. 27
( 35 .)


Kasuya also found a similar pattern in the Senate where an average of 33.3% of
opposition incumbents switched to the president™s party between elections (2001a, 23).
Calculating electoral volatility for the Philippines post-1986 is dif¬cult due to shifting
party alliances and the fact that candidates often run under more than one party label.
As a rule of thumb I treated each party in a temporary electoral alliance as a separate
party. For candidates that ran under more than one party label, I credited the votes to
the largest party with which the candidate was af¬liated (measured by vote share in
that election). In the few cases where candidates declared themselves independent but
also ran under a party label, those candidates were treated as independent.
Vote share data are not available for the 1987 election. For the 1992 election, the vote
shares data are available for only 174 of the 200 districts. For the 1998 election, vote
share data are based on parties™ performance in the nominal tier seats. An alternative
way to measure electoral volatility is to calculate changes in seat rather than vote
shares. Seat share data tells the same story “ greater volatility post-Marcos. During the
1946“69 period, volatility was 28.0 versus 45.4 from 1987 to 1998.
Building Party Systems in Developing Democracies

table 6.2. Lower Chamber Electoral Volatility

Country Time Span No. of Elections Mean Volatility
United States 1944“94 25 4.0
United Kingdom 1974“97 6 8.3
Uruguay 1974“94 3 10.4
Italy 1946“96 13 12.0
France 1945“93 14 18.3
Philippines I 1946“69 7 18.5
Argentina 1973“95 7 18.8
Venezuela 1973“96 6 22.5
Costa Rica 1974“98 7 25.0
Poland 1991“94 3 28.4
Brazil 1982“94 4 33.0
Thailand 1983“96 7 34.0
Philippines II 1992“98 3 37.3
Russia 1993“99 3 60.0
Sources: Mainwaring (1999); author™s calculations from Hartmann, Hassall, and Santos
(2001); Parliamentary Elections around the World (http://ww.universal.nl/users/dreksen/
election); Elections around the World (http://www.agora.stm.it/elections/); Centre for the
Study of Public Policy (http://www.cspp.strath.ac.uk//intro.html); Election Resources on

the Internet presented by Manuel Alvarez-Rivera (http://electionresources.org/).

There are other indications that party labels mean little to candidates.
Guest candidatures “ where a party invites a candidate from another party
to run under its banner without formally switching parties “ are not
uncommon. It is also not unusual for candidates to eschew attachment
to a single party, opting instead to run as an independent or as a joint
candidate “ a candidate running under more than one party banner. In the
1992 election, candidates with joint af¬liation or with no party af¬liation
whatsoever captured 7% of the House seats.5 In 1995, they captured
15%. Interestingly, nearly 7% of the 1995 seats went to candidates who
carried the banners for the government party and one of the opposition
Like candidates, voters do not place a high value on party label. Voters
frequently split their votes between candidates from different parties
(Mangahas 1998). Nowhere is this more evident than in the election of
president and vice-president. Filipino voters cast two separate votes, one
for a presidential candidate and one for a vice-presidential candidate.

This excludes candidates that ran as part of the formal LAKAS-LDP electoral alliance.
Philippines: Term Limits, Aggregation Incentives, No. of Parties 155

These votes need not be for candidates from the same political party.
Taking advantage of this rule, voters frequently split their votes between
candidates from two different parties. As a result, both the 1992 and
1998 presidential elections returned a president and vice-president from
different political parties. In 1998, the vote shares of presidential and
vice-presidential running mates differed by an average of 18.7 percentage
Filipino parties also tend to have less than national constituencies,
like their Thai counterparts. Few, if any post“Marcos parties could be
described as national parties. Parties have been unable to cultivate
lasting nationwide support (witness the high level of electoral volatility)
and most lack a national policy focus. Even during the pre-1972, two-
parties system, the Nacionalista and Liberal Parties are best seen as
shallow alliances of locally based and locally focused politicians, rather
than cohesive national political parties with distinct policy visions.
Politicians™ electoral fortunes depend primarily upon their ability to
deliver targeted bene¬ts to narrow constituencies rather than collective
goods to more national constituencies. In short, candidates, and their
respective parties, have a focus that is more local than national.
One indication of this subnational focus is the lack of a serious
national policy or ideological orientation by Filipino parties.6 Party
platforms are notable for their lack of distinctive ideological or national
policy content. An extreme example occurred in the run up to a recent
election. Several different parties, including parties in both the gov-
ernment and opposition, ended up hiring the same group of consultants
to write their party platforms. Because of the strong similarities across
all of the platforms the consultants adopted a simple rule to keep each
distinct “ use a different font for each.7 As this anecdote illustrates, the
major differences between parties are not differences over national
policy. Elections then are not battles between different ideologies or
party programs but rather struggles between personalities for the con-
trol of government resources.
To summarize, far from being cohesive unitary actors, political parties
in the Philippines are factionalized or atomized. Party switching occurs

