The Disadvantageous Effects of At-Large Elections on the Success of Minority Candidates for the Charlotte and Raleigh City Councils Paper by Grofman

    q- r- r{

H-f
tu

;

,7r
/\,)/,,t2,

//* p,r

,r'
gAa-rr* ,n'Pfr"l-

JLh
L*,



.\ rj'c" '-: .j'
: ,-.,.!. . I

,..;./ xi . ....

E t)ISADVANTAGEOUS EFFECTS OF AT-LARGE ELECTIONS

ON THE SUCCESS OF t'llINORITY CANDIDATES

FOR THE CHARLOTTE AND RALEIGH CITY COUNCILS*

J,r I

:,1..:'

ai;rul,q,r 
-':

f{,-L:.',.
,1 ,i . 'r;i.i'.:.,;'. *.i'i';i, ,

'.'':-.,,.:' ..;'; -'.t, .!'.. ."

...:'.,-,.;:

School of Social Sciences

: PIAINTIFF'S
EXHIBIT

,e9
Gr n,1"-s

C

[Jniversity of California, Irvine



THE DISADVANTAGEOUS EFFECTS OF AT-LARGE ELECTIONS

ON THE SUCCESS OF MINORIIY CANDIOATES

FOR THE CHARLOTTE AND RALEIGH CITY COUNCILS*

Bernard Grofman
Professor of Pol'ltical Sclence
School of Soclal Sclences
Unl vers 1ty of Cal'lf ornla, Irvl ne
Irvlne, Cal 1 fornla

May 20, .l983

*The portlon of thls research conducted March to Ju1y,.l982, was
supported by Natlonal Sclence Foundatlon Grant #SES 8l-07554, Polltlcal
Sclence Program. I am lndebted to the staff of the UCI School of Soclal
Sclences Word Processlng Center for typlng and table preparatlon. Data
were obtalned f rom or'lg'lna1 data sources by Sarah Crowder, Esq. The
vlews expressed are entlrely the responslbtllty of the author and are not
offlclal vlews of the Natlonal Sclence Foundatlon or the Unlverslty of
Callfornla, Irvlne.



I. Campalgn Expendltures 1n the Dlstrlct-Based

and At-Large Component of the Charlotte Clty Councll and

and Ralelgh Clty Councll Electlons in .l979 and lgBt

We would llke to test the hypothesls that at-large electlons are more

expenslve to run than dlstrlct-based campalgns. Intultlvely 1t would

seem very reasonable that at-'large electlons, lnvolvlng as they do iarger

constltuencles, would be more costly.l However, there are a number of

methodologlcal problems 1n emplrlca'l1y valldatlng what mlght appear

commonsenslcally obv'lous; even though the few avallable studies (e.g.,

Grofman 
.l982; Jewell .l982) all support the truth of the proposed

hypothes 1 s :

(l ) There are dlfferences 1n spendlng patterns between incumbents and

non-lncumbents. Moreover, those dlfferences are compllcated by the

conslderable lncumbency advantage 1n ralslng money versus the

countervalllng lesser need of h1gh1y vlslble lncumbents to spend money to

w1n electlons. Also the magnltude of the lncumbency advantage is often

dlfferent 1n at-large than 1n slngle member dlstrlct elect'lons.

(21 Both at-large and dlstrlct races contaln candldates who run with

llttle chance of vlctory (and wlth mlnlmal campalgn expenses), but the

number of such candldates 1s generally greater 1n at-large electlons.

lCampalgn funds are often spent somewhat dl fferently. 1n at-1arge
than 1n d'lstrlct electlons; for the latter, use of clty-w1de medla (e.9.,
radio, TV, clty newspapers) 1s less efflclent than for the former and
thls may reduce somewhat the cost advantages produced by the smaller
scope of dlstrlct-based campalgns.



(
2

(3) Many candldates largely flnance. clty councll campalgns through

the'lr own funds, and such personal resources vary w1de1y, lntroducing

ldlosyncratlc features whlch are hard to control for because of the smal'l

number of mlxed system electlons for whlch we have campalgn fundlng data

avallable for ana'lys1s.

