Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
This PDF is a selection from an out-of-print volume from the National Bureau of Economic Research
Volume Title: Business Cycles, Inflation, and Forecasting, 2nd edition
Volume Author/Editor: Geoffrey H. Moore
Volume Publisher: Ballinger
Volume ISBN: 0-884-10285-8
Volume URL: http://www.nber.org/books/moor83-1
Publication Date: 1983
Chapter Title: An Introduction to International Economic Indicators
Chapter Author: Geoffrey H. Moore
Chapter URL: http://www.nber.org/chapters/c0692
Chapter pages in book: (p. 65 - 92)
thetries
atorshemmdi- Chapter 6ly,chtt An Introduction to International
In- Economic IndicatorsIbeenn in--jied,i theLnce,eco-ime.OCU- For many years a system of leading, coincident, and lagging eco-re a- nomic indicators, first developed in the 1930s by the National Bu-from reau of Economic Research, has been widely used in the Unitednuch States to appraise the state of the business cycle. Since 1961 the cur-
rent monthly figures for these indicators have been published by theU.S. Department of Commerce in Business Conditions Digest. Simi-lar systems have been developed by government or private agencies inCanada, Japan, the United Kingdom, and more recently in manyother countries. Because of differences in content or methodology,however, these independent efforts do not provide comparable mate-rials. In 1973 the NBER Bureau began to develop an internationaleconomic indicator system (IEI) that would provide comparabledata, organized and analyzed in a comparable manner, for a numberof industrial countries. The Center for International Business CycleResearch at Rutgers University in New Jersey has continued thiswork since 1979. The research has demonstrated that such a systemcan be helpful in tracking an international recovery or recession, inrevealing factors that are holding back recovery or leading to reces-sion, in anticipating changes in foreign trade flows, and in providingearly warning of new inflationary trends. The Organization for Eco-nomic Cooperation and Development (OECD) and statistical agenciesin Canada, the United Kingdom, West Germany, France, Italy, Japan,and the United States have cooperated with the NBER and with theRutgers Center in compiling and analyzing the current data for this
Reprinted from International Economic Indicators: A Sourcebook, by Geof-frey H. and Melita H. Moore, Greenwood Press, forthcoming.
65
66 Business Cycles
system of indicators. The practical results of this research programare now available for use.
In this chapter the functions of the indicator system are explainedand the evidence summarized concerning its strength and weaknesses,and it is demonstrated how the system can be used to forecast busi-ness cycles, exports and imports, and inflation rates.
FUNCTIONS OF THE IEI SYSTEM
The first NBER study of business cycle indicators, conducted in1937 by Wesley C. Mitchell and Arthur F. Bums, had as its immedi-ate objective the use of indicators to signal a cyclical revival—that is,the ending of a recession, specifically the ending of the severe reces-sion in the United States that began in spring 1937. When the workwas taken up again after World War II, the objective was broadenedto include signals of a cyclical downturn, and NBER. studies com-pleted in 1950, 1960, and 1966 as well as a Commerce Departmentstudy in 1975 focused on both the beginning and the end of reces-sions. An international system designed along similar lines shouldsignal both peaks and troughs in each of the countries covered as wellas in several countries taken together. In short, an important func-tion of the IEI system is to detect a worldwide recession or recov-ery promptly. The importance of this function is underlined by thefact that international recessions—those in which many countriesparticipate more or less simultaneously—have been more serious thanlocalized recessions. One need only point to 1973-1975 and 1980-1982 to find examples of recessions that were both serious andinternational.
A second, and closely related, function of the indicator system isto measure the scope, severity, and unusual features of an interna-tional recession or recovery while it is in progress. For example, dur-ing the 1975-1976 recovery in the United States it became commonpractice, in reports devoted to the economic outlook, to compare thecurrent recovery with previous recovery periods in this country. Sim-ilar comparisons were made during the 1980 recession. News maga-zines, business journals, annual reports of corporations, governmentreports, and newspapers used this device as a method of appraisal.But few of the publications made such comparisons for other coun-tries, despite their relevance from a world point of view or their valuein the diagnosis of specific problems pertaining to other countries.One of the reasons is that the necessary information is not readilyaccessible. An international economic indicator system should en-
abledate.
AThederriand jabouingpleadjinventheyalsotry, ttoiit.Thpateiindicthese
Fowarnition.fore,inteninflatapprsystecludpolicfiscal
LEA(AND
Theleadieconcyclecipatiandceivedsive ITtion,
rAn Introduction to International Economic Indicators 67
grain able comparisons of this type to be made routinely and kept up todate.
uned A third function is to help appraise prospects for foreign trade.'sses, The leading indicators are sensitive measures of the general state ofbusi- demand. Although many other factors affect the volume of exports
and imports, demand is surely fundamental. A trade deficit can comeabout because of sluggish demand for exports from a country's trad-ing partners while its own demand for imports is growing. Since theleading indicators include such demand-related factors as new orders,
j in inventory change, hiring rates, and profitability, one can expect thatiedi- they would relate to the demand not only for domestic goods butat is, also for foreign products. Leading indicators for an importing coun-eces- try, therefore, should tell us something about how much it is likelyyork to import, or how much its trading partners are likely to export toened it. The international indicator system should therefore help us antici-om- pate changes in the flow of trade to the countries for which leadingaent indicators are available as well as changes in the trade balances amongces- these countries.ould Fourth, a system of international indicators can provide earlywell warning signals of an acceleration or deceleration in the rate of infla-•unc- tion. Inflation is in part a demand phenomenon, and, as noted be-coy- fore, many of the indicators are demand oriented. Inflation is also an• the international phenomenon. All countries experience it, and waves oftries inflation often occur at about the same time in many countries. Anthan appropriate set of international indicators should show how the price
system responds to and feeds back upon the rest of the economy, in-and cluding, of course, those variables that are under some degree of
policy control, such as the money supply, the flow of credit, or them is fiscal deficit.prna-dur- LEADS AND LAGS IN RECOVERYmon AND RECESSIONp theim- The international economic indicator system consists of groups ofaga- leading, coincident, and lagging indicators covering a wide variety ofient economic processes that have been found to be important in businessisa!. cycles. The leading indicators are for the most part measures of anti-un- cipations or new commitments. They have a "look-ahead" qualityilue and are highly sensitive to changes in the economic climate as per-les. ceived in the marketplace. The coincident indicators are comprehen-lily sive measures of economic performance: real GNP, industrial produc-en- tion, employment, unemployment, income, and trade. They are the
L —. — I
68 Business Cycles
measures to which everyone looks to determine whether a nation isserieprosperous or depressed. The lagging indicators are more sluggish inCOurtheir reactions to the economic climate, but they serve a useful pur-
pose by smoothing out and confirming changes in trend that are first Sona
reflected in the leading and coincident indicators. Moreover, their a 1indivery sluggishness can be an asset in cyclical analysis, because when
they do begin to move, or when they move rapidly, they may show m Si
that excesses or imbalances in the economy are developing or subsid- strating. Hence the lagging indicators can (and often do) provide the ear- pub]
liest warnings of all, as when rapid increases in costs of production procNoutstrip price increases and threaten profit margins, thus inhibiting I
ast]new commitments to invest, which are among the leading indicators.A conspectus of the U.S. indicators arranged according to the type each
madiof economic process they represent and the cyclical timing they ex-hibit is in Table 6—1. The compilation for other industrial countries rese
interis designed to represent substantially the same processes arranged in a eachsimilar manner. The degree of success in accomplishing this varies to bfrom one country to another, as shown in Table 6-2. fererThe attempt to duplicate the U.S. system abroad does not meanTithat all countries are thought to be alike or that other indicators forcould not be found that would serve equally well or better. Duplicat- dateing the U. S. system is not an ultimate goal but merely a practicable perfinterim target. The U. S. indicator system has the advantage of being dud'familiar to many users, and both its empirical properties and the eco- devihomic logic on which it was based have been thoroughly investigated turnby many scholars over a long period.' This logic seems applicable to. in aimany countries where free enterprise prevails. Orders placed for ma- leadschinery that is made to order are likely to lead machinery production respiin any market-oriented economy and are likely also to lead the pro- dure,duction of the goods the machinery helps to produce. Similarly, in ideniany enterprise economy, changes in the relations between prices and encecosts influence incentives to expand future output and to make capi- to 1.tal investments. In countries where there are markets for common ous i
stock, stock prices can be expected to be especially sensitive to Asuchanges in profit prospects as well as to changes in interest rates, and indehence to anticipate the effects of these changes on output, invest- 6—3.ment, and employment. TIThe selection of the U.S. indicator list as a target also advances the thoseobjective of providing sets of indicators as comparable as possible eachacross countries. Unless some attention is paid to this, comparisons of ccof cyclical movements in different countries are likely to become ponehopelessly confused. To cite one example, the index of leading mdi- selvecators published by the British Central Statistical Office includes a the i
—I- I -.
