|
The Thirteenth Election of India's Lok Sabha (House
of the People)
[CONTENTS]
In
mid-April 1999, the coalition government led by Prime Minister Atal Behari Vajpayee of the Bharatiya Janata Party
(BJP) lost a vote of confidence in the Lok Sabha, the lower house of Parliament, by just one vote.1
The BJP-led government fell because one of its alliance partners, the All India Anna Dravida Munnetra Kazhagam
(ADMK), withdrew the support of its 18-member Lok Sabha delegation. The ADMK is led by Ms. Jayalalitha Jayaram
(referred to simply as "Jayalalitha").2 Although Jayalalitha provided a few reasons for her threatened withdrawal
of support, most observers believe her motive was to avoid standing trial in a series of corruption cases, which
her alliance with the BJP had not prevented; she hoped that a Congress-led government would be of more help.
|
PARLIAMENT AND PRESIDENT
Parliament consists of two houses: the "lower" House of the People
(Lok Sabha), and the "upper" House of the States Rajya Sabha). The latter has members elected for fixed
six-year terms (with one-third elected every two years) by an electoral college consisting of the legislators of
a particular state, through a proportional representation system that almost invariably results in a state delegation
that reflects the balance of party power in that state legislature. The most important power the Rajya Sabha has
is the requirement that it must consent to constitutional amendments. The possibility that any single party will
in the future have a two-thirds majority in both houses is very, very small. (Even after the overwhelming defeat
of Mrs. Gandhi in the post-Emergency 1997 election, the Janata government was unable to pass constitutional amendments
without the consent of Mrs. Gandhi's Congress Party.)
The President of India is chosen for a five-year term by the members of
Parliament, casting half the total vote, with the other half cast by the members of the state legislatures. The
president must in almost all instances act on the instructions of the Council of Ministers (the cabinet), but,
in a power of increasing importance, has flexibility in choosing someone to form a new government, and in asking
that a prime minister prove his or her majority with a vote of confidence in the Lok Sabha. Constitutional precedents
are evolving, as in the present case where the powers of the "caretaker" government, and the role of
the president in this period, are a matter of dispute. K. R. Narayanan, a distinguished career officer in India's
foreign service, served as ambassador in Washington, and as a member of the Lok Sabha elected on a Congress ticket,
before he was elected first as vice president, and then president. He is India's first ex-untouchable, or "Dalit,"
to occupy that position.
|
A week later, having made a confident attempt to put together
a majority coalition, Sonia Gandhi of the Congress Party reported her failure. At the direction of the President,
K. R. Narayanan, the cabinet requested the dissolution of the Lok Sabha and the calling of fresh elections. Those elections will be held, with five separate polling days separated
roughly by a week, beginning on September 5, 1999. Counting of ballots will begin on October 6, and the (almost
complete) results will be known within 48 hours. In addition to the Lok Sabha election, there will be simultaneous
elections to the state assemblies of five states: Andhra Pradesh, Maharashtra, Karnataka, Arunachal Pradesh, and
Sikkim.
This election, the thirteenth since independence and the
third in three years, will once again be a massive exercise of political democracy.
Some 60 percent of the more than 600 million eligible voters will turn out to vote. In 1998 there were 176 parties
registered to contest 543 seats in 773,000 polling stations; 4,750 candidates stood, but on average there were
only 2.3 serious candidates per seat. Like every national and virtually every state election since independence,
this election will undoubtedly be "free and fair" by the most stringent definition. There is a mechanism
for having a repolling in the handful of constituencies in which there will be gross violations of procedure ("booth
capturing" and the like). Indian elections are becoming more peaceful: there were "only" 65 deaths
due to poll violence in 1998, compared to 213 in 1996.
This will be a "routine" election, not only because the procedures and campaign rituals are so familiar
to the Indian electorate, but also because there are no major issues or personalities that will produce anything
more than a marginal shift in the political pattern. As has been true of the four elections held since 1984, no
single party will win a majority of the seats. The next government, like those elected in 1989, 1996, and 1998,
will be a coalition.
That coalition–and the alliances of parties opposing
it–will emerge mainly from the dynamics of state-level politics.
As Yogendra Yadav, one of the most astute scholars of Indian elections, puts it, "the state [is now] the effective
unit of electoral choice. . . . Political loyalties, opinions and even social identities are now chosen at the
level of the state."3 Voters will cast their ballots mainly on the basis of their feelings
about the party coalitions' policies and ability to lead the country, and on the basis of what the first major
opinion poll of this election labeled "candidate image."4 Although
BJP activists are attempting to make this into a "Atal Behari Vajpayee versus Sonia Gandhi" "presidential"
contest, on the model of the four national elections held from 1971 to 1984, it is not at all clear that voters
will oblige them.
|
THE ELECTION COMMISSION
The conduct of elections is under the control of the Election Commission,
which has three members and a staff of civil servants, and can draft virtually unlimited numbers of government
officials for election duty. It is also empowered to deploy large numbers of security forces to ensure a peaceful
poll; the system of multiple polling days is due to the need to move sufficient police and paramilitaries to each
constituency to help prevent violence and vote fraud. As a legally autonomous body, it has historically functioned
without political interference. The commission determines the precise dates of polling and enforces the law on
campaign finance and practices. It also has promulgated a nonstatutory "code of conduct" that contains
rules for candidates, parties, and for the governments in power in Delhi and the states. These rules are designed
to neutralize the advantage incumbent governments have (to announce new programs, transfer bureaucrats, or cut
taxes, for example), but the Election Commission cannot formally penalize those who ignore it, as many do. The
commission's public declaration of "code of conduct" violations does carry weight, however.
|
Every Indian election reminds us forcibly that a deep-rooted,
strong democracy is possible in what at first glance seems like unfertile ground–enormous numbers of poor people,
40 percent illiterate, deeply divided by language and religion, with significant inequality derived from social
and religious norms. The gradual percolation of power downward toward the grassroots is also impressive. 5
Some relevant public opinion survey data are now available. 6 In answer to the question "Do you think your vote has an effect
on how things are run in this country, or do you think your vote makes no difference?" 59 percent of the respondents
in 1996 said their vote "has an effect" (up from 49 percent in 1971), and 21 percent said it makes no
difference. To the question of whether government would be run better "without parties or assemblies and elections"
the response was: 11 percent agreed, and 69 percent disagreed. All of this confirms the judgment of observers in
election after election that the commitment to democratic functioning in India has increased and is at a very high
level.7
The BJP government, and two other governments in the
last three years, collapsed due to the difficulties of keeping coalitions of diverse, not to mention venal and
opportunistic, politicians together. But we need not worry that "India" itself, or its political system,
is going to collapse in the near or medium-term future. It is important–as Indian voters more than anyone else
have constantly said–to move toward "stability" in government, so that certain policies can be reasonably
thought through and then given time to work. Indians
do not see the polity disintegrating or collapsing into chaos, even though they complain bitterly about increasing
corruption and "criminalization" of politics.
Indians are not as worried as some in the United States and
elsewhere seem to be about "stability" insuring a good climate for foreign investment, because they perceive
India's economic performance in the past fifteen years to be respectable, though less than expected. Foreign investors,
like Indian ones, will probably have to learn to live with predicting instability, and hedge their investments
accordingly.
