American Sociological - Social Sciences Faculty Web Pages

Civil Disorder and the Welfare Explosion: A Two-Step Process
Author(s): Sanford F. Schram and J. Patrick Turbett
Source: American Sociological Review, Vol. 48, No. 3 (Jun., 1983), pp. 408-414
Published by: American Sociological Association
Stable URL: http://www.jstor.org/stable/2095232
Accessed: 19/05/2009 15:36
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=asa.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].
American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to
American Sociological Review.
http://www.jstor.org
RESEARCH NOTES
CIVIL DISORDER AND THE WELFARE EXPLOSION:
A TWO-STEP PROCESS*
SANFORD F. SCHRAM
J. PATRICK TURBETT
State University College at Potsdam
Upon not finding a direct relationship between civil disorder and welfare growth in
American cities in the late 1960s, some analysts have rejected the Piven and Cloward
(1971) thesis that the expansion of welfare in the late 1960s operated largely as aform
of social control so as to recreate political stability. We hypothesize that the welfare
explosion in the late 1960s was in part the result of a two-step process in which civil
disorder impelled the national government to enact liberalizations of welfare policy
which in turn were most actively implemented by those states most wracked by
rioting. Analysis of the relative state growth rates in the number offamilies receiving
Aid to Families with Dependent Children suggests the plausibility of our hypothesis
and the Piven and Cloward thesis.
Piven and Cloward's(1971) thesis on the role
of welfare in westerncapitalistsocieties lies at
the centerof the debateover the welfareexplosion of the late 1960s and early 1970s in the
United States. Accordingto Piven and Cloward, the historical role of public-welfare has
been one of cyclical expansionand contraction
in responseto the alternatingneeds of the state
for political stability and acceptance of lowwage work by the poor. Duringtimes of political stability,public welfareprimarilyfunctions
to enforce on the poor the work norm of selfsufficiencyby restrictingaccess to aid. During
periods of unrest among the poor, public welfare functions largely to promotepolitical stabilityby easing the poor's access to aid. In the
1960s the growing political unrest of poor
blacks escalated into over 160 majorriots between 1965and 1968(Feagin and Hahn, 1973)
and promptedthe dramaticrise in welfare recipientsin the late 1960sand early 1970s(Piven
and Cloward, 1971:222-47).
Variousresearchershave providedempirical
supportfor the relationshipbetween civil disorder and welfare growth in the late 1960s
(Betz, 1974; Jennings, 1979; Isaac and Kelly,
1981).Critics,however, have rejectedthis thesis upon finding no relationshipbetween riottorn cities and increases in welfare (Durman,
1973; Albritton, 1979). Our analysis seeks to
resolve these conflictingfindings. We provide
evidence supportiveof the hypothesis that the
welfare explosion was in part the result of a
two-step process in which civil disorderin the
late 1960simpelledthe nationalgovernmentto
enact liberalizationsof welfarepolicy which in
turn were most actively implementedby those
states most wracked by rioting.
A TWO-STEPPROCESS
From one perspective, the relationship between civil disorder and welfare growth was
indirect rather than direct. Frequent and intense rioting in certain cities contributedto
nationalchanges in welfarepolicy which led to
welfaregrowthin all cities (PivenandCloward,
1979:1012;Isaac and Kelly, 1981:1259).We
augment this perspective with the possibility
that the states most wracked by rioting most
actively implemented these changes. Therefore, we suggest that the relationshipbetween
civil disorderand welfare growth was both direct and indirect.
For several reasons, we focus our analysis
on the growth in the number of families receiving Aid to Families with Dependent Children (AFDC). Duringthe 1960s, this program
had the broadest coverage and was the most
expensive of all programs,for the poor. It was
also the welfareprogrammost stronglyassociated with the poor, blackunderclasswhichwas
implicatedin the disorderof the 1960s(Albritton, 1983).For both the nationaland state gov.ernments, the structure of AFDC made it
convenient as a means of quelling disorder.
