phillips curve brings inflation down.

We didn't see that mechanism. The economy is still booming and we've just barely got interest rates above. So my sort of view is the Fed's jumping out in front of the parade here. The parade's ending on its own and the Fed is claiming victory. But I'm curious on your view.

And the last one, which will actually be a question, going forward, I think the Fed is thinking what's our new framework? Powell announced the search for the new framework. I think the ECB is doing the same. And I'm curious what you think on the new framework on two dimensions.

One, the joint fiscal, joint monetary policy and financial obviously have to talk to each other. You didn't mention the hilarious part of the UK where pretty much on the same day the monetary policy people said, we're going to quantitative tightening, and the financial stability people said, we're going to quantitative ease, and we're gonna do both at the same time.

Our stress tests were asking about interest rate declines as the monetary policy was asking about interest rates going up. Clearly, these two people need to talk to each other. But the second issue for the framework is, are bygones still going to be bygones? And I was very interested.

You mentioned the heretical possibility that maybe average inflation targeting works in the other direction. The American voter understands that price level is not inflation rate. Even if some of our politicians don't understand the difference between price levels and inflation rates, should a new framework going forward? First of all, it needs, the old framework was a beautiful Maginot line against deflation that forgot about, what if the Germans come through Belgium again?

Perhaps we need one that has an inflation part and a deflation part. And perhaps we need to think about how much average inflation targeting or price level targeting or whatever are bygones. Even 10% bygones are gonna still be bygones in the future. So the question there was your thoughts on the new framework?

 

>> Raghuram Rajan: Yeah, okay, let me start backwards. The old framework is precisely going to raise the question that you raised, right? Are bygones going to be bygones, or do we have to create a deflation in order to compensate for the excessive inflation in the past? And almost surely the answer is, we're going to change.

You've seen very few people talk about average inflation targeting at the Fed over the last few months. And I think that's consistent with the idea, it is history.

>> John Cochrane: I think they've rediscovered the Taylor rule.

>> Raghuram Rajan: Yeah, in some ways. On the UK, it's actually very interesting because they squared the circle.

The point you made, QT on one side with the monetary side and QE on the fiscal side, now, they can actually do that. Because they have two different committees and they can say one committee proposed QT and the other committee, though the Bank of England governor is the chair of both committees.

What was clever on the part of the UK was they essentially said this would be temporary. And actually the governor announced an end date, which was Friday. Obviously, the big question is what would have happened if they hadn't been successful by Friday? But I am given to understand they actually knew that if they did as much as they did in the interim period, that was enough to take care of the problem and they could actually exit on that Friday.

So it was a very targeted operation. But absolutely, this is the problem more generally. John, you asked me a question last time we talked on why the Fed couldn't keep intervening on a daily basis and provide the reserves that the market needed as and when it needed it.

Part of the problem is when you have a huge deficit, which is likely to happen when you withdraw reserves in a big way, but you haven't shrunk the deposits or the lines of credit as much, is that mismatch, that gap can become prominent at any point, creating the dash for cash at that point.

The safe way to deal with it is to reduce the cap by providing more permanent reserves, which means expanding your balance sheet once again. If you leave it to daily supply, then you make it a matter of, does the Fed recognize the problem every day enough to provide what is needed?

And we're talking hundreds of billions in supply rather than the few billions it has to do every day. In other words, it becomes a more semi permanent problem rather than a problem of short term disruption. And that's why I think the Fed, after September 2019, to expand QE once again.

We could talk about this more, but I think that's on why inflation has come down. I'm with you in that activity hasn't slowed as much. What seems to have slowed considerably are the supply disruptions that took place during the pandemic. And now you're seeing some action on rentals.

Also some of the surge in rentals that took place shortly after work resumed, I think that's quietening down. So it may well be some of all this quietens down without that proverbial slowdown, significant slowdown in activity. Will the Fed sort of stop at that point and say, mission accomplished?

I doubt it. I think it'll want to build some slack in the labor market.

>> John Taylor: Okay, Mike Boskin has a question.

>> Mike Boskin: Yeah, I have a couple of questions and observations. First is, since the Fed adopted its 2% target, improvements have been made in the measurement inflation that have reduced it by about 40 or 45 basis points.

