On Friday, March 1, Waller, who had accurately predicted the direction of the U.S. economy, proposed at the U.S. Monetary Policy Forum that the Federal Reserve should "buy short and sell long" on its balance sheet, which instantly ignited market risk sentiment, U.S. bond yields fell sharply, and gold prices and U.S. stocks rose sharply. Waller noted that the Fed should increase the proportion of short-term U.S. debt in its balance sheet and reduce its holdings of institutional mortgage supporting** (MBS) to zero.
The proposal is similar to the Fed's previous "operation twist." However, the OT implemented from 2011 to 2012 was "selling short and buying long", that is, selling short-term bonds and buying long-term U.S. bonds at the same time. ButWaller is "buying short and selling long" this time, which is considered by the market to be a "reverse twist operation".
Wall Street News has previously mentioned that Waller accurately judged a soft landing for the US economy two years ago and could be a strong candidate for the next Fed chairman. Therefore, the market is also paying close attention to his remarks. So, in what context does Waller mention "buy short and sell long"? What is the logic? What does he think of the Fed's QE and QT policies?
The following is a translation of the full text of the speech, which can be found on the official website of the Federal ReserveReflections on quantitative tightening, including a review of Global Quantitative Tightening: What Have We Learned? ".
Fed Governor Christopher Waller.
2024 U.S. Monetary Policy Forum, New York.
Thank you, it's a pleasure to be here. I'm excited to join this panel discussion on one of the policy actions that are being implemented by the world's largest banks: quantitative tightening (QT). I'd like to thank Kristin, Matt and Wenxin for putting together a great report that provides an overview of the impact of Qt in the Big Seven.
Quantitative easing, or QE, often referred to by banks as "massive asset purchases" (LSAPs), as a tool to increase monetary policy easing, while QT is used to tighten policy, a view that has changed over time. When it was used during and after the global financial crisis, QE was seen as a "non-traditional" tool in the bank's arsenal. However, QE has been used many times over the past two decades and for long periods of time when the policy rate is at the effective lower bound, so I would say it's no longer unconventional.
Given the role of QE and QT in the policy toolkit, it would be good for researchers and policymakers to examine how asset purchases work and discuss current issues related to their implementation. This article is a very timely and comprehensive review of countries' experiences with QE and QT. There's so much to be said in this work that it's hard to fully assess in our time today. Therefore, I focus my review on four points: (1) evidence of the asymmetry of the effect of QE with that of QT; (2) Enforcement of QE and QT in the United States; (3) the role of the announcement effect of QT; And finally, (4) who took the Fed's place to buy assets when we exited the market. I will then conclude with reflections on some of the issues the Fed faces in normalizing its balance sheet.
For me, one of the most interesting results of this article is that the announcement effect of QE is much greater than that of quantitative tightening. The authors found that the Qt announcement had a small but statistically significant effect on increasing bond yields – about 4 to 8 basis points. But in absolute terms, this effect is much smaller than the estimate that the QE announcement will lead to a decline in yields. The conclusion is that the interest rate effects of QE and QT are asymmetrical. To economists, this result may seem confusing – why would changing the symbol of an action lead to an asymmetric effect on ** and real variables?
Ever since the bank launched QE in response to the global financial crisis, its effectiveness has been debated in academia. One view is that when ** banks trade their zero-interest reserves for shorter-term Treasury bonds with zero interest rates, it has a very limited effect. Because these two assets seem to be almost perfect substitutes, changing the combination of these two assets cannot have a ** effect. The bank simply traded two $10 bills for a twenty-dollar bill. If this is true, then reversing such a transaction via QT has no effect either. It should be symmetrical.
Another view, based on market segmentation or preferred habitat theory, is that when banks pay with reserves, it reduces the amount of these that are available to private investors, which pushes the bar and lowers the interest rate. By lowering interest rates on longer-maturity assets, which have higher interest rates than reserves, banks can stimulate the economy in a similar way to lowering policy rates. But according to this logic, when QT reverses QE, the asset** should fall and the yield should rise by the same amount. Therefore, any positive effects obtained from QE are reversed when QT occurs. This suggests that QE and QT may cancel each other out in terms of benefits. But what is the point of carrying it out if there is no net benefit from the action? To illustrate this, let's say someone uses a ** drug, and they lose 80 pounds, but when the drug is taken away, they gain another 80 pounds. If there's no net benefit gain, what's the point of this exercise?
For me, for the net benefit of QE to be positive, there has to be an asymmetry of the effect of QE relative to QT. My thinking on this has long been guided by my ** conclusion with Alex Berentsen about the best stability policy, which is exactly what QE and QT should ultimately focus on. The gist of this argument is that when there is a sudden shock and friction in a transaction, banks can take actions such as injecting reserves to ease the friction or credit constraints and improve welfare. However, wait until the friction and impact have dissipated before revoking the injection, and the positive effect will not be reversed. For example, when a house catches fire, pouring water on the fire will extinguish the fire, which is of great benefit to all. But when the fire is extinguished, draining the water does not re-ignite the fire – the original benefit is not reversed. The point here is that QE is conducted in different market conditions than when QT occurred, so it is not surprising that the effect is different. The author's discovery of the asymmetry between the effects of QE and QT is not a mystery, but shows that the bank has done a good job in timing QE and QT to put society in a better state.
