Professional Documents
Culture Documents
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The Econometric Society is collaborating with JSTOR to digitize, preserve and extend access to Econometrica.
http://www.jstor.org
RUTLEDGE
OF
VINING
37
38
RUTLEDGE
1. REGIONAL
ECONOMIES
VINING
IN INTRANATIONAL
BUSINESS-CYCLE
ANALYSIS
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
39
40
RUTLEDGE
VINING
marked.It is also not true in generalthat "export" productshave nation-widemarkets,but it is the case in manyinstances,and at least the
structureof these regionalinterlocksmay be investigated.Finally, as
was indicated,a state is not what we should regardas an "economic
region" and will generallyinclude parts or the whole of several integratedtrade areas; and the behavior in the businesscycle of state figures will be a sort of average of the behaviorof the figuresof its componentregions.States comprisedofeconomicallysimilarregionsshould
show a similarityof behavior in the business cycle. Statisticsdescriptive of industriallocation by states should indicate how a particular
state mightbe expectedto respondin a bu.sinesscycle and should help
explainobservablegeographicalpatternsofbusiness-cycleresponse.
2. CERTAIN
RELEVANT
FEATURES
OF INDUSTRIAL
LOCATION
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
41
BEHAVIOR
E.
E
of localizationis computedas follows:
The coefficient
(3)
WeightedDeviation
fromUnity
(2)
Location
Factor
(1)
ela
ela
**
..............
Ea
E-
ei
**
.
.
...............................................
.
ena
en
of LocalizationCoefficient
El
E
. . . . .
01
el
el
Ea
1 -
ei
ela
Ea
ee
en
en
E-
E
ea
ena
E
nI
e;
eia
E
Ea
The algebraicsum is obviouslyzero,indicatingthe equality of the positiveand
negativedeviations.The absolute sum, disregardingsigns,varies fromzero to
two. If in each regionthe two percentagesare identical-i.e., ifthe industrywere
distributed-thenthe sum of the deviationsis zero. If all of
perfectlyuniformly
the givenindustryis located in a singleregion,and ifregionsand industriesare so
definedas to be ofnegligiblesize relativeto the total economy,thenthe absolute
sum of the deviationswillapproachtwo as a limit.
42
RUTLEDGE
VINING
BEHAVIOR
43
thesefeaturesofindustrial
location.For thispurpose,we arrayedfor
each industrial
classification
thestatefigures
forthepercentage
oftotal
employment
foundinthatindustry.6
For eachofthesearraystherange
limitsofeach quintileintervalwerelocated,andtheserangelimitsand
the medianfigurewerethendividedby theappropriate
nationalpercentage.The resultsofthisprocedure
are showninTable 1. This table
providescertainpointsonthearrayofstatelocationfactors
foreachindustrialclassification,
and theselatterarearrayedfromtopto bottom,
in theorderofwhatappearsto be thedegreeoflocalization.The table
showsapproximately
that whichFlorence'scoefficient
of localization
showsin a muchmorecondensedform,and forits awkwardbulkiness
it maypossiblyoffer
compensation
in theformofan enhancedvisual
impression
ofthecharacterofindustrial
localization.
For ourpurposes,thesedata are fraught
withshortcomings,
and it
maybe instructive
to referto the implications
of a fewof thesedeIf the subareasunderstudywereapproximately
of equal
ficiencies.
"economicsize" and smallrelativeto thetotalarea,ifthestandardsof
livingand consumer
tasteswereapproximately
uniform
as betweenregions,if industrial
classifications
werefineenoughdrawnin orderto
provideinternaluniformity,
thenwe mightpossiblyrationalizean expectationfora certaingeneralpatternto be exhibitedby thecolumns
in thistable,theindustries
beingarrayedin orderofdegreeoflocalization. The marketareas of different
productswould obviouslyshow
widedifferences.
Supposeweshouldmeasurethetypicalmarketarea of
a productby someaveragedistancebetweenthe pointat whichthe
valueis addedand thepointoffinalconsumption.
A frequency
districould
butionofthesevariates,typicaldistanceforeachdefined
product,
be constructed,
butit wouldbe difficult
to anticipatethegeneralshape
ofthisdistribution.