Nonetheless, each of these methodologlcal problems assoclated with

analyzlng comparatlve campalgn expendl tures across d1 fferent types of

electlon systems may be solved (or at least mlilgated) 1f (l) we

dlstlngulsh between lncumbent and non-lncumbent expendltures lZ) for both

lncumbents and non-lncumbents we focus on the expendltures of the wlnnlnq

candldates, and (3), we comblne data so as to obtaln a larger sample slze

and more rel lable data estlmates. ['.le shal I look at charlotte c1 ty

councll and Ra'le1gh clty councll campalgn expendltures patterns,

comblning 1979 and l98l data.

In Charlotte there were four at-large seats and seven dlstrlct seats

1n both the .l979 and l98l electlons (see Appendlces 1 and 2). Comblnlng

data for the two electlons we flnd wlnners at large averaged over $.I2,000

on campalgn expendltures (whether they were lncumbent or non-lncumbent);

whlle 1n the dlstrlct based e'lections, wlnning challengers spent on'ly

$5,8.l5 and wlnnlng lncumbents spent only $3,198 (see Table I ). Thus,

Table I about here

campalgn costs 1n Charlotte C1 ty councl 1 at-large electlons hrere, on

average, more than twlce those for dlstrlct electlons 1n that clty.



3

In Ralelgh, for both the .l979 
and the l98l e'lect1on, there were two

at-'large seats and f 1ve dlstrlct seats (see Append'lx 3 and 4). Comblnlng

data for the two e'lectlons we flnd lncumbent wlnners at-1arge spent an

average of $9,.l05 whlle lncumbent dlstrlct wlnners spent an average of

only $5,344; non-lncumbent at-1arge wlnners spent an average $1.l,925

whlle non-lncumbent dlstrlct wlnners spent on average only $5,2.l3. Thus,

at-'large campalgn costs 1n Ralelgh at-'large clty councll electlons were,

on average, roughly twlce those for dlstrlct electlons tn that c1ty.

Table 2 about here



(

II. Success of Black Candldates in the Dlstrict-Based

and At-Large Component of Charlotte Ci ty Councl 1

and Raleigh C1 ty Councl 1 E lectlons

The cons 1 derabl y h1 gher expendl tures requl red to run a succes sful

at-large race 1n Charlotte lmposes a burden on mlnorlty groups (such as

blacks) who are economlcally dlsadvantaged. Thls flnancial burden,

combined wlth raclal bloc voting whlch makes for a greater dlfflculty of

b'lack success 1n at-large race wlth a primarlly whlte electorate as

compared to a dlstrlct race wlth a prlmarily Black electorate (e.9.,

Charlotte Districts 2-3l,, has meant that Blacks are dlsproportlonately

excluded from the at-large councll seats 1n Charlotte. In the perlod

'1977-198.l , of the 2l dlstrlct seats contested, 81acks won 6 (28.6%t;

whlle of the l2 at-large seats contested Blacks won only 2 (16.1%1,

desplte the fact that there were more Black candldates for the four

at-1arge seats than for the seven dlstrlct seats. In the precedlng

perlod, .l945-.l9i5, 
under a pure at-large system, Black representation was

even 1ess, averaglng only 5.4% l[e111g and Mundt l98l; see a'lso Hei11g,
.l978; Hundt .l979).

As 1n Charlotte, Black electoral success 1n Ralelgh was conslderably

greater 1n the dlstrlct than 1n the at-large component of the clty

councll electlons 1n .l977-.l98.l. 0f the l5 dlstrlct seats contested,

Blacks won three (20.0%), whlle of the slx at-1arge seats contested,

Blacks won no seats (0.0%), desplte the fact that there were

proportlonally about as many Black candldates contesting the at-1arge



electlons as contestlng the dlstrlct electlons. Thls flndlng of greater

mlnorlty success 1n a dlstrlct-based system (or the dlstrict-based

component of a mlxed system) than under an at-1arge or mu'lt1-member

dlstrlct system has been repeated 1n a large number of munlclpallties and

other Jurlsdlctlons where there exlsts a substantlal mlnorlty populatlon

:' and patterns of polarlzed votlng (see esp. Engstrom and McDonald l98l;

,. Karnlg and l.lelch .l978, .I979; 
Grofman l98l; and overvlews of the

llterature 'ln Engstrom and McDonald 
.l984 forthcomlng and 1n Grofman

r982b).

"Indeed, few generallzatlons 1n po11t1cal sclence appear to be
as well verlfled as the proposltlon that at-1arge electlons
tend to be dlscrlmlnatory toward black Amerlcansr'(Engstrom
and Mc0onald, .l984 forthcomlng).

