Monetary Policy & Inflation | US
The Fed Balance Sheet: What Happens in May 2022?
- By Dominique Dwor-Frecaut
- 9 min read
Summary
- The Fed will likely shrink its securities portfolio starting in May. This is called quantitative tightening.
- Quantitative easing and quantitative tightening work at least as much through money creation as through the ‘portfolio balance effect’, which raises the price of risky assets.
- The Fed intends to rely mainly on reinvestment caps. That is, only reinvesting a limited share of maturing securities in its portfolio.
- Slower money creation is likely to be negative for asset prices and could also help control inflation.
Quantitative Tightening and the Fed’s Portfolio Balance Effect
Recent Fed communications suggest it will announce the start of quantitative tightening (or securities portfolio shrinkage) at next week’s FOMC meeting. In this explainer, I build on my high-level description of quantitative tightening from January. I go deeper into its mechanics and market implications.
What Is Quantitative Tightening?
Quantitative tightening is when a central bank (like the Federal Reserve) reduces the size of its balance sheet, typically by not reinvesting when bonds mature. It is a contractionary monetary policy that aims to decrease liquidity in the economy, normalise interest rates, and control inflation.
Quantitative tightening is the reverse of quantitative easing. Therefore, I will start by explaining the mechanics and impact of quantitative easing – or in Fed jargon, large-scale asset purchases (LSAP).
Enjoying this Explainer? Sign up to our free newsletter for more.
The Portfolio Balance Effect
The traditional explanation of the impact of quantitative tightening or easing is through the portfolio balance effect. When the Fed buys securities, it reduces the quantity of Treasury bonds held by the public, which lowers bond yields. The Fed believes this induces investors to seek riskier assets with higher returns such as corporate bonds and equity. In turn, this raises the price of riskier assets, lowers borrowing costs and raises aggregate wealth in the economy.
We cannot directly measure if the portfolio balance effect works as the Fed describes. Instead, we require models. As a result, a voluminous academic literature has developed, with central banks’ research departments usually finding evidence of significant and positive impact of quantitative easing on the economy.
The Problem With the Fed’s Theory
In reality, no obvious correlation exists between bond yields and Fed asset purchases (Chart 1). The Fed and other central banks practicing quantitative easing would likely argue that, everything else equal (economists’ most magic formula), quantitative easing has made bond yields lower. Disputing that is hard. Yet we cannot directly measure by how much lower. And whether lower bond yields benefit ‘Main Street’ rather than ‘Wall Street’ is not obvious either.
Much less attention has been paid to the impact of quantitative easing on money creation. Yet that could be as powerful as the portfolio balance effect, as we now explore.
Quantitative Easing Works Through Money Creation
Table 1 shows stylized balance sheets for the Fed, the US Treasury, banks and the private sector excluding banks.
Table 2 shows two quantitative easing scenarios. In Scenario 1, the Fed buys securities from the banking sector. In Scenario 2, the Fed buys the securities from the private sector outside of banks.
- In either scenario, the Fed pays for the securities by issuing short-term debt, i.e., reserves that are banks’ deposits at the Fed.
- Scenario 1: the Fed buys from banks.
- Banks’ holdings of Treasury securities fall, and their reserves increase by an equivalent amount.
- Banks’ total assets and liabilities are unchanged. The only change is the composition of their assets.
- Scenario 2: the Fed buys from the nonbank private sector.
- The nonbank private sector holdings of Treasuries falls, and their holdings of deposits rises.
- Banks’ reserves and deposits rise by an equivalent amount because the Fed pays for its purchases by crediting the account of the sellers’ bank at the Fed.
- Bank deposits get created by an amount equivalent to the Fed asset purchase.
Note that the Treasury balance sheet does not appear in my story. The implicit assumption here is that Treasury issuance, i.e., fiscal policy, is not influenced by quantitative easing. It is unrealistic. In FY2020-21, the US ran its largest peacetime budget deficits ever, 15% and 12% of GDP. I do not think this would have happened without the administration and Congress counting on large Fed buying of Treasuries.
Who does the Fed buy securities from in reality? Mainly from nonbanks. That is, Scenario 2 is much more prevalent than Scenario 1 (Chart 2). Quantitative easing therefore tends to increase the quantity of deposits in the economy.
Why should we care about deposit creation? It is because deposits are very liquid assets that can be used to purchase goods and services, or financial assets. That is, deposits form most of what is called broad money or M2 (Table 3). Note that the increase in deposits during 2020-22 is close to the Fed securities purchases of 17 ppt of GDP.
Quantitative easing is not the only driver of deposit growth. Other factors such as the demand for credit or banks’ ability to grow their balance sheet also play an important role. In the aftermath of the Global Financial Crisis, bank and household balance sheet repairs led to relatively low deposit growth despite large asset purchases by the Fed (Chart 3).
