A model election system

Material Information

A model election system
National Municipal League
Place of Publication:
New York
Publication Date:
Physical Description:
xi, 68 pages : ; 23 cm


Subjects / Keywords:
Voter registration -- United States ( lcsh )
Elections -- United States ( lcsh )
Elections ( fast )
Voter registration ( fast )
United States ( fast )
bibliography ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references.

Record Information

Source Institution:
Auraria Library
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
00875324 ( OCLC )
74160978 ( LCCN )
JK2160 .N35 1973 ( lcc )
324/.241/0973 ( ddc )

Auraria Membership

Auraria Library
Literature Collections

Full Text
National Municipal League

Carl H. Pforzheimer Building
47 East 68 Street, New York, N.Y. 10021
y \j


Copyright 1973 National Municipal League
Price $2.50

The Model Election System presented here is an extension of the
National Municipal Leagues major objective: the establishment of gov-
ernmental institutions and procedures that serve the public justly and
efficiently and are subject to popular control. This objective accounts for
the Leagues long-standing concern with the reform of election systems.
The League first established a committee on electoral reform in 1920.
In 1927 a successor committee on election administration, chaired by
the late Charles E. Merriam of the University of Chicago with Dr.
Joseph P. Harris serving as secretary, launched a comprehensive nation-
wide study of all aspects of election administration. This study, spon-
sored by the Social Science Research Council, resulted in publication of
a comprehensive report that gave particular attention to the need for
improving registration. Existing systems were then described as incon-
venient to the voter, expensive in operation and ineffective in prevent-
ing fraudulent voting. The report was published by The Brookings In-
stitution and included the Model Registration System and Model
Election Administration System proposed by the NMLs committee.
These models, published in 1927 and 1930 respectively, established the
agenda of electoral reform for more than a generation and were influ-
ential in providing guidelines for massive changes in American elec-
tions, particularly in the elimination of blatant fraud.
Now, almost two generations later, sweeping changes in voting rights
have raised new issues of election administration and voter registration.
The growing concern of the federal judiciary and the Congress with
these matters points to the need to review current practices and reeval-
uate earlier NML model proposals.

A grant from The Ford Foundation has made it possible for the NML
to undertake this new study of American election systems. At an early
phase a parallel program of the League of Women Voters Education
Fund, also financed by The Ford Foundation, identified administrative
obstacles to voting. In contrast, the National Municipal Leagues study
has sought legal reforms that would overcome these obstacles and fur-
nish guidance for developing an election system which would provide
maximum convenience for the voter and prevent fraudulent practices.
A special committee, including representatives of groups concerned
with obstacles to voting, provided assistance for the joint activities of
the LWVEF and the NML. Two League presidents have provided leader-
ship for the Election Systems Project. William W. Scranton, president
when the Project was launched and an active participant at all stages of
the undertaking, designated Wilson W. Wyatt as chairman of a special
advisory committee on election systems. In November 1972, Mr. Wyatt
was elected League president. He has guided the deliberations of the
advisory committees and chaired special sessions on election systems
at the 1971 and 1972 National Conference on Government. He also has
worked closely with the staff in the preparation of the Project reports.
The Project has been directed by Richard J. Carlson, who previously
served as research coordinator of the Illinois Constitutional Convention.
Major research assistance during the greater part of the project has been
provided by Jeanne Richman, former vice president of the League of
Women Voters of New York State. NML Assistant Director William
J. D. Boyd has been responsible for general supervision of the Project.
The broadly based advisory committee included election administra-
tors, political leaders, academic experts, civic reformers and others with
long experience in election problems. This committee has evaluated
Project studies and proposals. Special acknowledgement is due Profes-
sor Joseph P. Harris whose earlier work established the basis for League
programs in the field. He gave the Project the perspective of more than
40 years of intimate involvement in election problems. Former League
President Richard S. Childs also provided a link with the Leagues early
Many state and local election officials have contributed invaluable in-
sights into the practical problems of election administration. Members of
the Project staff visited 21 states and were greatly assisted by the re-
actions of election officials from 46 states at a special conference of the
National Association of Secretaries of State co-sponsored by the League
in New Orleans in February 1973. The Project also benefitted from the
Leagues similar role in the annual conference of the International Insti-

tute of Municipal Clerks and from its innumerable contacts with local
election officials. The relatively high rate of voter participation in
foreign democracies prompted comparative studies of election systems
in six countries, with particular attention given to the administration of
the 1972 Canadian federal election. Special acknowledgment is due to
J.-M. Hamel, the Chief Electoral Officer of Canada, and to the officials
in other countries who were most generous in their cooperation with
the League s staff.
Extensive assistance was provided by a number of special consultants.
Richard Scammon, Richard G. Smolka, James F. Blumstein, Robert R.
Outis, George Braden and W. Edward Weems, Jr., all made important
contributions. The League is ailso indebted to researchers who provided
assistance in making state-by-state analyses of election laws.
In addition to the Model the Election Systems Project includes a
major study of the costs of election administration and a volume of es-
says on major issues of election reform. Release of this series by the end
of 1973 will complete the major portion of the Projects publication
schedule. Its resources, however, will be available through the Leagues
continuing role as a clearinghouse. This will include periodic reporting
in the National Civic Review and the availability of up-to-date files
on developments in the field of election reform.
William N. Cassella, Jr.
Executive Director
National Municipal League

mwrmm-3 ,
?J t-fiw
.-*i*&.&& fxmimmfri.$pfe**twfe*tti
ifea** WMl#^;i^^-a.--teiiit pstitStfa.
ta4*>. .*.* $i>. ***/ a:s,; a^asM-
d£?v * .:*'>!' n*#^#-'li *#s:iepiaB'.
* Jti&m'M:'' -V;- -i ^a#if . -i ;4. .mim.*Mv-^um-kr^ Iflr^^^rfS f&Vfcs& &£§£
^.^rvSf. **v > :>i ^rfs*?
m, <$<#&$ #fe *-* && x^hMm-
W&#Mbr-.*:ifi* j^W43i*
: '.-'* ?. - wmvk
*p£ \* * *\>./'* , :. '
H* ,%<-'t?. <*. ^.^vifi/j js*tj
i*2'"VW ^ fr***^-*5
I .$# *4"!^ ;9'-*- j§K :;:*.* ^ >V.v-;^ -ti5F^^:.
,.' #&> &**(!%; ;**> ># ?'f dj%f'***( ;,i.. :.^|' '**
&¥$*' '#**- Vi #f i&fpfcM ?>x*&ii z&pf %**.-'&>> ;*>:£ii&i#** ' \
-,-i ;. <*$s-f^JU*w4* >*> ***& iM*>M0^'
^ ~U iSi'kj t&-'&jiVS- J^. ^^fif
If*.:*r^>'.f,v jni@gmw ffiwhwt;-tv -v
Ptir&t--1*' fr-ewrA $> C^s^-iu* * ;V.4 ,tK '^^'^.iMf^:.;
"^:::.VV- ' ' ; -. :' ' - -SIT''
ivfea-y.;^#*-' ii^- i4k.^" n- ot^i4*:'3*i^^'-:.-?~ ^-^-'
M$m'':frtet %& fomikski ;
4H $&$&*. sfrft yUked ^¥>'i§: Kr '^*00
rl. f- il '
.* ^ t^^r-w. tW' a.^' rT.^ i^-v
*%# ^'^^\.^:,'i^i^imim ihior-
- ., ... .H-"..-- '
1 '' .''i .: )" V- '
' :; f, r .
: ;'

s - ...
= T;: >.'.$; _.v;
' :.} .;: ',

Table of Contents
Foreword iii
Introduction: The Uses of a Model ix
1. The Need for Reform 1
2. A Model Election Administration System 10
3. A Model Voter Registration System 23
4. Absentee Voting 37
5. The Consequences of Complexity 44
6. The Federal Role in the Conduct of Elections 51
Appendix: Implementing Statute 57
Revised Suffrage and Elections
Article, Model State Constitution 63
Advisory Committee on Election Systems 66
Supplemental Remarks 68

V.v:*: y'
,^V;V* : >. ;.V-;

i 'x- :-i A r .v-4 -, V; ^
i *:^ U': J---; 't-?; i. : . --.U;
>Sf;: ;^;v : \ Wfi'^ ;'v v'
: ;
. /v; -X-
' '
: ~.f. r\ '' ^ ... .": ,i,:
> "i .- "'' .'.'^,-y V':V
' ' Vt;' -i: 1
i .:y:X \^y4'<--
Vy. .' '?' ...' 1
: ' i- : " ' v V r'v
. - > i<> 12- /

.;' i'r\ * y/'
. ; ',V-. '

* . > : t ,.
' ''. . '
./. -. ;;%',--v;
"r: ,':,fc;.
'-atit^tfcoD^o aids!


* *
> ... -
-n ' rS*' !>, ^ ~*("' '
booh/ -Is -JT '
/ / '
'..''/'y. - ; ' ; t. ;...
; eV/'$wht)>)4 h ,4*'
4 .
fe :':
'frY. '.'

. 'fry*'
T: UXfi.
..,- - v .
' :* _>/ J
&stktfitcQ: V. '. ,3
si*5K>?fa^5f^ >i> s4i i& ''# -srft UsT
;..\\ \ -'''"' ' 7 " ;- . .
IlyrAtHamStcpt*J' .lK5iSKfq/.'
ifeotteofi bon ""'
alas? \n\vz'0t
' "* 1', ' .'
ik> ~muv4iid&


''-XvV '-'.

- c -V.':i^r

The Uses of a Model
Strictly speaking, there can be no such
thing as a Model State Constitution
because there is no model state ...
From the introduction to the
Model State Constitution
Since its founding in 1894 the National Municipal League has
prepared model laws and systems in many areas of governmental con-
cern. The purpose of these models is to help citizens and governments
find practical solutions to their own problems and to enhance the federal
system by strengthening the structure of state and local governments.
The Leagues models are not tailored to the needs of any single state, yet
they may answer some need in every state. They establish goals, set
standards and propose solutions; they also bring the innovations of one
state to the attention of the others. Above all, they are designed to pro-
vide states with guidance in adapting the best current practices to their
particular needs.
The Leagues models emphasize flexibility because of the great diver-
sity that exists among the states. Although these differences have at

times allowed for the dominance of parochial interests, on balance the
sharing of governmental powers has kept the American political system
dynamic. In fact the major strength of the federal system lies in the flex-
ibility the states have to operate within varying conditions. The federal
government should avoid imposing measures that could depress the
states ability to develop solutions of their own. At the same time, states
must recognize that they have common tasks to perform. If states default
on their responsibilities they may lose the opportunity to solve problems
in ways that are compatible with their own institutions and traditions.
The health of the federal system depends on preserving a careful balance
between the two levels of government.
The Model Election System presents a comprehensive series of recom-
mendations, with appropriate alternatives, for the improvement of voter
registration and election administration among the states. Its central
recommendation is for full government responsibility for initiating the
registration of voters through a system of house-to-house canvassing.
However, states will not be able to implement the registration proposals
effectively nor provide uniform voting opportunities unless they restruc-
ture their systems for administering elections. The Model includes rec-
ommendations for an election administration system as well as a voter
registration system. An implementing statute which illustrates how the
preferred recommendations of the model system can be put into stat-
utory language is attached as an appendix. An updated version of the
Suffrage and Elections Article of the Model State Constitution, revised
to incorporate recent constitutional developments and congressional
legislation, is also included in the appendix.
In preparing the Model Election System the National Municipal
League set some limits on consideration of the many problems confront-
ing voters and election administrators. If the Model ignores issues asso-
ciated with the nominating process and with rules for campaign prac-
tices and financing, it should not be understood as a failure to recognize
their importance. The Model is addressed to administrative practices
that discourage voting; its primary objective is to develop procedures
that will expand the use of the franchise.
The Model is not offered as a prescription for uniformity among the
states but rather as a set of guidelines for reform. To the extent possible,
its proposals are discussed in the framework of alternatives that states
might use to accomplish their common tasks in a manner best suited to
their own customs and political preferences. It is to be hoped that by
avoiding rigidity the Model will encourage states to consider the sub-
stance, if not the exact form of the major recommendations summa-
rized below.

introduction: the uses of a model xi
A Model Election Administration System
1. To exercise its responsibility for providing uniform registration and
voting opportunities, a state should centralize authority over elec-
tions in an administrative office headed by a single officer of state
2. At a minimum' a state should finance voter registration, training
for election workers, and state-mandated meetings of election per-
3. The states Chief Electoral Officer should have general authority to
implement the law, establish rules for procedures and supervise the
election system.
4. Each state should establish an Election Council to provide partisan
balance in the administration of elections.
5. A single officer at the county level should be answerable to the Chief
Electoral Officer for the local administration of registration and vot-
6. A single official should be responsible for the conduct of elections
within each precinct or voting district.
A Model Voter Registration System
1. All voting precincts in each state should be canvassed annually or bi-
ennially to locate eligible voters and to add their names to the regis-
tration list and to remove the names of voters who no longer reside at
their registered address.
2. The statewide canvass should be conducted through door-to-door vis-
3. Canvassers should be qualified citizens carefully selected and
trained in their duties.
4. The statewide canvass should be followed by a short period during
which errors in the registration list may be corrected and the names
of voters overlooked during the canvass may be added.

' ' ^

, A'--:
fcfcl'SW A
hsm -H* '}%>::
^fv*:-v^--ua- %'*>?* * ^i4 k?v:;. A'Vvyifls
jg*& $&*>
.4.0^, fe s-v-. - At tho
*& .-*/: 4i ,V'. '^wr.-c-- /* o*** ytsti&sti^.-* 4j*s-
# sMft' ftMftftlfeh*
: A-4 ''?*#&<: ?y / a.£m>rV
f-i,-A :.. ,f . , ..'f -.v .- .<' .£ ,a> r^->.r,-SU*51H'?^Pwfe
*.!-*=:.- .,- A V. #11 *;i'i, ipisJf '
Ar-f ?^0Kk.!*..v" ,:' .'jy-.x^'4 ': ,uf* J&tO-a^iC--
. c^%l 1&J§9@^" jil'f
* r.-ju..' ,.. j'J ->...*.a*i aamtoh*. bartf'srwvj it^A.
&*- ^-y^'^-e- " k-.ic A-jfh :
flixi *&t0l :'ASritj^ ii'J* U**' .-:='' '^r; 4 / -i*~ _
"iA ;?:W: j>>i;:!ds;t*j^-gfc&**#*#{ M&tfki.
A<-A? f--Afv,^^tv*si&£ :p&kh\\ii flf^tme^.. h- i$;t. W &>,i*fi %^--fey
$pM-#> ^ mi ;& *%*&. '$&**; hi >ik- W$4t- .tmttma^
f fc:f -tl Wfc-iA ' . - - -

i V.'.i^i.- /i
V, wf
> ...'U
;-' . v^.'.-"
: :i: '
: istiv -,:.

The Need for Reform
In the 1972 presidential election the legally eligible electorate
was the largest in American history. A constitutional amendment had
added 18-, 19- and 20-year olds, and the United States Supreme Court in
a major decision had invalidated state laws that specified length of res-
idence as a qualification for voting. Congress, exercising its broad au-
thority to regulate federal elections, had passed legislation providing ex-
panded absentee voting and registration opportunities in voting for
President. Yet despite these developments nearly half the voting age
population failed to vote. It was the lowest turnout since 1948.
Some critics concluded that the reforms had made no difference. They
argued that the apathy of individual voters remained the basic obstacle
to voting. The results of the National Municipal Leagues two-year study
of election administration in the states suggests that there are other sig-
nificant causes of non-voting that do not support this conclusion. The
ejection of 1972 is evidence of the failure of existing procedures to regis-
ter a large percentage of the newly eligible voters and to facilitate their
participation in elections. The major premise of the Model Election Sys-
tem presented here is that the problem of non-voting in America is di-
rectly related to the machinery states have created for registering voters
and administering elections. The Model is designed to provide states
with guidelines to restructure their administrative systems so that
citizens have greater access to the electoral process.