We need to pay attention to this election even though
it seems likely to produce a loose coalition government, whose policies and programs will be not very different
from those of the last five years, under governments led by the Congress, the "United Front," and the
BJP. India's democracy is in and of itself of great importance–India has just reached the billion-person mark,
one-sixth of the world's population–but it also can serve as a guide, as similar countries democratize. The exciting
thing about this election is that it will make only a marginal difference to India. It is cause for celebration
that India's democracy is so entrenched that elections are routine, and that the domestic and foreign policies
of the various parties have begun to overlap to such an extent that wild swings are no longer likely.
There are underlying deep trends that we need to understand:
the increase in democratic participation, as more and more of the "backward" groups take to politics;
the emergence of the states as stable political systems, at the moment pivotal in national politics as well; and
the continuing reweaving of India's social fabric, with caste and religious communities playing very different
roles now than they once did. The currents of change may be slow-moving but they are powerful, and their influence
will be seen at the surface level that is a national election.
"Who will win?" is thus of less significance than
we tend to think. Judging the likelihood of one party alliance or the other getting a majority in parliament depends
on understanding the particular issues of this election. But that in turn depends on reviewing what we know about
why Indians vote the way they do. Is it because of their social background, particularly their caste, religious,
or social class identity? How important are issues such as who the people are being projected as candidates for
prime minister, or economic issues such as inflation? Why does the so-called anti-incumbency factor seem to play
so large a role?
The Pivotal Factor: State-Level Parties and Their Alliances
State-level party politics will determine the outcome of this election. (It
must be noted that the major states of India would be large countries in other parts of the world: the largest
fifteen states range in population from 20 million to 170 million people. Many of the states have distinct languages
and cultures.)
Thirty years ago state politics was a picture of coalition
governments of convenience collapsing with alarming rapidity, forcing frequent elections, with parties emerging
and dying and splitting and merging. Now states display a remarkable pattern of stability. The trend has been for more and more state governments to complete their full
five-year terms. Moreover, the states have weathered not only the era of their own chaotic politics but also the
efforts of Mrs. Gandhi, particularly, to centralize power in Delhi; India has moved toward a more genuine federal
system. Chief ministers now actively compete with each other–and with Delhi–for investment from India and abroad.
Many states now have two-party (or stable two-coalition)
or three-party systems. Oliver Heath has analyzed
data on seats won and vote share by parties over all elections. He has applied the political science concept of
"effective parties," which allows us to disregard the literally hundreds of parties that have essentially
no vote share and win no seats, to measure how many parties are truly "effective." Looking at data for
the national, state, and constituency level, he has shown that at the constituency level (and the state data run
parallel) the number of "effective parties" has averaged 2.7 (also the figure for the 1998 election).
At the state and constituency level, India is, as he puts it, "a collection of two-party systems." At
the national level, however, the number of effective parties has increased.8
|
THE FEDERAL POLITY
India is composed of 25 states and the "National Capital Territory
of Delhi" (which is a state in all but name), plus 6 "Union Territories" (small, central-government-administered
places, each with one Lok Sabha member). The states can be conveniently grouped by size of Lok Sabha delegation,
which is proportionate to each state's population, into seven "heavyweight" states, with 350 of the 543
seats (Andhra Pradesh, Bihar, Madhya Pradesh, Maharashtra, Tamil Nadu, Uttar Pradesh, and West Bengal); eight "middleweight"
states, with 157 seats (Assam, Gujarat, Haryana, Karnataka, Kerala, Orissa, Punjab, and Rajasthan); and the remaining
"lightweight" states (typically with one or two Lok Sabha members), of which the larger are Delhi, with
seven seats, Jammu and Kashmir, with six, and Himachal Pradesh, with four. See Appendix Table 2 for party strength
in each state.
|
What determines which parties develop alliances? The pivotal
factor is the dynamics of state politics and its relationship to political dynamics in Delhi: the pattern of opposition
in the state determines what alliances are possible for the national election.
To be sure, personal rivalries of party leaders also play an important part in getting parties together, but these
days ideological compatibility plays a decreasing role.
There are three poles around which party alliances form
at present (see Figure 1). In order of
vote share in the 1998 election, they are: Congress (25.8 percent), BJP (25.6 percent), and the "Third Front"
(15.6 percent), in which the Communist Party of India (Marxist) and the Samajwadi (Socialist) Party are the largest
components. The current BJP-led alliance, now formalized as the "National Democratic Alliance" (NDA),
would have had, including supporting parties not in the NDA, 39.3 percent of the 1998 vote.
The BJP has one major party as a firm ally, the Shiv Sena
in Maharashtra, and the lion's share of "regional" parties, many of whom at one point had ties to the
"Third Front." The DMK of Tamil Nadu is now allied with the BJP, whereas its arch rival, the ADMK of
Jayalalitha, has shifted its problematic support from the BJP to the Congress.
Figure 1:
The Congress has a stable set of allies in Kerala, as part of its "United Democratic Front" (UDF),
and one other major ally in Bihar, the RJD of Laloo Prasad Yadav, who was also once part of the third front.
A number of parties will not ally with the Congress
at the national level, for ideological or personal reasons, or because the Congress is their chief rival in their
particular states. The Samata
Party of India's present defense minister, George
Fernandes, a fairly firm ally of the BJP, was once part of the "Third Front." Fernandes has been at the
core of alliances based on "anti-Congressism," and will not ally with the Congress for that reason. The
Lok Shakti
of Ramakrishna Hegde, a very successful chief minister of Karnataka in the Janata Dal, and another old colleague
of Fernandes, is fairly firmly allied with the BJP. The Janata Dal (JD) split in August 1999 into the JD(U) and the JD(S). The Samata and the Lok
Shakti (LS) will contest under the JD(U) party symbol,
and are attempting to exert leverage vis-‡-vis the BJP within the NDA, and the coalition government they believe
will come.
The parties that will not ally with the Congress because
it is their major rival in the state include four major parties, all of them allied with the BJP for this election.
The Telugu Desam (TDP)
Party of Chandrababu Naidu in Andhra Pradesh is the largest. Naidu was the convenor of the United Front alliance,
but supported the BJP alliance as it put together its government in 1998, and is now allied with the BJP for these
elections but is not a part of the National Democratic Alliance (NDA). The Shiromani Akali Dal (SAD)
rules in Punjab, and the Asom Gana Parishad (AGP) governs Assam. The Biju Janata Dal (BJD) in Orissa is also directly confronting the Congress in that state;
when it split from the Third Front's Janata Dal in the 1998 election, it allied with the BJP.
Three parties have been formed by dissident Congress
party members. In two cases–the Tamil Maanila Congress
(TMC)
and the West Bengal Trinamool Congress (WBTC) led by (Ms.) Mamta Banerjee–they took the bulk of the state party
unit with them. These parties clearly compete for followers with the Congress, and so an alliance with the Congress
is not possible (although, given the history of similar parties, merging with the Congress after the election cannot
be ruled out). The TMC is not prepared to ally with the BJP, but the WBTC did so in the 1998 election.
The latest entrant into this group is the National Congress
Party (NCP)
of Sharad Pawar in Maharashtra. He was expelled from the Congress in mid-May 1999 when he called for Sonia Gandhi
to consider the difficulty inherent in her being born a foreigner and renounce her goal of becoming prime minister,
and support a constitutional amendment requiring that the prime minister be born in India. Pawar commands a large
following in Maharashtra. The Shiv Sena—BJP alliance is delighted to have a simultaneous state election with its
opposition split down the middle, as it was in 1996 (again because of Pawar). The alliance then won power in the
state (it lost badly in the 1998 Lok Sabha election when Pawar was back in the Congress fold). Pawar is exploring
alliances with Third Front parties, particularly the Socialist Party.