If it wereU-shapedit wouldmeanthata relatively
largeproportion
ofproducts
are soldat a pointcloseto theirrespective
at intermediate
dispointsofproduction,
a relatively
smallproportion
sold
tancesfromthesepoints,and anotherrelatively
largeproportion
in all cornersofthenationalarea. If it werea single-humped
distribution,it wouldmeanthatthereis a concentration
pointformarketarea
distancesand that extremedistanceson eitherside of the concentraton becomeprogressively
less frequent.Those products,or valueadditions,whose marketareas encompassthe entirenation must
adofcoursebe thehighlylocalizedindustries-some
sharpproduction
vantagebeingfoundat certainpointsorcenterswithinthelargerarea.
Aroundand amongthose primaryclusterswill have been built up
industries
withtradeareasofintermediate
length,
and witheachpopu6 The data analyzed werefromthe Sixteenth
CensusoftheUnitedStates: 1940.
Occupation,Income, by States,
The Labor Force-Employment,Unemployment,
Vol. IV. Washington,GovernmentPrintingOffice,1941.
44
RUTLEDGE
VINING
13.12.11.10.9. 8. 7. 6. 5. 4. 3. 2. 1.
Iron
Coal
Autos
Paper
Crude
Textile
Logging,
and Leather
Oil
Chemicals
and Machinery
Petroleum
and
Non-ferrous
and Mining
and
Steel
and
and
RATIOS
Mfg Transportation
Auto
Products
Allied
OF
Sawmils,
and
Metals
Coal
Leather
and
Allied
Natural
and
and
Their
Mfg
Products
Equipment
.
Gas EquipmentIndustrial
Products
Apparel
Products
.
PERCENTAGE
Planing
Products
MfgProducts
........
(Except
Products
Mills
.......
.
......
Classification
......
EMPLOYED
Autos)
IN
.....
STATE
TO
...........
..........................
...................
..........................................
..................
.............................
...............
.....................
...............................
........................
........................................
.......................
......................00
0.00
0.00
0.00
0.04
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
LessNo
Than
States
0.00
0.08
0.00
0.00
0.04
0.00
0.04
0.00
0.00
0.00
0.00
0.00
0.00
LessThree
Than
States
of
0.31
0.20
0.25
0.07
0.08
0.07
Less the
0.14
0.00
0.00
0.00
0.00
0.00
0.00 Than
States 20%
of
0.40
0.21
0.62
0.21
0.50
0.23
0.43
0.33
Less the
0.13
0.00
0.14
0.00
0.08 Than
States40%
of
0.21
0.62
0.50
0.50
0.28
0.25
0.57
0.50
0.29
0.13
0.00
Less the
0.00
0.08 Than
States50%
TABLE
1
PERCENTAGE
EMPLOYED
IN
NATION
BY
of
0.51
0.80
0.71
1.39
0.50
0.50
1.00
0.83
0.43
Less the
0.25
0.33
0.38
0.15 Than
States60%
of80
1.20
2.39
1.70
1.29
1.25
2.25
1.29
1.50
1.14
1.13
1.25
2.50
Less the
0.36 Than
States%
All
2.20
2.71
2.42
2.25
But
2.92
2.08
3.25
3.00
3.00
3.25
5.00
4.00
Less Three
0.62 Than
States
INDUSTRIES
IN
1940
All
9.00
8.69
One
But
5.28
3.04
3.14
4.25
7.14
10.33
Less
3.57
15.38
11.25
17.00
15.00Than
State
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
32.31.30.29.28.27.26.25.24.23.22.21.20.19.18.17.16.15.14.
Food
Other
Stone,
Retail Other
Finance,
BusinessUtilities and Domestic
Printing
Furniture
Wholesale
Agriculture
Clay
Professional
Amusement,
Government
and Mines
Trade
Construction
... and
Serv.
Personal
Transportation
and
and
Communications and
Trade
Kindred
Personal
and
Insurance,
Misc.
.
Glass
Service
Publishing
and
Recreation,
Related
Quarries
Repair
Service
Products
etc
and Wooden
Real
Products
...............
Serv.
................
Mfg
Services
(Laundries,
Allied
..........
Estate Goods
........
....... .
Mfg
......
Hotels,
(Including
etc.)
Auto
.
TABLE
Repair).
.
1
..................
.....................
.................
...............
........................
.........................................
........................................
.......................
...........................................
..............................
...........................................
.............................
...................
........................................
......................
. ..........
..............................................
.........................
