Podcast
Important Information:
What's the Fed Doing with its Balance Sheet Now?
Important Information:
Huberto Ennis provides an updated view of the Federal Reserve's management of its balance sheet and the flow of reserves between banks, which were the focus of a recent conference held at the Federal Reserve Bank of Richmond. Ennis is group vice president for micro, macro and financial economics at the Richmond Fed.
Related Links
Transcript
Tim Sablik: My guest today is Huberto Ennis, group vice president for micro, macro, and financial economics in the Research Department at the Richmond Fed. Huberto, welcome back to the show.
Huberto Ennis: Thank you for having me back, Tim. It's a pleasure.
Sablik: Two years ago, you came on the show to talk about the Federal Reserve's balance sheet and how the Fed used asset purchases to respond to disruptions from the COVID-19 pandemic. As we discussed at that time, the Fed had begun the process of unwinding some of those purchases. Now seemed like a good time to check in and update listeners on the latest developments.
Using the balance sheet as a policy tool is something that Fed economists study closely. Our department recently hosted a workshop on the topic with other economists from inside and outside the Fed system. You were involved in organizing that event. Can you tell us what your objectives were for organizing it?
Ennis: This was an idea that came out of conversations I was having with Ricardo Reis. He is one of our long-term consultants who, like me, is very interested in central bank balance sheet topics.
Organizing one of these events takes time, so probably we started discussing this in the fall of 2024. At the time, Ricardo and I were anticipating that in late 2025 the topic of the Fed balance sheet would be in a phase of active discussion. And, we were right. The idea was to gather experts from universities, other central banks, and the Federal Reserve System and spend a day together learning from each other about these topics.
Sablik: What were some of the key highlights that you took away from these discussions?
Ennis: One of the highlights for me was the fact that we had more than 20 colleagues from around the system that came to attend the events and participate in the Q&As and in the informal discussions during coffee breaks and the lunch we had. There were one or two economists from each Reserve Bank and several colleagues from the Board and the [Open Market Trading] Desk at the New York Fed. Many of them made substantial contributions during the day. These are all experts across the system thinking hard about these topics and their knowledge really showed when they had the opportunity to comment on what was being presented. As you know, these are pretty complex issues and it was really great to spend the day with thoughtful and very knowledgeable people.
Sablik: Dallas Fed President Lorie Logan participated in a panel discussion at the event. She introduced an interesting proposal for making some changes to the way that the Fed approaches monetary policy implementation. Can you tell us about her proposal?
Ennis: Sure. It's a very interesting proposal.
First, let me say that President Logan posted her remarks online and I would encourage your listeners to take a look at those. She provides a very clear explanation, much better than what I would be able to do here.
As you know, the FOMC implements and communicates monetary policy, using as a guidepost the interest rate in the federal funds market. When banks hold plentiful reserves, as has happened for much of the time in the last few years, they do not have [many] incentives to trade with each other in the fed funds market. In fact, [many] of those trades we regularly see in that market are between financial institutions that cannot earn interest on reserves — they are the home loan banks, the housing GSEs, and also banks that more often than not are foreign banks. This implies that the market is smaller and more susceptible to experiencing sharp changes when conditions change.
President Logan, during the workshop, suggested that the Fed should consider moving away from the fed funds market towards repo markets, a much bigger and more active short-term cash market. Her proposal is to start using a benchmark rate in that repo market as the new Fed guidepost for monetary policy implementation.
Sablik: What other topics were discussed at the workshop?
Ennis: We had a good mix of academic and policy presentations. Massimo Rostagno from the ECB [European Central Bank] came all the way from Europe to discuss issues related to the ECB balance sheet and policy. Ricardo Lagos from NYU presented frontier work he's doing on the determinants of banks' demand for reserves. Annette Vissing-Jørgensen, another leading expert on these topics, discussed how fluctuations in the Treasury general account impact the process of interest rate control, which is a critical aspect in the implementation of monetary policy. The Treasury general account is basically the checking account that the Treasury has in its bank, which is the Fed.
President Logan also participated in a panel with Anil Kashyap from Chicago Booth and Jeremy Stein from Harvard. The panel was moderated by Ricardo Reis, whom as I mentioned earlier is one of our long-term consultants and is also a chair professor at the LSE [London School of Economics].
These are just some of the highlights. We had eight speakers and all of the talks were really, really interesting.
Sablik: You also presented some work with our colleague, Alex Wolman, who has been on the show before. Can you describe that work and why it's important?