The last of the major parties to be accounted for here is
the Bahujan Samaj Party (BSP), the party of the "ex-untouchables" or dalits, led by
Kanshi Ram and (Ms.) Mayawati, which polled 20 percent of the vote in Uttar Pradesh in 1996 and 1998 and almost
5 percent of the national vote. Kanshi Ram has called himself an "opportunist": he will ally with any
party, or fight the election on his own, depending on what he sees will advance the interests of the dalits who
support him. He has been remarkably successful: Mayawati has served as chief minister of Uttar Pradesh in a power-sharing
agreement with the BJP, and implemented a number of programs designed to boost the self esteem of the dalits. The
BJP took office as agreed, but then engineered party splits that enabled it to do without BSP support. That "betrayal"
may well have contributed to the dramatic surprise of Mayawati casting the BSPfive votes to bring down the BJP
government in Delhi.
There are a formidable number of independents and tiny parties
that contest elections, but although their vote share adds up to a significant amount, and in certain states they
have a real impact, they have played a very minor role in alliance politics.
Party Characteristics: Geographic Spread, Membership,
Leadership
This description of the pattern of party alliances has
glossed over other characteristics of the parties. Most significantly, the geographic scope of the parties needs
to be spelled out clearly. There are now only two parties that have both national ambitions and national reach:
the BJP and the Congress. The BJP contested 50 percent
or more of the seats in 14 of the 18 major states, and didn't lose its deposit in any contest in nine of those.
It lost its deposit in all the seats in Kerala, but that was the only unmitigated disaster. The Congress contested
50 percent plus in 17 of the 18 (Bihar was the 18th). It lost its deposit in all the seats it contested in Tamil
Nadu, in 67 percent of those in West Bengal, and, shockingly, in 91 percent of the contests in Uttar Pradesh, completing
the precipitous decline in support that had begun with the 1989 election.
Some of the Third Front parties once had some national
presence, particularly the Janata Dal, but most wield serious power in only one or a handful of states. This includes the CPM, which is a serious presence only in West Bengal,
Kerala, and Tripura.
There are other parties, typically called "regional,"
which represent what might be more accurately called a "nationality"–a community aware of its separate
identity defined in terms of language, religion, or homeland, but which while often making claims on the government
for safeguards and increased autonomy does not claim to be sovereign.
These parties include the DMK, ADMK, AGP, Akali Dal, and TDP, as well as some parties that have lost considerable
support, such as the Jharkhand Mukti Morcha (JMM), a party claiming a homeland for tribes in east-central India.
Some parties have very strictly regulated membership–the
CPM in particular–but most have, below the level of formal holders of office in the party, a more loosely defined
membership composed of "political workers."
They also have links to affiliated organizations such as cultural groups, trade unions, and student organizations.
Party policies and alliance decisions must take into account not only the party's support base among voters but
also the loyalty of its political workers and affiliated organizations and their supporters. Political workers
are the backbone of a party's strength: they provide the enthusiastic participants in rallies and processions,
the manpower–they are overwhelmingly male and comparatively young–to organize events, put up posters and banners,
serve as polling agents on election day, and collect funds. A significant number have a firm commitment to the
party's ideology and programs, and that can prove to be a problem when alliances are formed. It is important that
parties differ significantly from each other in the character and commitment of the party workers.
Parties formed at different times have different leadership
generations. The founding generation of leaders of
the CPM are just now passing from the scene. The BJP, even more importantly, is still very much under the control
of leaders who grew up with the party after its founding in 1951 (Prime Minister Vajpayee and Home Minister Advani,
most prominently). The Congress, of course, has seen several generational changes, even leaving aside those of
the preindependence era: the last of Mrs. Gandhi's cohort are now serving mainly as advisors to Sonia Gandhi. Leaders
of the opposition to Mrs. Gandhi's rule, who came together in the Janata coalition (later party) of 1977 form the
core of the non-CPM "Third Front," but they remain in many cases leaders without much organizational
following. Other crucial founder-leaders include Bal Thackeray of the Shiv Sena and Kanshi Ram of the BSP.
A Foundation of Consensus: Economic and Foreign Policy
When the BJP government fell, and the Congress failed to
put together an alternative ministry, a critical piece of government business was left suspended: the passage of
the Union budget that had been introduced to considerable acclaim at the end of February. Because of a legal requirement,
the budget had to be passed by the about-to-disappear Lok Sabha. Even though the opposition had objected strongly
to certain provisions, the budget was passed within a week, after some careful negotiation by all the parties and
some minor changes, more or less by acclamation. A host of bills affecting the economy and other matters were left
to die with the government, and the business community mourned the loss of at least some of them.
How was it possible for parties who had been almost literally
at each others' throats a few days before to come together and pass the budget? The primary reason is that democracy
and democratic procedure in India has become second nature to all: it would be almost literally unthinkable for
even a quite venal and irresponsible politician to hold up a constitutionally necessary step. But the other reason
is more prosaic: the BJP, the Congress, and many parties
of the United Front had come to accept an economic program that differs in significant but not fundamental ways.
The agreement on fundamental economic policy among the parties,
quite apart from the necessity of passing the budget, is clear. The
economic reforms instituted in a major fashion in response to the economic crisis of 1991–though there had been
early moves in that direction over the previous decade–have essentially been accepted by everyone except the left
wing, represented mainly by the CPM (in whose party congress in 1998 Stalin's portrait was still proudly displayed). According to India
Today (May 10, 1999, p. 17), even the socialist parties
believe in measures such as privatization of public-sector undertakings (though not of the nationalized banks)
and "with 80 per cent of the 12th Lok Sabha concurring on economic policies, reversal of reforms is now a
non-issue."
Shortly after the Lok Sabha was dissolved, the army
discovered the presence of "intruders" from Pakistan well beyond the "Line of Control" in Kashmir,
along the road connecting Srinagar and Leh, near the town of Kargil.
It took the army two months of hard fighting, with significant numbers of Indian soldiers killed and wounded, plus
diplomatic pressure from even Pakistan's traditional friends, to force the withdrawal of the combination of Pakistan
army personnel and Kashmiri and non-Kashmiri mujahidin back behind the Line of Control. As can readily be imagined, political maneuvering quickly receded into the
background, into a patriotic determination not to do or say anything that would undermine the war effort.
Indeed, even in times of peace, partisan differences over
security and foreign policy are largely matters of nuance and timing: it is worth remembering that the Congress
government of Narasimha Rao was on the verge of testing a nuclear weapon in 1995 and the subsequent United Front
governments, even that of I. K. Gujral, did not renounce the "nuclear option." There is also considerable
overlap in party positions in foreign economic policy, particularly now that the BJP has governed in Delhi and
discovered the constraints stemming from India's international obligations to the WTO and other organizations.
