Sie sind auf Seite 1von 8

The Taylor Curve and the

Unemployment-Inflation Tradeoff
BY SATYAJIT CHATTERJEE

I
improve economic performance by
n the past, monetary policy options were
engineering some inflation in order to
described in terms of a tradeoff between the reduce the unemployment rate.
unemployment rate and the inflation rate, But by the early 1970s,
scientific support for a tradeoff between
the so-called Phillips curve. the rate of inflation and the unemploy-
Macroeconomists no longer view the Phillips curve as ment rate had ebbed. As a result of
advances in monetary theory and a
a viable “policy menu” because its use as such is clearer perception of monetary facts,
inconsistent with mainstream macroeconomic theory. economists recognized that a higher
inflation rate could lower the unemploy-
In the late 1970s, John Taylor suggested an alternative ment rate only temporarily. An expan-
set of options for policymakers to consider, one sionary monetary policy sustained over a
long period would, in the end, generate
consistent with macroeconomic theory. These only higher inflation with no reduction
alternative options involve a tradeoff between the in the unemployment rate.
Currently, the conduct of
variability of output and the variability of inflation. monetary policy respects this circum-
Satyajit Chatterjee explains the logic underlying this scribed view of the effectiveness of
monetary policy actions. The challenge
new variability-based policy menu and discusses its for policymakers is to determine how
implications for the conduct of monetary policy. best to carry out monetary policy when
people know that monetary policy
actions have only temporary effects on
the unemployment rate.
In thinking about how the Fed central banks can do has evolved over
One possibility is to refrain
should conduct monetary policy, it’s time and so have prescriptions for
from exploiting the temporary tradeoff
important to know what monetary conducting monetary policy.1 In the
between inflation and unemployment
policy can and cannot accomplish. 1950s and 1960s, monetary policy
and carry out monetary policy with
Without a clear idea of what is within options were formulated in terms of a
some desired long-run inflation target in
the reach of a central bank in terms of tradeoff between the unemployment
mind. For instance, Nobel laureate
controlling economic activity, it’s not rate and the rate of inflation, the so-
possible to make sensible choices called Phillips curve.2 Economists back
regarding monetary policy. then thought that the Fed could sustain 2
British economist A.W. Phillips documented
Scientific consensus on what a lower or higher rate of unemployment an inverse relationship between the rate of
by bringing about a higher or lower rate wage inflation for U.K. workers and the
unemployment rate in the U.K. for the years
of inflation. The implication was that if 1861-1957. In 1960, American economists Paul
Satyajit Chatterjee the unemployment rate associated with Samuelson and Robert Solow drew attention
is a senior economic to the inverse relationship between the rate of
price stability (that is, zero inflation) price inflation in the United States and the
advisor and
economist in the turned out to be too high, the Fed could U.S. unemployment rate, a relationship they
called a “modified Phillips curve.” The
Research Depart- qualifier “modified” has long since disap-
ment of the 1
See the article by Philadelphia Fed peared, and the Phillips curve is now
Philadelphia Fed. President Anthony Santomero in the First generally understood to represent the inverse
Quarter 2002 Business Review for more relationship between price inflation and the
discussion of this point. unemployment rate.