The Dimensions of Non-Voting
The percentage of voting-age Americans going to the polls in presi-
dential elections has declined from 63 percent in 1960 to 62 percent in
1964, 61 percent in 1968 and 56 percent in 1972. During this same pe-
riod over 80 percent of those voters who were registered usually voted.
The non-voting population is now larger than the entire electorates of
such democracies as France, England, Canada and Australia. There
were 39 million non-voters in 1960, 43 million in 1964, 47 million in
1968, over 61 million in 1972.
The United States has traditionally had the lowest turnout for national
elections among all the democracies. For example, the average turnout
for parliamentary elections in France and Great Britain since World
War II has been 80 percent, in Canada 76 percent, in Austria 95 percent
and in West Germany 86 percent. Taken together these other democ-
racies average nearly 84 percent turnout in national elections, or 24
percentage points above the comparable figure for American presi-
dential elections and about the same level of turnout as that among reg-
istered voters.
If the American record of voter turnout in presidential elections is
poor, then participation in elections for other units of government, those
usually considered closer to the people, must be described as dismal.
While more than one-third of our adults do not vote in presidential elec-
tions, more than half do not vote in congressional and state elections and
even fewer participate in local government elections. Indeed, it is not un-
common for the vote in municipal elections to drop to less than one-
fourth of the qualified voters.
It is difficult to reconcile our self-image as the most democratic of
democracies with our relatively poor showing at the polls. The search for
an explanation points to a significant difference between American elec-
tions and those of other democracies. Voting in the United States is a
two-step process. First the voter must travel to a registration office and
establish his eligibility. Only then will he be allowed to vote. Virtually
every other democracy places the burden for registration on govern-
ment, and it is a government agency that has the responsibility for locat-
ing voters and entering their names on the voters list. Canada conducts
systematic house-to-house canvasses prior to each election. In England
the town clerk is responsible for registering voters through house-to-
house or other sufficient inquiry. He compiles registration lists through
an annual mail canvass of all households, supplemented by selective
house-to-house canvassing. These systems of government-initiated reg-
istration make voting far more accessible to the citizen than in the
United States.

Development of Registration in the United States
Although the first registration laws were enacted as early as 1800, reg-
istration requirements were not the general practice until after 1860.
Early in American history most people lived in small, closely knit com-
munities or in rural settlements, and in most areas everyone knew his
neighbors. The problem of identifying eligible voters at the polls was
minimal. The quality of American society changed dramatically after the
Civil War with the growth of large cities and the beginnings of mass im-
migration. It was no longer easy to identify ones neighbors on election
day. These changes hastened the emergence of machine politics in the
bigger cities and the advent of large-scale election frauds.
As industrial expansion, urbanization and mass immigration changed
the nature of American life, states began to adopt registration laws. Re-
formers hoped the requirement of registration would reduce voting
frauds and lessen the influence of urban machines and their immigrant
clienteles. Without the safeguard of prior registration, non-residents
often appeared at the polls and successfully demanded the right to vote
and repeaters were often used to cast votes in more than one precinct.
When ineligible voters are allowed to vote, or if individuals cast more
than one vote, the legitimacy of the democratic process is undermined.
Registration before an election enables officials to permit only eligible
voters to participate on election day. Between 1860 and 1880 the older
northern states began to require registration in large cities, followed by
many southern and western states beginning in 1880. In other parts of
the country registration was extended to small cities and, in a few in-
stances, to rural areas. While other democracies were meeting the elec-
toral problems of industrial society with government-initiated registra-
tion systems, the American states placed the registration burden on
individual voters.
Most of the early registration laws were based on the theory that
the lists had to be frequently discarded and a new registration held in
order to purge voters who had died or changed residence. Such systems
were considered to be particularly appropriate for highly mobile urban
populations. But dissatisfaction with the great expense and incon-
venience of periodic registration led reformers to advocate permanent
personal registration. Under a permanent system a voter who registers
once remains on the list as long as he continues to reside at the same
address and votes. Registration is conducted in a central location
throughout the year and voter lists are kept current through various
purging techniques such as official death reports, transfers of regis-
tration, a house-to-house canvass and cancellation of registration for
failure to vote. Permanent personal registration is generally less expen-

sive than periodic registration and more convenient for the voter.
Large cities were again the testing ground for this reform. Boston has
had permanent personal registration since 1896, Milwaukee since 1911
and Omaha since 1913; Oregon enacted it statewide in 1915. Perma-
nent personal registration subsequently became a major issue during the
1920s. Local civic groups such as the Bureau of Municipal Research in
Philadelphia, the Citizens League of Cleveland and the Chicago Bureau
of Public Efficiency added registration reform to their programs. In Jan-
uary 1927 the National Municipal Leagues Committee on Election Ad-
ministration issued the first Model Voter Registration System which
contained a detailed set of specifications for a permanent registration
system. The central features of this system remained part of the League
program through four revisions, the latest in 1957. The League of
Women Voters of the United States placed permanent registration on its
legislative program in 1928 after three years of study. Permanent per-
sonal registration has gradually become the dominant practice among
the states. Every state except North Dakota now provides for some form
of permanent system, although in two states registration is not required
The shift from periodic to permanent systems of personal registration
was an important, if limited, advance in making the procedure more
convenient for the voters who, although they still had to make a personal
visit to a registration office, did not need to go back and reregister at reg-
ular intervals. The number of voters registered under permanent per-
sonal registration has been higher than under periodic systems and turn-
out also seems to have increased slightly. However, permanent regis-
tration has proven to be only a modest improvement over earlier systems
for two reasons. First, the purging practices necessary to keep the lists
accurate often force voters to reregister frequently, significantly reduc-
ing the major advantage of permanent registration. When voters are
purged from the lists for failure to vote every two years or in specified
elections, the word permanent is hardly an accurate description of
the registration system. Second, voters must still take the initiative.
Present Registration Systems Discourage Participation
Each year millions of eligible voters remain beyond the reach of even
the most ambitious and aggressive registration efforts. In 1968 no more
than 77 percent of the voting age population was registered. More than
30 million adults were unable to vote because they were not registered.
In the non-presidential election year of 1970, the percentage slipped be-
low 74, which meant there were over 32 million unregistered adults.2

In 1972 the registration percentage again rose to about 76; that still left
32 million voting age Americans unregistered.
Personal registration of any kind impedes voter participation. A study
of participation in 104 of the nations largest cities in the 1960 presiden-
tial election found that differences in registration rates explained most
of the differences, between cities in the number of voters who partic-
ipated in that election. Cities with higher rates of registration had
consistently greater turnout rates. The authors concluded that registra-
tion requirements are a more effective deterrent to voting than anything
that normally operates to deter citizens from voting once they have reg-
istered.3 Another study examined voter turnout in counties that adopt
personal registration systems. The study included such states as Ohio,
Missouri and Pennsylvania and found that as counties adopt personal
registration, participation drops 8 to 12 percent from previous levels.4
Participation among voters who do become registered is usually around
80 percent for the nation, indicating that people do vote in presidential
elections once they overcome the obstacle of personal registration.
Efforts to Expand Registration Have Been Ineffective
In most states local officials have broad discretion to provide their
communities with outreach registration opportunities through branch
offices, mobile registration units, deputy registrars, and evening and
weekend office hours. This is an important discretionary power because
election officials can control local registration rates by expanding or
contracting the range of opportunities available. Local officials do not
always use these powers in a constructive way. In some instances they
are hampered by staff shortages and budgetary limitations; in others
they may limit registration opportunities, either to protect established
political interests or because they feel that registration and voting
should be a rugged test of voter commitment. In either case the result is
an unacceptable constriction of voting opportunity.
Much of the current effort to bring registration opportunities closer to
individual voters depends on the volunteer activities of civic and political
groups who staff mobile registration facilities or serve as deputy regis-
trars. However well-intentioned these efforts may be, they often are lim-
ited by the goals and capabilities of the sponsoring groups. Political
parties and candidates are obviously interested only in seeing the names
of their supporters added to the rolls. Many nonpartisan groups direct
registration drives at selected groups such as blacks or college students.
Regardless of sponsorship, outreach efforts or registration drives in-
variably have failed to produce significant increases in the number of

registrants. Short-term gains are soon lost through population mobility.
Even communitywide registration efforts are limited by what is possible
under existing systems. A recent national study of voter registration
drives concluded that there seems to be a limit ... to what one can ac-
complish by methods that ultimately rely on initiative by the individual.
. . The data collected for this volume document a history of limited re-
turns, not commensurate with the effort and money expended to increase
the numbers of eligible voters within the structure built into the laws of
all but one (North Dakota) of the fifty states.5 Experience indicates
that registration drives are far more effective in training political vol-
unteers than in registering large numbers of voters.
The Prevention of Fraud
A registration system will be effective in deterring fraud only if it pro-
duces a list of eligible voters that is accurate and up-to-date. The prob-
lem has been well stated by Joseph P. Harris in a pioneering study of
voter registration in the United States: The existence of a defective reg-
istration, with inaccurate or padded lists, constitutes no protection
against frauds. On the contrary, such a registration is an open invitation
to fraud.6 Permanent personal registration has consistently failed to
measure up to this standard.
Candidates for public office who mail campaign literature to regis-
tered voters frequently complain that between 20 and 30 percent of
those whose names appear on official registration lists no longer live at
the address indicated. A 1971 national survey made by the League of
Women Voters Education Fund in cooperation with the National Munic-
ipal League revealed that approximately 50 percent of the organiza-
tions using registration lists found the lists to be inaccurate, and in half
those cases the inaccuracy was reported to be greater than 10 percent.7
Inaccurate lists are characteristic of permanent personal registration
because of the inherent difficulties of the purging process. Registrars
must spend as much, and often more, time and money on striking off the
names of ineligible voters as they do on adding the names of new regis-
trants. Frequent purging of the lists imposes on citizens the inconve-
nience of repeated reregistration. But, if purging is careless, inaccurate
lists may increase opportunities for fraud. Neither result serves the dem-
ocratic process.
To remedy these problems the Model Voter Registration System in
Chapter 3 seeks to combine registration and purging into a single pro-
cess accomplished through a system of door-to-door canvassing closely
supervised by state and local governments. Such a system should in-

crease dramatically the number of registrants in the most efficient and
economic manner while ensuring that registration lists are accurate and
A Leadership Vacuum
Registration reform under the guidelines of the Model will require
strong administrative leadership at the state level. Yet states have rarely
assumed such responsibility for the conduct of elections, as they have
for other state activities. Instead the responsibility has been shifted to
county and municipal governments. An estimated 7,000 governmental
units in the United States today have administrative responsibility for
elections. Lacking effective guidance from the state they are often left to
their own devices to interpret laws that may be vague or outdated. In
other instances, states try to provide for all contingencies through de-
tailed laws which require frequent amendment, often with little or no
attempt to integrate the changes into the existing legal framework. This
results in a set of ambiguous and contradictory provisions for local use.
While there are outstanding examples of creative and effective action
by local election administrators, many are unwilling to resolve ambigu-
ities themselves or even to use their discretionary powers in a construc-
tive way to expand voting opportunities. Some are understandably
reluctant to be innovators in a system which discourages local initiative,
while others are prevented from doing so by the rigidities built into the
election code.
The absence of a centralized system of administration also means that
the courts are continually being asked to intervene in situations which
should initially be subject to administrative remedies. The courts are sel-
dom in a position to balance administrative requirements with the
interests of the parties involved in an electoral dispute. Much current
election litigation could be avoided if the system were flexible enough
to deal with problems that arise and if administrators were accountable
for their actions.
Wide Variations in Voting Opportunity
State failure to supervise the conduct of elections means that there
will be wide variations within a state in opportunities available for reg-
istration and voting. These disparities are most troublesome in the regis-
tration process. State laws give local officials broad discretion to provide
branch and mobile facilities, to deputize registrars and to set hours at a
central office. The extent of the opportunities provided depends on local

resources and official willingness to serve the needs of the community.
Evidence suggests that many officials do not provide the full range of
opportunities allowable under law. The national survey by the League
of Women Voters Education Fund revealed the following situation:
In 29 percent of the communities where deputy registrars were allowed,
election officials failed to use this method to reach citizens. While only 10
states expressly forbid evening and Saturday registration, 77 percent of
the communities studied had no Saturday registration and 75 percent had
no evening registration in non-election months. Even during . the 30
days prior to the closing of registration, 38 percent of the communities
provided no additional hours for registration.
The study also reveals that many election officials fail to provide needed
services to assist voters and precinct election officers:
Although the law may neither require, suggest nor forbid it, an elec-
tion official might provide information to citizens concerning the election,
might conduct extensive training programs for all poll workers and might
provide bilingual clerks where needed. While such initiative would remove
many obstacles to voting, local officials have seldom acted in these areas:
only 11 percent of the local officials included in this study published
a voter information guide; 28 provided no training for poll workers; and in
approximately 30 percent of the registration places where bilingual assis-
tance was needed, local officials failed to provide this service.
The study concluded that Election officials clearly have the power to
make registration and voting procedures easier for citizens, but this
study has found that, by and large, they dont use it.8
The isolation of local election administrators has become a serious
liability in the face of recent federal legislation and judicial action. With
passage of the Voting Rights Act of 1965 and the amendments of 1970,
Congress has demonstrated increasing interest in legislating uniform
registration and voting opportunities among the states. In the last 10
years the United States Supreme Court has altered substantially the
ground rules for state voter qualifications. When Congress mandates
changes in absentee voting procedures for presidential elections or the
Supreme Court rules on the validity of state durational residence qualifi-
cations, thousands of local election officials across the country must
reach for ways to implement the changes with little guidance from the
state. Too often the result is confusion and ineffective implementation.
To overcome such difficulties, the Model Election Administration Sys-
tem proposes a centralized state system for the conduct of elections
with responsibility fixed in a single, visible state officialthe chief elec-

toral officerwho has adequate rule-making power. A single official in
each county or city should be answerable to the chief electoral officer for
the operation of the system. Structural reform of this kind would be an
essential first step in most states to make the voting process convenient
and equitable to citizens, and efficient and economic to government.
Richard J. Carlson, Personal Registration Systems Discourage Voter Participa-
tion, National Civic Review (December 1971), pp. 597-602.
--------, ed., Issues of Electoral Reform (New York: National Municipal League,
Democratic National Committee, Report of the Freedom to Vote Task Force
(November 1970).
J.-M. Hamel, Registering Voters in Canada: A Responsibility of the State,
National Civic Review (July 1972), pp. 336-340.
Joseph P. Harris, Election Administration in the United States (Washington,
D. C.: The Brookings Institution, 1934).
National Municipal League, Model Election Administration System (New York,
National Municipal League, Model Voter Registration System (New York,
1. Ohio and Wisconsin. North Dakota does not require prior registration.
2. U. S. Bureau of the Census, Statistical Abstract of the United States
(Washington, D. C.: U. S. Government Printing Office, 1970, 1971).
3. Stanley Kelley, Jr., Richard E. Ayres and William G. Bowen, Registration
and Voting: Putting First Things First, American Political Science Review
(June 1967), p. 362.
4. Waltern Dean Burnham, A Political Scientist and Voting Rights Litiga-
tion: The Case of the 1966 Texas Registration Statute, Washington University
Law Quarterly (Spring 1971), p. 346.
5. Penn Kimball, The Disconnected (New York: Columbia University Press,
1972), p. 290.
6. Joseph P. Harris, Registration of Voters in the United States (Washington,
D. C.: The Brookings Institution, 1929), p. 14.
7. League of Women Voters Education Fund, Administrative Obstacles to
Voting (Washington, D. C., 1972), p. 15.
8. Ibid., pp. 7-8.

A Model Election
Administration System
The conduct of elections in the United States is big business
with no one at the head of the firm. State failure to assume major
responsibility for supervising the administration of elections has left
thousands of county and city governments to cope with complex and
ambiguous codes. Election administration is an orphan of state govern-
ment, underfinanced and relegated to an obscure position. The periodic
nature of elections encourages state and local officials to treat them as
part-time concerns; these officials are often chiefly responsible for other
major duties. This high degree of localism also inhibits professionalism
and permits widespread variation in voting opportunities within a single
The Model Election Administration System also recommends central-
ized state control over the conduct of registration and voting, built
around two principles essential to the sound administration of elections:
First, the conduct of elections should be a separate and distinct govern-
mental responsibility and should be treated as such in the structure of
state and local government. Second, election officials must be account-
able to the public they serve. The visibility and accountability of an elec-
tion system is basic to its good management. Above all, voters must be
able to understand procedures if elections are to serve the public.

