Order Code 98-16 E
CRS Report for Congress
Received through the CRS Web
The Federal Reserve: Should Its Mandated
Goal Be Price Stability? The Issues and
Technical Problems
Updated March 25, 2005
Marc Labonte
Analyst in Economics
Government and Finance Division
Gail Makinen
Consultant in Economic Policy
Government and Finance Division
Congressional Research Service ˜ The Library of Congress
The Federal Reserve:
Should Its Mandated Goal Be Price Stability?
The Issues and Technical Problems
Summary
Some economists have long criticized the American model of central banking
for featuring multiple policy goals, discretion on the part of the central bankers as to
which goal or goals to emphasize, freedom in the choice of instruments to achieve
the policy goals, and rather vague accountability for policy failures if, indeed, these
can even be identified. Recently, the critics have urged that the multiple policy goals
of the Federal Reserve be replaced by a single goal of price stability. Critics believe
that central bankers tend to use their discretionary powers to achieve political as well
as economic objectives, notably to create “good times” through monetary expansion.
Since these “good times” do not last long, such a policy imparts a costly inflationary
bias to an economy and, hence, is not economically optimal over time. Among other
virtues, it is argued that a single goal would provide an explicit anchor for the
American monetary system. The proponents of a price stability goal are supported
by an array of economic theories and empirical studies.
The current model has strong support as well. Since an economy faces many
unforseen contingencies, supporters argue that giving central bankers multiple goals
and a high degree of discretion is optimal. They question whether a price stability
goal would be flexible enough to allow the Federal Reserve to remain the lender of
last resort to the U.S. financial system and to cope with short run stabilization
problems that beset the country at times. They note that the Fed has successfully
delivered price stability for over two decades under the current multi-goal regime.
Their position is also supported by empirical studies and theoretical arguments.
To formally replace the current multi-goal mandate of “maximum employment,
stable prices, and moderate long-term interest rates” with a single goal to maintain
stable prices would require an act of Congress. Members from both parties have
introduced such legislation in past Congresses. A number of countries have recently
made price stability the sole goal of their monetary policy. In practice, these
countries have not focused their monetary policy solely on price stability, but have
responded to changes in output as long as it did not undermine long-term price
stability. This arrangement has been coined “constrained discretion,” since it has left
central banks with significant freedom to set policy as they see fit.
The price stability goal, while simple and straightforward, raises a number of
technical questions about definition, in terms of a goal of inflation or constant prices,
whether a point or band target should be used, and the appropriate price index to
measure price stability. The goal may also place constraints on fiscal, debt
management, and exchange rate policies — policies not delegated to the Federal
Reserve. Accountability should be greater than under the current regime, but the
degree of accountability depends on how the goal is defined. Since it is infeasible
to expect the central bank to keep inflation right on target at all times, consideration
should be given to the exceptions granted to the goal and the permissible time
interval over which the targets must be met. But these exceptions in turn make
accountability more difficult. This report will be updated as events warrant.
Contents
The Case for Refocusing the Federal Reserve . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
(1) The Neutrality of Money . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
(2) Long and Variable Lags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
(3) Rational Expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
(4) Precommitment and Credibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
(5) The Natural Rate of Unemployment (the NAIRU) . . . . . . . . . . . . . . . . . 6
(6) The Desire for Greater Accountability . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
The Case Against A Single Goal for
Federal Reserve Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
(1) Is Money Neutral in the Long Run? Are Expectations Rational? . . . . . 10
(2) Does NAIRU Exist? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
(3) Long and Variable Lags (Again) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
(4) A Little Inflation Can Make Important Adjustments Easier . . . . . . . . . 13
(5) The Importance of Other Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
(6) The Need for Flexibility and Discretion . . . . . . . . . . . . . . . . . . . . . . . . . 15
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Technical Problems With Implementing a Price Stability Goal . . . . . . . . . . . . . 16
The Definition of Price Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Choice of Price Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Should There Be a Fixed Band Around the Target, and If So, How Wide
Should It Be? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Would Exceptions Be Allowed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
How Long Would the Central Bank Have to Achieve Its Goal? . . . . . . . . . 21
If a Target Were Changed or Missed, What Would Be the Optimal
Time Over Which to Return to the Target? . . . . . . . . . . . . . . . . . . . . . 22
What Incentives Are There, or Should There Be, for the Federal
Reserve to Achieve Its Announced Target? Or, What Type of
Accountability Should There Be? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Who Would Set a Target for Price Stability? . . . . . . . . . . . . . . . . . . . . . . . 23
How Could the Government Set a Permanent Target? . . . . . . . . . . . . . . . . 24
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
List of Figures
Figure 1: Phillips Curve, 1961-1969 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 2: Phillips Curve, 1970-2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
List of Tables
Average Annual Inflation Rate in the Industrial Countries, 1950-2004 . . . . . . . . 2
The Federal Reserve: Should Its Mandated
Goal Be Price Stability?
The Issues and Technical Problems
The 1970s stand out in the post-World War II era as the inflation decade. This
is evident from the data in Table 1 for the leading industrial countries. Inflation
served to motivate public policy in a number of countries. A major impetus of these
initiatives was to focus central banks on one major policy goal: the achievement of
price stability. In some cases this has taken the form of legislation specifying this
goal to the exclusion of all others. In other cases, the central bank was granted
greater autonomy in the expectation that this would lead to the desired outcome.
During recent years, Canada, the United Kingdom, New Zealand, Sweden, Australia,
and Israel, among others have adopted inflation targeting as the major goal of
monetary policy. And since the European Central Bank’s inception, price stability
has been its main objective.
These developments have not gone unnoticed in the United States. The ultimate
objectives of Federal Reserve (Fed) policy are currently specified in the Federal
Reserve Reform Act of 1977 as maintaining the “long run growth of monetary and
credit aggregates commensurate with the economy’s long run potential production,
so as to promote the goals of maximum employment, stable prices, and moderate
long-term interest rates.”1
Both Democratic and Republican Members of Congress have introduced
legislation that would replace the current multigoal mandate of “maximum
employment, stable prices, and moderate long-term interest rates” with a single goal
to maintain “stable prices.” In some proposals, “stable prices” is defined as a low
inflation rate. In others, the overall price level would remain constant.
An early example of Democratic efforts along these lines was the “Zero
Inflation Resolution” introduced by Congressman Stephen Neal of North Carolina
in 1989. In the 109th Congress, Representative Jim Saxton, Republican of New
Jersey, introduced the “Price Stability Act of 2005,” H.R. 498, “To mandate price
stability as the primary goal of ... monetary policy ....”
1 For more information, see CRS Report RL30354, Monetary Policy: Current Policy and
Conditions, by Marc Labonte and Gail Makinen; and CRS Report RS20949, Federal
Reserve: Recurrent Public Policy Issues, by Marc Labonte.
CRS-2
Table 1. Average Annual Inflation Rate in the Industrial
Countries, 1950-2004
(data in percentages)
1950-1959
1960-1969
1970-1979
1980-1989
1990-1999
2000-2004
United States
1.8
2.3
7.1
5.6
3.0
2.6
United
3.5
3.6
12.6
7.4
3.7
2.4
Kingdom
Austria
6.8
3.3
6.1
4.0
2.4
1.9
Belgium
1.9
2.7
7.1
5.1
2.2
2.1
Denmark
3.8
5.3
9.3
7.1
2.1
2.2
France
6.2
3.8
8.9
7.8
1.9
1.9
Germany
1.1
2.4
4.9
2.9
2.3
1.5
Italy
2.9
3.4
12.5
11.8
4.1
2.5
Netherlands
3.8
4.2
7.1
3.1
2.5
2.8
Switzerland
1.1
3.1
5.0
3.3
2.4
0.9
Canada
2.4
2.5
7.4
6.7
2.2
2.4
Japan
3.1
5.4
9.1
2.5
1.2
-0.5
Greece
6.5
2.0
12.3
20.1
11.1
3.4
Ireland
3.9
4.0
12.7
9.9
2.3
4.2
Portugal
0.7
4.0
17.1
18.2
6.0
3.3
Spain
6.2
5.8
14.1
10.6
4.2
3.2
Australia
6.5
2.5
9.8
7.6
2.5
3.3
New Zealand
5.0
3.2
11.4
12.5
2.0
2.3
Mean
3.7
3.5
9.7
8.1
3.2
2.4
Standard
1.99
1.08
3.28
5.04
2.17
1.02
Deviation
Source: For 1950-1989: Consumer prices compiled by the International Monetary Fund and reported
in Grilli, Masciandaro, and Tabellini. Political and Monetary Institutions and Public Financial Policies
in the Industrial Countries. Economic Policy. October 1991, p. 344. For 1990-2004: Consumer
Prices compiled by International Monetary Fund. Statistical computations made by authors of this
report.