Ennis: Yes, of course. We are working on a project that aims at improving our understanding of the process of diffusion of reserve balances across banks in the U.S. system. This is a topic that is often mentioned in policy discussion but, as far as I can tell, very little is known about it. And so, we are trying to make the initial steps on it.
We thought it might be good to start by taking a look at some data. And, that's what we are doing — looking at weekly data published by the Federal Reserve on balances held by groups of banks at a relatively high frequency, so we think it can be informative. We are basically looking at how reserve balances in the different groups change over time and how we can interpret those changes as giving information about the flows — how reserves move from some banks to other banks.
To the extent that some reserves are stuck at some banks, then the total quantity of reserves would need to be higher because other groups of banks may find themselves short. So, to know how much is sufficient reserves, you need to know how quickly those reserves tend to move from banks that have too many to banks that have too little.
Sablik: Even though this work is still preliminary, has it yielded any interesting insights yet?
Ennis: One finding that I think is pretty interesting is that you see those flows across groups of banks increase as the level of reserves become less plentiful, or abundant like the experts would say. So, when reserves are generally less abundant, they tend to get redistributed more quickly across banks. We see this in our data and I think it's intuitive. It also underscores a more general point — that the behavior of the different actors in this space can adapt to changing conditions in meaningful ways.
Sablik: When we last talked in 2023, the Fed had begun shrinking its balance sheet. More recently, the Fed has stopped that process and is starting to grow the balance sheet again to keep up with the growth in the economy and other factors. What motivated those decisions?
Ennis: The Federal Open Market Committee, the FOMC, announced in 2019 that it intended to implement monetary policy under a regime with an ample supply of reserves. By having sufficient reserves in the system, the way to exercise control over the level of short-term interest rates would primarily be through setting administered rates, such as the interest paid on reserves balances at the Fed, without any active management of the total supply of reserves.
In the last couple of years, as the balance sheet shrank, Fed liabilities had to decrease — initially balances at the overnight reverse repo facility were falling, but eventually also reserves started to fall. The Fed monitors market conditions and recently decided that reserves were starting to transition from being abundant to being ample, so the committee decided to stop shrinking the balance sheet.
Let me clarify the terminology here. The committee has stated that in the longer run, the Fed will hold no more securities than necessary to implement monetary policy efficiently and effectively. The words "ample reserves" have become the standard terminology to refer to the level of aggregate reserves consistent with that — not too large, not too small. Abundant reserves, instead, are levels that are larger than ample and considered unnecessarily too large.
Sablik: Can you briefly summarize how the balance sheet is expected to evolve from here?
Ennis: Right now, the Fed is adjusting balance sheet policy and reserves right now are leaving the abundant level and entering into the ample range. The idea going forward is that for the first few months in 2026, reserves are going to grow at a relatively quick pace, anticipating fluctuations in factors that influence reserves related to tax policy. And then, after around April, reserves will start growing at a more moderate pace that basically keeps up with the growth in the economy.
Sablik: Aside from the size of the balance sheet, several policymakers have recently discussed the composition of assets that the Fed should hold going forward. Why might this matter, and what factors will the Fed weigh when deciding what to hold?
Ennis: Yeah, that is a really interesting aspect of the discussion. The Fed has said that it intends to hold primarily Treasury securities in order to minimize the influence that its holdings of assets might have on the allocation of credit in the economy. But, even when deciding to hold only Treasury securities, there is a question about the maturity of those securities.
When the Fed conducts large asset purchases for monetary policy purposes — what is often called QE [quantitative easing] — it tends to buy long-maturity bonds to affect long-term yields and, in that way, economic activity. When there is no more need for QE and the Fed embarks in balance sheet normalization, it finds itself holding a portfolio of securities that is relatively tilted to the long-term assets that it bought during the QE.
In order to move towards normalizing the maturity composition of the Fed's asset portfolio, recent announcements say that the Fed will start buying mostly Treasury bills, which are securities of much shorter maturity. The pros and cons of the different possible compositions of the Fed balance sheet are currently actively debated within the Fed.
Sablik: Are there other questions around the balance sheet policy that you and your fellow economists are continuing to explore?
Ennis: One topic that we have not discussed but that is receiving a lot of attention among policymakers and researchers is the role and the design of different methods and tools for intervention that the Fed can use to enhance its ability to control short-term rates in the process of implementation of monetary policy. One of those tools is the standing repo operations, which just recently were redesigned to make them more effective. There is a lot of thinking going on about the way to structure those tools. This is a rather technical area, but it is really important stuff.
Sablik: Huberto, thanks for joining me today to update us on the Fed's balance sheet.