0.21
0.01
0.33
0.33
0.47
0.25
0.00
0.00
0.33
0.39
0.00
0.57
O.50
0.44
0.44
0.42
0.44
0.47
0.58
0.46
0.56
0.41
0.13
0.58
0.29
0.17
0.46
0.25
0.70
0.56
0.64
0.70
0.50
0.56
0.44
0.69
0.67
0.14
0.50
0.50
0.67
0.57
0.84
0.75
0.53
0.38
0.43
0.83
0.69
0.79
0.89
0.79
0.87
0.67
0.67
0.50
0.78
0.75
1.00
0.64
0.75
0.78
0.63
0.92
1.00
0.89
1.00
0.96
0.92
0.82
0.89
0.79
0.75
0.94
0.97
0.57
1.00
0.83
1.01
0.78
0.64
1.28
0.85
0.75
0.88
0.92
1.04
0.92
0.97
0.89
0.87
0.75
0.99
0.98
0.57
1.11
1.61
0.71
1.04
0.89
0.86
1.00
1.00
1.10
1.00
0.92
1.03
1.07
1.04
1.00
1.07
1.08
1.08
0.84
1.11
1.00
1.16
1.29
1.00
1.94
1.50
1.23
1.15
1.25
1.33
1.35
1.16
1.14
1.22
1.06
1.09
3.00
0.86
1.21
2.60
1.33
2.13
1.38
1.36
1.33
1.25
1.33
1.86
1.14
1.39
1.37
1.69
1.47
13.25
1.38
1.38
2.29
2.21
1.61
3.00
2.75
4.71
1.42
1.67
3.12
1.25
1.33
1.70
1.50
1.63
2.02
2.44
1.56
37.50
2.08
1.88
(concluded)
45
46
RUTLEDGE
VINING
lation aggregatewill be the service and other industrieswith the extremelyshorttrade radii. Thus, an ideal Table 1, it mightbe supposzd,
would showthe mosthighlylocalized industriesas lackingaltogetherin
manyregions.Only a fewregionswould show as greatrelativeemploymentin theseindustriesas does the nation,and a veryfewwould show
a relativelyhigh proportionof their employmentin these industries.
The productsof these industrieswould be importproductsfornearly
all of the regionsof the nation. As the industriesbecome less localized
(as the typical trade radii become smaller), fewerregionswould lack
the industriesaltogether,more regionswould move into the exporting
class, and the locationfactorsin the specializingregionswould become
smaller. The "ideal" Table 1, that is, mightsuggesta surface.Thus,
startingwiththe upperleft-handcorner(withthe mosthighlylocalized
industries)and movinghorizontallyacross the cumulativedistribution
of regions,this surfacewould move along the zero line leaving it relatively late but finallyclimbingvery rapidly to a height sufficientto
make thearea underthe traced-outline equal to unity7-the latterconditionalways being the case. For the next most highlylocalized industry, the surface would leave the zero line slightlyearlier and climb
somewhatless rapidlyto a heightsomewhatlower than the industiy
just preceding.At length,we should exhaust the list of industriesthat
are entirelylacking in at least one region,and as we move down the
arrayofindustrieswe shouldapproach that list ofindustriesthat would
be representedby a height of unity over the entirerange of regions.
That is to say, the firstcolumnswould show a continuousrise to the
neighborhoodof unity. Then fora while the columns would rise to a
maximumand then decline to the neighborhoodof unity.Finally, the
columnswould declinecontinuouslyand approach unity.
The conditionsto which the data in the actual Table 1 are subject
divergein great measurefromthose outlinedin the second sentenceof
theprecedingparagraph.The states showvast differences
in size so that
industrieshighlylocalized in large states show much smallermaximum
locationfactorsthan industriesequally highlylocalized in small states
Average income,living standards,and consumerhabits differimportantly fromregion to region so that employmentthat is necessarily
residentiaryis of varying relative importancefromregion to region.
The industrialclassificationsare verybroad and obviouslylack internal
homogeneityso that in certaincases the location factorsare rendered
quite ambiguousby the inclusionofuniformly
distributedemployment
7 If the regionswereofequal size in termsofemployment,
the simplemeanlocation factorwould be uniity.That is, the mean ordinateunderthe curvewould
be one, and the base of the figurewould extendfromzero to 100 per cent. The
mean ordinatewouldalso be unityifthe regionswerenot ofequal size, provided
thereis zerocorrelationbetweensize and the magnitudeofthelocationfactor.