26 Q3 2002 Business Review www.phil.frb.org


Milton Friedman has suggested that the that macroeconomists should be However, because of nonfundamental
Fed should endeavor to keep the money cautious about recommending any factors, the actual unemployment rate
supply growing at a constant rate, one particular policy rule too strongly until can deviate from the natural rate. The
consistent with long-run price stability or more is known about the effects that theory links these deviations to events
a modest level of long-run inflation.3 different combinations of inflation and that cause the actual inflation rate, at
In 1979, economist John Taylor output variability (on the Taylor curve) any given date, to diverge from the
suggested a different possibility.4 Taylor have on a typical household’s standard inflation rate expected for that date in
pointed out that the temporary tradeoff of living. earlier periods.
between inflation and unemployment The reasoning underlying this
was consistent with a permanent A PRIMER ON THE THEORY OF link goes as follows.7 In modern
tradeoff between the variability of THE NATURAL RATE OF industrial economies, it’s common for
inflation and the variability of output UNEMPLOYMENT workers to enter into employment
over time. At some point, policymakers The proposition that the policy contracts in which they agree to supply
face a choice between lowering the choices suggested by the Phillips curve as many hours of work as demanded by
variability of output at the cost of more cannot be sustained is a key implication their employers (within reasonable
variability in the inflation rate or of the theory of the natural rate of unem- limits) for an agreed-upon wage rate or
lowering the variability of the inflation ployment. Since the natural rate theory is salary. This contractually fixed wage
rate at the cost of more variability in Taylor’s point of departure in his search rate or salary reflects, in part, what
output. In his article, Taylor estimated for a sustainable tradeoff between infla- workers and employers expect the
the tradeoff between variability in tion and output, it’s best to begin with a inflation rate to be over the term of the
inflation and output for the U.S. brief description of this theory and its contract. If the inflation rate turns out to
economy.5 This “Taylor curve” displays implications for the Phillips curve. be as expected, employers demand (and
one set of options available to The theory of the natural rate workers supply) the normal level of work
policymakers when monetary policy of unemployment centers on the hours, and the overall unemployment
actions have only temporary effects on determinants of the unemployment rate. rate is close to the natural rate. If the
the unemployment rate. The theory makes a distinction between inflation rate turns out to be higher than
In this article, I will explain the fundamental determinants of the expected, employers buy additional
how policymakers can exploit a unemployment rate and nonfunda- work hours because the price at which
temporary tradeoff between the mental factors. Fundamental determi- they can sell their products is higher
unemployment and inflation rates to nants are factors that change slowly over than expected but the wage they must
consistently achieve particular inflation time, such as demographics, technology, pay for additional hours of work remains
and output variability combinations on laws and regulations, and social mores. contractually fixed. In this case the
the Taylor curve.6 Then I will discuss These fundamental factors determine utilization of labor rises, and the
what lessons about the conduct of the natural rate of unemployment. unemployment rate tends to fall below
monetary policy can be drawn from the the natural rate. Conversely, if the
Taylor curve. Taylor has argued that the inflation rate turns out to be lower than
very shape of the curve reveals the expected, firms lay some workers off
5 Taylor couches his arguments in terms of
general nature of the monetary policy variability of output rather than unemploy-
because the price at which firms can sell
rule that macroeconomists should ment but this difference is not important their products is now lower than
recommend to policymakers. I suggest because the two are closely related. expected but the wage they must pay
Macroeconomists often use a rule of thumb to
translate variability in output to variability in their workers remains contractually
the unemployment rate. The rule of thumb is fixed. In this case, the utilization of labor
3
Friedman stated his views in his 1967 that a 1-percentage-point reduction in the
presidential address to the American unemployment rate goes hand-in-hand with a falls, and the unemployment rate tends
Economic Association. The text of his address 3-percentage-point increase in output. This
appears in his 1968 article. rule of thumb, which appeared in a 1971
article by Arthur Okun, is referred to as
7 There are two variants of the natural rate
4
John Taylor is professor of economics at Okun’s Law. For the sake of comparison with
Stanford University and a renowned scholar the Phillips curve, later in the article I’ll theory. The text describes the variant
on issues concerning monetary policy. couch Taylor’s arguments in terms of the formulated, in part, by Taylor, which forms the
Professor Taylor has served as a member of the variability of the unemployment rate instead basis for Taylor’s subsequent work. Robert
President’s Council of Economic Advisers and of output. Lucas Jr. developed the other variant, which
is currently serving as Undersecretary for focuses on informational frictions rather than
6Economists refer to this tradeoff as a “policy employment contracts. Both variants appear
International Affairs at the U.S. Department
of Treasury. menu.” to be consistent with the evidence.