So maybe 1.4% wasn't so bad. That's number one. That's just a simple fact. They were also issuing a lot of different types of forward guidance, dot plots, etc, that led a lot of the banks to assume that rates were gonna stay low for a long time, which kind of reinforced the behavior.

And Michael Bordo might wanna comment later on the fact that we've had long periods. Now a lot's changed, we didn't have interest on reserves, we had Reg Q. Most of the credit is in the financial markets, not in the banks, etc. But we had long periods. We had 1% inflation and good growth.

So just want to come back to that question. It's getting closer to John. Not quite all the way. Maybe I'm heading that way asymptotically if I live that long. And then I think two other points I'd like specific questions I wanna ask. Number one is you kind of mentioned this framework, this rethinking, etc, that led to the relaxation.

But you were pretty quick and not fully complete on what led up to that, on their thinking about how they had to rethink, how they dealt with an era of low interest rates, etc, and how it became so asymmetric. I wonder if you want to add anything to.

Maybe it was reasonable to take a look at it, but were the conclusions at all reasonable? I mean, some of that was the work of our former student John Williams, by the way. So just kind of interested about that. And lastly, I think something you kind of inferred but didn't really talked about was we had central bank governors actively encouraging deficit spending.

And which is a big reversal from the traditional role central bank governors have played when they've lectured our Congress on not getting involved in so much deficit spending. Now, you can argue in the midst of a collapse like the government ordered lockdowns and COVID, it was temporarily a different situation.

But it seemed to me that they also relaxed their concerns about fiscal policy, and that was not a good signal. So I wonder if you have any sense about that and what the appropriate role of a central bank governor is when we talk about those two things.

>> John Taylor: Raghu, you wanna react to that?

 

>> Raghuram Rajan: Yeah, absolutely. So inflation, 1%, not so bad. Absolutely agree with you. And I think the big concern is galloping deflation. Not seeing that, even in Japan. And we have to rethink whether it's, we need to be in such a busy, especially when we don't have strong tools to elevate inflation when it's low on the framework.

I don't know the thinking that went into this. But I agree with John Cochrane that, in a sense, this became the wrong framework for an era of higher inflation. And perhaps, I don't know how much thinking went into this, assuming that inflation somehow the biggest problem we had to deal with was low inflation and they could stay that way.

I mean, I think with the benefit of hindsight, we will have something which is more robust to regime. And how that will play out I think would be very interesting. On the fiscal policy, I mean, I sort of threw this out as a possible concern. That is, with central banks not wanting to seem to be an enemy of the people, given how much they were involved in the rescue of the hated bankers and without too much of an explanation as to why that might be necessary, I suspect they were less willing to enrage Congress by saying we're gonna stand against unwarranted spending.

And of course, some of them, as you said, actively encouraged more fiscal, partly with the thought that it would take pressure off the central bank to try and elevate activity. Which some of them felt was necessary when they were the only game in town. So, I think I'm with you that especially at this time, when you see the deficit at a level which is unprecedented in non crisis times.

Combined with the Fed which is actually trying to slow activity, it seems this is textbook. No, no, they shouldn't be working at cross purposes.

>> John Taylor: Krishna, go ahead, Krishna.

>> Krishna Guha: Thank you very much. So masterful as always, Raghu, slightly depressing. I think if we look at the catalog that you detail.

Just a couple of thoughts, first of all, on the instruments. If QE doesn't really work, we don't like forward guidance because of all the problems that we've experienced with it. We're also not that thrilled about having governments jump in in the kind of way they did in this last go around.

What are we supposed to do if we are still in, or return to a low r star world when we suffer serious adverse shocks? Are we just assuming the economy is going to stabilize on its own? So what are the instruments, that we have with respect to shortcomings in the framework?

I think it's important to recognize that there are sort of multiplicative shortcomings. And you detailed those well, it seemed to me absurd that we should abandon preemptive forward looking policy. And that was really a separate question from whether we wanted some form of averaging, whether we'd been disciplined enough in terms of specifying or not.

When we think about what is consistent across all these regimes, it's the desire to stabilize inflation expectations at target. And I wonder if what we're sort of groping towards is some kind of focus that's a little less on the inflation path per se, as really trying to find a more disciplined way of managing expectations going forward.

