A Field Experiment on the Effects of Ethnicity and Socioeconomic Status on the Quality of Representation

• Author: Sarah Birch,Philip Habel
• Published date: 01 August 2019
• DOI: http://doi.org/10.1111/lsq.12230
• Date: 01 August 2019

389

LEGISLATIVE STUDIES QUARTERLY, 44, 3, August 2019

DOI: 10.1111/lsq.12230

PHILI P HA BEL

University of South Alabama

SARAH BIRCH

King’s College London

A Field Experiment on the Effects of

Ethnicity and Socioeconomic Status

on the Quality of Representation

The principle of equal representation is the cornerstone of modern de-

mocracy, yet there have been concerns that the benefits of representation can

be skewed, advantaging some over others. We argue that elected officials will be

more responsive to constituents whom they perceive as more like themselves and

more likely to be politically active. We examine inequalities in representation in a

parliamentary democracy where, intrinsic to the institutional framework, there is

a long-standing reputation for serving constituent needs: the United Kingdom.

We also advance the literature by focusing on the expediency of responsiveness

and the helpfulness of response. Drawing on a field experiment, we find both an

overall high level of responsiveness and helpfulness but also biases affecting MP’s

responsiveness. Our findings raise important questions about equal access to

representation, even in a political system where constituency service is the norm

and expectation.

In modern repre sentative democracie s, the principle of

equality provides a fu ndamental normative underpin ning for po-

litical institutions. All citizens are considered to have an equ al

right to contribute to shaping the public goods that result from

policy proces ses, and all citi zens can expe ct to benefit equa lly

from the outputs of the political syst em, with public ser vices ad-

ministe red impartially. This is how equal repres entation is under-

stood. Yet a number of studies, particu larly in the U.S. context,

have shown a disparity in re sponsiveness to cer tain groups or

sectors of societ y (Grose 2014). For example, in a series of stud-

ies, Butler (2014) finds th at U.S. politicians are more likely to

respond to coeth nic constituents, bui lding on earlier work with

similar f indings (Broock man 2013; Butler and Broockm an 2011).

Building on thes e and other studies, we arg ue that legislator

© 2019 Washington University in St. L ouis

390 Philip Habel and Sarah Birch

responsiveness to c onstituents’ personal c oncerns wi ll not fully

adhere to the democ ratic norm of impartiality but wil l instead be

sensitive to two factors, one strateg ic and the other psychologi-

cal. Strategic considerations suggest that repre sentatives should

be more responsive to constit uents whom they perceive to be

politically ac tive, as well as to constituents on whom the provi-

sion of servic e may have a persuasive ef fect vis-á-vis mobi lization

and even vote choice (Mayhew 1974; Searing 1994). Psychological

theory indic ates that politicians may exhibit ingroup bias in that

that may be more likely to respond to those citizens w ith whom

they share demographi c characterist ics. We go beyond previous

work in several importa nt ways. First, we disentangle the impact

of ethnicity and so cioeconomic status on legislator responsive-

ness. By focusing on a context where cla ss and ethnicity a re both

highly sal ient political cleavages, we are e asily able to signal both

character istics and to asce rtain their impact separately a nd in-

teractively. Second, we provide a more thorough asse ssment of

the quality of legisl ator responsiveness than h as been carr ied out

in previous studie s by examining overal l responsiveness together

with the temporal dy namics of respon siveness (through an event

history framework), as well as the helpful ness of response s. We

look at to whom Members of Parliament in the United K ingdom

respond, how fast they respond, and how helpful are their re-

sponses. Th ird, we focus on a parliamentary system. Given that

most democracie s are of the parliamenta ry type, res earch in this

context is espec ially needed to u nderstand the broader dyna mics

of democratic repre sentation. We do, however, examine a syste m

where the electoral i nstitution resembles that of the United States

and thus where legislators have a gre ater incentive to cultivate

a personal vote than in other parliamentar y systems, and where

responsiveness to c onstituent needs is well-establ ished and firmly

grounded in the cu lture of the legislatur e.

We adopt a field-experi ment approach, using email s to con-

tact MPs in the United Kingdom for help with a problem, but

varying random ly the occupation and putative eth nicity of the

emailer. The strategi c hypothesis is by the emai ler’s occupation,

either a professional position of an ac countant or a working-cla ss

position of a “builder” (construct ion worker). Despite a degre e

of class dealignment over the past several dec ades, the U.K.

party system re mains strongly class b ased, maki ng the occupa-

tion of the constituent a relevant marker of ele ctoral behavior.

Ingroup bias can be exam ined both through the occ upation of the

391A Field Experiment on Representation in the United Kingdom

constituent—where memb ers of Parliament may feel a stronger

affin ity to those in more professional car eers—and by the o sten-

sible ethnicity of the e mailer. Our find ings point to a high rate

of responsiveness and a h igh level of helpfulness, much greater

than previous studi es have shown. But we also find evidenc e of

inequalitie s in politicians’ provisions of constituent servi ce, par-

ticularly in di fferences in the efficienc y of responsiveness based

on the occupation and the ethnicity of the ema iler. Our theoreti-

cal and empiri cal contribution overall spe aks to broad and im-

portant concer ns about representation in democratic system s.

Responsiveness in Modern Demo cracies

Recent rises i n levels of economic inequ ality in many de-

mocracies have prompted c oncerns that the benefits of represe n-

tation may be skewed toward certa in sectors of society. We focus

on the servic e provided by elected repr esentatives to indiv idual

constituents who contact them, as have a number of recent stud-

ies in the United State s (Broockman 2013; Butler 2014; Butler and

Broockman 2011; White, Nathan, and Faller 2015) and to a lesser

extent, comparatively (Distel horst and Hou 2014; Meng, Pan,

and Yang 2017; Vaccari 2014). Helping citiz ens deal with prob-

lems and issues that arise is an important form of representation

that elected off icials deliver, offeri ng a useful means of gaug ing

whether the democratic p olitical process is effect ive in treating all

citizens equ ally. Moreover, focusing on the eff iciency and quality

of the provision of that service is important. Const ituents often-

times rais e concerns that ar e timely in nature, and i f politicians

and their staf f wait too long before intervening, any assistance

offered will b e less effectu al.

The job of a representative elected to a legislative ass embly

is a complex and time-i ntensive one, including resp onsibilities of

care to the constitu ents who elected him or her. There is a long tra-

dition—particularly in polities w ith single-member district elec -

toral systems— of representatives t aking on casework stem ming

from issues or persona l matters voiced by those in t heir district s

(Cain, Ferejohn, and Fiorina 1984, 1987; Fenno 1978; Norton and

Wood 1993; Searing 1994). Such service a ffords elected off icials

advantages beyond fulf illing a duty; providi ng casework can be

advantageous electoral ly as well (Cain et al. 1987). Helping citi-

zens deal with problem s can foster goodwill and c ontribute to a

reputation for competence. For citizen s of the same party, such