Carnegie Mellon University

Research Showcase @ CMU Tepper School of Business

4-1980

The Case for Gradualism in Policies to Reduce Inflation Allan H. Meltzer Carnegie Mellon University, [email protected]

Follow this and additional works at: http://repository.cmu.edu/tepper Part of the Economic Policy Commons, and the Industrial Organization Commons Published In Stabilization Policies: Lessons from the '70s and Implications for the '80s, Proceedings of a Conference, Center for the Study of American Business.

This Conference Proceeding is brought to you for free and open access by Research Showcase @ CMU. It has been accepted for inclusion in Tepper School of Business by an authorized administrator of Research Showcase @ CMU. For more information, please contact [email protected].

STABILIZATION POLICIES: Lessons from the '70s and Implications for the '80s

Proceedings of A Conference

CENTER FOR THE STUDY OF AMERICAN BUSINESS Working Paper No. 53 April 1980

•

THE CASE FOR GRADUALISM IN POLICIES TO REDUCE INFLATION Allan H. Meltzer I n f l a t i o n is usually defined as a sustained rate of increase in a broadly based index of p r i c e s .

Whatever meaning one gives to the im-

precise term "sustained," the past f i f t e e n years seem to meet the standard.

Both the a l l - i t e m consumer price index and the i m p l i c i t GNP

d e f l a t o r have increased in every quarter since l a t e 1965, and neither seems l i k e l y to reach a zero rate of change in the near future. Sustained i n f l a t i o n at the rates of recent years is r a r e , even i f not unique, in the h i s t o r i e s of developed economies.

It seems u s e f u l ,

at a conference summarizing the lessons of the seventies and drawing implications for the eighties to look back on the path we have trave l l e d and to explore the path we might take to restore price

stability.

I s h a l l use the opportunity to discuss some of what has been learned about monetary p o l i c y .

The l i s t is a long one, p a r t i c u l a r l y i f we in-

clude propositions that once were "known" but l a t e r forgotten or rejected in the years of Keynesian orthodoxy, so I shall not attempt to be complete. Any long-term gain from ending i n f l a t i o n depends on a negative r e l a t i o n between i n f l a t i o n and real output.

The most common reason for

suspecting that a gain w i l l occur is the observed association between i n f l a t i o n and changes in r e l a t i v e p r i c e s .

See Cukierman (1979).

The

Dr. Meltzer is Professor of Economics and Social Sciences at CarnegieMellon University. The author is grateful to Alex Cukierman and Jerry L. Jordan for helpful comments on an e a r l i e r d r a f t .

-127-

principal

problem for monetary policy at present is to achieve this

gain by ending i n f l a t i o n at minimum t r a n s i t i o n a l

loss of output.

Every six months, I j o i n with my colleagues on the Shadow Open Market Committee in recommending a policy of pre-announced, gradual,

sustained

reductions in the growth of money as a means of restoring price ity.

stabil-

A c l e a r statement of the reasons for a policy of this kind —

often c a l l e d gradualism - - has not been provided. tially fill

I will

try to par-

that gap and to relate the case for gradualism to some of

the lessons we have learned from recent experience with sustained inflation. The history of recent i n f l a t i o n is surrounded by myths that obscure the origins of the i n f l a t i o n and the reasons for i t s

persistence.

I begin with an account of the o r i g i n and an explanation of p e r s i s t ence.

Much of the case for gradualism depends on the way in which i n -

dividuals form a n t i c i p a t i o n s of the future.

I present one view of

rational expectations, in the sense of Muth (1961), and use this model of expectations to show how Federal Reserve p o l i c y procedures can convert real shocks into permanent changes in the rate of price change. Then I present the case for gradualism in a world in which persistent and t r a n s i t o r y changes in monetary policy cannot be i d e n t i f i e d q u i c k l y .

THE ORIGIN AND PERSISTENCE OF CURRENT INFLATION The most enduring myth about the origins of the current tion is that the i n f l a t i o n started during the Vietnam v/ar.

infla-

According

to a standard version of h i s t o r y , President Johnson rejected the recommendations of his advisers by refusing to choose between "guns and butter."

