Dilemmas of Electoral Clientelism: Taiwan, 1993
| Author | Charles Kurzman,Chin-Shou Wang |
| Published date | 01 March 2007 |
| Date | 01 March 2007 |
| DOI | http://doi.org/10.1177/0192512107075408 |
| Subject Matter | Articles |
Wang & Kurzman: Dilemmas of Electoral Clientelism 225
International Political Science Review (2007), Vol. 28, No. 2, 225–245
DOI: 10.1177/0192512107075408 © 2007 International Political Science Association
Sage Publications (Los Angeles, London, New Delhi, and Singapore)
Dilemmas of Electoral Clientelism:
Taiwan, 1993
Chin-Shou Wang and Charles Kurzman
Abstract. For many years, studies of electoral clientelism regarded
clients as the captive votes of patrons. In recent years, this conventional
wisdom has come under challenge, as scholars have come to recognize
the widespread noncompliance of clients. This article uses the case
of the 1993 Taiwan election to offer the fi rst ever systematic data on
noncompliance. Documents from the ruling Kuomintang (Nationalist
Party) campaign offi ce in one Taiwanese district, combined with district
electoral results, demonstrate considerable leakage in this instance of
clientelistic mobilization: at least 45 percent of voters who sold their
votes to the Kuomintang did not, in fact, vote for the Kuomintang’s
candidate. This article argues that clientelistic mobilization faced at least
four serious obstacles, including (1) broker scarcity, (2) factionalism,
(3) embezzlement, and (4) fi nancial limitations. These obstacles prevented
the Kuomintang from making full use of its broker organizations, even
as it devoted extensive economic and political resources and personnel
to the election.
Keywords: • Clientelism • Electoral corruption • Kuomintang • Taiwan
• Vote buying
The Study of Electoral Clientelism
There is no consensus in the study of electoral clientelism about the mechanisms
that bind clients to patrons. But for many years, there was a common agreement
in the fi eld that such mechanisms are highly effective. Patrons were thought to
mobilize and deliver “vote banks” or “blocks of votes” (Chubb, 1981: 80–1; Graziano,
1975: 33; Hagopian, 1996: 48–9; Rouquie, 1978: 25). Clients were often treated as
“captive votes,” whose support for their patrons was automatic and unproblematic
(Ames, 1994: 96; Coppedge, 1993: 262–3; Graziano, 1977: 370; Guterbock, 1980: 10;
Mouzelis, 1985: 337; Rouquie, 1978: 24–5). The client’s obedience was frequently
characterized as total and permanent, as in Carl H. Lande’s (1977: xxvii) introduction
to an infl uential anthology on political clientelism: “The client must attempt to
226 International Political Science Review 28(2)
pay part of his unrepayable debt in advance through a continuous display of
affection, deference and obedience to his patron. Even then, it is made clear to
him that he remains perpetually a debtor.”
But there is little empirical research on the effectiveness of clientelistic mobil-
ization. Indeed, in recent years increasing numbers of scholars have started to
recognize the importance of clients’ noncompliance with their patrons. A study
of Chilean elections in the 1960s, for example, notes that “while regidores
[clientelistic patrons] boasted that they could deliver many of ‘their voters’ to
the congressional candidate of their choice, they also admitted that many times
clients did not follow their directives” (Valenzuela, 1977: 83).1 A study of Colombia
notes that there is often “little guarantee that this aid [that is, patronage] would
translate into actual ballots” (Hartlyn, 1988: 173). A study of Brazil concludes that
clientelism can be “fraught with uncertainty and subject to constant challenge,
renegotiation and change” (Gay, 1999: 49; see also Gay, 1994, 1998). In addition,
recent work on Argentina has explored poor people’s perceptions of political
clientelism, perceptions that are often less than positive, concluding that clients
outside of a patron’s inner circle are not nearly so beholden as commonly
thought (Auyero, 1999, 2000a, 2000b). Frederic Schaffer’s introduction to a new
anthology on electoral clientelism adopts quite a different tone from Lande’s
1977 anthology: “vote buying is not as effective as one might suspect given the
heavy fi nancial and organizational investments that candidates and parties are
willing to make” (Schaffer, 2007).
But there has been no systematic study of clientelism’s performance during the
election process. This article addresses this gap with a case study of a clientelistic
system that was arguably in crisis: the Kuomintang (Nationalist Party) vote-buying
apparatus during the Taiwanese elections of 1993. A comparison of Kuomintang
vote-buying lists in one Taiwanese county with actual electoral results shows
considerable leakage in clientelistic mobilization: at least 45.4 percent of voters
who accepted Kuomintang money for their votes did not actually vote for the
Kuomintang’s candidate. In the process of mobilizing voters and brokers, this
article argues, the Kuomintang faced at least four serious obstacles that prevented
the Kuomintang from making full use of its broker organizations, even as it em-
ployed considerable economic and political resources and personnel in the 1993
election.
Captive Voters?
The clientelistic relationship runs in two directions. The patron appeals to the
client, and the client responds to the appeal. The literature on clientelism, while
stressing the importance of both aspects, has focused its research and debate al-
most exclusively on one side of the equation: what appeals do patrons make?
The literature identifi es three sorts of appeals: material, normative, and coercive
(see Schaffer and Schedler, 2007).
Material Appeals
The dominant approach to political clientelism has long been the “resource-
based” model. According to this model, the simplest and most important mechanism
of political clientelism is the exchange of material resources (Clapham, 1982: 2;
Lande, 1977: xv). “The interaction on which they [patrons and clients] are based
To continue reading
Request your trial