Related Links

---

• The Fed Is Shrinking Its Balance Sheet. What Does That Mean?
• Projecting the Evolution of the Fed's Balance Sheet
• What if the Federal Reserve Books Losses Because of Quantitative Easing?

Transcript

---

Tim Sablik: Hello, I'm Tim Sablik, a senior economics writer at the Richmond Fed. My guest today is Huberto Ennis, group vice president for macro and financial economics in the Research department at the Richmond Fed. Huberto, thanks for joining me.

Huberto Ennis: Thank you for having me, Tim. It's a pleasure.

Sablik: I recently wrote an article for our magazine, Econ Focus, about the Fed's balance sheet — how it has evolved over time and what the Fed is doing now. We'll put a link up to that article in the show transcript for anyone interested in reading more about that. Huberto has also been thinking about these issues for a long time, so I thought we could have more of a conversation about the Fed's balance sheet rather than a formal interview.

To start things off, I'll provide some background for the conversation. The size of the Fed's balance sheet has grown dramatically over the last 15 years. Prior to the financial crisis of 2008, it was less than $1 trillion and it more than quadrupled in size during that crisis. More recently, in response to the pandemic, the Fed increased its balance sheet again, topping out at nearly $9 trillion.

As with any balance sheet, the Fed's balance sheet consists of assets and liabilities. Prior to 2008, its assets were mostly Treasury securities and its liabilities consisted mostly of currency in circulation — the dollars that you or I carry in our wallets. When the Fed expanded its balance sheet through operations most commonly known as quantitative easing, or QE, it did so by purchasing Treasuries and mortgage-backed securities issued by government-sponsored enterprises. Our listeners have probably heard of some of these GSEs, which include Fannie Mae and Freddie Mac.

The Fed paid for these purchases by creating reserves on the liability side of its balance sheet. Reserves are cash balances that banks hold at the Fed and on which the Fed pays interest.

The reason the Fed engaged in this QE was to provide additional economic accommodation, after it had already reduced its policy interest rate as much as it could. By buying up longer term Treasuries and mortgage-backed securities, the Fed could reduce their supply, thereby increasing their price and lowering their yield. This would allow the Fed to lower long-term interest rates, providing more stimulus to the economy.

So Huberto, my first question for you is, now that the Fed has been using QE for a little over a decade, do economists have a good sense of how well it has worked in practice?

Ennis: Well, that's definitely an important question, one that is really difficult to answer conclusively.

For sure, the topic has received a lot of attention. The Fed needed extra tools when interest rates moved close to zero after the global financial crisis and again when the pandemic hit.

It is hard to know exactly how much the asset purchases helped the economy. There are multiple candidate channels, such as lowering long-term rates, signaling persistent monetary policy accommodation, increasing funding liquidity in financial markets. I'm sure we can come up with a few others.

These effects are hard to measure, and there's an active debate in the research community about their relative importance. But overall, I would say that the experience with balance sheet policy has been reasonable, not bad for sure.

Sablik: More recently, the Fed has begun shrinking the size of its balance sheet by allowing some maturing securities to roll off without being reinvested. I think one might reasonably expect that this quantitative tightening, or QT, would have the same effects as QE, just in reverse.

But there are some important differences between the two processes. QE often happens suddenly in response to an unexpected shock. For example, the Fed stepped in to purchase huge sums of Treasuries in March and April 2020, when the effects of the pandemic first hit the economy. In contrast, the Fed's approach to QT has been much more measured. It announced its plans for shrinking the balance sheet months in advance and that process is happening passively following a preset schedule. Why is the Fed taking this approach with QT?

Ennis: Well, before I address your question, I think it's important here to distinguish Fed interventions to deal with market functioning issues from the large asset purchase programs that happened in the first half of the last decade.

[The latter] were used as a policy response, an adaptation when interest rates became constrained by the effective lower bound — the inability to move rates further down once they are at or close to zero. When policy is limited that way, unconventional monetary policy, such as large asset purchases, really become an important alternative to the traditional interest rate policy. But generally, those purchases happen in a predictable and relatively stable, pre-announced pace.

As you said, the asset purchases that happened in March and April 2020 were abrupt, but those were motivated by stresses in financial markets — most critically, the Treasury market. I count those as a different kind of intervention. Of course, as the pandemic settled in and became a more lasting problem for the economy — with policy interest rates, again, down all the way to the effective lower bound — what started as a market functioning operation became also a more general monetary policy effort. This is understandable. The pandemic was an unprecedented situation and adjustments were being made as we learned.

At the same time, I think that an important lesson going forward is to make more of an explicit separation between these two types of interventions — the ones aimed at market function and the ones in increasing policy accommodation.

Sablik: Yeah, that's a really good point. It reminds me of what happened in the United Kingdom not too long ago, where the Bank of England stepped in to purchase government bonds to address disruptions in that market.

Ennis: That's right. There were different factors driving what happened in that situation versus what happened in the Treasury market in 2020. The aim of the Bank of England's intervention was to restore market functioning rather than increase accommodation. In fact, their general posture at the time was for removing accommodation. That made the distinction really stark.

Going back to your question about the Fed shrinking its balance sheet, I think the Fed approaches that as a natural process of normalization. It's true that the reduction of the Fed's portfolio may have some impact on financial conditions and long-term rates. But I think it's clear that the Fed wants to make short-term interest rate policy the principal tool for adjusting monetary policy right now.