www.phil.frb.org Business Review Q3 2002 27


to rise above the natural rate.8 average of unemployment rates over interesting aspect of the figure is the
The architects of the natural time is a good proxy for the natural authors’ labeling of the curve. As noted
rate theory took a stand on which unemployment rate and if the average at the bottom of the figure, Samuelson
events caused actual inflation to diverge of inflation rates over time is a good and Solow thought that this curve
from expected inflation. They attributed proxy for the expected inflation rate, the “shows the menu of choice between
these discrepancies to erratic monetary natural rate theory implies that a plot of different degrees of unemployment and
policy. They argued that when the the actual annual rates of inflation and price stability.” The authors’ labeling
monetary authority expands the money unemployment should trace out an suggests that if policymakers find the 5.5
supply unexpectedly, it makes aggregate inverse relationship. According to the percent unemployment rate correspond-
demand for goods and services rise faster theory, a year with a higher-than- ing to price stability (point A on the
than aggregate supply. This excess expected inflation rate should be a year curve) unacceptably high, monetary
demand causes the actual inflation rate
to rise above the expected inflation rate, The natural rate theory can explain why the
which, in turn, motivates firms to
increase the utilization of all factors of
data on inflation and unemployment can take
production, including labor. The the form of a Phillips curve but implies that the
increase in the utilization of labor leads
Phillips curve shows a short-run tradeoff
to a decline in the unemployment rate.
Conversely, when the monetary between inflation and unemployment.
authority unexpectedly contracts the
money supply, aggregate demand falls with an unemployment rate lower than policy actions could lower the unem-
short of aggregate supply. Now excess the natural rate, which, using averages ployment rate to 3 percent at the cost of
supply causes the actual inflation rate to of the two rates over time, implies that a an annual inflation rate of 4.5 percent
fall below the expected inflation rate, year with a higher-than-average (that is, move the economy from point
which, in turn, induces firms to reduce inflation rate should also be a year with A to point B on the curve).
the utilization of labor (and other factors a lower-than-average unemployment Although the natural rate
of production) and causes the unem- rate. In other words, there should be a theory accounts for the existence of a
ployment rate to rise. negative relationship between the Phillips curve in the data, the theory also
The Natural Rate and the inflation and the unemployment rates.9 implies that the Phillips curve shows a
Phillips Curve. Under certain condi- Figure 1 reproduces Paul short-run tradeoff between inflation and
tions, the natural rate theory can explain Samuelson and Robert Solow’s original unemployment, not one that can be
why the data on inflation and unem- estimate of the “modified” U.S. Phillips sustained over the long run. To see why,
ployment can take the form of a Phillips curve for the period 1933-58. The curve suppose that the natural rate of unem-
curve. Recall that the Phillips curve shows a negative relationship between ployment in the economy of Figure 1 is 5
refers to a negative relationship between the average annual rate of inflation and percent, and suppose that policymakers
the inflation rate and the unemploy- the annual unemployment rate. For want to lower the unemployment rate to
ment rate: During years in which the instance, at point B on the curve, an 3 percent. According to the natural rate
inflation rate is high, the unemployment inflation rate of 4.5 percent accompanies theory, the only way in which the
rate tends to be low; during years in an unemployment rate of 3 percent; at monetary authority can sustain an
which the unemployment rate is high, point A, an inflation rate of zero unemployment rate of 3 percent is by
the inflation rate tends to be low. If the accompanies an unemployment rate of generating actual inflation that’s higher
5.5 percent. than expected inflation. Initially, the
From the perspective of the monetary authority may succeed in
8
If employers indexed wage rates or salaries natural rate theory, however, the most generating higher-than-expected
to future inflation outcomes, the incentives inflation and get the unemployment
to demand additional work hours when the
inflation rate is higher than expected and to 9
It’s worth noting that the prediction of the rate below the natural rate. But
reduce work hours when the inflation rate is natural rate theory concerning Phillips curves eventually people will catch on to the
lower than expected would disappear. Thus, holds up when the natural unemployment
Taylor’s variant of the natural rate theory rate and the expected inflation rate are fact that the monetary authority is
leans rather heavily on the fact that most proxied by formulas more sophisticated than generating more than the expected
employers do not appear to index wage-rate simple averages of the rates over time. See, for
or salary contracts to inflation outcomes in instance, Figure 1.5 in Thomas Sargent’s 1999 amount of inflation, and employment
the future. book on U.S. inflation. contracts will begin to take the new