Since its inceptionAFDC has been a national*Directall correspondenceto: SanfordF. Schram state programfor which the national governand J. PatrickTurbett,Departmentof Political Sci- ment sets basic guidelinesand providesat least
ence, State UniversityCollege,Potsdam,NY 13676. half of the funding. It representedan income
408
AmericanSociologicalReview 1983, Vol. 48 (June:408-414)
CIVIL DISORDER AND THE WELFARE EXPLOSION
transfer programwhich was both relevant to
the populationassociated with the riotingand
at the disposal of the national government.
This was not the case with the other income
transferprogramof relevance to poor blacks,
GeneralAssistance (GA), which is exclusively
financed and administered by states. Also,
states had the incentive of havingthe national
governmentpay for at least half of the welfare
growthwhen it occurredunderAFDC. In fact,
in the early 1960s, northern states quickly
sought to save money by shiftingGA families
to AFDC under the AFDC-UP (Unemployed Parent)option, which was established
by the nationalgovernmentin 1961(Piven and
Cloward,1971:184-89).This type of shiftingof
welfarerecipientsprobablycontinuedas states
sought to have the nationalgovernmentshoulder as muchof the burdenof welfaregrowthas
possible.
The structureof AFDC enabledstates rather
thancities to play a pivotal role in the two-step
response by governmentto the disorder.The
nationalgovernment'sability to liberalizethe
poor's access to welfare was contingentupon
how state governments implementednational
changes. Historically, states have had substantial latitude in implementingAFDC (Albritton, 1983). While states did vary their responses, the nationalchanges in AFDC in the
late 1960swere broadrevisions in policy which
requiredeach state to implementthem evenly
throughout its jurisdiction (Patterson, 1981:
27-36). Such changes limited the extent
to which localities could vary their responses.
Of course in the 35 states where local governmentsdid not at this time participatein the
administrationof AFDC, local discretion was
probablyeven more limited.
The specific national changes in AFDC in
the late 1960slend support to this argument.
Much of the national government's actions
came by 1969-the year followingthe periodof
frequentand intense riotingby poor blacks in
many Americancities (1965- 1968).In the year
that followed, the national AFDC rolls increased an unprecedented36 percent as comparedto 17percentfor the entire decade of the
1950s(Rodgers, 1979:93).By 1969, the Social
Security Amendmentsof 1967,along with new
regulationsfor 1969, went into effect so as to
liberalizethe workingpoor's access to welfare
(Steiner, 1971). By 1969, the legal rights services of the nationalgovernment'sEconomic
OpportunityAct were having an impact. This
push for legal services for the poor culminated
in the landmark welfare case Shapiro v.
Thompson (1969). In this case, the United
States SupremeCourtabolished the residency
requirementsmany states had for determining
eligibilityfor welfare and made large pools of
409
previously ineligible, newly arrived poor automaticallyeligible for welfare. By early 1970,
the Courtruledin Goldbergv. Kelly (1970)that
poor people could not be denied or terminated
from welfare without "due process" and welfare agencies were prodded into being more
careful than in the past as to whom they kept
off the rolls. In 1969, a series of additional
revisions to the AFDC programexpanded its
coverage to various groups of poor persons,
includingolder children.
Although several of these national policy
changes began before 1969, most, if not all of
them, were not pushed by the national government until the rioting had already peaked.
All of these changes represented broad revisions in policy which requiredeach stat&to
implement them evenly throughoutits jurisdiction. Therefore, it is reasonableto suggest
that the national government liberalized
AFDC in response to the riots and once it did
so, riot-tornstates actively implementedthese
general policy changes throughouttheirjurisdictions.
Althoughthere may have been a relationship
between civil disorderand other social welfare
programs,in most instances this was not the
case, and in other instances the relationshipis
difficultto document.Manyof these programs,
such as UnemploymentCompensation,were
not of direct relevanceto the poor blackunderclass. Other programs, such as Medicaid,
which were relevant to this population, were
new, therebycreatingthe possibilityof mistakenly attributinggrowth rates common to new
programsto the disorderof the 1960s. Finally,
otherrelevantprograms,such as Food Stamps,
were to be fundamentallytransformedand expanded in coverage duringthe late 1960sand
early 1970s, thereby making study of the relationshipbetween their growth and civil disorder impossible (Rodgers, 1979:82-117). In
sum, government'sresponse to the civil disorder of the 1960smay have been in part, but
not exclusively, a two-step process of the national governmentliberalizingAFDC and the
states most wracked by rioting most actively
implementingthose changes.
