Jones and Mainwaring - The Nationalization of Parties and Party Systems

7/27/2019 Jones and Mainwaring - The Nationalization of Parties and Party Systems

1/28

THE NATIONALIZATION OF PARTIESAND PARTY SYSTEMS

An Empirical Measure and an

Application to the Americas

M ark P. JonesandScott M ainwaring

ABSTRACT

Political parties and party systems exhibit widely varying degrees of

nat ionalizat ion, tha t is the extent to w hich a pa rty receives similar levels

of electoral support throughout the country. The level of party nation-

alization has a prominent effect on such important factors as the

survival of democracy, the types of issues that dominate political

competition, legislative behaviour and public policy. In spite of its

importance, party nationalization has been neglected in the compara-tive politics literature. Our article makes two contributions. First, it

provides a measure of party a nd part y system nationa lization, ba sed on

the Gini coefficient, that is superior for comparative analysis to those

employed to date. Second, it utilizes these measures to analyse nation-

alizat ion in 17 democracies in the Americas, the first time nationa liza-

tion has been examined empirically outside the advanced industrial

democra cies. The measure underscores the w idely varying degrees in

nationalization across party systems, within party systems over time,

across parties within countries and w ithin part ies over t ime.

KEY WOR DS Latin America nationalization political parties

We address an under-ana lysed issue in the compa ra tive study of part ies and

party systems: their degree of nationalization. In the massive literature on

party systems, considerable attention has been paid to the extent of polar-

ization (Knutsen, 1998; Sani and Sartori, 1983; Sartori, 1976), the number

of part ies (Lijphart, 1994; Sar tori, 1976; Taagepera and Shugart, 1989) andthe level of institutionalization (Harmel and Svsand, 1993; Janda, 1980;

Mainwaring and Scully, 1995). With the exception of some work on the

United States, very little has been written on the degree of nationalization.

P A R TY P O L I TI C S V O L 9 . N o . 2 pp. 139166

Copyright 2003 SAG E Publications London Thousand Oaks New D elhi

1354-0688(200303)9:2;139166;030835
http://www.sagepublications.com/http://www.sagepublications.com/


2/28

It is time to address this lacuna. Parties and party systems vary signifi-

cantly in the degree to w hich they a re nationa lized. By a highly na tiona lized

part y system, w e mean one in w hich the major pa rties respective vote shares

do not d iffer much from one province to the next. In a w eakly na tiona lized

party system, the major parties vote shares vary widely across provinces.

The same logic applies to parties: highly nationalized parties have a

relatively even share of the vote across different geographic units, whereas

weakly nationalized parties have widely varying shares across geographic

units.

Party system nationalization reflects an important component of the

dyna mics of part y competition. Tw o systems may be similar in the number

of parties and the degree of polarization, the two dimensions Sartori (1976)

used to classify party systems in his classic work. Yet if the level of nation-

alization diverges sharply between these two systems, then the competitive

dyna mics are quite different. In a case of high nationa lizat ion, electora l com-petition follows a roughly similar pattern across the countrys sub-national

units. In a ca se o f low na tio na liza tion, the pa rties tha t fa re w ell in so me

sub-nationa l units are minor electora l competitors elsew here. O ur indicato r

of party system nationalization allows for precise measurement of such

differences across countries.

Moreover, two parties (either within the same system or not) may have

the same share of the national vote, but their political and electora l stra tegies

likely differ if one party w ins a roughly similar share of the vote in a ll sta tes

while the other is dominant in a few states and a secondary force in theothers. Our measure of party nationalization captures such differences.

In this article, we make three contributions. First and foremost, we

propose a new way to measure party systemnationalization and partynat ionalization. For measuring party na tiona lizat ion, w e ad vocate inverting

the G ini coeffi cient, a measurement of inequa lity across different units

widely used in other contexts. Most previous measures of nationalization

have significant shortcomings for comparative research. When clear and

careful measurement of concepts is possible, as it is in this case, it forms a

fundamental building-block of good social science. Our measures allow fora precise means of assessing nat iona liza tion a nd of comparing across parties

and party systems and across time. In addition, they reveal interesting

information about change over time within a given political system. If a

country moves markedly tow ard a more nat ionalized system, o r conversely

toward greater interstate differences, these phenomena deserve careful

at tention (Bra dy, 1985; C laggett et a l., 1984; Kaw ato, 1987; Stokes, 1967).

Our measure of party nationalization allows us to compare parties over

time, to see if part ies electo ra l increases and declines occur relat ively evenly

throughout the country o r a re geographically concentra ted, and to exa minew hether most part ies in the same country follow a similar patt ern or diverge

markedly in terms of nat ionalization.

Second, we provide data on the nationalization of party systems and

P A R TY P O L I TI C S 9 ( 2 )

140


3/28

parties in 15 Latin American countries, Canada and the United States.

Although this analysis is limited to 17 countries, our measures tra vel across

the great majority of countries without difficulty. The empirical data in this

article for the first time allow for careful and systematic comparisons of

party system nationalization across countries outside Western Europe and

the United States.

Finally, we call attention to an important phenomenon that has been

badly neglected in the comparative literature on party systems. One of our

principal findings is that party system nationalization varies greatly across

cases, with potentially important consequences for how democratic systems

function. Yet, notwithstanding great variance in party system nationaliza-

tion, the comparative literature on party systems has virtually neglected this

issue. Although variance in pa rty nat ionalization is an importa nt issue, w ith

the exception of the United States there have been few attempts to compare

across countries or time.1

We are aware of only two previous efforts to compare party or party

system nationalization across different countries (Caramani, 2000; Rose

and Urwin, 1975).2 Ironically, both previous efforts to compare party

nationa lization a cross different countries focused on the adva nced industrial

democracies, where (except for Canada, which is often not incorporated

into comparative analyses of the advanced industrial democracies) variance

across party systems has been considerably less pronounced than outside

this select group of countries. Our work indicates that scholars of parties

and party systems elsewhere neglect the topic of party and party systemnationalization at their peril. In short, this article furthers attempts to

expand the frontiers of the comparat ive understanding of party systems w ith

the hope tha t scholars w ill begin to pa y more systemat ic attention to differ-

ences in party system nationalization and their consequences.

Conceptual Issues and Measurement

A first essential step is clarifying w hat w e mean by nat ionalization o f partiesand party systems. In the literature on the United States, the concept of

nationalization has been used in two different ways. Schattschneider

(1960) and Sundquist (1973) utilized it to assess the extent to w hich sub-

nat ional units approximate na tiona l vote patterns; in this case, the absolutelevel of support for different part ies across sub-na tiona l units is the relevant

measure. In contrast, Stokes (1965, 1967), Brady (1985) and Bawn et al.

(1999) examined how closely the direction of electoral change in sub-

national units approximated the national pattern of electoral change (i.e.

the sw ing from one party to others).3

In their w ork, the vote shift from o neelection to the next rather than the absolute level was relevant. Claggett et

al. (1984) and Kaw ato (1987) used both meanings of na tionalization and

clearly distinguished betw een them.

J O N E S A N D M A I N WA R I N G : TH E N A TI O N A L I Z ATI O N O F P A R TI E S

141


4/28

Both conceptions of na tionalization a re meaningful, but the term na tion-

alization of parties or the party system should be reserved for the former

concept. Both Stokes and Brady measured not the nationalization of the

party system, but rather the nationalization of electoral trends (or swings).

The na tiona lization of electora l sw ings might bear a n empirica l relat ionship

to the nationalization of the party system (or, stated differently, to the

absolute levels of electoral support parties win across sub-national units).

