Bill Nelson on the Fed’s Policy Tools in the Post-COVID Economy

Bill Nelson is a Chief Economist and Executive Vice President at the Bank Policy Institute, and formerly a Deputy Director of the Division of Monetary Affairs at the Federal Reserve Board, where his responsibilities included monetary policy analysis, discount window analysis, and financial institution supervision. Bill also worked closely with the Bank for International Settlements on liquidity regulations. Bill is a previous guest of Macro Musings, and he returns to the podcast to discuss the Fed’s increasing role in credit policy, the prospects for yield curve control and negative interest rates, and why makeup policy would be uniquely suited to the challenges presently facing the economy.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Bill, welcome back to the show.

Bill Nelson: Thank you.

Beckworth: Well, Bill we've been interacting a lot. We were on the conference with Cato and Mercatus. But today, I want to switch gears and look at what the Feds have done and moving forward, again, what we might see the Fed do. And I have to say you're one of two people that has been highly requested by our listeners. So, Peter Stella was the other one, and we had him back on so you're back on by popular demand. So, I wanted to let you know that, and people really appreciate you, too, and I think it has to do something to the fact that you guys really know the plumbing of the monetary system really well and they really enjoyed all the insights from last time we talked, which was late last year about the time of the repo market stress, and we walked through some of those issues. So, we'll touch on some of those again, today, but we're going to cast a wider net and talk about Fed policy in general what they've done.

Beckworth: So, I want to begin by summarizing, if I can, what I see the Fed has done and then I want to throw it back over to you and you can flesh it out, you can critique my summary and I have some questions, but I'm glad to have you on because you know what the Fed thinks. You're a former Fed insider. You can at least understand where they're coming from and it help us as listeners understand, maybe what they're thinking. So, if I had to summarize what the Fed's done, I would have three broad buckets. So, the first bucket I would say would be just traditional monetary policy or you'd technically say, unconventional monetary policy, but everything from cutting interest rates down to zero lower bound, which for the Fed is zero to 0.25%, forward guidance, large scale asset purchases. We know the Fed's balance sheet now is now near $7 trillion and about $4.2 in treasury holdings, $1.9 trillion in mortgage backed securities. So, they're doing those what I consider traditional monetary policy features and they've said in their June meeting, they're going to continue buying treasuries and mortgage backed securities going forward. So, that'd be the first bucket, traditional monetary policy.

Beckworth: The second bucket, I would label liquidity to a distressed financial system or liquidity efforts, and that would include everything from their efforts in the repo market to tweaking the discount window and bank's access to funding from the Fed to their efforts with what I would call shadow banks. So, all of these facilities, they've invoked for liquidity purposes under 13(3) of the Federal Reserve Act, so the primary dealer credit facility, the money market funds liquidity facility, commercial paper facility, dollar swap lines, and so forth. That group there, I think you can see it as an attempt to provide liquidity to funding pressures in the shadow banking system. It's similar to a discount window, but for all those financial firms that don't have access to it. So to me, that's a reasonable thing they're doing. It's part of their mandate of financial stability. That'd be the second bucket.

Beckworth: The third bucket and final one is, at least in my mind, they're going beyond liquidity provision. I know this is a debatable point, but in my mind, they're going beyond liquidity provision and they're getting into credit policy. And I know the Fed has called it liquidity provision as well, but the ones I'm thinking of, of course, are the primary credit corporate facility, the secondary market credit corporate facility, the municipal facility, the various main street facilities where they're getting into those markets as well. And I want to stop there with my summary and ask, is that a fair summary or division of the buckets or would you do it differently?

Nelson: No, I think that that's a very astute division, in fact, and I think you identified a really interesting aspect of what they are doing, which is in terms of seeking to directly target private sector, interest rates and asset prices, that's kind of a new development relative to the last time around, as opposed to getting rid of liquidity premiums in those asset prices.

Beckworth: Yeah, so the liquidity facilities, I understand, and I know some are critical of those as well, but I understand at least why the Fed is doing them. There's a run on funding on dollars in the shadow banking system that has to be dealt with or we're going to have a far worse situation. To me the more controversial issue, the more contentious issue is the credit policy the Fed is doing because in theory, Congress and Treasury, Fiscal Policy should be done by them, and many would call this the Fed veering into the fiscal policy lane by doing credit policy. And I guess, the question is why? Why are they doing this? I'm going to throw two theories out there. And I want you to evaluate them.

Why Has the Fed Engaged in Credit Policy?

Beckworth: My first theory is, I think it's a pretty common one, is they're the only game in town or they're the go-to player that the government uses because Congress can't get around to doing things itself or Treasury doesn't have the in-house expertise, so you hear this story a lot, the Fed has to step up because Congress and Treasury aren't doing their part, which seems reasonable at some level. A second theory and this is one that I just came across and I got this from Robin Wigglesworth of the Financial Times and he had an article titled “Long live Jay Powell, the new monarch of the bond market.”

Beckworth: And his argument is that more and more of global markets are being funded by bond markets, so the securitization is becoming more and more important to capitalism, less reliance on banks, this is his argument. And this is a trend kind of where the global economy is going, and as a result, central banks, not just the Fed, but like the ECB, the Swiss National Bank, they all have been buying corporate securities for some time, the Fed's kind of late to the game, but the Fed's jumping in.

Beckworth: So, his point, now he's not giving a reason for why the Fed is intervening, but I take his point and apply it in this manner. His point is that the increased securitization of finance around the world is requiring the Fed to adapt and move into a space that previously wasn't comfortable doing or didn't do in the past if that could just deal with banking and bank-like firms, but now has to deal with the bond market given the nature of global capitalism. So those are the two theories I'm going to throw out there. Maybe you have a better one, but could you evaluate what they are?

