As the US Federal Reserve (Fed) contemplates a potential start to monetary policy normalization later this year, I’d like to focus on the evolution of the Fed’s policy toolkit and discuss the various mechanisms available as they attempt to exit a highly accommodative policy stance. In thinking about the potential for higher rates and the long, winding journey to this point in the cycle, I was reminded of a movie released around the time of the last Fed tightening cycle. O Brother, Where Art Thou? follows the meandering adventures of three prison escapees as they set out to recover a buried fortune. A global financial crisis, a protracted period of slow growth, and nearly a decade of easy monetary policy have left many front end investors asking “O Yield, Where Art Thou?” They also should be asking themselves if the Fed’s policy tools are adequate to navigate a clear passage toward higher rates.
June 30, 2004 (3,989 days and counting)
The Fed’s last tightening cycle began at the June 30, 2004 Federal Open Market Committee (FOMC) meeting — almost 11 years ago. At that time, the committee believed that “robust underlying growth in productivity was providing ongoing support to economic activity”. They would go on to raise rates 17 more times in the two years that followed, taking the target fed funds rate from 1.00% to 5.25%. While the notion exists that tightening cycles are common, in the last 30 years there have only been four periods of rising fed funds target rates.
In prior cycles, the Fed has relied solely on its traditional policy tool, setting a fixed target fed funds rate in-line with its objectives of maximum employment and stable prices. The Fed relied on the Open Market Trading Desk to conduct market operations that would adjust the level of reserves in the system. The Fed could move the fed funds effective rate by increasing or decreasing the supply of reserves. However, as early as the 1990s, the Fed began to question the effectiveness of using the level of reserves to target its policy rate.
As the Fed’s balance sheet expanded during and after the financial crisis, their traditional rate setting framework which relied on a baseline level of reserve balances no longer worked. The structural short in reserves evaporated as excess reserves ballooned to $2.4 trillion. The Fed was compelled to revisit prior consideration of a “channel” or corridor system for setting monetary policy. In such a system, the target rate would be set within a corridor established by the Fed, thereby using several rates simultaneously to implement policy. The key advantages of the corridor system were clear — it could help the Fed achieve its target rate independent of the size of its balance sheet.
While the Fed has not yet formally adopted a corridor framework, they began establishing the tools that would allow for the setting of such a corridor in late 2008. As an initial step, the Fed began paying interest on excess reserves (IOER). By paying interest on reserves, the Fed initially sought to establish a lower bound for the target fed funds rate. A lower bound was needed to allow the Fed to continue to expand its balance sheet in pursuit of its dual mandate. This move was not fully effective and divergence in the spread between the IOER rate (set at 0.25%) and the fed funds effective rate led to the introduction of the Reverse Repurchase Agreement (RRP) facility in September of 2013. By incorporating an expanded set of counterparties (most of which are unable to lend at the IOER rate), the RRP has proven more effective in maintaining a floor on front end rates. IOER and RRP represent the two critical pieces of the corridor framework for fed funds. In practice, we have seen a corridor system used very effectively in a number of non-US markets.
After more than a decade of money market investors asking, “O Yield, Where Art Thou?,” recent movements in US interest rates suggest that market participants believe liftoff is imminent, thanks in part to a strengthening economy, but also their confidence in the evolution of the Fed’s policy toolkit.