In brief

• The Federal Reserve (the Fed) will be active in 2022 as policy direction shifts to combat above-target core rates of inflation and full employment that is likely to be achieved much sooner than current Fed expectations.
• We expect the committee to begin raising rates this quarter and start to reduce the size of its balance sheet in the second half of this year. The withdrawal of liquidity will be of concern to markets, but we think ultimately palatable at least through 2024.
• Tighter policy should lead to higher long-term rates, creating a challenging environment for fixed income investors. Investors should continue to remain short duration and embrace credit within portfolios. 

The outcome of the January Federal Open Market Committee meeting affirmed the committee’s hawkish shift telegraphed from its members in recent weeks. With several rate hikes already being priced by the markets this year—we expect four quarterly rate hikes beginning in March—focus is turning towards the Fed’s bloated balance sheet.

The removal of liquidity could be much more disruptive for markets than rate hikes. The Fed’s Treasury and mortgage-backed securities (MBS) holdings have risen by USD 3.0trillion and USD 1.3trillion to USD 5.7trillion and USD 2.7trillion, respectively. This is more stimulus than the previous three rounds of quantitative easing combined. The Fed faces the complicated task of communicating and removing this accommodation.

In this note, we outline our expectations for what the shift from quantitative easing (QE) to quantitative tightening (QT) will look like and implications for markets.

The greater risk is to not tighten

The economic conditions for tighter policy in the U.S. are very clear. Inflation is at multi-decade highs and the unemployment rate has declined faster than anticipated. We expect the rate of inflation to fall as supply chain disruptions are resolved and the spike in prices in certain reopening categories of spending fade. However, rising wages and shelter costs are likely to keep the core rate of inflation above the Fed’s target. Meanwhile, the strength in employment growth in the U.S. and the slow recovery in participation by workers should see the unemployment rate continue its downward trend. Outside of an exogenous growth shock, there is little that would prevent the Fed from raising interest rates at its March meeting, just days after its QE program officially ends.

From QE to QT

The different inflation and labor market dynamics in this cycle compared to the last have increased speculation of a truncated period between the first rate hike and the run-off in the Fed’s balance sheet. In the 2015 hiking cycle, there was a two-year gap between the first rate hike and when the Fed began to run down its balance sheet. This time, that period could be as short as four months. We think it’s possible a formal announcement regarding the balance sheet could come as early as June, with implementation commencing in July.

The previous tightening cycle should not serve as an exact template for this one, but we expect that the Fed will adopt a similar approach at an accelerated pace. This means a passive run-off in Treasuries and MBS as they mature without reinvestment. Given the lumpy profile of Treasury and MBS maturities, the Fed will likely apply a cap on the value of assets each month to smooth the process, and any amount in excess of the cap is reinvested.

Achieving a faster reduction suggests higher caps that increase over time. A plausible scenario could see the Fed gradually increase the cap of maturing securities by USD 25billion per meeting beginning in July (USD 15billion in Treasuries, USD 10billion in MBS). This implies a max cap of USD 100billion in maturing securities per month beginning in 2023, double the pace from 2018. At this rate, USD 1.37trillion in Treasuries and USD 770billion in MBS would be removed from the balance sheet by 2024 (Exhibit 1).

A steeper yield curve

A peculiarity of markets over the opening stages of this economic cycle is why long-dated bond yields are not higher and the yield curve steeper. The Fed seems intent on pushing up longer-term interest rates and has the flexibility to adjust the balance sheet unwinding to achieve this. The short end of the yield curve has risen with rate expectations, but movements in the longer end of the curve have been more muted, leading to a flatter curve. We don’t believe that the Fed would actively sell bonds to push up longer-dated bond yields, but a shrinking balance sheet should see longer-dated bond yields rise. 

Exhibit 1: Fed balance sheet reinvestment and runoff based on increasing monthly cap

Source: J.P. Morgan Asset Management; (Left) J.P. Morgan Securities; (Right) Bloomberg.
Data reflect most recently available as of 24/01/22.

Isolating the impact of QT on the bond market is challenging given the variety of factors influencing long-term rates. However, it is notable that from October 2017 to November 2018, the Fed reduced its then USD 4.5trillion balance sheet by approximately USD 700billion and the nominal 10-year UST yield rose from 2.3% to 3.2%. 

Honey, I shrunk the balance sheet

QT should push long-term rates higher, however, the key question for investors is how much can the balance sheet shrink before it impacts risk assets and broader financial conditions?

The size of the Fed’s balance sheet is dictated not by its assets but its liabilities. The largest liabilities for the Fed are currency in circulation, the Treasury General account (TGA), excess reserves and the reverse repo facility.

