Structured Debt

Tapering without the tantrum

Structured Debt

Tapering without the tantrum

John Kerschner, Head of U.S. Securitized Products, Nick Childs Portfolio Manager and Thomas Polus Fixed-Income Trader discuss the outlook for mortgage-backed securities as the U.S. Federal Reserve prepares to “taper” its purchases.

Key Takeaways

  • The Fed will soon begin to taper its purchases of U.S. Treasuries and mortgage-backed securities.
  • We think the Fed learned from the 2013-2014 "taper tantrum." While bouts of volatility may arise, the Fed's efforts to better communicate intentions should help to mitigate volatility overall.
  • MBS may actually benefit from tapering if they are favored over “risk assets,” including corporate bonds, as the tapering process unfolds.

The amount of fiscal and monetary support provided to the U.S. economy in 2020 broke records and, in our view, set new precedents for intervention in the financial markets by the U.S. Federal Reserve (Fed). But a lot has changed in a year. Both macro and microeconomic fundamentals have improved, corporate bond markets have rallied toward historical spread lows, and securitized products have recovered broadly. As a sustained economic recovery appears to be the market’s base case, it is now contemplating when and how fast the Fed’s monetary stimulus will be withdrawn. We have written much about our expectations for interest rate policy, but what effect will reducing the current pace of quantitative easing (that is, the direct purchase of securities in the open market) have on credit markets generally, and the mortgage market specifically?

The Fed was aggressive in 2020

Before COVID-19 struck, the Fed owned about 23% of the outstanding agency mortgage-backed securities (MBS) . As COVID spread, the Fed started buying agency MBS1 and Treasuries in the open market to provide liquidity to financial institutions, and to help lower mortgage rates. While the liquidity provided short-term medicine, lowering mortgage rates was a longer-term strategy to provide consumers the opportunity to refinance their mortgages at cheaper rates, thereby putting more disposable income in their pockets. The strategy worked. Mortgage rates plummeted as Treasury yields fell, refinancing spiked2, and the Fed aggressively bought the new mortgages.

Figure 1: Fed holdings of agency MBS

Fed Holdings of Agency MBS

Source: Bloomberg, as of 31 August 2021.

As of late September, the Fed owned over 30% of the outstanding agency MBS market, or approximately $2.5 trillion3. And the Fed is still buying. Estimates vary, but it seems likely the Fed will end up owning around 40% of the market before reaching the point where the volume they are buying is lower than the rate at which their holdings mature – that is, paid off or “pre-paid” as homeowners refinance or move. Regardless of where that point is, whether it is somewhat higher or lower than 40%, it is less of a concern to the market then how the Fed plans to ultimately reduce their exposure.

The market remembers 2013-2014

The Fed instituted a similar, but smaller scale, MBS and Treasury purchasing program during the Global Financial Crisis (GFC). Unfortunately, their attempt to unwind that program, or “taper” their purchasing, was something of a debacle. When the markets caught wind that Treasury purchases would slow, 10-year yields spiked around 1.0% in just a few months. The period came to be known as the “taper tantrum.”

But, in our view, the volatility in bond markets did not happen because the Fed pivoted to a more restrictive policy, but because they failed to give the markets time to digest their plan. Volatility is, almost by definition, dependent on there being a surprise. In 2013, the market leapt to a worst-case conclusion because the Fed failed to communicate, in advance, their intentions. Just as raising official policy rates can have dramatic market implications if it comes as a surprise but can result in orderly market moves if expected, a successful tapering needs successful communication.

We think the Fed learned their lesson from 2013-14 and has thus been much more intentional about telegraphing their thinking in advance. Chairman Jerome Powell said, now somewhat famously, as long ago as June that “you can think of this meeting as the talking-about-talking-about meeting4.” While he was referencing the Federal Open Market Committee meeting that had just taken place, where interest rate policy is set, his point was clear: Before the Fed will even talk about tightening monetary policy, they will first talk about talking about it. Put differently, they will take this process in steps, will keep the market informed, and the market (starting from June) should consider just the first step taken. The result was that the 10-year bond yields rose very briefly, only to fall over the next month.