Governors and Regional Bank Presidents of the Federal Reserve are perceived
to have mixed views on making inflation the sole goal of monetary policy, with
Chairman Alan Greenspan perceived to be opposed to a Congressionally mandated
CRS-3
inflation target.2 William J. McDonough, former president of the New York Federal
Reserve Bank, said:
“It is often said that there is a worldwide community of central bankers.
I certainly feel that way. Central bankers in all countries share a number
of concerns. Perhaps the most important of these is the desire for price
stability. While central bankers may differ in the way they seek price
stability — differences grounded in our respective histories, customs, and
institutions — the goal we all strive for is no less important.”3
While the purpose of this proposal is straightforward, it raises many technical
issues which this report will consider. As a preface to this discussion, the report will
begin with pro and con cases given by economists who either favor or oppose the
legislation focusing the Federal Reserve on an exclusive goal of achieving price
stability.
The Case for Refocusing the Federal Reserve
The ultimate purpose of refocusing the Federal Reserve on a price stability goal
is to increase the amount of real goods and services available to the nation, not, as the
late Prof. James Tobin reminded us, because “Price or inflation stability is . . . an
ultimate social good.”4 Thus, it must be shown that inflation has a pernicious effect
on economic growth, the efficiency with which the economy works, the choices
available to Americans to satisfy their needs and wants, or on employment. Such a
case can be made.5 But should price stability be the sole goal of monetary policy?
Proponents make that case based on five powerful strands of economic theory and
empiricism, as well as one political argument.
(1) The Neutrality of Money
The basic case made by economists for refocusing the Federal Reserve is built
on a very old economic doctrine known as the “Neutrality of Money.” This is the
view that the influence of money and changes in the money supply are neutral with
respect to changes in the real economy where economic growth, employment, real
2 Vivien Lou Chen, “Fed’s Debate on Inflation Targets May Shape Post-Greenspan Era,”
Bloomberg News, Mar. 22, 2005.
3 William McDonough, “A Framework for the Pursuit of Price Stability,” Economic Policy
Review, vol. 3, no. 3 (Aug. 1997), p. 1.
4 Prof. Tobin’s full quote is “As Jacob Marschak gently reminded Henry Wallich in a
memorable Yale seminar years ago, prices are not in anybody’s utility function. Price or
inflation stability is not an ultimate social good, but must be justified as an instrument that
will deliver more utility-laden goodies to the society.” See James Tobin, Panel Discussion
in J. C. Fuhrer, ed., Goals, Guidelines, and Constraints Facing Monetary Policymakers,
Federal Reserve Bank of Boston, June 1994, pp. 232-236.
5 The case is made in CRS Report RL30344, Inflation: Causes, Costs and Current Status,
by Mark Labonte and Gail E. Makinen.
CRS-4
interest rates, and relative prices are determined. These depend on such factors as the
choices individuals make between leisure and work, the technical means by which
labor and capital are combined, and the saving/investment decisions by economic
agents. Money, on the other hand, influences only money things such as the price
level, money wages, the money value of output, and the nominal or market rate of
interest. Since this is money’s primary economic effect and a changing price level
can have harmful effects on an economy, the doctrine of the neutrality of money can
serve as a powerful rationale for focusing monetary policy on achieving price
stability.6 7
(2) Long and Variable Lags
A second element supporting refocus on price stability is based on the empirical
finding that changes in the money supply can affect the pace of economic activity and
prices with a lag that is both long and of a variable length (i.e., a given change in the
rate at which the money supply grows does not always affect the pace of economic
activity and prices within the same length of time). This may be due to changes in
the underlying structure of the economy as well as to changes in policy regimes, as
might be expected to occur when a country moves from a system of fixed to a system
of flexible exchange rates.
Because of the long and variable lag of monetary policy, changes in policy
undertaken today can have their effect on the economy after the underlying cause of
the original disturbance may already have corrected itself. If so, countercyclical
monetary policy could be destabilizing. For that reason, some economists argued
against using monetary policy to promote such goals as full employment. Rather,
they argued, it should be geared to producing stable prices, since money’s lasting
effect on the economy is on nominal magnitudes.
(3) Rational Expectations
Third, there were developments in specifying how economic agents formed their
expectations. Expectations, especially of inflation, are important in many forward
looking price-setting activities in market economies, such as wages, the prices of
individual goods and services, and interest rates. The revolution in this area occurred
with the introduction of so-called theory of rational expectations.
Rational expectations is the theory that economic agents make use of all relevant
information, including information about monetary policy, in formulating their
expectations about the future. Wage earners, for example, would strike a wage
6 The neutrality of money was viewed by early economists and some of their later followers
as a long run proposition. In the shorter run, variations in the growth rate of the money
supply could affect the growth rate of real output and employment.
7 The classic work exploring money’s effect on the economy is Milton Friedman and Anna
Schwartz, A Monetary History of the United States 1867-1960 (Princeton University Press,
1963). See also, Christina D. Romer and David H. Romer, “Does Monetary Policy Matter?
A New Test in the Spirit of Friedman and Schwartz,” in O.J. Blanchard and S. Fischer, eds.,
NBER Macroeconomic Annual 1989 (Cambridge, MA: MIT Press, 1989), pp. 121-170.
CRS-5
bargain with an employer only after considering what monetary conditions would
likely prevail over the period of the employment contract.8 Rational expectations do
not mean that individuals are always right. It only means that they do not make
systematic mistakes.
This method of forming expectations has a powerful implication. It implies that
systematic monetary policy, or that expected by economic agents, can have no effect
on the real sector of the economy since it would have been anticipated by economic
agents and become a part of their market behavior. Thus, systematic monetary policy
is also neutral in the short run (as well as in the more general case of long run
neutrality). If monetary policy does affect the pace of economic activity in the short
run, it must be because it comes as a surprise — it is unanticipated and
nonsystematic. Its nonneutral effects will last only until economic agents incorporate
it into their wage, price, and interest rate decisions (this leads to the so-called
misperceptions theory of business cycles).
The question might arise why the Federal Reserve would want to spring
monetary surprises on the economy. A major reason given in the literature is that it
yields to political pressures to boost economic activity — create good times — prior
to presidential elections. This notion of a political business cycle enjoys some
support among economists. It should also be noted that a surprise based monetary
policy is not an optimal policy since ultimately the cost of avoiding inflation reduces
welfare and output is no higher than it would have been in the absence of the surprise
inflation. In the jargon-rich language of economists, this is also known as the “time
inconsistency problem.”
The notion of rational expectations supports focusing Federal Reserve policy on
the single goal of price stability. This is because, according to rational expectations,
the only way the Federal Reserve can alter employment is by engineering a surprise.
Surprise changes in monetary policy can, at best, have only a short run effect on
employment. The longer run effect is only on the inflation rate and inflation has
harmful side effects on the economy.
(4) Precommitment and Credibility
The discussion above stresses the importance of misperceptions by economic
agents as a cause of business cycles. If errors of predictability and errors of
understanding are an important part of misperceptions, then making monetary policy
more predictable, consistent, and understandable should reduce errors and
misperceptions. Furthermore, this theory suggests that over time surprises will
8 While this may seem to the reader as the natural thing for individuals to do, in empirical
approaches to expectation formation, economists often reasoned that economic agents would
merely extrapolate the past in forming notions about the future. Thus, expectations about
the future rate of inflation were taken to be some weighted average of past inflation. In this
formulation, economic agents would neglect some available relevant information about the
forces theory suggested caused inflation (e.g. the rate of growth of the money supply) and,
instead, form their notions about the future by looking only at the past actual rate of
inflation. This was regarded as irrational.