The President delayed"asking Congress for increased taxes

-128-

(or for smaller expenditures for r e d i s t r i b u t i o n ) d e f i c i t to overstimulate the economy in 1967. has been i n t r a c t a b l e .

and allowed the budget

Since 1967,

inflation

According to some estimates, ten or more years

of recession would be required to eliminate i n f l a t i o n by monetary and fiscal

policies.1 The facts do not correspond to this capsule h i s t o r y .

The rate of

increase of consumer prices reached the 3 to 42 range at least a year before the Vietnam d e f i c i t s .

Spending by the federal government in dol-

lars of constant purchasing power remained 3 to 5% below the 1962 level during most of 1965.

Budget d e f i c i t s and government spending did not

s t a r t the i n f l a t i o n or encourage the Federal Reserve to expand in 1965 or 1966. 1966.

The budget had a small surplus in 1965, and a small d e f i c i t in

The Federal Reserve slowed the growth rate of the monetary base

late in 1966 in a sudden burst of concern about r i s i n g i n f l a t i o n .

The

1967 d e f i c i t of more than $13 b i l l i o n comes a f t e r these f i r s t steps to slow i n f l a t i o n and much too late to explain the s t a r t of the i n f l a t i o n . A surtax was added to the income tax in 1968, so the Vietnam defi c i t proved to be temporary.

By late 1968, the budget again was in

surplus, and the surplus persisted in 1969.

The 1969 surplus of $8.5

b i l l i o n is one of the largest of the past t h i r t y years in real as well as in nominal terms. To sustain the thesis that the Vietnam d e f i c i t s started the current i n f l a t i o n , one must not only ignore the problem of the timing of

^ee Perry (1978) for a more complete statement of t h i s view and for an extreme form of the argument that i n f l a t i o n is i n t r a c t a b l e . Perry's P h i l l i p s curve implies that i t costs S200 b i l l i o n d o l l a r s of real output for each percentage point reduction in the rate of i n f l a tion.

-129-

the s t a r t of i n f l a t i o n , on which I commented e a r l i e r , but must accept the improbable proposition that six quarters of wartime d e f i c i t generated anticipations that were i r r e v e r s i b l e .

Credulity is

strained

further when the 1967 d e f i c i t is expressed in constant d o l l a r s to compare with the d e f i c i t s in e a r l i e r and l a t e r years.

The 1967 d e f i c i t

is

almost identical to the 1958 d e f i c i t when both are expressed in d o l l a r s of the same purchasing power. of sustained i n f l a t i o n .

The 1958 d e f i c i t did not i n i t i a t e years

On the contrary, i n f l a t i o n f e l l

from the 3 to

4% range of 1956-57 to the 1 to 2% range in 1958-59 and to less than

U

by 1961. The 1975 nominal budget d e f i c i t of $70 b i l l i o n i s four times larger than the d e f i c i t s of 1958 and 1967 when the three are expressed in d o l l a r s of comparable purchasing power.

The 1975 d e f i c i t i s not

followed by a balanced budget or a surplus but by sustained d e f i c i t s . Yet, most broad measures of the rate of price change declined in 1976. The GNP d e f l a t o r rose by less than 4.5%, on average, f o r the f i r s t three quarters of the year, and the consumer price index rose by less 2 than 5% f o r the year as a whole. The proximate cause of the s t a r t of the current i n f l a t i o n i s the monetary policy of the early 1960s.

I n f l a t i o n p e r s i s t s because policy

continues to sustain anticipations of future i n f l a t i o n by producing persistent i n f l a t i o n .

Bursts of a n t i - i n f l a t i o n p o l i c y , and announce-

ments of firm commitments to reduce i n f l a t i o n , are not followed by p o l i c i e s that reduce money growth. 2 The decline in the rate of i n f l a t i o n affected more than just food prices as is sometimes claimed. The wholesale price indexes of consumer finished goods rose by less than 2.5:: f o r the year.