By the time foreign-policy issues reach the level of the
voter, differences are even harder to discern. The public opinion surveys rarely ask about general foreign policy
issues, probably because the "don't know" responses would be quite large. Questions are asked about India-Pakistan
relations, but the answers may well reflect attitudes on the domestic issue of Hindu-Muslim relations. Asked to
react to the statement that India should develop friendly relations with Pakistan, 45 percent of the respondents
agreed, 18 percent disagreed, and 38 percent gave "don't know/no opinion" responses. The party differences
are not very striking: 17 percent of Congress and Left Front partisans disagreed with the statement; 12 percent
of National Front (the Janata Dal and others) and BSP voters disagreed, while 23 percent of voters favoring the
BJP alliance disagreed (1996 data; p. 149). Given the BJP's strong if not strident anti-Pakistan rhetoric for the
last decades, that is a surprisingly low figure.
Why Voters Vote the Way They Do
Social and Economic Identity
India's democracy has been far from static in the
decades since independence. But voters have all along
favored candidates and parties for reasons that will not surprise anyone. They vote on issues; they vote for (or
against) particular people because of their enthnicity or moral character or charisma or a host of other reasons;
they vote "tactically," taking into consideration party alliances and other such matters; they vote on
their immediate material and other interests; and they vote on the more generalized grounds of their view of how
the country is being governed (in which the economic and international situations definitely are important). They
vote for parties because they best represent the mix of those motivations–particularly, according to most analysts
of the Indian political scene, for parties that represent, in some fashion, the voters' ethnic or religious community–or
simply because they have "always" voted for that party.
Social Ties, Particularly Caste
Citizens now do not vote according to the orders of caste leader, political patron, or
even husband; nor are their votes typically "bought" in any direct way.
There is a genuinely secret ballot, and voters long ago learned that whatever promises they made and whatever money
they took, they could vote as they wished in the polling booth. In a handful of places voters are paid–or, more
commonly, ordered with the threat of force–to stay away from the polling booth altogether.
|
CHANGING ELECTORAL SYSTEMS
Political scientist Yogendra Yadav of CSDS has identified three "electoral
systems": the first operated in the first four Lok Sabha elections, in the era of the dominance of the Congress
party; the second began with Indira Gandhi's "Garibi Hatao" ("Get rid of poverty") 1971 campaign;
and the third system with the 1989 election that brought V. P. Singh to power. The third electoral system has four
distinctive features. The first is the "participatory upsurge" among women, scheduled castes, and scheduled
tribes, which has brought them much closer to the other, upper groups in society in terms of voter turnout. The
second is the penetration of the belief systems, particularly concerning social justice, of these groups into political
discourse. The third feature is the continuing change in the relationship of social cleavages (around caste and
religious community) to voting, with a far greater weight given to state-wide general groupings of castes than
before. The fourth feature is the primacy of politics at the state level. Yogendra Yadav, "The Third Electoral
System," Seminar 480: Cruel Choices (August 1999), pp. 14—20.
|
The survey data reveal some important facts that run counter
to the conventional wisdom on voter behavior. Women
in 1998, when asked who they consulted before voting, overwhelmingly said, "no one" (65 percent); another
17 percent consulted their husbands; and 11 percent other members of the family ( India
Today, February 16, 1998, p. 42). In 1996, 75 percent
of the sample said they were not guided by anyone in their voting decision; only 16 percent of the men admitted
to being guided by someone else (p. 97). Of the 25 percent who sought advice, only 7 percent sought it from caste
and community leaders (70 percent took it from their family members)(p. 98); that is ,
less than 2 percent of the electorate got direct advice on how to vote from caste and community leaders.
The survey analysts correctly note that these results
would seem to call into question the cherished notion that leaders and parties draw on "vote banks" of
particular castes and religious groups. On the other hand, the survey data show that social background, defined
in terms of caste, caste category, or religion, is a good indicator of voting behavior. For example, Muslims do not vote for the BJP (7 percent voted for the BJP plus
allies in 1998; Muslim support for the BJP on its own must be even less). The congruence of the dalit vote with
the BSP vote in Uttar Pradesh and areas of Madhya Pradesh next to Uttar Pradesh is remarkably high. Survey data
from the 1996 U.P. state assembly elections show that with Uttar Pradesh's BJP chief minister of the lodh
caste, 78 percent of the lodhs voted for the BJP (p. 133).
|
ON A PERSONAL NOTE
My own experience in talking to voters during election campaigns (in 1969,
1984, 1989, 1996, and 1998) makes me trust this result. Politicians in their speeches often refer to the voters'
"valuable vote"–and voters do believe that even if their vote represents a very small piece of power,
it is often the only piece of political power they can exercise. They understand that they can gain benefits by
keeping their voting intentions well hidden from those who ask; the politicians complain that they have to make
promises without having any assurance that the quid pro quo will materialize, because if they don't make the promises,
the vote will surely go against them.
|
But we have to be careful about carrying "social base"
analysis too far. The most important survey data show the change over time. In 1971, 51 percent of the respondents
agreed that it was "important to vote the way your caste/community does" (30 percent disagreed), but
in 1996 the percentages were reversed: 51 percent disagreed with that statement (29 percent agreed). The same percentage
(25) in both 1971 and 1996 believed that their community "votes for one party," but 22 percent in 1971
believed that they vote for different parties, while that figure went up to 48 percent in 1996. In 1996, a convincing
64 percent said that there is "no political party" that "cares specially for" the interests
of their caste or community. ( India Today, February 2, 1998, p. 38) In 1998, "caste and community"
was seen as an issue by only 5.5 percent of the respondents in one poll (Frontline/Centre for Media Studies Poll of January 1998; Frontline, March
6, 1998) and ranked last of nine issues in another (India
Today/ORG-MARG Poll; India
Today, January 5 1998).
|
CHANGING "VOTE BANKS"
The most devoted followers of Indian politics–politicians and political
workers, journalists, and some academics–invariably use caste and religious community for their analysis of political
support and political trends. "How can we get more of the Muslim vote?" asks the politician; "the
rajputs are unhappy with the ministries they've been given," says the political worker; "the BJP's OBC
[other backward classes] vote has been picked up from an ailing United Front," writes the academic. All use
the term "vote bank." A vote bank in its original meaning and reality (during Yadav's "first electoral
system") was a deposit of votes that a caste leader could deliver reliably to a patron or an ally. As the
reality of the secret ballot became clear, and as leaders used delivered votes mainly to benefit themselves, the
constituents of the "vote bank" (the citizens themselves) dissolved it. "Vote bank" then came
to mean a stratum of voters defined by their caste identity that could be mined for support, again by caste leaders.
In these two incarnations, a vote bank referred to a narrowly defined group, the jati (and its equivalent among
Muslims, the sect, or the biraadari), which is defined by the requirement–easily maintained in a society where
virtually every marriage is an arranged marriage–that people marry within their caste. In the years since the 1989
election, the concept of vote bank has been stretched to cover a very few caste and community groupings within
which there is enormous variation and even antagonism among jati and religious sects, so that its utility is highly
suspect. It is hard to understand the justification for always putting Shias and Sunnis in Uttar Pradesh together
as "the Muslims," for example. The now-standard groupings, and the estimate (for the first two categories)
of their percentage in the electorate are: Upper Castes (16 percent), OBCs (44 percent), Scheduled Castes [Dalits]
(15 percent), Scheduled Tribes [Adivasis] (8 percent), Muslims (12 percent), and other minorities [mainly Christians
and Sikhs] (6 percent).
|
All the evidence points to the fact that these respondents
are correct: members of particular castes (and especially members of the large caste groupings such as "Dalit"
or "Other Backward Classes") can be found voting for every party.