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
47
BEHAVIOR
and highlylocalized employmentwithinthe same industrialclassification. Nearly all of our-industrialclassificationswould show in some degree this ambiguity.But the industriesranked fourteenth,fifteenth,
seventeenth,and twenty-secondin Table 1 are especially outstanding
instances of internalheterogeneity.Clay and brick-makingmaterial
forregionsas large as states is almost "ubiquitous" materialand the
heavy transportationcharges result in certain subclasses of "Stone,
Clay, and Glass Products" manufacturingbeing residentiary.On the
other hand glass manufacturingappears to be relativelyhighly lominingis very highlylocalized, but other
calized. Nonferrous-metals
subclasses of "Other Mines and Quarries" are residentiary."Agriculture" is no more a single industrythan is "Manufacturing,"and the
production of certain extremelyimportantagriculturalproducts is
highlylocalized, while certain other subclasses are residentiary.Such
"Food and Kindred Products" manufacturiesas bakeries and ice
but othersubclasses show highlocalization.
plants are residentiary,
Nevertheless,Table 1 showsroughlythe characteristicsof our hypothetical surface.Were we to omit the ambiguous classificationsnoted
above, the similaritywould be almost striking,consideringthe other
discrepanciesin the underlyingconditions.The degree of localization
shades offgradually fromthe employmentsthat are virtually independentof local demand,throughthose that produce exportableproducts forwhichthe local demand is not negligible,to those finallythat
are virtuallywhollydependentupon local demand. From this table a
roughclassificationmay be made and a fairlygood conceptionmay be
obtained of the relativeimportanceof the "carrier"industrieswithina
given state. Referringto the table, we may set apart the followingindustriesas the "passive" industries.At the rightof each industrythe
medianstate percentageof total employmentis given:
Retail Trade ...............
. .................
Professionaland Related Services...............
Business Service and Repair Services............
Domestic Personal Service .....................
OtherPersonal Services (Laundry,Hotels, etc.). . .
Transportation............................
Construction.................................
..
Utilities...........1
Wholesale Trade ...........
Government.................................
Communications...........
..............
Food and Kindred Manufacturing
Amusement,Recreation,etc....................
Finance, Insurance and Real Estate .............
Printingand Publishing.......................
13.8%
7.4
1.9
4.5
3.4
5.0
4.5
2.3
3.5
0.8
2.0
0.7
2.4
0.9
54.2%
48
RUTLEDGE
in
VINING
thestates
groups,
at
These
and
1931
1930
1932
1933
1935
1934
1941
1936
1942
1940
1939
1937
1938
these are
following
these
not
one.
limitsin
Year
Yer
Q
all
Forwas
intervals
so
UINTILE
1st
69.0366.8159.9189.9083.15strictness
example,
interval
Quintile
represent
of
INTERVALS*
104.95-113.11
108.58-116.84
102.73-105.57
73.75
81.23
92.80-111.29
96.18-103.14
96.30
99.85-104.99
89.26
105.88-109.04
113.42-116.29
82.64%
the thequintile
IN
insignificantly
65
limits
small
intervals.
as
groups,
7 to shown
THE
2d
89.55-
AS
81.5182.9074.1996.44-
ANNUAL
interval
Quintile
by
109.61-110.31
103.25-104.72
113.24-115.53
111.59-112.76
75.85
120.11-127.48
91.38
97.38
83.38
106.18-106.87
105.12-106.46
Actually,
116.93-119.38
render
contain
86.27%
these
the
but
8
arbitrary
groups.
object
ARRAYS
PERCENTAGES
OF OF TABLE
3d
YEAR
the was
states
ANNUAL
83.6676.2686.3591.51Some
to
and
interval
Quintile
7
STATE
97.71-100.61
112.98-119.35
127.87-130.36
105.24-106.93
77.53
85.33
119.53-122.23
placing
93.75
110.42-112.46
106.50-107.34
107.12-109.07
115.59-117.58
PREVIOUS
90.38%
of divide
a discretion
contain the
INCOME
was
12.state
4th
states
in
of
One
85.4593.9377.7190.94one
interval PAYMENTS
exercised
each
Quintile
group
119.45-124.21
131.79-142.81
122.68-126.41
107.00-108.46
109.24-110.53
95.49
112.81-114.53
80.76
88.23
100.67-103.44
107.41-108.87
117.82-119.84
where
"quintile
annual
the
contains
array
but
7 interval"
into
variation
92.96%
EXPRESSED
5th
80.8288.50-
interval
five
Quintile
rather
states.