CRS-6
become less and less effective at stimulating the economy, until they ultimately
become counterproductive. Conversely, the theory suggests that monetary policy
changes, such as disinflations, could be faster and less costly if credibility were
greater. It is thought that under a credible central bank individuals might change their
inflationary expectations more quickly, making the economy more flexible as a
result. Thus, a monetary policy based on precommitment and credibility should be
conducive to economic stability. A single goal such as price stability, it is argued,
can increase the clarity and understandability of monetary policy by “anchoring”
expectations and, thus, contribute to this end.9
Figure 1: Phillips Curve, 1961-1969
6
1969
5
)
1968
4
(CPI
e
3
1967
1966
ion Rat 2
flat
1965
In
1964 1963
1
1962
1961
0
3
4
5
6
7
8
Unemployment Rate
Source: Bureau of Labor Statistics
(5) The Natural Rate of Unemployment (the NAIRU)
In the 1960s, there was a broad consensus in macroeconomics that a relationship
existed between inflation and unemployment known as the “Phillips Curve.” This
theory posited that a rise in inflation would lead to a predictable fall in
unemployment, and vice versa.10 There was, as shown in Figure 1, considerable
empirical support for this notion.
9 For example, see Ben Bernanke, et al., Inflation Targeting (New Jersey: Princeton
University Press, 1999), p. 20.
10 It should be noted that a long run trade-off of unemployment for inflation violates the
neutrality of money for it suggests that inflation (a monetary phenomenon) can have a
permanent effect on employment (a phenomenon that according to the neutrality doctrine
is determined exclusively in the real sector of the economy). Thus, the notion of the NAIRU
is embodied in the concept of the neutrality of money.
CRS-7
In the late 1960s, two economists, Professors Edmund Phelps of Columbia
University and Milton Friedman of the University of Chicago, independently rejected
the Phillips Curve framework and in its place put forth the notion of the non-
accelerating inflation rate of unemployment or NAIRU as a definition of the
unemployment rate consistent with the full employment of labor.11 12 This refers to
an unemployment rate consistent with a stable rate of inflation. In the long run, the
economy will return to the NAIRU with any inflation rate, be it zero or any positive
number, and so there is no permanent tradeoff between inflation and unemployment.
Phelps and Friedman suggested that the empirical finding shown in Figure 1
occurred because individuals during the 1960s had not anticipated the inflation that
had occurred because they based their expectations upon the 1950s when inflation
was low. Once they built into their expectations the inflation that had occurred (both
Phelps and Friedman wrote before the rational expectations revolution), Phelps and
Friedman predicted the stability of the Phillips curve would vanish. The curve would
shift up and to the right. The only way monetary policy could then keep the
unemployment rate below the natural rate would be to engineer continual surprises
— keep accelerating the inflation rate.
The contribution of Phelps and Friedman was important and prescient; what
these economists predicted seemed to come to pass. As the data plotted in figure 2
show, the tradeoff that is apparent in Figure 1 vanishes after the 1960s. The large
amount of dispersion in the data suggests that there is no stable tradeoff between the
two variables. If anything the relationship becomes slightly positive — as the
unemployment rate fell, so did the inflation rate.
The NAIRU cannot be influenced by monetary policy because it is determined
in the real sector of the economy by such things as the work/leisure choices of
individuals. Thus, the function of monetary policy, it is argued, should be to keep
aggregate demand growing at a rate consistent with price stability.13 If aggregate
demand grows at a rate consistent with price stability, then the NAIRU (or full
employment) will prevail, making the NAIRU concept consistent with making price
stability the sole goal of monetary policy.14
11 See Edmund Phelps, “Phillips Curves, Expectations of Inflation, and Optimal Inflation
Over Time,” Economica, NS, vol. 135 ( 1967), pp. 254-281, and Milton Friedman, “The
Role of Monetary Policy,” American Economic Review, vol. 58 (Mar. 1968), pp. 1-17.
12 Economists have always been both uneasy about and somewhat vague in defining what
is meant by “full employment.” Clearly, it has never meant a zero unemployment rate. For
some economists who believed in a permanent trade-off between inflation and
unemployment, full employment had, at best, an ambiguous definition. Other economists
were content with accepting an arbitrary rate of 4%, such as embodied in the Humphrey-
Hawkins Act (also known as The Full Employment and Balanced Growth Act of 1978).
13 This, however, does not by itself support a monetary policy geared to producing a zero
rate of inflation since the economy can be at its NAIRU at any constant rate of change of
prices. Other factors, such as the losses to an economy from a positive rate of inflation must
be invoked to support a monetary policy pledged to justify price stability.
14 It should be noted that the NAIRU was not expected to be constant across time.
(continued...)
CRS-8
Figure 2: Phillips Curve, 1970-2001
14
12
10
(CPI)
te
8
Ra
n
6
tio
a
4
fl
In
2
0
3
4
5
6
7
8
9
10
Unemployment Rate
Source: Bureau of Labor Statistics
(6) The Desire for Greater Accountability
In addition to the economic arguments presented above, there is a closely related
political argument for making price stability the sole goal of monetary policy. There
has long been dissatisfaction voiced with the accountability of the Federal Reserve
for the macroeconomic performance of the economy. The political independence
granted to the Fed combined with the imprecise and oftentimes conflicting goals it
is mandated to achieve means that there is little chance for congressional criticism
of its performance to have a concrete effect on future policy decisions.15
Twice a year, the Federal Reserve reports to the Congress on the state of the
economy and monetary policy. At these hearings, the Chairman of the Federal
Reserve Board presents a review of the current state of the economy and projections
for the future course the economy is expected to take over the coming 18- to 24-
months. This has prompted the Nobel Prize winning economist, James Tobin to
declare: “It is disingenuous for the FOMC [Federal Open Market Committee] to
14 (...continued)
Changing labor market conditions, changing demographics of the labor force, changes in
labor legislation, and changes in institutions governing labor, among other changes, it was
argued, should be expected to change NAIRU. See CRS Report RL32774, A Changing
Natural Rate of Unemployment:Policy Issues, by Marc Labonte.
15 For more information, see CRS Report RL31056, The Economics of Federal Reserve
Independence, by Marc Labonte.
CRS-9
forecast or ‘project’ the economy, pretending that they have no control over it.”16
When the economy behaves differently from these projections, little effort is exerted
in the reports to explain why. When the effort is made, the explanation frequently
attributes it to unexpected events. Federal Reserve policy is seldom, if ever, the
culprit.
Formal accountability is weak in this system. No governor of the Federal
Reserve has ever been removed from office for any reason, although removal is
statutorily permissible “for cause.” Some have not been reappointed, however, and
during hearings the Members of Congress have not been hesitant in voicing
displeasure with the performance of the economy, especially during economic
downswings and periods of inflation.
Proponents of a price target argue that a target would make the Fed more
accountable. The Fed would no longer be able to justify its decisions by pointing
arbitrarily to the achievement of one of its goals, while disregarding its failure to
meet other goals. The target could be crafted in such way that failure to reach the
target led to explicit remedial actions, as discussed below. Yet a target would not
undermine the Fed’s independence, they argue, because it would not lead to political
interference in the day-to-day decision-making of the Fed.
Summary
Proponents of a single price-stability goal base their case on several basic tenets
of economic theory. First, money is “neutral” in the long run, meaning it cannot
affect real economic activity. It can only affect inflation, which in excess is harmful
to the efficient market allocation of resources; this makes price stability the natural
goal of monetary policy in their eyes. Second, unemployment tends to a “natural
rate,” which is dictated by labor market conditions and policies. Since monetary
policy cannot affect this natural rate of unemployment, the Federal Reserve cannot
be held responsible for achieving a goal of full employment. Third, people have
rational expectations and cannot be systematically fooled by monetary “surprises.”
This implies that the economy will function most smoothly if the monetary policy is
given a predictable “anchor” such as price stability so people can make decisions
with some degree of certainty about the future path of policy. It is claimed that this
anchor will make the Fed more accountable for its actions and will make its decisions
more credible, which, in turn, will make policy more effective.
For some proponents, a price-stability goal is desirable because they believe
discretionary monetary policy has done more to destabilize than stabilize the business
16 James Tobin, “Panel Discussion,” in J. C. Fuhrer, ed., Goals, Guidelines, and Constraints
Facing Monetary Policymakers, Federal Reserve Bank of Boston, June 1994, p. 235. Tobin
goes on to declare: “I would like to see the report contain the consensus of the FOMC as to
the macroeconomic path they will use their powers to achieve over coming quarters and
years.” Some suggest this would add accountability to the present regime. The Federal
Open Market Committee is the principal policy committee of the Federal Reserve. Its’
voting members consist of the Board of Governors and five of the presidents of the regional
Federal Reserve Banks.