At the local and state level, one finds particular parties that attract the votes of particular communities–for
example, the Socialist Party in Uttar Pradesh draws heavily on the votes of midlevel agricultural castes. Parties
with national reach, however, frequently draw on different groups in different states. To give one example: the
Congress got 41 percent of the "upper caste" vote in West Bengal and Kerala (where it competes with CPM
led fronts) and 7 percent of the "upper caste" vote in Uttar Pradesh and Bihar ( India Today,
February 2, 1998, p. 39).
In India as a whole, moreover, political polarization
along social cleavages has declined. It is less and
less true that knowing the caste of a voter lets you reliably predict the party he or she will vote for. If there
are two parties competing in a constituency it does not necessarily mean anymore that they represent two distinct
social groups.
Class Identity
Why then is "social base," properly applied, a
good indicator of voting behavior? The main reason is that caste identity is "shorthand" for other things,
particularly economic position. It is true that parties nominate candidates to take into account large blocs of
"socially" defined people. Usually this means that all the candidates come from the predominant caste
or community group. If there is no predominant caste, a candidate from a caste that has nonconfrontational relations
with other comparable castes might be chosen. The calculation of likely caste-derived support is a matter of political
savvy, and is often mistaken, particularly since there are no reliable data on how many members a particular caste
has.
"Social base" is closely correlated with a whole
host of other factors: most important, wealth (or if you wish, "class"). As a (very) rough rule: the
lower the caste the poorer the person. Education and caste/community are also clearly correlated: the higher the
caste, the higher the average educational level.
There is no doubt that the first cause of this correlation has been caste and community: certain castes were prevented
from obtaining education and were consigned to positions of economic inferiority and kept there, often by the use
of force. And particularly when it comes to the dalits (which means "the oppressed") an often-brutal
hierarchical ideology remains an important factor in politics as well as social and economic life. Dalits, because
they are dalits, are prevented from prospering and gaining educational advantage, and not just in the countryside.
Castes are often associated with particular occupations,
and certain groups were able to take advantage of India's economic changes–the green revolution, the development
of certain industries–because they were in the right socioeconomic niche at the right time. At the same time, politics
provided the opportunity for many groups to mobilize in large numbers to redirect government resources toward themselves,
thus reinforcing certain caste or caste-group identities.
With the exception of certain interests castes or communities
have in common–for instance, the interest Muslims in north India have in promoting the recognition of the Urdu
language–interests are likely to cut across caste lines, and conform to economic or other social divisions.
The survey data on the relationship of class and caste also
reveal that "class-based support for the BJP is so significant that as the SC [Scheduled Castes], i.e., dalits
(a group which is quite hostile to the party) move up, occupying higher rungs on the class ladder, their preference
for the BJP also increases, from merely 10.7 percent from the very poor, to 19.0 percent among the poor, to 25.3
percent among the middle class, to 40.8 percent among the upper class. More or less similar trends have been noticed
for the OBC and ST [Scheduled Tribes] as well" (p. 139).
Indian "classes" are defined typically by outsiders
who consider not only wealth in land, income, and occupation, but also consumption patterns (particularly when
identifying the "middle class"). Classes as such have not been notably active in politics. Classes are
divided vertically by caste and other social cleavages. At the village level–and 70 percent of India still lives
in the countryside–patron-client structures are common, in which landowners mobilize economically and socially
dependent men (some of whom may be of subordinate castes) to compete with other landowners with similar groups.
Not infrequently these competing and confrontational factions (villagers often call them by the English word "party")
are headed by brothers.
Electoral issues and the allocation of economic benefits
by the government can often be understood in class terms.
Land-owning classes have an interest in the continuation of the heavy subsidies for fertilizers and electricity
for irrigation pumpsets. Poor people favor the "food for work" and other employment-generating programs.
Unfortunately, we don't have survey data that would establish the precise connections of those interests with voting
behavior.
Corruption was one electoral issue that once was thought
to be related to class. The 1989 election's core
issue was corruption, and some interpreted the increased vote share of the BJP in that election as a reflection
of the increasing influence of a middle class increasing rapidly in size. In the 1991 election, when the BJP increased
its vote share yet again, one interpretation held that this reflected the popularity of the BJP-promoted revivalist
Hinduism among the middle class, as shown by the great popularity of the TV serialization of the two great Indian
religious epics. This question has not been pursued by scholars, and the easy equation of the middle class and
the BJP has been dropped by columnists and other pundits.
Perennial Election Issues
Group characteristics–caste, class, education, gender,
and others–can predict voting behavior to some extent. Yet voters focus on "issues" and candidates give
speeches that deal mainly with issues. Some issues are hardy perennials: "corruption," "stability,"
and inflation. "Leadership" becomes an issue particularly when there are competing prospective prime
ministers.
The 1996 survey allowed respondents to name issues spontaneously,
rather than react to a menu of choices. In that survey, the top-ranked issues were: corruption (named by 20.9 percent
of the respondents), stability and good government (10.4 percent), "price rise" (7.0 percent), unemployment
(5.9 percent), and poverty (5.1 percent) (p. 109).
What the percentages suggest is that when voters are
not given a menu of issues, no particular issues stand out in an "ordinary" election. It is also worth noting that "economic reform"–which was
after all a major feature of the Narasimha Rao government–does not appear on the list (though "Babri Masjid"
[in Ayodhya, demolished during Rao's tenure] ranks ninth, with 3.1 percent of the respondents mentioning it).
The preference for a "stable government"
does not mean necessarily the government of a single party: governing coalitions have been stable in West Bengal
and Kerala (where there are alternating CPM-led and Congress-led coalitions). But coalitions have more often meant,
first at the state level (from 1967 through the 1980s), and then at the national level (in 1977—79; in 1989—91;
and in 1996—99) an uncertainty of policy enactment and program implementation that has disturbed voters.
Coalitions have been unstable, with governments falling and
new cabinets being formed after only a year or two in office. They have been prone to greater shifting of ministers
(the political heads of departments, and the most coveted post in Indian politics). That has meant not only changes
in policy but also, far too often, a demand for a bribe to keep a government contract or other benefit, when the
predecessor minister had already received a bribe.
In such a period of uncertain policy and uncertain program
implementation, businessmen cannot plan and development programs get replaced by a newly named, slightly different
one. Bureaucrats are capriciously transferred because they are not to the (latest) political authority's liking,
with the consequence of short terms in an office. (Just as you learn the job you are gone.) Money is wasted on
short-term splashy projects rather than long-gestation ones.
Unstable governments thus affect citizens of all occupations and classes. Voters have consistently punished defecting
politicians–in the period before the antidefection law put a stop to the most blatant changing of sides–who have
brought down governments. Whether they will punish defecting parties or their allies (the ADMK, for example, in
this election) remains to be seen.
"Price rise" is clearly a major concern to voters.
Elections have allegedly been won on the issue of the price of onions. We don't have recent studies on the precise
way in which inflation affects voting–for example, is there a threshold, such as "double-digit inflation"
(which in fact has been quite rare), that tips the balance? Is there a time lag, so that it is really the economic
conditions of a few months before that is meaningful for the voter? Or conversely, does it make a difference if
the election is held just before a harvest (the time of greatest scarcity and high prices) or just after?