110.89-121.13
109.04-115.31
115.43-145.81
146.16-166.72
95.63-101.40
120.41-124.65
109.13-117.57
128.35-140.24
124.31-136.53
88.62
104.32-113.77
95.79
93.38-101.24%
than
equal
between
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
49
50
RUTLEDGE
VINING
in the sense that the demand forits productsis externalto the state.9
Thus, wereall states "typical" in thisrespect,thenatureoftheproducts
of around one-thirdof the employmentof a state would determinethe
sensitivityof that state to national cyclicalforces.
Withtheabove generalizationsregardingindustriallocationin mind,
we may approach our principal problem. Our questions now are: Do
states fall systematicallyinto groups of states showing significantly
160-
120110
100 /_
60706050-
1930 1931 1932 1933 19'3' 1935 1956 1937 1936 1T99 19'0 1911 1912
FIGURE
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
51
PATTERNS
OF CYCLICAL
FLUCTUATIONS
This problemof classifyingregionsinto groupsshowinginternaluniformityof cyclical behavior is, so far as we know, a problemthat has
not been studied extensively.'0It is a problem,however,that would
seem to introduceinterestinganalogies with other fieldsof analysis.
Just as in all problemsof the analysis of variance, we have here a set
of variates that may in principlebe classifiedin accordance withcertain attributes.On the basis ofsuch classifications,
the variationwithin
groups of regionscould be compared with the variation between the
groupssuch that patternsof cyclicalbehaviormay be discernedif they
exist.The systemsofclassificationthenmay yieldadditionalinsightinto
the processinvolvedin the interregional
diffusionoflocalized economic
shock.
As a matterof fact,the methodof studythat we have adopted up to
the presentdevelopmentofour inquiryis onlyanalogous to an analysis
ofvariancein a quite roughand backhanded manner.The data that we
have now at hand would hardlyjustifyan elaborateanalysis,but a simpler method can at least suggestan outline of furtherinquiry. Table
2 and Figure 1 show by years the quintile intervalsof state income
paymentsexpressedas percentagesof the precedingyear." We have
firstendeavoredto findout if the shiftsamong the occupants of these
quintileintervalsgive evidence of a regionalpatternof behavioras the
different
phases of the business cycle develop. These intervalsrepresent segmentsof the horizontalscales of our frequencydistributionsof
state percentagechanges,and if the individualstates appear to be distributedat randomfromyear to year withinthese rangesthenwe may
10 Withoutmakinga pretenseof documentingthe discussionof this problem
of regionalvariation of economic fluctuations,we should at least like to call
attentionto two sets of stimulatingarticlesin this field:the papers by D. G.
Champernowne,"The Uneven Distributionof Unemploymentin the United
Kingdom, 1929-36," Reviewof EconomicStusdies,Vol. 5, February, 1938, pp.
93-106 and Vol. 6, February, 1939, pp. 111-124; and the articles by H. W.
Singer,"The Process of Unemploymentin the Depressed Areas (1935-1938),"
ibid., Vol. 6, June,1939, pp. 177-188, and "Regional Labour Markets and the
Process of Unemployment,"
ibid., Vol. 7, October,1939,pp. 42-58.
11The data used in the computationofthesepercentagesare the same as those
used in the analysis describedin the cited articlein the July,1945,ECONOMETRICA. The estimates of state income payments for 1929 through1939 are
those publishedand discussedin the Surveyof CurrentBusiness,Vol. 22, 1942,
pp. 18-26; the data for1940 through1942 are fromthe June,1943,issue,Vol. 23,
pp. 10-22.
52
RUTLEDGE
VINING
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
53
BEHAVIOR
TABLE 3
QUINTILE
Year
Ark.
Miss.
1930
1931
1932
1933
1934
1935
1936
1937
1.938
1
2
5
3
4
4
5
1
5
3
1
5
5
1
1
5
3
5
3
5
1
2
5
1
5
5
1939
1940
1941
1942
Ala. Tenn.
1
1
4
3
5
3
5
1
3
2
4
5
4
Ga. Ken.
1
2
4
4
5
2
4
3
3
4
4
5
2
1
2
5
4
5
3
2
1
4
2
4
4
4
1
3
3
3
3
4
4
3
2
3
2
2
3
La.