CRS-10
cycle in the past. They base their case on the temptation for monetary surprises and
the long and variable lags in policy effectiveness that make successful discretionary
policy unlikely. Other proponents acknowledge that the responsible application of
monetary policy is helpful in the reduction of economic instability, but would not see
responsible stabilization policy as inimical to a price stability goal in most cases.
They would be likely to agree with Bernanke and Mishkin’s characterization of
monetary policy under a price-stability goal as “constrained discretion” in practice.
In this characterization, central banks would be free to stabilize the business cycle as
long as long-run price stability is not placed at risk in the process.
The Case Against A Single Goal for
Federal Reserve Policy
It is, perhaps, best to begin the case against a single goal for the Federal Reserve
with the proposition that while the current monetary system does not have an explicit
anchor such as would be provided by a fixed exchange rate or a legislated inflation
target, it does have an implicit anchor. The Federal Reserve has been extremely
reluctant over the past two decades to let the U.S. inflation rate rise above 4%
without intervention. Rates above 4% seem to bring on monetary tightening of the
type that often leads to a cyclical downturn. Thus, in practice, the adoption of an
inflation target cannot be supported on the grounds that the Fed has neglected to
pursue the goal of price stability. The burden of proof should be on proponents to
show that the Fed’s past performance could have been improved — or there is reason
to believe that future performance could be improved — if a price stability regime
had been in place.
The case against an exclusive price stability goal can be subdivided into six
parts:17
(1) Is Money Neutral in the Long Run? Are Expectations
Rational?
While most economists believe in theory in the neutrality of money as a long run
proposition, some also agree that for all practical purposes, over any reasonable time
horizon, money is not neutral. Changes in the growth rate of the supply of money
can, over such a time horizon, according to this view, have significant and lasting
effects on the growth of real output and employment.
This perspective was well stated by Prof. Richard N. Cooper:
...the strong and sometimes helpful working hypothesis of the economics
profession [is] that in the medium to long run, money supplies affect only price
levels, not the real side of economies, so that central bank action can only
17 As will be explained below, these six parts of the case against a price stability focus are
not mutually exclusive. For example, an opponent of the view that money is neutral could
hardly believe in NAIRU.
CRS-11
influence prices in the long run. This working hypothesis through repetition and
use has come to be accepted as fact, as a structural characteristic of actual
economies. It is a dangerous assumption, largely because it is rarely questioned.
The evidence is ample that it is false in the short run that runs for several years.
The best that can be said about the empirical evidence over longer periods is that
with sufficient imagination by the estimators, the hypothesis cannot be rejected
— a very weak test on which to base important policy decisions.18
Those rejecting the neutrality thesis believe that a stable Phillips curve does
exist over reasonable time periods and ought to be exploited, for they argue that the
costs to an economy from unemployment far exceed the costs due to inflation (for the
rates of inflation experienced by the United States in the post World War II period).19
From the above, this can be seen as an assault on views such as those that business
cycles are due to misperceptions of the actual course taken by inflation and on the
concept of the NAIRU.
Other economists question outright the practical significance of rational
expectations. They point out that there is really little evidence that American
business cycles are due to misperceptions of inflation (see page 5) and there is
equally little evidence to support the view that the Federal Reserve somehow yields
to political pressure to create booming economic conditions just before presidential
elections (the so-called political business cycle or that the Federal Reserve engages
in policies that are “time inconsistent.”).20 Former Presidents such as Gerald Ford,
Jimmy Carter, and George H.W. Bush might agree since they did not have the best
of economic conditions when they faced re-election. There is something quite
fundamental in this criticism that should not be overlooked. The proponents of
refocusing the Federal Reserve on a single goal of price stability do so because of
their view that the continuation of inflation is due largely, if not entirely, to the self-
interested short-term focus of politicians aided and abetted by the discretionary
choices made at the Federal Reserve. This criticism is aimed both at this explanation
for inflation and its goal for reform, the conduct of monetary policy to achieve a
single goal.
(2) Does NAIRU Exist?
Some question the entire concept of NAIRU. They point out that the behavior
of the U.S. economy in the late 1990s is at variance with the widely held view that
for the United States NAIRU is about 6.0%. If this estimate is correct, the United
States should have experienced a rising rate of inflation since the unemployment rate
18 See Richard N. Cooper, “Panel Discussion,” in J. C. Fuhrer, ed., Goals, Guidelines, and
Constraints Facing Monetary Policymakers, Federal Reserve Bank of Boston, June 1994,
p. 192.
19 For a statement of this view, see James Tobin, “Inflation and Unemployment,” American
Economic Review, vol. 62, no. 1 (Mar. 1972), pp. 1-18. For a more recent exposition, see
James Galbraith, “Time to Ditch the NAIRU,” Journal of Economic Perspectives, vol. 11,
no. 1 (Winter 1997), pp. 93-108.
20 See, for example, Alberto Alesina, “Politics and Business Cycles in Industrial
Democracies,” Economic Policy, vol. 1 (Spring 1989), pp. 58-98.
CRS-12
was below 6.0% from August 1994 through 2001. But the inflation rate followed no
trend during those years, fluctuating between 1.6% and 3.4%.21 These critics are also
likely to claim that monetary (and fiscal) policy may in fact influence the long run
unemployment rate, contrary to the assertions of the neutrality of money and the
NAIRU. They argue that the future employability of people is, in part, determined
by their experience with unemployment. Thus, severe short term downturns may
affect the longer term unemployment rates of some countries.22
Nevertheless, those economists who are critical of NAIRU must explain the data
patterns observed in Figure 2. An oil price shock (or supply shock in general) will
cause both unemployment and the rate of inflation to rise. Thus, some of the
observations can be explained in this way. Others can be explained by the efforts of
the Federal Reserve to reduce the unemployment caused by the supply shock (which
should reduce unemployment while accelerating the ongoing inflation rate).
(3) Long and Variable Lags (Again)
As it happens, one of the arguments used in favor of a price stability goal can
also be used against it. The technical difficulties in implementing monetary policy
may be as problematic for a price stability goal as for countercyclical policy.
The Federal Reserve does not directly control the price level. Rather, it controls
only the monetary and credit conditions of the country that influence changes in
aggregate demand. And it is the interaction of changes in demand with changes in
supply that affect the price level and the rate of inflation. To the extent that monetary
policy operates with lags that are long and of variable length, maintaining price
stability can be a difficult task. Were a shock to the economy to move inflation away
from the target, policy lags would prevent the Fed from returning inflation to the
target immediately. As explained below, the lack of direct control is not a fatal
problem. Much would depend on how a law would be written, that is, how price
stability is defined and the time horizon over which stability is to be achieved. It may
be that the lags pose no fundamental problem.
21 This is true for the CPI or a stripped down version of the CPI know as the “core” index.
The two price indexes from the GDP accounts fluctuated between 1.2% and 2.3% during
that period.
22 As noted above, those who believe in NAIRU recognize that it is subject to shifts over
time as conditions in labor markets change. These shifts, however, are supposed to be
independent of changes in monetary policy. There is a major development in some of the
European countries that the critics of NAIRU cite as evidence against the concept.
Supporters of NAIRU estimate that for the countries in the Euro Area, the NAIRU has risen
from about 2.5% in the 1950s to perhaps in excess of 8.0% in the 1990s. They do not have
an adequate explanation why this has happened, but some economists hold out the
possibility that it may be related in part to the longer run employment consequences of the
monetary and fiscal policies followed in these countries with an excess emphasis on price
stability. See C.A.E. Goodhart, “Central Bank Independence,” The Central Bank and the
Financial System (MIT Press, 1995), pp. 60-71. See also a symposium entitled “The
Natural Rate of Unemployment,” in The Journal of Economic Perspectives, vol. 11, no. 1
(Winter 1997), pp. 3-108.
CRS-13
Another possible way to deal with the difficulties posed by the long and variable
lags is to use a system of intermediate targets such as the monetary aggregates.