Logically, how ever, the concept of party system nat iona lization should refer

to the structure of the party system, not to whether electoral swings are

similar a cross districts.

To measure the nat iona lization of parties, w e subtra ct the G ini coeffi cient

from 1. The G ini coeffi cient is a w idely used measure of ineq ualities across

units (Creedy, 1998; Shryock et al., 1976); it is, for example, the most

commonly used measure of income inequality. It ranges from 0 in cases of

perfect equality across all units to 1 in cases of perfect inequality. As usedhere, the G ini coefficient a ssesses the extent to w hich a party w ins eq ual vote

shares across all the sub-na tiona l units. A G ini coeffi cient o f 0 signifies that

a part y received the same share of the vote in every sub-na tiona l unit. A G ini

coefficient of 1 means that it received 100 percent of its vote in one sub-

national unit and 0 percent in all the rest. 4 We subtra ct the G ini coeffi cient

from 1 so that a high score indicates a high level of nationalization. We call

this inverted G ini coefficient the Part y Na tiona lization Score (PN S).5

The PN S has tw o primary purposes. First, it can tra ce changes in a part ys

level of nationalization over time. Of particular interest is verifying therelationship between electoral ascension or decline, on the one hand, and

the PNS, on the other. Second, the PNS can be used to compare parties

w ithin the same system or a cross systems. Tw o pa rties may ha ve the same

share of the vote but diverging levels of nationalization, reflecting very

different competitive situations. These differences are likely to affect the

parties electoral and congressional strategies.

In ad dition to being w idely know n a nd used, the G ini coeffi cient is tech-

nically superior to most existing alternatives. Like any measure of nation-

alization that merits serious consideration for cross-national analysis, it isnot country-specifi c; a G ini coefficient o f 0.15 in one country is compara ble

to a G ini coefficient o f 0.15 in another.6

Our measure of party nationalization is based on variance across states,

provinces, departments, administrative regions or parishes in parties elec-

toralperformance. Let us be clear about what this concept does notmeasure. A party might have pronounced cross-state differences in pro-

gramma tic chara cter and social ba se and yet w in the same share of the vote

in all of the sta tes. O n our measure, this party w ould be perfectly nat ional-

ized, notwithstanding the internal differences across states. Thus ourconcept of a nat ionalized party d oes not imply one that is programma tically

homogeneous across sub-national units. Nor does it imply a party that is

highly cohesive or disciplined at the national level. A nationalized party as


142


5/28

defined here could be a loosely knit organization of individual political

entrepreneurs who a re part of the same nat ional orga nization.

Building on the nationalization score for individual parties, we also

develop a measure of the nationalization of part y systems. To create this

measure w e multiplied the nationa lization score (PN S) for every pa rty by its

share of the national valid vote, and then summed this product for all the

parties. We call this weighted PNS the Party System Nationalization Score

(PSNS).7 The contribution of every party to the PSNS is thus proportionate

to its share of the vote. It is useful to have such a summary expression of

the level of nationalization of the party system for the same reason that it

is useful to have a summary expression such as the effective number of

part ies: it enables schola rs to tra ce changes over time within the same system

and to compare across countries.

Some Uses of the Measures

The PNS and PSNS allow researchers to measure nationalization precisely,

to undertake more informed comparisons across countries and over time,

and to locate cases on an important dimension of party systems. Our

measures could also be useful for several other purposes in comparative

political inquiry. For example, the degree of party system nationalization is

relevant for the most widely used measure of legislative electoral rules

(average district magnitude, or a lterna tives such as effective magnitude andeffective threshold) and of the national party system (the effective number

of electora l part ies). Where a pa rty system is highly na tiona lized, the use of

these measures is perhaps appropriate. In contrast, where a party system

possesses an intermediate or low level of nationalization, national-level

measures such a s the average district magnitude and the effective number of

electoral parties very likely suffer from validity problems.8 The effective

number of parties is still a meaningful indicator for cross-national com-

parisons and for understanding some aspects of national-level dynamics.

The nat iona l-level da ta , how ever, ca n ma sk major intra -country differences.In a weakly nationalized system, a portrait at the national level may be

meaningless in understanding sub-national dynamics, which are important

in most w eakly na tiona lized pa rty systems.

In the remainder of this section, we note four hypotheses that political

scientists have proposed about the effects of different levels of pa rty system

or party nationalization that could be more effectively tested through use of

the PNS or PSNS. First, Schattschneider (1960) linked the nationalization

of the party system to voters orienta tions. H e argued tha t in highly na tion-

alized party systems, national factors may be more important in forgingbonds between voters and parties. Conversely, in less nationalized party

systems, sub-national factors may be more salient in creating such bonds.

These hypotheses are not true by fiat; the nationalization of parties, as


143


6/28

defined here, is analytically distinct from whether voters are responding to

national issues. The measures of party nationalization in the United States

have also been important in understanding partisan realignment (Brady,

1985; Kawato, 1987), and the PNS could serve a similar function in com-

parative politics.

Second, the degree of na tiona lization reflects, and ma y have conseq uences

for, legislative careers and for executivelegislative relations. In highly

nationalized party systems, national issues are likely to be central in legis-

lators careers. Executives might have greater ability to forge legislative

coalitions on the basis of national issues and to negotiate with a few key

national party leaders. In a patchwork or weakly nationalized party

system, sub-national issues are likely to be more important in legislative

careers. Brazil illustrates this point, as Ames (2001) and Samuels (forth-

coming) have convincingly argued. Under conditions of w eak pa rty na tion-

aliza tion, the centra l party leadership may be less able to speak for the entireparty and to deliver its legislative support. This hypothesis also is not true

by fiat; it merits more systematic research. A good measure of nationaliza-

tion could help stimulate new research on such subjects and is essential for

the empirical testing thereof.

Third, differences in nationalization likely have public policy conse-

quences. Decisions related to national transfers to sub-national units,

administrative reform, and subsidies may be strongly influenced by the

degree of part y system na tiona lizat ion (G ibson and C alvo, 2000; Samuels,

forthcoming; Stepan, 2001). Where a partys base of support is relativelyconstant across geographic units, it may be more likely to treat all units

equally.9 In contrast, where its support varies widely across geographic

units, we would expect it to base its decisions in part on the degree of

support it receives in specific geographic units. These public policy conse-

quences seem particularly likely if a weakly nationalized vote pattern falls

along regional lines; that is, if a governing party fares markedly better in

some regions than others.

Fina lly, D iamond (1988), R eynolds (1999) and Stepan (2001) argue that

in new democracies where pronounced ethnic, national or religious cleav-ages coincide w ith territory, the nat ionalization o f some major pa rties may

be a key facto r in preserving democracy. In such contexts, it ma y be salutary

that some parties with sufficient electoral appeal to help form a national

government articulate programmatic countrywide concerns. In a weakly

nationalized party system in a new democracy with profound ethnic,

national or religious cleavages, parties will likely orient their electoral

messages tow ard specific ethnic, na tiona l or religious groups. Stepan (2001)

contends this pat tern may lead to w edge nationa l and ethnic politics rather

than to the politics of accommodation, to use Lijpharts (1977) term.10

H eargues (p. 331) tha t if the goa l is the consolida tion o f democra cy in a multi-

cultural or multinational polity, a strong case can be made that the exist-

ence of statewide parties is useful. The PSNS could help assess hypotheses


144


7/28

about the relationship between party system or party nationalization and

democratic stability in multi-ethnic societies.

The utility of our measures does not rest on whether these four hypoth-

eses are correct. Ra ther, the measures are useful because they a llow for

systematic comparison, show differences in pat terns of electora l competition

and pave the way for the testing of these and other hypotheses.

