Monetary Rules
Left to their own devices, the Federal Reserve’s Federal Open Market Committee will more often than not adopt a policy of “easy money”. Rather than adopting a monetary policy that stabilizes the value of the dollar, the FOMC has preferred to use a policy they think will drive unemployment down. The way in which they ran their Quantitative Easing (QE) program might actually have caused deflation with little or no economic stimulation (see here and here and here and here and here and here). Â
This sounds crazy since usually you think increasing the money supply a great deal beyond the GDP growth rate will be inflationary. The basic story told by the linked posts above goes something like the following. We postulate that the Federal Reserve buys a huge amount of treasury bonds and mortgage-backed securities from banks, paying them a large amount of new money. However, the Fed then pays interest to banks on excess reserves they place back into the Fed, luring about 80% of the QE money back into reserve. Only a trickle finds its way into the economy, buying stocks and mortgages.  QE can work to stimulate the economy and raise nominal interest rates from zero only if the Fed also has a commitment to higher future inflation, i.e. only if the Fed allows a lot more of that new money to find its way into the economy. (Beware! I am spinning a Keynesian narrative in the last sentence.) If it does not (and it has not), and if this policy is extended in time, businesses and the public are persuaded that inflation will not come from QE. With these expectations reinforced, people and companies begin to reduce their expenditures and save their money instead. This general trend then causes deflation.
Whether or not this narrative really describes the effect of QE, there is no denying the fact that the U.S. and Japan have been in a deflationary environment under QE (see here and here and here). In other historical periods (I have in mind the 1970s and the 2007-2008 crisis), the Fed has tried to stimulate the economy, only to generate inflation or stagflation, or to inflate asset bubbles. It would seem we can get stable money only if Congress imposes some kind of rule that limits Federal Reserve authority to  create money. The reasons why the Federal Reserve should concentrate only on keeping the value of a dollar as constant as possible were explored in our last post on monetary policy. Stable money should be the only Fed mandate.
A number of monetary rules for maintaining monetary stability have been proposed. They come in two flavors: rules in the first type are targeting based rules, and rules with the second flavor are based on market  relationships. First, as they are somewhat simpler in conception, let us consider the targeting rules.
Friedman’s k-percent rule:
Proposed by Milton Friedman, the k-percent rule would increase the money supply by a fixed percentage every year at a rate that could not be changed by the Federal Reserve (or the central bank in another country). The rate k could be determined by the legislature or by some other rule. One drawback would be that k might not be able to adjust to sudden or unexpected changes in the economy, making rapid adjustments to supply and demand imbalances impossible.
Taylor’s interest rate rule:
With this rule, a target would be set for the short-term interest rate. If the market-determined interest rate was smaller than the target rate, the Fed would adjust the money supply by open market operations (The buying or selling of government bonds to member banks), by changing the federal funds rate (The interest rate at which member banks lend to each other, usually over-night and uncollateralized), or by changing the reserve requirements for member banks (i.e. the percent of a bank’s assets that must be deposited with the Fed). If the market rate is smaller than the target rate, the amount of money in the system would be adjusted downward to raise market interest rates toward the target. If the market rate is larger than the target, money would be injected into the system to lower market rates. The target rate would be calculated from a formula  derived from the assumption that the nominal interest rate is equal to the real interest rate plus the inflation rate if the inflation rate is equal to the targeted inflation rate and the real output of the economy (i.e. GDP) is equal to the potential output. See here for more. If inflation and output are above desired levels, the target interest rate increases and money would be withdrawn from the economy to raise market interest rates. If they are below desired levels, the target rate decreases and the money level is increased to lower market interest rates to the target. Since the formula for the target rate could be calculated at any time, the money level could be adjusted as frequently as deemed necessary. It has the drawback that it could be costly to implement by requiring timely measurement of the needed variables for the target interest rate formula.
McCallum’s feedback rule:
This monetary rule is actually a family of rules that adjusts the interest rate k in Friedman’s k-rule in order to bring macroeconomic variables to desired levels. See this Wikipedia article for an example that targets a desired inflation rate and a desired GDP growth rate and uses the observed velocity of money. As with Taylor’s rule, it could be costly to implement by requiring timely measurement of the needed variables.
Inflation targeting rule:
This rule targets a given inflation rate and adjusts the money supply every pre-determined period to reduce the inflation variation from the targeted rate. See this article for more. Benefits should include reduced inflation volatility, reduced inflationary impact of shocks, and the setting of a constant expectation of a given inflation rate in society. The possible drawbacks are ones that  a Keynesian would expect:
- Restricted ability of the central bank to respond to crises or unforeseen events,
- A policy that would not be explicitly responsive to unemployment, undesirable exchange rates, and other macroeconomic variables besides inflation.
- Potential instability in the event of large supply-side shocks.
A neoclassical economist would say these kinds of problems should not be addressed by monetary policy anyway, but typically by tax cuts and reduction of economic regulations. These are the provinces of the political branches of government, the Congress and the executive.
In addition to these targeting rules, there are at least three proposed market-based rules that are institutional changes. They could possibly lead to greater economic stability than the target rules.
Market-based nominal income targeting:
The idea of this monetary rule is to tie the level of money supply to market expectations of future GDP. The central bank would be restricted to buying and selling an unlimited amount of a derivative financial asset. This derivative asset would have a market value dependent on the actual level of nominal national income, set in advance in a derivative contract, and would control the actual price at which the central bank would buy or sell the derivatives. The difference in the dollar value of derivatives bought with the derivatives sold by the central bank in a given period would determine the increase/decrease to the money supply in that period. The benefit of this kind of rule is that it would use the knowledge as well as the profit-seeking incentives of the individuals that make up the entire market. See here and here for details on this type of rule.
Commodity Standards:
A renewed gold-standard would be one example of this kind of rule for determining the money supply. A quantity of some kind of commodity would be set as equal to one dollar. The political barriers against returning to a gold standard or to any other kind of commodity standard are probably insuperable. Nevertheless, such a standard does have the desirable properties of being self-correcting and stabilizing.
Free banking:
Harking back to the 19th century, this kind of system would allow anyone to open a bank and issue bank notes backed by the assets of the bank. Such a system might well be stabilizing in the same way that nominal income targeting would be, but without the backup of a Federal Reserve, we could expect periodic bank runs and all the ills the Fed was originally created to prevent. Also transition costs would be extremely high.
The targeting rules would be much easier to adapt to our modern institutions of central banks than would be the market-based rules. A Friedman rule would probably be rejected by a majority of economists as too inflexible for handling unexpected economic developments. Any of the remaining three – Taylor’s interest rule, McCallum’s feedback rule, and the inflation targeting rule – would be far superior to the discretion of a Fed that has given us repeated asset bubbles, inflations, and stagflation.
Views: 2,606