Summary of Remarks made by Governor Edward M. Gramlich
February 27, 1998
In this speech Governor Gramlich addresses the issues that arise when the question of whether the Federal Reserve Board should use rules in the conduct of monetary policy is contemplated. He tells how the Fed once based policies on monetary aggregates and that now many suggest they base them on the federal funds rate. The Fed currently votes on monetary policy based on the judgement of its members.
There are several different types of policy rules he discusses. The first is an unconditional rule, for example, having monetary authorities increasing the money supply a certain percentage each year or base a rule on some target objective such as rigid prices and reduce the inflation rate to a certain level. Another intermediate approach could be called a feedback rule. Using this approach, policy objectives are specified in the rule and authorities would react in a regular way to the discrepancies between actual values and the target levels of these variables.
Rules also vary in how restricting they are. They can be as extreme as being mandated by Congress or self-imposed. Either way, they can include exceptions for special circumstances. It is also possible for the rules to be simple and informal in order to guide some of the members in their votes.
There are numerous advantages and disadvantages for rules. A disadvantage of the rules is that they must be oversimplified and authorities do not benefit from them. Also, there are several monetary objectives that conflict with one another. Rules based on one objective may oppose the intentions of another rule. Sometimes rules may not work or work only under certain circumstances.
There are also some benefits to rules. One is that policy is clear, regular, and consistent. Rules can help in guiding, both quantitatively and in behavior, when information is conflicting or when there is political pressure. Rules can determine in advance ways to help stabilize the economy and can use complicated lag patterns. They are helpful when authorities are unsure of how to remedy a situation.
Conservative economist Milton Friedman supported the unconditional rule that if the Fed simply increases the money supply by approximately four percent a year; the rate would lead to long-run price stability. Liberal economist Paul Samuelson argues that the Fed should not only control the money supply but also the interest rate. Friedman s constant rate of growth (CROG) standard explains that if the trend growth of real output is on the order of three percent per year, the trend growth of money of approximately four percent would permit for some inflation and account for changes in velocity.
CROG has some possible problems. There could be sudden increases in the demand for money caused by changes in the creation of money system or foreign exchange. There is also no room for short-term monetary changes built into CROG. Passivists, such as Friedman, believe there is no reason to have discretionary policy actions- activist policy is either too little (too much) or too late (too soon).
A different approach used by some industrialized nations is inflation targeting. The main purpose of this method is to stabilize prices -which ultimately is the goal of monetary policy. Policy makers have no ability to control or make decisions with any other purpose in mind. This may sound constricting but actual inflation targeting strategies have historically been more flexible. The number of pros and cons of this type of policy is equal to other types of policies. People can understand and see how the policy directly relates to stabilizing prices. But it also takes away many privileges of the central bank.
The main difference between inflation targeting and nominal income targeting is that they respond to shocks differently. Nominal income will not respond to price shocks as much as inflation. If there are output productivity shocks, these shocks could change nominal income and pressure the central bank to expand or contract even if inflation is at equilibrium. It is difficult to determine if price or productivity shocks will be larger and more noticeable and if nominal income targeting will or will not improve on inflation targeting.
John Taylor came up with an intermediate approach that manipulates the federal funds rate, which can be easily controlled. Using the federal funds rate directly eliminates the effects of shocks in the demand for money. Taylor s rule is expressed in the formula:
PFR= r* + p + .5y + .5(p-p*)
where PFR is the recommended federal funds rate in nominal terms, r* is the equilibrium funds rate in real terms, p is the actually rate of inflation, y is the deviation of output, and p* is the desired inflation.
In Taylor s rule, the adjustment coefficients are already include stabilizing components. There are indirect properties through the behavior of long term bond markets. The Taylor rule also presumes that monetary policy is independent, that the federal funds rate can vary without having any effect.
There are uncertainties concerning all these objectives, especially when concerning NARIU. The Fed must understand where NAIRU is because uncertainty carries over to output gap uncertainty. When using the equilibrium funds rate, the Fed must determuine the real federal funds rate, even if inflation and output are on target.
Because of these uncertainties it is very difficult to determine which of the policies is the best. Rule making may become so advanced that monetary rules may become useful in the conduct of monetary policy. For now, they will continue to help guide policy makers to avoid large mistakes that will effect the whole country.