“You can be sure of succeeding in your attacks if you only attack places which are undefended. You can ensure the safety of your defense if you only hold positions that cannot be attacked.” – Sun Tzu, The Art of War
“I have absolutely no doubt that when the time comes to reduce the size of the balance sheet we’ll find that a whole lot easier than we did when expanding it” Sir Mervyn King, February 2012
In September, 2017 the Fed announced that it would start paring back its multitrillion-dollar balance sheet. The reduction will start modestly with USD 6 billion in Treasury bonds and USD 4 billion in mortgage backed securities a month. By the end of 2018 the pace of reduction is expected to reach USD 50 billion a month.
Ever since the Fed indicated its intention to gradually begin selling some of its bonds portfolio, opinions on the possible consequences of the balance sheet reduction have been divided. The equity bears argue that quantitative tightening will lead to a severe tightening of monetary conditions causing a recession and a stock market crash. The bond bears on the other hand argue that the added supply of Treasury bonds, at a time when foreign central banks are retreating from the Treasury market, will overwhelm the market, causing bond prices to fall sharply. While there are others that curiously argue that just as quantitative easing was expected to be inflationary but ended up being deflationary, quantitative tightening, which is expected to be deflationary, will cause inflation to spike.
We wonder, however, if quantitative tightening will turn out to be a non-event?
Prior to the global financial crisis, the Fed through the Federal Open Market Committee (FOMC) used open market operations – the buying and selling of government securities – as its primary tool to regulate money supply in the economy. The open market desk at the Federal Reserve Bank of New York would buy securities to increase the level of reserves held by banks – increasing money supply – or sell securities to remove reserves – to reduce money supply. There were also no interest payments on excess reserves.
As banks, in the ordinary course of business, try to maximise income on the given level of available funds, the opportunity cost for holding excess reserves prior to the crisis was high. Instead of holding excess reserves, banks, by lending or investing their assets, attempted to maximise earnings by reducing liquidity to as close to the statutory minimum as possible. According to the Federal Reserve Bank of Cleveland, from 1959 to just prior to the financial crisis, the level of reserves in the banking system remained stable, growing at an annual average of 3.0 percent over the period – in line with the growth rate of deposits. Excess reserves’ share of total reserves, outside periods of extreme uncertainty or economic stress, was also stable, rarely exceeding 5.0 percent.
Under the circumstances prior to the financial crisis, banks did not hold surplus funds needed to buy bonds from the Fed. Their only option to generate funds was to sell other assets or call in loans. So when the Fed sold bonds to the banks, the banks sold other financial instruments, driving down the price of financial assets and pushing up interest rates. This selling tightened monetary conditions and served to cool economic activity.
In response to the global financial crisis, the Fed injected large amounts of reserves into the banking system and introduced new mechanisms that encouraged banks to increase their level of excess reserves. Since December 2008, the Fed pays interest on all banking reserves, including excess reserves, thereby increasing the marginal benefit of holding excess reserves as compared to before the crisis. At the same time, the heightened levels of risk in other securities and lending combined with an increasing regulatory burden reduced the marginal benefit of the alternatives to parking money at the Fed. The decision to pay interest on banks’ excess reserves also ensured that the mass injection of liquidity did not result in short-term rates falling below zero, thereby putting a floor on short-term rates.
Interest on Excess Reserves vs. Fed Funds Rate and 90 Day T-Bill Discount Rate
Source: Bloomberg
A number of liquidity-easing programs to alleviate some of the stress in the financial system were also implemented by the Fed. The largest of the liquidity programs implemented, quantitative easing, involved the Fed purchasing Treasury securities, federal agency debt and mortgage-backed securities primarily from non-banks. These purchased assets were then converted into deposit liabilities at the banks. As a consequence of these asset purchases and their conversion to deposit liabilities, excess reserve balances at the Fed expanded greatly and as of 25 October, 2017 stood at USD 2.14 trillion.
Excess Reserves of US Depository Institutions (USD million)
Source: Federal Reserve Bank of St. Louis
To undertake quantitative tightening, the quantitative easing process will have to be reversed. This will involve the Fed selling the securities it purchased to banks in order to absorb the excess funds in their current accounts. As bond bears postulate, this selling of government bonds should, in theory, cause their price to fall, driving up interest rates. In practice, however, if the Fed does not shrink its balance sheet by more than the level of excess reserves in the banking system, quantitative tightening will not be a “tightening” of monetary policy. In a tightening phase, the Fed would be selling bonds to banks to soak up market liquidity and reduce the volume of money circulating. However, for any level of quantitative tightening up to the USD 2.14 trillion in excess reserves, the funds are already available in the banks’ accounts with the Fed. As banks will not need to raise funds elsewhere, the operation, in our opinion, is unlikely to have the negative impact of a standard tightening operation, and interest rates should not rise as a direct consequence of quantitative tightening.
The next question that comes to mind then is why undertake quantitative tightening now? The answer to that, we think, lies in the very reason many feared quantitative easing would cause inflation to spike. Quantitative easing did not result in high levels of inflation as there was little demand for debt financing, other than for corporate share buybacks and mergers & acquisitions. As long as there is no demand from would be borrowers, no amount of quantitative easing will result in inflation. However, as we argued in The Case for a Pickup in US Inflation, the capital expenditure cycle may be picking up and that may spur an expansion in credit issuance.
Of the aggregate reserves of depository institutions held with the Fed, only around USD 127 billion, as of 25 October, 2017, are required to satisfy reserve balance requirements. Accordingly, the USD 2.14 trillion in additional reserves can potentially support a level of money supply much larger than that exists today. By absorbing these excess reserves, the Fed will reduce the possibility of a drastic expansion in bank lending and money supply that could destabilise the economy.
Investment Perspective
Much like Pavlov’s dog, we, as market participants, have been conditioned by the pairing of a neurologically potent stimulus, such as directional market action, with a neutral stimulus, such as the Fed’s policy decisions, to elicit a response. The nature of our response is ultimately conditioned by our most formative experiences within markets. The direction of market action that was most recurrent subsequent to the application of a particular Fed policy during this time is most likely to become our default expectation whenever the policy is implemented again.
The market regime experienced by the vast majority of market participants today is the one where banks’ excess reserve levels were modest and the Fed selling bonds resulted in interest rates rising. Quantitative tightening involves selling of bonds by the Fed, ipso facto, financial conditions will tighten and interest rates will rise so sell bonds. We consider this to be a systematic error caused by conditioning under an altogether different market regime.
Quantitative easing was a compelling reason to buy long-dated government bonds. Quantitative tightening, on the other hand, is not a sound reason to be selling or shorting long-dated Treasury securities.