The survey data can tell us about longer-term trends
of views of economic betterment. The data are analyzed
according to gender, caste, education, and other variables; we will just look at the economic class differences
(see Table 1 in
Appendix). Only half of even the "upper"
class is satisfied with their financial situation, so it is hardly surprising that economic issues are prominent
in elections. It is also important to note that half of the "very poor" and "poor" expect their
children to have better opportunities, which suggests that even the most deprived might use their vote to that
end, if that would help.
"Leadership" is another significant issue.
Analysts of the Indira Gandhi era, from the 1971 election through the 1984 election, have emphasized the "presidential"
nature of parliamentary elections: that is, voters seemed to favor a party (or coalition) that had a clearly designated
prime minister-to-be. The desire for a "stable
government" is obviously linked to that preference. To have a "government that works" (Mrs. Gandhi's
slogan in the 1980 election) one needs a leader who can rule both his own party and the country.9 It
is not clear, however, how much the leader matters to the ordinary voters in post-1989 politics. The prime minister
designate does matter a great deal to the party workers: slogans using the name of the leader are the most enthusiastically
shouted.
The public opinion polling in the last election does
not help very much here, although the contest between Atal Behari Vajpayee and Sonia Gandhi was prominently discussed
in the media, and the BJP made a major effort to "project" Vajpayee as prime minister. In the most authoritative opinion poll, Vajpayee was named by 21.5
percent and Sonia Gandhi by 13.6 percent, and the then prime minister, I. K. Gujral, by 3.4 percent (India Today,
February 23, 1998, p. 20; the same poll reported that 32 percent had "never heard of" Sonia). Other polls
had much larger figures in the same order, and taken together, the polls certainly suggested that the BJP was heading
for a victory, but we have no evidence of how important that contest of presumptive prime ministers was. The first
poll of this election has voters favoring Vajpayee over Sonia Gandhi by 51 to 33 percent (with all others, including
Manmohan Singh, at two percent or less).10
Loyalty to Leaders and Party
The qualities of local leaders, and particularly those
who are contesting in the voters' constituency, that are ultimately the only thing actually being voted for (in
India's parliamentary system) can matter a great deal.
Elections in which a "wave" of public support for a party meant that even a "lamppost" candidate
could be elected on a surging party's ticket are rare; and there are few genuinely "safe" seats for any
party.
Although incumbents consistently win half the time, a Member
of Parliament who pays no attention to his or her constituency, who doesn't even bother to visit, is a real liability.
Members of Parliament spend a great deal of time, starting very early in the morning, meeting with constituents,
in Delhi and in the constituency. Personal attractiveness also matters: being able to speak well–sincerely, eloquently,
with humor–is valued. In certain constituencies, a reputation for integrity is critically important; in others,
voters are more tolerant of a candidate who "plays the dirty game" of politics, as long as he or she
shares the benefits with them. And nationally known leaders–particularly prime ministers–can get themselves elected
from constituencies far from their home.
Voters also have loyalty to parties, and vote "tactically."
Alliances can work only if a party can, with reasonable reliability, "deliver" its votes to an alliance
partner. (The mechanism of an alliance is seat-sharing:
assigning constituencies to the alliance partners.) One explanation of why the BSP has been reluctant to join alliances
is that while its leaders can deliver BSP votes to other parties, other parties' voters, because of caste prejudice,
seem reluctant to vote for BSP candidates. These are seen as representing dalits (though in this election, fewer
than a quarter of the BSP candidates in Uttar Pradesh are dalits). When a party that had some support "gives"
a seat to another in a seat adjustment, the recipient certainly gains votes, but rarely as many as would be calculated
by adding the vote shares in the previous election. The longevity of some alliances–the Left Front in Bengal, the
Left Democratic Front, and the United Democratic Front in Kerala–is clearly based on the willingness of voters
to vote for the alliance partner of their own party.
"What Have You Done For Me Lately?"
The mass of Indian voters–who are overwhelmingly residents
of villages and small towns, and slum areas of the cities–are very directly concerned with getting immediate material
benefit from the government. There is no doubt that
"what have you done for me or to me lately?" is very much in the foreground as voters make their choices.
Parties have been very successful in elections when they promise things such as cheap, even free, electricity for
farmers' irrigation pumpsets. Since electioneering, let alone the political ad, have not yet made it onto Indian
TV in any significant way, open-air meetings are still the main way parties and candidates communicate with voters.
The speeches are filled with details of development projects and other government programs of particular interest
to people of that area, not to speak of promises to get a bridge built, or a school, or a health center. One doubts
that voters expect much in the way of delivery on these promises, but they do expect the promises to be made, for
politicians to show concern for the very direct needs of their constituents. These
concerns may be the most important thing for the voters making their choices in an election, like this one, without
major issues that stir people.
Most of the programs that touch peoples' lives continuously–education,
health, law and order, electricity supply–are state-government programs. Rural development programs also are administered
through the state, sometimes along with newly energized institutions of local government, even if the money comes
from Delhi. Thus the politician who has most influence on getting voters material benefits is a local one, with
strong links to the ruling party in the state.
The "Anti-Incumbency" Factor
One vital factor affecting voting is the "anti-incumbency"
factor (known in the United States as "throwing the rascals out").
There is certainly a pattern in some states of alternating ruling parties: in Kerala (Left Democratic Front CPM-led
versus United Democratic Front (Congress-led), Himachal Pradesh (BJP vs. Congress), Tamil Nadu (DMK vs. ADMK),
Rajasthan (BJP vs. Congress), Delhi (BJP vs. Congress), and many others. There are, however, a few significant
examples of ruling parties not being turned out: most important, the Left Front (led by the CPM) has been ruling
West Bengal continuously since 1977. An impressive example was the reelection of Ramakrishna Hegde's Janata government
in Karnataka in 1985, only a few months after the Congress had swept the state in the Lok Sabha election.
Recently, incumbent governments were reelected in Bihar and
Madhya Pradesh. There is some suggestion that these states were "better governed" (even in Bihar, where
Laloo Prasad Yadav was credited with preventing communal rioting in the aftermath of the demolition of the Babri
Masjid), and the voters rewarded the ruling parties for that. "Anti-incumbency" is clearly not an iron
law.
The 1998 election was seen as an "anti-incumbency"
poll. It is not simply that the United Front parties, who were governing in Delhi with the support of the Congress,
lost votes, and the Congress along with them. More important, it was the party that was in opposition to a ruling
party in a particular state that tended to benefit in
that state, improving its numbers in the state's
delegation to the Lok Sabha. Thus the Congress "won"
in Rajasthan where the BJP was in power (it has since lost to the Congress in a state election). And the BJP-BJD
alliance "won" in Orissa, where a Congress government is in power.
Looking at each state in turn provides strong evidence
for this pattern holding true. It is not at all clear
what the reasoning of voters might be here: just how does it help to vote "against" those in power in
the state (who are going to continue governing there) in a national election? Is it to send them a warning? Unfortunately,
the question has not been asked in the surveys.
At the constituency level, the picture is not so clear. Parties tend to give incumbents the chance to defend their seat:
83 percent of MPs elected in 1996 ran again in 1998. Seventy-two percent of BJP incumbents won (compared to 33
percent of nonincumbent candidates) and 55 percent of Congress incumbents won (versus 23 percent of the nonincumbents).