-130-

mmm

CHART 1

Rate of Growth of the Monetary Base ( 3 Year Moving Average)

% 9.0

o>

> 8.0

o> c

I

7.0

V. ' >% ro

^

0> c

o

4.0

Q> JZ

3.0 5 o

6

2.0

o o

o

or

l

1.0

1

1 1 1 I

1955

••i

^mmmmmmm

I

1 '

1960

I

I

I

1 » t

1965 Year

I

1 I

1970

1 1 1 1 l

1975

wmmmm*

l

1

M

1980

Chart 1 uses a twelve quarter moving average of the growth of the adjusted monetary base as a measure of the long-term e f f e c t of monetary policy.

Using this measure as an index of the sustained thrust of

monetary p o l i c y , we can divide the monetary history of the past twentyf i v e years into f i v e episodes.

The f i r s t , from 1955 to 1960, has a low

average rate of monetary growth, 1.1%. transition.

The twelve quarter moving average r i s e s s t e a d i l y toward

the 5.5% range.

In the t h i r d period, 1964-71, the growth of the base

remains in the neighborhood of 5.5%. transition,

The second i s a three-year

The fourth period i s a one-year

1972, during which the maintained growth of the base moves

from about 5.5% to 8.5%.

Since 1973, thé moving average of the base

has grown at a maintained rate of about 8.5%. A number of studies, including my own Meltzer (1977), suggest that i n f l a t i o n follows money growth with an average two-year lag.

The

mean of the three-year moving average ending in year t , shown in Chart 1, is an unweighted average centered in year t - 1 .

I f we impose a two-

year lag, i n f l a t i o n in year t+1 i s influenced by the twelve quarter rate of growth of the monetary base ending in year t . sistence,

To measure per-

I have computed the standard deviation of the percentage

rates of change of the consumer price index and the percentage rate of change of money wages for the years 1956-61, 1965-72 and 1974-78 that correspond to the two-year lag of prices behind the maintained growth of the monetary base. 3

The data are shown in Table 1.

S mi f ? r i c e a n d w a 9 e change are one-year averaaes of the consumer r i c e in t l l n w P dex for six-month spans and averag^ hourly before ige? 8 " S 1 X " m 0 n t h s p a n s f r o m B C D - Wa 9 e are not available

-132-

TABLE 1 Mean ( u ) and Standard Deviations Years (t)

Rate of Price Change t+1

Growth of Adjusted Monetary Base in t a

(a) Rate of Wage Change t+1

u

a

u

1955-60

1.1

.18

1.9

1.00

N.A.

1964-71

5.7

.44

4.0

1.26

5.9

1.13

1973-78

8.4

.31

7.5

2.42

8.0

.79

6.4

.85

7.7

.36

Omitting 1974

o

The data show a tendency f o r the standard deviation of the rates of change of money and wages to f a l l

in recent years.

Removing the

e f f e c t s of the o i l shock, by omitting 1974, further reduces the standard deviations.

The standard deviations of the rates of change of

wages and prices are not s t a r t l i n g l y d i f f e r e n t from the standard deviations of the maintained growth of the adjusted base.

The persistence

of rates of price change from year to year appears to be related to the persistence of maintained rates of money growth. To examine further the r e l a t i o n between the persistence of money growth and the persistence of i n f l a t i o n , Table 2 compares the two quarter average rates of growth of base money to the quarterly averages of the rates of change of prices and wages used in Table 1.

As before,

I imposed a two-year lag of rates of price change behind rates of money growth.

The data now suggest that the v a r i a b i l i t y of base money growth

is of approximately the same magnitude as the v a r i a b i l i t y of the rate . 4 or wage change.

The standard deviations of the rate of price change,

The time periods for the base d i f f e r from those in Table 1 because Table 1 has a three-year moving average. I have kept the periods for rates of price and wage change the same as in Table 1.

-133-

however, are not c l o s e l y related to the standard deviations of rates of base money growth.