N.C.
S. C.
2
4
4
2
4
1
3
3
5
2
1
3
3
1
3
5
5
4
1
1
3
4
1
4
5
5
5
1
4
2
4
5
5
3
4
1
4
4
pointsofAlabama. In the followingyear,1939,whileArkansasand Alabama are within1 percentagepoint of each other,Mississippi is about
5 pointsabove these othertwo. In 1940,Arkansasand Mississippistay
together,but Alabama is about 6 points above the others.The rest of
the states show discrepanciesfromtime to time, yet all are clearly
pulled toward the same general vicinitywithinthe array,and in the
cases of most of the discrepanciesreferenceto the percentagefigures
will show a close similarity.In 1931, the declinesregisteredby the first
threestates weresomewhatmoremarkedthan in the restof the group.
In 1933,Northand South Carolina werecompletelyout ofline showing
rises of some 12 per cent while the others of the group were showing
Year
N.Y.
Mass.
R.I.
1930
1931
1932
4
4
3
4
5
5
4
5
4
5
4
4
3
4
3
1934
1935
1936
1
1
1
1
1
1
1
1
1
1
1
2
1
2
2
4
2
2
2
1
1
1933
1937
1938
1940
1941
1942
2
1
1
1939
N.J. Vt.
3
2
2
1
2
2
4
1
N.H.
Me.
Md.
Mo.
4
5
3
5
5
4
5
5
4
3
3
3
2
1
2
1
3
2
2
1
1
1
2
1
1
1
1
2
3
2
5
4
3
3
3
1
4
2
54
RUTLEDGE
VINING
55
are regardedas the "regions." When the behavior of more appropriately defined"regions"-isanalyzed thereis evidence that the satisfaction of certainconditionswill resultin groupingsthat followindustrial
lines with less regardforgeographicalcontiguity.In the case of Missouri the conformanceto this patternshould perhaps be explained on
special grounds,and these groundswill be presentedwhen the attempt
is made to rationalizethe grouppatterns.
In Group III we place Indiana, Ohio, Michigan, Illinois,and Pennsylvania. Connecticutand Delaware bear a strongfamilyresemblance.
West Virginia and Wisconsin show resemblances,but marked discrepancies are in evidence. The quintile patterns for this group are
shownin Table 5.
TABLE 5
QUINTILE
Year
Ind.
Ohio
Ill.
Mich.
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
2
2
1
3
3
4
4
5
1
5
3
5
2
2
2
1
3
3
3
4
4
1
5
3
4
2
2
2
1
1
3
3
3
5
1
5
3
2
1
2
2
2
1
5
5
4
5
1
5
5
4
2
4
4
3
2
2
2
3
4
1
4
4
4
2
2
5
3
2
3
2
4
4
1
5
5
1
1
2
4
3
4
4
1
4
3
2
1
3
2
2
2
2
2
3
3
5
3
3
2
1
3
3
2
56
VINING
RUTLEDGE
These other states of the group are not so far out of line with one
anotherin 1933 as the-quintilelocations would suggest. The range of
variationforall 48 states was 24 percentagepoints,while the range of
variationforGroup III states exceptingMichiganwas 5 per cent.
Referenceto the percentagefiguresforthe firstfivestates indicates
few other discrepanciesof importance.Illinois was low in 1934, and
Illinois and Pennsylvania were somewhatlow in 1941. But it can be
TABLE
QIUINTILE
PATTERNS
OF THE
STATES
WITHIN
GROUP
IV
Year
Wash.
Ore.
Calif.
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
3
2
2
4
3
2
5
2
4
2
4
5
5
2
3
2
3
4
4
5
2
4
4
3
4
4
5
3
3
3
3
2
5
3
4
2
4
3
3
said, subject to the qualificationsregardingMichigan in 1933 and regardingthe apparent relationshipbetween Pennsylvania and Group
II, that the fivestates adhere very closely to a group patternand occupy the same vicinityof the frequencydistributionyear afteryear.
The otherstates show certaindivergencies.Connecticutoccupied a
Group II positionduringabout the firsthalf of the period.Since 1937,
it has conformedcloselywiththe patternof Group III. Delaware also
is attractedto the Group II pattern.West Virginiashows markeddiscrepancies,but its percentagepatternclosely resemblesthat of Pennsylvania. Wisconsin shows a moderate resemblance,but its position,
like Missouri's,tendsto remainin the centerof the arrayor towardthe
lowerend of the array of the absolute rates of change.