THE NEED THESIS
The welfareexplosion cannot be explainedby
the major alternative thesis that welfare expandedto meet increasingneeds.' Poor blacks
who migratedout of the south to the north
I Gr0nbjerg(1977) argues that the welfareexplosion was largelyattributableto an increasinglyliberal
commitmentby our society to accordingwelfare to
all who needed it. For a critiqueof this perspective,
see Schramand Turbett(1983).
410
increasinglyfound unemploymentand a need
for welfare (Beardwood, 1968; Patterson,
1981). This process, however, began after
WorldWarII, and sharpincreases in the welfare rolls were not registered until the latter
half of the 1960s (Piven-and Cloward,,1971).
Nor can the dramaticgrowth in the rolls be
explainedby increases in poor black, femaleheaded families accompanyingthe black migration to northern cities (Moynihan, 1965).
Between 1960 and 1972, the proportion of
nonwhite families headed by single females
rose from slightly one-fifth to one-third(Levitan et al., 1975:114). Yet, if all the new
female-headedfamilies in the period between
1959and 1966had received welfare, this would
only account for about ten percent of the increase in the AFDC programduringthe same
period (Lurie, 1968).
In any case, the idea that increases in poor
blackfemale-headedfamilies translatedirectly
into increases in the AFDC rolls is fundamentally misleading, for it assumes that eligible
poor families automaticallyapply and get accepted for AFDC. This hardly was and still is
not the case, althoughapplicationrates of poor
people for AFDC increased in the late 1960s
and early 1970s (Doolittle et al., 1977). In the
early and mid-1960s,only about one-halfof all
eligible poor families were receiving AFDC
(Piven and Cloward, 1971:186).Welfareagencies now as then avoid publicizing the
availability of AFDC and invoke a series of
practices from fraud investigations to long
waitingperiodsin orderto keep the rolls down.
An expanded pool of poor families eligible for
AFDC created only thepotential for increases
in the AFDC rolls. Yet, it was not largeenough
to create the welfareexplosion and was insufficient by itself in doing so.
AMERICAN SOCIOLOGICALREVIEW
for which Cronbach's Alpha was .90.3 With
this index we can more effectively assess the
extent to which states were wrackedby rioting
than if we were to rely simplyon the numberof
riots states suffered. Our measuresof need include the percent of families poor and femaleheaded, the percent unemployed, the percent
residingin central cities, and the net black inmigrationbetween 1960and 1970. We also include measures of the state's capacity to finance welfaregrowth:medianincome and per
capita revenue.4 Interparty competition as
measuredfrom 1956to 1970(Ranney, 1971)is
includedto account for the tendency of political parties in competitive states to pursue liberal welfarepolicies in a quest- for votes (Dawson and Robinson, 1963).5 Welfare growth
from 1960-1964 is includedto account for the
possibility of postriot welfare growth being a
continuationof welfare growth duringthe period precedingthe intense rioting. And lastly,
we include the average annualcommunityaction expendituresper poor person in the state
for 1965-1969to assess the extent to which the
efforts of communityaction programsmay ac-
3Our index of the severity of riotingis the summationof the standardizedz-scores for the number
of civilians killed and injured, the number of law
officerskilledandinjured,the numberof arrests,and
the numberof arsons duringthe riots between 1965
and 1968for each state. Isaac and Kelly (1981) include durationin their index of riot severity; however, the length of time of any riot is not a good
indicatorof riot severity in that many riots had long
periods of inactivitybetween violent incidents (see
National Advisory Commissionon Civil Disorders,
1968:589-631).
4All data except that on rioting come from the
Statistical Abstract (1960-1973). Except for net
black in-migration,all need and capacity indicators
are averaged. Unemploymentis averagedfor 1960,
DATA ANALYSIS
1965,and 1970,and the others are averagedfor 1960
Ourdataanalysisemploys a series of measures and 1970.