Short-term v a r i a b i l i t y of the rate of price

change r e f l e c t s more than the v a r i a b i l i t y of monetary growth.

TABLE 2 Mean (y) and Standard Deviations

Period

Two quarter moving average of growth of monetary base

(a)

Standard Deviations

Period

quarterly average rate of change over six-month spans Consumer prices

t

u

CT

1954-59

1.1

0.87

1963-70

5.7

1972-76

8.2

t+2

(a)

u

a

1956-61

1.5

1.61

1.10

1965-72

4.00

0.91

1974-78

8.2

Money wages y

a

1.34

5.9

1.19

2.61

8.0

0.90

The data f o r 1963-70 and 1972-76 include several periods in which i n f l a t i o n was given "highest p r i o r i t y " as a goal of public p o l i c y . Careful inspection of the data shows that periods of slower growth of the base coincide with these announcements in 1966, 1969-70 and 1974-75, but none of these periods of slower growth is long enough to have any marked e f f e c t on the standard deviation of the growth rate of the base. Table 2 shows that the standard deviation of the two quarter moving growth rates is independent of the rate of growth of the base and not very d i f f e r e n t in the three sample periods. The data suggest two reasons for the persistence of i n f l a t i o n and the slow response of i n f l a t i o n to changes in the growth rate of money. F i r s t , short-term rates of price change are r e l a t i v e l y v a r i a b l e , so people have d i f f i c u l t y separating the e f f e c t s of money growth from other influences on short-term price changes.

-134-

This is p a r t i c u l a r l y the

case for recent years, when announced changes in o i l prices have had considerable influence on measured rates of price change and t h e i r variability. not l a s t .

Second, the commitment to a n t i - i n f l a t i o n p o l i c i e s does

People are unwilling to buy long-term contracts based on the

assumption that the slower rate of money growth" w i l l p e r s i s t long enough to reduce the trend rate of i n f l a t i o n .

In the next section, I

o f f e r an explanation of the r e l a t i o n between the v a r i a b i l i t y of money growth and the persistence of

inflation.

THE BASIC INFERENCE PROBLEM5 Each week the Federal Reserve reports the growth rates of various monetary aggregates.

Market participants try to i n f e r the future

course of money growth, i n t e r e s t r a t e s , prices and exchange rates from the announcement.

Their problem, and ours as economists, i s to sepa-

rate t r a n s i t o r y changes in money growth (or other variables) from pers i s t e n t changes.

I c a l l t h i s problem of separating permanent or per-

s i s t e n t changes from ephemeral or transitory changes the basic

infer-

ence problem because i t arises for most economic variables and i s a major problem for people making decisions. To i l l u s t r a t e the problem, suppose that in a given week the announced change in money i s large r e l a t i v e to past changes.

Few ob-

servers w i l l use the observation for a single week to predict the growth path, and fewer s t i l l w i l l predict an equiproportionate change in the rate of i n f l a t i o n .

Let the increased rate of money growth per-

s i s t , for a month or two, and the balance of opinion w i l l s t a r t to

5

This section owes a large debt to Brunner, Cukierman and Meltze

(1979).

-135-

change.

More observers w i l l

i n f e r that there has been a persistent

change in the growth rate of money. The e f f e c t of the f i r s t week's observation on market p r i c e s , interest rates and exchange rates d i f f e r s from the e f f e c t s of a change that is perceived to be permanent.

Although the.change in money is

re-

ported, and therefore is known, the correct inference to be drawn from the information is uncertain because the content of the information is uncertain.

A rational

investor who uses a l l available

information,

must f i r s t decide what he knows; that is to say, he must decide how much of the changes he has observed can be expected to p e r s i s t . This view of the world in which monetary and other p o l i c i e s operate d i f f e r s in an important way from the usual model of rational expectations developed by Lucas (1975) and others.