28 Q3 2002 Business Review www.phil.frb.org


THE TAYLOR CURVE: A
FIGURE 1 TRADEOFF CONSISTENT WITH
NATURAL RATE THEORY
Phillips Curve for U.S. If the Phillips curve cannot be
used as a policy tool, is there any
tradeoff between inflation and unem-
ployment that can? Taylor argues that
there is. Like the Phillips curve, this
alternative curve also concerns the
relationship between inflation and
unemployment but focuses on the
variability of inflation and the variability
of unemployment.
To develop these variability-
based combinations, Taylor takes the
view that there are other nonfunda-
mental events, besides erratic changes in
monetary policy, that cause the actual
unemployment rate to deviate from the
natural rate. For instance, if consumers
become unduly pessimistic about their
This figure shows the menu of choice between different degrees of unemployment and price
stability, as roughly estimated from American data from 1933-58. Adapted from Paul A.
prospects for future income and,
Samuelson and Robert Solow, “Analytical Aspects of Anti-Inflation Policy,” American consequently, reduce their spending,
Economic Review (Papers and Proceedings), 50, May 1960, pp. 177-94. Used with permission. the economy can end up in a situation
where aggregate supply will exceed
aggregate demand at prices that firms
expected to prevail. In this situation, the
higher rate of inflation into account. chosen inflation rate, it will be consistent downward pressure on prices will make
Once that discrepancy between actual with the natural rate of unemployment. the actual inflation rate fall below the
and expected inflation disappears, the To summarize, the genesis of expected inflation rate and the utiliza-
unemployment rate will rise again to 5 the Phillips curve lies in studies of the tion of factors of production will fall and
percent. Thus, unless the inflation rate is historical relationship between the the unemployment rate will rise. Con-
continuously different from what people growth rates of wages and prices and the versely, if consumers become unduly
expect, the unemployment rate will unemployment rate. Although the optimistic about prospects for future
return to the natural rate. negative relationship between inflation income and, consequently, increase
The natural rate theory implies and unemployment exists in the their spending substantially, prices will
that for the monetary authority to keep historical data (for that matter, in more be higher than expected and the
the unemployment rate permanently recent data as well), macroeconomists utilization of factors of production will
below the natural rate, it must continu- no longer believe in a long-run policy rise and the unemployment rate will fall.
ally stay ahead of people’s expectations tradeoff between inflation and unem- Given the possibility of such
of rising inflation by generating inflation ployment. The natural rate theory events, the central idea underlying
at an ever-rising rate. Put differently, the persuaded most macroeconomists that Taylor’s variability-based tradeoff is that
only unemployment rate that’s consis- it’s impossible for a monetary authority policymakers can choose the degree to
tent with nonaccelerating or to achieve any unemployment rate which monetary policy is used to buffer
nondecelerating price inflation is the other than the natural rate without the unemployment rate against
natural unemployment rate. This also eventually having either accelerating or nonfundamental disturbances. For
implies that the inflation rate associated decelerating inflation. Although the instance, if consumers become unduly
with the natural rate is a matter of policy Phillips curve describes a genuine pessimistic about the future and the
choice. Within limits, it can be anything pattern in the data, the reason under- actual inflation rate turns out to be
the monetary authority wants it to be, lying the pattern implies it cannot be lower than expected, the monetary
since once people come to expect the viewed as a policy menu. authority can then expand the money

