Quantitative Tightening Is About to Ramp Up. What It Means for Markets.

Now, with inflation rampant, the Fed is unwinding this liquidity via a process known as quantitative tightening, or QT. In June, the central bank started to shrink its portfolio by letting up to $30 billion of Treasuries and $17.5 billion of mortgage-backed securities, or MBS, roll off its balance sheet, or mature without reinvesting the proceeds. The amount will double this month and effectively kicks in Sept. 15, as Treasuries are redeemed midmonth and at the end of the month.

QT is as ambitious as its impact is uncertain. At full-throttle, the pace of balance-sheet tightening will be much more aggressive than in the past, and come at a time when interest rates are rising quickly. What could go wrong? Potentially, a lot, suggests Joseph Wang, a former trader on the Fed’s open-market desk and author of the Fed Guy blog and Central Banking 101. Wang explains what’s at stake in the edited conversation that follows.

Barron’s: How will quantitative tightening unfold, and how will accelerated redemptions affect the market?

Joseph Wang: When the economy wasn’t doing well, QE put downward pressure on interest rates and increased liquidity in the financial system. Now the Fed wants to tighten financial conditions. QT increases the amount of Treasuries available to investors while also reducing their cash holdings. Mechanically, the U.S. Treasury issues new debt to an investor and uses the issuance proceeds to repay the Treasuries held by the Fed. The Fed receives that cash and then simply cancels it—the opposite of what happened during QE, when the Fed created cash out of thin air.

When you’re increasing the supply of bonds into a market that isn’t very liquid, and when the marginal buyer is changing as the Fed steps back, you’re going to get volatility. Markets haven’t priced in just what that means. We will likely see higher fixed-income yields. Higher yields affect equities in a few ways. There is the portfolio-rebalancing impact, whereby losses on the bond side of a portfolio would prompt an investor to sell some equities to rebalance. QT also reverses the risk-on effect of QE, which occurred when many investors looking for yield moved into riskier assets or longer-dated Treasury bonds.

This is happening at a time when Treasury issuance is high. Why does that matter?

Market pricing is determined by supply and demand, and in the coming years, there is going to be a tremendous supply of Treasuries coming from two sources. One, there are the budget deficits the U.S. government is running. While the deficit will shrink a little this year compared to last, the Congressional Budget Office says the trajectory is basically a trillion dollars a year of Treasury issuance for the foreseeable future. The second source of additional supply is QT. Together, these will increase the supply of Treasuries to historically high levels of around $1.5 trillion this year and next. Before Covid, net supply was about $500 billion.

On the demand side, the marginal buyer is changing as the Fed extricates itself from the Treasury and mortgage markets. The hedge funds aren’t there. The Fed isn’t there. And the banks aren’t there. We won’t have to go through a phase of price discovery. Keep in mind the context: Treasury-market liquidity is weak right now. There is some fragility, and it will likely be stressed as QT ramps up.

Speaking of the marginal buyer, who will fill the void as the Fed steps back? Can these markets function without the Fed?

I’m not sure who the new buyer will be, which is why I think there could be significant volatility in interest rates. But new buyers can be manufactured through policy. One way is through a Treasury buyback program, where the Treasury becomes a large buyer of Treasuries. The Treasury department recently floated this idea. Another way that new buying by banks could be encouraged is through regulation changes, wherein regulators reduce the capital requirements of banks, thus encouraging them to buy more government debt.

But the point is that if issuance is growing by a trillion dollars a year, it’s hard to say that there will ever be enough marginal buyers. We are locked in a world where there will always be QE, because the Fed will have to ultimately become the buyer again. The growth in Treasury issuance is faster than the market can handle by itself.

Consider that over the past 20 years, the amount of Treasuries outstanding has more than tripled, but the average daily volume in the cash market has grown far slower. That is inherently unstable. It’s like a stadium that keeps getting larger while the number of exits remain the same. When a lot of people need to get out, as happened in March 2020, then the market has issues.

Fed officials say they don’t know much about how QT will play out. Why is that?