There, people are

uncertain about whether the changes they observe are the r e s u l t of shocks that change r e l a t i v e prices or shocks that change the absolute price l e v e l ; once information becomes a v a i l a b l e , there is no doubt about i t s meaning. Given the speed with which information becomes a v a i l a b l e , the confusion between aggregative and r e l a t i v e changes cannot be the principal source of confusion.

The main aggregates in our models — money,

debt and d e f i c i t s or GNP, prices and output — are observed within a month or a quarter.

Once they are observed, the confusion between ab-

solute and r e l a t i v e changes disappears. The permanent-transitory confusion does not disappear when data are published.

The p r i n c i p a l

uncertainty that individuals face a r i s e s ,

in this model, from an i n a b i l i t y to properly interpret information, not f

rom lack of information.

People observing the price index must decide

-136-

à n9e

f.L V>9t

»11 j

Of

«fe,

J * Of eh, not A

c°nf-,

us

Pr-oKj

"

&/n

*Pph.

ât

h

can *n be

y

di

°às

S>

n

' rhe

^e

V .

J's xp f

Of

f n

*t

fn tàe

*tt 3

cOur.

sta

^s

str

H

fnf(

tf¡*

.

C

Of

in

'ng s

"ent s^va

Of

'Pe, 'ctt } °ns ¿o,«s not s0 j

for

oh 7

>t

^ce

er.

Of

or

Pe/'

/e

t

•9fj on

" O " V

*f0
9

Of Xt>

'ty

co

»«ftf

tn _

Of

f

«7/ ,

i'n

Cf>at

ftp/

•13?.

y -

*n g

using a d i s t r i b u t e d lag of past observations is an optimal method of forecasting.

The reason is that r e p e t i t i v e observation of an aggregate

are required to learn whether a permanent change has occurred.

If per-

manent changes are frequent, and t r a n s i t o r y changes are infrequent, a change in X is more l i k e l y to be treated as permanent soon a f t e r occurs.

it

At the opposite extreme, transitory changes are frequent and

permanent changes are r a r e , so i t i s optimal to observe a r e l a t i v e l y long series of observations before concluding that a permanent change Z has occurred.

In more technical terms, the larger the r a t i o

the

f a s t e r people c o r r e c t l y i n f e r that a permanent change has occurred; the smaller the r a t i o , the larger is the number of observations required to sustain the inference that a permanent change has occurred. We can put more content into the terms "frequent" or "infrequent" by using the computed standard deviations f o r the two quarter and threeyear moving averages in Tables 1 and 2 to estimate the r e l a t i v e

vari-

ance of permanent and t r a n s i t o r y components and to find the implied length of the lag in reaching rational judgments about permanent shocks. The permanent variance of the growth rate of the monetary base i s set equal to the variance of the three-year growth rates.

The two quarter

moving average growth rates include both permanent and t r a n s i t o r y components.

We assume that permanent and t r a n s i t o r y variances are inde-

pendent and compute the t r a n s i t o r y variance by subtracting the variance of the twelve quarter average from the variance of the two quarter average.

Muth (1960, pp. 302-4) shows that the best (minimum variance)

linear estimator of the permanent value of a variable can be computed from past actual values using the variances of the permanent and

-138-

t r a n s i t o r y components.

For the problem at hand, the c a l c u l a t i o n s

for

the three periods of r e l a t i v e l y constant growth of the monetary base show that the r e l a t i v e variances of the growth rates of the base are: 1955-60

1964-71

1973-78

2 .04

.14

.19

These ratios imply very d i f f e r e n t lags in the adjustment of the expected growth of the base.

In 1955-60, only 55% of the adjustment of

expectations occurs within three years.

The reason i s that the very low

variance around the three-year average growth of base money obscures the change in the maintained rate of growth, when i t occurs.

Rational

individuals i n t e r p r e t most of the permanent change as t r a n s i t o r y and fail

to adjust f u l l y for several years.

In the two remaining samples,

the variance of the permanent component is higher r e l a t i v e to the variance of the t r a n s i t o r y component.

Expectations adjust more quickly; Q

more than 95% of the f u l l adjustment occurs in the f i r s t three years. Expectations of i n f l a t i o n are related to the growth of money that individuals expect to be maintained.