To exercise its authority for providing uniform registration and vot-
ing opportunities, a state should centralize responsibility over elec-
tions in an administrative office headed by a single officer of state
Each state should establish a separate State Office of Elections
headed by a Chief Electoral Officer (C.E.O.) who should exercise the
broad supervisory^ and administrative power necessary to implement
state law. The manner of selecting the C.E.O. will be governed by the
constitutional and political imperatives of each state. In keeping with the
integrated executive concept of the Leagues Model State Constitution,1
this Model recommends that the Chief Electoral Officer should be ap-
pointed by the governor, subject to the advice and consent of the Senate,
and be subject to removal only for cause and after appropriate hearing
procedures. The C.E.O. should serve for a term of five years or more to
encourage his political independence and to give him time to develop the
expertise his office requires. In any case, he should be a full-time official
with no responsibility other than elections.
In most states, an elected or appointed secretary of state has authority
over elections. If the State Office of Elections is to be under the secre-
tary of state, he should be required by law to appoint a special individual
to head it. This would relieve the secretary of the need to divide his at-
tention between elections and other duties. The creation of a separate
office for elections is essential to its visibility whether that office is
located with the governor or with some other state official.
It should be recognized that any method of selecting governmental
officials will have political overtones. While elective posts are more di-
rectly subject to such pressure, appointments made by elective officials
are not wholly free of partisan implications. Certainly, degrees of parti-
sanship are apparent among election officials in many states, a condition
frequently accepted as an inevitable reflection of the party interests at
stake in the election. Thus it is important for a state to distinguish be-
tween the methods for selecting its Chief Electoral Officer, which may
vary, and the attributes of his office, which should be constant. The
selection of the Chief Electoral Officer should not depend on his political
views and policies but on his ability to carry out the responsibilities dis-
cussed below and on his expertise in election procedures.
Several approaches are available to insulate the state administration
of elections from partisan influence. The visibility of the Chief Electoral
Officer is perhaps the primary restraint. Longer terms, better funding
and more emphasis on trained staffs would promote the professional
standards that tend to offset purely partisan considerations in adminis-

tration. Bipartisan controls, discussed in a subsequent section, may also
be necessary.
Americans tend to have little confidence in the ability of the states to
separate the administration of elections from the heat of campaigns.
Few if any states have yet achieved the attitude prevalent in foreign
democracies that election procedures may be impartially supervised as a
government service. The widespread practice of giving administrative
authority to cumbersome boards representing multiparty interests re-
flects a commonly felt need to guard against abuses. Admittedly, it
may take years of effective upgrading of election administration before
most states would be willing to reconsider present structures. In the
meantime, the nature of the safeguards a state raises against any mis-
use of political power will vary according to its political tradition.
The Chief Electoral Officer should have general authority to imple-
ment the law, establish procedures and supervise the state election
system. This authority should include the following specific powers:
The Chief Electoral Officer should provide written standards and
directions to county election administrators for carrying out local
responsibilities for registration and voting.
The Chief Electoral Officer should provide rules for compiling and
maintaining voter registration lists. In addition, he should set standards
for selecting such personnel as precinct workers and canvassers, and
should provide criteria to be used locally in selecting voting equipment
and supplies. The C.E.O. should design the procedures by which local
election officials are to carry out their duties throughout the year and
are to process and count the ballots on election day. All rules and regu-
lations should be in writing and well publicized.
State election officials agree generally on the need to upgrade election
procedures by providing more technical guidance to local officials, par-
ticularly in such areas as the utilization of electronic data processing
techniques. Differences of opinion arise over the degree of authority
state officials should have in imposing standards. The Model assumes
that effective implementation of the standards requires a state election
office with true executive power. There is another view, however, that
the state could improve local election administration effectively by pro-
viding operating models and advice. The existence of statewide guide-
lines might improve the conduct of elections in many states, even
where the C.E.O. lacked the necessary power of enforcement.
The state should arrange for periodic conferences of county election

The Chief Electoral Officer should call a conference, financed by the
state, of all county and city election administrators at least once a year.
This would give every administrator access to the same interpretation
of state law and rules, and would also give each an opportunity to ex-
plore and resolve any ambiguities or objections to state directives. Also
the C.E.O. would have the opportunity to communicate directly with
county officials on legislative action or on new court decisions, estab-
lishing a single interpretation to be applied uniformly across the state.
Finally, the conference would permit county and city election adminis-
trators to share experiences and learn about new techniques and would
give the C.E.O. greater insight into the workability of state rules.
In several states, election officials meet informally to exchange infor-
mation but few states require such conferences. Most states and local
election officials have no continuing opportunity to discuss common
problems or ways to implement legal changes. The absence of such for-
mal channels of communication seriously impedes uniform application
of state law.
The state should develop training programs for precinct personnel
and other election employees.
State policies are no more effective than the people who execute them.
To promote equality of electoral opportunity for all its citizens a state
must be able to set performance standards for election employees. This
can be accomplished only if a central source lays out all training programs
and has the power to supervise their uniform implementation through-
out the state.
While the training should be conducted by the county administrator
or his staff, the state election office should design the programs, and
provide the materials and directions for delivering instruction. Because
the training would be a function of the state system, it should be fi-
nanced by the state.
Few states provide standard training for election workers. Many states
publish instruction manuals, but the quality varies enormously and in
few cases does the state mandate their use. In effect it is a matter of local
option whether the county election chief carries out training at all. One
difficulty commonly encountered is that election day workers are often
recruited through the parties on short notice and there is little time to
qualify them for the duties they perform.
The Chief Electoral Officer should be responsible for providing voter
The C.E.O. should provide and supervise the dissemination of voter

information on the time and places for state elections, procedures for
casting ballots, explanation of state offices and issues, services available
to the voter such as assistance at the polls, absentee voting opportuni-
ties, challenge procedures and an explanation of the function of the
precinct workers the voter is likely to encounter on election day. Many
states carry out one or more of these services but few have succeeded in
providing all the necessary information in a form that is readily under-
standable to all voters. Voters would benefit from the official distribu-
tion of such information along with a sample ballot.
Too often the county election office lacks funds to supply voters with
useful information on local elections. Each voter is entitled to basic
knowledge about local elections, but the information he needs will vary
from locality to locality. The costs of providing it should be shared by
the state and localities, and the state should supervise the development
of information programs. In addition the C.E.O. should publish and dis-
tribute to each county maps of all election districts for state and fed-
eral office so that the county administrator can establish precinct bound-
The Chief Electoral Officer should report regularly on all matters
pertaining to elections.
Regular reporting by the C.E.O. is a key method for establishing the
accountability of his office. Certainly the legislature and the public are
entitled to an annual report of all election expenditures. Even in a year
with few or no elections, the administrative costs would be of general
interest. On other election matters, a state may wish to provide for
longer reporting intervals. For example, states that hold elections every
two years may not wish to require annual reports on registration and
voting activities. A state may even determine that a two-year interval
is necessary to give the C.E.O. time to evaluate election procedures and
to make recommendations for legislative change. Under any circum-
stances, the Chief Electoral Officer should provide reporting forms to
the localities so that he may depend on comparable information from all
areas in making his own determinations.
The Chief Electoral Officer should set budgeting and accounting
standards for election expenditures throughout the state.
Information about current levels of local spending on elections is hard
to come by.2 Few if any states require local offices to report and, in any
event, most of them have election responsibilities as an adjunct to their
other duties and may not separate out all costs. Some city or county elec-

tion authorities simply forward bills for their expenses to the local appro-
priating body for payment. In other places a budgetary limit forces elec-
tion officials to curtail voter services. Arbitrary ceilings on expenditures
can have obvious effects on the voter if, for example, election of-
fices cannot afford to stay open evenings or Saturdays.
Most states simply have no way of showingor knowinghow much is
spent on elections. With no basis for comparison states will have a hard
time evaluating the fiscal implications of administrative reform. Thus
the power of expenditure review may be a critically important financial
function of the state election office.
Complete state administrative control would include the authority to
budget for all functions of the system and the responsibility to pay for
them. In strict logic, a state thus should assume responsibility for all
election costs, but current realities demand consideration of alternative
approaches. At a minimum, innovations mandated by the state should
be financed by the state, including procedures recommended by the
Model for the statewide canvassing system, training programs and the
election conference.
A state could finance other election costs initially and charge them
back to the counties on the basis of the number of local offices at stake.
This reflects the view that localities should share in the cost of elections
at which contests for local offices are also at stake.3 Alternatively, a
state could set a standard of reimbursement prorated according to popu-
lation, or it could set subsidy levels for county personnel and equipment
utilized in the state election. Or the state could pay all the costs of certain
elections. A general election once a year could be a state fiscal responsi-
bility, while local or special elections held at other times might be an
entirely local expense. These are some of the ways states and localities
could work together to meet their respective fiscal concerns.
Even if it does not underwrite election costs a state should have a role
in overseeing expenditures by requiring county administrators to submit
local election budgets for review and comment, and by requiring them to
report their spending at the end of each year. Budgeting is basically a
planning process and accounting is the procedure through which expen-
ditures are reviewed. The state should provide standards for budgeting
and report forms for accounting to help localities improve their pro-
cedures. Cost information which the state receives from the counties
would enable it to establish a standard of comparability for spending,
which in turn would allow the C.E.O. to comment constructively on local
spending. States may have the means to improve administrative prac-
tices at the local level simply by knowing what various services cost.

The Chief Electoral Officer should have the power to investigate
violations of the election law, nonperformance of election duties,
and irregularities in the conduct of elections.
The power of investigation is important to the effectiveness of the
C.E.O.s administrative authority. His responsibility for overseeing
compliance with the election laws and rules of procedure is likely to en-
courage accountability on the part of his staff and by county officials.
His investigative role will also tend to increase the visibility of the state
office as the proper place for citizens to lodge complaints about the
workings of the system. The current anonymity of most election bu-
reaucracies often heightens a feeling of ineffectiveness among disad-
vantaged voters and discourages reform. The C.E.O. should therefore
establish systematic procedures for receiving and hearing complaints
against any election practice or official, and should make the avail-
ability of these procedures known to the public. The Chief Electoral
Officer has a variety of tools necessary to the investigative task, includ-
ing his supervisory authority over county and other election officials and
his responsibility for budget review.
Each state should establish an Election Council to provide partisan
balance in the administration of elections.
A major goal in election administration should be a strong, profes-
sional system that does not dilute the two-party system. However, pat-
terns of partisanship in some states may lead reformers to look skepti-
cally on efforts to separate the C.E.O. from party influence. The defini-
tion of the C.E.O.s responsibilities will depend to some degree on each
states ability to achieve professional administration, but as a general
rule his duties should not include decision making likely to impair his
credibility as an impartial administrator. To this end the Model recom-
mends that each state establish an Election Council as a political ad-
junct of the State Election Office. It is intended to provide an institu-
tional check on the C.E.O. in a way familiar to American political
In some states authority over elections is vested in a single bipartisan
body. Such states are already assured of political balance but they sacri-
fice the advantages of efficiency and accountability associated with a
single administrative officer. Under this structure it is often impossible
for the public or for the election board to distinguish its political from its
administrative function. The effect of the Models proposal for a State
Election Office headed by a Chief Electoral Officer working in tandem
with a partisan Election Council is to make these functions separable.

The Election Council may be either an advisory or a policy-making
body. The goal of the Model is an open, impartial and efficient election
system. To provide for it, a state will need professional organization and
centralized administration. To the extent that the administration may
be influenced by bipartisan interests, however, its authority must be
tempered by opposing interests. The Election Council as proposed should
fortify the C.E.O.s Administrative integrity in the public view by provid-
ing visible assurance that the conflicting political interests of the state
will act as a check on each other. Where public confidence is high in the
ability of the C.E.O. to free himself from partisan domination, the Coun-
cil might act as his advisory arm. Where the appointment of the C.E.O.
has uncertain political implications in the public view, the legislature
might want to give the Council final decision-making power in resolving
the questions assigned to it. In either case, the visibility of a bipartisan or
multipartisan board with quasi-judicial functions and the publication of
its recommendations on contested issues would give the state election
office political balance without eroding its administrative authority.
Bipartisan representation in election administration has long been used
to prevent abuses by any one political interest.
A state should define the Councils function through the statutory
duties assigned to it. Thus, in some jurisdictions the Councils principal
responsibility may be to investigate alleged fraud and irregularity in the
conduct of elections. In that case a state might want to give the Council
power to administer oaths, issue subpoenas, summon witnesses and com-
pel the production of books, papers, records and other evidence in con-
nection with its inquiry. If warranted, the Council could turn its findings
over to the state attorney general for prosecution. Such states might
choose to require the C.E.O. to file with the Council a list of all com-
plaints received in the state office, empowering the Council to investi-
gate and act on them. While the unbridled investigatory power of the
Council could permit it to harass the administrative authority, both of
the above approaches would provide undoubted political checks. In
other states the Councils function could be limited to challenges to
nominating petitions, recounts and certification of election results; or it
could include other politically sensitive responsibilities outside the scope
of the Model such as supervision of campaign financing and disclosure
laws. Clearly, the duties assigned to the Council will vary among the
states. If there are states where partisanship does not intrude on election
administration, the Council might even be unnecessary.
The Election Council should represent the major political interests.
In some states its membership would be limited to representatives of the

two major parties. In states where third or fourth parties show sub-
stantial strength at the polls it should be reflected in the composition of
the Council. Its members should be elected at party conventions, or
selected by party chairmen or state party committees. To encourage
manageable meetings the Council should not exceed seven members.
They should serve four-year, overlapping terms to assure adequate time
to develop expertise and to provide continuity in membership. Opposing
partisan interests should have alternating power to help give different
political interests an effective voice in Council decisions and at the same
time protect against deadlock.
For example, in a two-party state the Council would have five mem-
bers. Each party would have two members at all times. Although each
would serve a four-year term, the appointments would be staggered so
that each party would name a new member every two years. The chair-
man should serve a two-year term and the position should alternate be-
tween the two parties every two years. The short-term domination of
each party might then tend to discourage an abuse of power by either,
since it would be subject to quick retribution.
If a state should decide on a Council with purely advisory power and
entrust ultimate responsibility for implementing its recommendations
to the C.E.O., the composition of the Council becomes slightly less crit-
ical and an even number of members might be more acceptable. In
such states the C.E.O. should be required to account to the Council and
to the governor if he fails to implement the Councils recommendations
on any matter before it. Council meetings should be geared to the polit-
ical calendar and the C.E.O. should have the power to convene addi-
tional sessions as he deems necessary.
A single official within each county should be responsible to the
Chief Electoral Officer for the local administration of registration
and voting in all elections.
The most important member of the election system is the county of-
ficial who implements state policies and procedures.4 He holds the key
position as the link between the State Office of Elections and the voter,
and is traditionally the most accountable member of the system. Precinct
workers have few supervisory functions and the state officials are too
distant. The county administrator registers voters, receives applications
for absentee ballots and hears complaints about irregularities. He has an
essential role in supplying information to the voters for local election
issues and candidates and for publishing voting hours and polling loca-

In his capacity as agent of the state, the county administrator must
have some discretion in applying the law and administrative rule to local
needs. He should be primarily responsible for carrying out the states
training program, for assuring the presence of adequate staff, equipment
and supplies, for overseeing the registration process, the preparation of
the ballots and the poll facilities, and for seeing that municipal and pre-
cinct performance levels meet state standards. Policies governing
county procedures and rules governing county election staffs should be
in writing for the convenience of local employees and for the information
of any party organization or voter who wishes to consult them.
In addition to the duties he performs on behalf of the state, the county
official has important independent functions. These include drawing
precinct lines within state guidelines. State law should set the upper
and lower limits of the number of voters in each precinct and require
an adjustment after each reapportionment to ensure that each precinct is
wholly contained within any larger district. This would ease the countys
administrative task in preparing ballots and voting machines, and in
simplifying information the voter needs when he gets to the polls.
The county administrator should also have power to appoint precinct
personnel to help him maintain uniform methods of operation through-
out his jurisdiction. Since the bipartisan presence is desirable at the
polling place (see below), it seems appropriate for the administrator to
make these appointments from lists submitted to him by the county
party organizations. He should, however, be able to apply state merit
standards and reject party nominees who do not meet them.
It is at the county level that the greatest need exists to improve stan-
dards of professionalism. Because of the episodic nature of elections,
the county must employ many part-time workers to conduct registration
and to perform election day duties. The temporary nature of his staff
places a real burden on the county administrator. It is the states re-
sponsibility to improve local operations by upgrading the county office
through professional guidance, fiscal assistance and training programs.
The most effective way to achieve an integrated chain of command
might be state appointment of county administrators.6 In the case of
elections the responsible individual is now often an elected or appointed
official with other responsibilities. As long as he can achieve efficient
management of registration and voting and a high level of visibility
for his role as election administrator, the manner of his selection is of
secondary importance. The primary consideration is that responsibility
for the conduct of elections be vested in a single individual in each
county. Certainly this objective is compatible with local selection, which

