Revisiting Majority-Minority Districts and Black Representation

• Published date: 01 June 2018
• DOI: 10.1177/1065912917738574
• Date: 01 June 2018
• Author: William D. Hicks,Carl E. Klarner,Seth C. McKee,Daniel A. Smith
• Subject Matter: Articles

https://doi.org/10.1177/1065912917738574

Political Research Quarterly

2018, Vol. 71(2) 408 –423

© 2017 University of Utah

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1065912917738574

journals.sagepub.com/home/prq

Article

Ever since the United States moved toward a more inclu-

sive and democratic system of elections, the color of rep-

resentation has been a major concern of American politics

scholars (Whitby 1997). Particularly in the American

South,1 the end of Jim Crow and the subsequent massive

re-enfranchisement of African-Americans via the 1965

Voting Rights Act (VRA) made it evident that eventually

there would be many local settings where blacks would

finally have the ability to elect one of their own (Bullock

and Gaddie 2009; Davidson and Grofman 1994; Valelly

2004). Indeed, in the five Deep South states (Alabama,

Georgia, Louisiana, Mississippi, and South Carolina)

containing the region’s highest African-American popu-

lations, all of their upper and lower legislative chambers

now consist of majority-black Democratic state legisla-

tive delegations that face off against much larger and

almost entirely white Republican delegations controlling

these legislatures (McKee and Springer 2015).

In this context, descriptive representation—meaning

the election of black legislators who represent the inter-

ests of mostly African-American constituencies (Swain

1993; Tate 2003)—is a major feature of the contemporary

American political landscape, especially in southern

states where the black electorate is substantial. The rise in

black representation, however, has raised some notable

normative issues (Canon 1999), specifically in terms of

party politics. The work of Epstein and O’Halloran (1999,

2000, 2006) makes it clear that with the emergence of the

southern Grand Old Party (GOP), the growth in the num-

ber of majority-minority districts has come at the direct

electoral expense of white Democrats, as the attendant

increase in the number of neighboring majority-white

districts now greatly favors Republicans (Black 1998;

Hill 1995; Kousser 1999; Lamis 1999).

The crowding out of white Democrats in electoral

politics would not (perhaps) be that great a concern if it

did not also contribute to an overall reduction in the size

of Democratic delegations. But the replacement of white

Democrats with black Democrats has been anything but

one-to-one (McKee 2010; Petrocik and Desposato 1998),

738574PRQXXX10.1177/1065912917738574Political Research QuarterlyHicks et al.

research-article2017

1Appalachian State University, Boone, NC, USA

2University of Florida, Gainesville, USA

3Texas Tech University, Lubbock, USA

Corresponding Author:

Daniel A. Smith, Department of Political Science, University of

Florida, 234 Anderson Hall, Gainesville, FL 32611, USA.

Email: dasmith@ufl.edu

Revisiting Majority-Minority Districts

and Black Representation

William D. Hicks1, Carl E. Klarner2, Seth C. McKee3,

and Daniel A. Smith2

Abstract

What is the minimum black population necessary to elect African-American state lawmakers? We offer the most

comprehensive examination of the election of black state legislators in the post-Thornburg v. Gingles (1986) era. We

begin by charting changes in the partisan affiliation of state legislators and the percentage of black legislators from

1971 to 2016. This descriptive assessment is undertaken according to important regional (Non-South and South) and

subregional (Rim South and Deep South) contexts in American politics. We then perform multivariate analyses of the

likelihood of electing black legislators across three periods following the marked increase in the creation of majority-

minority districts (1993–1995, 2003–2005, 2013–2015). Because of sectional variation in the partisan strength of the

major parties, the probability of achieving black representation is significantly different depending upon whether a

contest occurs in the Non-South, Rim South, or Deep South, with the latter constituting of the highest threshold of

black population necessary to elect an African-American. By merging an original dataset on state legislative elections

with the most complete evaluation of the factors shaping the election of black lawmakers, our findings shed new light

on minority representation and how sectional differences greatly affect the electoral success of African-Americans.

Keywords

majority-minority districts, black representation, state legislative elections, Deep South, Rim South

Hicks et al. 409

as Republicans have been the primary beneficiaries of the

increase in black representation. In many instances, pack-

ing minority voters into fewer districts has directly con-

tributed to Republican takeovers of congressional and

state legislative delegations (Hill 1995; Hood and McKee

2013), and particularly so in the South as compared with

the rest of the country (Black and Black 2002).