Taken together, the non-BJP, non-Congress major-party incumbents were successful 50 percent of the time, while
only 29 percent of their nonincumbent counterparts won.
Voting Behavior
Whatever it is that determines why people vote the way
they do, we know that many don't make up their minds until very late in the campaign. According to the survey data, 28 percent of the respondents had decided whom
to vote for before the campaign began. One-third decided a few days before the end of the campaign and 17 percent
on polling day (India Today, March 16, 1998, p. 52).
For the observer, the reticence of the voter and the making
up of minds just before polling day make it difficult to predict results. The last "wave" election at
the national level–where there is a last-minute surge of support and a bandwagon effect turned the election around–occurred
in 1980. But the factors that affect voting behavior are so diverse, and the variation among state electorates
so great, that changes during the course of the campaign have been significant. In 1998, the "late swing"
to the BJP was almost 5 percent, and in certain states even higher (13 percent in Tamil Nadu). And we don't know
why.
Although most voters are not actively engaged in the
election, a great many are well informed about the issues and competing parties and candidates. The survey data tell us that in 1996 the campaign barely touched
most voters, even though 42 percent were visited by a party worker or other canvasser: 65 percent were "not
at all" interested in the campaign, and 83 percent attended no election meetings (pp. 92, 93, 96). Politics
is certainly discussed, especially by men, in informal gatherings, as the campaign gathers momentum, and posters
and banners appear along with processions and loudspeaker vans. According to an index that measured the ability
to name the prime minister, chief minister (of the state), member of parliament, and previous member of parliament,
40 percent of the electorate had "high information," while 29 percent had "no information"
(p. 106).
Indian voters turn out in numbers that continually
surprise the outside observer. All adults over the
age of 18 are put on the electoral rolls by government officials (although many are missed). Although polling booths
are typically conveniently located, the infirm and elderly can find it difficult to walk to the booth and stand
for a long time in a queue. In many parts of India, women are rarely seen outside their own home; separate queues
for women are a common practice. In the first parliamentary election, 45.7 percent of the electorate voted; in
1998, the figure was 62.1 percent. In recent years, the major gains in voting turnout have come from disadvantaged
groups.
Trends and Scenarios of the 1999 Election
As of this writing (mid-August 1999), the BJP and its
allies, now called the "National Democratic Alliance," seem to be in the best position in their competition
with the Congress and its allies and a loosely grouped "third front," which is entirely a gathering of
state-based parties.
Foreign Policy
It is not clear how the Kargil (Kashmir) crisis will figure
in the campaign. There is some sentiment against spoiling the national nonpartisan support of the military and
the (caretaker) government's foreign policy moves with arguments about who should be held responsible for the failure
to anticipate Pakistan's moves and other issues. The BJP may not get a direct advantage from the surge of patriotic
fervor (the soldiers killed are universally referred to as "martyrs"); shortly after testing India's
nuclear weapons in May 1998, which was also greeted with patriotic outpourings from (almost) all sides, the BJP
lost some critical by-elections.
Foreign policy, even national security, has never
penetrated very far as an electoral issue, and in this case, it is not even clear that Prime Minister Vajpayee,
constrained by being a caretaker head of government, has been perceived as the leader of the national effort. The BJP is likely to gain from the enthusiasm of its political workers
over the government's forcing a Pakistani withdrawal. The political workers may, in turn, use this success in BJP
strongholds to reinforce its Hindu nationalist message.
Economic Performance
The economy has been doing well, particularly considering
the consequences of the sanctions imposed after the nuclear tests.
The 1998—99 record grain crop, and the drop in food prices (combined with the inflated prices of a year ago, against
which the inflation rate is measured) has meant an inflation rate measured by the wholesale price index of less
than 2 percent (in mid-August). Since "price rise" is a perennial election issue, so low an inflation
rate–even though there are some measurement problems that suggest the real rate is closer to 5 percent–should favor
the party in power.
The GDP, according to the Central Statistical Organization
(report of June 30, 1999) grew by 6 percent in 1998—99 (the Indian fiscal year ends on March 30), and real per
capita income by 4.2 percent. Industrial growth has also seen a spurt, and the foreign exchange position is comfortable.
The Bombay Stock Exchange index reached a record high in July, after four years. The most recent budget of the
BJP government was widely welcomed when it was introduced in February, and the BJP will undoubtedly take credit
for the strong economy. But it is also clear that the policies it has pursued are virtually indistinguishable from
those of its predecessor governments, and it has given up some of its more strident demands for "self-reliance,"
which may alienate some of its political workers. In any case, economic policy has not been an issue in Indian
elections, at least recently.
Atal Behari Vajpayee versus Sonia Gandhi
The BJP clearly plans to make leadership a major issue in
the campaign, making pointed comparisons between the experienced prime minister designate, Vajpayee, and the inexperienced
leader of the Congress, Sonia Gandhi. Whether or not the BJP formally makes Sonia's foreign birth an issue, we
can be sure that among the political workers, and thus in much local canvassing and public meetings, it will be
a major theme. The issue has, of course, already been explicitly raised by Sharad Pawar and others as they moved
to form the National Congress Party.
Public opinion clearly favors Vajpayee. In the 1998 election,
Sonia kept the Congress from being beaten even more decisively. Before her claim to be prime minister was made
explicit, 54 percent agreed that she should "lead the Congress" (72 percent among Congress supporters)
(ORG-MARG poll; India Today, January 5, 1998), and only 24 percent felt her Italian birth was a political
liability (Outlook,
January 26, 1998). The irony here is that it is the most educated citizens who object to a "foreigner"
being prime minister; more traditional Indians believe that a wife takes on not just the name of her husband, but
literally becomes part of his lineage. Sonia has been a wife and daughter-in-law and widow who has visibly, and
sincerely, adopted Indian customs, and that has made her acceptable as an Indian (and not just as an Indian citizen) to a lot of voters.
Sonia's lack of political experience, and her unwillingness
until recently to expose herself to uncontrolled political dialogue in public (for example, in a press conference)
is another matter. That has probably cost her support among the middle class in particular, though she has clearly
developed a rapport with many Congress workers. Sonia's control of the Congress party should count with voters
looking for a leader who can govern.
She did not do well in the maneuvering following the vote
of confidence, and suffered the embarrassment of having to admit to overestimating her support. Some have claimed
that she seemed willing to become prime minister through backroom bargaining, while denying an ambition to "take
power."
The polls available in mid-August make the position
clear. In the Outlook—Centre for Media Studies poll, The BJP and its allies had a 39 percent
vote share, but 51 percent of the respondents chose Vajpayee as "the best leader to be the next prime minister
of the country," with Sonia getting 33 percent (Outlook, August 16, 1999, p. 26). The Times
of India poll had the BJP and allies favored by 44
percent of the respondents. Without being prompted, 57 percent chose Vajpayee and 27 percent Sonia in answer to
the same question (Times of India, August 14, 1999). The India
Today—Insight poll is in agreement: the BJP and allies
get 41 percent support, with Vajpayee chosen as the person "who will make the best prime minister" by
50 percent of the respondents compared to Sonia's 26 percent (India
Today, August 21, 1999, pp. 15—16).
The Third Front parties, as of this writing, have
not put forward a prime-ministerial candidate, and their record of divisiveness while in government certainly does
not help with voters concerned about "stability."