So I just appreciate your comments on.

>> Raghuram Rajan: Yeah, I think they're sort of related, right? I mean, what can you do in a low r star world? I would say you can keep interest rates as low as you can get, maybe even go negative, though we don't know if some of the benefits of going negative outweigh the costs.

The Swedes were quite willing to move out of negative quickly because they didn't find it particularly attractive. But what I'm saying is, it's not clear to me you have much else in the way of instruments without starting to interfere in pricing. Once you enter into various markets or tying your hands, once you go into some kind of long term guarantees, we will not raise interest rates until thus and such happens, guidance etc.

And we've seen the consequences when the regime shifts on you. But we haven't seen you independently being able to move the regime through all this activity. I would say there's a fair case to going back to boring banking, central banking. Where you say, I've done what I could and now its up to everybody else.

Now, who is everybody else? Well, fiscal, there's much less room going forward as you know, given the amount. I mean there is still the old IMF saw, which is structural reforms. And clearly with green, with some of the community issues, the underdevelopment in various communities and so on.

There is a possibility of structural reforms also helping elevate growth rates. Hopefully, I mean we don't get back into a low r star world because of everything else that's going on. Including hopefully some of the positive stuff, which is productivity gains from all the good technology that is coming.

But my quick answer would be, do what you can. But don't get into, don't try a whole lot of magic things. Maybe there are some problems you can't solve, and you have to leave it to the others, especially the political authority to do what they can. They have fiscal room.

If they don't have fiscal room, also contemplate structural changes.

>> John Taylor: Okay, next question, sorry, behind you.

>> Male Speaker 3: I want to come back to the statement that the central bank should do better by doing less. And ask about sort of a few throwaway sentences that you had about the federal home loan banks and the recent banking problems in earlier 2023.

Which is maybe the central banks that try to do what it's supposed to do well, before it ventures into other things. So that it seems to have failed pretty miserably on bank supervision regulation both in 2008 and now again in 2023. So I wanted to hear more your thoughts about, you said there's a trillion dollars Washington issued by the federal home loan banks to address financial weaknesses.

And what are the implications for that, of sort of using other state type banks to deal with a problem? Which seemed to obfuscate what really is going on and maybe hide the failures at the central bank. And hence what that's going to mean and the pressures that might put on monetary policy in the future.

I wanted to hear more of your thoughts on that.

>> Raghuram Rajan: Well, great question, and I may be relying on some of your work here without acknowledging it. But I think the worry is that we've crowded out the private markets for banks. The interbank to unsecured bank market no longer exists, right?

And when we look at aggregates, we feel comfortable, okay, there's enough liquidity in the system. But liquidity is not about aggregates, as you well know. It's about where the liquidity is located, and does it move from the place where it's located to where it's not? And as you have seen in the Silicon Valley problem, liquidity moved from the banks to the money markets and to the big banks, leaving a hole in the small and medium banks.

And the healthy banks weren't willing to fill that hole by lending in. And this is something that Daryl has also documented on a daily basis, the kind of liquidity holding by banks, knowing what might happen. Given this problem, one has to do everything to revive that interbank market, have them sort of on a daily basis understand each other's risks.

That can also be a source of discipline, but it can also be a source of mutual insurance when bad stuff happens. Now that that market It doesn't exist and there are the repo markets. But if the repo markets also start blowing out, the lender of last resort becomes the more immediate lender, in this case through the backdoor, the home loan banks.

So, I do think there's a good case for re-examining why these markets for liquidity have dried up. How do we revive those markets and how do we not let central intervention crowd out those markets?

>> John Taylor: So, Jeff Lacker.

>> Jeff Lacker: Yeah, thank you. Thank you for your work, Raghu.

And I'll start with two cheers for boring central banking. But my comment and question has to do with the safety net. The core of your analysis and narrative is banks risk management decisions, balance sheet management decisions, and how they're affected by central bank balance sheet policies over the last decade or two.

I would think that the core of analysis of how banks choose would be their expectations about the likely implications for them of adverse circumstances. And in that connection, I think the likelihood that the central bank or some other official entity is gonna rescue their creditors and them, it would likely loom large.