Nelson: Sure. So, I think it's a really interesting question, so I agree with you. Addressing illiquidity in a financial market a natural thing for a central bank to do. It's something that they were designed to do from the start and it's something that they can do very well, because as an institution, they themselves don't face liquidity risk, so they can effectively go in and purchase assets or make loans without concern that they themselves will come under liquidity pressure, which is often the limiting factor on why the forces of arbitrage don't get rid of those things.

Bill Nelson: And, moreover, the transactions themselves, they had the advantage of generally they are of the nature that you're buying low and selling high. You're going in and you're purchasing an asset, it's undervalued, its spread is too high and through your very action, you're bringing that spread down. And while central banks aren't in the business of making money, it does help with the credit protection. It automatically sorts of builds in a layer of credit protection that you have this natural gain to what you're doing.

Bill Nelson: Now, in the case, I think the thing that really struck me really is the secondary market corporate bond facility. So, the primary market facility is kind of like the CPFF, which was a primary market facility, in that you could envision it as a mechanism to address a liquidity problem. You'd be saying, I mean, typically, the Fed and other central banks have sought to charge rates that were below whatever was the prevailing rate in turmoil, but above normal market rates and you step in and you say, "I'll buy those bonds or buy that CP," at issuance at a rate which is above normal rates, but below where they are now and that sort of brings about greater comfort and liquidity in that instrument.

Bill Nelson: But to go in and buy as the Fed is doing now corporate bonds in the secondary market, that's quite a different thing because you're not providing funding to the corporation. What you're doing is you're supporting corporate bond prices generally in the marketplace. And it's interesting, I went back in preparation for this podcast and reminded myself why the Fed said it was taking that action in its letter to Congress. There's always very interesting go-to places if you want to look and drill into their different actions. They have to write a report to Congress about each of the 33 facilities, and they said that they were doing so because basically funding, there was a lack of funding in the corporate bond market and they were addressing that and at the same time, there was elevated funding because of lack of income because of coronavirus.

Bill Nelson: But, yeah, the corporate bond market has come back like gangbusters as I'm sure you know. Its spreads have come in, issuance has been extremely brisk. There's really no evidence out there now that there's a shortage of credit supplies from the corporate bond market, but when Jay Powell was questioned about this by Congress, I think that the line has been, while we said we were going to do this that contributed to the recovery and we have to deliver on what we said we would do or we will give up on the gains that we got. It's logical, but it's a bit worrisome because particularly in this case, [the Fed] didn't really say what their plans were specifically, they didn't say we're going to buy this much or that much. They just said, ‘we're going to start buying them.’ And so it kind of raises the question then of, where does it stop?

[The Fed] didn't really say what their plans were specifically, they didn't say we're going to buy this much or that much. They just said, ‘we're going to start buying them.’ And so it kind of raises the question then of, where does it stop?

Bill Nelson: I mean actually the same issue arose in the Treasury purchases and in the MBs purchases that the Fed is now purchasing each month 80 billion in treasuries and 40 billion in agency MBS, considerably more than they were doing and QE3, which was a program that really blew up their balance sheet and it's not even really clear why they're doing it. I mean, they stepped in to address very severe dislocations in those markets. They said they're not doing it to try to lower Treasury MBS rates, which is understandable because those rates are very low and they've said that those markets have recovered, they're functioning fine. But there's some sense that, but we feel that we need to continue to do this to keep them functioning fine. But again, it's very difficult, puts the Fed in a difficult position to articulate a stopping rule.

Bill Nelson: All that said, and to get back to maybe your proposition, it is true that other central banks have started purchasing corporate bonds as a means of influencing economic outcomes for other central banks for the ECB, for instance, at its foundation, it was purchasing government securities that was considered the radical and dangerous action, it was built to be very similar to the Bundesbank and Bundesbank always felt that the worst thing that you could do is buy government securities.

Bill Nelson: And one advantage of being so old is that I've been involved in a lot of interesting things at the Fed and one of them was actually in 2000, it's hard to believe, but everybody seriously believed that the federal debt was going to be paid down and there was a big program at the Fed to identify other assets that the Fed could purchase. I mean, basically, the Federal Reserve is an engine for acquiring assets that match currency. Currency is the Fed's liability and they also have to worry about reserves in order to conduct policy, but they have to have some assets in order to stand opposite those things. And now currently and for a very long time, those assets have been Treasury securities on and off MBS, but if those things were going to go away then the question arose, what would we buy instead?

Basically, the Federal Reserve is an engine for acquiring assets that match currency. Currency is the Fed's liability and they also have to worry about reserves in order to conduct policy, but they have to have some assets in order to stand opposite those things. And now currently and for a very long time, those assets have been Treasury securities on and off MBS, but if those things were going to go away then the question arose, what would we buy instead?

Bill Nelson: So, one of the things that was looked at, that was most involved, which was actually making discount window loans, in size, and that ended up becoming that, I mean obviously the problem went away. Congress found a way to spend that money, but then when the crisis hit, we actually went back to some of those ideas and one of the ideas was to auction off discount window loans and that became the TAF, which was an instrument that was used in the past crisis. But another thing that was seriously looked at was buying corporate bonds. At that time, the view was that that would require a change in the law for the Fed to do that, but that in principle, under those circumstances, if you bought a broad-based bundle of high-rated corporate bonds that might be a relatively non distortionary way to purchase assets.