5 Our measureof interpartycompetitionis based
in order to test for the two-step thesis while
takinginto accountotherfactors, such as need, on Austin Ranney's(1971) scale for party competiin AmericanStates. Ranney'sscale is based on
which may have promoted AFDC growth in tion
datafor 1956to 1970for all states and includesmeathe states. We used data on the frequencyand suresfor each state of: (1) the state senate seats held
intensity of rioting in the states between 1965 by the Democrats; (2) the average percentage of
and 1968-the periodof most intense riotingin popularvotes won by DemocraticgubernatorialcanAmerican cities in the 1960s (United States didates;and (3) the percentageof all terms for govSenate, 1968).2Fromthese data we constructed ernor,senate, andhouse in whichthe Democratshad
a standardizedriot severity score for each state control.To calculatethe scale, Ranneyaddedthese 3
measures-together and divided by 3. This yields a
score for each state rangingfrom .0000 to 1.0000
2 We recognizethat there was sporadicriotingby
(representingtotal Democraticsuccess), with .5000
the poor precedingand followingthe 1965-1968pe- representing balanced two-party competition. To
riod. Yet, 1965-1968was undoubtedlythe periodof convert Ranney's scale to one in which balanced
most intense riotingby the poor and therefore,our competitionwas on one end of the scale, we subdatashouldbe representativeof the extent to which tracted the Ranney score for each state from .5000
states were wrackedby riotingin the 1960s.
and took the absolute value.
CIVIL DISORDER AND THE WELFARE EXPLOSION
411
Table 1. Factors Affecting AFDC Roll Growth 1969-1972for the 50 AmericanStatesa
Severity of Rioting
MedianIncome
CAP Expenditures
Black Migration
InterpartyCompetition
% CentralCity
AFDC Growth, 1960-1964
% Unemployed
Per CapitaReyenue
% Poor Female-headedFamilies
Increases in AFDC Families per 1000 Poor Families
1969-1970
1969-1972
b
Beta
b
Beta
.09
'.14
.54*
.38*
.0002
.38*
.0006
.48*
-.0004
-3.3
.0002
.18
-.24
14.1
.001
7.12
-.04
-.13
.002
-9.5
.06
-.17
.33*
.04
-.11
.19
.10
.16
.0001
2.36
-.22
.96
-.004
-11.69
.08
.28
-.05
.01
-.17
-.13
.30*
.44*
.10
.0006
R2=.73
# of Riots
MedianIncome
CAP Expenditures
Black Migration
InterpartyCompetition
% CentralCity
R2=.52
.044
.0002
- .0008
- .06
.0013
-1.83
.0002
-.22
-.07
.29
-.05
-7.25
.00004
1.58
AFDC Growth, 1960-1964
-.054
% Unemployed
Per CapitaRevenue
% Poor Female-headedFamilies
20.51
.002
9.97
-.03
R2=.59
.28*
.17
.22
.33*
.49*
.05
-.13
-.04
.18
-.17
-.04
13.3
- .0026
-11.2
- .11
.09
-.12
R2=.49
a
All figures representthe result of simple multipleregression.
* F is significantat .01 level.
the increasein the numberof familiesreceiving
AFDC per 1000poor families.7
Table 1 presents the results of multipleregression equations which assess the relative
effects of each of the above independentfactors on welfare growth, controlling for the
other independentvariables.The effects of severity and frequency of rioting were assessed
in separateequationsbecause of their high intercorrelation(r = .80), therebymakingit difficult to distinguishtheir relative effects simultaneously. The standardizedregressioncoefficients for the severityof riotingindicateit to be
the most importantpredictorof 1969-70AFDC
roll.growth and the second only to medianincome for 1969-72 roll growth. The frequency
of riotingis second only to medianincome as a
predictorfor AFDC increases for both 19691970 and 1969-1972. Also of particular
significanceis the fact that measures of need
generallyare not importantfactors in explaining the variationin state welfaregrowthfor the
periodsunderstudy. Differencesin the level of
We recognizethatexpendituresare a crudemea- civil disorder are more important than dif-
count for variations in state increases in the
welfare rolls.6
We examine two time periods for assessing
variation in our dependent variable-AFDC
roll growth following the 1965-1968 period of
intense rioting. We measure change in the
AFDC rolls for both the first year (1969-1970)
and the first three years (1969-1972) immediately following the riot period. As we have
mentioned, 1969-1970 is the year of unprecedented AFDC growth; it is also the first full
year after the most intense rioting by the poor
in the 1960s and the first year after the national
government took a series of dramatic steps to
liberalize the poor's access to welfare. For
these reasons, it merits our attention. We examine AFDC growth for 1969 to 1972 because
it measures such growth from the time the national government liberalized access to AFDC
in the wake of intense rioting until the AFDC
rolls began to level off. We measure growth as
6
sure of effort by communityaction programsto increase services to the poor;however, no other measures are at our disposaland state communityaction
' We do not use the simplepercentincreasein the
expendituresperpoorpersondo correlatewell in our
state data (r = .52) with increasesfrom 1969to 1972 numberof familiesreceivingAFDC because percent
in the numberof familiesreceivingAFDC per,1000 change measuressuch as this are heavily dependent
on the base values.