Those expectations aren't binary. They aren't zero, one. They've evolved over time, significantly since the 70s and 80s, as you well know. And it seems plausible that it's not an accident that the problems we saw earlier this year arose on the soft fringe of the set of banks and banking institutions that were viewed as too big to fail down in the regional, sort of the moderate to small size regional banks.

I wonder what your view about that and its connection to your analysis is. But also I think that if you're talking about the lessons learned for central banks, you might have something to say about whether they should address the continuing ambiguity and their continuing unwillingness to provide clarity about the extent to which the safety net is gonna be deployed for various institutions.

 

>> Raghuram Rajan: Yeah, I mean, look, I couldn't agree with the import of your question more. That is, we've just verified that there is no bank which is too small to fail, I mean, or fail in a excessively harsh fashion with depositors running and so on,. We're gonna sort of come in and help out, and in this case, help out all the depositors.

I think losses haven't been imposed on uninsured deposits in any of these failures. Correct me if I'm wrong, but I think that was the case. And that creates a concern that maybe the lesson from Lehman is let no one fail. And if that is the case, then you're putting an enormous weight on getting the supervision right.

Otherwise, it's a direct sort of transfer from taxpayers to these entities. I would also worry about the non-bank system that many players are taking risks which could accumulate, and if a tail event occurs, they're really out on a limb. I mean, 100 to 1 leverage and so on.

And of course, the bet is at those times the Fed will come in and help out. And so you get a lot more leveraging of that kind that is happening. So I'd say this is still a problem we haven't solved and certainly something we need to think long and hard about.

How do we impose losses? Luigi Zingales and I wrote a piece which I think seems pretty straightforward. Which is we should have let at least one banks' uninsured depositors take losses. Maybe bail out the rest, but have the first one at least take losses. There's a paper in the Journal of Finance which says this, and we thought it was a good idea.

But you have to let some of the losses fall where they should. Let me stop there.

>> John Taylor: Steve, go ahead.

>> Steven Davis: Thanks, Raghu, two observations and a question. So, first picks up on something John said in which Mickey Levy has documented in his paper presented at the monetary policy conference last year at Hoover.

And that is the perception errors and the forecast errors made by the Fed over the past 2 or 3 years. So I would think that that would lead to a considerable loss of confidence in the Fed's credibility, not with respond the usual hawkishness dovishness dimension, but just their technical competence.

Okay, so that's an assertion. You can agree or disagree with it. The second observation is we've just lived in the past three and a half years through the biggest swings of inflation in the United States and many other advanced economies in recent decades. I would think that that would cause many economic agents to start paying more attention to monetary policy and inflation statistics than they did in the past.

So then the question with these two observations is, what is this loss of confidence in the technical competence of the Fed and the increased attention by many economic agents to inflation and monetary policy? What does that mean for the conduct of monetary policy going forward, including possibly the design of a new framework?

 

>> John Cochrane: Could I just add, there was also a brilliant Steve Davis comment at that last conference where Davis pointed out that the Fed was in a catch 22 problem. It wants to stop inflation. If it issues forecasts that there's gonna be inflation, then inflation is gonna break out.

So it cannot issue honest forecasts. It wasn't just technical problems. There was an incentive problem.

>> Steven Davis: Thank you, John.

>> John Taylor: Raghu?

>> Raghuram Rajan: Yeah, I was just trying to pull up a slide which reflects the point that Steve just made. It's taking too long. Let me just, here it is.

Can you see this page? You see the graph on the right-hand side, which is Fed forecasts of PC projection, which constantly has it coming down neatly over the forecast horizon. And we know that that was excessively optimistic. That's just reaffirming what you said. What is interesting is despite the Fed being wrong in forecasting, when you look at medium term indicators of credibility 5 year forwards and so on.

It's been rock solid at 2 and change, 2.3 when I last checked. But it had gone up to 2.45 at some point and came down after that. So the Fed has somehow maintained confidence that it'll get things done despite the failure of forecasts. Is the point then that these forecasts don't matter?

People think that they're going to do it by the seat of their pants and somehow they'll get it done and we don't really need to worry about it. I don't know, but you raise a very good question. Should they have lost credibility in the process? And they may not have, at least as of now.