Case Selection: Countries,Geographic Units and Elections

Although our measure has broad applicability, this article empirically is

limited to the United States, Canada and 15 Latin American countries.

Within the Americas, our selection criteria w ere fourfo ld. First, w e included

only countries that held at least three consecutive democratic lower houseelections. Second, we included only democratic lower house elections held

since 1979, that is since the beginning of the third wave of democratization

in the region (Huntington, 1991). Third, w e included only medium to la rge

countries (i.e. those with populations greater than 2 million). Fourth, a

handful of countries or elections w ere excluded due to our inability to ga ther

the necessary electora l da ta (D ominican Republic 197890) or to problems

associat ed w ith the w ay the ava ilab le da ta are reported (C olombia, Ecuador

1998, Venezuela 2000).

Although it would be possible to examine presidential and senate elec-tions, lower house elections form the best starting point. In terms of party

nationalization, presidential elections in multiparty systems have an idio-

syncrat ic logic because of the importa nce of individual cand idacies, because

of party coalitions, and because there are greater, yet widely varying, incen-

tives for stra tegic voting. A part y could be q uite na tiona lized in presidentia l

elections because of the dra w ing pow er of a particular cand idate and/or

because of interparty coalitions, yet be weakly nationalized in lower house

elections (the Brazilian PSDB [Pa rtido d a Social D emocracia Brasileira] was

an example in 1994 and 1998). We use low er house ra ther than sena te elec-tions because severa l countries have a unicamera l legislature or a sena te tha t

is not popularly elected. Furthermore, many senates are renewed via partia l

renovation, and some districts do not vote in every senate election year.

Table 1 provides summary informa tion a bout the political institutions in

these 17 countries. The information reflects the current sta te of a ffa irs as of

the most recent lower house election included in our study, although with

few exceptions there was little change in these institutions during the time

period under analysis.

We used geogra phic units of roughly comparable political sta tus. For thesix federal republics, we utilized the state (Brazil, Mexico, United States,

Venezuela) or province (Argentina, Canada) as the geographic unit. 11 For

Bolivia, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Peru and


145


8/28

146

Table 1. Political institutions in 17 countries (as of the most rec

N o. of members

Federal or N ame of unit N o. of in the lower Elect ions

Country unitary of analysis units house included E

Argentina Federal Province 24 257 19832001 M

Bolivia Unitary D epartment 9 130 19851997 M

Brazil Federal State 27 513 19861998 M

Canada Federal Province/Territory 13 301 19802000 S

Chile Unitary Region 13 120 19892001 B

Costa Rica Unitary Province 7 57 19821998 M

Ecuador Unitary D epartment 21 82 19791996 M

El Salvador Unitary Department 14 84 19942000 MG uatemala Unitary D epartment 23 113 19901999 M

H onduras Unitary Department 18 128 19812001 M

Jamaica Unitary Parish 14 60 19801997 S

M exico Federal State 32 500 19942000 M

Nicaragua Unitary Department 17 90 19902001 M

Peru Unitary D epartment 26 180 19801990 M

United Sta tes Federal Sta te 50 435 19802000 S

Uruguay Unitary Department 19 99 19841999 M

Venezuela Federal State/Territory 24 207 19831998 M

Note: The federal district in Argentina , B raz il, M exico and Venezuela is included since residents of these d(unlike their counterparts in the United States). One US state employs Single-Member Majority Runoffrespectively, 12 (every 4 years), 22 and 20 a ddit iona l members are elected using a separat e ballot f rom a nRepresentation, C = Compensatory, NC = Non-Compensatory.


9/28

Uruguay the departments w ere our unit o f a nalysis, w hile for C hile, C osta

Rica and Jamaica we employed the administrative regions, provinces and

parishes, respectively.

Because of the strat egic behaviour by elites and voters tha t t akes place in

single-member plurality d istricts, use of these districts (for C anada , Jamaica

and the United States) as our unit of analysis would provide less valid and

reliable measures of party support than use of the larger territorial units.

Furthermore, employment of these single-member districts would result in

a much wider variance in the number of units per country in the analysis,

w hich given the nature of the G ini coeffi cient is not desirable.

Although the G ini coeffi cient is superior to the alternat ives, it might not

w ork w ell for comparisons of countries w ith a w ide range in the number of

geographic units. Fortunately, the sensitivity to the number of units is

modest given the relat ively limited ra nge in the number of units w e use in

this analysis, from a low of 7 (Costa Rica) to a high of 50 (United States),w ith a median of 19. C ombined w ith reporting PNSs at tw o decimal places,

the effect on our interpretat ion ca used b y these differences in the number of

units is modest. The use of these administrative units is less arbitrary and

suffers from less measurement error than the use of alternative measures

such as deciles or quintiles of the rank-ordered sub-national administrative

units.

In the tables on partyna tiona liza tion (Tables 4, 6 a nd 7), because of

spatial constraints, we limit the presentation to parties that received a

minimum of 5 percent of the national vote in the respective election. Incontrast, the data on par ty systemnationalization are based on all the

parties as well as independent candidates.

The Nationalization of Party Systems

The bot tom row of Table 2 gives the average PSNS for the 17 countries. The

level of nationalization varies markedly across party systems. The systems

of Argentina, Cana da , Peru and especially Bra zil and Ecuador a re much lessnat ionalized tha n tho se of the six most na tiona lized countries. The data thus

confi rm previous scholarly w ork that emphasized the importa nce of provin-

cial-level politics in Argentina, Brazil, Canada and Venezuela and of

regiona l-level politics in Ecuad or. Bolivia a nd G uatemala a lso ha ve w eakly

nationalized party systems.

Mexico, the United States and El Salvador are intermediate cases with

PSNSs far higher than those of Ecuador and Brazil but lower than those of

most unitary Latin American countries. Although the literature on the

nationalization of the United States party system has emphasized its feder-alized nature, according to our measure, during this period it ranked as an

intermediate (not low ) case of na tionalization amo ng our 17 cases.

The party systems of Chile, Costa Rica, Honduras, Jamaica, Nicaragua


147


10/28

148

Table 2. Party system nationalization in 17 countries, 19792001: PSNS, Elect

(in boldface)United El

Ecuador Br azil Per u A rgentina Canada G uatemala Venezuela Bolivia M ex ico States Salvador N i car agua

0.62 0.65 0.70 0.79 0.78 0.74 0.86 0.80 0.84 0.82 0.87 0.86

(1979) (1986) (1980) (1983) (1980) (1990) (1983) (1985) (1994) (1980) (1994) (1990)

0.54 0.51 0.79 0.81 0.83 0.74 0.85 0.77 0.79 0.85 0.83 0.85

(1984) (1990) (1985) (1985) (1984) (1995) (1988) (1989) (1997) (1982) (1997) (1996)

0.47 0.51 0.62 0.74 0.78 0.80 0.72 0.78 0.83 0.83 0.85 0.88

(1986) (1994) (1990) (1987) (1988) (1999) (1993) (1993) (2000) (1984) (2000) (2001)

0.63 0.61 0.76 0.62 0.61 0.72 0.83

(1988) (1998) (1989) (1993) (1998) (1997) (1986)

0.55 0.62 0.67 0.85(1990) (1991) (1997) (1988)

0.59 0.68 0.63 0.85

(1992) (1993) (2000) (1990)

0.58 0.79 0.87

(1994) (1995) (1992)

0.60 0.72 0.87

(1996) (1997) (1994)

0.73 0.86

(1999) (1996)

0.59 0.80

(2001) (1998)

0.82

(2000)

0.57 0.58 0.70 0.72 0.72 0.76 0.76 0.77 0.82 0.84 0.85 0.86


11/28


12/28

with respect to the peaks provided by the presidential years (1984, 1988,

1992, 1996). This finding indicates the na tiona lizing effect tha t presidential

elections can have on congressional elections.