The expected growth of base money

can be reduced permanently only i f the actual growth of base money i s reduced.

The speed of adjustment of expected to actual growth can be

reduced, al so, i f the v a r i a b i l i t y of the growth rate of the base is reduced.

For example, i f the Federal Reserve reduces the variance of the

two quarter growth rate to equal the variance of the twelve quarter

"Transitory" variances are computed from two quarter moving averages, so two quarters are used as one period when computing the lags.

-139-

growth r a t e , 85% of the adjustment of expectations about the permanent growth occurs in the f i r s t year.

Expectations of i n f l a t i o n respond

more rapidly to monetary p o l i c y ; the length of the lag of i n f l a t i o n behind money growth d e c l i n e s . It i s , no doubt, a mistake to use these numbers as precise mates of the expected length of the lag.

esti-

Fortunately, the p r i n c i p a l

implications do not depend on the precision with which we measure the speed of adjustment of expectations.

I f short-term p o l i c i e s are less

v a r i a b l e , the speed of adjustment increases.

Faster adjustment of ex-

pectations lowers the length of time between changes in the growth rate of the monetary base and changes in the expected growth of the base and, therefore, in the expected rate of i n f l a t i o n .

The shorter the

lag, the smaller, c e t e r i s paribus, i s the persistence of A r e l a t e d , but d i s t i n c t ,

inflation.

implication explains why short-term

changes in the growth rate of the base have l i t t l e e f f e c t on maintained inflation.

The larger the t r a n s i t o r y variance of the growth rate of the

base, given the long-term or permanent variance, the longer i s the lag. Short-term reductions i n the growth rate of the base have l i t t l e

effect

on long-term expectations i f the short-term growth of the base is highly v a r i a b l e .

The real costs of reducing i n f l a t i o n are higher,

under these circumstances. r i s i n g unemployment.

The costs take the fom

of recession and

Recession encourages the Federal Reserve to s h i f t

to a policy of monetary expansion thereby reinforcing expectations

that

the maintained average growth rate of the base w i l l not be reduced. Chart 1, above, shows that past periods of a n t i - i n f l a t i o n policy have, in f a c t , had l i t t l e e f f e c t on the maintained growth rate of the base.

-140-

The calculations

in Tables 1 and 2 imply that the lag in the

formation of expectations is shorter how than in the f i f t i e s .

The data

suggest, however, that the reason f o r the shorter lag is the increase in the measured variance of the permanent component, not a reduction in the measured variance of the t r a n s i t o r y component.

THE POLICY PROBLEM The Federal Reserve can reduce the short-term variance of the growth of the monetary base by adopting targets expressed in terms of the base.

Reserves and currency, the uses of the base, are approxi-

mately equal to the sum of reserve bank c r e d i t and international serves.

re-

With f l o a t i n g (or adjustable) exchange r a t e s , the Federal

Reserve can control the two quarter growth rate of the base by c o n t r o l l i n g the stock of Reserve bank c r e d i t .

To control the base the Federal

Reserve need not solve an impossible or even a d i f f i c u l t problem.

All

they must do is control the asset side of t h e i r balance sheet. As i s well-known, the Federal Reserve cannot control both i n t e r est rates and the growth rate of the base.

By specifying short-term

targets in terms of values (or ranges) of the Federal funds r a t e , the Federal Open Market Committee surrenders control of short-term changes in the base.

The problem of separating permanent and t r a n s i t o r y

changes helps to explain how loss of short-term control of the base contributes to persistent movements of the base even i f the dominant shocks in the economy are r e a l , not nominal shocks. To i l l u s t r a t e the problem, I use the three equation, model based on Brunner, Cukierman and Meltzer (1979).

-141-

All

equilibrium variables

are natural logarithms. classical (1) with l t ,

Production or output, y t ,

i s given by a neo-

production function y t = ufc + 6 l t the number of man hours of labor and u t a productivity shock;

6 is the e l a s t i c i t y of output with respect to labor.