poor families.
412
ferences in need in explainingstate variations
in AFDC roll increases after 1969.
Severity appearsto be much more strongly
associatedthan frequencyof riotingwith welfare growth for 1969- 1970; however, they
seem to be equally associated with AFDC
growth for 1969-1972. The data suggest that
intense rioting had acute effects promoting
dramatic welfare growth in riot-torn states
once the nationalgovernment liberalizedaccess to AFDC in 1969. The frequency of rioting's effects on welfaregrowth was greaterfor
1969-1972 than in the year following the intense riot period. Nonetheless, the frequency
of riotingseems to have been less of a factor
thanriot severity in explainingwelfaregrowth.
Replacingfrequencywith severity of riotingin
the equations increases the total variance explained 14 percent for welfare growth 19691970and 3 percent for 1969-1972.8
The multiple regression results in Table 1,
however, are insufficient by themselves to
substantiatethe two-step hypothesis. For instance, should our multiple regression
equationsproduce similarresults, indicatinga
relationship between rioting and welfare
growth for a period preceding 1969, then we
could not conclude that riot-torn states were
more likely than other states to increase their
welfare rolls only after the national government had liberalizedAFDC in 1969. In other
words, it may be that riot-tornstates tended to
have higherwelfareincreases thanother states
for reasons other than variationsin state reactions to nationalwelfare policy changes.
Table 2 casts doubt on this possibility by
indicating that variations of state welfare
growth during the period of intense rioting
(1965-1968) were unrelated to the frequency
and intensity of rioting but were related to a
numberof otherfactors.9The resultsin Table 2
lend some support to the idea that riot-torn
states were likely to undergo welfare growth
more so than other states only after the na-
AMERICAN SOCIOLOGICALREVIEW
Table 2. Factors Affecting AFDC Roll Growth,
1965-1968 for the 50 American Statesa
Increases in
AFDC Families
per 1000 Poor
Families
1965-1968
Severity of Rioting
Median Income
CAP Expenditures
Black Migration
Interparty Competition
% Central City
AFDC Growth, 1960-1964
% Unemployed
Per Capita Revenue
% Poor Femaleheaded Families
b
Beta
.0073
-.00003
.0009
3.95
-.00006
-.33
.36
7.91
.006
.06
-.09
.11
.22
.17
-.12
.26*
.16
.78*
R2 =
#of Riots
Median Income
CAP Expenditures
Black Migration
Interparty Competition
% Central City
AFDC Growth, 1960-1964
% Unemployed
Per Capita Revenue
% Poor Femaleheaded Families
.35
10.29
.01
.00004
.0009
4.07
-.00007
-.39
.33
8.98
.006
.69
.11
-.10
.11
.23
-.18
-.14
.24*
.19
.79*
9.83
.33
R2 =
.69
a
All figures represent the result of simple multiple
regression.
* F is significant at the .01 level.
tional government liberalized welfare policy in
1969, after the period of intense rioting (and
not before).
Additional support for this two-step process
can be gleaned from Table 3, which shows that
the frequency and severity of rioting in a state
in any one year were most often not good predictors of welfare growth the following year.
8 Isaacand Kelly (1981)foundin theirexamination Only the severity of rioting in 1965 was related
of nationaldata that the frequency ratherthan the to welfare growth the following year, and this
severityof riotingin the nation in any one year was relationship may be the product of the fact that
the betterindicatorof nationalwelfaregrowthin the a small number of states (5) suffered a few
intense riots in 1965, and these states tended to
subsequentyear.