www.phil.frb.org Business Review Q3 2002 29


supply to counteract the higher down toward the previously expected By choosing how aggressively to combat
unemployment that results from the path, the monetary authority has to variability in the inflation rate, the
disinflationary shock. Similarly, if tighten monetary policy more than what monetary authority determines where
consumers become unduly optimistic would be needed to keep the unemploy- on this curve to locate. A policy of
about the future and the actual inflation ment rate at the natural rate. The aggressively combating deviations in the
rate rises faster than expected, the additional monetary restraint raises the inflation rate from a given target path
monetary authority can then contract unemployment rate above the natural will put the economy on a point like B,
the money supply to counteract the rate and, therefore, adds to the where the variability of output is
negative unemployment effect of the variability of the unemployment rate. relatively high but the variability of the
inflationary shock. But it also works to bring the inflation inflation rate is low. Conversely, a less
The important point to note is rate back toward the pre-shock level aggressive policy of combating
that such buffering is not inconsistent and therefore serves to lower the deviations in the inflation rate from a
with the natural rate theory because the variability of the inflation rate. given target path will put the economy
monetary authority is not trying to Furthermore, the more quickly the on a point like A, where the variability
create unexpected inflation or deflation monetary authority aims to bring the in output is low but variability in the
on a sustained basis. On the contrary, inflation rate back down to the pre- inflation rate is relatively high.11
the monetary authority is acting to offset shock level, the more variability it will
variability in unemployment caused by a inflict on the unemployment rate. THE TAYLOR CURVE AND THE
discrepancy between actual and This then is the tradeoff facing CONDUCT OF MONETARY
expected inflation. Various events can policymakers, according to Taylor’s POLICY
cause actual inflation to deviate from theory. To reduce the variability of the Taylor posed the problem of
expected inflation, so there is a scope for inflation rate, the monetary authority the best way to conduct monetary policy
beneficial monetary policy actions that’s must be willing to tolerate increased in the following way.12 Is there any
entirely consistent with the natural rate variability in the unemployment rate. particular point on the Taylor curve
theory. Two ingredients seem necessary for such that’s likely to be acceptable to all
The Unemployment- a tradeoff to exist. First, there must be policymakers?
Inflation Variability Tradeoff. Taylor disturbances (other than erratic Suppose that some
notes that successful buffering of the monetary policy actions) that cause the policymakers are more concerned about
unemployment rate against nonfunda- actual inflation rate to deviate from the variability in the inflation rate and
mental disturbances can dampen the expected inflation rate.10 Second, any others about variability in the
variability of both the inflation and the change in the inflation rate must tend to unemployment rate. In that case, the
unemployment rate. However, he also be persistent. It’s this property of point where Figure 2 curves sharply,
argues that at some point, further persistence that leads to a situation point C, is the variability combination for
reduction in the variability of the where the variability of the inflation rate which there is likely to be consensus.
unemployment rate can come only at can be lowered only at the expense of The reasoning goes as follows.
the expense of more variability in the greater variability in the unemployment Policymakers more concerned about
inflation rate. rate. output variability are not likely to agree
The problem is that a change To summarize, Taylor has on variability combinations that lie to
in the inflation rate tends to persist over developed an inflation and output the northwest of point C because they
time. For instance, if the inflation rate tradeoff consistent with the natural rate would be giving up a lot in terms of
rises because of some unexpected event, theory. His tradeoff involves the
all else remaining the same, the inflation variability of the inflation rate and the 11
The bowed-in shape of the curve indicates
rate will tend to be higher in the future. variability of output, which, recall, is that policymakers face a form of “diminishing
This means that even if the monetary closely related to the variability of the returns.” To bring about a given level of
decline in output variability, policymakers
authority undertakes monetary policy unemployment rate. Figure 2 shows must accept larger and larger amounts of
action to fully offset the unemployment what this tradeoff looks like for the U.S. inflation variability (and vice versa). The
effects of, say, a positive inflation shock, existence of such “diminishing returns” seems
plausible, although the exact reasons for it lie
it’s left facing a path of future inflation 10
Such disturbances could be due to in the character of the macroeconomic model
that’s higher than the path that consumers’ undue optimism or pessimism used by Taylor.
about their future earning prospects. More
everyone expected to prevail prior to the generally, any disturbance that results in 12
This description draws on Taylor’s 1999
shock. To nudge the inflation rate back pricing mistakes by businesses would qualify. article.

30 Q3 2002 Business Review www.phil.frb.org


output variability for meager gains in policymakers are more leery of inflation natural rate theory. But there remains a
inflation stability. Analogously, volatility and others more leery of second, equally important, step: to
policymakers more concerned about volatility in the unemployment rate. determine how the economic welfare of
inflation variability are not likely to This second assumption, the typical household varies across
agree on variability combinations that lie however, is troublesome. In effect, different points on the Taylor curve.
to the southeast of point C because they Taylor treats a policymaker’s preferences
would be giving up a lot in terms of for inflation stability over output stability VARIABILITY AND ECONOMIC
higher inflation variability for meager or vice versa in the same way an econo- WELFARE
gains in output stability. Consequently, mist would treat a person’s innate prefer- At present, not much is known
as long as there is some diversity of views ences for, say, apples over oranges. But about the economic welfare conse-
about the relative demerits of inflation surely preferences about inflation and quences of different variability
and output variability, the combination output variability must derive from some combinations on the Taylor curve.
for which there is likely to be consensus understanding of the relative merits of Furthermore, the connection between
is somewhere in the vicinity of point C. output and inflation stability, an economic welfare and different degrees
Taylor recommended a policy understanding that ultimately must (or of variability of inflation and output is
rule that gives equal weight to stabilizing should!) have some connection to how sufficiently complex that we cannot be
inflation and output. In particular, his output and inflation variability affects certain how economic welfare will
rule recommends that the Fed lower the the welfare of working households. change as we move from a point like A
fed funds rate by half a percentage point This consideration suggests on the Taylor curve to points like B or C.
when real GDP falls below potential that the derivation of the variability Turning first to the economic
GDP by 1 percent and that it raise the tradeoff is an important first step for the welfare effects of inflation variability,
fed funds rate by half a percentage point satisfactory resolution of the question of observe that variability of the inflation
if actual inflation rises above its target which monetary policy rule to adopt. rate will be most harmful if it affects the
path (of 2 percent) by 1 percentage Taylor’s variability tradeoff defines the real value, or purchasing power, of a
point. This policy rule has come to be choices that a monetary authority faces, household’s earnings. During periods of
known as the Taylor rule. Taylor choices that are consistent with the higher-than-expected inflation, growth
recommended this rule, in part, because
it was simple. As he notes in his 1999
article (p. 47), this “[p]olicy rule was FIGURE 2
purposely chosen to be simple. Clearly,
the equal weights on inflation and the The Taylor Curve
GDP gap are an approximation
reflecting the finding that neither
variable should be given negligible
weight.” 13
Taylor’s policy recommenda-
tion hinges on two important assump-
tions. His first assumption is that the
selection of a policy rule (or, equiva-
lently, the selection of a variability
combination on the Taylor curve) will
occur through a democratic process.
Given this assumption, Taylor views the
economist’s job as proposing a policy rule
that’s most likely to command
consensus. His second assumption is that
he takes for granted that some