The impact of shifting to QE and QT on interest rates is usually focused on term premiums. There are three key elements to asset purchases that affect term premiums: (1) the expected path of the QE, including the volume and timing of purchases; (2) the length of time the bank expects to hold the additional; and (3) the expected path of Qt, including the amount and timing of redemptions, which largely depends on the size of the bank's eventual holdings (and reserve balances). These expectations are formed once the asset purchase plan is announced, creating a term premium effect on interest rates, or TPE.
Over time, TPE will change based on the passage of time and any updates that the public expects from the various components I have just mentioned. Let me touch on three factors that influence the expected path of asset purchases and interest rates. These factors are something to consider in future policy decisions.
First, there are two ways to implement QE, and they have different effects on interest rates. These are what I call closed or open QE plans. A closed-end QE program involves announcing the purchase of assets in a fixed inventory for a fixed period of time. An example of this type of asset purchase program was launched by the Federal Reserve in March 2009. OpenQE simply gives the monthly purchase volume, but there is no calendar end point, so the expected plan size is not specified. A range of economic conditions may be prescribed to reduce or end purchases, but when this occurs is not entirely possible**. Here, consider the Fed's recent asset purchase program.
At the time of the asset purchase announcement, it will be easier for the market to fully price the closed-end plan as both its purchase volume and end date are given, while the pricing of the open-ended plan will depend on the market's expectation of the evolution of the economy. So, if you want to have a specific impact on interest rates on the date of the announcement, you may prefer a closed-end program, or you may realize that you need more guidance on the expected path of an open-ended program.
Over time and the evolution of the economy, the two programs work differently. Over time, people may prefer an open-ended plan because it responds dynamically to the evolution of economic conditions. As conditions improve or worsen, the program can be suspended or extended, unlike closed programs. But of course, the criteria set in an open plan must be carefully considered. As I said in a recent Feds Note and several presentations, the criteria for starting QT in 2020 may have been too strict and did not allow the committee to start tapering as early and gradual as possible. It is very difficult to set the right standards in advance to create the flexibility needed to respond to changes in economic and financial conditions.
The second factor influencing the asset purchase path is that it is important that QE is followed by QT that can be trusted. If QE is seen as merely a permanent injection of money into the economy, then it could trigger inflation. This was widespread in 2009**, but inflation did not happen. Why? In my view, this is because the Fed has credibly committed to withdrawing the injected reserves at a later date. The advance commitment to QT is the reason for injecting reserves into the economy without triggering inflation or other long-term distortions in market pricing. Therefore, when starting an asset purchase or weighing how to deal with asset loss or sale, it is important that the bank commits to normalizing its balance sheet.
The third factor is that it is important to tread carefully when the bank is approaching the end of QT and the desired level of sufficient reserves. The endpoint should be related to the banking system's expectation of demand for reserves. In the US, we saw pressure on money markets in the fall of 2019 when the level of reserves was reduced to July during the normalization of the balance sheet, and then there was a large issuance of Treasuries in September. Reserve levels may be a little too low. Learning from our experience and trying to understand how the need for reserves changes over time shows that it is wise to move carefully towards the end of Qt.
So, even though QE is an open-ended plan, QT is more likely to resemble a closed-end plan. Banks usually have an idea of how big they want their balance sheet to be at the end of Qt; Therefore, once the speed of Qt is announced, the market should be able to effectively price the entire plan when the plan is announced. After that, the actual execution of Qt only validated the beliefs of market participants at the time of the announcement. That's why many refer to QT as merely a reserve that is not needed for emissions, which should be as interesting as watching the paint dry.
Now let me turn more directly to the authors' ** and their two findings. First of all, as I mentioned earlier, they found that the QT announcement of the ** bank had only a small impact on interest rates. To carry out this analysis, the authors conducted an incident study around the QT announcement, which required them to identify the "surprises" in the QT announcement. As the author admits, this is not a simple task. My comment here is to point out why the QT announcement surprise is challenging when considering the US example.
Let me review the evolution of the Fed's Qt communication in the spring of 2022 and consider how the various communications affected the expected path of Qt. Recall that QE ended in March 2022. Heading into April, the market may expect the Fed's redemption path to be similar to the 2017-2019 QT program. 9 The plan is to implement redemptions in phases over a 12-month period, eventually allowing up to $30 billion in Treasury bonds and $20 billion in mortgage-backed (MBS) to be redeemed each month. On April 5, 2022, then Vice Chairman Lael Brainard noted in a speech that the balance sheet would shrink faster compared to 2017-2019; Specifically, she said that "the maximum limit and the time to reach the maximum limit will be significantly shorter compared to 2017-2019." "The next day, the Federal Open Market Committee (FOMC) minutes provided additional information on the expected maximum monthly limit and phase-in period, saying that participants generally agreed on a three-month phase-in and $60 billion and $35 billion limits for Treasuries and **MBS.