may even give more visibility to the post. In many instances a state can
exercise effective control over local election administration by financing
a major portion of the process, such as registration.
On behalf of local selection, it may be argued that the ability of each
county to name its own election official guards against a concentration
of political power that could result if the C.E.O. were able to name all
administrators from the ranks of his own party. In any event, there is no
conflict between a county official selected or appointed locally but
answerable to a state office. In many, if not most, areas of governmental
concern county officials act as agents of the state, carrying out its pol-
icies and laws in such areas as health, welfare and criminal justice.
A single official in each precinct should be responsible for the con-
duct of elections locally.
Almost inevitably, the weakest link in the election system is at the
precinct level where temporary employees are underpaid and required to
work long hours. It is difficult to recruit people for such duties and even
more difficult to insure their presence at a training program. Neverthe-
less, the quality of their performance will depend on the quality of their
training. Under these circumstances, the existence of at least one trained
official to oversee poll operations is essential. The county administration
should designate a single individual in each precinct to supervise elec-
tion day activities and be accountable for the conduct of voting.
The supervisory official should be responsible for seeing that precinct
workers identify voters, instruct voters on procedures when necessary,
watch over machines and enforce regulations against electioneering at
the polls. He should be responsible for supervising the inspection of ma-
chines and equipment before the polls open, for monitoring the closing
of the polls, and for counting and recording the vote where that is a local
function. He should also report on irregularities and malfunctions at the
The precinct supervisor would be a temporary employee but, like other
election officials, he should be barred from holding public or party office.
Many if not most election laws permit local officials to serve in minor
capacities such as village trustee, assessor or commissioner of deeds.
These so-called minor posts normally hold considerable local political
significance and the conflict should be avoided. However, public em-
ployees such as teachers or postal clerks should be able to serve as poll
workers if it does not interfere with the performance of their other duties.
There also appears to be substantial reason for continuing the com-
mon state practice of requiring precinct personnel to be designated on a

bipartisan or multipartisan basis, primarily for ensuring that the parties
out of power have a presence at the polls. In most states the bipartisan
approach seems also to have succeeded in establishing public confidence
in even-handed treatment at the polls. The Presidents Commission on
Registration and Voting Participation (1963) recommended a biparti-
san policy at the precinct levels as an effective deterrent to fraud. Cer-
tainly polling places 4eem to be an appropriate location for representa-
tion of the political interests involved in an election, and the presence
of opposing parties tends to discourage temptation to intimidate voters.
It is strongly recommended that bipartisan representation in election
administration not be extended to the clerical staff of a state or local
election office. In addition it should be pointed out that bipartisan or
multipartisan considerations in the recruitment of election day person-
nel can coexist with merit selection procedures if the party role in re-
cruitment is limited to submitting lists of names to the hiring authority.
Finally, all poll workers should be clearly identified by function. If
their assignment varies during the day, their identification labels should
reflect the change. This would help voters direct questions to the appro-
priate official and protect them against the interference of unauthorized
personnel. To prevent unwarranted pressure on the voter, party repre-
sentatives should also be clearly identified.
Office of Federal Elections, United States General Accounting Office, A Study of
Election Difficulties in Representative American Jurisdictions: Final Report
(Washington, D. C.: January 1973).
Richard G. Smolka, The Voting Rights Act: Unfinished Business for 1972,
National Civic Review (January 1972), pp. 15-19.
1. In the Model State Constitution the governor is the only popularly elected
officer of the state and is its effective administrative head, appointing and re-
moving all department chiefs.
2. The most current study of the costs of registration and voting is Richard G.
Smolka, The Costs of Administering American Elections (New York: National
Municipal League, 1973).
3. The same logic would also call for greater financial contribution by the
federal government to pay for its share of the cost of federal elections (see Chap-
ter 6).
4. To avoid confusion the Model refers consistently in this chapter to county
officials. The recommendations with respect to this official may be understood
in some states to apply to both county and city election administrators. The

language used in the text reflects the traditional legal role of the county as the
primary administrative arm of the state.
5. In separately conducted city, village and town elections, local officials may
assume such responsibilities as preparation of their own ballots, receipt of
absentee ballot applications and preparation of information on local candidates
and issues.
6. Delaware and North Carolina are two examples of this administrative
structure. In Alaska all election personnel are state employees.

A Model Voter Registration System
Current methods of registering voters often do not work well for
voters or election administrators. Each year millions of eligible voters go
unregistered despite vigorous and well-intentioned efforts by public
officials, civic groups and political parties. At the same time election of-
ficials must cope with a system that does not permit effective planning
or provide the control necessary for efficient administration.
Citizens must now register through a single central office open only
during normal business hours or through branch offices, mobile registra-
tion and deputy registrars. In either case the volume of registrations is
still dependent on the initiative of individual voters. Administrators can
control the convenience of registration, but they cannot directly control
the volume. Officials are usually committed to a fixed range of facilities
which must be staffed and paid for whether they are used or not. The
most extreme example of this inefficiency comes under systems which
require one or two officials to conduct registration for several days in
each voting precinct prior to an election. This often results in excessively
high costs, since the same level of staffing and facilities is used regard-
less of the number of voters to be registered. But under present condi-
tions this may be an important way to provide convenient opportunities
to register.
Registering voters, however, is only half the story. Once voters are

enrolled, the lists must be updated through periodic purges. Most states
remove the names of registrants who fail to vote at least once in a given
period of time, such as two or four years. Registrants are usually notified
of the cancellation by mail and given the opportunity to reregister. The
severity of a states purge law has a direct bearing on the number of
voters to be registered. Frequent purges will result in more current and
accurate lists, but they also increase the volume of registrations that
officials must process and may seriously inconvenience voters. What
the system confers at one point, it often takes away later.
These problems will persist until the administrative burden of locat-
ing, identifying and enrolling eligible voters becomes a responsibility of
state government. The Model Voter Registration System recommends
that states assume this responsibility by conducting a statewide canvass
of all households. A closely supervised system of door-to-door canvass-
ing should produce more accurate lists at a lower cost per registrant
than most existing systems,1 and should help administrators to make
more efficient use of their resources. Most importantly, the resulting
lists should contain all but a handful of the eligible voting population.
This chapter outlines the procedures which implement the concept of
government-initiated registration most effectively. Alternatives are
discussed under each major recommendation.
All voting precincts in the state should be systematically canvassed
every one or two years to register all eligible voters and to remove
the names of voters who no longer reside at their registered ad-
dress. The canvass should be conducted through door-to-door
The most desirable canvassing method is systematic door-to-door vis-
itation carried out on a statewide basis.2 This represents the most ex-
tensive commitment a state can make toward an equitable system of
voter registration. Personal visits by trained canvassers to each house-
hold in the state should produce the most accurate list of eligible voters
possible at the lowest cost per registrant. Door-to-door visitation also
allows both registration and purging during the canvass. Combining
these two functions has a distinct advantage. Since one procedure re-
places two, there should be some savings in cost. Purging would also be
done more systematically than it is done now in many jurisdictions,
resulting in more accurate lists.
Door-to-door canvassing is a common occurrence in American life.
Political parties and candidates, of course, rely to a great extent on this
technique. Meter readers for electric and gas utilities make such can-

vasses on a regular basis. One-third of the states now authorize can-
vassing for the purpose of purging registration lists. An extension of this
authority would allow these states to canvass affirmatively to add the
names of qualified voters to the register. A citywide door-to-door can-
vass was the practice in San Francisco and other California cities 50
years ago. Its purpose was to add the names of eligible voters to the list,
not delete the names jaf registrants who have changed residence. Hawaii
and Alaska now permit door-to-door canvassing to enroll eligible voters,
and Canada and Great Britain have long done so.
The canvass should be timed to meet the requirements of a states
election calendar. In those states in which there is a general election
each year, an annual canvass will be necessary to compile an up-to-date
registration list. In other states a biennial canvass may be sufficient. The
date should fall during a period when activity is at a minimum, such as
winter or early spring. This period might logically come prior to a pri-
mary election, but the registration list will be most accurate immediately
after the close of the canvassing period; and some states might decide
to hold the canvass prior to a general election. Whatever the date, it
should remain fixed on the election calendar so that the citizens know
exactly when to expect it.
Canvassers should be required to seek only information necessary
to establish a citizens voter qualifications and to identify him at
the polls.
The information required of prospective registrants should be kept to a
minimum and should be directed specifically toward state voter qualifi-
cations such as age, citizenship and residence. Under some systems of
door-to-door canvassing the information required for registration need
not come directly from the person to be registered but may be supplied
by other reliable sources such as other occupants of the same household,
apartment managers or neighbors. This technique is referred to as non-
personal registration and is used during the nationwide enumeration
that precedes each Canadian federal election. Canvassers are required
to record only the name, occupation and address of each qualified voter
they enumerate. This information often comes from other members of
the household. A similar practice under a personal registration system is
allowed in Colorado where a citizen who registers in person may also
register other members of his immediate family.
Under some systems of mail registration not every registrant is re-
quired to sign and fill out the form himself. In Great Britain, where
registration forms are mailed annually to every household throughout

the country, the head of the household is instructed to list by name and
age every eligible voter living there. Under the Texas mail registration
system any member of a family can register any other member who is
eligible to vote by filling out the prescribed form.
A form of non-personal registration was common to the first systems
adopted by many states in the late nineteenth century. They have since
fallen out of favor because of past abuses. Under these early systems un-
supervised precinct election officials, chosen by local party organiza-
tions, made up lists of eligible voters purely from their own personal
knowledge. Often the lists were padded with tombstone names that
enabled party agents to cast illegal votes on election day. The possibility
of abuse is eliminated under a system of trained canvassers subject to
the safeguards built into the Model.
If a personal application by each registrant is not required under a
door-to-door canvass, then a signature on the original application cannot
be obtained and a signature comparison at the polling place is not pos-
sible, at least for first-time voters. This is an important consideration
because of the significance which is often attached to the ability of pre-
cinct officials to compare a signature obtained on election day with the
signature on the original registration record as a means of preventing
impersonation. The signature comparison has always been a part of the
Leagues model system of permanent personal registration. The require-
ment now exists in 21 states. In nine other states the voter must sign the
poll book but no formal comparison is required by law. No signature is
required in the other 20 states. The signature comparison safeguard may
be most relevant to the absentee voting process in which ballots are
requested by mail. The signature obtained at the time of registration al-
lows both the application and the ballot to be checked against the
original registration record to ensure that the person registered is the
person voting absentee. The election laws of 32 states provide for some
form of signature comparison in the absentee voting process. In 16 of
these states the signature on the absentee ballot is compared to that on
the original registration form, and in the other 16 the ballot signature
is compared to the one on the absentee ballot application.
Simplicity in procedural requirements encourages citizen partic-
ipationthe simpler the requirements, the greater the incentive to partic-
ipate. This need is nowhere greater than in the registration process. If
local conditions allow it, the canvass should be conducted on a non-per-
sonal basis under appropriate rules determined by the Chief Electoral
Officer.3 However, if a signature comparison is considered desirable,
two alternatives are possible under the door-to-door canvassing system.

If a state chooses to require a signature on the original registration rec-
ord, canvassers who are unable to make personal contact with a poten-
tial registrant should enter his name and address on a signed, postpaid
registration form and leave it at his residence with instructions to
complete, sign and mail the form to the registration office.4 A signature
could also be obtained after the canvass by requiring voters to sign the
poll sheet on election day. Although the signature comparison could not
be made at the first election, it could easily be done for succeeding elec-
The utility of the signature test has traditionally been its preventive
value, since actual comparison at the polls is seldom done with any reg-
ularity. The very act of signing ones name in public creates an instinc-
tive apprehension presumed to deter impersonation. This psychological
safeguard is possible without the need to obtain a signature on the orig-
inal registration record. If a strict comparison is judged necessary, then
a signature could still be obtained during the canvass, although this
requirement might dilute the effectiveness of the canvassing procedure.
The signature comparison requirement should also be considered in
the context of a states system for processing registration information
and producing precinct lists. Five states now maintain a statewide com-
puter registry,5 and it is likely that similar systems will soon be adopted
by other states. Copies of registrations taken by local officials are sent
to a central location to be placed on a single statewide computerized list.
The purpose of the system is to promote standardization of records and
procedures, to make purging more effective and to facilitate the transfer
of registrations between local jurisdictions. A major benefit of this
system is its ability to produce an accurate precinct register, in the form
of a computer printout, quickly and economically. The signature com-
parison requirement is not fully compatible with a statewide computer
registry and prevents full use of its capability. To allow for a complete
signature comparison, the original signed registration form should be
kept in a binder in each voting precinct, and a precinct register produced
by computer would then be an unnecessary duplication.
The variation that exists in state laws governing the signature com-
parison suggests that it is not always needed as a safeguard. For these
reasons the Model recommends that a signature test be required only
when a state carefully determines that it is necessary to guarantee the
integrity of the electoral process.1*
The signature comparison test which was an essential element in the
Leagues 1927 Model continues to be considered necessary by Joseph P. Harris.
See his supplemental remarks in the Appendix.

The duration of the annual canvass will depend on the techniques
used. The area to be canvassed should be small enough to be cov-
ered by a canvasser twice within the period of one week.
It should be feasible to complete a door-to-door canvass in approx-
imately 40 days, including time for additions and corrections.6 The San
Francisco canvassers of the 1920s made two sweeps of the city between
January and March. The Canadians conduct a nationwide door-to-door
canvass in one week. An additional period of time is devoted to cor-
recting the registration lists in a series of public hearings. The entire
process takes 49 days. In Great Britain the annual mail canvass sup-
plemented with targeted door-to-door canvassing, including time for ad-
ditions and corrections, takes two and one-half months.
With a door-to-door system, each area should be small enough so that
canvassers could comfortably cover it twice in one week. A longer
period might seriously inhibit efforts to hire a sufficient number of the
temporary personnel likely to be available, such as college students and
housewives. The Canadian experience indicates that an area having 250
to 300 eligible voters is about the optimum size for a thorough door-to-
door canvass, although in many instances population shifts increase the
size of some polling divisions to 500 voters. Precincts in which voting
machines are used may be too large to be easily canvassed as a single
unit. Where possible they should be subdivided for that purpose, al-
though not necessarily for voting.
Door-to-door canvassers should personally visit each household. Un-
less satisfied that no qualified voter remains unregistered within a
household, the canvassers should make two visits, once during the day
(9 A.M. to 6 P.M.) and once at night (7 P.M. to 10 P.M.). When canvass-
ers are unable to communicate with anyone in the household from
whom they can obtain the required information, they should leave a card
stating the day and the hour they will make their second visit. The card
should also contain their names, addresses and phone numbers. The can-
vassers should leave a notice of registration for those people whose
names they will add to the list.
Canvassers should be nominated by the political parties but should
be selected by and responsible to the county administrator of elec-
tions. They should be required to attend a comprehensive training
session prior to each canvass.
1 o insure an impartial registration process canvassers in each precinct
should be named to represent the two major political parties. The
county administrator should make the selection from lists supplied by

the county party organizations. Canvassers may be the same people who
serve as election judges at the polling place. Their increased familiarity
with those eligible to vote may provide an additional safeguard against
fraud on election day. Each canvasser, however, should be clearly re-
sponsible to the county administrator who should be free to reject or
replace anyone nominated by the parties if he states his objections. Each
canvasser should be required to sign the list of registered voters pro-
duced by his efforts. States which do not use the bipartisan or multi-
partisan system can easily rely on a single canvasser in each canvassing
Regardless of the manner of selection canvassers should be required
to attend training sessions developed by the Chief Electoral Officer and
supervised by the county administrator. Canvassers should receive
additional compensation for the time spent in training. They should be
required by law to carry appropriate identification and to canvass the
entire area to which they are assigned. Canvassers should be paid for
each new registration they make and also for each deletion or correc-
tion. Since new registrations may involve considerably more work than
deletions, canvassers should be paid a higher rate for adding names to
the list. Compensation should be high enough to attract a sufficient num-
ber of canvassers.
If a state requires party registration, voters may enroll in the party
of their choice at the time of the canvass or in a supplementary mail
In 22 states party enrollment is a prerequisite for primary voting. In
other states no party affiliation is officially recorded, or party enroll-
ment is based on primary participation. The canvassing procedures rec-
ommended here are, for the most part, compatible with any direct pri-
mary system. Door-to-door canvassers could obtain a declaration of
party affiliation from each eligible voter whose name is added to the list.
This procedure, however, should be accompanied by safeguards. Per-
sonal contact with each registrant should be required and the declara-
tion should be affirmed by his signature. If it is deemed impractical or
undesirable for canvassers to solicit declarations of party affiliation,
canvassers could be required to leave a party registration form at each
household and to notify the registrant that he may enroll in the political
party of his choice by filling out the form and returning it by mail or in
person to the county administrator.
The statewide canvass should be followed by a short period for
claims and objections during which errors in the list may be cor-