Real aggregate

spending is always equal to output, y t , and depends on expected or per manent income, yP, on the real rate of i n t e r e s t and on shocks to aggregate demand, t

The anticipated rate of i n f l a t i o n is the d i f f e r -

v

ence between the logarithms of the price level anticipated f o r next period ( t P t + 1 ) and today's prices ( p t ) . is

The market rate of i n t e r e s t

it. (2)

y t = a+byP

+ c[1t

- (tPt+rPt)]

«t

+

b >0 ; c 0

making^^ora^ustmenis66 " "

^

ten

-142-

"S ° f

9r

°Wth

PateS

0f

mone

*

^

The three equations form an augmented IS-LM model.

The p r i n c i p a l

novelties are the d i s t i n c t i o n between permanent and current income and the introduction of permanent and t r a n s i t o r y shocks. ut,

e t and

The three shocks,

have permanent and t r a n s i t o r y components, but people are

not able to d i s t i n g u i s h the permanent and t r a n s i t o r y components when observing the shocks. 2 CT

up

For example, u t = u£ with known variances

2 and

A UQ»

nonT,

a l d i s t r i b u t i o n s and expected values Eu^ and EAU£

equal to zero. Substituting eq. (1) Into (2) and (3) and solving f o r 1 t reduces the system to two equilibrium r e l a t i o n s .

The money market equilibrium

or LM, in eq. (-4) and the IS curve, eq. (5) r e l a t e i t to the three shocks, to the p r i c e level and to other v a r i a b l e s .

For the current

a n a l y s i s , I t r e a t y£ and l t as given and independent of the shocks. 1 0 (4)

6 i t - B + Tt - pt -

Yut

+ 9e t - Y * l t - ( l - Y ) y P -

(5)

c 1 t = c ( t p t + 1 - p t ) + u t - e t + 6 l t - b yP - a

a

During most of i t s existence, the Federal Reserve used the market i n t e r e s t rate (or some surrogate l i k e the level of free reserves) as the operating target.

Suppose the Federal Reserve sets the target

in-

terest rate at i Q and supplies or absorbs base money to keep i* = i t 0«

no7Q\ k

solution is given in Brunner, Cukierman and Meltzer

io..?H J î ï ' f £ n g t h e , 1 a l ? o r m a r k e t equations. The additional d e t a i l Eïii.« Î e î! t h e u c o n c 1 u s 1 0 n s ° f ^ i s discussion. The o r i n c i p a l d i f ferences that have been neglected are the dependence of yP on the exand the v'aîues o? tï hheI rre aJ l \ r h*o c V ^ of l ^ o n the acîual ks î L°! ? The reader who i s disturbed &y the p a r t i a l s bst t te shiïîl S l " Permanent and actual values of shocks real t h a t from Î o'

a n d e S p o n

x H ; " s o f IS

F r the anal si ° y 's k a n d LM t 0 t h e sh

that follows what matters is o c s cause 1 t to d i f f e r

-143-

The stock of base money B + interest rate at i

0,

pends on the real

shocks.

(6)

changes only as required to maintain the

which i s to say that the stock of money now de-

Yt * * ( e t , u t )

Equations (4) and (5) are shown as s o l i d l i n e s in Figure 1. slope of LM from eq. (4) is p o s i t i v e in the i , p plane. IS is -1.

The price level

is p .

The

The slope of

The p o l i c y of f i x i n g i n t e r e s t

rates,

temporarily at i , makes the i n t e r e s t rate pre-determined at i .

Mone-

tary policy keeps the i n t e r e s t rate constant by changing money.

When-

ever there are real shocks to productivity or to spending and the demand for money, the Federal Reserve changes the stock of money enough to hold i n t e r e s t rates fixed u n t i l

i t decides that the shock is per-

manent. Consider the e f f e c t of a negative productivity shock, du^ < 0. From (4) and (5) we compute the e l a s t i c i t i e s

as* -

a n d

'LM

^

4