9 Comparingthe unstandardizedb's for AFDC be states which were undergoing above avergrowth1960-1964in Tables 1 and 2 revealsthat state age welfare growth at that time. Although we
variationsin welfaregrowthfor 1%5-1%8were tied have evidence indicating that rioting in 1965
to variationsin growthfor 1%0-1964,but state vari- was directly related to welfare growth the folations in welfaregrowth after the rioting were not. lowing year, the data in Table 3 generally lend
This shift suggests that somethingoccurredto alter
the 1960-1%8patternof welfaregrowth.These data support to the idea that rioting was not directly
buttressour argumentthat the states most wracked related to AFDC increases until after the naby rioting tended to have the largest AFDC roll tional government liberalized the program in
growth only after the period of intense rioting and 1969.
The relationships we found between the freonly after the national government liberalizedthe
programin 1%9.
quency and intensity of rioting and welfare
CIVIL DISORDER AND- THE WELFARE EXPLOSION
413
Table 3. Correlationsand StandardizedRegressionCoefficientsfor Frequencyand Severity of Riotingin
One Year and WelfareGrowththe Following Year for the 50 AmericanStatesa
Frequencyof Rioting
Severity of Rioting
Increasesin AFDC Families
in the Base Year
in the Base Year
per 1000 Poor Families
Beta
Simple r
Beta
Simple r
1965-1966
.05
.02
.48
.39*
1966-1967
.15
-.14
.21
-.11
1967-1968
.49
.01
.23
-.10
1968-1969
.01
.09
.07
-.05
a All Betas are the result of simple multipleregressionsin which the other variablesin the equations in
additionto the severity or frequencyof riotingare the same as the ones reportedin Tables 1 and 2.
* F is significantat the .01 level.
growth for states most often do not hold for
cities of 200,000or more in population.If we
limit our analysis to cities. with populationsin
1965of 500,000or more, we do find a relationship between the severity of rioting in a city
and welfaregrowth 1969-72 (r = .34). Beyond
that, however, the links between rioting and
welfaregrowthwhich we found for states were
not present in cities. These data lend some
supportto the idea that it was states, not cities,
which had the fiscal capacity and authorityto
vary their responses to rioting; and once the
nationalgovernmentacted to liberalizeAFDC,
riot-torn states implemented these changes
more actively than others so as to ease the
poor'saccess to welfare. Once a riot-tornstate
so acted, the welfarerolls were apt to rise at an
above-averagerate in both riot and nonriot
cities in that state, thereby diminishingthe
probabilitythat riot-torncities would tend to
have higher welfare roll increases than other
cities.
(1979) failed to find a direct relationshipbetween civil disorder and welfare growth in
cities, they took this as a refutationof Piven
and Cloward'sthesis that welfarelargelyoperates as an agent of social control. The absence
of a direct relationshipbetween civil disorder
by the poor and welfare growth in American
cities in the late 1960s does not refute this
thesis. In the late 1960sand early 1970s, welfare may have performed a social-control
function by responding to civil disorder
through a two-step process of national liberalization of welfare policy and active implementation of such changes by riot-torn
states.
REFERENCES
Albritton,Robert B.
1979 "Social amelioration through mass insurgency? A reexaminationof the Piven
and Clowardthesis." The AmericanPolitical Science Review 73:1003-1011.
1983 "Subsidies: welfare and transportation."
CONCLUSION
Pp. 373-414 in Virginia Gray, Herbert
Although our findings are not unambiguous,
Jacob, KennethVines (eds.), Politicsin the
the general pattern in our data is consistent
AmericanStates: A ComparativeAnalysis.
with the idea thatthe relationshipbetweencivil
Boston: Little, Brown.
disorderand welfare growth in the late 1960s Beardwood,Roger
1968 "The Southernroots of the urbancrisis."
and early 1970swas the productof a two-step
Fortune78(2) (August 1968):80-87.
process. Rioting by the poor exacerbated a
volatile political climate compelling national Betz, Michael
1974 "Riots and welfare:are they related?"Soin welfare policy, which in turn
liberalizations
were most actively implementedby the most
riot-tornstates. Ourfindingsoffer supportfor
the Piven and Cloward thesis that welfare
largelyoperatesas a formof social controland
in the late I%Osit expandedas partof government efforts to recommit a rebellious poor
populationto the existing social order. These
findings also complement a growing body of
literatureon the role and functionof welfare in
capitalistic, industrialized societies which
argues that welfare performs a social-control
function for the state (e.g., O'Connor, 1973;
Wolfe, 1977;Gough, 1979).