In addition, as discussed in my inflation explainer, an increase in the quantity of money does not always lead to CPI inflation. And when it does not lead to CPI inflation, it tends to be supportive of financial asset prices. This is because quantitative easing gets implemented when the Fed policy rate (the Feds Fund Rate) has hit zero and cannot get lower. So the abundance of deposits is accompanied by very low interest rates, incentivising investors to move into higher-yielding assets.
Over the past year, however, inflation has picked up. And it is likely that the high level of deposits, i.e., broad money, has played an enabling role. Let us turn to how the Fed will shrink its securities portfolio.
Shrinking the Fed Securities Portfolio in 2022
During 2020-22, the Fed purchased the equivalent of 17ppt of GDP in securities (Table 4). In January, the Fed stated it would reduce its securities holdings by ‘adjusting the amounts reinvested of principal payments received from securities held in the System Open Market Account’ (SOMA, another fancy acronym that basically means the 12 regional Federal Reserve Banks’ holdings of securities used in monetary policy implementation). As we will see, that strategy is not very different from outright securities sales.
In the March minutes, the Fed further announced that:
- Monthly reinvestment caps would be $60bn for Treasuries, $35bn for mortgage-backed securities (MBS).
- The cap would be phased in over three months ‘or modestly longer if market conditions warrant.’
- When Treasuries principal repayments fall below the $60bn cap, the difference with the cap is to be made up by not reinvesting TBills.
- Because MBS redemptions are likely to remain well below the cap, and because in the long term the Fed wants to hold only Treasury securities, the Fed intends to proceed with outright sales of MBS ‘after balance sheet runoff is well under way’.
A Quantitative Tightening Scenario for the Fed in May 2022
Table 5 shows a quantitative tightening scenario. I have assumed that caps increase by $20bn/month and $10/month, then $15bn for Treasury securities and MBS respectively.
Treasuries
- In May, June and July, the amount maturing (redemptions) is above the cap and the amount reinvested equals the difference between caps and redemptions. All maturing TBills are reinvested.
- In July, the amount maturing is below the cap. There is therefore no reinvestment. In addition, the Fed does not reinvest the portion of the TBill maturing equal to the difference between the coupons cap and redemptions or $18bn.
MBS
- The amounts expected to mature are negligible during May-August. This is because MBS are securities collateralized by a pool of residential mortgages, where the typical maturity is 30 years and therefore has long expected maturity.
- In reality, because US borrowers have the option to prepay their mortgage, the effective maturity of mortgages and therefore MBS has tended to be much shorter.
- However, mortgage prepayments are likely to be much less important going forward. Mortgage rates are historically high and rising than during the past four decades of falling rates.
- This is why the Fed is planning to sell its MBS rather than letting them roll off the balance sheet.
Quantitative Tightening Will Reduce the Growth of Deposits
The balance sheet changes we discussed above will play in reverse with quantitative tightening (Table 6).
Step 1
When Treasury securities held by the Fed mature, the Fed securities holdings fall and the TGA (Treasury deposits at the Fed) falls by an equivalent amount. The same happens on the Treasury balance sheet.
Step 2
The Treasury issues new debt that the nonbank sector buys.
- The Treasury assets and liabilities increase by an equivalent amount as the Treasury deposits the proceeds of its debt issuance at the Fed.
- The nonbank sector increases its Treasury holdings and decreases its bank deposits by an equivalent amount.
- Banks pay the Treasury on behalf of the buyers by transferring reserves to the TGA and debiting the accounts of the buyers: banks reserves and deposits decline by an equivalent amount.
- The Fed total assets and liabilities are unchanged, but the composition of liabilities has changed: reserves have fallen, and the TGA has increased by an equivalent amount.
- Note that if the Treasury had sold its debt to banks instead, there would have been no change in banks’ total assets/liabilities (or those of the nonbank sector). Instead, the composition of banks assets would have changed with Treasuries increasing and reserves decreasing by the same amount.
Step 1 and 2 together reduce the size of the Fed balance sheet and that of banks.
How different would outright sales of Treasuries be? Not that different. The Fed would lower its holdings of securities and reserves by an equivalent amount. The buyer would pay for the Treasuries by running down her deposits. If the buyer was not a bank, bank deposits and reserves would fall by an equivalent amount. If the buyer was a bank, there would be no change in the size of banks’ balance sheet but a change in the composition of their assets.
So quantitative tightening is likely to slow the growth of deposits and possibly the demand for goods and services as well as for financial assets. However, as explained above, quantitative easing is not the only driver of money growth. For instance, interest rates on residential mortgages and consumer durables are negative in real terms, which is supportive of credit creation and money growth – a topic for another explainer.