(



III. Summary

We examlned the campalgn expendlture patterns for the at-large and

dlstrlct components of Charlotte and Ralelgh, North Carollna clty councll

electlons and found that successful at-large electlon campalgns are more

expens'lve to run than successful dlstrlct campaigns. We then looked at

the success of black candldates 1n recent Charlotte and Ralelgh c'lty

councll races and found dramat'lcally greater success for black candldates

runnlng 1n the dlstrlct-based elections than for those runnlng for the

clty-w1de seats. In reduclng thelr llkellhood of obtalning offlce 1f

they do seek 1t, and,/or 1n lncreaslng the amount of money whlch must be

spent to achleve offlce, at-large electlons in Charlotte and Raielgh had

a d1 scrlmlnatory effect on Elack candldates, when compared w1 th dl strict
electlons 1n the same clties.



Tab'l e I 
I

Campaign Expenses:

Charlotte Clty Councl1, .l979-.l98.l

Wlnnlng Incumbents2

At-larqe 0lstrlct

W1 nni ng Non-Incumbents

At-larqe Di strlct

$.l8,.l42 l{one
l9 ,.l00

$.l 8,621I 979
expendi tures

$ 554
I,684
I,907 (N=2)
2,699
5,784
2,9.l4
5,675

$3,03'l (N=5 )average ( N=2 )

$5,706
4,945

$5,326

l.Jlnnlng Incumbents

At-larqe D1 strlct

$3'll9
I,936
2,717 average

$l 8,452 4,531 ( N=2 )
I 9 .669 4,900

$.I9,06.| (N=5) $3,433 (N=5)

l'l1 nni ng Non-Incumbents

At-larqe D1 strlct

$7 ,0.l 4
5,292

$8,7.l 7

2,91 3
I 981
expend 1 tures

average (N=2)

$6,.I 53 ( N=2 ) $5,8.l 5

I 979 and
I 981
comb 1 ned

average
(N=4)

l.llnning Incumbents

At-larqe Dlstrlct

$.l 2,.l 94 ( N=.l 2 ) $4,.I 98

Wi nni ng Non-Incumbents

At-larqe Dlstrlct

$l 2,387 ( N=2 ) $5,81 5(N=9)

lThere were not enough wlnnlng black candldates to make 1t feaslble to
separately tabulate by race of candldate. The raw data on whlch thls table
was based 1s provlded as appendlces to thls research note.

2In .l979 and l98l all lncumbents running for reelectlon to the Charlotte
Clty Councll won reelectlon. In .l979 9 of ll lncumbents sought reelectlon;
tn .l98.l, 7 of ll did.



(
Table 2l

Campalgn Expenses:
Ralelgh C1 ty Counci 1, .I979-.I98.|

l.ll nni ng Incumbents

At-larqe Dlstrlct

t,'l'l nn 1 ng Non-Incumbents

At-larqe Dlstrlct

1 979
expendl tures

average ( N=l )

$3.598
$3, 598

$.l5,723
4,lg7

257
5.048

$ 6,304
$t 0,0.l 6

$t 0,0.I 6
$8.962
$8 ,962

I 981
expend'ltures

average ( N=l )

$l 4,6l l
$l 4,6l l ( N=4 )

$5,3.I0.l.301

$4,383 (N=l )

l,Jl nnl ng Incumbents

At-larqe Dlstrlct

!.ll nnlng Non-Incumbents

At-larqe Dlstrlct

$.13,834 $.l,463
$.l3,834 (N=l ) $.|,463

(
Wl nn i ng Incumbents

At-larqe 0lstrlct_

$9,'105 (N=8 ) $5,344 (N=2 )

Hlnnlng Non-Incumbents

$.l 
.l 
,925 ( N=2 ) $5, 2.l 3

lThere were not enough wlnnlng b'lack candldates to make'lt feaslble to
separately tabulate by race of candldate. The raw data on whlch thls table
was based 1s provlded as appendlces to thls research note.