Intermediate targets can provide much useful information about the thrust of Federal
Reserve policy since they are the link between Federal Reserve action and the
ultimate goals of policy. In the case of the monetary aggregates, however, the value
of the information they provide about Federal Reserve intentions has decreased
considerably in the 1980s and 1990s.23 Nevertheless, the case for using an
intermediate target and what is required to make it work is well stated by Bernanke
and Mishkin:
If credibility building is an important objective of the central bank, and if there
exists an intermediate target variable — such as a monetary aggregate — that is
well controlled by the central bank, observed and understood by the public and
the financial markets, and strongly and reliably related to the ultimate goal
variable, then targeting the intermediate variable may be the preferred strategy.24
(4) A Little Inflation Can Make Important Adjustments Easier
Economic systems are subject to a variety of shocks, some of which require
changes in real magnitudes as the system returns to equilibrium. Supply shocks can
pose particularly serious problems. When they involve a reduction in aggregate
supply (such as the OPEC cut-off of oil supplies in 1973), they often require a fall in
real wages in order to restore full employment. Because of the pervasiveness of
contractual arrangements in the U.S. market economy, it is argued that the fall in real
wages can be accomplished with less loss of output and increase in unemployment
if the Federal Reserve allows the price level to rise rather than force an increase in
unemployment to bring about a fall in money wages. In this case, a little inflation is
thought to ease the return to full employment.25 For example, Akerlof, Dickenson,
and Perry derive a model based on the assumption of some downward nominal wage
rigidity and show with U.S. data that below a certain inflation rate a permanent
tradeoff exists with unemployment.26 In this paper nominal wage rigidity holds even
though expectations are assumed rational. Of course, an inflation target could avoid
this problem if the numerical target were set high enough.
23 For a discussion, see CRS Report RL31416, The Monetary Aggregates: Their Use in the
Conduct of Monetary Policy, by Marc Labonte and Gail Makinen.
24 See Ben S. Bernanke and Federick Mishkin, "Inflation Targeting," op. cit., p. 112.
25 Interestingly, this case can also be made by those who believe in the neutrality of money,
the NAIRU, and rational expectations. They also realize that supply shocks often imply
reductions in real wages and one-time increases in the price level may be the least cost way
to accomplish this even within the confines of their model.
26 The notion of downward nominal wage rigidity is popular among many economists. An
early statement of this proposition and the reasons for it can be found in J.M. Keynes, The
General Theory of Employment, Interest, and Money (Harcourt, Brace and Co. 1936), pp.
12-15. See George Akerlof, William Dickens, and George Perry, "The Macroeconomics of
Low Inflation," Brookings Papers on Economic Activity, vol. 1, 1996, pp. 1-76. In the first
section of this paper, the authors provide a great deal of evidence for a belief in the
downward rigidity of nominal wages.
CRS-14
(5) The Importance of Other Goals
Those who reject changing the ultimate goal of Federal Reserve policy point out
that a central bank has a number of responsibilities that are not necessarily
encompassed in a price stability goal, even if it were to give up its counter-cyclical
role. In particular, a central bank is responsible for the integrity and solvency of the
payments system which includes its role as a lender of last resort to the financial
system. An important reason for establishing the Federal Reserve was to deal with
financial panics that had periodically gripped the United States.27 Our central bank
was to serve as a “lender of last resort” to the financial system in time of trouble to
avert a serious destabilization or even collapse. This important role for the Federal
Reserve might be precluded by a narrowly written law mandating a single goal of
price stability. Similarly, a literal and narrow interpretation of a price stability goal
could needlessly increase the volatility of output and unemployment. Critics would
argue that if this outcome is not desirable, then the goal of full employment should
not be eliminated. They add that most central banks that have made price stability
their sole goal have continued to employ some counter-cyclical policy when they
deem it consistent with long-run price stability. Thus, these foreign central banks do
not practice the pure price stability goal that they are mandated to follow.
Second, monetary policy is not the only policy a nation has. Most governments
have fiscal policies, debt management policies, and even exchange rate policies. In
the United States, responsibility for these policies has not been delegated to the
Federal Reserve. There is no doubt that a goal of price stability for monetary policy
can constrain these other policies. It may make it impossible to achieve certain fiscal
positions, to intervene in the foreign exchange market should this prove necessary,28
or to deal with any attempt by creditors to refuse to renew their holdings of maturing
federal debt or to purchase new debt to finance an existing federal budget deficit.29
27 It was the financial panic of 1907 that set in motion the serious effort to reestablish a
central bank in the United States. The Federal Reserve failed to adequately handle the
financial panic of 1929-1933 that brought about a collapse of the U.S. banking system. In
recent years, Fed watchers have generally applauded its efforts to deal with the failure of the
Continental Illinois Bank in Chicago, the Mexican debt crisis of 1982, and the terrorist
attacks of Sept. 11, 2001, all of which, it was feared, could have had serious destabilizing
consequences for the U.S. financial system and economy.
28 The issue of a permissible range of exchange rate variation becomes less relevant in a
system of flexible exchange rates. Nevertheless, the Federal Reserve could, under its
current mandate, intervene in the foreign exchange market should the dollar come under
extreme selling pressure (or should disorderly markets develop). The possibility of this
happening has been heightened over the years as the United States has moved from the
position of an international creditor to international debtor.
29 There have been historical episodes when the Federal Reserve has had to enter financial
markets to support the price of U.S. government securities when customers could not be
found for the issues that were offered. A price stability goal could compromise this type of
support by the Federal Reserve should it be required.
CRS-15
(6) The Need for Flexibility and Discretion
There has been a long and continuing debate in monetary economics over
whether monetary policy should be conducted by rules that limit Fed decision-
making or by allowing the Fed to exercise discretion.30 The debate over the
desirability of refocusing the Federal Reserve on a price stability goal is often cast
in the terms of this discussion with the new goal being seen as a rule.
Critics argue that all macroeconomic contingencies cannot be spelled out in
advance. Unforeseen circumstances can arise that cannot be accommodated within
the framework of a simple target. For example, would a target have limited the Fed’s
reaction to September 11?31 Since targets cannot accommodate all contingencies, the
judgment of central bankers arguably should prevail in deciding how to conduct
monetary policy. Individuals with this view likely believe that the judgment of Paul
Volcker and Alan Greenspan has produced a better performing economy over the
past two decades than could have been achieved if their hands had been tied by a goal
mandating price stability.32
Critics would also argue that the empirical evidence has been unable to
corroborate the prediction that more discretionary power leads to poorer economic
performance, and has even found the opposite to be true. Several studies have
compared the response of the German economy and the U.S. economy to shocks.
Since the German central bank was presumed to “inspire greater confidence” than the
Federal Reserve because of its history of low inflation, Germany should have
experienced a smaller loss in real output relative to the United States in response to
a given reduction in the inflation rate. Yet the evidence seems to suggest that the
United States has experienced smaller losses.33
Bernanke and Mishkin have argued that in practice the price stability goal does
not impose a rigid rule on central bankers. Rather, the legislation that has been
enacted in foreign countries is more appropriately viewed as a case of “constrained
discretion,” in which central banks are given a goal but have wide latitude in
determining how the goal is met. The central bank in this arrangement is referred to
as having “operational independence.” Bernanke and Mishkin argue that inflation
targeting has many of the advantages of rules and discretion, with few of the
drawbacks. If they are correct, a target may not limit the Fed from acting on its best
30 See CRS Report RL31050, Formulation of Monetary Policy: Rules vs. Discretion, by
Marc Labonte.
31 The Federal Reserve responded to the terrorist attacks of Sept. 11, 2001, by immediately
flooding the financial markets with liquidity with the goal of averting a possible financial
panic.
32 There are those who point out that under the leadership of both Volcker and Greenspan,
the Federal Reserve has pursued a policy of low inflation.
33 See Guy Debelle and Stanley Fischer, "How Independent Should a Central Bank Be?" op.
cit., pp. 202-204; and Adam Posen, "Central Bank Independence and Disinflationary
Credibility : A Missing Link?" Federal Reserve Bank of New York Staff Report, No. 1.
May 1995.
CRS-16
judgment as much as some critics fear.34 But this, then, raises the question of what
purpose a price stability goal would serve if broad discretion is still allowed to
subjugate it to other goals.