13 This rule will not put the economy on point Adapted from John B. Taylor, “Estimation and Control of a Macroeconomic Model with
C on the Taylor curve, but it will deliver Rational Expectations,” Econometrica, 47 (5), 1979, pp. 1267-86. Used with permission.
similar variability in inflation and output.

www.phil.frb.org Business Review Q3 2002 31


in nominal compensation will lag growth use Okun’s rule of thumb that a 1- ability of experiencing unemployment. If
in the general level of prices, and real percentage-point increase in the we ignore for now the inflation
compensation will decline (recall that unemployment rate corresponds to a 3- variability effects of the two policies, it
this decline in real compensation is the percentage-point drop in output from follows that all households will benefit
reason firms expand hiring during trend, points A and B on the Taylor under the second policy, relative to the
periods of surprise inflation). Conversely, curve would roughly correspond to first, when the unemployment rate is 4
during periods of lower-than-expected unemployment rate variability of about percent but will lose under the second
inflation, households will experience 1/3 and 1-1/3 percent, respectively. policy, relative to the first, when the
faster growth in real compensation. Fluctuations in the unemploy- unemployment rate is 6 percent.
These fluctuations in real ment rate affect households in two Economic research has shown that the
income inflicted by variability in ways: the probability of job loss for gain will be less than the loss so that,
unexpected inflation cannot be good for employed members and the probability overall, households will be economically
households. But how bothersome of job gain for unemployed members. worse off under the second policy as
variability in inflation is depends on how For instance, during a recession, when compared to the first. However, this
much variability in unexpected inflation the unemployment rate is relatively research has also shown that the
it leads to. The important point here is high, the probability of job loss for predicted loss can be quite small.14 If this
that the high variability of inflation at a employed workers is also relatively high, is the case, the important consideration
point like A in Figure 2 need not imply a and the probability of job gain for in comparing the two policies may turn
high variability of unexpected inflation. unemployed individuals is relatively low. out to be the policies’ effects on inflation
The logic of the Taylor curve suggests Thus, all individuals face a higher risk of variability rather than unemployment
that some of it will come from variability unemployment. Conversely, during an rate, or output, variability.
in expected inflation. But variability in economic expansion, the probability of But this is not, by any means,
expected inflation need not have the job loss for employed workers is relatively the only possibility. The economic
same effect on economic welfare as low, and the probability of job gain for welfare effects of unemployment rate
variability in unexpected inflation. For unemployed workers is relatively high. variability depend importantly on the
one thing, firms and workers have the Hence, all individuals face a lower risk details of how the fluctuations in the
opportunity to alter compensation terms of unemployment. If a policy rule re- unemployment rate affect an indi-
in response to changes in inflation that duces the variability of the unemploy- vidual’s probability of experiencing
are expected to happen. Arguably, the ment rate, it will reduce fluctuations in unemployment. If we drop the assump-
disruption caused by changes in the risk of unemployment. tion that a lower or higher unemploy-
inflation that are expected to happen is To make matters concrete, let’s ment rate implies that all households
likely to be less than the disruptions suppose that the monetary authority is face a proportionately lower or higher
caused by unexpected changes in comparing two policy rules with the probability of experiencing unemploy-
inflation. Therefore, to assess the effects following properties. Under the first ment, the outcome may be different. In
of inflation variability on households, we policy, the unemployment rate is particular, if an increase or decrease in
need information on how the mix predicted to be (almost) constant at, say, the unemployment rate makes the
between expected and unexpected 5 percent, and under the second policy probability of experiencing unemploy-
inflation variability varies as we go from it’s predicted to fluctuate, with equal ment rise or fall proportionately more for
a point like B on the Taylor curve to a probability, between 6 percent and 4 people who are currently jobless, the loss
point like A. At present, this knowledge percent from one period to the next. in economic welfare from following the
is lacking. Observe that the average unemploy- second policy will be larger. Also,
Turning to the economic ment rate is 5 percent under the second unemployment rate variability may not
welfare effects of output variability, policy as well. be the only important consequence of
consider, again, points A and B on the The effects of these two output variability; greater output
Taylor curve. At point A, variability in policies on economic well-being will variability may adversely affect the
output is much lower than at point B. depend on exactly how these policies investment decision of firms and thereby
Why is this relevant? One obvious affect an individual’s probability of reduce the long-term growth rate of
answer is that output variability goes experiencing unemployment. Suppose worker productivity and wages.
hand-in-hand with variability in the that a lower or higher unemployment
unemployment rate, which is of im- rate implies that all households face a 14 For details on this point, see my Business
mediate concern to households. If we proportionately lower or higher prob- Review article.