In Group IV are the PacificCoast states-Washington, Oregon,and
California.The adherenceto a group patternis somewhatcloser than
the quintile interval shown in Table 6 would suggest. Until 1941, a
rangeof8 pointswould includeall threestates in any year,and excepting 1930 and 1934 a range of 4 points would suffice.In 1941 and 1942
Washingtonand Oregongained at a more rapid rate than did California. The tendencyforthis groupappears to have been forrelativestability,the positionin the array tendingto be the middle ground.The
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
57
BEHAVIOR
exceptionsare 1936 and the two war years when the group occupied
the upper reachesof the arrays.
In Group V are placed certain political divisions that have shown
a markedtendencyforexpansion.The percentagefigureswhen placed
alongside of each other do not give as clear an impressionof consistencyas do the quintilefigures,but even here the conformanceto a
patternis in evidence. In seven of the years a range of 5 percentage
pointsincludesall the states,in threemoreyearsa rangeof8 pointswill
suffice,and in the remainingthreeyears 14 points separate the lowest
from the highest percentage of the three. Obviously, expansionary
forcesof an unusual characterhave been in operationin these three
areas. It does not follow,of course,that the similarityin the reactions
TABLE 7
QUINTILE
Year
Dist. of Colum.
Va.
Fla.
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
5
5
5
1
2
4
5
2
5
1
3
2
3
3
5
5
2
4
2
3
2
5
4
5
5
3
4
4
4
2
4
4
5
4
5
5
4
2
3
58
RUTLEDGE
VINING
1941
1931
1940
1933
1942
1939
1932
1930
1938
1937
1936
1935
1934
4 1 1 1 3 5 24 1 5 3 1 1
4 4 1 1 5 5 3 3 3 5 4 2 2
Y
e
a
r
Okla.
Tex.
-
5 3 1 2 2 5 4 3 4 4 1 2 2
_
_
_
Ariz.
-
5 3 5 1 4 5 35 3 4 1 2 3
_
_
_
Utah
-
3 1 3 4 3 4 5 4 4 5 2 4 3
_
_
_
N.M.
Q-UINTILE
3 1 1 2 2 5 4 3 2 4 3 3 4
_
_
Colo.
-
4 3 3 1 2 5 5 3 5 5 2 1 3
1 2 5 3 2 3 1 5 5 4 1 1 1
3 5 5 4 1 5 1 5 1 5 3 1 2
5 1 5 5 1 4 25 4 1 4 5 5
PATTERNS
_
_
_
Idaho
THE
TAB3LE
2.
2 4 5 1 3 4
5 1 3 1 2 1
5 1 5 1 1 5
3 4 1 5 5 5
5 1 5 1 1 5
_
Mont.
_
STATES
N.D.
_
b
_
WITHIN
Nev.
_
_
2 2 2 3 5 1
OF
GROUP
_
Wyo.
VI
_
_
Neb.
_
_
_
_
Iowa
_
4 4 2 3 5 1
5 5 4 1 1 5
S.D.
_
5 5 4 1 1 4 2 4 3 4 1 2 5
_
Kan.
_
2 1 1 2 4 3 3 5 2 3 2 3 4
Minn.
_
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
59
60
RUTLEDGE
VINING
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
61
62
RUTLEDGE
VINING
UNDERLYING
AND AMPLITUDE
REGIONAL
OF CYCLICAL
VARIATION
IN TIMING
FLUCTUATION
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
63
64
RUTLEDGE
VINING
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
65
66
RUTLEDGE
VINING
these states are highly specialized in producingraw materials. Oklahoma is dominatedby cotton and petroleumproductionand refining.
Iowa is predominantlyagriculturaland specializes in corn and livestock. Nebraska's concentrationis in corn,livestock,and wheat. South
Dakota is as predominantlyagriculturalas Arkansas and Mississippi,
depending,however,upon wheat, corn, and livestock. Montana and
North Dakota are dependentupon wheat and livestock. Montana, in
addition, has nearly 10 per cent of total employmentin miningand
metal refining,the principalmineralinvolved being copper. Idaho is
dominated by livestock agriculture,nonferrousmetals mining,and
lumber.Nevada is principallya miningstate, 15 per cent of its employmentbeing engaged in the miningof nonferrousmetals. Mining
employsas many people as does agriculturein Nevada, and what agriculturethereis, is mainlyrange livestock.
