The microeconomic perils of monetary policy experiments.

• Author: Calomiris, Charles W.

The Federal Reserve (Fed) tells us that its experiments in quantitative easing (QE) are stimulating the economy, and financial news services obediently echo that message. But there is reason to believe that, under current unusual market conditions--especially near-zero interest rates and tightening prudential regulatory requirements--Fed actions may be having little effect, or even effects opposite to those the Fed intends. Because of changes in how the tools of the Fed work under the current unusual circumstances, raising interest rates and shrinking the Fed's balance sheet have little effect, or even a positive effect, on economic activity. That is especially true when one adopts an appropriate medium-term perspective on monetary policy and takes into account the benefits of avoiding the destabilizing potential consequences for asset markets of the Fed's current mortgage-backed securities (MBS) purchases.

The argument for raising interest rates at this time, however, is not mainly one about incremental accommodating or tightening; rather, it is about restoring predictability to monetary policy by reviving the federal funds market. It is essential to return to a situation where (based on decades of empirical evidence) die Fed can use changes in the federal funds rate as its policy tool and confidently project whether its incremental policy actions are stimulative or contractionary. We should not continue under the current unusual circumstances, in which the Fed and the markets cannot tell whether incremental actions by the Fed constitute stepping on the gas pedal or the brake.

I propose a modest 2 percentage point increase in the federal funds rate, along with other measures that on balance would probably have little immediate effect on the economy. Those actions, however, would ensure a speedy return to a normal policy environment--where bank reserves are once again "scarce," where the federal funds rate takes on its traditional usefulness as a gauge of monetary policy, and where Fed actions would have much more predictable consequences.

Background

Monetary policy affects the economy through a variety of "transmission mechanisms." For example, when the central bank expands its balance sheet through securities purchases, there is an increase in "high-powered money" and reserve holdings of banks at the central bank. If commercial banks maintain a constant fraction of their deposits as reserves, the central bank's expansion of securities purchases will create an expansion of bank deposits, which are used to fund bank assets such as loans. This process of deposit and loan expansion that may result from central bank securities purchases operates through what is sometimes called the "loan-supply" transmission mechanism of monetary policy.

Loan-supply changes are only one of the ways monetary policy affects the economy. Purchases of securities affect the economy through channels other than the expansion of the supply of bank loans, such as changes in market interest rates. If the central bank is targeting a particular class of assets in its purchases, the consequent subsidy the Fed is providing to some assets (through its willingness to absorb risks related to the term structure or the mortgage market) may also affect the relative prices of securities. Recent MBS purchases by the Fed, or long-term Treasury purchases, are best seen as a form of fiscal--not monetary--policy, which subsidizes certain risks and thereby favors certain investments. For example, Fed purchases may have important effects on the MBS spread over Treasuries or on the term structure of Treasuries, which may affect investment in housing.

Because exchange rates reflect the forward-looking value of the dollar relative to other currencies, changes in the current or prospective supply of dollars (controlled via the Federal Reserve's ultimate monopoly over the supply of high-powered money) via securities purchases, or other actions or statements by the Federal Reserve, also can affect exchange rates, which in turn influence the supply and demand for exports, imports, and international capital flows.

Further complicating the analysis of monetary policy effects are the numerous tools the central bank can employ to influence markets through each of these channels. The Federal Reserve's purchases and sales of securities are one tool, but the Fed also sets its discount rate (through which it lends funds to member banks), varies its reserve requirement (which can influence the extent to which an expansion of high-powered money results in changes in deposits and loans), determines the interest on reserves (higher interest reduces bank loan supply by encouraging the accumulation of excess reserves), and promulgates regulations that affect banks and other intermediaries' abilities to supply loans and deposits or engage in repurchases (repos). For example, with respect to regulatory influences, when the Fed recently increased bank capital requirements by setting a "Supplementary Leverage Requirement," which required repos to be included in the definition of bank leverage for some banks, the market supply of repos decreased (Allahrakha, Cetina, and Munvan 2016).

Apart from all of those current actions, the Federal Reserve can also influence markets by issuing "forward guidance" about its future intentions with respect to any of those actions, either through speeches or explicit forecasts of the future path of interest rates and other key variables that it can influence.

During some periods (e.g., from 1983 to 2001), the complexity of the monetary policy transmission mechanism did not pose major problems for predicting the influence of monetary policy. That was the case because (1) the Fed's actions were limited to targeting the federal funds rate, (2) the1 primary tool was open market operations, (3) the Fed's activities were confined almost exclusively to the purchase and sale of Treasury bills, (4) important regulatory policies (capital and liquidity requirements) were known and not subject to dramatic change, and (5) the Fed seemed to be following an implicit rule that linked its federal funds target to current levels of inflation and unemployment.

In the current environment, however, it is very hard to know how to gauge the consequences of Fed actions, many of which make use of policy instruments that have not been used in the past. No empirical record exists from past Fed behavior from which to form reliable estimates of the consequences of current Fed behavior. Furthermore, these unprecedented policies are interacting with a unique economic environment (most obviously, one in which nominal interest rates have remained near zero for many years and regulatory policy is subject to constant change). The combination of a unique environment and the use of many new tools (quantitative targeting of the Fed's balance sheet, as in QE1, QE2, and QE3; Fed involvement in the repo market; Fed setting of rates of interest paid on reserves; Fed guidance statements about likely future policy), operating through many potential channels of...