Although we have not systematically collected data for other regions, the

norm in Western Europe is a high degree of nationalization (Caramani,

2000; Rose and Urw in, 1975). The Rose and Urw in indicato r w e consider

reliable (see below) indicated that the United States had a less nationalized

party system than any Western European country. In our population, by

contrast, the United States party system is more nationalized than that of

a majority of the other countries. Furthermore, some elections in develop-

ing countries such as India and Malawi have yielded less nationalized party

systems than have elections in any country in our population.

Low Nationalization Countries12

We now shift from party systems to part ies as the unit of ana lysis. Table 4

presents da ta on party nat ionalization in Ecuador, Brazil, Peru, Argentina,

Cana da , G uatemala, Venezuela and Bo livia.

Ecuadors major parties possess low average PNSs, with three among the

11 least nationalized major parties (45 total) in our population (Table 5).

Ecuadors smaller parties are even less nat iona lized, w ith severa l among the

least nationalized of our entire party population. These smaller parties

account for a substantial portion of the vote; in none of the eight electionsdid the parties winning 20 percent or more of the vote account for more

than half of the national vote.

Brazil is another outlier; the national party system is a pastiche of state

party systems. State-level politics is very important, and regional and state

differences in the party system have long been salient. With the partial

exception of the PM D B (Partido do M ovimento D emocrt ico Bra sileiro) in

1986, Brazils parties have won widely diverging vote shares across the 27

units of the federation. Three of the 11 least nationalized major parties are

Bra zilian (Table 5).Peru provides the largest inter-election PSNS change, from 0.79 in 1985

to 0.62 in 1990, as the traditional Peruvian party system disintegrated.

Perus na scent democra tic experience ended short ly thereaf ter, w hen Presi-

dent Alberto Fujimori staged a coup in 1992. In 1990 all parties possessed

relatively low PNSs, with Fujimoris C90 (Cambio 90) registering an

extremely low (especially for a governing party) 0.45.

Argentina is another case of low nationalization, but its pattern of party

federalization diverges from the Brazilian, Canadian, Ecuadoran and

Peruvian. For much of this period, two main parties, the PJ (PartidoJusticia lista) and UC R/ALIANZ A (Unin C vica Radical/Alianza ), both of

which have intermediate to high PNSs, coexisted with several powerful

provincial parties, which normally compete in only one province and hence


150


13/28

have extremely low PNSs. Although each provincial party accounts for

limited w eight in the PSNSs presented in Table 2, generally averaging

betw een 0.5 percent a nd 1.5 percent o f the nat iona l vote, cumulatively these

parties have won a meaningful share (between 5 percent and 8 percent) of

the national vote in most elections.

In the 1980s, the Canadian party system had a level of nationalization

only slightly lower than that of the United States. Since 1993, however,

Canada has approximated Brazils PSNSs. In the 1990s, two new forces

emerged a t the nationa l level and dramat ically low ered C ana da s PSNS. The

BQ (Bloc Quebecois) (PNS 0.08), which competes only in Quebec, is the

sole single-province party in the 17 countries ana lysed here tha t obta ined 5

percent or more of the national vote. The Reform Party (RP), which first

contested national elections in 1988 (garnering a meager 2 percent of the

vote), a lso has a very low PN S, reflecting its concentra tion in the West. The

RP drew many of its votes from the historic base of the PC (Progressive Con-servatives), w hich saw its support dissipate in the West, but remained com-

petitive in the Maritime provinces.

The remarkable changes in the Canadian party system in 1993 were

associated with the second largest single inter-election PSNS change (tied

with Ecuador) in the 68 inter-electoral periods covered in this study, from

0.78 in 1988 to 0.62 in 1993. In contrast to the 1980s, when two of the

countrys three major parties (PC and LP [Liberal Party]) were relatively

nationalized, in the 1990s, of the five major parties, only the LP remotely

approached a significant nationwide presence.Guatemalas two largest parties during this period, the PAN (Partido de

Avanza da Na ciona l) and FRG (Frente Repblicano G uata malteco), became

increasingly na tiona lized o ver t ime. This was especially the case for the FRG

w ith a PN S of 0.91 in 1999 (up from 0.72 in 1990). With one minor excep-

tion, the other Guatemalan parties had low levels of nationalization.

Until 1993, the Venezuelan party system ranked in the high nationaliza-

tion category. In the 1990s, however, the decline of the two parties (AD

[Accin Democrtica] and COPEI [Comit de Organizacin Poltica Elec-

toral Independiente]) that had dominated Venezuelan politics for theprevious quarter century, along with constitutional reforms that decentral-

ized pow er to the sta tes, resulted in a ma ssive PSNS decrease (the largest

aggregate shift, 0.25, in our population), pushing Venezuela into the low

nat ionalization ca tegory.

Most of Bolivias major parties possess intermediate levels of national-

ization. In contrast, CONDEPA (Conciencia de Patria) is a weakly nation-

alized pa rty; its support is heavily concentra ted in the Department of La Pa z.

CONDEPA helps drag the countrys party system from the intermediate to

low classifi cation. After C ana da s BQ, C O ND EPA possesses the low est PNSof the major part ies in our ana lysis (Table 5).


151


14/28

152

Table 4. Party nationalization (PNS) in the low nationalizatio

Country Elect ion year Pol i t ical part ies

Ecuador CFP PSC ID PLRE Other1a Other2b

1979 0.82 0.48 0.61 0.55 0.53 0.481984 0.46 0.59 0.77 0.34 0.47 0.57/0.1986 0.58 0.65 0.71 0.61 0.46 0.371988 0.60 0.68 0.831990 0.53 0.79 0.431992 0.74 0.67 0.38 0.591994 0.77 0.69 0.49 0.341996 0.79 0.50

Brazil PM DB PFL PT PSDB PPBc PTB1986 0.84 0.71 0.60 0.411990 0.63 0.59 0.64 0.53 0.47 0.371994 0.69 0.55 0.67 0.55 0.52/0.50 0.401998 0.74 0.68 0.69 0.70 0.64 0.53

Peru APRA APd IU CD C90 IS1980 0.71 0.821985 0.86 0.76 0.82 0.731990 0.75 0.69 0.61 0.45 0.69

Canada LP PC ND P BQ RP1980 0.82 0.84 0.721984 0.82 0.92 0.701988 0.75 0.91 0.671993 0.81 0.81 0.52 0.08 0.521997 0.86 0.72 0.73 0.08 0.54

2000 0.82 0.62 0.68 0.08 0.55

Argentina PJ UCR Othere ALIANZA1983 0.89 0.881985 0.87 0.91 0.511987 0.87 0.88 0.371989 0.88 0.86 0.52


15/28

153

Table 4. (continued)

Country Elect ion year Pol i t ical part ies

Argentina PJ UCR Othere ALIANZA(continued) 1991 0.88 0.75 0.24

1993 0.88 0.81 0.361995 0.84 0.77 0.691997 0.82 0.821999 0.81 0.37 0.872001 0.83 0.46 0.76