At-larqe Dlstrict.l979 
and

I 98.l average ( N=2 )
comb'lned



Appendl x 1

Charlotte Cl tv Councl I .l979 
Campalqn Exoend'l tures

tJ*

* - lncumbents
Source: Repor t

At Larqe

Dave Berryh111
Llnwood 8oy1es
Betty Chafln
Franklln Burnham
John Hardlng
Rudy Hendrlcks (B)* Pat Locke
t{lll1e Payne (8)
Marnlte Shuford (B)
Herberty Spaugh

0l strlct I

(D)

Dlstrlct 2

$.l 8,.l 42..l 
.l

7,793.24
5,706.95

236. I 6
8,649. 2l
2,079 . 

.l 
3

4,945 .96
I 98. 46

8,875.99.l9,100.21

554.09

D

R

0

D

D

R

R

D

D

D

(D)
(D)

(0)
(R)

(D)
(R)

H

Oon Carrol I

Charlle 0anne11y (B)
Sy1vla Jordan

Dlstrlct 3

l.l * Ron Leeper (8) (D)

Dlstrlct 4

I ,684.40'l 
,201 .92

I 8,264.56
5,784.64

I ,907 .84

2, 699 .48
I 0,768.62

tl * Laura Frech
Elllott Newcombe

John Q. Burnet te
trl * ['1lnette Trosh

Dlstrlct 5

Dlstrlct 6

James Breneman
Sandy Judson
George Seldon

Dlstrlct 7

Tom Cox
Gene Golding

H = wlnner (B) =

of the Board of Electlons

(0)
(D)
(R)

I ,383.60
2,967 .51
2,9.l4.76

5,675..l3
655.00

(R)
(D)

Black



't0

Appendl x 2

Charlotte Cl tv Councl I l98l Electlon Campalqn Expendl tures

At Larqe

Dave Berryhl 1 1

Lee Roy Epps
Harvey Gantt (B)
Brenda Hllliard
Jlm Hutchlnson
l,lar 1 e Jordan
Pat Pat ton
Sue Myrick
Ed Peacock
Blll Payne (8)
Rlckey Reid (B)
Herber t Spaugh

Denn 1 s Cudd
M1 ke Fennel I
Mark Grlffin
Buddy Reid
Parn Patterson

G'lenn Fogle
Laura F rech

$l 8,452.71
996.41

7,0.l3.77
I,69l.ll

371.34
I I 4.00

4,271 .25
I I ,496 .0.l
5,29.l .63

r 42 .00
422 .42

19,669.52

(

(

(

(

(

(

Dlstrlct I

(D)
(D)
(R)
(D)
(D)

Dlstrlct 2

Dlstrlct 4

(R)
(D)

Dlstrlct 5

D)
R)
D)
D)
R)
D)
R)
R)
R)
D)
D)
D)

1 ,232 .44
I ,209.77

233.0r
7 ,646 . 23
2,9.l3.73

t{

l.l

t.J

Charlle Dannelly (8) (D)
Sy1v1a Jordan (D)

0lstr1ct 3

Franklln Durham (B) (D)
Ron Leeper (8) (D)
Eugene Marsh (B) (D)

3, I 
.l8.94

536.08

57.00
I ,936. I 3

208. I 2

D)
D)
R)

3,3.l9..l6
2,776.57

7 ,840 .34
767. I 7

4,530.72

* - lncumbent
Source: Repor t

G. tl. Hartsell
l.larc Helm
M1 net te Trosch

!,l = wlnner (B) = Black
of the Board of Electlons



il
(

Appendlx 2 (continued)

Dlstrlct 6

Ann Maxwell (D

George Seldon ( R

Distrlct 7

Nell Flsher
George Kal lam
Ralph McMl I lan

5,596.66
4,799.73

D

D

R

I,653.04
I 08 .00

8,71 6 . 78

(



12

(
Append 1 x 3

0ctober 9,.I979 Ralelqh Cltv Councll Electlon

Campalqn Expendltures

At Larqe

w

[r|*

Sandy P. Babb
Spurgeon Cameron (B)
Theora M1 tchel I Rourk ( B )

Averay C. Upchurch
James B. I'lomble (appointed to

$ 5,442. 
.l 

8
2,02.l.99

233.00
10,016.52

lncumbency) 3,597.95

Dlstrlct A

I'landa Canada
Leslle L. Koehler
Tony Jordan

Dlstrlct B

Oak1ey Herring
tl * Edward A. Walters

Dlstrlct C

l.J Arthur J. Ca'lloway (8)* t.Jllliam R. Knlght (8)
Mayso Tomllnson, Jr. (B)

. Dlstrlct D

tl * Mlrlam P. Block
0. D. Rai nes

Dlstrlct E

l.l * John A. Edwards
Robert R. Gardner

* = lncumbents H =wlnner (B) = Black
Source: Report of the Wake County Board of Electlons