rected and the names of voters overlooked during the canvass may
be added.
After the canvass is completed the county administrator should pre-
pare a list of all eligible voters. This list should be printed on a precinct-
by-precinct basis with a central list kept in the county office.7 The county
administrator should ensure that the lists are available to parties and
candidates. Each voter should also have an opportunity to see that his
name has been included and to challenge the names of those he believes
to be ineligible. A special period for claims and objections should be
provided for this purpose. The county administrator should include in-
structions for making claims and objections in the voter information
material that is regularly distributed by his office.
Although most eligible voters will be entered on the rolls during the
canvass, there should be a special period when those missed can add
their names and when voters incorrectly listed can be challenged. This
period for claims and objections will help groups wishing to police the
registration process to make sure that the final list is as accurate as pos-
sible. Eligible voters choosing not to be on the list could also take action
to have their names removed.
Following the period for claims and objections the county administra-
tor should audit the final list by taking a random sample of the names
and checking the validity of their registration. A representative sample
can be easily drawn. The audit would serve two purposes. First, it would
encourage an honest effort among the canvassers by putting them on
notice that their work is subject to a central check. Second, it would pro-
vide added insurance that the final list is as accurate as possible.8
After the close of registration the county administrator should make
the list available at cost, or without charge, to recognized candidates and
political parties. It should not be available to other government agencies
for any non-electoral purpose, and present constraints against the use
of the list for commercial purposes should be continued. Although there
is no foolproof way to prevent commercial interests from using a copy of
the list, such a prohibition would be in keeping with the concept of the
electoral process as a public activity operated in the public interest. This
principle was upheld recently in a successful suit brought by the state of
Canvassers should be deputized to serve as registrars in their pre-
cincts throughout the year.
The mobility of the American people is so great that additional regis-
tration opportunities must be available outside of the canvass if an ac-

curate list is to be maintained. To meet this need a central registration
office should be open and available to voters throughout the year. Can-
vassers should be deputized by the county administrator to serve as
registrars in each precinct until the next statewide canvass. The pres-
ence of a registrar in each neighborhood will provide new residents
who arrive after the canvass is completed with a convenient opportunity
to add their names to tl^e list. If canvassers are unable to serve as deputy
registrars, the county administrator should name replacements from
nominees submitted by the political parties. Neighborhood registrars
should become a familiar feature of the community, making the process
an integral part of local affairs. This familiarity and visibility should
greatly facilitate the success of the canvass as well as neighborhood
registration throughout the year.
Voters who are not registered during the statewide canvass and are
unable to register in person at other times should be able to register
by mail throughout the year.
Under the Voting Rights Act Amendments of 1970 each state is re-
quired to provide absentee registration opportunities to persons qualified
to vote for President. This provision was intended to overcome the re-
strictions found in current systems. Even under government-initiated
registration systems, however, there will always be a small category of
voters not registered during the canvass who are unable, because of age,
illness or some other reason, to register with local officials between
canvasses. Absentee registration should be available to these individuals
for state and local as well as for federal elections.
Canvassing by Mail
The absentee registration provision described above is not apt to be
heavily used under a door-to-door system with adequate opportunities to
register between canvasses. However, where climate or terrain make a
door-to-door canvass difficult, some jurisdictions might adopt the mail
technique as a first-line system to initiate registration. Under such cir-
cumstances the mail registration would be an alternate means of can-
vassing not to be confused with the absentee registration opportunities
described in the preceding paragraph.
Although mail canvassing may be an appropriate alternative in some
areas, it has three major liabilities when compared to door-to-door sys-
tems. First, used alone, it will not reach the eligible voting population as
effectively as door-to-door canvassing.

Second, a mail system will require more processing steps than door-
to-door canvassing, thereby increasing both costs and the possibility of
administrative error. The distribution and receipt of a mail form is only
the beginning of the administrators responsibility. Many returned forms
will be illegible, incorrect or incomplete and information would have to
be obtained from the registrant in a second mailing or other contact.
Each registration must also be sorted by election district and precinct at
a central office, another administrative step that would be avoided if a
trained canvasser obtained the complete information.
Finally, a mail system would be subject to the same selective use by
interest groups that is characteristic of current systems of registration.
Some groups that now engage in registration activities would undoubt-
edly distribute mail forms to their constituencies and ensure that they
are filled out and returned. For these reasons a mail canvass must be
recommended only as an intermediate step between individual initiative
and full government responsibility for registration.
Despite these limitations, registration by mail may be useful if com-
bined with door-to-door canvassing. Great Britain relies initially on an
annual mail canvass followed by selective door-to-door canvassing to
cover areas where the response rate has been low or to correct forms
that are improperly filled out. If a state decides to combine the tech-
niques, specific criteria should be established to determine which one is
to be used in a particular area. Either technique might be used if the
registration levels achieved are substantially the same. Since the two
approaches will often produce different results, however, safeguards
should be enacted to prevent officials from choosing on the basis of an-
ticipated political effect.
Expanding Registration Opportunities Under Present Systems
States which do not implement a door-to-door canvassing system to
register voters should nevertheless provide maximum registration op-
portunities for voters. Several techniques to accomplish this purpose are
suggested below.
(1) Closing Dates. Voters should be able to register as close to elec-
tion day as possible, but in no instance should registration close more
than 30 days before an election. Registration is closed prior to an elec-
tion for administrative purposes, to allow local officials to prepare lists
for use on election day. However, the period immediately prior to an
election is when voter interest in campaigns is greatest and when unreg-
istered voters are most likely to enroll. Thus administrative requirements

often seriously restrict registration opportunities. Certainly, late closing
dates would be more feasible if local officials made maximum efforts to
register voters throughout the year, thereby reducing the last-minute
rush. If the outreach techniques suggested below are widely im-
plemented, registration might be closed as late as one week before
election day, depending on the size of the jurisdiction and its administra-
tive capability. /
(2) Selective Door-to-Door Canvassing. States which do not assume
responsibility for door-to-door canvassing should authorize and encour-
age local officials to do so. Trained volunteers could be used under a
systematic plan developed locally. Canvassing could be done throughout
a locality or limited to areas characterized by low levels of registration.
(3) Deputy Registrars. State law should allow local officials to dep-
utize any eligible voter to register other eligible voters. The state should
encourage local officials to develop training programs for deputy regis-
trars and to make maximum use of volunteer efforts.
(4) Branch and Mobile Registration. In many jurisdictions voters
must travel great distances to register at a central location. Such re-
stricted opportunities are serious obstacles to full participation. States
should allow local officials to establish branch offices throughout their
jurisdictions, utilizing firehouses, libraries and other public offices
where registration could be taken year round. Localities might also dep-
utize an official in each high school so that young voters could become
registered close to their eighteenth birthday.
(5) States Should Not Cancel a Citizens Registration For Failure To
Vote In Any Period Less Than Four Years. Under the permanent per-
sonal registration systems of some states, individual registrations are
cancelled if a registrant fails to vote every two, or sometimes three
years, forcing people who vote only in presidential elections to reregister
every time they plan to vote. Such requirements are discriminatory and
undesirable. Cancellation provisions for failure to vote should require no
more than one vote every four years and should include the opportunity
to maintain registration by return notification within a reasonable period
of time.
(6) Absentee Registration. Citizens who are ill, physically disabled
or otherwise unable to register in person should be allowed to register
by mail. States should provide convenient postcard forms that may be
returned postage free.
(7) No Registration. States which do not now require registration in
some areas should continue this policy unless or until they encounter
evidence of election fraud. Preferably, states should direct their reform

energies to the expansion of registration opportunities in other areas or
to the consideration of eliminating registration in additional communi-
ties where it may not be needed.
A Note on a Practical Standard of Performance for
a Registration System
The Model Voter Registration System incorporates a set of procedures
that should allow state and local governments to identify accurately the
population eligible to vote within their jurisdiction. However, the pro-
cedures themselves are less important than the results they are intended
to produce. Any system of registration should be judged ultimately on
the basis of an explicit standard of performance. The determination of
that standard should be based on a number of considerations.
Placing responsibility on state government for initiating registration
is the fundamental recommendation of the Model and the prime stan-
dard against which an existing system must be judged. As the alterna-
tives outlined in the recommendations suggest, the government can
meet this responsibility in a number of ways. Therefore some additional
standard is necessary to evaluate the various techniques available. The
most useful standard is a percentage of the eligible voters registered.
The percentage of the voting age population registered fluctuates
widely from state to state and from election to election. In 1968, a
presidential election year when interest in politics is usually at a peak,
the national average was 76.7 percent.9 In 1970, a nonpresidential year,
the national average decreased somewhat to 73.8 percent.
These figures are not acceptable. Our institutions can be modified to
produce a considerably better level of results. Recent constitutional
amendments, court decisions and congressional acts have removed state
restrictions on voter qualifications to such an extent that virtually every-
one over 18 can qualify to vote. The major exceptions in most states are
convicted felons who are imprisoned or who have not had their civil
rights restored, persons adjudged mentally unsound and aliens. Data
gathered by the U. S. Rureau of the Census indicate that these legally
ineligible groups constitute approximately 4 percent of the voting age
A registration system capable of locating the residence of every eligi-
ble voter could conceivably register 96 percent of the voting age popula-
tion. In practice, however, even the most efficient system will fall sub-
stantially short of this ideal. There will always be a small number of
otherwise qualified voters with no fixed residence. The physical land-

scape of our cities is constantly changing as new buildings are con-
structed and old ones torn down. With population movement so fluid,
it is highly unlikely that any practical registration system could keep
track of every eligible voter. Much has been made of the difficulties of
canvassing in urban areas. Access to large apartment buildings some-
times presents a problem; fear of street crime is another. The other side
of the coin is that high-density areas may in some instances make can-
vassing easier by reducing the leg work required. Certainly, canvassers
who know the precinct will perform more efficiently in any community,
rural or urban. Whether or not the difficulties or advantages of urban
canvassing balance each other out, the net result of a door-to-door sys-
tem is virtually certain to be better than the present system of voluntary
registration. An efficient registration system incorporating the appro-
priate canvassing methods should register at least 90 percent of the
voting age population.
Under the provisions of the Model the register should be most
accurate immediately after the canvass. The register should equal or
exceed the 90 percent standard. If no further steps are taken to update
the list, population mobility would reduce its accuracy by 10 to 15 per-
cent or more within 12 months. The efforts of deputy registrars in each
precinct should keep the list around the 90 percent level until the next
statewide canvass.
William G. Andrews, American Voting Participation, Western Political Quar-
terly (December 1966), pp. 639-652.
Angus Campbell, Philip E. Converse, Warren E. Miller and Donald E. Stokes,
The American Voter (New York: John Wiley and Sons, Inc., 1960).
J.-M. Hamel, Registering Voters in Canada: A Responsibility of the State,
National Civic Review (July 1972), pp. 336-340.
Joseph P. Harris, Registration for Voting in San Francisco, National Munic-
ipal Review (April 1926), pp. 212-218.
V. O. Key, Jr., Politics, Parties and Pressure Groups, Fifth Edition (New York:
Thomas Y. Crowell, 1964), pp. 597-649.
National Municipal League, Model Voter Registration System, Revised Edition
(New York: 1957).
Lloyd B. Omdahl, Fraud Free Elections are Possible Without Voter Registration,
Bureau of Governmental Affairs, University of North Dakota (September
Report of the Presidents Commission on Registration and Voting Participation
(Washington, D. C.: U. S. Government Printing Office, 1963).
U. S. Bureau of the Census, Statistical Abstract of the United States (Washing-
ton, D. C.: U. S. Government Printing Office, 1970, 1971).

U. S. Bureau of the Census, U. S. Census of Population and Housing: 1970,
Data-Collection Forms and Procedures (Washington, D. C.: U. S. Government
Printing Office, 1971).
Meyer Zitter and Donald E. Starsinic, Estimates of Eligible Voters in Small
Areas: Some First Approximations, American Statistical Association: Pro-
ceedings of the Social Statistics Section 1966 (Washington, D. C.), pp. 368-
1. For a discussion of the cost of various registration techniques see Smolka,
The Costs of Administering American Elections.
2. Mail canvassing is discussed on page 31.
3. An exception to this recommendation applies to closed primary states. In
such states the Model recommends that a signature be required if declarations
of party affiliation are recorded during the canvass. See page 29.
4. This procedure is similar to the one suggested for party enrollment in some
closed primary states. See page 29.
5. Alaska, Delaware, Kentucky, South Carolina and Virginia.
6. See the implementing statute in the Appendix for a suggested schedule.
7. In the growing number of states that now maintain computerized regis-
tration lists, the results of the precinct canvass could be sent directly to the cen-
tral state registration office.
8. Existing sources of information might also be used to update the list
directly after the canvass or at other times during the year. For example, resi-
dent aliens must register with the federal government each year through the
Post Office. The list might be matched against these records to check for at least
one class of ineligible voters. Similar checks might be made against the death
reports from bureaus of vital statistics or against the records of local utility
companies. An updating system based on Post Office change of address forms
has been used with great success in Allegheny County (Pittsburgh) Pennsylvania.
The Post Office regularly provides the county with change of address forms filed
by mail patrons. The county then cancels the voters registration at his former
address and sends him a notice of suspension form which also allows him to
transfer his registration if his new residence is within the county. Since 1961 an
average of 90,000 changes have been processed each year. Because the changes
are filed mostly by families, an estimated 200,000 potential voters are affected
9. In considering official registration figures, particularly for individual states
or counties, it is important to remember that the actual number of registrants
tends to be seriously inflated under a permanent registration system because
of inadequate purging practices. The averages noted here have been computed
from state data reported in the Census Bureaus Statistical Abstract. In many
U. S. counties the number of registered voters often exceeds the voting age pop-
ulation. A 1971 report from Illinois, for example, revealed that 21 of the states
102 counties had registration rates of more than 100 percent of the voting age
population. Most of the counties were in rural central and southern Illinois. Big
Voter Discrepancies in 21 Counties, Champaign-Urbana Courier, April 23

Absentee Voting
Absentee ballots invariably complicate the administrative task
far beyond their usual contribution to the total vote cast. The steps in-
volved in applying for, issuing, casting, receiving and counting absentee
ballots have significant administrative consequences and raise thorny
questions about safeguarding the secrecy of the ballot and the integrity
of the election result. Because of administrative problems and high
costs, some administrators may be less than eager to provide for liberal
absentee voting procedures. These considerations must be taken into
account. Yet, in a society as mobile as ours, provision must be made for
voters who expect to be absent from their city or county on election day.
Failure to do so disfranchises people away for business and pleasure as
well as those who are physically unable to go to the polls.
Since only a few states compile the total number of absentee ballots
issued and cast, it is difficult to determine precisely how many are cast
nationally for any class of elections. The limited evidence available sug-
gests that in most states no more than 4 or 5 percent of the total vote cast
is by absentee ballot, although in some areas it may be as much as 10
percent or more. Absentee voting increased somewhat in 1972, and this
should continue to hold as a result of the twenty-sixth amendment and
the federal Voting Rights Act Amendments of 1970 (VRAA). The en-
franchisement of 18- to 20-year-olds for all elections meant that large
numbers of college students used absentee ballots to vote at their home

addresses. The VRAA mandated national standards that expanded op-
portunities to vote absentee in presidential elections. The imposition of
federal requirements left some states with two classes of absentee voters
one for presidential and one for state electionsmaking a single set of
standards an increasingly desirable reform.
This chapter does not attempt to provide guidelines for a complete sys-
tem of absentee voting, a procedure that involves administrative com-
plexities outside the scope of the Model. The limited recommendations
that follow are intended simply to propose absentee voting opportunities
within the design of the Model, and to take account of some of the major
problems for both administrators and voters.
Absentee voting provisions should apply to both primary and gen-
eral elections.
Because elections in so many regions of the country are decided in the
contest for party nomination, absentee voting opportunities should be as
readily available in primaries as in general elections. Competition be-
tween the two major parties is sharpest and most visible at the level of
presidential politics. It tends to decline, however, with the size of the
political unit, dropping steadily from the national arena to statewide
contests for U. S. senator, governor or other state officials, and declining
to a low point in local government elections. In effect this sometimes
means that in small political units the candidate of the areas dominant
party is virtually certain to be elected. And, since their nomination
commonly is controlled by the local party organization, the election
of the party leadership, as determined in primary elections, also as-
sumes critical importance in local politics. Finally, the only voice that
most rank and file party members have in the naming of their partys
presidential candidate is in the selection of delegates to a national
nominating convention. In most states these too are chosen at a primary.
Party elections should therefore be as accessible to the voters as any
other election.
Absentee voting should be available to any qualified voter who ex-
pects to be away from his county or city on election day or who is
ill or physically disabled.
For presidential elections the VRAA provides that anyone who is ab-
sent from his county or city for any reason on election day is entitled to
vote by absentee ballot. No state should have more restrictive require-
ments for eligibility. In addition, states should extend such privileges to