• The cash in circulation creates a floor on the size of the balance sheet (the Fed cannot remove hard dollars from the system). There is USD 2.2trillion in circulation.
• The TGA acts as the U.S. Treasury’s checking account at the Federal Reserve, it currently sits at USD 490billion, but is likely to rise to USD 650 to 700billion in the coming months and maintain that level.
• Commercial banks have to hold a certain level of reserves to meet regulatory requirements and often hold more than required. Total reserves are currently USD 3.9trillion. We expect reserve balances to fall, but too low banking reserves could result in overly tighten financial conditions, something the Fed will wish to avoid. 
• The reverse repo facility is another USD 1.9trillion but could fall back to a historical average level of roughly USD 250billion. The reverse repo system is used when there is excess liquidity and not enough collateral in the system. As liquidity is drained through QT, this should slow AUM growth in money market funds, which have been the primary users of the Fed’s reverse repo facility.

Overall, the Fed will need to operate with a larger balance sheet to meet its liabilities so is unlikely to return to pre-pandemic levels. The Fed could potentially reduce its balance sheet by up to USD 3.0trillion in the coming years, however, a smaller decline closer to USD 2trillion is more likely.

Exhibit 2: Balance sheet could fall by USD 3trillion at most
Source: Federal Reserve Board, J.P. Morgan Asset Management. Currency in circulation projection is based on the trend growth rate from December 2015 to today. The reserve balance projection reflects the assumption that the interbank market could operate smoothly with a minimum of USD 2.0trillion in reserves. Current Fed liabilities are based on the average level for the week ending 19/01/22. Data reflect most recently available as of 25/01/22.

Fixing the plumbing

The 2017 balance sheet unwind highlighted the complexities of the U.S. financial plumbing and the reduction in liquidity that caused spikes in repo rates. There may be a concern of a repeat scarcity in liquidity. However, the Fed has introduced a Standing Repo Facility, which can provide as much as USD 500billion of cash overnight to the banking system and serve as a backstop to ensure adequate liquidity. While this facility may not be needed, its existence may be reassuring to markets similar to the lending facilities which were made available in the early days of the COVID-19 pandemic.

De facto rate hikes

With a QT sample of just one from 2017, it is difficult to judge whether a two-year time frame from rate hike to asset unwind is the right length. However, the potential for an accelerated tightening in short-term rates this time around and the growing role of the balance sheet in monetary policy points towards the  “policy substitution effect”.

The Kansas City Federal Reserve has estimated that every USD 675billion reduction in the balance sheet is equivalent to a 25bps rate hike¹. This suggests that a USD 2trillion reduction in the balance sheet over the next two years is the equivalent to 75bps of policy tightening. This de facto tightening may explain some, but not all, of the market view that the official policy rate in the U.S. is not expected rise much beyond 1.5% in the coming years.

Investment implications

Nothing is guaranteed in financial markets, but the previous experience of balance sheet reduction at least provides the Fed and market participants with some knowledge of what to expect. However, the faster pace of unwind, the larger size of the balance sheet as well as prevailing economic conditions all have the potential to create volatility in markets as the Fed rotates relatively quickly from QE to QT.

Investors may wish to consider the following:

• Underweight agency MBS: The Fed is shifting from taking down 65% of MBS supply in 2021 to just 12% this year. Accelerated taper and subsequent tightening will leave large amount of issuance for private markets to absorb, putting widening pressure on spreads.
• Overweight credit: Still robust earnings growth this year supports the outlook for credit even as Treasury yields rise. With defaults at multi-decade lows, high yield may be preferred over investment grade given the shorter duration profile and relative yield attractiveness. However, tight spreads implies lower returns from carry over capital.
• Shorten portfolio duration relative to benchmark: Balance sheet unwind points to increasing longer-dated bond yields. With this backdrop, there is little to be gained from holding duration beyond its role as a hedge against an economic shock to growth.
• Steeper curves: The potential for a steeper yield curve driven by rising real yields supports the continued outperformance of value over growth sectors in the equity market. However, as the U.S. economy moves more firmly into mid-cycle and valuations on growth stocks fall, investors will be reminded of the secular themes that have driven the longer-term outperformance in growth.

One final note, with the President’s nomination of Sarah Raskin as Vice Chair of Supervision and Lisa Cook and Philip Jefferson to serve as Governors, U.S. President Joe Biden has nominated five of the seven Board seats. While Democrats continue to try and push further fiscal initiatives through reconciliation, some might suspect these new nominees—if confirmed in the Senate—could shift policy in a more dovish direction in the years ahead. However, even the resident doves on the committee have had to concede that rising inflation and a falling unemployment rate warrant a removal of accommodative policy. Therefore, the appointment of three perceived policy doves should not materially alter the path of policy tightening over the next couple of years. 

¹ Source: Forecasting the Stance of Monetary Policy under Balance Sheet Adjustments; A. Lee and Troy Davig, Federal Reserve Bank of Kansas City

09a2222701023759