In the past, when analysts such as Albritton
cial Problems 21:345-55.
Dawson, RichardE. and James A. Robinson
1963 "Interparty competition, economic variables and welfare policies in American
states." Journalof Politics 25:265-89.
Doolittle, Frederick, Frank Levy and Michael
Wiseman
1977 "Themirageof welfarereform."The Public
Interest47 (Spring):62-87.
Durman,Eugene
1973 "Have the poor been regulated?Towarda
multivariate understanding of welfare
growth."Social Service Review47:339-59.
Feagin, Joe and HarlanHahn
1973 GhettoRevolts:The Politicsof Violence in
AmericanCities. New York: Macmillan.
414
Gough, Ian
1979 The Political Economy of the Social Welfare State. London: Macmillan.
Gr0nbjerg,Kirsten A.
1977 MassSociety andthe Extensionof Welfare,
1960-1970.Chicago:Universityof Chicago
Press.
Isaac, Larryand WilliamR. Kelly
1981 "Racial insurgency,the state, and welfare
expansion:local and nationallevel evidence
fromthe postwarUnitedStates." American
Journalof Sociology 86:1348-86.
Jennings, EdwardT., Jr.
1979 "Civil turmoil and the growth of welfare
rolls: a comparativestate policy analysis."
Policy Studies Journal7:739-45.
Levitan, Sar A., William B. Johnston and Robert
Taggert
1975 Still a Dream: The Changing Status of
Blacks Since 1960. Cambridge,MA: Harvard University Press.
Lurie, Irene
1968 An EconomicEvaluationof Aid to Families
with Dependent Children. Washington,
D.C.: The BrookingsInstitution.
Moynihan,Daniel Patrick
1965 The Negro Family:The Case for Political
Action. Washington,D.C.: United States
Department of Labor, Office of Policy
Planningand Research.
National Advisory Commissionon Civil Disorders
1968 Reportof the National Advisory Commission on Civil Disorders.Washington,D.C.:
U.S. GovernmentPrintingOffice.
O'Connor,James
1973 The Fiscal Crisis of the State. New York:
St. Martin's.
Patterson,James T.
1981 The WelfareState in America, 1930-1980.
Durham, England: British Association of
AmericanStudies.
AMERICAN SOCIOLOGICALREVIEW
Piven, Frances Fox and RichardA. Cloward
1971 Regulatingthe Poor:The Functionsof Public Welfare. New York: Vintage.
1979 "Electoral instability, civil disorder, and
relief rises: a reply to Albritton." The
American Political Science Review
73:1012-19.
Ranney, Austin
1971 "Parties in state politics." Pp. 82-121 in
HerbertJacob and Kenneth Vines (eds.),
Politicsin the AmericanStates: A Comparative Analysis. Boston: Little, Brown.
Rodgers, HarrellR., Jr.
1979 Poverty Amid Plenty: An Economic and
Political Analysis. Reading, MA:
Addison-Wesley.
Schram, SanfordF. and J. PatrickTurbett
1983 "The welfare explosion: mass society vs.
social control." Unpublishedpaper available from the Departmentof Political Science, State University College, Potsdam,
N.Y. 13676.
Stiner, GilbertY.
1971 The State of Welfare. Washington,D.C.:
The BrookingsInstitution.
United States Departmentof Commerce
1%0- Statistical Abstract of the United States.
1973 Washington,D.C.: United States Departmentof Commerce,Bureauof the Census.
United States Senate, Government Operations
Committee, Permanent Subcommittee on Investigations
1968 Staff Study of MajorRiots and Civil Disorders, 1965-1968.Washington,D.C.: U.S.
GovernmentPrintingOffice.
Wolfe, Alan
1977 The Limits of Legitimacy: Political Contradictions of ContemporaryCapitalism.
New York: Free Press.