anyone otherwise qualified to vote who is unable to go to the polls be-
cause he is aged, ill or disabled. Anyone voluntarily confined to an in-
stitution should also have his right to vote protected through absentee
There should be no requirement for notarization of the absentee
ballot application or the absentee ballot.
For most voters a requirement that a notary public certify an absentee
ballot application is an unnecessary inconvenience and serves no useful
purpose. Notaries are usually available only during working hours and
even then few are conveniently close to most voters. Furthermore, the
cost to the voter of paying for the notarys services amounts to a small
poll tax.
No proof of inability to attend the polls should be required. The ill and
the physically disabled should not be required to obtain a doctors certif-
icate in order to secure an absentee ballot. The few states that require
such proof impose an unnecessary impediment and a serious inconve-
nience on physically disabled voters. The fee physicians charge for pro-
viding such certificates is another burden for the absent voter.
No special application form should be required to obtain an absen-
tee ballot.
Some states require that a special application form be filled out by the
prospective absentee voter before a ballot will be sent. This adds an
extra step to the process and expands the minimum time period within
which an absentee ballot can be requested, received and counted. States
should accept any qualified voters written or personal request for an ab-
sentee ballot.
Applications for absentee ballots should be accepted up to seven
days before an election and absentee ballots should be counted if
received by the time the polls close.
The Voting Rights Act Amendments of 1970 mandate these require-
ments in voting by mail for President. They should apply to all other
elections as well. Unnecessary administrative complications result if a
state creates two classes of absentee voters by establishing one set of
deadlines for the application and receipt of ballots for federal elec-
tions and another for state and local elections. During the 1972 general
election, when the VRAA were in effect for the first time, some states
with two sets of deadlines failed to provide a special absentee ballot
form for President and Vice President only. This raised the strong pos-

sibility that ballots received after the more restrictive state deadline
would not be counted, even for President and Vice President. Central
counting of absentee ballots, as recommended below, should allow
states to extend more easily the federal deadlines to state and local
To preserve the secrecy of the ballot and to prevent fraud, absen-
tee ballots should be returned to and counted at the central elec-
tion office which issued them rather than distributed to each pre-
The same safeguards that attend the counting of regular ballots should
be applied to absentee ballots. Requiring precinct personnel to count ab-
sentee ballots complicates their workload unnecessarily and may jeopar-
dize the secrecy of the ballot. For example, if absentee ballots are sent to
the precinct for counting, the number handled may be so fewonly one
perhapsand the handling so informal that it becomes easy to recognize
how an individual voted. The absent voters choice may also be more sus-
ceptible to identification in precincts where a voting machine is used and
the absentee ballots are the only paper ballots received.
The counting and recording of absentee ballots requires special train-
ing that is difficult to administer to temporary poll workers. The possi-
bility of error is compounded when the slow and tedious task is carried
out after a long day at the polls.
To circumvent duplicate voting, the voter who has been mailed an
absentee ballot should be required to turn it in if he later wishes to vote
in person at his precinct on election day. If he has already mailed in his
absentee ballot but wishes to vote in person, he should be permitted to
do so only by casting his ballot at the central office where his absentee
ballot has been, or will be, received and counted, so that it may be iden-
tified and destroyed.
Every state should adopt all recommendations made in the Federal
Voting Assistance Act of 1955 as amended.
The Federal Voting Assistance Act of 1955 was passed by Congress to
encourage absentee voting and registration by members of the armed
forces and the merchant marine, their spouses and dependents. Cover-
age of the act was extended in 1968 to include United States citizens
temporarily living abroad and their spouses and dependents. The act
recommends that each state accept a prescribed form, the Federal Post
Card Application (FPCA), as an application for an absentee ballot and,
if the applicant is not registered, as a simultaneous application for regis-

tration for all special, primary and general elections. The act provides
that the post card application and voting materials from local election
officials may be sent postage free. The recommendations contained in
the act, however, are not binding on any state. In September 1971 the
Department of Defense reported that 42 states accepted the FPCA as
an absentee ballot application and an application for registration for
members of the armed forces.1 Only 17 states accepted the FPCA from
U. S. citizens temporarily living abroad. Each state should make the pro-
visions recommended by the act an integral part of its election law.
American citizens who reside outside the territorial limits of the
United States and are otherwise qualified to vote should be able to
register and vote at least in federal elections where they last re-
sided in the U. S.
The 1.5 million American citizens residing abroad who are currently
discouraged from participating in state and national elections should be
extended the opportunity to vote, at least for national office. These
civilians usually are subject to U. S. tax laws and other obligations of
American citizenship and should have a voice in the affairs of the
country. In an age when so many Americans pursue commercial and
educational interests overseas it is unfortunate that most are disfran-
chised in a majority of states by laws which require physical evidence
of their residence in the state to vote.2 By failing to define residency
neither the Federal Voting Assistance Act nor the Voting Rights Act
Amendments of 1970 provide a sufficient legal mechanism for enfran-
chising qualified voters living overseas who are not members of the
armed forces or federal employees. The states continue to have exclu-
sive authority to define residence for purposes of voting, although they
may not require a particular length of residence. Many states now define
residence to exclude people who do not maintain a home or other
evidence of physical presence in the state. Interpretation of the law can
vary considerably within a state. Prior to the 1972 presidential election
some local registrars were accepting absentee ballot applications from
former residents residing abroad while their officials in other parts of the
state were refusing to accept such applications.3
To enfranchise qualified voters residing overseas and eliminate dis-
criminatory interpretations of the law each state should provide a formal
procedure through which these citizens can vote in national elections
and in state and local elections as well if they consider it desirable. State
law should define residence to include qualified voters living overseas
and should be based on the Federal Post Card Application as recom-
mended by the Federal Voting Assistance Act.

Other Problems
Even if the foregoing recommendations were to be implemented, the
voting privilege would be difficult to exercise in some situations. For
example, 11 states and the District of Columbia hold runoff primaries
and in several of the states the outcome is tantamount to election. Yet
often the interval between the first primary and the runoff is insufficient
to print new ballots, issue new applications to the original list of absen-
tees, mail out ballots to those who actually apply and receive the ballots
for counting by election day. Under these circumstances states might
consider allowing at least a month, and preferably six weeks, between
the primary and the runoff. States may want to consider sending absen-
tee ballots for the runoff to all those who are on the request list for the
primary or provide an opportunity for a voter who requests an absentee
ballot for the first election to request one for the runoff at the same time.
This would, however, involve the risk of having excess ballots in the mail
if a voter has returned to his home precinct between the initial primary
and the runoff. A state could safely send a ballot for both elections only if
it also provided for the receipt and counting of absentee ballots at the
central office (see above). Central counting would then permit the state
to destroy the runoff ballot conveniently if the voter has returned
and wishes to cast his ballot in person. Another obstacle to participation
occurs in localities that nominate candidates at caucuses instead of at
primaries. Absentees would obviously be unable to participate in this
kind of election.
National Municipal League, Model Civilian Absentee Voting Law, Fifth Edition
(New York: 1970).
1. Thirty-three states also accept the FPCA for their spouses and dependents;
48 states accept the FPCA for members of the merchant marine but only 33 ex-
tend the same privilege to their spouses and dependents. The act also recom-
mends that registration be waived for persons covered by the act who return to
the United States after the close of registration. Seventeen states have imple-
mented this procedure for members of the armed forces, 12 for members of the
merchant marine and none for civilians temporarily residing abroad. Altogether
the act contains 12 detailed recommendations. In 1971 only three statesIowa,
Montana and North Dakotahad implemented all 12. The Department of De-
fense estimates that only 26.5 percent of the armed forces personnel covered
by the act voted in the 1970 elections. U. S. Department of Defense, The Federal
Voting Assistance Program, Eighth Report, September 1971.

2. In testimony before the Senate Committee on Post Office and Civil Service
on October 12, 1971, the Bi-Partisan Committee on Absentee Voting estimated
that 42 states made it virtually impossible for civilians residing abroad to
vote because of confusing statutes and court decisions, voter verification of
residence forms, the lack of absentee registration opportunities or outright dis-
qualification of the overseas voter.
3. Richard Severe, Americans Overseas Face Denial of Vote Under Law.
The New York Times, September 10, 1972. As this article points out, the sit-
uation is further complicated by the insistence of a handful of states that over-
seas voters also pay state income taxes. Some voters here, the author writes,
have sent in applications only to receive state tax forms in the return mail.

The Consequences of Complexity
The Model Election System is designed to facilitate voting by
making registration a government responsibility and by promoting the
efficient conduct of election day activities. The changes recommended
here should encourage greater numbers of citizens to exercise their
franchise, but there is no insurance that they will cast an informed vote.
Other elements of the election system affect the quality of individual
participation. Many of these reflect constitutional, historical or political
circumstances which the Model does not address, but which any discus-
sion of electoral reform must at least take into account.
The length of the ballot and the number of elections held point to the
continuing popularity of the mistaken notion that the more decisions the
electorate makes the better off it is. There comes a point, however, when
the length of the ballot complicates the electoral task beyond the voters
ability to cope with it intelligently and the frequency of elections ex-
hausts his civic patience. It is equally appropriate to consider whether
electoral practices designed to meet some interest other than the voters
have an adverse effect on participation. Is the form of the ballot, for
example, calculated to aid the voter or the candidates and parties?
Nowhere in the world is the voting task as complicated as it is in the
United States. While the laws of other democracies restrict elections to
principal policy makers, Americans have always been eager to expand
their role in the conduct of government. In addition to the hundreds of

thousands of public officials we elect each year, many state constitutions
and city charters require the submission of complex questions to the
electorate. The number of elections in this country reflects our implicit
faith in the responsiveness of local institutions. The legions of office-
holders who present themselves to the electorate each year are also an
outgrowth of the Jacksonian belief in the inherent wisdom of the
common man and our distrust of formal concentrations of power. Many
of our institutions are designed to keep government functionaries at all
levels continuously answerable to the people. This pattern will be diffi-
cult to change in spite of its impracticality as a device for effective gov-
ernment. There is a growing body of evidence, however, to suggest that
our uses of popular control have outstripped our ability to make in-
formed electoral decisions.
Length of the Ballot
The number of candidates and issues a voter faces at each election is
determined by political policies unrelated to the conduct of elections,
but the consequences of the long ballot present formidable hazards for
administrators and voters alike. To deal with scores or sometimes
hundreds of choices some states use extremely long ballots and then
allow the voter less than three minutes to record his decisions. When
the voter is confronted with an overwhelmingly complex choice, how-
ever, he may simply abdicate his responsibility for making independent
electoral judgments as indicated in a recent study by the United States
General Accounting Office:
In an Ohio Democratic primary election of May 1972, the ballot for del-
egates and alternates to the national convention contained 285 names. Be-
cause of the long ballot, the secretary of state directed that every jurisdic-
tion use a paper ballot for these candidates. This ballot permitted the voter
to cast a vote for a slate of delegates pledged to a presidential candidate
by making a single X or to vote on each delegate individually. The option
of individual voting was virtually ignored. . The voters in Ohio, when
confronted with this ballot, chose to cast their ballots for slates committed
to a presidential candidate. Even the most well known political leaders
failed to draw a significant number of individual votes.1
The voter may also lose interest when the ballot demands judgments
on matters beyond the competence of the average citizen. Eleven days
after a 1966 election only 8 percent of the voters in New York City
polled by the Citizens Union could recall the name of a single judicial
candidate. No one judge was named by more than one percent of the
voters except for the chief judge of the court of appeals whose election

was uncontested and whose name appeared on all four party lines. In his
case a record 4 percent of the voters in Syracuse recalled his candidacy.
Numerous other polls have been conducted to show that voters can sel-
dom name the minor offices that appear on the ballot, much less the
candidates who seek to fill those posts. Nevertheless, voters continue
to be asked to select officials to run highway departments, to audit the
books of state and local governments, and to serve as calendar clerks in
county courthouses.
Besides the crowds of candidates who appear on our ballots, legis-
lative policy or initiative procedures often require voters to decide on a
bewildering number of ballot issues. In 1972 California submitted 22
questions to referendum, Washington 24 and Rhode Island 15. In No-
vember 1970 there were 53 constitutional amendments on the ballot in
Louisiana and they were all defeated. No matter how valiantly a state
tries to distribute voter information on these issues, many well-inten-
tioned citizens will not have the time or interest to give to this voting
task. Faced with an impossible number of choices, or impossible kinds
of decisions, voters tend to make judgments in an arbitrary manner,
selecting candidates at random or using traditional clues such as party
affiliation. This has the effect of relegating the decision-making process
to party organizations or other powers which control the nominations.
Worse still, it may discourage citizens entirely about the act of voting.
The long ballot also causes logistical problems on election day. Voters
across the country have had the common experience of waiting in long
lines in front of polling places as their fellow citizens struggled with
huge ballots inside the booths. A random sampling of hundreds of news-
paper reports after the 1972 elections produced the following anec-
From the Melbourne (Fla.) Times:
The long wait didnt personally bother me, but it appeared to upset the
elderly citizens and those young voters participating in the election process
for the first time, one woman reported.
After waiting in line at Precinct 81 for two hours and ten minutes, he
was still number 40 in the line. He became so annoyed he went to the
registration desk and turned in his registration card.
From The Houston (Texas) Post:
Many hadnt read the amendments before coming in, said one election
judge. They got angry if you didnt give them enough time to decide
about them when voting. In one precinct a man who had minor trouble in
a booth was overheard asking, What do I do now? In unison, the long
line of voters shouted back: Vote!

From the Providence (R.I.) Journal:
They applauded in Exeter when one voter emerged from behind the
curtains after spending 15 minutes making his selection. . .
In spite of efforts by the state Board of Elections to educate voters on
the numerous referenda questions . poll workers said it appeared that
voters spent a lot of time reading the referenda questions.
It is fair to conclude that the simpler the ballot the better the voters
performance will be. As a corollary, the more significant the choices
presented the more incentive the voter will have to participate. Most
important, the voter may focus on the issues more easily when he can
identify them clearly on a ballot limited to major offices.
Ballot Forms and Voting Mechanisms
Since voters may have little or no information about many candidates
or issues on a long ballot, their behavior is often influenced by the
ballot arrangement and the voting system in use. There are two major
ballot formats used by the states. The most common is the Indiana or
party column ballot, used by 34 states, on which candidates are grouped
in columns under a party label. Sixteen use the Massachusetts or office
block ballot which groups the candidates by the office they seek. The
office block ballot encourages individual choices on each office and this
increases the rate of ticket splitting. On the other hand the Indiana bal-
lot encourages voting by party and creates greater straight ticket voting.
The argument is often made that the office block ballot is desirable
because it encourages choices based on the qualifications of individual
candidates rather than party allegiance. Studies have shown that the of-
fice block format does indeed result in voters splitting their votes more
frequently than with the party column format. However, the effect is not
uniform for all voters. Ballot form seems to make little difference in the
proportion of split or straight ticket voting among highly motivated
voters or those who identify strongly with a political party. The office
block ballot has its greatest impact among poorly motivated voters and
those who do not identify strongly with a party. This ballot form is also
associated with higher levels of roll off or the failure of voters to mark
their ballots for lesser offices. States that go from the party column to the
office block arrangement not only experience greater ticket splitting but
the total vote for offices listed further down on the ballot is reduced as
well.2 It is unlikely that ticket splitting and roll off would occur as often
as they do if voters could focus their attention on a limited number of
important candidates and issues.