Summary
Critics of proposals to make price stability the sole goal of monetary policy
argue that there are other important goals that monetary policy can and should
accomplish. At the extreme, critics argue that money is not neutral and can affect
unemployment over relevant time horizons, and a little inflation makes adjustments
easier. While few economists may agree with these views today, there are many who
would nevertheless agree that the short-term stabilization of the business cycle is a
meaningful goal of monetary policy that should not be sacrificed in the pursuit of
price stability. They would also argue that the Fed’s lender of last resort function is
essential for maintaining a sound and stable financial system. These critics believe
that the economy is too complex for monetary policy to be committed to one simple
goal. It is impossible to foresee every contingency, so discretion is necessary to
allow experts to use their best judgment. They might agree with certain price
stability proponents that “constrained discretion” is the optimal form of monetary
policy, but they would argue that multiple goals are the best way to make certain it
is achieved.
Having presented both the case for and against the proposal to refocus the
ultimate goal of Federal Reserve policy, this report now explores a number of the
technical issues that would likely be raised should Congress decide to adopt a
singular goal of price stability for the Federal Reserve.
Technical Problems With Implementing
a Price Stability Goal35
Beginning in the late 1980s, a number of countries imposed a goal of price
stability on their central banks. Their experience will be used in the exposition to
follow because it demonstrates that the manner in which the legislation is written in
those countries has either compounded or simplified the technical problems noted
34 See Ben S. Bernanke and Frederic S. Mishkin, "Inflation Targeting: A New Framework
for Monetary Policy?" Journal of Economic Perspectives, vol. 11, no. 2 (Spring 1997), pp.
97-116.
35 The following discussion draws heavily from C.A.E. Goodhart, and Jose Vinals,
"Strategy and Tactics of Monetary Policy: Monetary Examples from Europe and the
Antipodes" in Goals, Guidelines, and Constraints Facing Policymakers, op. cit, pp. 139-
187; Guy Debelle and Stanley Fischer, "How Independent Should a Central Bank Be?,"
ibid, pp. 195-221; and Richard Dennis, “Bandwidth, Bandlength, and Inflation Targeting:
Some Observations,” Reserve Bank of New Zealand Bulletin, vol. 60, no. 1, 1997, pp. 22-26.
CRS-17
below. Mishkin and Schmidt-Hebbel identify 17 countries that currently use a goal
of price stability; in addition, the European Central Bank has such a goal.36
The Definition of Price Stability
The Chairman of the Board of Governors of the Federal Reserve, Alan
Greenspan, once said that “price stability exists when inflation is not considered in
household and business decisions.” Although the utility of this definition for policy
formulation can be challenged, it does raise the question of whether price stability
should be defined in general terms or in terms of a quantified numerical target.
Should the former be pursued legislatively, some would suggest using terms “of
reasonable price stability.” Congress might amend the Federal Reserve Reform Act
of 1977 to require the long term growth of monetary and credit aggregates
commensurate with stable prices, thereby dropping goals of potential production,
maximum employment and moderate long term interest rates. However, some would
argue that, in practice, the goal of maintaining stable prices has already dominated
Fed policy under Paul Volcker and Alan Greenspan. If that were the case, making
price stability the sole goal of monetary policy would lead to no changes in policy
unless a numerical target was set.
For those preferring numerical targets, discussions about the definition of price
stability usually center on whether the goal should be defined in terms of keeping the
value of a price index stable or keeping an inflation rate stable. The advantage of the
former, it is claimed, is that economic agents would know with certainty the long run
value of the price level and it would be an immense aid in planning a variety of
economic activities. The disadvantage is that every deviation of the price level from
its legislated value would have to be corrected and this, it is conceded, could lead to
bouts of deflation and introduce a great deal of volatility into the pace of economic
activity and employment.37
Alternatively, Congress could define price stability as a rate of inflation. And
the rate could be a given amount (a so-called point target) or a permissible range, for
example, between zero and 2.5%. The advantage claimed for this alternative is that
bygones would be bygones in the sense that rates of inflation that deviated either
from the point target or the permissible range would not have to be corrected in
subsequent periods. While this would reduce the volatility of economic activity and
36 The experience of some of these countries is examined in detail in CRS Report RL31702,
Price Stability as the Sole Goal of Monetary Policy: The International Experience, by Marc
Labonte and Gail Makinen. See also Frederic Mishkin and Klaus Schmidt-Hebbel, One
Decade of Inflation Targeting in the World: What Do We Know and What Do We Need to
Know?” National Bureau of Economic Research, Working Paper 8397, July 2001.
37 Although this is typically assumed, it need not be the case. A price level target could be
allowed to rise over time, so that prices did not remain constant, but past deviations from
the target would have to be corrected. This would result in a positive rate of inflation in
most years, but years of higher than average inflation would need to be offset by years of
lower than average inflation. However, some of the random price shocks that affect the
overall price level would be likely to cancel each other out over time.
CRS-18
employment over time, it would make uncertain the longer run value of the price
index and this may undermine the putative beneficial effects from this legislation.38
How price stability is defined would appear to be quite crucial to any legislative
effort in this area. All countries that currently impose a price stability goal on their
central banks do so in terms of an inflation range (e.g., Canada 1-3%, New Zealand
0-2%) rather than a price level target.39
Choice of Price Index
An ideal index should, at a minimum, be timely, accurate, not subject to
revisions, and readily understood by the public. These characteristics largely exclude
the two price indexes that come from the series on Gross Domestic Product because
they are subject to numerous revisions. This leaves the CPI which is published
monthly, widely reported in the news media, understood by the public, and not
subject to revisions. As presently formulated, however, it (as well as many other
prices indexes) is subject to a number of problems or biases which may make it a
poor candidate to accurately measure the true price level or the true rate of inflation.
In particular, some economists believe the CPI overstates inflation, so an inflation
target of 0% as measured by the CPI might result in forcing the Federal Reserve to
deflate the economy. Many economists believe this would be harmful to the goal of
maintaining full employment in the presence of sticky prices.40
Some have argued that the use of a target that included volatile commodities
such as food and energy would make monetary policy destabilizing. They argue that
the Fed should instead target a “core inflation” measure which excludes these
commodities. This argument is buttressed by the fact that energy shocks have often
destabilized growth in the past, and making monetary policy react to their effect on
“headline” inflation could compound the destabilization, as discussed below.
Occasionally, the core and headline rates diverge for long periods of time, so a focus
on core could diverge from the price stability goal. For example, headline inflation
exceeded core in five out of six years between 1999 and 2004. Another issue would
be whether to use a price index that includes imported goods or goods and services,
the supply of which are more vulnerable to disruptions and whose price is more
sensitive to changes in the exchange rate than goods and services in general.
38 Dittmar, Gavin, and Kydland demonstrate that uncertainty about the future path of prices
becomes much greater if the central bank continues to respond output volatility under an
inflation target, a possibility that will be discussed below. Robert Dittmar, William Gavin,
Finn Kydland, “Price-Level Uncertainty and Inflation Targeting,” Federal Reserve Bank of
St. Louis Review, July 1999, pp. 23-33.
39 See Mishkin and Schmidt-Hebbel, op. cit.
40 It would be possible to draft legislation in terms of a price index that would be allowed
to trend upward by a given percent per year or an inflation rate per period fixed in terms of
a range whose value would be determined by the upper bound of the estimated bias. For a
discussion of these biases, see Mark Wynne and Frank Sigalla, "A Survey of Measurement
Biases in Price Indexes," Journal of Economic Surveys, vol. 10, no. 1, 1996, pp. 55-89.
CRS-19
All countries that impose a numerical price stability goal on their central banks
use a CPI index. Most use the full CPI. The remainder use a CPI less a number of
items such as food, energy, excise taxes, and home mortgage costs.
Should There Be a Fixed Band Around the Target,
and If So, How Wide Should It Be?
All agree that any target set should be both demanding and credible. Two
approaches have been taken to achieve these ends. One has focused on the selection
of a point target (for either a price level or a rate of inflation). This target is specified
in law with the understanding that some deviations about the point are unavoidable.
However, the precise range of these deviations is left unspecified. A second
approach has involved the specification in law of the permissible range or band in
which the price level or rate of inflation may fluctuate, such as 0% to 2% per year.
It is understood that such a law does not impose on the central bank any obligation
to keep the rate at the mid-point of the range. Any point within the range or band is
equally good from a policy perspective. It is possible to view these two approaches
in the following way. A point target can be thought of as the mean value of an
unspecified range while a fixed range or band can be thought of as a specified range
without a mean value.