Furthermore, this partisan trade-off triggered by drawing

majority-minority districts also entails a trade-off between

descriptive versus substantive representation (Cameron,

Epstein, and O’Halloran 1996; Epstein and O’Halloran

2006). Although there is some disagreement with respect

to what substantive representation exactly encompasses

and how it should be measured (see Grose 2011), it is

hard to deny that white Democrats are more responsive to

minority interests than are Republicans (LeVeaux Sharpe

and Garand 2001, 2003; Overby and Cosgrove 1996).

Hence, the overall reduction in elected white Democrats

may result in the paradox of “less” representation of

black interests if majority-minority districts foster

Republican legislative majorities (Lublin 1997b).

We revisit the matter of majority-minority districts

and black representation because several features still

have not received their due. First, the extant literature is

dominated by congressional analyses. Yet, because of

the size of state legislative delegations and the election

of hundreds more black legislators, there is considerably

more analytic purchase when evaluating black represen-

tation in state legislative elections. Second, race has been

and continues to be the driving factor for the election of

black legislators (Grofman 2006; Lublin et al. 2009), but

it is not the only factor. With this in mind, most existing

scholarship fails to include many controls when assess-

ing the likelihood of electing a black lawmaker (Lublin

1997a and Lublin et al. 2009 are exceptions). With the

most comprehensive dataset on state legislative elec-

tions, we rectify concerns over underspecified models by

including numerous factors that potentially affect black

officeholding. Third, nearly all previous studies offer

only a static snapshot of black electoral success in a

given year, either immediately prior to or following a

decennial redistricting. In contrast, we look dynamically

at how the threshold for electing black state lawmakers

may fluctuate over time.

Finally, although much of the literature on black repre-

sentation makes the appropriate distinction between the

Non-South and South because of sectional variation in

black electoral success rates, few have taken the next step

of distinguishing between Rim South and Deep South

states (but see Black and Black 2002; Bullock and Gaddie

2009). As we will demonstrate, the Deep South subregion

drives the disparity in the probability of electing a black

legislator; in the heart of Dixie, where black populations

are the greatest, it is most difficult to achieve black

representation because these electorates are the most

racially polarized (Black and Black 2012; Grofman 2006;

Hood, Kidd, and Morris 2012; McKee and Springer

2015; Valentino and Sears 2005; White 2014).

Our study proceeds as follows. First, we document

changes in the partisan profiles of state legislators and

the share of black lawmakers from 1971 to 2016. Our

descriptive data stress that regional (Non-South and

South) and subregional (Rim South and Deep South)

changes to the partisan affiliations of state legislators is

the broader context in which the attainment of black rep-

resentation occurs. The most telling development in

black representation transpired with the Supreme Court’s

ruling in the 1986 North Carolina redistricting case,

Thornburg v. Gingles. In the wake of this decision, which

compelled North Carolina and other states, particularly

those in the South, to greatly expand their number of

majority-minority districts (Cunningham 2001), there

has been a marked increase in the share of black state

legislators. Our data analysis begins in earnest in this

post-Thornburg era of legislative politics. We offer pre-

liminary evidence of the relationship between the elec-

toral success of African-Americans and the black

population in their state legislative districts according to

where these lawmakers reside: Non-South versus South,

and Rim South versus Deep South. We then move beyond

these descriptive presentations to perform several multi-

variate analyses that model the likelihood of electing

black lawmakers across regions and subregions. We con-

clude by discussing the important political and represen-

tational implications of our findings.

Partisan Change and Black State

Legislators, 1971–2016

The basic delineation between the Non-South and South

reveals some palpable differences in the party affiliations

of state legislators from 1971 to 2016. Although some

scholars have further subdivided regions within the Non-

South (see Black and Black 2007), we adhere to the more

common Non-South/South bifurcation because variation

in the likelihood of achieving black representation is typi-

cally expected to be substantially different between these

two major regions, not within Non-South subregions.

However, because of greater racial polarization in voting

patterns between the Rim and Deep South subregions, we

do emphasize the significance of this division.2

In the figures that follow, the time-series run from

1971 to 2016. As is the case for all the data presented in

this section, we weight state legislative seats so that the

upper chamber is proportional to the lower chamber in

the computation of these partisan splits (see the online

appendix for further details). Combining state house and

state senate chambers, Figure 1a displays the partisan