7,3.l 0.56
2,.l08.64

15 ,7 22 .69

.l 
,879 .7 5

4,.l86.70

8,962.1 7

6,945.86
376.26

256. 54
8t 4.00

5,048.98
I , I 03.34



13

(

*
*

Appendl x 4

0ctober 6. l98l Ralelqh C1 tv Councl I Electlon

Campalqn Expendltures

At Larqe

Sandra P. Babb
Ronald Kent Campbel'l
Avery C. Upchurch
James B. tr,lomble

$l 3,834. 35
3,464.66

I 4,6.l I .00
5,469.25

Arthur J. Calloway
Mayso Toml i nson, Jr

Dlstrlct A

Tony Jordan
l.l1l11am C. Young

0lstr1ct B

Edward A. t'lalters

Dlstrlct C

(B)
. (B)

Dlstrlct D

( no 1 ncumbent )

G.0an1el Coates
L. E. Colv'ln, Jr.
R. E. Deans
Rober t Hoy
Walter M. Keller (lncludlng run-off)
Charles G. Meeker ( lncludlng run-off)
Stephen G. Rea
Deslree Hhlte (B)

D1 s tr 1c t E

7 ,276.39
55.00

3,644 . 64

5,309 . gt
776.76

5 .00
430 .00

I ,664.40
I 52.40

I ,463.00
I ,567 .26.l,793.02

2,006 . 36

.|,30.l.06!'l * John

W = w1 nner
Source: Repor t

A. Edwards

lncumbent (B) = Black*-
of the Wake County Board of Electlons



l4

(

References

Engstrom, Rlchard L. and Mlchael [). HcDonald. "The Electlon of Blacks to

Clty Counclls: Clarlfylng the impact of Electoral Arrangements on

the Seats/Populatlon Relationshlp. " The American Polltlcal Sclence

Review, Vol. 75, No. 2 (June l98l ), 344-354.

Engstrom, Rlchard L. and Mlchael D. HcDonald. "Effects of At-Large vs.

Dlstrlct Electlons on Raclal Representatlon. " In B. Grofman and

A. LlJphart (Eds. ), Electoral Laws and Thelr Polltlcal Consequences.

New York: Agathon Press, .l984 forthcomlng.

Grofman, Bernard. "Alternatives to S1ng1e-Member Plurallty Dlstrlcts:

Lega'l and Emplrlcal Issues." Pollcv Studles Journal Speclal Issue on

Reapportlonment, Vol. 9, No. 6 (Apr11 .l981 ), 875-898. Reprlnted 1n

B. Grofman, A. LlJphart, R. McKay, and H. Scarrow (Eds.),

Representatlon and Redlstrlctlnq Issues. Lexlngton, MA: Lexington

Books , I 982, I 07 -l 28 .

Grofman, Bernard. Report on the constltutlonsallty of the Hawail

Reapportlonment Commlsslon's proposed state leglslatlve

redlstrlctlng. Prepared testlmony 1n Travls et al. v. Klng, U.S.

Dlstrlct Court for the State of Hawa11, Harch 23-24, 1982(a),

photo-offset.

Grofman, Eernard. "Hard vs. At-Large Electlons, Part II: A Revlew and

Crltlque of Twenty-Three Studies .l970-.l98.l." 
Unpubllshed manuscrlpt,

1982(b).



t5

He11ig, Peggy. "The Abandonment of'Reform'1n a Southern City:

0utcomes of a Return to Dlstrlct Politlcs." Dellvered at the Annual

Heetlng of the Mldwest Polltlcal Sclence Assoclatlon,.l978.

He1'119, Peggy and Robert J. Mundt. "Do Dlstricts Make a Dlfference?',

Urban Interest (Aprll, l98l), 62-75.

Jewell, Malcolm E. "The Consequences of Slngle- and Multi-Member

0lstr1cting. " In B. Grofman, A. LiJphart, R. McKay, and H. Scarrow

(Eds. ), Representatlon and Redlstrlctlnq Issues. Lexlngton, MA:

Lexl ngton Eooks, I 982 , 129 -l 35.

Mundt, Robert J. "Referenda 1n Charlotte and Ra1elgh, and Court Actlon

ln Rlchmond: Comparatlve Studles on the Rev'lval of Dlstrict

Representation." Paper presented at the annual meeting of the

Amerlcan Polltical Science Associtlon, .l979.