rAn Introduction to international Economic indicators 69
)fl S series on interest rates treated invertedly—that is, a rise in rates ish Ifl counted as a depressing factor, and vice versa. This is not an unrea-pur- sonable position to take, but in the U. S. classification interest ratesfirst are treated on a positive basis and are included among the lagging;heir indicators (see Table 6—1). It is recognized that at times a rapid riseThen in such indicators can be interpreted as an adverse development. Ahow straightforward comparison of the U.S. and U. K. leading indexes asbsid- published in each country would run afoul of this difference in
procedure.tion Nevertheless, it is obvious that the system should not be held inLting
a straitjacket, and that adaptations to the way business is done inors. each country and to the particular statistical data available should betype made as more experience with the system accumulates and additional
ex- research is conducted. Perhaps two systems will evolve, one in whichtnes international comparability is strictly maintained, and one in whichin a each country's own data and cyclical response mechanisms are usedanes to best advantage—always avoiding, as far as possible, arbitrary dif-ferences in methodology.iean The acid test of the plan to assemble comparable sets of indicators
.tors for each country according to the U.S. system lies in whether suchCa
- data behave in the way U. S. experience has led one to expect. Toab e perfo this test long-run trends were fitted to each indicator, in-eing cluding those for the United States, cyclical turning points in theeco- deviations from trend were identified, a chronology of growth-cycleite turns for each country was set up to represent the peaks and troughse 0 in aggregate economic activity (after allowance for trend), and the
leads and lags of each trend-adjusted indicator were measured withion respect to these growth-cycle turns. The trend-adjustment proce-
dure, although subject to difficulties of its own, was essential to theidentification of cyclical movements in countries that had experi-enced almost continuous rapid growth through the period from 1948to 1973. Computer programs, carefully monitored to rule out dubi-
to ous results, helped to enhance the objectivity of the data processing.and A summary of the findings on cyclical timing based on compositeest- indexes constructed from each group of indicators, is given in Table
6-3.the The leading indexes constructed from indicators corresponding toible those classified as leading on the basis of U.S. data, lead as a rule in
each of the other countries except France. The coincident indexes,me of course, show virtually no lead or lag, because they and their corn-idi. ponents are used to determine the growth-cycle chronologies them-es a selves. The lagging indexes lag. Significantly, because the grouping of
the indicators is based on U.S. experience only and not on experi-
I
Tab
le 6
—i.
Cro
ss-c
lass
ifica
tion
of U
.S. I
ndic
ator
s by
Eco
nom
ic P
roce
ss a
nd C
yclic
al T
imin
g.
Eco
nom
ic P
roce
ss
Cyc
lical
Tim
ing
Lead
ing
Rou
ghly
Coi
ncid
ent
Lagg
ing
Empl
oym
ent a
ndun
empl
oym
ent
Ave
rage
wor
kwee
k, m
anu-
fact
urin
g.N
ew u
nem
ploy
men
t in-
sura
nce
clai
ms,
inve
rted
Non
farm
em
ploy
men
t.U
nem
ploy
men
t, in
verte
dLo
ng-d
urat
ion
unem
ploy
men
t,in
verte
d
Prod
uctio
n, in
com
e,co
nsum
ptio
n, a
ndtra
de•
New
ord
ers,
cons
umer
goo
dsan
d m
ater
ials
a.
Gro
ss n
atio
nal p
rodu
ct a
Indu
stria
l pro
duct
ion
Pers
onal
inco
me
aM
anuf
actu
ring
and
trade
sale
saFi
xed
capi
tal
inve
stm
ent
Form
atio
n of
bus
ines
sen
terp
rises
Con
tract
s and
ord
ers,
plan
tan
d eq
uipm
ent a
Bui
ldin
g pe
rmits
, hou
sing
Inve
stm
ent e
xpen
ditu
res,
plan
t and
equ
ipm
ent a
Inve
ntor
ies a
nd in
ven-
tory
inve
stm
ent
Cha
nge
in b
usin
ess
inve
ntor
ies a
Bus
ines
s inv
ento
ries a
Pric
es, c
osts
, and
prof
itsIn
dust
rial m
ater
ials
pric
ein
dex
Stoc
k pr
ice
inde
xPr
ofits
aR
atio
, pric
e to
uni
t lab
orco
st, n
onfa
rm
Cha
nge
in o
utpu
t per
man
hour
,m
anuf
actu
ring,
inve
rted
Mon
ey a
nd c
redi
tC
hang
e, c
onsu
mer
inst
allm
ent d
ebta
Com
mer
cial
and
indu
stria
llo
ans o
utst
andi
nga
Ban
k in
tere
st ra
tes,
busi
ness
loan
s
-
C)
r)
0
Ban
k in
tere
st ra
tes,
busi
ness
loan
s
1
Not
es to
Tab
le 6
-1a
a In
cons
tant
pric
es.