The Fed’s intentions for tapering MBS purchases have followed a similar path. As such, the market’s estimates for when it will begin, and at what rate, have progressively narrowed. Simply put, the Fed does not want to surprise the market because the consequences of doing so are too well understood. And, thus, should the market be wrong in its expectations, we expect the Fed would (gently) nudge the market in the right direction.

Expectations are for a late 2021 start, and a modest pace

Our view is that the Fed will most likely announce the date they will start tapering their MBS purchases after their November meeting. We expect them to say they will begin soon after that meeting and will start by reducing purchases of MBS by an additional $5 billion each month. This view is a fairly consensus position insofar as most market participants believe the start date will quickly follow the announcement date and the estimates of the dollar amount are similarly narrow.

Given the volume of bonds currently owned and being purchased, we expect the Fed’s tapering will last less than a year, but getting the size of their MBS portfolio to their target could take a few years to complete. We think the Fed would like to get back to owning closer to 20% of the market, near the level they held pre-COVID. To be clear, we do not expect the Fed will ever outright sell MBS in the open market – which could raise supply concerns – rather, they will buy fewer bonds over time while letting the ones they do hold mature or prepay.

How quickly they will get back to 20% depends, in part, on the outlook for the economy, but also on the outlook for interest rates, given how sensitive prepayment rates are to government bond yields. If Treasury yields see a sustained rise on either fears of inflation or a recovery in real rates, mortgage rates will follow, which could dampen the speed at which they prepay. Equally, if Treasury rates stay in a more stable range, many homeowners who should or could refinance are more likely to have already done so, also dampening the prepay speed. Given our base case estimates that the U.S. economy will continue to recover, returning to trend growth in the next couple of years, and that interest rates will rise but largely in an orderly fashion, we do not expect dramatic changes in prepayment rates. As such, we expect a relatively modest decline in the Fed’s MBS holdings over the coming years.

MBS, perhaps counterintuitively, could benefit from tapering

As markets first began to digest the Fed’s intentions to taper purchases, MBS struggled. While total returns of mortgage indices were supported by declining Treasury yields in recent months, spreads did widen as expectations for tapering grew5. That they have stabilized in the third quarter suggests the market has reached a consensus on its expectations for the pace of Fed tapering, and their comfort level with it.

It is hard to price a “fair” value for the effect tapering will have on the MBS market given the uncertainty around the direction of interest rates and other factors such as the demand from banks. U.S. banks are one of the largest purchasers of MBS and their demand is highly correlated to the level of their deposits as MBS are generally seen as relatively secure assets to invest in over shorter time periods. However, our internal modeling suggests that MBS spreads at current levels are attractive relative to other credit markets, particularly investment-grade corporate bonds.

In our view, withdrawing liquidity from the MBS and Treasury markets is more likely to have a greater impact on corporate bonds and other “risk assets” broadly. MBS, in contrast, are still widely seen as “safe-haven” assets with relatively low correlations to corporate bond returns. If riskier assets do see spread widening as the Fed withdraws liquidity, we think investors may find MBS’ approximately 0.9% yield advantage over U.S. Treasuries (while carrying the same credit rating) attractive6. Conversely, if a reduction in the Fed’s purchasing of Treasury securities cause their yields to rise (even if orderly), this would raise mortgage rates commensurately, thereby lowering the speed at which they would prepay – another positive for MBS.

Figure 2: Investment-grade and high-yield corporate bond spreads relative to MBS spreads
Investment-Grade High Yield Corporate Bond spreads relative to MBS spreads

Source: Bloomberg, as of 31 August 2021.

Finally, we should remember that in addition to buying MBS and Treasuries, the Fed bought corporate bonds and – for the first time – signaled its willingness to buy high-yield corporate bond exchange-traded funds. Their overall message was clear: Not only will the Fed provide liquidity by purchasing Treasuries and MBS, but it will also directly support the corporate bond markets. The Fed, in our view, wanted to drive yields down on both Treasuries and MBS to encourage investors to take more risk. And, as if to underscore the point, they would lower the risk implied in corporate bonds by making it clear they would be a buyer of last resort. Insofar as the Fed encouraged investors to leave the relative safety of Treasuries and MBS in 2020, a reversal of that encouragement could see more demand for MBS in 2022.