G uatemala PAN FRG f D CG g UCN M AS Otherh

1990 0.77 0.72 0.80 0.88 0.73 0.511995 0.86 0.86 0.76 0.621999 0.88 0.91 0.68

Venezuela AD COPEI M AS LCR CONVERG M VR1983 0.86 0.90 0.801988 0.85 0.92 0.66

1993 0.72 0.83 0.60 0.64 0.741998 0.76 0.62 0.58 0.70

Bolivia AD N M NR M IR i Otherj COND EPA UCS1985 0.88 0.84 0.72 0.771989 0.88 0.86 0.88 0.69 0.331993 0.80 0.93 0.66 0.42 0.831997 0.82 0.82 0.77 0.45 0.82

a 1979: PCE, 1984: FADI, 1986: FADI, 1990: PSE, 1992: PCE, 1994: PCE, 1996: MUPP-NP.b 1979: CID , 1984: FRA/PD , 1986: FRA, 1992: PUR , 1994: APRE.c PD S in 1986 and 1990, P P/PPR in 1994.d FREDEM O alliance in 1990.e

1985: PI, 1987: UCEDE, 1989: ADC, 1991: UCEDE, 1993: MODIN, 1995: FREPASO, 1999: AR, 200f Presented in alliance with the PID and FUN in 1990.g The DC G , UC N, and PSD presented in alliance in 1995.h 1990: MLN, 1995: FD NG , 1999: D IA-URN G .i Presented in alliance with the ADN in 1993.j 1985: MNR I, 1989: IU, 1993: M BL.


16/28

Intermediate Nationalization Countries

Despite formal federalism, the Mexican political system was highly central-

ized until the 1990s. It is thus unsurprising tha t M exico ha s a more nation-alized party system than Argentina, Brazil and Canada. In contrast to the

situation in the latter countries, Mexican state governments are compara-

tively bereft of resources.

Mexico is the country with the sharpest differences in nationalization

from one major party to the next. The most nationalized party, the PRI

(Pa rtido Revolucionario Instituciona l), is tied as the tenth most nat iona lized

of our major pa rties (Table 5). In contrast, the PRD (Pa rtido de la R evolu-

cin Democrtica) is the tenth least nationalized. Mexicos PSNS thus

conceals important cross-party differences (Table 6).Although a substant ial literature has correctly emphasized the federalized

character of the United States party system (Kawato, 1987; Mayhew,

1986), the United States party system is more nationalized than half of the


154

Table 5. Ra nking of the 11 most na tionalized a nd least nationa lized ma jor part ies

Average Average vot e

Rank ing Party PN S share (% ) Country

Top eleven PUSC 0.96 40 Costa Rica

PLN 0.95 45 Costa RicaARENA 0.95 39 El Salvador

CONCERTACIO N 0.94 53 Chile

PLH 0.94 48 H onduras

PNH 0.93 45 H onduras

PNP 0.93 54 Jamaica

JLP 0.93 45 Jamaica

PC 0.93 34 Uruguay

ALIANZ A 0.92 37 Chile

PRI 0.92 42 M exico

Bottom Eleven ID 0.70 14 Ecuador

PRD 0.69 20 M exico

ND P 0.67 14 Canada

PSC 0.65 18 Ecuador

PT 0.65 11 Brazil

PFL 0.63 15 Brazil

PSDB 0.59 14 Brazil

PRE 0.55 14 Ecuador

RP 0.54 21 Canada

CONDEPA 0.40 14 BoliviaBQ 0.08 12 Canada

Note: Includes only parties (45 total) that averaged 10% or more of the vote and competed inat least t hree elections during t he ana lysis period. For addit iona l informa tion, see Tab les 4,6 and 7.


17/28

systems in this population. But if we compare the PNSs of the Democratic

(D P) and Republican (RP) parties w ith those of the major pa rties in Co sta

Rica, Honduras and Jamaica, the United States parties are indeed quite

federalized.

The Salvadora n party system underw ent a striking tra nsformat ion during

this period. The largest party in the 1980s, the PDC (Partido Demcrata

Cristiano), suffered an electoral haemorrhage. By 1997 the FMLN (FrenteFara bundo M art para la Liberacin Na ciona l), a revolutionary group in the

1980s, had become El Salvadors second largest party, merely 2 percent

behind ARENA (Alianza Republicana Nacionalista). The PDC collapse

occurred very unevenly across the country: the party moved from a rela-

tively high PNS to a low one betw een 1994 and 1997. The FM LNs grow th

w as more even; its PNS rose only slightly, f rom 0.84 in 1994 to 0.88 in 2000.

High Nationalization Countries

The two parties that have dominated Nicaraguan politics since 1990, the

FSLN (Frente Sandinista de Liberacin Nacional) and PLC (Partido Liberal


155

Table 6. Party nationalization (PNS) in the intermediate nationalization countries

Election

Country year Poli t ical part ies

Mexico PRI PAN PRD

1994 0.94 0.79 0.671997 0.91 0.74 0.71

2000 0.91 0.84 0.69

United Sta tes DP RP

1980 0.82 0.83

1982 0.86 0.84

1984 0.84 0.83

1986 0.84 0.83

1988 0.87 0.85

1990 0.87 0.861992 0.90 0.90

1994 0.87 0.91

1996 0.88 0.90

1998 0.83 0.84

2000 0.83 0.85

El Salvador ARENA FM LN PDC PCN CD /C DUa

1994 0.96 0.84 0.87 0.78 0.47

1997 0.96 0.85 0.75 0.75

2000 0.94 0.88 0.71 0.72 0.68

a CD in 1994 and CD U in 2000.


18/28

Constitucionalista), have relatively high PNSs. As is the case in the other

Central American countries, however, the Nicaraguan minor parties are

much less nationalized than the dominant parties.In response to the tw o-member district legislat ive electora l system

employed in Chile, the major Chilean parties formed two large and rela-

tively stable national coalitions. The Concertacin (CONCERT) includes


156

Table 7. Party nationalization (PNS) in the high nationalization countries

Election

Country Year Pol i t ical part ies

Nicaragua FSLN PLC a

1990 0.88 0.871996 0.87 0.87

2001 0.89 0.89

Chile CONCERT ALIANZ A PAIS PC

1989 0.93 0.96 0.47

1993 0.94 0.94 0.68

1997 0.94 0.86 0.76

2001 0.94 0.92 0.67

Uruguay PC PN FA NEb

1984 0.94 0.92 0.721989 0.91 0.92 0.71 0.75

1994 0.92 0.90 0.79 0.72

1999 0.94 0.87 0.85 0.83

Costa Rica PLN PUSC PU FD

1982 0.96 0.95 0.73

1986 0.95 0.97

1990 0.94 0.97

1994 0.96 0.97 0.63

1998 0.94 0.94 0.81

H onduras PLH PNH

1981 0.93 0.91

1985 0.94 0.93

1989 0.95 0.96

1993 0.93 0.92

1997 0.95 0.93

2001 0.94 0.95

Jamaica PNP JLP

1980 0.93 0.95

1989 0.95 0.94

1993 0.94 0.92

1997 0.96 0.93

a UNO alliance in 1990.b PGP in 1989.


19/28

four centre and centreleft parties. The Alianza por Chile (ALIANZA), as

it is currently named, includes two centreright parties along with a sub-

stantial number of independents. We therefore used coalition (rather than

party) results. With the partial exception of the ALIANZA in 1997, these

tw o major coa litions have consistently ha d very high PNSs.

Uruguays two traditional parties, the PC (Partido Colorado) and PN

(Pa rtido Na ciona l), ha ve consistently had relatively even vote patterns across

the countrys 19 departments. In contra st, the leftist FA (Frente Amplio) had

a markedly less nationalized pattern until 1994. It fared much better in

Montevideo than in the interior. The FAs ability to win more votes across a

broader range of departments was a key factor in its electoral growth from

21 percent of the national vote in both 1984 and 1989 to 31 percent in 1994

and 40 percent in 1999. In contrast to what has occurred in most cases, the

PNs electoral decline after 1989 (39 percent), to 31 percent of the vote in

1994 and 22 percent in 1999, has yet to produce sharply low er PN Ss.Costa Rica, Honduras and Jamaica have the most nationalized party

systems. The respective two dominant parties in these countries are highly

nationalized, garnering relatively equal shares of the vote in all districts.