I'm sorry, Steve, what was your second question?

>> Steven Davis: Other one was just about economic agents, or many economic agents, not financial market participants, but individual households, smaller businesses, less interest rate sensitive sectors. You would expect, they've just lived through this unusual episode in the history of their lives.

And you'd expect they'd be paying more attention to monetary policy and inflation now on a regular basis than they did in the past. What does that mean for the conduct of monetary policy?

>> Raghuram Rajan: I think they would have become more sensitive to inflation, right, and the old statement that when you don't really care about where it is, that's the right level, that it doesn't impinge on your daily decisions.

And now they're seeing it matter when they go to the grocery store. The other thing, to John's earlier point about inflation coming down, I wonder if, what about catch up? Do we worry about workers wanting to catch up in their wages, given that they've experienced a pretty significant bout of inflation and over that period.

The real wage growth was negative. Is there going to be a period of catch up and how does that play out? I don't know. I just think that there will be more sensitivity, and I think we may be a little optimistic if we think it's all done and gone.

 

>> John Taylor: Larry Schembri.

>> Larry Schembri: Thanks, John. Thanks for the excellent presentation. I look forward to reading the book. The question I have is just getting back to John Cochrane's comment about the next round of frameworks. And as you know in many inflation targeting jurisdictions, especially in the smaller ones, central banks have this control range around the inflation target.

And the US and the ECB has avoided having a control range. It just seems like having a control range, for example, one to three, like for example we had at the bank of Canada, just gives you a more robust framework to deal with low and high inflation targeting regimes.

That may be the way of getting the robustness you need to still remain preemptive, but at the same time being able to tolerate inflation below target when you're in the low inflation regime, I think that's a great point. And actually, even the ECB had an asymmetric target for a while, and this was up to two, but they didn't make a big deal if it was below two.

And somehow that changed to make it symmetric around two. And maybe that was the post pandemic anxiety about trying to enhance economic activity through monetary policy. So I'm entirely with you that a range makes a lot of sense. It allows the central bank to be less hyperactive, so long as things are broadly within the range and likely to drift towards the middle.

And it can be a little more relaxed, that may be the way to deal with this issue.

>> John Taylor: David Malpass, David?

>> David Malpass: Yeah, can you hear me? Hi all. Thank you, John. And thanks, Raghu. And so I have comments and then a question. Very good that you explained the duration impact.

That was significant, and also very good. You explained the difference in size of banks regarding their dependence on uninsured bank deposits and also the importance on the interbank market. So I want to come to that. In addition to the liquidity arbitrage that's going on among the banks, there's a leverage ratio that's constrained.

And it's constrained differentially among big banks versus small banks. And so one thing I observe is the size of the arbitrage that's going on on the liquidity side is massive. We're talking hundreds of billions of dollars in order to accomplish the off balance sheet instruments that we're working on.

And the Fed is trying to make all of that arbitrage safe. So a key question, or a key point that you were making is, how do we impose losses in order to have some discipline within the market? My observation is right now, there isn't discipline, that everything is safe because the Fed is drawing kind of a red line around all arbitrage kinds of activities.

So my question is on, you didn't mention the leverage ratio. Do you see that it's constraining different sized banks differently? And thanks.

>> Raghuram Rajan: Yeah, I mean, thanks, David. I think some people have raised this issue. And for some time, I think there were some exemptions given to the banks on the leverage ratio applying to high quality securities like reserves.

And then that was reinstated, and that's when the money market funds came in in a big way to absorb reserves. I mean, look, I think to the extent that the leverage ratio impinges on safe activities, it's worth re examining, because that was not the purpose of the leverage ratio.

It was to inhibit risky activities. But once you start a whole range of exemptions, then you go back to the old problem. You have risk based capital, and who knows what is risky? The wrong mortgage backed, AAA rated mortgage backed securities turned out to be more risky than we thought.

So there's some intermediate point where you recognize safe assets don't necessarily need to attract the leverage ratio, but somehow you still prevent banks from being levered 50 to 1.

>> John Cochrane: Yes, so I'm concerned about the strategic framework. And Jay Powell says the Fed is going to undertake this review in the second half of next year.