Sour
ces:
The
list
and
cla
ssifi
catio
n is
subs
tant
ially
the
sam
e as
that
pre
pare
d in
196
6 an
d pu
blis
hed
in G
eoff
rey
H. M
oore
and
Juliu
s Shi
skin
, Ind
icat
ors o
f Bus
ines
s Exp
ansi
ons a
nd C
ontra
ctio
ns (N
ew Y
ork:
NB
ER, 1
967)
. The
chi
ef m
odifi
catio
n is
that
thos
e se
ries m
arke
d w
ith a
are
conv
erte
d to
con
stan
t pric
es. T
he ti
min
g cl
assi
ficat
ion
for e
ach
serie
s is t
he sa
me
as sh
own
in B
usi-
ness
Con
ditio
ns D
igie
st fo
r all
turn
s (se
e Ta
ble
1, c
olum
ni, i
n an
y re
cent
issu
e), e
xcep
t as f
ollo
ws:
Une
mpl
oym
ent i
s unc
lass
ified
(U) a
t all
turn
s in
BC
D b
ecau
se it
lead
s at p
eaks
and
lags
at t
roug
hs, b
ut h
ere
it is
cla
ssifi
ed ro
ughl
y co
inci
dent
, as i
n th
e19
66lis
t. Fo
ur se
ries t
hat h
ere
are
in c
onst
ant p
rices
are
show
n in
BC
D o
nly
in c
urre
nt p
rices
: cha
nge
in c
onsu
mer
inst
allm
ent d
ebt,
inve
stm
ent e
xpen
ditu
res f
or p
lant
and
equ
ipm
ent,
com
mer
cial
and
indu
stria
l loa
ns o
utst
andi
ng, a
nd c
hang
e in
out
put p
er m
an-
hour
, man
ufac
turin
g, in
verte
d, w
hich
is th
e co
nsta
nt p
rice
equi
vale
nt o
f lab
or c
ost p
er u
nit o
f out
put.
The
cons
tant
pric
e se
ries
are
assi
gned
the
sam
e cl
assi
ficat
ion
as th
e cu
rren
t pric
e se
ries.
0 0 c-s.
Lead
ing
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
2.0
New
une
mpl
oym
ent c
laim
s a3.
0 N
ew o
rder
s, co
nsum
er g
oods
b4.
0 Fo
rmat
ion
of b
usin
ess e
nter
pris
es5.
0 C
ontra
cts a
nd o
rder
s, pl
ant
and
equi
pmen
tb6.
0 B
uild
ing
perm
its, h
ousi
ng7.
0 C
hang
e in
bus
ines
s inv
ento
riesb
8.0
Indu
stria
l mat
eria
ls p
rices
9.0
Stoc
k pr
ice
inde
x10
.0 P
rofit
sb11
.0 R
atio
, pric
e to
labo
r cos
t12
.0 C
hang
e in
con
sum
er d
ebtb
Coi
ncid
ent
13.0
Non
farm
em
ploy
men
t14
.0 U
nem
ploy
men
t rat
ea15
.0 G
ross
nat
iona
l pro
duct
b16
.0 In
dust
rial p
rodu
ctio
n17
.0 P
erso
nal i
ncom
eb18
.0 M
anuf
actu
ring
and
trade
sale
sLa
ggi r
ig19
.0 L
ong-
dura
tion
unen
iplo
ymen
ta20
.0 P
lant
and
equ
ipm
ent i
nves
tmen
tb21
.0 B
usin
ess i
nven
torie
s22
.0 P
rodu
ctiv
ity c
hang
e, n
onfa
rma
23.0
Bus
ines
s loa
ns o
utst
andi
ngh
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
2.0
New
une
mpl
oym
ent c
laim
s a3.
1 N
ew o
rder
s, du
rabl
e go
odsb
na.
New
ord
ers,
equi
pmen
tbN
onre
side
ntia
l bui
ldin
g pe
rmits
bR
esid
entia
l bui
ldin
g pe
rmits
Cha
nge
in b
usin
ess i
nven
torie
sbIn
dust
rial m
ater
ials
pric
esSt
ock
pric
e in
dex
Prof
its'>
Rat
io, p
rice
to la
bor c
ost
Cha
nge
in c
onsu
mer
deb
tb
Nor
ifarm
em
ploy
men
tU
nem
ploy
men
t rat
e a
Gro
ss n
atio
nal e
xpen
ditu
resb
Indu
stria
l pro
duct
ion
Pers
onal
inco
me>
Ret
ail t
rade
b
Long
-dur
atio
n un
empl
oym
enta
Plan
t and
equ
ipm
ent i
nves
tmen
tbB
usin
ess i
nven
torie
s1Pr
oduc
tivity
cha
nge,
man
u fa
ctur
inga
Bus
ines
s loa
ns o
utst
andi
ng'
Inte
rest
rate
s, bu
sine
ss lo
ans
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
2.0
New
une
mpl
oym
ent c
laim
sa3.
1 C
hang
e in
unf
illed
ord
ers C
n.a.
n.a.
6.0
Res
iden
tial b
uild
ing
perm
its7.
0 C
hang
e in
stoc
ksb
8.0
Raw
mat
eria
ls p
rices
9.0
Stoc
k pr
ice
inde
xna
.11
.0 R
atio
, pric
e to
labo
r cos
tn.
a.
Non
farm
em
ploy
men
tR
egis
tere
d un
empl
oyed
, num
bera
Gro
ss d
omes
tic p
rodu
ctio
nbIn
dust
rial p
rodu
ctio
nna
.18
.1 R
etai
l sal
esb
Long
-dur
atio
n un
empl
oym
ent>
'Pl
ant a
nd e
quip
men
t inv
estm
ent1
Bus
ines
s inv
ento
riest
Prod
uctiv
ity c
hang
e, n
onfa
rm>'
Bus
ines
s loa
ns o
utst
andi
ngb
Inte
rest
rate
s, bu
sine
ss lo
ans
Uni
ted
Stat
esU
nite
d K
ingd
omW
est G
erm
any
-..
Tab
le 6
—2.
A C
lass
ifica
tion
and
List
of I
nter
natio
nal E
cono
mic
Indi
cato
rs.
Uni
ted
Stat
esC
anad
aFr
ance
Ca C)
Ca
C,)
5.1
5.2
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.1
19.0
20.0
21.0
22.1
23.0
24.0
13.0
14.1
15.0
16.0
19.0
20.0
21.0
22.0
23.0
24.0
T5Tf
ls-io
smw
-'o.rt
ninT
ru1-
ng—
-—-—
24.0
inte
rest
rate
s, bu
sine
ss lo
ans
1:u
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
- --
----
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
Uni
ted
Stat
esU
nite
d K
ingd
omW
est G
erm
any
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
14.0
15.0
16.0
17.0
18.0
Con
tract
s and
ord
ers,
plan
tan
d eq
uipm
entb
Bui
ldin
g pe
rmits
, hou
sing
Cha
nge
in b
usin
ess i
nven
torie
s'In
dust
rial m
ater
ials
pric
esSt
ock
pric
e in
dex
Prof
itsb
Rat
io, p
rice
to la
bor c
ost
Cha
nge
in c
onsu
mer
deb
tb
Une
mpl
oym
ent r
ate
aG
ross
nat
iona
l pro
duct
"In
dust
rial p
rodu
ctio
nPe
rson
al in
com
e1M
anuf
actu
ring
and
trade
sale
s
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
n. a
.n.
a.
New
com
pani
es re
gist
ered
Bus
ines
s fai
lure
sN
ew o
rder
s, en
gine
erin
g in
dust
ryb
New
ord
ers,
cons
truct
ionb
Hou
sing
star
tsC
hang
e in
stoc
ks b
Bas
icm
ater
ials
pric
esSt
ock
pric
e in
dex
Prof
its b
Rat
io,
pric
e to
labo
r cos
tIn
crea
se in
hire
pur
chas
e de
btb
13.0
Em
ploy
men
t, in
dust
ry
Reg
iste
red
unem
ploy
ed, n
umbe
raG
ross
dom
estic
pro
duct
bIn
dust
rial p
rodu
ctio
nPe
rson
al in
com
eR
etai
l sal
esb
Long
-dur
atio
n un
empl
oym
enta
Plan
t and
equ
ipm
ent e
xpen
ditu
re b
Bus
ines
sin
vent
orie
sbPr
oduc
tivity
cha
nge,
indu
stry
aIn
dust
rial a
nd a
gric
ultu
ral l
oans
outs
tand
ing'
Inte
rest
rate
s, bu
sine
ss lo
ans
(Tab
le 6
—2.
con
tinue
d ou
erle
af)
•1.