On those two days in April, the market may have updated their expected QT plan for faster and larger redemptions. This change will be associated with a less negative term premium effect, implying higher Treasury yields. In the two days leading up to the vice chair's speech and the FOMC minutes, the 10-year Treasury yield rose 19 basis points — 12 basis points on the day of her speech and another 7 basis points on the day of the FOMC minutes — for a total of 37 basis points that week.
About a month later, on May 4, 2022, the FOMC communicated its "plan to reduce the size of the Fed's balance sheet." The program was in line with the April FOMC minutes and was little changed on the day and week of the 10-year Treasury yield (down 4 basis points on the day of the announcement and 2 basis points over the five-day period). Therefore, when conducting an incident study, it may be difficult to estimate the full impact of the QT announcement simply by looking at the official announcement of the QT program.
Let me turn to the second point, about what types of investors have increased their holdings because the Fed has reduced their holdings. When banks stop buying assets and start reducing their balance sheets, a common question is, who will take the bank's place to buy? I always replied, "Does it matter?" "If the bond market is broad and deep, there will be a lot of buyers – no need to worry about who will buy the debt. However, if the bond market is not broad or deep, then the actions of the banks may have an adverse and unwanted impact on the functioning of the market and the market. This will affect how quickly banks can reduce their balance sheets and whether they can do this passively or proactively.
It's also important to say that knowing buyers helps to understand the transmission of Qt to assets** and interest rates. Does it matter if a bank or non-bank financial institution is buying? Does it matter if it's a hedge**, retirement**, or actual family buying? Before fully understanding the answer to this question, there needs to be a better understanding of why this question is being raised.
The authors focus on the decline in total holdings of banks and find that households and brokers are the main investors who absorb redemptions. For my discussion today, I decided to dig deeper into the U.S. Financial Account in two ways. First, I decided to look at each type of ** (Treasury and **MBS) separately. For treasuries, I also found that since the start of QT in 2022, households have increased their market share, and brokers have also increased their share. For MBS, not only has the market share of these two types of investors increased, but so has the market share of the money market.
Second, I delved deeper into the family category. In the current classification, the U.S. Financial Accounts' household category includes hedging**. The Federal Reserve Board is working to separate hedges in this data set**. During this period, the Board of Directors released separate data on the domestic hedge** balance sheet. Using this supplementary data, I found that it was not hedged** responsible for the increase in household market share. This means that the increase is being driven by other home investors: actual families and nonprofits.
What do I think of this finding? My interpretation is that this reinforces the broad and deep view of the demand for Treasuries – that buyers are not a handful of deep-pocketed, savvy investors, but the American public. Therefore, the speed of outflow is not a problem. As we have seen in the current phase of Qt, the monthly outflows of up to $95 billion have not caused significant stress in financial markets – which would have surprised many people a few years ago, given the widespread concerns about QT prior to 2022.
Let me conclude with a few comments on what I think the Fed should be headed for as it continues to normalize its balance sheet. By "normalizing", I mean reducing the size of the balance sheet, but keeping enough assets to manage monetary policy using a system of adequate reserves.
As the Fed continues with its QT program, I support further thinking about how much more to redeem**. We have an overnight reverse repo facility that absorbs over $500 billion, and I think that money is excess liquidity that financial market participants don't want, so that tells me that we can continue to reduce our holdings for a while.
In addition, it is important to remember that we now have a standing repurchase agreement facility (SRF). The SRF acts as a backstop for the money market because it accepts Treasury bonds as well as **MBS and puts reserves into the banking system. This facility may allow banks to reduce reserve levels to levels that would have been without the facility and may provide a signal that reserves are close to adequate.
Chair Powell has already noted that the FOMC will begin discussions at this month's FOMC meeting on slowing our QT, which will help us transition to what we see fit. A change in our QT pace will occur when the committee makes its decision, the timing of which will be independent of any change to the policy rate target. Balance sheet planning is about getting the right level of liquidity and approaching "adequacy" at the right rate. They do not imply any hint at the stance of interest rate policy, which is concerned with influencing the macroeconomy and fulfilling our dual mandate.
Given the long-term issues associated with the Fed's portfolio, I would like to mention two things. First, I'd like to see the Fed's MBS holdings go to zero. **MBS holdings have been slow to decrease, averaging around $15 billion recently in recent times, as interest rates on underlying mortgages are very low and prepayments are very small. I think it's important to continue to reduce those holdings.
Second, I would like to see a shift in Treasury holdings to a larger share of short-term Treasuries. Before the global financial crisis, about one-third of our portfolios were Treasury bills. Today, Treasury bills are less than 5% of our Treasury bond holdings and less than 3% of our total ** holdings. The shift to more Treasury bills would bring the maturity structure closer to our policy rate – the overnight Federal** rate – and allow our income and spending to rise and fall as the FOMC raises and lowers the target range. This approach can also assist with future asset purchase programs, as we can allow short-term** rollovers off the portfolio without increasing the balance sheet. This is something that the FOMC will need to decide in the coming years.
In conclusion, let me be clear, this is a great article that will be a major reference for researchers and banks. The author's analysis will certainly have a longer shelf life than my discussion of it.
Wall Street news, welcome **app to see more.