A trend toward greater intervention?

Central bank policy evolves over time, but the trend has been to provide ever greater support for financial markets. The famous Greenspan “put”7 gave confidence to equity markets that there was some level of drawdown that the Fed would not tolerate. If things got too bad, Greenspan would bail them out. Subsequent central bankers echoed this approach. During the GFC, the Fed extended support to the banking system (resulting in much public criticism) and to financial markets broadly through large-scale purchases of Treasuries and MBS. In time, this kind of “quantitative easing” – a way of providing liquidity to financial markets without lowering interest rates – became acceptable central bank policy.

Today, we don’t believe the Fed will lower its holdings of MBS to zero. Instead, we think both the GFC and the COVID-19 crisis showed the Fed that purchasing MBS was a useful tool for supporting the economy and moderating market volatility in times of stress. At the risk of oversimplification, we believe the Fed sees managing their MBS portfolio as akin to managing policy rates. They can, with relative ease, manage money supply through the mortgage market while also impacting the most significant rate in consumer borrowing.

Is the Fed likely to provide the same support, or even more, in the next crisis? The trend would suggest the answer is yes. But, regardless of how much the Fed’s intervention in bond markets may grow in breadth or depth out of a desire to support markets, we believe it makes sense for them to remain active in both the U.S. Treasury market and the MBS market. The logic is simple: Quantitative easing has proven to be an effective and uncontroversial policy tool, and with interest rates globally uncomfortably near or through zero, the Fed needs all the tools it can get. It may also help that the Fed needs congressional approval to buy corporate bonds but does not to buy Treasuries, or MBS.

Tapering, without the tantrum

The Fed deserves credit for learning from mistakes and being willing to first “talk about talking about” their plans to both raise interest rates and taper their bond purchases. It will still be a change, but with time, change can be understood and absorbed without excessive volatility. Because tapering does not mean the Fed is or will remove all of its support for the mortgage market, MBS spreads have had only to price the uncertainty of how, not the cost of what. As there is now consensus about the how, in our view most of the potential widening in MBS spreads due to tapering has already taken place.

Meanwhile, we expect a rise in demand for MBS from other participants, including U.S. banks and asset managers broadly. In our estimates, many asset managers have favored corporate credit over mortgages (demonstrated in their relative outperformance) in recent quarters, and thus are more likely to add back mortgages to their portfolios as tapering slowly drains liquidity from the bond markets. In our view, “risk assets,” including investment-grade and high-yield corporate bonds, have potentially greater exposure to the risk of wider spreads or lower returns as the tapering process unfolds. And, with interest rates more likely to rise, or at least be stable, than to fall again, we think the wave of prepayments that swept the mortgage market in 2020 has seen its peak for this cycle. Because falling prepayments are generally supportive for MBS, our forecast suggests yet another reason MBS should remain supported through the coming tapering.

1 Bloomberg, as of 31 August 2021.

2 Between 21 February 2020 and 6 March 2020 the U.S. Refinancing Index rose more than 200%.

3 Bloomberg, as of 23 September 2021.

4 As reported by Pension & Investments, 16 June 2021.

5 Bloomberg, as of 31 August 2021.

6 Bloomberg, as of 31 August 2021. Represents the Bloomberg U.S. MBS Index versus the duration-equivalent U.S. Treasury note.

7 The term Greenspan Put is a reference to policies put in place by Federal Reserve Chair Alan Greenspan to help halt excessive stock market declines. It is derived from a put option, which is a kind of product sold in the options market that allows an investor to sell a security at a pre-agreed price. A put option can be used to protect its holder from a fall in a securities value, because they have the right to sell it at the pre-agreed value, regardless how low its price may fall.

These are the views of the author at the time of publication and may differ from the views of other individuals/teams at Janus Henderson Investors. References made to individual securities do not constitute a recommendation to buy, sell or hold any security, investment strategy or market sector, and should not be assumed to be profitable. Janus Henderson Investors, its affiliated advisor, or its employees, may have a position in the securities mentioned.


Past performance does not predict future returns. The value of an investment and the income from it can fall as well as rise and you may not get back the amount originally invested.


The information in this article does not qualify as an investment recommendation.


Marketing Communication.