And when we look back at the process that led to the strategic framework in August 2020, it was all asymmetric thinking that led to an asymmetric interpretation of its dual mandate, getting rid of the preemptive, the average inflation adjusted targets and all that. And in fact, I went back and read Jay Powell's speech.

He gave that Chicago fed seminar that culminated their strategic planning. And in 19 paragraphs, he focused on the effective zero bound in 10 of the paragraphs. And then all of the papers given at that conference, that's an exaggeration. Most of the papers were about how to deal with the effective zero bound and what kind of strategies would work best.

So the question is, will the Fed change from its asymmetric thinking as it considers how to review the strategy? I question that because the Fed has this tendency to keep things the way they are. And, in fact, recently we've heard some Federal Reserve members say, well, the new strategic plan is just fine.

We just didn't interpret it. Our tactics were wrong, but let's keep things the way they are. And so I'm very concerned about just the Fed's whole approach to how it thinks about a strategy. And it gets to this issue, does it just let bygones be bygones? Can they now do a victory lap that inflation has come down and go on their merry way?

 

>> Raghuram Rajan: I hope not. I mean, I have to believe that they will be informed by some of what went wrong, whether it's adequate for what you're saying. It's hard to say, I don't have a lot to add on this. I just hope that they think carefully. I mean, I really hope they do.

Again, nothing to add to that.

>> Male Speaker 4: Raghu, I have a question. You're speaking as an experienced central banker in a very important part of the world, but you didn't talk about that part of the world very much. It's on a lot of our minds. What's going on with China, what's going on with Russia, what's going on with India?

Could you say just a couple words about that? Should we adopt? You say there's a problem with rules or strategies, but what do you think about the global situation? Could you say a few words about that?

>> Raghuram Rajan: You're talking about the Indian economy and its relations right now rather than Indian monetary policy?

 

>> Male Speaker 4: Yes.

>> Raghuram Rajan: Yeah, now, look, India had a very bad pandemic. And it's recovering reasonably strongly at this point, but it's a two-phase recovery. What you hear about the big firms, the celebration, the stock markets, it's hit some record high at this point. Inflation has been relatively quiet, but there's a hurting part of the economy, the small and medium sectors, some of the rural areas.

The stress and unemployment rates are quite high. I mean, we have around 6.5% growth at this point. But that's not creating enough jobs in the economy and certainly not making for the backlog of jobs at this point. So when you look at India, it's sort of huge contrast.

There's some really impressive firms, really impressive achievements. I just went to a small firm in the south which was making 3D-printed rockets, essentially gonna send up one satellite at a time on a customized basis. And they were all gung ho about it, enthused by becoming Elon Musk. So that's on one side.

And then you have the pandemic hit education everywhere. But in India, which already had bad levels of education, on average, the fifth grade student can't do math. At the second grade level, around 50% of them can't do that. That suggests that the pandemic hit education really hard. My worry is this government that we have is much more focused on capital investment, on infrastructure build out, which India does need.

And much less focused on improving human capital, which may be sort of India's greatest deficiencies with respect to the future lie. We have 1.5 million engineers being generated every year, but about half of them are unemployable. And a big fraction of the remaining can do low level jobs rather than high quality engineering jobs.

The silver lining is the best trained are doing fantastically well. Goldman Sachs has a huge office in India and there are 1,600 global capability offices like that, employing 10,000 people at a time. So that's a big source of employment and growing, but that's sort of the tip of the iceberg.

There's a lower level. That's the bifurcated part that I talked about, which I worry very much about, because that's where the social conflict starts coming on the political side. Modi just won a state election. I mean, he has mesmerized the electorate, and he's given enough in the form of direct transfers, freebies, to make people's lives a little better.

Even if they don't have jobs, they have some money coming from the government. So all in all, he's probably gonna get reelected. In my view, it doesn't speak well for democracy going forward, but that's the state of the nation as it is.

>> John Taylor: Let me just mention we still have copies of this beautiful book, Monetary Policy and Its Unintended Consequences by Raghuram Rajan.

And thank you so much for sharing your ideas with us about this very important issue. Thank you, Raghu.