That makes sense to me. When tightening monetary policy, there is really no constraint on what the Fed's interest rate policy can do, no effective bound at play. So, it makes sense to use the tool that we know best. We know how it works, how it affects the economy. I would say the Fed wants to normalize the size of its balance sheet but not necessarily use it as an active tool for policy.

Sablik: Yeah, that's another important distinction. The Fed doesn't need to shrink its balance sheet to tighten policy. It can do that just through raising interest rates. So that being the case, what do you think might be the Fed's reasons for shrinking its balance sheet now?

Ennis: Well, the way I see it, the size of the Fed's balance sheet is a byproduct of monetary policy actions taken during the pandemic and before for the purpose of improving market functioning, and subsequently to support economic conditions as the economy struggled with the impact of the pandemic. Now, the economy has mostly recovered and asset purchases are no longer a relevant tool. It's time then to start the process of normalization, gradually reducing the size of the balance sheet.

There are practical reasons to do it. But also, perhaps importantly, there are philosophical reasons for what we might call in minimalist approach to central banking. If the central bank has not a good reason to play a role, then just stay out of it. Let the private sector work through things. Trust the market. On the more practical side, I think it's good to reduce the size of the balance sheet now and be in a better position down the road in case unconventional monetary policy becomes relevant again.

Sablik: Right. Related to that minimalist approach, the Fed has also said that its long-term goal is to hold mostly Treasuries on the asset side of its balance sheet. Getting to a portfolio of mostly Treasuries could take a long time at the current pace that mortgage-backed securities are rolling off the balance sheet.

Some Fed officials have suggested that the Fed might need to consider selling those securities at some point to speed up the process. What are the potential challenges with actively selling securities?

Ennis: First and foremost, the Fed does not want to disrupt market function by actively selling securities in large quantities. The issue here is that the Fed, in its interventions in the financial markets, is a large player. It can move markets.

To avoid that, the approach has been to let the portfolio run down naturally as the securities mature. But like you said, that's likely to take a long time. MBS securities are of relatively long maturity, so not many of them roll off each month. Also, prepayment of mortgages that could accelerate that process have decreased significantly recently. Nobody wants to refinance these days, of course.

More generally, though, I think that the Fed will want to minimize the impact that its actions have on market functioning and on price determination. As we said before, it's good that the Fed leaves the market to work, undistorted. Fed intervention can scramble those market signals that tell investors where it's best to put their money. So, to the extent that the Fed does not have a good reason to play a role and influence price formation — something that could result from large asset sales — it is good that it stays on the side and minimize its impact.

Aside from that, there's a communication aspect of Fed action. If the Fed were to start selling securities, you want to be clear about how people need to interpret those sales. That's always tricky.

Sablik: One side effect of the Fed's large balance sheet is that the Fed earns interest on the assets it holds. It also pays interest on some of its liabilities, such as reserves. For much of the past decade, the Fed has earned more than it paid out and it remitted any profits above its operating costs back to the Treasury. But now that the Fed has been raising interest rates on reserves, it is actually paying out more than it earns on its assets.

William English and Donald Kohn, who were both formerly at the Fed's Board of Governors, recently wrote a post for the Brookings Institution, arguing that the temporary losses on the Fed's balance sheet don't pose any problem for its operations. As long as the Fed's earnings eventually turn positive, it can resume remittances to the Treasury. Do you agree with that?

Ennis: Yes, I think that makes sense. Of course, the time to assess the pros and cons of going into a tightening cycle with a large balance sheet is the time when that balance sheet build-up was happening. Now, the Fed has the large balance sheet and the financial implications are what they are. In other words, unconventional monetary policy involves some benefits but also some costs. The whole package needs to be evaluated from the start.

Having said this, I think that to the extent that the Fed can now reduce the size of their balance sheet in the background, without much influence on the rest of the situation, that seems helpful to me.

Sablik: The Fed hasn't said exactly how much it plans to shrink its balance sheet, but it has stated its intent to maintain an "ample" level of reserves. What does ample reserves mean?

Ennis: Monetary policy implementation is really about what the Fed does to implement the monetary policy directives of the Federal Open Market Committee. As it turns out, a system with ample reserves is able to better maintain interest rate control at the level that the Fed intends to target. Ample here means that banks have sufficient reserves to handle payment flows and other liquidity needs without much difficulty.

Sablik: Okay. So how will the Fed know when it is getting to the point where reserves are becoming less ample?

Ennis: In principle, you may see more fluctuations in short-term money market rates, more great volatility in general. However, the Fed now is well positioned to avoid that kind of volatility. Instead, what we might see is that some of the money market facilities that the Fed put in place recently — such as the standing repo facility or even the longstanding discount window — start attracting more activity from market participants.

I anticipate it will be a fairly smooth process. We'll probably see, first, deposit facilities such as the reverse repo facility decrease in importance, and then the lending facilities such as the standing repo facility getting participation. When we're close to the minimum level of ample reserves, the Fed can smoothly transition to a secular growth in its balance sheet, one that would accommodate the increased demand for currency in circulation and reserves in banks which are associated with the natural growth in economic activity and the US economy in general.

Sablik: Well, we could probably spend another 20 minutes or an hour talking about these issues, but we probably should stop there. Huberto, thanks so much for joining me today to unpack some of these topics about the Fed's balance sheet.

Ennis: It was my pleasure, Tim. Thank you.

Sablik: As always, I'll encourage listeners interested in keeping up with the research we're doing on this and other topics to head over to our website, Richmondfed.org. And if you enjoyed this conversation, please consider leaving us a rating and review.