Thus, these states, like the Group I states, are specialistsin agricultural and the otherextractiveindustries.A relativelyhigh proportion
of the employmentof these states is dependentupon externaldemand,
and thislatterwould appear to have a highshort-runincomeelasticity.
The situationof these raw-material-producing
states would seem to be
somethinglike that of the marginalfacilitiesof a manufacturing
firm.
For much of theirproduction,demands are only forthcoming
at high
levels of national incomeand high levels of national investmentactivity. When investmentactivity slackens, the use of the raw-materials
output of these mountain,plains, and southeast states declines drastically. This decline in the primaryindustriesis then transmittedto
those residentiaryindustriesproducingforlocal demand.
Two observationsof Keynes in his General Theoryare of especial
pertinenceregardingthe ratherexceptionalviolenceof cyclicalfluctuation in these areas in which the primaryindustriesare predominantly
agriculturaland mineral.Keynes raises, in order to explain away, an
apparent paradox. He says:
We have seen above that the greaterthe marginalpropensityto consume,the
greaterthe multiplier,and hence the greaterthe disturbanceto employment
to a given changein investment.This mightseem to lead to the
corresponding
paradoxicalconclusionthat a poor communityin whichsaving is a very small
proportionofincomewill be moresubject to violentfluctuations
than a wealthy
communitywheresaving is a larger proportionof income and the multiplier
consequentlysmaller.
REGIONAL
PATTERNS
OF BUSINESS-CYCLE
BEHAVIOR
67
dC/dY
68
RUTLEDGE VINING
industrysuch
whereagricultureis the predominant
a community
in
will
importance
be
of
overwhelming
changes carry-over
compared
fluctuations.
GroupsII and VI
withanyotherusualcauseofinvestment
thatare described
but the effects
are made up of suchcommunities,
forthesestates.A preby Keyneswouldbe even moreintensified
outputwouldrepdominant
partofthevalueoftheirentireagricultural
forthem(in the senseadoptedin
expenditures"
resent"investment
discussionsof the internationaltrade multiplier),and carry-over
changeswouldbe accompaniedby pricechanges.Keynesgoes on to
fluctuation,
exsuggestthatin thepastperhapsno causeofinvestment
ceptwar,
of agriwas in any way comparablein magnitudewithchangesin the carry-over
culturalproducts.Even today it is importantto pay close attentionto the part
played by changesin the stocksof raw materials,both agriculturaland mineral,
oftherateofcurrentinvestment.I shouldattributethe slow
in the determination
has been reached,mainly
rate of recoveryfroma slump,afterthe turning-point
to the deflationaryeffectof the reductionof redundantstocks to a normal
level.... Sometimes,indeed,the reductionof stocks may have to be virtually
completedbeforeany measurabledegreeof recoverycan be detected.
It is commonunderstanding
in tot%lconsumer
nowthata reduction
demanddoes nothave to take place in thelatterphaseofan upward
in orderthatthereshoulddevelopa reductionin
businessmovement
decision.All thatis required
demandforoutputsubjectto inventory
is thecompletion
depletedinventories
ofthebuildingup ofpreviously
ofconso thatthedemandforoutputis loweredto theneighborhood
sumertaking.Or again,it is not necessarythat consumerdemand,
thatis, nationalincome,risein thelatterphaseofa downwardmoveand otherdurableraw
mentin orderthatthedemandforagricultural
materialoutputshouldrise.It is onlynecessarythatthatportionof
in invenconsumption
thathas been providedfroma disinvestment
of these
A development
be reducedsufficiently.
toriesor carry-over
erratic
ofrelatively
mustsurelylead to an expectation
considerations
ofturning
points
andviolentfluctuations
and ofan initialdevelopment
withintheGroupI and GroupVI states.In fact,theworstofan assortseemsto be the
so faras thebusinesscycleis concerned,
mentofworlds,
lotoftheseareas:a lowaveragepropensity
to consumehomeproducts
and dependenceuponinterre(thatis, a highdegreeofspecialization
to consumehomeproducts
gionaltrade);a highmarginalpropensity
ofdemandforimports);a highlysensi(thatis,a lowincomeelasticity
multiplicand
(thatis, a highinviolentfluctuating
tiveand relatively
ofdemandfortheexportproducts).
comeelasticity
University
of Virginia