Bill Nelson: So, currently, the Fed has its interpreted its ability to lend to an SPV that in turn purchases things as long as it's capitalized by the Treasury very broadly in terms of what it could buy and that seems to include treasuries and even exchange traded funds is as you know, but in any case, it's certainly a legitimate means to influence the economy. It's just not one that's been used in the U.S. I wonder, I mean, I don't like it. I find it to be an unnecessary intervention of the Fed into the private sector, when the markets were not sufficiently disrupted to warrant that or briefly, they were but still, but that said, and something that we may talk about in a bit, which is, well, what are they going to do if things get bad?

Bill Nelson: One advantage of this program is that it's scalable. When they start wanting to looking for dials to turn. This is pretty modest right now, but this is a dial that they could definitely turn up and it doesn't require that the market not be working in order to actually influence the economy.

Beckworth: Now, if they're going to do this permanently though, the corporate bond purchases, they would have to change the law, though, right? They're just using 13(3) in emergency setting.

Nelson: And Cares Act.

Beckworth: And Cares Act, yeah.

Nelson: Equity investment, absolutely. Yeah. This could not be done. Well, I don't want to say it vividly. It seems to me that it cannot be done under existing law without an emergency and without and some source of equity investment in the SPV.

Beckworth: Okay. Let me go back to the Robin Wigglesworth idea. Let me just push that a little bit more. Do you think there's any merit in his notion that because securitization, because bond markets are playing a larger and larger role, no matter what the Fed wants to do, no matter what Congress wants to do, they're going to be nudged in that direction. If they want to be lender of last resort or market maker of last resort in the midst of a crisis, they're going to be forced into private bond markets. Do you think that has merits? It's a bigger force than any one institution.

Nelson: So, let me be an economist about this and to give you sort of two sides to the answer.

Beckworth: Okay.

Nelson: So, on the one side, I don't think that it is necessary for the Fed to act directly on an instrument in order to affect the pricing of that instrument. Monetary policy has traditionally operated in a very small market, Federal Funds Market in the United States to influence the short-term, effectively risk-free rate in that very small market with very light touch by the Federal Reserve. And those rates were well transmitted into repo markets and into other trip markets and since long-term rates are really a reflection of the expected path of short-term rates, they were also transmitted into longer term rates. And so, the Fed was completely capable of moving short-term and to a great extent, long-term rates up and down to conduct monetary policy and to get the aggregate demand outcomes that are one.

Bill Nelson: That's true, regardless of whether or not credit intermediation takes place largely through capital markets or largely through banks, because the interest rate channel of monetary policy works perfectly effectively that case. Now, increasingly, and I think that this is a result of the move to a floor-based system that we've had a lot of enjoyable conversations about, people perceive monetary policy as something that happens when the Fed is basically operating on the margin in whatever security that it wants to be influencing. That's novel and not necessary, but I think that there's sort of a widespread view that that's kind of how it's done now.

Bill Nelson: And so, I can see where one would therefore be led to the conclusion that since credit intermediation takes place in corporate bonds, asset backed securities, the Fed has to operate in corporate bonds and asset backed securities in order to bring about the changes in aggregate demand it wants to bring about to achieve its monetary policy objectives. That's just simply not the case. The Fed can have central banks generally operate through influencing risk-free rates, and that should work. Now, the only place that I would sort of step in and say, on the other hand here though, is that there has been over time over the last decade, a significant reduction in the liquidity of those markets as it's become more costly for primary dealers, rather for broker dealers to intermediate in those markets as various regulatory changes have taken place and there seems to be an increasing need for the Federal Reserve to step in potentially be in the market maker of last resort on both sides of those markets. That's not a necessary development.

Bill Nelson: And as one who prefers that the private sector solve problems whenever the private sector can, I don't think it's a welcomed move. I think it's a dangerous outcome for the Federal Reserve, but there has been an increasing trend over time in which the Federal Reserve is taking actions that increased its involvement in the financial system. I think it's bad for the financial system. I think it's bad for the Federal Reserve.

There has been an increasing trend over time in which the Federal Reserve is taking actions that increased its involvement in the financial system. I think it's bad for the financial system. I think it's bad for the Federal Reserve.

Beckworth: That's a great point that the Fed's own choices have led it down this path where it may have to intervene in these markets, so different regulatory touch, different operating system, all those things can push it in a direction where the markets can do most of the heavy lifting instead of the Fed and I'm very sympathetic to that point.

Beckworth: All right. So Bill, let me switch gears a little bit and touch on some of the plumbing related to these Fed actions. So, we've talked about the big buckets, we've talked about the Feds veering into credit policy lane with some of these corporate bond purchases, but I want to drill down a little bit more into the Fed's actions in the actual implementation of monetary policy. So, I briefly mentioned it tweaked discount window usage and overnight drafts. Can you speak to those? What has the Fed done that has materially changed how banks can interact with the Fed?

Recent Changes to Discount Window Usage and Overnight Drafts

Nelson: Okay, great. So, there are some very interesting things about what the Fed did to change their lending or to the discount window, but let me come back to that. So, in some sense, the Fed is continuing to conduct monetary policy the way it did before this pandemic crisis, in the manner that it sort of backed into in the last crisis and officially said about a year and a half ago that it was going to continue with, which is to oversupply reserve balances. Those are the deposits of banks to the Fed, the Fed's main liability, so that it pushes overnight rates down to two or a bit below the interest rate that the Fed pays on those reserve balances.

