Finally, the Fed has now adopted an operating regime for monetary policy commonly called a floor system. This system has several features, but a key element is that the Fed’s balance sheet is no longer tied to monetary policy.4 The primary instrument of monetary policy in a floor system is an administered rate, the interest paid on reserves (both excess and required). The balance sheet or the volume of reserves is not crucial to the setting of monetary policy as long as demand is satiated at the interest rate paid on reserves.
The rationale offered by the Fed for maintaining a large balance sheet, or as they refer to it, one with “ample reserves,” has focused on (1) simplicity and improved efficiency in the control of short‐term interest rates, including a reduced necessity of actively managing the supply of reserves through interventions, and (2) improvements in financial stability and the increase in demand for reserves induced, in part, by new regulations such as the LCR (liquidity coverage ratio). My view is that this new operating regime is not as simple or as beneficial as advertised and political economy concerns pose risks to the Fed’s independence. More specifically, I argue that:
• Operational benefits are dubious;
• The role of the balance sheet in conducting monetary policy remains unclear;
• Important governance questions surrounding the operation of the system remain unresolved; and
• Political economy issues that put at risk Fed independence loom large for the institution.
So, let me briefly touch on each of these concerns.
The Fed has argued that the new system is simple and successful. This conclusion is far from obvious. From my perspective, it looks like a jerry‐rigged apparatus held together with belts and suspenders. The floor system has a lot of moving parts. As currently implemented,
• It specifies a target band for the federal funds (ff) rate determined by the FOMC;
• It requires the Board of Governors (BoG), not the FOMC, to set the interest on reserves (IOR);
• It has required the development of a reverse repo program (RRP) to support the lower end of the ff band, as the IOR has proved to be a leaky floor, and it still requires interventions using standard repos to manage the balance sheet to ensure the upper end of the ff is maintained;
• Moreover, the Fed is pursuing new regulations to prevent a private bank from competing with the RRP by forcing it to accept a lower IOR than other banks.
Another argument for the floor or ample reserves regime is that maintaining a buffer of highly liquid reserves enhances financial stability. What evidence is there to support such a claim? Is the balance sheet now a tool of financial stability rather than monetary policy?
The creation of reserves involves the Fed purchasing Treasury securities in the open market and creating bank reserves. Why does substituting reserves for Treasuries increase liquidity or financial stability? Both are highly liquid and qualify in meeting the LCR. Moreover, only banks can hold reserves while Treasuries can be held by a broader range of financial institutions and the public at large. How does the substitution of one highly safe and liquid asset held by the public for another, which can only be held by banks, improve stability? Of course, if the Fed chose to finance the reserve creation by purchasing more risky assets, the action comes with additional costs and risks born by the taxpayer and ultimately constitutes a form of fiscal policy or credit allocation by the central bank.
So, what about interest rate control? The Fed has argued that the floor system leads to better control of the short‐term rate. Yet, the volatility of short‐term rates during the precrisis regime never seemed to interfere with the implementation of monetary policy or the transmission of the funds rate target to other short‐term rates. Interestingly, the old system worked well when total reserves were about $40 billion and fluctuated by relatively small amounts. Recent turmoil in the repo market has led many to conclude that the current system, with reserves of about $1.6 trillion (about 4,000 percent more reserves than precrisis), falls short of the amount necessary to stabilize short‐term rates.
Granted the regulatory environment has changed. The LCR, for example, has increased the demand for reserves, and supervisory pressure seems to have biased banks into holding reserves rather than Treasuries. The Treasury has changed the way it manages its cash balances, relying more on deposits at the Fed than private banks, thus altering the need for reserves by the depository institutions. Changes in the way LCR is computed and shifts in government balances has resulted in more uncertainty and greater volatility in the demand for reserves than in the precrisis regime. This seems to have led to the Fed intervening in a massive way to absorb shocks. Ironically, this is exactly the sort of action the Fed had been hoping to avoid by adopting the floor system.
Another operational consequence of the expansion of reserves is that the interbank market for reserves, or fed funds market, has been largely eviscerated. The system is awash in reserves without much incentive for day‐to‐day trading. Without an efficiently operating funds market, reserves are not moving to where they are needed, trading volume is quite small, a few large banks hold the great majority of reserves, and, even on a day when reserves were seemingly scarce, such as during the repo turmoil in September 2019, reserves were not being traded and consequently not helping address the bottlenecks that were arising. Thus, the undermining of the funds market would appear to have had unintended consequences. The Fed seems to have become the de facto market‐maker for reserves rather than the banks.