There is evidence to indicate that voting may be discouraged in some
elections by complicated voting mechanisms. A 1968 study at the Uni-
versity of Michigan indicated, for example, that voting machines tend
to inhibit voting by persons low on the sociological scale. Supporting the
findings of two earlier Michigan studies, the writer found that partic-
ipation levels in each of the four referenda elections were consistently
higher in the paper ballot precincts. ... In no election, however, did
participation levels in the machine precincts reach those in the paper
ballot precincts.3
Although in some instances paper ballots may be the simplest for vot-
ers, they are not widely used in this country for several reasons. One im-
portant consideration has been the considerable savings that may be
realized with voting machines and electronic vote counting systems.
These systems allow officials to consolidate precincts and reduce the
total number of precinct personnel used to conduct voting. Equally im-
portant are the complications that stem from ballot length. The handling
of paper ballots is both fatiguing and time consuming to election officials
who must count the vote after the polls close. In addition, paper ballots
are more susceptible to human error and fraud since they are easily
miscounted, mismarked or otherwise spoiled. European countries have
been more successful with their use largely because there is normally
only one office in contest at each election. The generally higher turnout
levels in European countries are attributable to short ballots and less fre-
quent elections as well as to government-initiated registration systems.
In the last 10 years increasing numbers of jurisdictions have adopted
voting systems in which punch card ballots are counted by computer or
in which special optical scanning systems are used to count paper
ballots. These systems seem to reflect the belief that new technologies
can deal effectively with the complications associated with ballot length,
and they do appear to have solved some administrative problems. How-
ever, until ballots are shortened considerably there is a good chance that
American voters will be confused regardless of the voting system.
Another sensitive question relating to the form of the ballot concerns
the rotation of names. This technique is intended to give every candidate
an equal chance to benefit from the preferred first position on the ballot.
Rotation of names protects the candidates, under ordinary circum-
stances, from any built-in disadvantage based purely on ballot position.
However, the effort to be fair sometimes puts the candidates interest
above that of the voters. For example, where rotated ballots are used,
having sample ballots may confuse rather than inform the voter who
would ordinarily expect to see in the voting booth what he has been

shown on the sample ballot. Similarly, the guides that election officials
or public service groups may wish to provide would not help voters.
Perhaps of greater significance is the administrative disruption often
caused by rotation of names, particularly if the deadline for resolving
challenges to nominating petitions is very close to the election date. A
recent study of administrative difficulties in seven cities by the Office of
Federal Elections came to tl^e following conclusion regarding ballot
In each metropolitan area visited, ballot rotation, where required by law,
posed major difficulties. Printing of the ballot, preparation of the voting
machine and tallying the result, regardless of the method of voting used,
are made much more complicated and expensive by name rotation require-
ments. Preparation of the ballot for the printers becomes a timely and
tedious process. Personnel must lay out the ballot for each precinct includ-
ing sometimes several unique ballots per precinct. Large numbers of print-
ing proofs must be checked and rechecked and many short and separate
printing runs are required. The margin for error is obviously increased
manyfold. Errors result in candidates not appearing on the ballots at all
in some precincts as well as incorrect tabulations of results.4
Election Calendars
The number of elections as well as the number of ballot choices may
also directly affect voter participation. Somewhere in this country an
election is taking place every month of every year. Worse, in many states
an individual voter may be called on to go to the polls five or six times a
year. The average Californian goes to the polls at least three times an-
nually to elect some 50 candidates who directly affect his taxes and
everyday welfare. In odd-numbered years, a suburban New Yorker may
go to the polls four times for village, school, primary and general elec-
tions quite apart from special elections that may take place in special
districts or for bond issue referenda. These constant demands on the
voters attention are almost sure to be attended by a diminution of his
In an attempt to insulate local government from the influences of na-
tional politics, many municipal elections have been scheduled in non-
presidential years or at times other than November in a presidential
year. The separation of local and national elections shortens the ballot
required for each election and may help focus public attention on local
issues, but these benefits are achieved at some cost. Separate local elec-
tions place an additional financial burden on local governments and,
most importantly, voter turnout is reduced substantially.
Primary elections are particularly prone to calendar change. The party

in control of the legislature may have a special interest in making the
primary as late as possible one year to diminish the opportunities for
campaigning by insurgents; or they may schedule it as early as possible
in the year after an apportionment to permit incumbents the longest pos-
sible time to make themselves known in their new districts. Changing
the time of the primary can thus be an annual legislative exercise. To
protect the public against political manipulation of the election calendar,
each state should permanently set the time of its primaries.
The most logical uniform time to hold primaries is in the late spring.
Every four years, delegates to the national party conventions must be
elected no later than June. To achieve uniformity and reduce the number
of elections, primaries should be geared to a date that would accommo-
date the election of delegates to these conventions. Most political con-
testants argue that an early primary serves only to increase the difficulty
and length of the campaign. There is a real question, however, whether
this argument is directed primarily to the interests of the voter or to
those of the candidate and the political parties. The conceded difficulties
of long campaigns may perhaps be dealt with through better controls on
campaign practices and spending than on shifting election dates to the
confusion of the voter.
Richard S. Childs, Civic Victories (New York: Harper and Brothers, 1952).
National Municipal League, Model State Constitution, Sixth Edition (Revised)
(New York, 1968).
Donald G. Zauderer, Consequences of Ballot Reform: The Ohio Experience,
National Civic Review (November 1972), pp. 505-507, 520.
1. Office of Federal Elections, United States General Accounting Office, A
Study of Election Difficulties in Representative American Jurisdictions: Final
Report (Washington, D. C.: January 1973) p. VII-8.
2. Jack L. Walker, Ballot Forms and Voter Fatigue: An Analysis of the Office
Block and Party Column Ballots, Midwest Journal of Political Science (Novem-
ber 1966), pp. 448-463.
3. Norman C. Thomas, Voting Machines and Voter Participation in Four
Michigan Constitutional Revision Referenda, Western Political Quarterly
(September 1968), pp. 409-419.
4. Office of Federal Elections, United States General Accounting Office, A
Study of Election Difficulties in Representative American CitiesPreliminary
Report (Washington, D. C.: September 15, 1972), p. VII-7.

The Federal Role in the
Conduct of Elections
The purpose of this chapter is to describe the expansion of fed-
eral involvement in the conduct of elections, to consider the proper divi-
sion of government responsibility for the conduct of elections, and to
suggest how federal and state governments can cooperate to improve
election administration. The federal government should continue to pro-
vide strong leadership in setting national standards for the expansion of
voting opportunities and incentives to the states for the improvement of
election administration and voter registration procedures. At the same
time the states should be free to meet national standards by developing
procedures compatible with their institutions and traditions, and with
the characteristics of the population they serve.
An Expanding Federal Role
State authority over elections has developed in the absence of a clear
federal constitutional guarantee of the right to vote. In the years before
the Civil War states took the initiative in extending the suffrage, making
this country the first major democracy to enfranchise all white male
adults. Since the Civil War, the federal constitution has become increas-

ingly specific about protecting voting rights. Eight of the 14 amend-
ments enacted since that time have expanded the suffrage, from the
fourteenth amendment through others prohibiting state restrictions on
the right to vote because of race, sex, failure to pay a poll tax and age.
The impact of these constitutional guarantees has been significantly
advanced by the federal courts. Using the fourteenth and fifteenth
amendments, the U. S. Supreme Court has invalidated such racially
discriminatory practices as the grandfather clause (1915), the white pri-
mary (1944) and the poll tax in state elections (1966). In the early 1960s
the Supreme Court began to apply the equal protection clause of the
fourteenth amendment to state laws governing legislative apportion-
ment and voter qualifications. Its historic series of legislative districting
decisions gave significant new emphasis to the franchise with the doc-
trine that equal value must be given to each vote. Then, in a less noted
but just as revolutionary series of decisions, the court began to apply
a strict standard of equal protection review to state laws governing eligi-
bility to vote. When a state acts to deny the franchise to a group of cit-
izens, the court ruled it must do more than demonstrate that its laws are
reasonably related to state policy goals, the traditional judicial test. In
voter qualification cases, states must demonstrate a compelling inter-
est in limiting the franchise and must use the least restrictive device
available. Applying this standard the court has invalidated state laws
excluding military personnel from voting, whether or not they could
qualify as residents (1965); the poll tax in state elections (1966); require-
ments that voters in school board elections be parents or property own-
ers (1969) and requirements that voters on bond issues be property
owners (1969). In one of its latest rulings the court invalidated state
laws requiring that citizens be residents for a specified period of time
before becoming eligible to vote (1972).
Through these decisions the Supreme Court has severely limited the
license of the states to restrict the franchise. Moreover, the court, by
giving each vote equal value, has placed an individuals political rights
under much the same constitutional protection as his civil rights. In ef-
fect the court has established a set of national standards for certain por-
tions of the electoral process. Congress also influences the conduct of
elections in two major ways; it can protect voting rights in all elections
from inequitable infringement and it can regulate the conduct of federal
elections. In 1957 Congress enacted the first of a series of civil rights
laws to eliminate discrimination against black voters. Congressional ef-
forts to protect voting rights reached a peak with the Voting Rights Act
of 1965 which suspended literacy tests and authorized the United States

Attorney General to send federal examiners to register voters in areas
covered by the act. Then in the Voting Rights Act Amendments of 1970
Congress mandated national standards for participation in presidential
elections by eliminating durational residence requirements and estab-
lishing national standards for absentee voting and registration. The 1970
amendments also included a provision lowering the voting age to 18 in
all elections. Although the Supreme Court subsequently ruled that Con-
gress had no authority to lower age qualifications for state and local
elections, it upheld the 18-year-old standard in federal elections and set
the stage for adoption of the twenty-sixth amendment which enfran-
chised young voters in all elections. The amendment was ratified in
record time as states rushed to avoid the administrative hardships of im-
plementing different voter qualifications for federal and state elections.
The Proper Distribution of Authority
Now that eligibility barriers to voting have been substantially over-
come, reform initiatives at the federal level are beginning to focus on
administrative and procedural obstacles such as registration. In 1972
and 1973 legislation was introduced in the Congress that would create a
national postcard registration system for federal elections but with fi-
nancial incentives to states that accepted it for state elections. Other
plans have been introduced to involve the federal government directly
in assuming responsibility for the conduct of voter registration, as well
as proposals to provide states with grants-in-aid to expand registration
These proposals reflect an important shift in congressional concepts
of federal responsibility for elections, first evident in the 1970 Voting
Rights Act Amendments. Until the passage of that act the federal role
had been directed at setting standards and policing state actions that
might restrict individual rights. Thus when Congress amended the Vot-
ing Rights Act in 1970, establishing rules for voting for President, there
was great confusion among the states. To give but one example,
federally-fixed deadlines for receiving absentee ballots for President
came into conflict with state laws. The new federal law departed from
the traditional approach in which procedures for voting are left to the
states, or more precisely to local officials acting under state laws. For in
spite of its undisputed power to regulate its own elections the federal
government does not conduct elections anywhere.
To a large degree the increasing federal interest in administrative pro-
cedures is a consequence of state inaction. The administration of elec-

tions has long been relegated to a back seat among state priorities and a
strong impetus is needed to encourage serious reform. There is a con-
tinuing need for federal involvement in regulating federal elections and
in setting standards for all elections. Federal intervention, moreover,
could provide the uniformity that is often the key to equal voting oppor-
tunity. However, there are some difficulties to consider.
Centralization of rule-making power could mean rigidity as well as
uniformity if a single national system were substituted for the diversity
of current state election practices. For example, North Dakota and small
communities in two states have no registration requirements at all, an
approach that may be appropriate in places with a stable population. A
uniform national registration system should not impose registration in
areas where it is not really needed nor exclude other possibilities for
A national system might tend to concentrate on the election of the
President, at the expense of other contests for congressional as well as
state and local offices. It could also have a significant impact on state
primary elections which go to the heart of the states widely varying
party systems. For example, present state distinctions between open and
closed primaries might be hard to maintain under a federal registra-
tion system that did not keep concurrent party enrollment records.
Different state calendars for holding primaries would have to be
changed to meet the schedule of the national nominating conventions.
Thus there is some danger that a federally administered system geared
to federal contests might neglect state and local interests. In view of the
growing clamor for public action and community control there is a real
need to encourage citizen activity in politics at the state and local level.
States might well be encouraged to overhaul their own systems to stim-
ulate voter participation in the far more numerous state and local elec-
tions. The prospect of dual bureaucracies at the state and federal levels
could cause great duplication of effort and expenditure as well as voter
The national and state governments should not be regarded as com-
petitors for authority but as two levels of government that can comple-
ment each other in efforts to meet the growing demands on both. The
resources of state and local governments should supplement national ac-
tion where necessary and relieve the national government of having to
divert its resources and energies to activities that could be handled as
well or better at other levels.
In spite of the drawbacks to national administration there is a growing
need for reform pressures from the federal government. The strength of

the federal system may lie in the freedom that states have to develop
new approaches to solving problems, but electoral reform among the
states has been rare. Several avenues are open to the federal govern-
ment to improve election administration without jeopardizing the state
administrative role. The federal government should continue to pursue
its traditional role as the protector of an individuals right to vote. The
Voting Rights Act of 1965, extended for an additional five years in 1970,
has been a dramatically effective vehicle for increasing black voter
registration in the South. Congress should continue to extend the act and
pass other legislation necessary to protect voting rights from unconsti-
tutional infringement by individuals or governments. The United States
Department of Justice should be encouraged to enforce vigorously the
provisions of the Voting Rights Act and to prosecute violations of other
federal voting rights statutes. Congress can encourage innovative prac-
tices by providing grants-in-aid to states for specified activities:
1. Implementation of statewide systems of door-to-door or mail
registration under the supervision of a state agency;
2. Expansion of personal registration opportunities through in-
creased registration hours and locations, widespread use of
deputy registrars and absentee registration;
3. The creation of statewide training programs for local election
administrators and precinct personnel;
4. The establishment of a state association of election adminis-
trators with regularly scheduled conferences;
5. The design and implementation of electronic data processing
programs or other management techniques to modernize voter
registration and election administration procedures.
The federal government should also promote better communication
among state and local election officials and encourage continuing
studies of the problems of election administration through the Office of
Federal Elections of the United States General Accounting Office. The
Office of Federal Elections was created to monitor the campaign expen-
ditures reports required under the Federal Election Campaign Act of
1971, but the act also directed the Comptroller General to serve as a
national clearinghouse for information in respect to the administration
of elections and to enter into contracts for the purpose of conducting
studies of the administration of elections. This responsibility might
prove to be one of the most helpful functions of the office. There is a
critical need for greater exchanges of information among state and local
election administrators and the development of comparative studies of
the problems of conducting elections.

A Model Election System
Implementing Statute
The Implementing Statute is based on the preferred recom-
mendations of the Model Election System. Page numbers in parentheses
throughout the statute refer to alternative approaches discussed in the
text of the Model. Each state is encouraged to examine these alterna-
tives in the light of its own institutions and traditions. The Implement-
ing Statute does not provide substitute provisions to be used for alterna-
tive approaches.
Administration of Elections
Section 1. State Department of Elections.
In order to ensure uniform registration and voting opportunities
throughout this state, there is hereby created in the Office of the
Governor a Department of Elections to supervise the conduct of elec-
tions. (See pp. 21-23.)
Section 2. The Chief Electoral Officer
The Department of Elections shall be administered by a Chief Elec-
toral Officer who shall be appointed by the Governor, with the advice
and consent of the Senate, for a term of five years. The Governor, upon
notice and after hearing, may remove the Chief Electoral Officer for
cause. In the event of a vacancy in that office, the Governor shall ap-

point a successor with the advice and consent of the Senate to serve for a
term of five years from the date of appointment. The Chief Electoral
Officer shall hold no party or other public office.
Section 3. Duties of the Chief Electoral Officer
The Chief Electoral Officer shall:
a) superv ise the compilation of a state list of registered voters, includ-
ing door-to-door canvassing of all voting precincts in the state;
b) develop and implement training programs for all election officials
in the state and establish employment standards for all levels of person-
c) set statewide standards for polling places and voting times;
d) prepare information for voters on registration and voting proce-
dures and on statewide candidates and ballot issues;
e) publish and distribute to each county a political calendar, a man-
ual on challenge procedures, and a map of all election districts for state
and national office in that county;
f) convene an annual State Election Conference of County Adminis-
trators of Elections to discuss uniform implementation of state election
g) prescribe the forms of ballots and the form and wording on the
ballot of state referendum questions, issues and constitutional amend-
h) investigate nonperformance of duties or violations of the election
law by election officers or other persons;
i) require such reports from county administrators as he deems neces-
j) receive nominating petitions for state office and certify as to their
k) certify all election results;
l) maintain an accounting procedure for all election expenditures in
the state; and
m) prepare and publish biennial reports on the conduct and costs of
registration and voting in the state, including a tabulation of election
returns and such other information and statistics as he may deem appro-
priate. (See pp. 26-27.)
Section 4. Rule-Making Powers of the Chief Electoral Officer
In carrying out his duties and otherwise to ensure uniform registration
and voting opportunities throughout the state, the Chief Electoral Of-
ficer shall issue such rules and regulations as he deems necessary. All
such rules and regulations shall be available to the public, shall be in
writing, and shall have the force and effect of law throughout the state.