1Sh
ort-h
our w
orke
rs5
2.0
Une
mpl
oym
ent a
pplic
ants
an.
a.4.
1 In
solv
ent e
nter
priS
esa
5.1
New
ord
ers,
inve
stm
ent g
oods
b
Res
iden
tial c
onst
ruct
ion
orde
rsb
Inve
ntor
y ch
ange
bB
asic
mat
eria
l pric
esd
Stoc
k pr
ice
inde
xIn
com
e fr
om e
nter
pris
ebR
atio
, pric
e to
labo
r cos
tC
hang
e in
con
sum
er d
ebtb
13.1
Em
ploy
men
t, m
inin
g an
dm
anuf
actu
ring
14.0
Reg
iste
red
unem
ploy
men
t rat
ea15
.0 G
ross
nat
iona
l pro
duct
b16
.0 In
dust
rial p
rodu
ctio
n17
.0 D
ispo
sabl
e in
com
e'18
.1 M
anuf
actu
ring
sale
sb18
.2 R
etai
l tra
deb
n.a.
20.0
Pla
nt a
nd e
quip
men
t exp
endi
ture
b21
.0 B
usin
ess i
nven
torie
s22
.1 P
rodu
ctiv
ity c
hang
e, in
dust
ry a
23.1
Ban
k lo
ans o
utst
andi
ngb
24.1
Inte
rest
rate
s, la
rge
loan
s
'.4 0' c..
Lead
ing
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
2.0
New
une
mpl
oym
ent c
laim
s a3.
0 N
ew o
rder
s, co
nsum
er g
oods
b4.
0 Fo
rmat
ion
of b
usin
ess e
nter
pris
es4.
14.
25.
15.
26.
07.
08.
0 9.0
10.0
11.0
12.0
Coi
ncid
ent
13.0
Non
farm
em
ploy
men
t
6.1
7.0
8.0 9.0
10.1
11.0
12.0
"4 I.
14.0
15.0
16.0
17.0
18.1
19.0
20.0
21.0
22.0
23.1
24.0
Lagg
ing
19.0
Lon
g-du
ratio
n un
empl
oym
enta
20.0
Pla
nt a
nd e
quip
men
t inv
estm
ent'
21.0
Bus
ines
s inv
ento
riesb
22.0
Pro
duct
ivity
cha
nge,
non
farm
a23
.0 B
usin
ess l
oans
out
stan
ding
b
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
Tab
le 6
-2. c
ontin
ued
Uni
ted
Stat
esIta
lyJa
pan
Lead
ing
1.0
Ave
rage
wor
kwee
k, m
anuf
actu
ring
1.1
Mon
thly
hou
rs, i
ndus
try1.
1O
verti
me
wor
ked,
man
ufac
turin
g2.
0 N
ew u
nem
ploy
men
t cla
imsa
3.0
New
ord
ers,
cons
umer
goo
dsb
3.1
n.a. Cha
nge
in u
nfill
ed o
rder
s'n.
a.n.
a.4.
0 Fo
rmat
ion
of b
usin
ess e
nter
pris
es4.
1D
ecla
red
bank
rupt
cies
a4.
1B
usin
ess f
ailu
resa
5.0
Con
tract
s and
ord
ers,
plan
tan
d eq
uipm
ent'
n.a.
5.0
New
ord
ers,
mac
hine
ry a
ndco
nstru
ctio
nb6.
0 B
uild
ing
perm
its, h
ousi
ng7.
0 C
hang
e in
bus
ines
s inv
ento
riesb
6.0
Res
iden
tial b
uild
ing
perm
itsn.
a.6.
17.
0D
wel
ling
units
star
ted
Cha
nge
in in
vent
orie
sb8.
0 In
dust
rial m
ater
ials
pric
esn.
a.8.
0R
aw m
ater
ials
pric
es9.
0 St
ock
pric
e in
dex
9.0
Stoc
k pr
ice
inde
x9.
0St
ock
pric
e in
dex
10.0
Pro
fitsb
n.a.
10.0
Prof
itsb
11.0
Rat
io, p
rice
to la
bor c
ost
12.0
Cha
nge
in c
onsu
mer
deb
tb11
.0R
atio
, pric
e to
labo
r cos
tn.
a.11
.012
.1R
atio
, pric
e to
labo
r cos
tC
hang
e in
con
sum
er a
ndho
usin
g de
bth
Coi
ncid
ent
13.0
Nor
ifarm
em
ploy
men
t13
.0N
onfa
rm e
mpl
oym
ent
13.0
Non
farm
em
ploy
men
t,re
gula
r wor
kers
14.0
Une
mpl
oym
ent r
atea
15.0
Gro
ss n
atio
nal p
rodu
ct"
14.0
15.0
Une
mpl
oym
ent r
atea
Gro
ss d
omes
tic p
rodu
ct'
14.0
15.0
Une
mpl
oym
ent r
atea
Gro
ss n
atio
nal e
xpen
ditu
res
16.0
Indu
stria
l pro
duct
ion
17.0
Per
sona
l inc
ome'
18.0
Man
ufac
turin
g an
d tra
de sa
lesb
16.0
18.1
Indu
stria
l pro
duct
ion
n.a.
Ret
ail s
ales
b
16.0
17.1
18.1
Indu
stria
l pro
duct
ion
Wag
e an
d sa
lary
inco
me
Ret
ail s
ales
'La
ggin
g19
.0 L
ong-
dura
tion
unem
ploy
men
t't20
.0 P
lant
and
equ
ipm
ent i
nves
tmen
t'21
,0 B
usin
ess i
nven
torie
sb20
.121
.0
n.a
Plan
t and
equ
ipm
ent e
xpen
ditu
resb
Bus
ines
s inv
ento
riesb
20.0
21.0
n.a.
Plan
t and
equ
ipm
ent e
xpen
ditu
reb
Bus
ines
s inv
ento
riesb
22.0
Pro
duct
ivity
cha
nge,
non
farm
't23
.0 B
usin
ess l
oans
out
stan
ding
bna
.n.
a.22
.123
.1Pr
oduc
tivity
cha
nge,
indu
stry
aTo
tal l
oans
out
stan
ding
b24
,0 In
tere
st ra
tes,
busi
ness
loan
s24
.0In
tere
st ra
tes,
busi
ness
loan
s24
.1In
tere
st ra
tes,
busi
ness
loan
s
C.,
j
-.,
-_o_
_,._
..___
,.---
.— .—
. ..-
3—
23.0
Bus
ines
s lo
ans
outs
tand
ing
na.
231
Tota
l loa
ns o
utst
andj
ngb
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
24.0
Inte
rest
rate
s, bu
sine
ss lo
ans
24:1
Inte
rest
rate
s, bu
sine
ss lo
ans
Not
esto
Tab
le 6
-2a C
,
a T
reat
edin
verte
dly
in th
e co
mpo
site
inde
xes.
bIn
cons
tant
pric
es.