Bill Nelson: Currently, the Fed is paying 10 basis points on those reserve balances in the Federal funds rate. The overnight rate that banks lend to each other is about eight or nine basis points. So, that really hasn't changed. I mean the difference is if anything their growth of, I mean, reserves have simply grown enormously and they are now even more over supply. Reflecting the discount window as the Fed often, I mean, in terms of the discount window as the Fed often does, almost always does in response to crisis, one of the first things that it does is it says, the discount, now meaning the way that it lends to commercial banks is open and operating and it often reduces the discount rate as it did and it lowered the discount rate to the top of its target range 25 basis points.

Bill Nelson: And importantly, it indicated that rather than lending overnight, it was going to lend at 90 days and those loans would be sort of renewable at the request of the borrower. What that did was that that enables banks to replenish their holdings of liquid assets in a manner that boosts their liquidity coverage ratio, which is the main liquidity requirement that banks face. So, the liquidity coverage ratio is sort of the ratio of liquid assets to projected outflows under stress. Normally in the past if the Fed lent to you overnight, there would be an outflow that corresponded to the money that you had.

Bill Nelson: Now, since the term is well beyond the 30-day window of the LCR, there's no outflow. And you can always just roll that loan over and can repay it early, it never gets within the 30-day window. So effectively, the Fed was giving a tool to the banks to allow them to be confident that they could convert their discount into collateral and that there's I think, 1.6 trillion in discount sitting at the Fed unused, almost all of it very illiquid assets that don't count towards their liquidity ratios. The Fed was giving the banks a tool that they could use to boost those liquidity ratios considerably by taking those actions with the discount window. It didn't really have much of an effect if anything on monetary policy.

Beckworth: Okay. Let me ask this question. I was going to wait until later to talk about it. But let's talk about it now on the show. So both of us would like to see a world where the Fed has a smaller balance sheet, fewer reserves, something like a corridor operating system. Is that fair? I shouldn't ascribe it to be?

Nelson: Yeah, well, that's right.

Beckworth: Okay. Yeah, I think we're both big fans of the corridor operating system. And one of the challenges before this crisis in terms of that happening, were these new regulatory requirements that incentivize banks to hold more reserves. And I know one of the observations was banks don't go to the Fed and get these overdrafts, they don't tap the discount window, therefore, they're holding dear the reserves they do have for these regulatory purposes. I'm just wondering if we were to go back to a world where we had a corridor operating system as our ultimate goal, so we were shrinking the Fed's balance sheet, could the Fed keep these changes permanent as part of the journey to that destination, so that banks wouldn't be clamoring for so many reserves and make it hard for the Fed to shrink its balance sheet?

Nelson: Yeah, I think so, but let me back up and take a run at that. So, you're right. You and I have been one of the sort of the old holdouts, wanting the Fed to go back to conducting policy the way it did in 2007 and before. And large part on my part, because it just involved with much less intervention as Fed with a much lighter touch, less involved in the financial system worked perfectly well, in my view. I do now kind of think that it's going to be a long time before we're going to really reasonably have this debate again, because there's such a vast quantity of reserves balances out there, and those things are just going to keep going up that this almost becomes sort of, I forget the right word for it, but the conversation that we can have for intellectual curiosity, but when-

Beckworth: An academic debate?

Nelson: Yeah, but that said, so as we've discussed and I've written and talked about, I mean, one of the things that sort of locked, so the Fed did kind of get locked into a world where it was oversupplying reserves in part because I would argue that those reserves created their own demand. Banks got used to having those reserves, supervisors got used to the banks having the reserves and supervisors got used to the banks not running daylight overdrafts, even though that used to be a matter of no concern whatsoever. And for all of those reasons, as the Fed started to shrink its balances back, it's hard to remember now, but not that long ago, that in September of last year, the Fed found that even though it thought it had a long way to go, it actually encountered a lot of resistance with quite a high level balance sheet in part because it looked like everybody was very reluctant to face any kind of risk of becoming short in reserves haven gotten used to it.

The Fed did kind of get locked into a world where it was oversupplying reserves in part because I would argue that those reserves created their own demand. Banks got used to having those reserves, supervisors got used to the banks having the reserves and supervisors got used to the banks not running daylight overdrafts.

Bill Nelson: And I don't want to go back into the whole litany of reasons and stories that I've heard from bankers and others about what built in that desire. But in order to get the Fed, in order to get banks more comfortable moving further down their holdings of reserves, part of the thing that would have to happen is that they would have to feel comfortable knowing that they would have a source of reserves that they could go to if they needed it. And this kind of a longer term discount window that was available on request would be one way to do that.

Bill Nelson: I've actually written about a way to sort of formally do this that would have the advantage that it would sort of make banks pay for the privilege of having what's effectively a line of credit at the Fed and by sort of setting up a special deposit and lending account with rates constructed, so that banks to borrow, put the money on deposit, but then effectively have to pay a fee. That's another interesting topic perhaps for another day and perhaps in that distant future when they actually are really shrinking the balance sheet again. But importantly, the banks would need to feel that they could replenish their balance sheet on demand. And that's where this discussion of having like a standing repo facility was kind of attractive, because that could operate intraday, it would increase banks comfort with the knowledge that if they need the funds, they could always repo Treasury securities to get the funds.

Bill Nelson: And so, since Treasury securities and reserves count equally, in theory, in interest rate in liquidity regulations, the trick would be to get banks to be more comfortable holding treasuries rather than reserves and that could be facilitated by making them very comfortable with the knowledge that if they need the cash and the immediate liquidity that's only provided by reserves, they could do so with the Federal Reserve. That gets into the whole history of the fact that banks don't want to operate the Federal Reserve and that's a 70-year problem, that's a real problem that comes in a lot, but one of the many problems that will be ultimately have to be addressed in order to get that balance sheet down once more.

