Test de chow

Carnegie-Rochester Conference Series on Pub/k Policy 29 (1988) 137- 168 North-Holland

I-l-ED STATES:
ROBERT E. LUCAS, JR.* The University of Chicago

TITATIVEREVIE

eltzer*s research career has been so productive and so varied that it would be an act of folly, not friendship, to attempt to review it in a single paper. Yet I do want to talk about his research on this scholarship is carriedoccasion, for research is what Allan's career is mainly about, and I want to do SO in detail, because details are the way out. Accordingly, I will focus my attention mainly on a single paper, one oney: The Evidence from Time Series," published

that has influenced my own thinking on monetary economics a great deal, Meltzer's "The Demand for Meltzer's in the Journal of Political Economy in 1963."Demand for ?4oney" was one in a series of his empirical hich involved joint research with later contributions by studies in monetary economics,

Karl Drunner. It followed earlier work by Latane and others, especially Friedman, and helped to stimulate closely related Laidler and others.' Friedman's (1956) terms, to demonstrate The shared objective of this research program was, in that thedemand for money is a

"highly stable function" of a limited number of variables, to discover the most useful, operational measures of money and these other variables, and (again citing Friedman) to work "toward isolating the of monetary behavior."
numericdl

'constants'

eltzer's paper was the first to estimate an income

4

This like to Leonardo helpful discussion

paper Leiderman,was prepared Bennett and/or Conference.

for

the

November ,+ 1987 Milton Rosen, draft. Sherwin

Carnegie-Rochester Lars Peter and Thomas Sargent I also excellent

Conference. iiansen, Lawrence from Robert

I would
king, for Summers

thank

John Cochrane,

Thomas Cooley, McCallum, P.S.

Friedman,

discussions at the

comments on an earlier Eswar-Prasad

benefittedresearch

a stimulating

provided

8ssisfam?.
and Laidler (1966). and for

‘Two Of course, research, Friedman some of

important this (1959). and especially

sequels other the See

to work earlier

this on

paper money

are

Brunner was of

and M?lfzer closely Friedman

(1963) to and

demand

related (1956)

rther
his

cOnteW.)QrarY
students,. (1987)

contributions(1977) and, more

Laidler

recently,

IvDcCallum and

Coocfriend

the

relevant

background.

0 167 - 2231/88/$3.50

@ 1988 Elsevier Science Publishers B.V. (North-Holland)

(or wealth) elasticity and an interest elasticity SimultaneoUSly
series data

f’rom

time

from a single country (the U.S.).
be to

The objective of the present
to

paper

will

review andreplicate these results,

reconsider

how they

might be interpreted theoretically, and to see how well they stand up to the 25 years of new data that have become available since questions of economic policy. eltzer wroteAn estimated money demand function provides answers to two important The income elasticity, in a setting in which long run real output growth is both fairly predictable andinsensitive to changes in monetary policy, provides the answer to the question: What rate of growth of money is consistent with long run price stability? The interest elasticity is the key parameter needed to answer the question: What are the welfare costs to society of deviations from stability? long run price Purely qualitative answers to these questions, along the lines

of "Inflation ratesare significantly related to money growth rates" or "Inflation reduces welfare" are interesting and useful, perhaps, but surely propositions such as "An Ml growth rate of 3 percent per year will bring about price stability" or "A ten percent annual inflation rate has a social cost equivalent Though answers to the to a 0.5 percent of an of decline in that real income" are more interesting and, if...