CC
hang
ein
net
bal
ance
of s
urve
y re
spon
ses.
0
d N
otin
clud
ed in
lead
ing
inde
x fo
r Wes
t Ger
man
y.N
ole:
Ser
ies n
umbe
rs a
re b
ased
on
the
U.S
. lis
t.Th
e di
gits
afte
r the
dec
imal
indi
cate
whe
ther
the
serie
s is s
ubst
antia
lly th
e sa
me
asth
e U
.S. s
erie
s (0)
, or o
nly
roug
hly
equi
vale
nt(1
or 2
).So
urce
: Cen
ter f
or In
tern
atio
nal B
usin
ess
Cyc
le R
esea
rch,
Rut
gers
Uni
vers
ity.
C, 0 0
0
rL.
—E
e-•-
,. —
Tab
le 6
-3.
Cyc
lical
Tim
ing
of E
cono
mic
Indi
cato
rs d
urin
g G
row
th C
ycle
s, S
even
Cou
ntrie
s.
Sou
rce:
Cen
ter
for I
nter
natio
nal B
usin
ess C
ycle
Res
earc
h, R
utge
rs U
nive
rsity
.
A
1!M
edian Lead
(—)
orLa
g(+
), in
Mon
ths
Uni
ted
Stat
es' C
anad
aUnited
Kingdom
West
Germany
Fran
ceIta
lyJa
pan
A v
erag
es:
Six
1948-
1948-
1952-
1950-
1951-
1956-
1953-
Counlries
Seven
198!
1981
1981
1981
1981
1977
1975
Except U.S.
Countries
At g
row
th-c
ycle
pea
ksLa
ggin
g in
dex,
inve
rted
-15
—13
-24
—12
—12
—14
—14
—15
—15
Lead
ing
inde
x—
2—
2—
10—
70
—9
—4
—5
—5
Coi
ncid
ent i
ndex
+10
00
+20
—1
00
Lagg
ing
inde
x+5
+5+7
+2+2
+2+8
+5+5
At g
row
th-c
ycle
trou
ghs
Lagg
ing
inde
x, in
verte
d-1
1—
16-1
9—
18—
16—
7—
14-1
5—
14Le
adin
g in
dex
-2-4
-9—
20
—6
—5
—4
—4
Coi
ncid
ent i
ndex
00
00
0+3
00
0La
ggin
g in
dex
+6+4
+8+4
+5+1
0+8
+6+6
An Introduction to International Economic Indicators 77
ence in the country itself, the sequence of turns among the leading,coincident, and lagging groups in each country corresponds roughlyto the sequence in the United States. The detailed results show thatthis sequence has been repeated at virtually every turn in each coun-try. Moreover, this consistency includes the tendency for the turnsin the lagging indexes to precede opposite turns in the leading in-dexes, corresponding to the economic logic noted previously (com-pare the top line with the second line in each panel of the table).The sequences do not appear to differ systematically from one coun-try to another: hence it is appropriate to average them (see the lasttwo columns of the table). The average sequence is set forth sche-matically in Figure 6-1. Since the growth-cycle chronologies and therecorded leads and lags are based on trend-adjusted data, the risingand falling phases are roughly symmetrical, as are the intervals intowhich they are subdivided by the turns in the indexes.
Nevertheless, it is true that there are wide variations in the lengthsof lead or lag from one cycle to another or from one indicator toanother. The system is neither simple nor mechanical. But the his-torical record is available to help guide current interpretations, andit appears to support the basic hypothesis underlying the scheme,namely, that the U. S. indicator system is broadly applicable overseas.
The composite indexes referred to in Table 6-3 have been com-puted using a method employed for some years by the U. S. Depart-
Figure 6—1. Average Sequence of Cyclical Turns in Three Composite Indexesduring Growth Cycles, Seven Countries.
PEAK,LEAOING-1 PEAK, rPEAK,LAGGING
r TROUGH,COINCIDENT \ COINCIDENT /I I TROUGH COINCIDENTI TROUGH, !
[LAGGING/
TROUGH, LEADING
-6 MOS. 10 MOS. 5 1IIOS. 5 MOS. 14 MOS. 4 NOS.
RECOVERY AND EXPANSION SLOWDOWN AND CONTRACTION
(
21 MOS. 23
GROWTH CYCLE-'44 MOS.
Source: Table 6-3.
4
78 Business Cycles
ment of Commerce. The indexes are constructed so that their trend Figuri
rate of growth during 1966-1976 is equal to that of real GNP for the Coun
country concerned during the same period. The procedure correctsfor the rather haphazard long-run trends that are likely to resultfrom combinations of indicators that,, despite efforts to obtain corn-parabiity, are not precisely the same in the several countries. In addi-tion to the indexes with trend equal to the trend in GNP, indexesare available with the long-run trend eliminated. These depict thegrowth cycles discussed previously. The trend rates of growth in theindividual indicators are of interest in themselves for the purposes ofanalyzing each country's long-run rate of growth.2 Finally, short-run rates of growth in the indexes have been compiled, based onchanges over successive intervals of six months or twelve months.These rates also depict the growth cycles, but they do not dependupon any trend-fitting procedure and hence avoid the uncertaintythat is inevitably attached to bringing such trends up to date.
Although the procedure ensures that the long-run trend in the in-dexes will be approximately the same as the trend in real GNP, thefluctuations in the indexes are larger than those in GNP, partly be-cause most of the components are more sensitive than GNP, partlybecause most of them are monthly rather than quarterly. Anotherreason is that the average month-to-month change (without regardto sign) in each country's industrial production index is used as astandard by which to adjust the month-to-month change in theindex, and industrial production usually undergoes wider swings than SIX I
GNP. The indexes thus provide measures of economic performancebased not on a single indicator but on a group of significant indica-tors that are relatively homogeneous with respect to cyclical timing.As a consequence of both the cyclical homogeneity and the varietyof economic data included, the indexes are relatively free of themonth-to-month irregularities that beset most economic time series.
Figure 6-2 compares the leading and coincident indexes for theUnited States and for six other countries combined (Canada, UnitedKingdom, West Germany, France, Italy, and Japan) during 1972-1982. In the combined index each country's index is weighted by thecountry's GNP in 1970 (expressed in U.S. dollars). Figure 6-3 showsthe growth rates for the same indexes.3 Both charts demonstrate thecapacity of the leading indexes to keep a few months ahead of thebroad measures of economic performance contained in the coinci-dent indexes.
Another illustration of the kind of sequence that can be expected * Spe
by monitoring the leading and the coincident indicators, covering a aCan
longer historical period, may be found in Figure 6-4. Here the Sow
An Introduction to International Economic Indicators 79
endthe
ectssuitom-ddi-'xesthethes ofort-
onths.endnty
in.thebe.
rtly,heriardas athe
hanncelica-ing.ietythe
ries.theted72-i theowsthethenci.
ktedfig ath
Figure 6-2. Leading and Coincident Indexes, United States and Six OtherCountries, 1972-1982.
Specific cycle peak or trough.a Canada, United Kingdom, West Germany, France, Italy, Japan.Source: Center for International Business Cycle Research, Rutgers University.
1970: 100
72 73 74 75 76 77 78 79 80 81 82
80 Business CyclesT
Change).