They represent six of our nine most na tiona lized pa rties (Table 5).

Previous Measures of Nationalization

Claggett et al. (1984), Schattschneider (1960), Sundquist (1973) andKawato (1987) devised good measures of party nationalization for the

United Stat es, but they a re appropriat e only for tw o-party systems. C laggett

et a l. measured the sta ndard deviation (SD ) of the D emocrat ic Partys (D P)

share of the vote at the county level. SDs perform poorly for comparing

parties that win very different mean shares of the vote because they are not

comparable across parties with different vote shares. They tend to attribute

higher values to large parties and lower values to small ones because there

is likely to be less variance in absolute terms w ith a low er share of the vote.

Because most democracies (those with two-party systems are rare excep-tions) have relevant parties that win widely different vote shares, the SD is

a poor option for comparative research.

Schattschneider (1960: 7896) measured the DPs deviation from 50

percent, expressed as an absolute value, in every sta te and summed this value

for all states. It is possible to follow this procedure for other countries by

taking the deviation (as an absolute value) from a given partys mean share

of the vote in all states or provinces; this is one of the measures that Rose

and Urwin (1975) used. But in a multiparty system this measure greatly

overstates the nationalization of small parties compared to large ones.13

Sundq uist (1973: 3327) and Kaw ato (1987) devised measures of na tiona l-

ization based on the DP vote share, but focusing on one party is suitable

only in two-party systems.


157


20/28

Although the measures of nationalization used for the United States do

not work for most other countries, the debate about nationalization in the

United Sta tes usefully highlighted the importance of the subject. C uriously,

in comparative politics little has been written on this subject; the works by

Rose and Urwin (1975) and Caramani (2000) are the main contributions.

They both devised three different measures of na tiona lization. Bo th used the

SD, w hich, a s noted, performs very poorly for comparing parties of uneq ual

sizes. Rose and Urwin also used an Index of Variation (IV), which repro-

duces the modification of Schattschneider discussed in the previous para-

graph. As noted, it ha s serious short comings for compa ra tive research. Rose

and Urw ins Index o f C umulat ive Regiona l Inequality (ICRI), also used by

Caramani, is their only solid indicator. It is constructed by calculating the

percentage of a partys national vote won in a given territorial unit, sub-

tracting that territorial units share of the nationa l vote, taking the absolute

value, summing these absolute values for a ll units, and d ividing by tw o. Thismeasure performs well but has the disadvantage of being less well known

and used tha n the G ini coeffi cient. Their three measures perform very differ-

ently; the United States ranks as the least nationalized party system among

20 countries on 2 of them but as the most nationalized on the third. Such

profound discrepancies reflect the serious problems with the SD and IV as

measures of party nationalization. In addition to the SD and ICRI,

Caramani employed the CV, the shortcomings of which were discussed

earlier.

Conclusion

The extent of party and party system nationalization is an important topic

that has been neglected by the scholarly literature, except for the United

States. This is in part because the mainstream theoretical literature on

parties and party systems has focused on the advanced industrial democra-

cies, w hich w ith a few exceptions have fairly nat ionalized pa rty systems. But

when we turn our attention to Africa, Asia and Latin America, a muchgreater proportion of the party systems are w eakly na tiona lized. H ence, the

importance of analysing variance in nationalization is greater for these

regions than for the advanced industrial democracies. It is impossible to

understand many party systems in the world, including those of such large

and important countries as Argentina, Brazil, Canada, India and Russia

w ithout pay ing attention to the w idely divergent vote shares that parties w in

in different sta tes or provinces.

We have proposed a means of measuring nationalization that travels well

across part ies, countries and time. This art icle is the fi rst to provide empiri-cal informa tion about pa rty system nat ionalization outside of the ad vanced

industrial democracies. Our work makes clear that party system national-

ization varies markedly across countries.


158


21/28

Issues of measurement provide a key underpinning for exa mining import-

ant substantive issues in the social sciences. It is worth constructing new

measures if: (a) the subject of measurement is important enough to merit

sustained substantive discussion; (b) it is measurable; and (c) social scien-

tists gain a useful tool by being able to measure the phenomenon precisely.

We believe that all three conditions obtain. Our measures enable us to

compare across party systems; measure stability and change in nationaliza-

tion within a given party system; compare across parties in one or more

countries; analyse the relationship between electoral growth or demise and

party nationalization; and assess the relationship between realignment or

dealignment and nationalization. They should be useful for scholars

studying various aspects of the party system and its relationship to factors

such as voters orientations, legislative careers, executivelegislative

rela tions, public policy a nd democratic sta bility in multi-ethnic societies.

Our measures could also be useful for tracking and comparing differentpolitical processes. Consider the transitions from hegemonic party systems

to democratic politics in Brazil (197485) and Mexico (19882000). These

transitions were exceptional in the degree to which elections dictated the

transitions pace. In both countries the pro-regime parties held remarkably

similar positions: each had barely lost its majority in the lower house but

retained a solid majority in the senate. But beyond this important similarity

the nationalization data reflect a difference. In Brazil, the wealthiest states

tilted strongly to the opposition by 1974 and became overw helmingly domi-

nated by it in 1982, w ith the PNS of the governing party, the PD S (PartidoD emocrt ico Social), fa lling to 0.41 in 1986. In M exico, in sharp contra st,

the PR I remained competitive in a ll 32 sta tes. The PRIs PNS remained high

(0.91) even in 2000. The PNS calls attention to these differences between

otherwise similar cases and captures their magnitude.

Although for reasons of space a statistical analysis of the sources of

variance in party system nationalization must be left to a future study, three

potential explanatory factors leap out in this set of countries. First, the

federa l countries tend to have low er nationa lization scores, reflecting greater

variance in parties electoral performances across the sub-national units.The correlation between a binary federalunitary variable and the PSNS is

0.36. Federalism gives incentives for parties to orga nize and compete at the

stat e level and tends to foster more differences tha n unita ry systems in inter-

state patterns of electoral competition.

Second, the more fragmented party systems tend to be the least national-

ized. The correlation between the PSNS and the effective number of elec-

toral parties is a remarkable 0.87, notwithstanding the fact that the two

issues are obviously conceptually and operationally discrete. 14 In this

universe of countries, the larger parties tend to be more nationalized andthe smaller parties less nationalized. In fragmented party systems, small

parties win most of the vote, hence the powerful tendency towards low

PSNSs. The legislatures in the low nationalization countries are not only


159


22/28

generally fragmented along partisan lines; in addition, these different

partisan forces tend to have distinct geographic constituencies, adding one

more potential complication to coordination efforts in these assemblies.

Third, t here is a notew orthy inverse relat ionship betw een electora l vola til-

ity and party nationalization. However, the driving force behind this

relationship is most likely the previously noted correlation between frag-

mentation and nationalization.