Regardless of what approach is taken, a question arises about the width of the
band within which prices might fluctuate. If it were quite wide, the public might
perceive it as not very demanding and this could undermine credibility. But if it were
too narrow, the regime could suffer credibility problems because the band might
frequently be inadvertently breached.41 Achieving the proper balance could be
problematic — estimates of how wide a band would have to be for the central bank
to stay within the band 95% of the time range from 3 to 14 percentage points.42 At
the bottom of this selection is the type of shocks likely to be faced by an economy,
the type of price index that should be used, the lags inherent in monetary policy that
hamper control, and the need to maintain credibility.43 In practice, the widths have
been set to about 2 to 3 percentage points.
There is something more substantial in selecting the width of the band that is
frequently absent from the discussion on the desirability of focusing a central bank
on a single goal of price stability. Most economists currently hold the view that the
harm inflicted on an economy from inflation comes not so much from inflation itself
41 If inflation targeting is interpreted as targeting the forecast of future inflation, it may make
more sense to use a point target than a band. In this case, when the forecast of future
inflation exceeded the point target monetary policy would be tightened and when it was
below the target policy would eased.
42 Richard Dennis, “Bandwith, Bandlength, and Inflation Targeting: Some
Observations,”Reserve Bank of New Zealand Bulletin, vol. 60, no. 1, 1997.
43 Svensson argues that a band is necessary if the central bank wishes to pursue any output
stabilization and still achieve its target. Lars Svensson, Inflation Forecast Targeting:
Implementing and Monitoring Inflation Targets, National Bureau of Economic Research,
Working Paper 5797, Oct. 1996.
CRS-20
as from a variable rate of inflation. If inflation could be fixed at some moderate but
constant percent per year and held there, it might do little damage. Economic
calculations, on the other hand, can be severely handicapped by an inflation rate that
is highly variable. For that reason, the width of the band of permissible variations of
the price level or rate of inflation becomes much more important for it constrains the
possible variations in the price level or rate of inflation and, thus, the damage
inflicted on the efficient operations of the economy.
Would Exceptions Be Allowed?
It is often said that inflation is a monetary phenomenon caused by too many
dollars chasing too few goods. While this is true in the longer run, in the short run
movements in a price index can be due to more than just movements in the supply
of money. Shocks to the turnover rate of money and the supply of goods and services
available to a nation can have an influence on prices and the rate of inflation. Shocks
to supply can come about through changes in such factors as domestic productivity,
unusual weather conditions, and international flows of both trade and capital.
Some of these shocks are random, meaning that their average value over
extended periods of time is zero. The other shocks can be longer lasting in nature
and nonrandom in character. Both types of shocks would bear on the width of a band
that would either be specified if a fixed band were legislated or tolerated if a point
target were legislated.
Demand shocks do not pose a serious problem for a single goal regime focused
on price stability, or, for that matter, a multi-goal regime focused on price stability
and output stabilization. In a demand shock, prices and output move together so
monetary policy can be used to offset both problems at once. For example, after a
fall in consumer confidence, both output and inflation would be expected to fall. In
this situation, expansionary monetary policy would be consistent with both
maintaining price stability and stabilizing output.
It is the supply type shocks that have been highlighted in the literature as
imposing the greatest difficulty to achieving a price stability objective because output
and inflation move in opposite directions in a supply shock. Supply shocks are of
several types. The most commonly mentioned are the OPEC-type oil price shocks.
These are often called terms-of-trade shocks. With a singular goal of price stability,
an OPEC-type shock could force the Federal Reserve to deflate all other prices in
order to keep to the goal. This would lead to a rise in unemployment, especially in
the short run.
Terms-of-trade shocks can also occur for other reasons, especially in response
to international movement of capital. When a country is the recipient of a net inflow
of foreign capital, its exchange rate will appreciate and the price of foreign goods will
fall relative to domestic goods so that a trade deficit will occur. If imported goods
are in the price index, other things constant, the index will decline. Under a constant
price level target the Federal Reserve could be required to inflate the value of
domestic prices (its policy actions could depend on the time period over which it was
required to meet its goal). It might be required to do the same thing to prevent a
CRS-21
negative rate of inflation (i.e., a deflation), if the target were specified in terms of an
inflation band.
Another type of supply shock could occur if the United States decided to add or
substitute a consumption-based tax such as a VAT for the current income-based tax.
This is an option for the fundamental tax reform that the President has proposed. In
some of the countries that impose a numerical price goal on their central bank, such
a tax substitution or an increase in the VAT rate is allowed as an exception to the
goal.
The above discussion raises a general issue about exceptions to the goal. If too
many events that cause prices to change were made exceptions to a price stability
goal, confidence could be undermined and the directive to the central bank would be
significantly diluted. While some exceptions might be desirable, too many might
make the goal of price stability indistinguishable from current practice. One way to
get around the issue of exceptions is to set a fairly wide range or band in which prices
fluctuations are permitted or tolerated. As the width of the band is increased, the
setting of exceptions becomes less important. However, this is done with the
knowledge that if the band is too wide, credibility in the regime is undermined.
Most countries have chosen not to make a list of formal exceptions. In the other
countries, exceptions are made for shocks originating in terms-of-trade changes,
supply disruptions, and changes in excise taxes and interest rates. That such
exceptions are allowed is testimony to the importance of supply shocks to the general
ability of central banks to reach numerical targets.
How Long Would the Central Bank Have to Achieve Its Goal?
To answer this question, one should have some appreciation for what is
involved. The Federal Reserve cannot now rely on a direct and stable relationship
of a monetary aggregate to aggregate demand and the price level or rate of inflation.
Because of this, it manipulates short term market interest rates in an effort to shift
aggregate demand and, ultimately, the price level and rate of inflation. Success
requires technical expertise, good models of the economy (models that capture the
structure of how monetary variables interact with the real economy), some degree of
patience, and, because policy operates with a lag that is long and of variable length,
a long time horizon. Also hampering the success of the operation is that the relevant
interest rates that matter for aggregate demand are the unobservable real or inflation
adjusted rates. Thus, success in meeting a goal of price stability would likely depend
on the time horizon over which the goal would need to be met. The technical
considerations involved seem to support a time horizon longer than one or two
quarters. Among the eight countries specifying a numerical price goal, six specify
a time horizon of at least one year.
Were a numerical target selected that was significantly different from the
inflation rate prevailing at the time, there might also need to be a transition period
between the implementation of the new regime and the realization of the new target.
Otherwise, the sharp shift from the prevailing inflation rate to the targeted rate could
temporarily destabilize the economy.
CRS-22
The nature of the lags inherent in monetary policy and the uncontrollable shocks
to inflation and output raise the question of how literally the Fed should pursue its
goal under an inflation target. Because inflation and output do not always move in
opposite directions — notably in the case of oil shocks — a single-minded
concentration on stabilizing inflation over short periods of time could potentially lead
to a significant degree of volatility in output and unemployment. Arguably, countries
with inflation targets have interpreted the target as an intermediate goal in practice
for that reason. As a result, they try to minimize short-run fluctuations in output
because of its medium-run effect on inflation, even though this may move inflation
further from its target in the short run. But critics could argue that this behavior begs
two questions. First, are the gains in accountability of an inflation target regime lost
if the central bank can always claim to be aiming for the medium run? And more
importantly, what is the purpose of claiming to have a “sole goal” to monetary policy
if, in practice, central banks continue to pursue an unemployment goal in the short
run?
If a Target Were Changed or Missed, What Would Be the
Optimal Time Over Which to Return to the Target?
If the target were missed, it might have no consequences if the amount of
overshooting or undershooting were small and unlikely to persist. There are other
misses, however, that the monetary authorities could decide required corrective
action. If corrective action is required, then the goal should be a smooth transition
back to the target. Sudden and possibly large changes in monetary variables can have
large and disruptive changes to output, employment, interest rates, and the
international exchange value of the dollar. The purpose of a price stability goal
should be to increase the amount of goods and services available to the public, not
cause extreme volatility to real income, employment, and financial markets.
Alternatively, the time period to re-establish the goal should not be so long as
to be meaningless. This would undermine confidence in the new regime. Thus, any
legislation refocusing the Federal Reserve would be expected to pay particular
attention to this issue.