Figure 6-3. Growth Rates in Leading and Coincident Indexes, United Statesand Six Other Countries, 1972-1982 (Twelve-Month Smoothed Percentage
PERCENT
- z -
LU
C.)oE w
a,
-5
-_c t.s
-o- 5,;(D
I:Note: Arrows indicate rate of change, 1966-1976, in the indexes and in real GNP. g,3Canada, United Kingdom, West Germany, France, Italy, Japan.Source: Center for International Business Cycle Research, Rutgers University.
72 73 74 75 76 77 8078 79 81 82
r'1 :5,
C)
m z -4
C) 0 0 -4 CD a) -4 0• 0) 0 0 0. cc a) -4 0
(aI
ccC
-fl
(CC
CD
Fig
ure
6—4.
Gro
wth
Rat
es in
Lea
ding
Inde
x an
d in
Indu
stria
l Pro
duct
ion,
Wes
tern
Eur
ope,
106
-)98
2 t,T
wel
ve-m
onth
smoo
thed
per
cent
age
chan
ge).
j
a
56 5
7 58
59
60 6
162
63
64 6
5 66
67
68 6
9 70
71
72 7
3 74
75
76 7
7 78
79
80 8
182
aU
nite
d K
ingd
om, l
Nes
t Ger
man
y. F
ranc
e, It
aly.
binc
lude
s in
add
ition
to th
e ab
ove
four
cou
ntrie
s: A
ustr
ia, B
elgi
um, F
inla
nd, G
reec
e, Ir
elan
d, L
uxem
bour
g, T
he N
ethe
rland
s, N
orw
ay, P
ortu
gal.
Spa
in, S
wed
en, a
nd S
witz
erla
nd.
Sou
rces
: Lea
ding
Inde
x, C
ente
r fo
r In
tern
atio
nal B
usin
ess
Cyc
le R
esea
rch;
Indu
stria
l Pro
duct
ion
Inde
x, O
rgan
izat
ion
for
Eco
nom
ic C
oope
ra-
tion
and
Dev
elop
men
t.
Perio
dC
over
ed
Num
ber o
f
Tota
l
Ave
rage
Lea
d (—
)or
Lag
(+),
in M
onth
sLe
ads
Coi
ncid
ence
Lags
United States
1958—1979
10
20
12
-3
Canada
1958—1979
63
211
—1
United Kingdom
1957—1979
53
311
—4
West Germany
1959-1979
8—
210
-4
France
1959—1979
71
513
-3
Italy
1958-1979
91
111
-4
Japan
1958—1979
10
21
13
—2
Tota
l, se
ven
coun
tries
55
12
14
81
—3
Com
posi
te in
dexe
s:W
este
rñEu
rope
a19
56-1
979
91
111
—4
Six
coun
tries
bex
clud
ing
Uni
ted
Stat
es19
72-1
979
40
04
—2
Sour
ce: C
ente
r for
Inte
rnat
iona
l Bus
ines
s Cyc
le R
esea
rch,
Rut
gers
Uni
vers
ity.
A
m0-
f,<
0_c
çcn
o'C
D5
CD
5C
)C
)0.
'1C
DnC
I00.
I
ti'C
D .
CD
P<
0
0 5
C)
C)
P '"
—. C
'•i'
.U
,.ç
,p
-j
Tab
le 6
—4.
Lead
ing
Inde
xes
and
Indu
stria
l Pro
duct
ion,
Sev
en C
ount
ries
(Lea
ds a
nd L
ags
at T
urni
ng P
oint
sin
Rat
es o
f Cha
nge)
.to C
.., (a Pc C,
Not
e: T
he ra
tes o
f cha
nge
in b
oth
the
lead
ing
and
indu
stria
l pro
duct
ion
inde
xes a
re p
erce
ntag
e ch
ange
s ove
r tw
elve
mon
ths,
smoo
thed
, as e
xpla
ined
in th
e te
xt.
aT
hele
adin
g in
dex,
com
pile
d by
the
Cen
ter,
incl
udes
the
Uni
ted
Kin
gdom
, Wes
t Ger
man
y, F
ranc
e an
d Ita
ly, w
eigh
ted
byth
eir r
eal G
NP,
in U
.s. d
olla
rs, i
n 19
70. T
he in
dust
rial p
rodu
ctio
n in
dex,
com
pile
d by
OEC
D, i
nclu
des t
he a
bove
four
coun
tries
plu
s Aus
tria,
Bel
gium
, Lux
embo
urg,
Fin
land
, Gre
ece,
Irel
and,
The
Net
herla
nds,
Nor
way
, Por
tuga
l, Sp
ain,
Swed
en, a
nd S
witz
erla
nd, w
eigh
ted
by g
ross
dom
estic
pro
duct
orig
inat
ing
in in
dust
ry.
bC
anad
a,U
nite
d K
ingd
om, W
est G
erm
any,
Fra
nce,
Ital
y, Ja
pan.
In th
e le
adin
g in
dex,
com
pile
d by
the
Cen
ter,
the
coun
-tri
es a
re w
eigh
ted
by th
eir r
eal G
NP,
in U
.S. d
olla
rs, i
n 19
70. I
n th
e in
dust
rial p
rodu
ctio
n in
dex,
com
pile
d by
the
U.S
.D
epar
tmen
t of C
omm
erce
, the
cou
ntrie
s are
wei
ghte
d by
gro
ss d
omes
tic p
rodu
ct o
rigin
atin
g in
indu
stry
, in
U.S
. dol
lars
,in
197
5.
An Introduction to International Economic Indicators 83
growth rates in the composite leading index for four European coun-tries are compared with the industrial production index for OECDEurope. Industrial production is one of the most widely used eco-nomic indicators around the world, and it is one of the componentsof our coincident indexes. Many countries do not have quarterly esti-mates of gross national product but do have monthly figures on in-dustrial production. Covering manufacturing and mining and oftenincluding utilities and construction activity, the index pertains to asector of the economy that is highly sensitive to business cycles.
For all these reasons it is important to watch the industrial pro-duction index, and to contemplate where it is going next. Here iswhere the leading indexes can be helpful. The comparison in Figure6-4 reveals that the swings in the growth of industrial productionin Europe correspond to those in our leading index with considerablefidelity. Detailed study shows this to be true also of the data for eachcountry. Furthermore, the turns in the leading index growth ratesusually precede those in industrial production and are easier to recog-nize because the leading indexes are smoother and less influenced byerratic movements. As Table 6-4 shows, leads outnumber lags by
2 nearly four to one. The average lead time for all seven countries coy-ered by the table is three months.
The economic basis for this relationship is that the components ofthe leading index represent actions of an anticipatory nature that areespecially sensitive to changes in cost and profit prospects and thestate of demand. New orders for equipment, contracts placed forconstruction work, and housing starts are obvious examples. Thelength of the average workweek is one of the first adjustments madewhen manufacturers detect a shift in demand and reduce or increaseovertime work or the number of part-time employees. By combininga dozen or so of such measures into a single index, the idiosyncrasiesof any single one of them are muted, and the result is a leading indexpossessing the properties described. These properties do not make itan infallible instrument for appraising the outlook in any country.But when a businessman judges the future of his company, he doesnot regard the current state of his orderbooks as an infallible guideeither. That does not stop him from wanting to know what statethey are in. It is the same with the leading indexes. They reflectactions that are likely to affect production a few months hence, andconsequently are useful guides to the state of the market.