The way that QT plays out will depend on moving parts, and a lot of it is beyond the Fed’s control. First, there is the uncertainty around what the Treasury issues. It could issue lots of longer-dated Treasuries, which the market will have more difficulty digesting, or more shorter-dated Treasury bills, which the market can more easily digest. Depending on what the Treasury is doing, the market may have to digest a lot more duration, which would be disruptive in a Treasury market where liquidity is already thin.

Where liquidity comes out is also beyond the Fed’s control. When the Treasury issues new securities, they can either be purchased by cash investors, like banks, or levered investors, like hedge funds. When they’re purchased by levered investors, the money that goes to fund them most likely comes out of the Fed’s reverse repo facility, or RRP, an overnight lending program that you can think of as excess liquidity in the financial system.

If the newly issued Treasuries are purchased by levered investors, that results in draining liquidity that the financial system doesn’t really need, so the impact is neutral. But if the newly issued securities are purchased by cash investors, someone is taking money out of the bank and using it to purchase Treasuries to repay the Fed. In that case the banking sector loses liquidity, which can be disruptive, because it’s possible that someone, somewhere is dependent on that liquidity. That is what happened in September 2019 when the repo market seized up and the Fed had to add more reserves.

It seems you are worried that something will break again. Why?

It’s impossible for the Fed to know how the liquidity will be drained out of the financial system. But we can look at who is buying today, and buying is almost all coming from the banking system, as opposed to parties such as hedge funds. The RRP has been steady at around $2 trillion since the beginning of the year. So it seems like QT is going to be draining liquidity out of the banking sector rather than the RRP.

That is the opposite of what the Fed wants. Officials have assumed they could ramp up the pace of QT aggressively, because they see a lot of liquidity parked in the RRP. What they may not understand is that how liquidity gets drained is beyond their control. And at the moment, as noted, it is coming out of the banking system.

Fed Chairman Jerome Powell said in July that QT would last between two and 2½ years. That suggests the Fed’s balance sheet will shrink by about $2.5 trillion. Is that a realistic assumption?

The Fed thinks of QT as being limited by how much liquidity the banks need to operate well. They feel that the balance sheet could drop by around $2.5 trillion, and it would be fine. But remember, the Fed doesn’t have a lot of control over how the liquidity is drained. It seems they want the banking sector to have above $2 trillion in reserves. Right now, the banking sector has about $3 trillion. The only way QT can proceed as currently forecast is to ensure that the liquidity is drawn more evenly out of the financial system—meaning more liquidity coming out of the RPP versus the banking sector. If the Fed can’t find a way to achieve that balance, it might have to stop early. But there are ways that they can make this work.

What are they?

There are two primary solutions to the problem of too much liquidity draining from the banking sector while a lot remains in the RRP. First, the Fed can do what it did in the fall of 2019 and start buying a lot of Treasury bills. From the Fed’s perspective, purchasing bills isn’t the same as QE. They are basically exchanging short-dated assets for reserves, which are also short-dated assets, intentionally adding reserves into the system without affecting interest rates.

Second, and more likely, the Fed could work with the Treasury. If the Treasury conducts buybacks by issuing shorter-dated bills and uses the proceeds to purchase longer-dated coupons, it would move liquidity out of the RRP and into the banking system because the Treasury bills would be purchased by money-market funds with money held in the RRP. The sellers of the coupons to the Treasury would then be depositing the funds in a commercial bank. Moving money from the RRP to the banking system would allow the Fed to proceed with QT without having to worry about liquidity in the banking system falling by too much.

Will the Fed wind up selling mortgage-backed securities?

Its plan is to roll off a maximum of $35 billion a month in agency mortgages, but it estimates that it will only be able to do about $25 billion a month. Unlike with Treasuries, whose principal all gets paid on the maturity date, mortgage loans can be prepaid. For example, if someone who owns a home refinances, they take out a new loan to repay an old loan. In selling a house, they might using the proceeds to repay the mortgage. That all slows as mortgage rates rise.

Selling MBS is another tool the Fed has to tighten financial conditions, but it doesn’t seem like they want to deploy it. In the past few months, the housing market has softened significantly. I would imagine they want to see how this plays out before further tightening financial conditions in housing.

Thanks, Joseph.

Write to Lisa Beilfuss at [email protected]