Party nationalization varies even more dramatically than party system

nat ionalization. Individual pa rties range from almost perfectly na tiona lized

(the Costa R ican PUSC [Pa rtido Unida d Social Cristiana] having three times

registered a nationalization score of 0.97) to perfectly provincialized (the

Canadian BQ has won 100 percent of its votes in Quebec). A low level of

nationa lization does not necessarily result from a delibera te effort of parties

to cast state or regional appeals. In these countries, a deliberate focus on

state or regional appeals characterizes very few parties.One finding stands out in the variance across individual parties: small

parties tend to have substa ntially low er na tiona lizat ion scores. Large part ies

(those that win over 30 percent of the vote) almost always compete, and

have a chance to win seats, in virtually all of the sub-national administra-

tive units. It w ould be unusual for a party to w in 30 percent or more of the

vote if it were a minor party in a substantial number of provinces. Con-

versely, most small part ies in our population w on w ildly divergent shares of

the vote in different provinces.

This difference likely stems in part from the distinct incentives for nation-alization facing large and small parties in presidential democracies. Large

parties can realistically compete for the countrys most important electoral

prize, the presidency. Because the president is elected by a national vote, the

party cannot afford to lack a presence in any administrative unit and thus

invests resources in the development and maintenance of a nationw ide party

structure. This party structure in turn presents candidates in the legislative

elections (especially when the presidential and legislative elections are con-

current) regardless of their chances of success. In contrast, sma ll part ies lack

a realistic chance of winning the presidency and thus have a weaker incen-tive to develop and maintain a party structure, present candidates and

provide these candidates with campaign resources in districts where they

have little chance of success.

A related pattern that emerged consistently is the relationship between

party growth or decline and nationalization. Almost invariably, significant

electora l grow th w as a ccompanied by a PN S increase and significant decline

by a decrease.

More remains to be done on these issues, but we believe that our

measures, the PNS and the PSNS, and our empirical information on 17countries, pave the way for future research.


160


23/28


24/28

162

State represents C

what share of Column 2 sh

all states (i.e., divided by sum Cumulat ive st

Partys share of 1 divided by of column 2 in share of co

State vote in state no. of states) al l states states Y i X

(1) (2) (3) (4) (5) (6

A 0.00 0.25 0.00 0.25 0.

B 0.10 0.25 0.125 0.50 0.

C 0.30 0.25 0.375 0.75 0.

D 0.40 0.25 0.50 1.00 1.

Sum = 0.80


25/28

Notes

We are grateful for comments provided by Ana Mara Bejarano, Daniel Brinks,

Pradeep Chhibber, Matthew Cleary, Michael Coppedge, Jim Granato, Fran

Hagopian, Wonjae Hwang, Herbert Kitschelt, William Reed, Aseema Sinha,

Ashutosh Varshney, t w o ano nymous review ers, and panel participants a t the 2000Meeting of the M idw est Political Science Association and the 2001 C ongress of t he

Latin American Studies Association. Support for this research was provided by the

National Science Foundation (SES 9911140).

1 Brady (1985), C laggett et a l. (1984), Ka w at o (1987), Schatt schneider (1960),

Stokes (1965, 1967) and Sundquist (1973) measured the nationalization of

parties or electoral trends in the United States. Their contributions are valuable,

but t heir measures do not w ork w ell except in t w o-party systems.

2 Although scant comparative work exists on party system nationaliza tion, some

scholars have addressed related issues. Observing that electoral systems createincentives no t a t the na t iona l level, but ra ther a t the dist rict level, Cox (1997)

analysed the wa ys in which district-level polit ical competit ion leads to varying

party system patterns at the national level (see also Chhibber and Kollman, 1998).

Our endeavour is rela ted . Although we do not a t tempt to expla in howdistrict-

level competit ion is linked to the national party system, differences in nat ional-

ization reflect how closely results from one district mirror those of others.

3 Baw n et al. (1999) used the term electora l cohesiveness rather tha n na tiona l-

ization. They defined electoral cohesiveness as the extent to which the electoral

fates of incumbent candidates of the same party are tied together (p. 300).

Because their measure is based partly on an incumbents previous margin ofvictory, it w orks only in democracies w ith persona l voting.

4 This is the asymptotic result. In tho se ana lyses w here the G ini coeffi cient is most

commonly employed (i.e. small sample analyses) the limiting result is not

obt ained. In our a nalysis this small sample property is most nota ble when a party

wins 100 percent of its vote in a single electoral district.

5 For additiona l informa tion on calculat ing the G ini coefficient, see the Appendix.

6 As a diagnostic, w e calculated tw o a lternative measures of nat ionalization: a

rat io measure (RM ) and t he coefficient o f va riation (CV). The RM compares the

percentage of the vote won by a party in the top quartile of states to that won

in the bottom quartile. While this measure is easier to interpret, it is technicallyinferior because it arbitrarily selects two data points and ignores the rest. The

RM provided results similar t o the G ini coeffi cient, but it did so in a less

consistent ma nner. The CV w eights the sta nda rd deviation through t he mean. It

provided results substant ively indistinct from those obt ained using the G ini

coefficient. But the Gini coefficient has two advantages over the CV. It is better

known and equally good, and it ranges from 0 to 1. The CV has the disadvan-

ta ge of being less easily interpretable than the G ini coeffi cient because its values

have no intrinsic meaning (i.e. no upper limit). For example, w hereas the G ini

coefficient ranges from 0 to 1, the mean CV for 17 Western European countries

from 1918 to the 1990s ranged from 0.52 to 3.11 (Cara mani, 2000).

7 PSN S= PN Si* Vi), where Viis the vote share of the ith party.

8 This is a pa rticular problem for econometric ana lysis that employs the effective


163


26/28

number of parties as a dependent variable and average district magnitude as an

explana tory variab le and poo ls countries/elections w ith varying levels of pa rty

nationalization.

9 Also relevant here are factors such as the percenta ge of a party s nat ional-level

vote accounted for by a province and the percenta ge of a pa rtys legislat ors w ho

come from a province.10 In the 1960s, N igeria exemplifi ed the perils of a party system w ith highly region-

alized parties that simultaneously arose from, and reinforced, ethnic and

religious identities. Ra ther tha n t rying to build bridges across ethnic, t erritorial

and religious lines, as w ould ha ve been necessary for parties that aspired t o ha ve

a nat ionalized electora l base, party leaders cultivated their ow n ethnic and t erri-

toria l ba ses to t he exclusion of others; they inflamed ethnic tensions rather tha n

seeking accommodation. The resulting conflicts contributed to the breakdown

of democracy in Nigeria in 1966 (D iamo nd, 1988) and to a lesser degree in 1983.

11 Na tiona l territories w here citizens vote in na tiona l legislative elections are

referred to a s provinces/sta tes.12 In Tab les 4, 6 and 7 w e group the 17 countries into cases of low, intermediat e

and high nationalization based on the PSNSs. These groupings are purely

inductive. With a different set of cases, some regrouping would result.

13 Assume a country ha s 20 states. If Pa rty A (PA) w ins a mean vote share of 50

percent, the highest possible devia tion w ould be 50 percent 20 = 1000 percent.

This would result if PA w on 0 percent o f the vote in some sta tes and 100 percent

in the others. If Party B (PB) wins a mean share of 5 percent of the vote, the

highest possible deviation would be (95 percent 1) + (5 percent 19) = 190

percent. This would result if PB won 100 percent in one state and 0 percent in

19. PA w ould have the same nationa lizat ion score as PBs theoretical maximum(based on winning 100 percent of the vote in one state and 0 percent in 19) if

PA won 40.5 percent of the vote in 10 states and 59.5 percent in the other 10.

Yet PA would be obviously and profoundly more nationalized than PB. In fact,

it would be impossible for PB to be less nationalized, a fact that is captured by

the Gini coefficient.

14 This correlat ion w ould most likely be lower in Western Europe, where despite

considerable variance in the effective number o f parties variance in na tiona liza-

tion is low er than in the Americas.

15 If the descending vote shares are used, then the G ini coefficent is the absolute

value of t his process of subtra ction.

