in

The Fed is not doing QE. Here’s why that matters

Risk assets started this year on a tear. Before the coronavirus outbreak unsettled investors, global equity prices had risen by more than 10 per cent in three months while credit spreads were near record lows.

This left institutional investors wrongfooted, as many were positioned conservatively back in the autumn. To some, the explanation is obvious: no one expected the US Federal Reserve to restart “quantitative easing” a few months ago.

The problem with this explanation is that it is wrong. The Fed is not doing QE. To see why, it is worth revisiting how this policy is supposed to work.

In simple terms, QE can operate through three different channels. The first works by increasing the volume of reserves in the banking system. These are the balances that commercial banks hold at the central bank to settle transactions between themselves. Raising the amount of reserves in the system could encourage banks to increase their exposures to the rest of the economy, boosting the broader money supply.

The second channel works by altering the mix of assets held by investors. For example, if the Fed buys long-term Treasury bonds, it reduces the private sector’s exposure to interest rate risk. If some investors are required to hold these assets for regulatory purposes, a reduced supply should raise their price and lower yields. This in turn could support other asset prices, as a lower discount rate makes future streams of income more valuable today.

Finally, QE might influence investors’ expectations about monetary policy. Buying bonds could make the Fed’s commitment to keeping rates low that much more credible. This forward guidance should suppress interest rate expectations along the yield curve, elevating asset prices further.

The Fed’s attempt to quell problems in the repo market — where investors borrow cash in exchange for collateral such as US government debt — works only through the first of these channels, and even then, only partially. Since the middle of September, the Fed has lent up to $240bn to banks through repo operations and bought roughly $230bn of Treasury bills.

Ordinarily, this would raise the value of reserves in the system by an equivalent amount, $470bn. Yet reserves have increased by just $220bn. The difference is mostly due to the Treasury overfunding the federal deficit, rebuilding the balance on its account at the Fed and withdrawing cash from the commercial banking system. In effect, this has offset 50 per cent of the increase in the Fed’s balance sheet.

More importantly, this is probably the weakest of the three channels outlined above. Reserves rarely constrain banks’ ability to lend. So it should be no surprise that the US banking system has not multiplied up the Fed’s injection of reserves. Instead, banks have increased their assets at a slower pace. Nor are there signs that other investors have exploited the Fed’s calming influence on the repo market to leverage up their positions. Aggregate repo volumes look to have declined since September.

The second channel outlined above does not apply in this case because reserves at the Fed are functionally equivalent to T-bills. Both are risk-free liquid assets, providing a similar rate of return, so the mix of risk in investors’ portfolios is unchanged. The only important difference is that T-bills settle a day later than reserves, meaning banks need to tap the repo market if they need to access intraday liquidity.

The third channel is also moot — the Fed is not using its balance sheet to guide expectations for interest rates. More important were comments from Jay Powell, the Fed chair, in October, when he said it would take a “really significant move up” in inflation for the Fed to raise rates. The best that can be said about the Fed’s actions is that they have headed off a tightening of monetary conditions.

Perhaps we are getting too hung up on the details. A lot of clients have put it to us that if enough investors believe that what the Fed is doing is stimulus, then it is. Maybe so. The trouble with this logic is that it has been crafted after the event. As markets have rallied, investors have searched for a reason why. We struggle to remember anyone arguing that the Fed’s actions would boost risk assets back in September or October. It seems more like a classic case of rationalisation after the fact.

Rather than obsessing about fluctuations in the size of the Fed’s balance sheet, then, investors might be better off focusing on those things that have changed more fundamentally in recent months, and ask themselves the following questions. Will the stabilisation in global economic data result in a more pronounced upturn? Can corporate earnings recover? Have trade tensions been permanently put to bed? What impact will the coronavirus outbreak have?

Attributing recent market moves to the Fed’s actions is alluring. But it could leave investors wrongfooted again when the central bank pares back its interventions later in the year.

The writer is chief economist at Absolute Strategy Research in London


Source: Economy - ft.com

Ethereum Developers Reveal Holdup With Serenity in Reddit AMA

Singapore AML Framework Can Attract Crypto Businesses, Not Chase It Away