The only exception to this is parties on the Left “ which have generally performed
poorly at the polls “ and some new party list parties each of which can capture a
maximum of three seats in the House.
Interview with political consultant, July 2000. Anonymity requested.
Building Party Systems in Developing Democracies

regularly, and party labels carry little weight for either voters
or candidates. In addition, party constituencies are more local than
national. In the words of one scholar: “Far from being stable, program-
matic organizations, the country™s main political parties are nebulous
entities that can be set up, merged with others, split, resurrected, regur-
gitated, reconstituted, renamed, repackaged, recycled or ¬‚ushed down the
toilet anytime” (Quimpo 2005). This characterization of the Philippine
party system is consistent with a lack of party system institutionalization,
as de¬ned by Mainwaring and Scully (1995). Recall that Mainwaring and
Scully de¬ne institutionalized party systems as those in which (a) there is a
regular pattern of electoral competition, (b) parties have stable roots in
society, and (c) parties have organizations that “matter.” While a stable
two-party system was the norm prior to 1972, parties in the post“Marcos
period have yet to exhibit regular patterns of competition. As the electoral
volatility ¬gures suggest, party fortunes vary greatly from election to
Political parties also lack stable roots in society. One indication of this
is the high degree of electoral volatility. Another is the average age of the
largest parties since the return of democracy. One might have expected
that a return to democracy would bring a return to prominence of the
Nacionalista and Liberal Parties “ the two parties that dominated the
pre“martial law period. In many Latin American countries, there was
just such a continuity of political parties before and after periods of
authoritarian rule. In the Philippines, however, a return to democracy
brought with it a whole host of new parties. The average age of parties
with at least 10% of the House vote in the 1995 election was less than 6
years. Table 6.3 places this ¬gure in a comparative context. Though the
old Liberal and Nacionalista Parties were revived, they have yet to win
more than a handful of seats in any of the post“Marcos elections.
Finally, parties have yet to develop party organizations that “matter,”
Mainwaring and Scully™s third criterion. Parties remain centered around
notable individuals and function almost solely as electoral vehicles. As a
result, parties are noticeably devoid of any lasting organizational struc-
ture. In between elections, parties hibernate, with very little in the way of
ongoing connections to party “members.” A 1997 study of 10 Filipino
parties found that every party considered mass member recruitment a top
priority; however, none of the 10 parties were able to produce a mem-
bership list, suggesting that “political party memberships . . . are as ¬‚uid
Philippines: Term Limits, Aggregation Incentives, No. of Parties 157

table 6.3. Years Since Founding of Parties with
10% of the Lower Chamber Vote, 1996

Country, Election Year Average Age
United States, 1996 154
Uruguay, 1994 115
Argentina, 1995 54
Costa Rica, 1994 47
France, 1993 43
Chile, 1993 40
Italy, 1996 39
Venezuela, 1993 29
Thailand, 1996 20
Brazil, 1994 13
Philippines, 1995 6
Sources: Mainwaring (1999); author™s calculations.

as the party system itself.” (Carlos 1997a, 220) The internal governance
structure of parties is also notoriously weak. Members who deviate from
the party line (when there is one) are rarely sanctioned. In fact, of the
major political parties that were active during the 1980s and 1990s, only
one has ever employed a party whip or similar institutions to compel
members to toe the party line and protect the party label “ Marcos™s KBL
Party (Kilusang Bagong Lipunan) (Carlos 1997a, 224). Finally, respon-
sibility for and control of ¬nancing is very decentralized. Campaign
contributions generally ¬‚ow directly from the donor to candidate
(or faction leader), totally bypassing the formal party organization (de
Castro, Jr. 1992; Carlos 1997a).
There are a variety of historical, sociological, and institutional
explanations for why the Philippines party system developed as it did.
To begin with, the characteristics of party system partly re¬‚ect the
Philippines™ experience with colonialism and state building as well as the
social structure in place as it transitioned to elected government. A weak
central state has been the historical norm in the Philippines. In contrast
to much of the rest of Southeast Asia, precolonial Philippines lacked
major kingdoms able to exercise control over large areas.8 The dearth of