References

Ames, Barry (2001) The D eadlock of D emocracy in Brazil. Ann Arbor, MI:

University of Michigan Press.

Baw n, Kat hleen, G ary W. C ox and Frances Rosenbluth (1999) M easuring the Ties

that Bind: Electoral Cohesiveness in Four Democracies, in Bernard Grofman et

al. (eds) Elections in Japan under the Single N on-T ransferable Vote. Ann Arbor,

M I: University of M ichigan Press.Brady, David W. (1985) A Reeva lua t ion o f Rea lignments in American Polit ics:

Evidence from the House of Representatives, American Politi cal Science Review

79: 2849.


164


27/28

Cara mani, D aniele (2000)Elections in Western Europe since 1815: Electoral Resul ts

by Consti tuencies. Oxford: Macmillan.

Chhibber, P radeep a nd Ken W. Kollman (1998) Party Aggregat ion a nd the Number

of Parties in India and the United States, American Pol it ical Science Review92:

32942.

Claggett, William, William Flanigan and Nancy Zingale (1984) Nationalization ofthe American Electorate, American Poli ti cal Science Review78: 7791.

Cox, G ary W. (1997) M aking Votes Count: Str ategic Coor dination in the Wor lds

Electoral Systems. C ambridge: C ambridge University P ress.

Creedy, John (1998) The D ynamics of I nequality and Povert y: Comparing Income

D istributions. N orthhampton, M A: E. Elgar.

Diamond, Larry (1988) Class, Ethnicit y, and D emocracy in N igeri a: The Failur e of

the Fir st Republ ic. Syracuse: Syracuse University Press.

Gibson, Edward L. and Ernesto Calvo (2000) Federalism and Low-Maintenance

C onstituencies: Territoria l Dimensions of Econo mic Reform in Argentina , Studies

in Comparative International D evelopment35: 3255.H a rmel, R obert a nd La rs Sv sa nd (1993) Pa rty Lea dership a nd Pa rty Insti-

tut ionalisa t ion: Three Phases of Development , West European Po li t ics16:

6788.

Huntington, Samuel P. (1991) The Third Wave: D emocratization in t he L ate

Twenti eth Centur y. Norman, OK: University of Oklahoma Press.

Janda , Kenneth (1980) Polit ical Part ies: A Cross-N ational Survey. New York: Free

Press.

Jenkins, Stephen P. (2001) ineqdec0.ado. STATA Command. University of Essex.

Kawato, Sadafumi (1987) Nationalization and Partisan Realignment in Congres-

sional Elections, American Pol it ical Science Review81: 123550.Knutsen, O ddbjrn (1998) The Strength of the Pa rtisan C omponent of LeftRight

Identity: A Comparative Longitudinal Study of LeftRight Party Polarization in

Eight West European Countries, Part y Politi cs4: 531.

Lijphart, Arend (1994) Electoral Systems and Party Systems: A Study of Twenty-

Seven Democracies, 19451990. New York: Oxford University Press.

Lijphart, Arend (1977) D emocracy in Plural Societi es: A Comparative Explor ation.

New H aven, CT: Yale University Press.

Mainw aring, Scott P. (1999) Rethi nk ing Part y Systems in t he Thir d Wave of D emo-

cratization: The Case of Brazil. Stanford, CA: Stanford University Press.

Mainwaring, Scott and Timothy R. Scully (1995) Introduction: Party Systems inLat in America, in Scott M ainw aring a nd Timothy R . Scully (eds)Building D emo-

cratic I nsti tut ions: Part y Systems in L atin America. Stanford, CA: Stanford

University Press.

Mayhew, David R. (1986) Placing Part ies in American Poli ti cs: O rganizati on,

Electoral Sett ings, and Government A ctivi ty i n the Twenti eth Centu ry. Princeton,

NJ: Princeton University Press.

Reynolds, Andrew (1999) Electoral Systems and Democrati zati on in Sout hern

Africa. Oxford: Oxford University Press.

Rose, R ichard a nd D erek W. Urw in (1975) Regional D ifferentiat ion and Political

Unity in Western Nations, Sage Series Number 06007. Beverly Hills, CA: Sage.Samuels, David J. (Forthcoming) Careerism and its Consequences: Federalism,

Elections, and Poli cy Making in Br azil. C ambridge: C ambridge University P ress.

Sani, Giacomo and Giovanni Sartori (1983) Polarization, Fragmentation, and


165


28/28

Competition in Western Democracies, in Hans Daalder and Peter Mair (eds)

Western Eur opean Party Systems. Beverly Hills, CA: Sage.

Sartori, G iovanni (1976)Part ies and Part y Systems: A Framework for Analysis. New

York: Cambridge University Press.

Schattschneider, E. E. (1960) The Semi-Sovereign People: A Realists View of

D emocracy in Ameri ca. N ew York: H olt, R inehart and Winston.Shryock, H enry S., Ja cob S. Siegel and Associates (1976)The M ethods and M ateri als

of D emography. New York: Academic Press.

Stepan, Alfred C . (2001) Tow ard a New Compara tive Politics of Federalism,

(Multi)Nationalism, and Democracy: Beyond Rikerian Federalism, in Alfred

Stepan, Arguing Comparative Politi cs. Oxford: Oxford University Press.

Stokes, Donald E. (1965) A Variance Components Model of Political Effects, in

John M . C launch (ed.) M athematical A ppl ications in Polit ical Science. Dallas, TX:

Arnold Foundation.

Stokes, Donald E. (1967) Parties and the Nationalization of Electoral Forces, in

William Nisbet Chambers and Walter Dean Burnham (eds) The Ameri can Part ySystems: Stages of Pol i ti cal D evelopment. New York: Oxford University Press.

Sundquist, James L. (1973) D ynamics of the Part y System: A li gnment and Reali gn-

ment of Poli ti cal Part ies in the Unit ed States. Washington, D .C .: Bro okings Insti-

tution.

Taa gepera, Rein and M at thew Soberg Shugart (1989) Seats and Votes: T he Ef fects

and D eterminants of E lecto ral Systems. New H aven, CT: Yale University Press.

MARK P. JONES is Associate Professor in the Department of Political Science at

Michigan State University. His research focuses on the manner in which electorallaws and other institutions influence party systems, elite and mass political

behaviour, and representation. His recent publications have appeared in American

Journal of Political Science, Comparative Political Studiesand the Journal of

Politics. He is currently conducting an NSF funded study (with David Samuels) of

split-ticket vot ing in presidentia l democra cies.

ADD RESS: D epart ment of Political Science, M ichigan State University, East La nsing,

Michigan 48824-1032, USA. [email: [email protected]]

SCO TT MAINWARING is D irector of the Kellogg Institute for Internat iona l Studies

and Eugene Conley Professor of G overnment a t the University o f No tre Da me. H is

latest books are Rethi nk ing Part y Systems in t he Thir d Wave of D emocrati zati on:

The Case of Brazil (Stanford University Press, 1999); Presidentialism and

D emocracy in L atin Ameri ca(Cambridge University Press, co-edited, 1997); and

Bui ldi ng D emocrati c I nsti tut ions: Part y Systems in L atin Ameri ca(Stanford

University Press, co-edited, 1995).

ADDRESS: Kellogg Institute, University of Notre Dame, Notre Dame, Indiana

46556-5677, USA. [email: [email protected]]

Paper submitted 10 August 2001; accepted for publication 25 January 2002.


166

Jones and Mainwaring - The Nationalization of Parties and Party Systems

Documents

Transcript of Jones and Mainwaring - The Nationalization of Parties and Party Systems