Beckworth: All right, Bill. That's been great looking at what the Fed has done, some of the Fed’s plumbing, but let's look forward because we were hoping for a V-shaped recovery. There's some signs that maybe we won't have it and if that's the case, what can the Fed do? What if we go into like more of a W-shaped recovery? We have another little dip going on into the year as the cases begin to surge up if maybe the unemployment insurance runs out there, there's concerns about that, that things might begin to slow down, in that case, what can the Fed do?

Beckworth: To facilitate this question. I want to look at the June FOMC minutes that just came out and they talked through some of the tools they could use in the case of a need for further easing and probably, the most striking conversation that was in the June FOMC minutes was a conversation about yield curve control. And I guess the general takeaway, I think most people read this this way and that is, they kind of shot it down. There's a lot of skepticism about it, a lot of uncertainty. Yes, the Bank of Japan's doing it, yes, the Reserve Bank of Australia is doing it, and yes, there's even some history of the Fed doing it during World War II, but many people have looked at this read of the minutes and said, "They're probably not going to do that." Do you think that's right?

Prospects for Yield Curve Control

Nelson: That's how I read them, too. Yeah. And so just for those who may not be familiar with it, basically yield curve control is when the Fed says, "We're going to target the Treasury rates out the yield curve at some level out to some yield and as you noted, there have been some other countries that have used it. Fed has used in the past and the Fed did actually look pretty seriously into yield curve control during the past crisis and recovery, slow recovery. And I'd be happy to send you the links to the FOMC documents and you could put them on your show notes if you want.

Beckworth: Definitely.

Nelson: The problem and this came up in the minutes with yield curve control. One of the big problems is that exit is very difficult. So, I'm generally not in favor of the Fed announcing a rate target for a market that it doesn't really control rather than, say, acquire. "We're going to buy this amount of long-term Treasury securities," they can do that. If they want to control rates, I think it's very similar to episodes where central banks are trying to control exchange rates, which kind of ultimately, it always ends. The fate of every glass is to be broken, right?

Beckworth: Right.

Nelson: And it's a real problem in the case of yield curve control when you get to the end of your period of yield curve control. So, suppose you say, we're going to buy and sell 10-year Treasury securities in an amount sufficient to keep the 10-year Treasury rate at 1%. And so if the rate goes above 1%, we'll buy and buy and buy until it goes back down to 1%, but then, people begin to suspect that you're going to announce that you're done doing this in a few months. Suddenly, it becomes very profitable to purchase Treasury securities. The demand for 10-year Treasury securities goes up, and it keeps going because everyone knows that eventually you're going to stop with the premise, the rate is going to go on. I think I probably got those backwards, but so you're going to want to-

Beckworth: Interesting.

Nelson: You don't want to short the securities. Ultimately, you're going to sort of run. There's going to be a run on the Fed. So, the way to address that problem is to couple the yield curve control with date-based guidance. So, there's usually just kind of two kinds of forward guidance for what you're going to do with your short-term rate. There's three types. There's qualitative, you can say, "Well, we're going to keep it at zero for a substantial, for a considerable period of time," or there's date-based, "We're going to do it, we're going to keep it at zero until 2021 December," or there's threshold-based guidance, which is where they seem to be going or what they call... well, anyway, threshold based guidance, "We're going to keep it at zero until certain economic conditions are met or exceeded at least until."

Nelson: If you use date based guidance, you can say, "We're going to keep interest rates at zero for two years and we're going to at the same time target the yield curve out to two years." And then a month from now you say, "Well, we're going to keep interest rates at zero for one year and 11 months and we're going to target the yield curve Treasury securities with maturities of up to one year in 11 months." And that way, your promise on rates sort of slowly collapses into zero along with your premise on yields and you're not forced to defend a rate control level that you're not simultaneously delivering with your short-term rates.

Nelson: But now is not a good time at all for date-based floor guidance and that's, I think, in large part because as the committee has repeatedly said, "The odds look very uncertain." There's kind of two ways that the outlook could go, less and less true every day, but nevertheless, there was two ways that the economy could go. Sort of recovery on second half of this year and things would largely be not quite, but close to where they were or there could be a second wave of the virus, things could shut down again and we could be in a very severe problem.

Now is not a good time at all for date-based floor guidance and that's, I think, in large part because as the committee has repeatedly said, the odds look very uncertain.

Nelson: And you don't want to announce a date that's so short, that it works well in the good outcome because people won't find that very comforting, but you also don't want to announce a date that's so long that it's kind of the right date for the really bad outcome because then if you end up with a good outcome, your hands are tied for a very long period of time and longer than you would like by any law.

Nelson: So since I'm sure they probably decided that date-based guidance isn't really the way to go that could be part of why they decided that yield curve control isn't the way to go and that sort of brings up threshold-based guidance, well, conditional guidance, which we can discuss next if you want, if that's what you want.

Beckworth: Well, that's fascinating. I hadn't thought about that. So, they have to be careful because the market might try to play them with yield curve outcome.

Nelson: They will. I mean, who wouldn't?

Beckworth: Yeah, I know for sure. So, let me ask a related question to that. It's one thing for maybe the RBA, the Reserve Bank of Australia, to do this on a two-year, I think it's two-year yield. The Bank of Japan, if I understand correctly, is targeting a 10-year yield, but they're actually doing it to keep it up, not to push it down because they were worried about financial disintermediation with banks and stuff, so it's a little bit easier in that case, to play that game. And I guess the bigger question I have, though, is, given that the U.S. Treasury Market is the most important one, it's a big market, it's a global market, it might be really hard to implement. I mean, you're fighting a huge market. It's not like the 1940s when the Treasury Market was a lot smaller, maybe the Fed intervened. I mean, I guess my concern would be it could require a really, really large balance sheet of the Fed to be engaged with the market this size, is that a reasonable concern?