What Incentives Are There, or Should There Be, for the
Federal Reserve to Achieve Its Announced Target? Or, What
Type of Accountability Should There Be?
The adoption of a single goal of price stability defined with some arithmetic
precision would likely increase accountability compared to the current system, where
success is evaluated in subjective terms. But that raises the question of what
appropriate recourse should be taken if the Fed were to miss its target. Failure to do
anything would undermine public confidence in the new regime. Alternatively, to
penalize the governing board for missing a legislative requirement might be self-
defeating if the failure was unavoidable.
The nature of unavoidable shocks to the economy argues for a medium run
target, yet accountability is further weakened if the central bank aims to meet a target
CRS-23
only over the medium run.44 For example, if the Fed is mandated to target inflation
one year in the future, it is difficult to evaluate whether or not it is pursuing a policy
today that will meet the goal.45 It can be punished retroactively for missing the goal
today that it set last year, but it can always claim that it missed its goal for “reasons
beyond its control” and “it will do better next year.” Acknowledging the nature of
unavoidable economic shocks, some inflation target proponents argue that central
banks should be able to change inflation targets on a regular basis. This would
reduce accountability further since missed targets could then be revised away in the
future.
Formal accountability for meeting price stability targets varies considerably
among those countries that have imposed such a goal on their central bankers. In a
few countries, the central bank is required to write an open letter to the finance
minister explaining why the target was missed and what measure have been taken to
rectify the situation. Only New Zealand links the tenure of the head of its central
bank to achieving the inflation target. In most other countries, no explicit sanctions
for missing the target are given. Some have proposed that the salaries and, possibly,
bonuses of the governors might be linked to achieving the price stability goal.
Who Would Set a Target for Price Stability?
In many parliamentary systems of government, this is presented as an important
but unsettled issue. It is important because if the government alone sets the target,
it is thought to underscore the dependent position of the central bank and, it is
argued, may undermine central bank credibility. This may be less important in a
country such as the United States where central bank independence is well
established. The joint setting of the goal is thought to enhance credibility for it
would commit the government to the goal and make it more difficult to be critical of
the central bank in achieving the goal. In practice, most countries have set the goal
jointly.
In the United States, the Constitution vests monetary policy in Congress.
Congress, in turn has granted the Federal Reserve broad operational independence,
but maintained responsibility for oversight and determining the goals of monetary
policy. If Congress chose to set a target for price stability, it would have two general
options. First, it could specify the target in general terms such as directing the
44 A price level target may result in greater accountability than an inflation target because
under the former, policy decisions made today would be strongly influenced by whether the
target was missed in the past, information that is unambiguous and transparent. This is in
contrast to an inflation target, where policy decisions are strongly influenced by forecasts
of the future, which are harder for outsiders to evaluate. See Charles Carlstrom and
Timothy Fuerst, “Monetary Policy Rules and Stability: Inflation Targeting versus Price-
Level Targeting,” Federal Reserve Bank of Cleveland Economic Commentary, Feb. 2002.
45 Lars Svensson argues that if the Fed made its forecasting model public, over time
outsiders could infer from the forecasting errors whether departures from the target were
unavoidable or due to a shortcoming on the Fed’s behalf. See Lars Svensson, Inflation
Forecast Targeting: Implementing and Monitoring Inflation Targets, National Bureau of
Economic Research, Working Paper 5797, Oct. 1996, p. 12.
CRS-24
Federal Reserve to achieve reasonable price stability. This would allow the Federal
Reserve discretion in implementing the law (e.g., choosing the specific numerical
inflation target). (Of course, Fed implementation could be done in consultation with
relevant congressional committees as is now the case in the semi-annual monetary
policy hearings). Second, Congress could set the target range and direct the Federal
Reserve to achieve the goal.
How Could the Government Set a Permanent Target?
This question bears on the credibility of a monetary policy focused on a single
price stability goal. Many writers on this subject have expressed the concern that if
the government could override any prior decision on the target, it would undermine
the confidence of economic agents that price stability would remain the central goal
of monetary policy. For example, there could be opportunistic changes in the
numerical target in order to “pump up” the economy to meet short-term political
objectives.
While it is clear that credibility would be enhanced if it could be ensured that
changes to prior legislation would not take place, this is not possible in the American
political system. The possibility of change is always present. That government has
the power to change laws is, of course, the essence of democracy. It may also be the
essence of good economic policy.
Conclusion
The American model of central banking has distinctive attributes. The U.S.
Congress delegates to a central bank its power to “coin money and regulate the value
thereof.” In doing so, it specifies a variety of goals that monetary policy should
achieve that can be viewed as mutually inconsistent. This lack of goal independence
is, however, arguably superficial since the Federal Reserve has long been allowed to
pick and choose the one or ones on which it will place the greatest emphasis during
any given time period. The Federal Reserve has also been given complete instrument
independence in the sense that no constraints are placed on the monetary powers
available to it to achieve its ends. The exercise of monetary policy is completely at
the discretion of the Federal Reserve.
A growing number of economists argue that a more desirable regime would be
one in which the central bank is directed to achieve a single goal, price stability.
While this regime would restrict the goal independence of the Federal Reserve,
instrument independence would continue as it is now.
Most economists would agree that monetary policy has been highly successful
in the past 20 years. Proponents of a single price stability goal for monetary policy
must contend with the time-honored adage “if it ain’t broke, don’t fix it.” While
proponents are likely to agree that monetary policy has been a success in the last two
decades, they would attribute that success to the Fed’s decision to focus single-
mindedly on price stability. Making price stability the sole goal of monetary policy
would institutionalize this success, preventing any potential departure from this
CRS-25
philosophy under future Fed chairmen and insulating current policy from political
pressure. With the current regime of broad discretion, there is always the potential
for the Fed to spring opportunistic monetary surprises which would lead to an
inflationary bias that would create long-term harm for short-term gain. Furthermore,
they would argue that the Fed pays lip service to goals that are contradictory and
unattainable — thereby avoiding criticism — while focusing on only one goal. This,
they argue further, is a potential threat to the credibility, transparency, and
accountability of monetary policy. In proponents’ eyes, a price stability goal would
lead to an improvement on all three of these fronts, whereas the current multi-goal
regime leads to uncertainty, opacity, and subjectivity. Some proponents are
motivated by a desire to end discretionary policy, while others view a price stability
goal as “constrained discretion.” The latter believe that monetary policy can play a
useful role in reducing the volatility of the business cycle, as long as constraints are
present in the form of a price stability goal to prevent high inflation and monetary
surprises.
Critics would contend that price stability proponents underestimate the
complexity of monetary policy and the broad and varied effects it has on the
economy. While most would acknowledge the salutary effects the Fed’s pursuit of
price stability has produced, they would disagree that this is the only policy goal the
Fed can and should pursue. Instead, they would argue that the Fed has proven in the
last two decades that price stability can go hand-in-hand with a monetary policy that
minimizes the excesses of the business cycle and maintains the soundness of the
financial sector. If a price stability goal is interpreted as precluding the Fed from
pursuing these other two goals, then they would argue that the economy would suffer
as a result. For example, a price stability goal could have limited the Fed’s ability to
ease policy in response to recent oil shocks and the attacks of September 11.
Alternatively, if the price stability goal is interpreted as a regime of “constrained
discretion” which still allows for the stabilization of output, then critics would view
the current multi-goal mandate as more appropriate. Furthermore, they would argue
that the complexity of the economy means that policy must rely on expert judgment,
and the Fed has proven in the past two decades that discretion can be pursued
responsibly.
In conclusion, the price stability goal, while simple, is deceptively so. The long
and variable lags in policy effectiveness and unpredictable nature of shocks to the
economy mean that the Fed’s control over inflation is imprecise and delayed. For
that reason, the Fed could not reasonably be expected to keep inflation on a point
target at all times, should a price stability goal be adopted. This implies
accountability would not be as straightforward as proponents might hope.
Legislation can address this problem explicitly. Possible remedies for the problem
include allowing inflation to stay within a range, targeting core rather than headline
inflation, permitting exceptions when inflation would be allowed to miss its target
under pre-determined circumstances, and targeting forecasted rather than
contemporaneous inflation. But critics would argue that none of these solutions
really solves the inherent complications that makes the price stability goal
impractical.