Section 5. The State Election Council
a) There is hereby created a State Election Council consisting of a
chairman and four members and four alternates.
b) Prior to the first day of January next following the first guberna-
torial election held after the effective date of this statute, each major
state political party shall, under procedures provided for in its rules,
designate two members and two alternates, one each designated to
serve for four years and one each to serve for two years. (The Council
may also be designed to represent more than two parties as suggested
on pp. 32-33.) Thereafter members and alternates shall serve for four
years beginning the first day of January following each biennial state
election. Members and alternates shall be succeeded upon the expira-
tion of their terms by members and alternates designated by the polit-
ical party which named the members and alternates whose terms ex-
pired. Each alternate shall serve in the stead of a designated member at
any meeting of the Council which that member does not attend. In the
case of a vacancy, an alternate shall succeed his designated member for
the remainder of his term. If the office of an alternate becomes vacant,
the political party which named him shall designate a new alternate to
serve for the remainder of his term.
c) After each gubernatorial election the political party whose candi-
date was elected shall, under procedures provided for in its rules, desig-
nate a chairman to serve for two years. At the end of two years, the other
major political party shall, under procedures provided for in its rules,
designate a chairman to serve for two years. In the absence of the chair-
man at a meeting of the Council a member from the same political party
shall preside. If the office of chairman becomes vacant, the political
party which named him shall name a chairman to serve for the remain-
der of his two-year term.
d) The State Election Council shall meet at least twice during any
year in which a general election is held. One meeting shall be held on
the day following the last day provided by law for challenging candi-
dates who seek to have their names on the ballot at a primary election
for state office. A second meeting shall be held not later than seven days
following the general election. The Council shall also meet at the request
of the Chief Electoral Officer; at the written request of a majority of its
members; and at any other times that the Council at its first meeting may
designate as regular meeting days. Members and alternates shall be
compensated only for expenses for attending meetings.
Section 6. Powers and Duties of the State Election Council
a) It shall be the responsibility of the State Election Council to inves-

tigate fraud and other irregularities in the conduct of elections. (See
pp. 30-33.) The Council may undertake such investigations on its own
initiative; at the request of the Chief Electoral Officer; on the complaint
of any candidate, political party, independent nominating committee, or
voter; or on the complaint of any county administrator. The Council
shall have the power to administer oaths and to issue subpoenas for the
appearance of witnesses and the production of books, papers, records
and other evidence in connection with such investigations. The Council
shall report its findings to the Attorney General.
b) The State Election Council shall report annually to the Chief Elec-
toral Officer on its activities and recommendations.
Section 7. County Administrator of Elections
In each county there shall be a County Administrator of Elections who
shall be responsible to the Chief Electoral Officer for the proper admin-
istration of state laws and policies concerning registration and voting.
(See p. 33 concerning cities.)
Section 8. Duties of the County Administrator of Elections
The County Administrator shall:
a) compile a list of eligible voters from the canvass established by law
and state administrative rule;
b) procure and distribute supplies required for registration and voting
in the county;
c) prepare and disseminate voter information, including a sample
ballot, as prescribed by the Chief Electoral Officer;
d) draw precinct lines in accordance with state guidelines and dis-
tribute maps to each precinct showing precinct boundaries;
e) appoint canvassers, poll inspectors, and other personnel required
for the conduct of registration and voting in accordance with standards
established by the Chief Electoral Officer. For each precinct, the County
Administrator shall appoint two poll inspectors and two canvassers of
opposing parties, to be chosen from lists of registered voters submitted
by each of the states two largest recognized political parties under pro-
cedures provided for in their rules. A person may serve as both poll
inspector and canvasser;
f) designate one canvasser in each precinct as a Precinct Deputy Reg-
istrar, and one poll inspector in each precinct as Precinct Supervisor.
A Precinct Supervisor and a Precinct Deputy Registrar shall have com-
pleted the state training program for precinct workers and canvassers,
respectively. A person may serve as both Precinct Deputy Registrar and
Precinct Supervisor;

g) remove for cause any employee under his jurisdiction and appoint
a successor;
h) carry out training programs for all county and precinct election
officials as prescribed by the Chief Electoral Officer; and
i) receive and handle complaints referred to him by any voter or pre-
cinct official involving circulation of petitions, challenges to voters, ac-
tions of election officials, or irregularities of any kind in registration and
voting. He shall refer complaints to the Chief Electoral Officer, to the
State Election Council, or to the proper prosecuting authority, as may be
Section 9. Duties of the Precinct Supervisor
The Precinct Supervisor shall supervise the precinct election officials
and shall be responsible for:
a) the adequacy of poll facilities and supplies;
b) order at the polls;
c) enforcement of regulations against electioneering;
d) proper instruction of voters on voting procedures;
e) requiring all precinct personnel to bear visible identification;
f) prompt reporting to the County Administrator of complaints, chal-
lenges, or irregularities; and
g) authorizing, subject to the approval of the County Administrator,
the use of paper ballots in the event of voting device malfunctions.
Section 10. Financing Elections
The state shall assume the costs of:
a) voter registration;
b) the State Election Conference;
c) administration of the State Department of Elections and any re-
view, investigation or proceeding conducted by the Chief Electoral Of-
d) the State Election Council and any investigation or proceeding it
may conduct;
e) the development and implementation of training programs for
county and local election officials; and
f) the election of state officials.
Section 11. Budget Review
The Chief Electoral Officer shall require annual budget reports from
each County Administrator of Elections. The Chief Electoral Officer
shall review county election budgets and make appropriate recom-
mendations to the county governing body concerning expenditures for
registration, elections, and administration, including supplies, equip-

nient and personnel. The Chief Electoral Officer may revise county
election expenditures for any costs assumed by the state pursuant to
Section 10.
Voter Registration
Section 1. The Statewide Canvass
A list of registered eligible voters shall be revised biennially by a door-
to-door canvass conducted in each precinct in the state. (Or annually.
See p. 42.) The door-to-door canvass shall begin seventy days before the
date established for the primary election for state offices and shall be
completed within fifteen days thereafter. In each precinct the canvass
shall be conducted by the two canvassers appointed for that precinct by
the County Administrator of Elections.
Section 2. Procedure for Canvassing
Canvassers shall visit each household in their precinct to verify the
names of those registered at that address, to register unregistered eligi-
ble voters residing at that address, and to remove the names of persons
no longer registered at that address. If canvassers are unable to obtain
full and accurate information at a household on their first visit, a second
visit shall be made. One such visit shall be made between 9 A.M. and 6
P.M. and the other between 7 P.M. and 10 P.M. At each household the
canvassers shall leave a Notice of Registration for each resident
whose name is on the registration list. Any member of a household who
registers may register any other eligible voter resident in that household.
If a second visit will be necessary, the canvassers shall leave a Notice
of Visit card stating the day and hour that they will make a return visit.
The card shall include the name, address, and telephone number of the
canvassers. If, on a subsequent visit or visits, the canvassers are unable
to complete the registration of all eligible voters in that household, the
canvassers shall leave a Registration Information card containing the
information necessary for personal registration. Upon completion of the
canvass the canvassers shall prepare and certify a list of the registered
voters in the precinct and deliver the list to the County Administrator of
Section 3. Notice of Cancellation of Registration
Upon receipt of the list of registered voters from each precinct, the
County Administrator shall send a Notice of Cancellation of Registra-
tion to each voter previously registered whose name does not appear
on the new precinct list.

Section 4. Preliminary List
Within fifteen days following completion of the canvass in a county,
the County Administrator of Elections shall prepare an official Pre-
liminary List of Registered Voters. Copies of this list shall be available at
cost to all legally qualified candidates for office and all political parties.
Any citizen who claims to be an eligible voter may examine the list at
the office of the County Administrator of Elections. No one shall be
permitted to make an unofficial copy of the list. During a period of ten
days following preparation of the Preliminary List of Registered Vot-
ers, any eligible voter resident in the county may add his name to, delete
his name from, or challenge any name on the list. The County Adminis-
trator of Elections shall ensure that a reasonable number of opportuni-
ties are available to the voters of the county to challenge names on the
list. No additions or deletions of names shall be made after the close of
such period of ten days. Within five days following the close of such
period, each County Administrator of Elections shall forward the cor-
rected Preliminary List of Registered Voters to the Chief Electoral Of-
Section 5. The State List of Registered Voters
Not less than twenty days prior to the date of the primary election, the
Chief Electoral Officer shall certify the corrected Preliminary List of
Registered Voters for all counties as the State List of Registered Voters.
Section 6. Continuous Registration
Eligible voters may register or reregister at any time in person at the
office of the County Administrator of Elections or with the Precinct
Deputy Registrar of the precinct in which such voter resides. Each Coun-
ty Administrator of Elections shall maintain open office hours in
his county sufficient to ensure reasonable registration opportunities
throughout the year.
Revised Article on Suffrage and Elections
Since the sixth revised edition of the National Municipal
Leagues Model State Constitution was published in 1968 several
changes in Article III, Suffrage and Elections, have been made neces-
sary by constitutional amendment, federal legislation and U. S. Supreme
Court decisions. The Suffrage and Elections Article presented here has
been rewritten to take these changes into account. The voting age has

been set at 18 to match the requirements of the twenty-sixth amendment
to the U. S. constitution ratified on June 30, 1971. No reference is made
to durational residency requirements in accordance with the U. S. Su-
preme Courts ruling in Dunn v. Blumstein 405 U.S. 330 (1972) that
such calendar waiting periods are unconstitutional. However, the phrase
bona fide has been appended to the word resident to emphasize, as
the court did, that states retain authority to require bona fide residence
for voting purposes even though they may not require voters to have
been residents for a specified length of time. The section on literacy
tests has been deleted in accordance with Section 201 of the Voting
Rights Act Amendments of 1970 which imposes a nationwide ban on
such a test or device until August 6, 1975.
Article III
Suffrage and Elections
Section 3.01. Qualifications for Voting.
Every citizen of the age of 18 years and a bona fide resident of the
state shall have the right to vote in the election of all officers that may be
elected by the people and upon all questions that may be submitted to
the voters; but the legislature may by law establish disqualifications for
voting for mental incompetency or conviction of felony.
Disfranchised for conviction of a felony: Ark III 1, Mo VIII 2.
This section is based on the assumption that the broadest possible
participation in the electoral process is desirable and that voter quali-
fication requirements should be kept to a minimum. The minimum man-
datory requirements included here relate to age and bona fide residence
in the state. The legislature also has the discretionary power to disqual-
ify otherwise qualified voters if they are mentally incompetent or have
been convicted of felonies. The wording of the section prohibits the im-
position of other requirements such as, for instance, property qualifica-
tions or the payment of poll taxes.
Section 3.02. Legislation to Prescribe for Exercise of Suffrage.
The legislature shall by law define bona fide residence for voting pur-
poses, insure secrecy in voting and provide for the registration of voters,

absentee voting, the administration of elections and the nomination of
Legislature to Prescribe: H II 4, Ky 147, 155; similar Alas V 1, RI
Amend XXXV.
Secrecy: Del V I; similar Alas V 3, Ariz VII,. Cal II 5, Colo VII 8, Ky
147, 155, NM VII 1, NY II 7, ND VI 29, Pa VIII 4, Utah IV 8, Va II 27,
Wash VI 6, Wyo VI 11.
Absentee Voting: Cal II 1, Conn VI 6, Del V 4A, H II 4, Kan V 3,
and 15 others.
Nominations: Legislature to enact laws to secure fairness in nam-
ing party candidates: La VIII 4, Miss III 247.
This provision expresses in simplified language what most state con-
stitutions wrestle with in a most tortuous manner. The legislature is re-
quired, and is assured of all the necessary authority, to build a statutory
framework essential for a proper electoral system. The many specific
details of election administration are thus left to legislation.
It may be that some of the matters listed here are implicit in the gen-
eral legislative power and therefore need not be enumerated in the
Model, but narrow interpretations of constitutional provisions in this
area have served to prevent the legislature from acting in regard to some
of these matters, particularly as to absentee voting.
The section directs that the legislature define specifically residency for
voting purposes and protect the voters important right to cast his ballot
in secrecy. The legislature also is charged with responsibility to provide
systems for voter registration and absentee balloting.
The requirement that the legislature provide for the process by which
candidates are nominated places an important phase of the total election
system under the same constitutional regulation as the actual election
Further Readings: Paul J. Piccard, Citizen Control of Government,
in John P. Wheeler, Jr., ed., Salient Issues of Constitutional Revision
(New York: National Municipal League, 1961); James F. Blumstein,
The Supreme Court and Voter Eligibility, and Richard Claude, The
Federal Voting Rights Acts, in Richard J. Carlson, ed., Issues of Elec-
toral Reform (New York: National Municipal League, 1973).

National Municipal League
Advisory Committee on Election Systems
Wilson W. Wyatt, Chairman*
President, National Municipal League
Louisville, Kentucky
Theodore M. Berry
Cincinnati, Ohio
Jack G. Blue
County Clerk
Alameda County
Oakland, California
Earl Blumenauer
Member, Oregon State House of Rep-
Portland, Oregon
Mary Bradfute
Former President, League of Women
Voters of Pennsylvania and New
Boca Raton, Florida
Richard S. Childs
Honorary Chairman, Executive Com-
mittee, National Municipal League
New York, New York
George H. Gallup, Chairman
American Institute of Public Opinion
Former President, National Municipal
Princeton, New Jersey
Stephen K. Galpin, Manager
Community & Government Relations
General Electric Company
New York, New York
Joseph P. Harris
Professor of Political Science Emeritus
Berkeley, California
Donald G. Herzberg
Dean of the Graduate School
Georgetown University
Washington, D.C.
Penn Kimball
Graduate School of Journalism
Columbia University
New York, New York
Charlotte Roe
Executive Director
Frontlash, Inc.
New York, New York
Richard Scammon, Director
Elections Research Center
Washington, D.C.
William W. Scranton
Former Governor of Pennsylvania
Former President, National Municipal
Scranton, Pennsylvania
Richard G. Smolka, Director
Institute of Election Administration
The American University
Washington, D.C.
Marjorie R. Spear
Former President, League of Women
Voters of California
San Diego, California
Julia Davis Stuart
Former President, League of Women
Voters of the U.S.
Spokane, Washington
Amalia Toro
Office of the Secretary of the State
Hartford, Connecticut
Richard F. Treadway
Director, Field Services
U.S. Department of Commerce
Boston, Massachusetts

Gus Tyler, Director
Politics, Education & Training
International Ladies Garment
Workers Union
New York, New York
Fay H. Williams
Chairman, League of Women Voters
Education Fund-NML Election
Systems Project Advisory Commit-
Indianapolis, Indiana
In addition to the above, the following individuals provided assistance to the
NMLs Election Systems Project as members of the advisory committee for the
survey of administrative obstacles to voting conducted by the LWVEF:
Gino Baroni, Director
Center for Urban Ethnic Affairs
Washington, D.C.
Charles A. Barr
Director of Political and Economic Ed-
Standard Oil Company
Chicago, Illinois
Robert L. Bennett, Director
American Indian Law Center
Albuquerque, New Mexico
Lucy Wilson Benson, Chairman
League of Women Voters Education
Fund and President, League of
Women Voters of the U.S.
Washington, D.C.
Willie Campbell, Chairman
Special Research and Projects
League bf Women Voters Education
Washington, D.C.
Benjamin S. Hite
Alhambra, California
Dolores C. Huerto, Vice President
United Farm Organizing Committee
Delano, California
Barbara Jordan
Member, U.S. House of Representa-
Houston, Texas
John Perkins, Assistant Director
Committee on Political Education
Washington, D.C.
William J. Pierce
School of Law
University of Michigan
Ann Arbor, Michigan
Matthew A. Reese
Matt Reese & Associates
Washington, D.C.
John Lewis, Executive Director
Voter Education Project, Inc.
Atlanta, Georgia
John Sayre
Republican National Committee
Washington, D.C.
Althea Simmons
New York, New York
Ruth L. Sims
Former President, League of Women
Voters of Connecticut
Hamden, Connecticut
Anne Wexler
Common Cause
Washington, D.C.
Denotes present or former member of the National Municipal League Council.

Supplemental Remarks
Joseph P. Harris, whose work established a basis for prior League models on
voter registration and election administration, feels that the signature com-
parison should not be optional but should be required in each state. His state-
ment follows:
Each voter should be required to sign the registration record. In states which
require party registration by voters to vote in primary elections this information
should be recorded by the voter when he signs his registration record, thus
avoiding the necessity for a separate party registration. In the conduct of door-
to-door registration the canvassers should register voters whom they interview,
and should leave a registration record form at the residence of other eligible
voters with instructions for the voter to complete the form, sign it and mail to the
registration office in an attached postpaid envelope.
States which require voters to apply in person to register usually require
the voter to sign the registration record. His signature, in combination with the
requirement in most states that voters shall sign the poll list when they apply to
vote, provides the means for an instant identification of voters at the polls, and
is undoubtedly the most effective available safeguard against voting frauds.
It may be objected that it would be difficult to secure the signatures of voters
in a door-to-door registration, for many will be absent when the canvassers
call. This difficulty is readily overcome by the procedure suggested above. It is
not too much to ask of the voter that he sign and return his registration by mail
to the local registration office.
The requirement that the voter sign his registration record provides a needed
protection against fraudulent voting of absentee ballots, for the voters signa-
ture may be compared with the signature on his application for an absentee bal-
lot, as well as on the returned ballot. When the voter signs his registration
record he may also register his party affiliation in states where such registra-
tion is required, avoiding the expense and bother of maintaining a separate
party registration. When the voter signs his registration record he may supply
the identifying information, such as the place of his birth, the date, and his so-
cial security number if required, which are needed if computers are used to
process registration records. For these several reasons it is highly desirable to
require voters to sign their registration records, and it should be mandatory in
urban areas.

Carl H. Pforzheimer Building
47 East 68 Street, New York, N.Y. 10021