Nelson: I agree with you. I share that concern, and so do FOMC participants based on the minutes of the most recent meeting. Some of those that's in the list of reasons why they were sort of hesitant to go this route. One was that they were concerned that it would require uncomfortably massive intervention on the part of the Fed.

Beckworth: Okay. So, let's say the economy does turns south, that things do look worse. This is a tool that probably won't be used. I mean, so it's one thing right now to say we're not going to use it as the FOMC, but if things get really bad, might they still pull it off the shelf and use it or do you think it's put away for the good?

Nelson: I don't think that it's a tool that will be particularly helpful for them, even under those circumstances. In some sense, it's a tool that's naturally suited to kind of the opposite problem. So, in 2003, '04 and then again in 2009, '10, '11, the concern was that as signs of an improving economy where mass financial market participants took onboard signs of improving the economy, the interest, the path for short-term rates starts to move up and longer term rates start to move up, that removes financial combination, the removes the stimulus provided by financial conditions, and that could choke off the incipient recovery, stomping on green shoots is often related, so it's discussed.

Nelson: And so when you're in a period where you're still trying to support the economy and really makes sure it gets going and gets out of whatever bad situation it's in and you don't want rates to start moving up and choke off that momentum, that's sort of a good time to say, "No, we're really serious. We're going to keep rates low for a while longer than you think and long enough to really get the economy moving like a rocket." And that might be a time for sort of date-based guidance and something along these lines. But in this situation where kind of like we are now where people expect overnight rates to be near zero for a very long time, it doesn't really help a lot to say, "Oh, by the way, we're going to make sure it rates are zero for a period of time." It doesn't add much.

Beckworth: Yeah. It doesn't add much power to the existing stance of policy. Okay. So, yield curve control in the present circumstance is not very practical or useful. What about negative interest rates?

Prospects for Negative Interest Rates

Nelson: So, just in terms of what are the tools available to the Fed and what might the Fed do if the economy remains weak and looks like it's going to remain weak for a long time. First and foremost, I really don't think that this is a situation where the answer is monetary policy. It seems like the answer to this problem is fiscal policy. Ultimately, with rates near zero, the Fed is naturally moving to sort of second tier of things to do and people are going to be without income. The solution is to give people income to get through this problem and smooth out those expenditures across the taxpayers and across time. I mean, that's the solution.

Ultimately, with rates near zero, the Fed is naturally moving to sort of second tier of things to do and people are going to be without income. The solution is to give people income to get through this problem and smooth out those expenditures across the taxpayers and across time.

Nelson: But if we were to focus since exclusively on what the central bank can do, I just want to talk a little bit about the risk of a deflationary spiral. So, the Fed doesn't mention this very much and that's, I think wise, because you don't really want to talk about very bad outcomes a lot. You want to be transparent and honest, but you don't want to talk yourself into a bad situation. But, there've been a lot of talk over the last four or five years about how extraordinary is that inflation hasn't moved up, despite how low the unemployment rate got. In some sense, the real extraordinary thing, I guess, it did pretty extraordinary, but the extraordinary thing that preceded it was that inflation didn't move down in the great financial crisis, Great Recession period whence the unemployment rate moved up to 10 and inflation didn't start moving down.

Nelson: And both of those episodes indicate that inflation expectations which kind of anchor the inflation rate are very well anchored. That in the relationship between inflation and the unemployment rate is pretty flat, quite flat and so that's a good thing. At least, the well-anchored inflation expectations is a good thing. It allows a more effective monetary policy basically, without some of the problems associated with it. But inflation has been running below 2% on average, most of the time for about a decade. It's moved down more. If people start doubting the Fed's ability to hit its 2% target and inflation expectations move down, then, with nominal interest rates stuck at zero, real interest rates move up and so monetary policy is being tight.

Nelson: So, the Fed wants to provide stimulus, it can't provide enough and the unemployment rate stays stubbornly high, the inflation rate starts to move down. That brings about a monetary policy tightening, despite the Fed doing nothing. The unemployment rate goes higher, inflation goes down further, monetary policy is being tightened further, and then you end up and potentially, inflation could go negative. That's a deflationary spiral and that's a very dangerous situation that was dodged in 2010, '11, and I hope it's dodged again, but it may not be. And if the Fed sees that as a serious risk that's when I think they're going to dig deeper into their toolkit and do some additional things.

Nelson: So, one of those things that they mentioned a lot in the minutes, which seemed to be an LSAP, Large Scale Asset Purchase and the Fed distinguishes its Large Scale Asset Purchases from it. It doesn't call them QE because it's not about providing reserves, it's about buying longer term securities and pushing down long-term rates. I think that that would be a sort of an instinctive natural thing to do. And as we mentioned earlier, as discussed earlier, they're already purchasing assets at a greater pace even though they say it's not an LSAP. And moreover, term premium that's sort of the part of longer term rates that can't be explained by the path of short- term rates are at record lows and have been there for a while by the Fed's models. So, there may not the lot that they can do, but that's sort of one of their standard extraordinary tools that I'm sure that they would go to.