32 Q3 2002 Business Review www.phil.frb.org


CONCLUSION suggested by the Phillips curve, Taylor’s economic theory. Taylor’s development
An intelligent choice of tradeoff is also concerned with and elucidation of this variability-based
monetary policy requires knowledge unemployment and inflation, but it tradeoff is clearly an important advance
about what monetary policy can or focused on the variability of both the in monetary policy thought. Still, the
cannot accomplish. In the past, unemployment rate and the inflation Taylor curve does not resolve the
monetary policy options were described rate. (Actually, Taylor focused on question of which monetary policy rule
in terms of a tradeoff between the output variability instead of unemploy- to adopt. That decision requires some
unemployment rate and the inflation ment rate variability, but the two are understanding of how the welfare of
rate, the so-called Phillips curve. very closely related.) working households is affected by the
Macroeconomists no longer view the In particular, Taylor argued different combinations (of variability of
Phillips curve as a viable “policy menu” that policymakers face a tradeoff inflation and unemployment rates) on
because its use as such is inconsistent between the variability of inflation and the Taylor curve, an understanding that,
with mainstream macroeconomic the variability of the unemployment at present, is lacking. We hope that
theory. In the late 1970s, John Taylor rate. Unlike the Phillips curve, the future research will fill in this gap in our
suggested an alternative tradeoff for Taylor curve displays a tradeoff knowledge. BR
policymakers to consider. Like that consistent with mainstream macro-

REFERENCES

Chatterjee, Satyajit. “Why Does Phillips, A.W. “The Relation Between Sargent, Thomas J. The Conquest of
Countercyclical Monetary Policy Matter?” Unemployment and the Rate of Change of American Inflation. Princeton, NJ:
Federal Reserve Bank of Philadelphia, Money Wage Rates in the United Kingdom, Princeton University Press, 1999.
Business Review, Second Quarter, 2001. 1861-1957,” Economica, November 1958,
pp. 283-99. Taylor, John B. “Estimation and Control of a
Friedman, Milton. “The Role of Monetary Macroeconomic Model with Rational
Policy,” American Economic Review, 58, Samuelson, Paul A., and Robert M. Solow. Expectations,” Econometrica, 47 (5), 1979,
March 1968, pp. 1-17. “Analytical Aspects of Anti-Inflation pp. 1267-86.
Policy,” American Economic Review (Papers
Okun, Arthur M. “Potential GNP: Its and Proceedings), 50, May 1960, pp. 177- Taylor, John B. “Monetary Policy Guidelines
Measurement and Significance,” in 94. for Inflation and Output Stability,” in
American Statistical Association, Benjamin F. Friedman, ed., Inflation,
Proceedings of the Business and Economics Santomero, Anthony M. “What Monetary Unemployment, and Monetary Policy.
Statistics Section; reprinted in Arthur M. Policy Can and Cannot Do,” Federal Cambridge, MA: MIT Press, 1999.
Okun, The Political Economy of Prosperity. Reserve Bank of Philadelphia, Business
Washington DC: The Brookings Institution, Review, First Quarter, 2002.
1970.

www.phil.frb.org Business Review Q3 2002 33