RECESSION-RECOVERY PATTERNS
Once an historical chronology of business cycles or growth cycles hasbeen established and a collection of indicators assembled, it becomes
I
84 Business Cycles
possible to compare the current behavior of the indicators with theirpatterns of change during the corresponding stages of previous cy- Figuredes. Although no two business cycles are exactly alike, there arefamily resemblances, and they can be employed systematically toevaluate the current situation and glimpse what lies ahead.
Methods of making such comparisons have been employed formany years in the United States. The Commerce Department's Busi-ness Conditions Digest regularly carries charts of this type, and theRutgers Center for International Business Cycle Research issues pe-riodic reports utilizing the technique.4 The comparisons help one toanticipate what is typical business cycle performance and to observewhether current performance is in line with it or not. Although everybusiness cycle has its own surprises, partly because of policy actionstaken, many developments are not surprising, having occurred manytimes before. Hence a look at what has happened during the laterstages of past business cycles is also a look ahead.
An example of how this works in practice is provided by Figure6-5. The behavior of the U.S. leading and coincident indexes duringthe recession that started from the business cycle peak of July 1981is compared with their average pattern during the seven precedingrecessions. These seven can be divided into two groups: four reces-sions that were relatively sharp (1949, 1954, 1958, 1975) and threethat were mild (1961, 1969, 1980). Such a chart gives one a quick.grasp of how mild or severe a current recession is, how it comparesin duration, whether the usual signs of recovery are developing, andso on.
IMPLICATIONS FOR FOREIGN TRADE
The effect upon the economy of one country of a slowdown in eco-nomic growth in other countries is likely to be most visible in thatcountry's exports. The volume of exports depends upon the trend ofeconomic activity in the country to which the exports go. Ordinarilythis is measured by gross national product or industrial production,both of which are among the coincident indicators. Since the leadingindicators, as we have seen, usually anticipate the movement of thecoincident by several months, the leading indexes for the tradingpartners may also anticipate the movements in exports to them. Anumber of tests of this hypothesis have been made, using the leadingindexes to forecast the rate of change in the volume of trade to andfrom particular countries or groups of countries, and for trade as awhole as well as for various commodity groupings. The results of this resentsresearch show that a substantial proportion of the year-to-year indexe
eco-thatd ofri1y1on,dingthe
ing.Aing
andas athis
year
An Introduction to International Economic Indicators 85
Figure 6-5. Patterns of Current and Previous Recessions.heircy-
' to
forUSi-thepe-
e toerve,eryiOnSanyater
gurering981dingces-hreeLuCk
aresand
U.S. LEADING INDEX (IEI) 1970=100
147AVERAGE,3 MILD RECESSIONS'45 I
147
I /1451437/141
- \ VERAGEI
'39
137
135
131RECESSION
JUNE 82
135CESS IONS 133133 . CURRENT
131
129I
129
-6 —4 -2 0 2 4 6 8 10 12 14 16 18
MONTHS FROM BUSINESS CYCLE PEAKJAN. 81 JUL. SI JAN. 82 JUL.82 JAN.83
STANDING AT JULY 81 PEAK =143.0 1970=100
U.S. COINCIDENT INDEX (lEl)
138
136
'34
1970 100
138
AVERAGE,3 MILD RECESSIONS -,
132
136
130
134
128
I 32
CURRENT /RECESSION-A'
AVERAGEI 4 SEVERE
RECESSIONS
\JUNE8211/
130
126 I - I II I _ I I I_I il—6 -4 —2 0 2 4 6 8 10 12 14
MONTHS FROM BUSINESS CYCLE PEAKJAN.81 JUL.81 JAN.82 JUL.82
STANDING AT JULY 81 PEAK = 137.7 1970=100
128
126
IS
JAN.83
16
Note: The vertical line at zero represents business cycle peak dates. The current and pre-vious business cycles are aligned so that their peaks fall on this line. The horizontal line rep-resents the level of the data at business cycle peaks. The historical data are converted toindexes with their business-cycle peak levels equal to the current business cycle peak level.
86 Business Cycles
changes in trade flows can usually be accounted for in this manner.Figure 6—6 displays the results of one such test, where the percen- Figuitage rates of growth in U.S. exports to all countries, after allowancefor changes in prices, are compared with the prior changes in anexport-weighted leading index for the six countries outside theUnited States. Despite the fact that exports are affected by manyother factors not explicitly taken account of in this simple model,the method tracks the major swings fairly well.
The same method can be employed to forecast the exports of anycountry, developed or developing, that trades with the industrialcountries for which we have leading indexes. We have already ob-tained similar results for the exports of the United Kingdom, WestGermany, Japan, and for a major group of developing countries.Naturally this method by itself has serious limitations, since it ig-nores other factors that influence the quantities of goods exported.Changes in exchange. rates, tariffs and other barriers or incentives totrade, supply conditions, and pricing policies are taken into accountonly insofar as they affect the leading indicators for the importingcountries. Yet, in view of the importance of trade flows and tradebalances in the economic relations among nations, even a modestcontribution to our economic intelligence in appraising trade pros-pects is worthwhile.
IMPLICATIONS FOR INFLATION
Growth cycles are closely associated with the rate of inflation. In-deed, as far as U.S. experience is concerned, declines in the rate ofinflation have been associated with virtually every slowdown or con-traction in real economic growth and have not occurred at othertimes. Both parts of this proposition are important. Declines in therate of inflation have not been as rare as is commonly believed, butthey have occurred only at times of slower economic growth, neverat times of rapid growth. The proposition appears to be true in othercountries as well as in the United States.
The international indicator system is helpful in examining the evi-dence, and so is the concept of the growth cycle described earlier.This distinguishes periods of rapid growth from periods of slowgrowth by reference to a long-run trend. Trend-adjusted data rise aslong as the short-run rate of growth exceeds the long-run rate. Theydecline as long as the short-run rate is less than the long-run rate. NotThe peaks and troughs in trend-adjusted data, therefore, delineate Note:periods of rapid and slow growth relative to the trend rate. ber 1
expo
Billrons of 1972 Dollars
An Introduction to International Economic Indicators 87
Figure 6—6. U.S. Exports and an Export-Oriented Leading Index.
1970 100
240
230
220
210
200
190
180
170
160
150
140
130
120
r
iner.rcefl-'anCen an
theiarlyodel,
anystrialz ob-tWesttries..t ig-rted.es toountrtingradedest)ros-
. In-Le ofcon.theri the
:'but
ever)ther
evi-rlier.1ow
ast'heyrate.eate
¶
I
1973 74 '75 '76 '77 '78 '79 '80 '81 '82
CHANGE IN LEADING INDEX
0 SUBSEQUENT CHANGE INEXPORTS
0%
25
20
15
10
5
0
-5
I I I I I I I I-101972 1973 1974 1975 1976 1977 1978 1979 1980 1981
-73 -74 -75 -76 -77 -78 -79 -80 -81 -82
* Not available,Note: For 1972—1973, the change in the leading index is based on the ratio of the Decem-ber 1972 index to the average index during the preceding twe1ve months, The change inexports is based on totals for 1972 and 1973, Similar calculations are made for other years,
r
88 Business Cycles
For the United States, a chronology of growth cycles based ontrend-adjusted data for the physical volume of aggregate economicactivity is used in Figure 6-7 as a backdrop against which to exam-me the movements in the rate of change in two price indexes. Theindex of industrial materials prices—that is, prices of metals, textiles,rubber, and the like—shows an especially close relation to the growthcycle. Downswings in the rate of change in these prices occurred inevery period of slow growth or recession, and upswings occurred inevery period of rapid growth. Often, as in 1956 and 1959, the down-swings began before the onset of the slow-growth periods. This priceindex is one of the leading indicators in Table 6—1; here it leads notonly the growth cycle but also the rate of change in the consumerprice index (CPI), the bottom line in Figure 6-7. The CPI, whichof course includes the prices of services as well as commodities, re-sponds to the growth cycle as well, but often with a lag of a year ormore. The lags have been so long, especially in recent years, thatsometimes the rate of inflation in the CPI has risen almost through-out the period of slow growth or recession, giving the erroneous im-pression that slow growth had no influence on inflation.