Nelson: The second thing they can do, I think they probably would do, that we were talking about earlier would be to ramp up their corporate bond purchases. I think that they would see that as a means to provide additional stimulus effectively, just like an LSAP, but they're trying to put negative pressure on corporate bonds first. So, they could also engage in a funding for lending program. So, other central banks have effectively offered banks super cheap, long-term funding, as long as the banks lend, it's sort of reward for lending. The Fed hasn't done that this time around and the track record of those programs hasn't been very good, honestly, but it does seem like that's something that the Fed could consider doing.

Nelson: It could couple that with the term auction facility, which it didn't, has not yet owed. I mentioned this earlier, this is their auctioning of discipline. And it could do something like, "We're going to auction discount window loans, we're going to have a stop out rate of -1%, and we're going to only allow you to bid up to the amount that you've increased your small business lending or your business loan or some targeted amount over the last month." And then that would give banks at a margin, a very strong incentive to increase those things in order to earn that very low rate and you wouldn't need to have those loans as collateral. You could use all discount window collateral, so it can be super safe.

Nelson: I think the problem fundamentally that the economy is facing is not illiquidity. Illiquidity is something a central bank can solve. It's risk. It's sort of unbounded almost, unquantifiable risk, and because this is an unprecedented situation for the economy, the economy could do very poorly, losses could be high. And the only way you're going to solve a problem like that is well, is I'm going to contradict myself, that would be to take on some of that risk and it just doesn't look like the Fed and more importantly the Treasury is willing to take that stance.

The problem fundamentally that the economy is facing is not illiquidity. Illiquidity is something a central bank can solve. It's risk. It's sort of unbounded almost, unquantifiable risk, and because this is an unprecedented situation for the economy, the economy could do very poorly, losses could be high.

Nelson: Now, you could in this kind of funding for lending approach attempt to in effect, convince banks to take on more risk by offering them the very low rates, but it's very hard to do that. And you also come back to the fact that banks are very reluctant to borrow and they're also very, very reluctant, completely unwilling to participate in anything that could be cast as a bailout of the banks. Now, this wouldn't be a bailout. The banks would be doing the bidding of the government because the government is trying to convince them to make more loans. The banks don't need the money, right?

Beckworth: Yeah.

Nelson: But nevertheless, you could see how this could be cast as a bailout. And I think, it would be difficult to get banks to do that. It's possible, especially if it's an auction and the guy next to you is getting this cheap funding. It might work, but I'm not that optimistic. And then the last thing that you mentioned, that I mentioned would be negative interest rates. So, as it's true around the world, central banks zero isn't necessarily the lower bound. The Fed and the central bank can push rates below zero. The evidence is that those actions are at best modestly effective and at the same time, they can damage the banking system or the credit intermediation system. So, I'm pretty sure that that's why the Fed has been quite definitive that they're not going to go that way. And I think that that's certainly the plan not to pursue negative interest rates. But in the event of a dangerous deflationary spiral, I think all bets are off, and that will become a possibility.

Beckworth: Okay, Bill, so with negative rates how about the Fed doing it this way: the Fed lowers interest in excess reserves to say -0.5% and then it lowers its discount rates say -1%, which goes back to this what you're mentioning earlier, the banks could borrow and they're all negative, but you're still making it profitable for banks to get funds from the Fed, as you described earlier, and then make those loans out for the real economy, but you're introducing negative rates, but in a manner that preserves for them some semblance of profitability. So, they make the loans to the customers, presumably, there's something in it for them as well because the discount rate would be below the interest in excess reserve amount.

Nelson: Right, right, right. I see where you're going with that. I haven't really thought about that. So, the Fed would be losing money on those transactions, right?

Beckworth: And it would be subsidizing banks. I mean, it would-

Nelson: It would be in effect subsidizing banks. Although, I think that the level of reserve balances would vastly exceed the level of discount window bonds unless the Feds tries to sell-

Beckworth: That's a good point.

Nelson: -its huge portfolio of Treasury and agency securities. So, it would be countering to some extent, the money that it's sucking out of the banking system through those negative interest rates on reserve balances with these negative loans, but only insofar as the Fed's balance sheet was constructed to a large extent with loans. And so, you could have a two-tiered system, maybe where you pay zero on reserves up to the point where, the ECB has all these different sorts of tiers in place, to try to sort of soften the blow. And I'm sure that there would probably be a way to do that.

Nelson: And ultimately, in the end, I'm just a bit of, I think monetary policy works through the interest rate channel and not through the credit channel or any kind of monetary aggregate, but basically, when the rates are lower aggregate demand goes up because it looks more attractive to buy that refrigerator now or that car or for your plant to put in that vertical lay because all of those things have a better net present value than they did before and that's going to be true if you move rates from one to negative one, just as it's true if you move rates from five to four to negative to three.

Nelson: So I just think that it's just very difficult to move rates down, it causes a lot of distortion within the financial system, and ultimately, it can weaken your banks which can have an offsetting effect and it could be a considerable effect, basically because only banks, banks have to hold reserves and only banks have to hold reserves. So, if you're charging a negative rate on those reserves, you're penalizing the banks normally, so they can't get out.

Beckworth: And that's fair and that's why negative rates are so controversial. Banks, profitability can be impaired and again, the answer that would be, "We'll find a way to preserve their net interest margins." If you lower all rates proportionally, so that they're still making money and they're lending. But of course, that's easier said than done, what I suggested was one way to possibly do that. The other thing, of course, is that many people have a hard time with negative rates just period. They think it's artificial in some manner, as opposed to thinking "Well, maybe our star or the natural rate has fallen for reasons based on the economy itself."