The only exceptions to this were the Muslim kingdoms in parts of the southern
Building Party Systems in Developing Democracies

large political units enabled the Spanish to quickly conquer most areas of
the Philippines. However, the new colonial administration was never
able to exercise strong and centralized political control over the islands “
relying instead on Catholic priests to represent its authority in most areas
due to a chronic shortage of men and money (Andaya 1999). At the
same time, a new class of large provincial landowners emerged in the
Philippines that, in the vacuum of Spanish authority, gradually came to
dominate much of provincial life. This land-owning elite, known as the
oligarchs, became the patrons atop numerous patron“client networks
spread throughout the Philippines (Tancango 1992).
It was into this environment that the United States stepped when it
replaced Spain as the colonial power at the beginning of the twentieth
century. Several of the U.S. colonial government™s decisions had the
unintended consequence of hampering the development of a more
institutionalized, cohesive, nationally oriented party system (Hutchcroft
and Rocamora 2003). First, even though the United States installed
democratic institutions in the Philippines, it did very little to build up a
strong central administrative bureaucracy. As a result, political and
economic power remained spread among the various large land-owning
elite throughout the country.
Second, the decentralized and fragmented nature of political life was
reproduced at the national level via the early introduction of parties and
elections in the Philippines (Hutchcroft and Rocamora 2003). As the
political system was thrown open to electoral competition, those in the
best position to compete for elected of¬ce were the oligarchs. The nat-
ural building blocks for their electoral machines and political parties
were the pervasive patron“client networks. As a consequence, the oli-
garchs were able to use elections as a means of acquiring and strength-
ening political power, ¬rst locally, then nationally via congressional
elections (Land 1965; Wurfel 1988; Hutchcroft and Rocamora 2003).
Political parties and Congress quickly became the domain of these
powerful locally based interests, rather than a forum in which mass
interests could be articulated and national policies debated.9 Con¬‚icts
and competition between oligarchs manifested themselves via party
switching or intra-party factionalism. In sum, the parties that came to

For an analysis of the policy consequences of this arrangement, see Sidel (1996) and
Hutchcroft (1998).
Philippines: Term Limits, Aggregation Incentives, No. of Parties 159

dominate the political system were not cohesive parties with national
constituencies, but highly fragmented or atomized parties with narrow,
particularistic constituencies.10
Historical and sociological variables are helpful for understanding the
early development of the party system. However, they cannot completely
account for why key features of the party system have endured in the
Philippines. In fact, many of the sociological and historical factors have
varied over time in a way that would seemingly support the emergence of
a more institutionalized party system. By the 1960s, traditional patron“
client networks were breaking down, beginning ¬rst in and around
Manila and then spreading to other areas of the Philippines (Wurfel
1988). Likewise, a new class of business elite had emerged to challenge
the power of the oligarchs. This business elite (largely Manila-based)
had interests that were very different from the traditional landed-elite
(Hawes 1992).
One could argue that path dependence might account for the sticki-
ness of the party system in the face of these changes. However, given the
political, economic, and social upheaval of the Marcos era, it is not
dif¬cult to imagine that new paths were at least possible after his fall
from power. The extended presidency of Ferdinand Marcos accelerated
the relative decline of the oligarchs as he sought to centralize political
and economic authority while empowering a new class of cronies
(Hawes 1992).11 To oust him from power, opposition political parties
joined together to back Corazon Aquino for president. They were sup-
ported by the mobilized mass of the Filipino populace. Yet this mass
mobilization, greater centralization, and the relative decline of the oli-
garchs did not lead to the creation of large, mass-based national parties.
Nor did the coming together of different opposition groups to overthrow
Marcos translate into more cohesive parties post-Marcos. Instead, the
party system that emerged was similar in most respects to the pre-1972


. 27
( 35 .)