Watching both price indexes together, and bearing in mind theirdifferences in sensitivity and tendency to lag, enables one to see thatgrowth cycles have pervasive influences upon the price structure. Thechange one sees in the consumer price index (as, for example, thedecline in its rate of increase from autumn 1974 to spring 1976) isa lagged response to or reflection of similar developments in corn-modity markets that react far more promptly to changes in demandpressures or supply conditions.
Corresponding data and growth cycle chronologies for the sixother countries covered in the international indicator system suggestthat similar relations are to be found in these countries. Conditionsthat produce rates of economic growth greatly in excess of long-runtrend are conducive to an acceleration of inflation, while conditionsthat make for slow growth or recession are conducive to a reducedrate of inflation or even to deflation. When ordering is brisk andorder backlogs accumulate, sellers have opportunities and incentivesto raise prices, and buyers are less averse to paying them. Costs ofproduction tend to creep up, labor turnover increases, control overefficiency and waste tends to decline. New commitments for invest- C,,
ment are made in an optimistic environment, building up demand forlimited supplies of skilled labor and construction equipment. Creditto build inventories is more readily available and is in greater de-mand, even if higher interest rates must be paid for it, thereby raisingcosts. Labor unions see better opportunities to obtain favorable con-
-t -
--••
----
-- a
' i Dw
— -
- -
J 'S
-- -
'-
wp-
1
Fig
ure
6-7.
Rat
es o
f Cha
nge
in L
eadi
ng In
dex
and
in T
wo
Pric
e(T
wel
ve-M
onth
Sm
ooth
ed P
erce
ntag
e C
hang
e).
Inde
xes
durin
g G
row
th C
ycle
s, U
nite
d S
tate
s. 1
9S5-
'982
—4
Not
e: S
hade
dar
eas
repr
esen
t slo
wdo
wns
in e
cono
mic
gro
wth
as
dete
rmin
ed fr
om tr
end-
adju
sted
mea
sure
s of
agg
rega
te o
utpu
t,in
com
e, s
ales
, and
empl
oym
ent.
Line
s co
nnec
t cor
resp
ondi
ng p
eaks
and
trou
ghs
in th
e ra
tes
of c
hang
e in
the
thre
e in
dexe
s.S
ourc
es: L
eadi
ng In
dex,
Cen
ter
for
Inte
rnat
iona
l Bus
ines
s C
ycle
Res
earc
h, R
utge
rs U
nive
rsity
. Ind
ustr
ial M
ater
ials
Pric
e In
dex:
194
8—19
61.
Dep
artm
ent o
f Com
mer
ce. B
urea
u of
Eco
nom
ic A
naly
sis;
196
2 on
, Jou
rnal
of C
omm
erce
. Con
sum
er P
rice
Inde
x, U
.S. D
epar
tmen
tof
Lab
or,
Bur
eau
of L
abor
Sta
tistic
s.
C
I
55 5
6 57
58
59 6
0 61
62 6
64
65 6
6 67
68
69 7
0 71
72 7
3 74
75
76 7
7 78
79
80 8
182
90 Business Cycles
tract settlements, and their members are more willing to strike to get NOTthem. All these conditions apply to more and more firms and indus-tries and produce upward pressure on more and more prices. Indeed, 1,one of the principal factors underlying a rising rate of inflation in the shipgeneral price level is not just that some prices rise in big jumps but 2.that more prices rise at more frequent intervals. Eban
During periods of slow growth or actual decline in aggregate eco-nomic activity, the opposite conditions prevail. Firms and industriescut back their output, reduce or eliminate overtime, shave costs, give indebigger discounts off list prices, reduce inventories, repay bank debt, methand postpone new investment projects or stretch out existing ones. than,Quit rates decline, indicating that workers feel they must hang on to twe1their jobs, and labor demands for pay raises become more conserva- weatitive. Interest rates drop. As price increases become less widespread COUtS'
and less frequent, and as more price cutting takes place, the rate of flativinflation declines, of th
Since many of the processes sketched here are represented among rate.
the leading and lagging indicators, they can also be employed to remonitor inflation, evaluate its twists and turns, and judge its pros- 4pects. We hope, therefore, that this potential use for the indicators ter sipresented will not be overlooked.
FURTHER RESEARCH AND DEVELOPMENT
A continuing research and development program is essential if thesystem of international economic indicators described is to be usedeffectively and improved. As already noted, the Rutgers Center forInternational Business Cycle Research is conducting such a program.Coverage of the indicator system has already been extended to manyadditional countries, including Sweden, the Netherlands, Belgium,Switzerland, Australia, South Korea, and Taiwan. Attention is beinggiven to speedier access to data, new types of graphic displays, andother analytical tools. Methods of trend-adjusting current data arebeing tested,as well as methods of defining early warning signals ofrecession and recovery. Detecting an international recession prompt-ly and measuring its scope and severity, appraising trade prospects,and getting early warning signals of new inflationary trends are mat-ters of vast consequence to the peoples of the world. If the contentsof this book contributes to these tasks, it will have served its end.
An Introduction to International Economic Indicators 91
get NOTES TO CHAPTER 6
[us-
ed 1. For a list of NBER publications that explain the behavior of and relation-the ships among particular indicators, see Chapter 21.
2. The method of trend-fitting is described in Charlotte Boschan and Walterbu Ebanks, "The Phase-Average Trend: A New Way of Measuring EconomicGrowth," In Proceedings of the Business and Economics Statistics Section,
C0 American Statistical Association, 1978.nes 3. The growth rates are calculated by taking the ratio of the current month'siive index to the average index for the twelve months ending six months ago. Thisbt, method produces a rate of growth (or decline) that is similar to but smootheres. than, the percentage change in the index from the same month a year ago. Theto twelve-month average smooths away erratic factors, such as a strike or unusual
rva- weather, that may have affected the year-ago figure. The method does not, ofead course, smooth out any erratic influences affecting the current month. An alter-
of native method is to take the ratio of the current months's index to the averageof the twelve immediately preceding months, and express the result at an annualrate. This gives a more up-to-date estimate of the growth rate than the one
ong shown in Figure 6-3, but also is more affected by erratic movements in the cur-I to rent month.rOs- 4. Recession-Recovery Watch has been a bimonthly publication of the Cen-tors ter since 1979.
theised
forram.anylum,leingand
pare1s ofnpt-pets,nat-tents
r
su'
ouratebyhascenofJanmeaimme
elal
in i
P01cenforon!:age
RepmenhacpI°