Beckworth: So, I agree it's challenging. And finally, negative rates run up against at some point the true lower bound because of physical cash. At some point, it becomes worthwhile to hold physical cash, there's storage costs at some point. The storage cost might be less than the amount that would you would lose at a bank. So, there are definitely roadblocks to implementing negative rates and that's why some people have called for really radical proposals, which I don't want to go there like eliminating cash all together. I definitely am not in that camp.

Beckworth: All right, so we've talked about yield curve control, we've talked about interest rates, potentially negative interest rates. Let me throw out one last possibility that was also talked about in the June FOMC minutes, and that was the adoption of makeup policy, and you've touched on this really already, but allowing rates to stay at zero even if inflation goes above 2% temporarily. Again, the goal is still to keep long run price stability. So maybe over the next decade, you still have 2% inflation, but you temporarily have some catch up, so you let the economy run hot. You don't sweat the little surge in inflation, and things are better.

Beckworth: And I know you mentioned earlier that this recession was a real shock, a supply side shock, but there's also evidence that's spilled over and become there's demand damage as well and to the extent there is, this is maybe one way to make up for that is to run the economy hot, keep rates zero for longer, tolerate an inflation overshoot. So, what do you think of makeup policy in general first and secondly, as it applies to the current situation?

Makeup Policy 

Nelson: So, as you're aware, the Fed has been engaged in this consideration of its of its framework, not the corridor or floor kind of framework, but actually how it articulates and pursues its monetary policy for seems like it's been a couple of years now, at least a year or so and a really, sort of what that discussion in some sense all boils down to, I would say is, should the Fed target and hit an inflation rate of 2% and effective let bygones be bygones, it will always be seen to be 2% regardless of what's happened in the past, or should the Fed be targeting something like average inflation.

Nelson: So it says, "Well, if inflation has been low, I'm going to try to get it to be a little high, so that for some window or some average, it's at about 2%," or there are more radical versions of what is effectively that which would be something like nominal income targeting. You say, "We're going to target the growth of nominal income at trend and we'll make up any gap, any lost nominal income in the future." I find nominal income targeting and price level targeting which is a similar variant, I think that does present almost insurmountable communication difficulties because it's very hard to convince people, to explain to people what they're doing. Other people disagree. A lot of people are big fans. I just think it would be very hard to communicate.

Nelson: And there's pretty convincing research, I think out there that says, inflation average targeting doesn't really do that much and that kind of seems like to me, it's not going to make that big of a difference if you say we're addressing 2% on average versus 2% spot. But it does seem quite possible to me that they're waiting to finish that consideration before they change their forward guidance. So, the forward guidance right now is basically it's at zero and we're going to keep it at zero until the economy have recovered from this.

Nelson: And this is a good situation for having forward guidance, because of the uncertainty that we discussed of the form, "We're going to keep rates at zero until the unemployment rate gets down below some level and inflation gets up above somewhat." Now, I would argue that this is an especially good time to simply have an inflation target and that's because the only reason not to get the unemployment rate down as low as you can get it is that ultimately it could cause inflation. And right now, because of the aggregate supply damage that you mentioned, it's really hard to say what the natural rate of unemployment is. The unemployment rate was running well below 4% before this.

I would argue that this is an especially good time to simply have an inflation target and that's because the only reason not to get the unemployment rate down as low as you can get it is that ultimately it could cause inflation. And right now, because of the aggregate supply damage that you mentioned, it's really hard to say what the natural rate of unemployment is.

Nelson: If you were to announce an unemployment side to your threshold-base guidance, it's very hard to know what to announce, right? So, I suppose you said we're going to keep rates at zero until the unemployment rate gets back down to where it was at 3.7%, but the supplies that are damaged has actually moved the NRU up to 6.5%, then you've promised to get the unemployment rate far below the NRU and that puts you in a very difficult situation. You're really going to regret that.

Nelson: But on the other hand, suppose you said, "Well, right now is the time. We're going to be really providing on stimulus, we're going to certainly get it down at least below seven." That could be very dispiriting for people, when they could be saying, "Wow, you've already given up." About three percentage points of unemployment. So, I think a preferable approach would be, say something like, and meanwhile, the Fed has done a lousy job of achieving its inflation target. It's been on the low side of it, really, for a decade, at least, right? And so if the Fed were to say, "We're going keep interest rates at zero until the 12-month average inflation rate is above two for six months," that would be quite a promise and it would be one that's sort of incentive compatible, and it would be one that's robust to not really knowing what the NRU is, what the natural rate of unemployment is.

Nelson: So, and this is a very natural step for them to take, combined with the conclusion of their reconsideration of whether how they're articulating their policy, how they're pursuing monetary policy, because it really involves this average inflation issue quite a bit. Now, I kind of hope that they're not like waiting to September, to take big changes in order to just conclude their internal study, because if this is the right thing to do, and if they think it's the right thing to do, then they should get on with it because we're in a very severe recession, extraordinarily severe recession and this isn’t the time to wait to take appropriate action, just to finish up your internal stuff.

Beckworth: No, I agree with that entirely and I liked the way you framed that you don't want to commit to some R star or Y star or U star that you don't really understand, but you are wanting to commit to inflation running at least at 2% or maybe a little bit above for a certain amount of time. Some persistence there so you know that you are running the economy hot, you know you've reached that point. So, that would be great and I would love to see it happen. I would like to see some manifestation of makeup policy emerge sooner rather than later. Well, with that, our time is up. Our guest today has been Bill Nelson. Bill, thank you so much for coming back on the show.

Nelson: It's a real pleasure, David. Thank you for having me.

Photo by Eric Baradat via Getty Images