They do have some leaders with impressive credentials: the veteran chief minister of West Bengal, Jyoti Basu, and
a former finance minister (P. Chidambaram), but also some whose national vision, and integrity, are questionable.
The Anti-Incumbency Factor
It is not clear how the anti-incumbency factor will affect
the BJP coalition. On the one hand, it didn't really
last long enough in power to make much of an impact, nor were its members bedeviled with corruption scandals (aside
from Jayalalitha, and her departure was in some sense provoked by Vajpayee's inability to help her escape trial).
On the other hand, the government lasted long enough that it can no longer claim to be the one party that had not
been "given a chance" to govern; it now has a record it will have to defend. It is not clear whether
the BJP's need to compromise with its alliance partners on all issues dear to its political workers' hearts, except
for the nuclear tests, has cut into the support of that crucial group. There has been no movement to abolish Article
370 of the Constitution, which gives Kashmir special status; toward formulating a Universal Civil Code (abolishing
the separate Muslim and Hindu laws governing marriage and inheritance); and most of all, toward building a temple
to Ram on the site of the demolished Babri Masjid. And it is clear that the BJP is not going to come to power on
its own, at least in this election, so the political workers will have to work hard for the party knowing that
their core agenda will not be adopted even if the National Democratic Alliance wins. Whether they do so remains
to be seen.
Figure 2:
State-Level Political Dynamics
At the heart of any evaluation of the election prospects
of the various parties and alliances is a state-by-state analysis.11 The
anti-incumbency factor as it applies to the three broad alliances is a useful point of departure.
Looking first at the BJP alliance, it would seem that there
is not much to be gained via the anti-incumbency factor. In Tamil Nadu, the
BJP is allied with the DMK, a party almost certainly
subject to an "anti-incumbency" reaction as a ruling party in the state. The TMC provides a plausible
alternative to voting for the ADMK and Jayalalitha. A major ally since 1998, the TDP in Andhra Pradesh, where state
assembly elections are being held together with this parliamentary election, also is a ruling party, as is the
Shiromani Akali Dal, in Punjab. The BJP itself is a ruling party in Uttar Pradesh and, with the Shiv Sena, in Maharashtra.
It returned to power in Gujarat in February 1998, but lost state power to the Congress in Rajasthan and Delhi in
November 1998. In the latter two states the anti-incumbency factor is not likely to play a role, and in Madhya
Pradesh, where the ruling Congress was returned to power, the BJP is unlikely to be able to gain much, since it
already holds 75 percent of the state's seats.
Figure 3:
On the other hand, the BJP has allies who are in opposition
to ruling parties in other states. The Samata in Bihar, the Biju Janata Dal in Orissa, the Trinamool Congress in
West Bengal, and the Lok Shakti in Karnataka all stand opposed to ruling parties. However, the alliance has already
benefited from the "anti-incumbency" factor in the 1998 elections in those states. And in Karnataka,
the split of the Janata Dal, with the chief minister's new JD(U) joining the NDA, has muddied the picture. The
states mentioned here have 470 of the 543 Lok Sabha seats, but it is not clear that the BJP alliance stands to
gain many seats on the basis of the "anti-incumbency" factor.
Considering the anti-incumbency factor from the Congress
point of view, and counting only those states where it has a major presence, it seems likely to be helpful in gaining
some seats. The Congress opposes ruling parties in
Andhra, Assam, Gujarat, Karnataka, Kerala, and Punjab, with a total of 143 seats. The Congress at present holds
56 of them. Its position in Maharashtra after the split of the party is unclear. The Congress split should help
the ruling coalition in the state. The number of seats the Congress holds where it is in power (Madhya Pradesh,
Rajasthan, and Delhi) is 29 of the total 72.
"Third Front" parties may be able to gain
from "anti-incumbency" only in Maharashtra, Tamil Nadu, and Uttar Pradesh.
The CPM and its allies have incumbency strength in West Bengal and Kerala. Both the Socialist Party and the BSP have
an apparently solid "vote bank" (in the old sense) in Uttar Pradesh. Were they somehow able to combine,
with OBCs happily voting for dalits and vice-versa, they could reduce the BJP's seats by a large amount. At this writing, a quite optimistic prediction finds the Third Front
with far fewer than 100 seats. Former Third Front parties, now allied with the BJP or the Congress, however, may
do quite well, which may give the Third Front some hope of prying them loose in the maneuvering that will follow
the election, assuming another "hung parliament."
The simultaneous elections to the state assemblies in Maharashtra,
Andhra Pradesh, and Karnataka add a major factor to the analysis of those states, which have 118 seats (20 percent
of the total Lok Sabha): the voters' added attention to the record of the party ruling in the state. Anti—state-government
vote in Maharashtra is split between the Congress and the NCP, but one cannot rule out voters choosing, say, the
NCP candidates in the state elections (since Pawar was an effective chief minister) and the Congress candidates
the Lok Sabha. In Andhra, the BJP alliance with the TDP is a loose one, with seats being shared and no common program,
and whether that will insulate the BJP from possible "anti-incumbency" feeling is unclear. In Karnataka,
a state-level BJP all set to ride the anti-incumbency factor to victory has been discomfited by the split in the
ruling Janata Dal and the national BJP's acceptance of the JD(U) chief minister as an ally, under the NDA umbrella.
Congress seems set to benefit in both state and national elections there.
Congress versus BJP
The electoral battle centers on the BJP-led alliance and
the Congress, and it will probably occur mainly in the context of state-level politics. The BJP and the Congress each received 25 percent of the vote in the 1998
election, a historic high for the BJP and a historic low for the Congress. It is tempting to plot the two trajectories
and conclude that the BJP is likely to gain in strength, while the Congress will continue to disintegrate. The
recent polls cited above support such a conclusion.
It is worth remembering, however, that the Congress
has read several of its own obituaries–in 1967, 1977, and 1989–and has somehow managed to reemerge. It seems to some that the BJP has reached some sort of a limit based
on its ideology of Hindu nationalism, because of the need to work with allies who follow "secular" policies,
and may begin to lose support from its core following, including that of its committed political workers. The 1999
Lok Sabha election result, one can be pretty sure, will not settle the question.
Figure 4
Outcomes
Let us conclude by sketching out four possible outcomes and the implications of each, in decreasing order of probability,
in a rough rendering of "informed opinion" in India in mid-August.
The most probable is a strengthened BJP-led government,
one with 300—350 seats. Somewhat less likely are two other outcomes: a BJP-led coalition about as strong as the
last one; or a weak Congress-led coalition, perhaps supported by some "Third Front" parties "from
the outside." Finally, and very unlikely, a "Third Front" coalition supported "from the outside"
by either the BJP (as in 1989) or the Congress (as in 1996).
A "third front" government would probably look
like the two that followed the 1996 election: a weak prime minister spending too much time negotiating with members
of the government. The focus of its efforts would be on "social justice" programs, including clear policies
for the protection of minorities. Its economic policy would tend to harken back to the era of state-led development,
although not to the point of attempting to reverse the economic reforms. Particularly if it were supported by the
Congress, such a government would be quite antagonistic to what is seen as the hegemonic role of the United States
in the world: many in the Third Front parties condemned the NATO bombing of Serbia.
A weak Congress-led coalition would bring in a government committed to the continuation of economic reforms,
provided that its outside support is not that of the CPM. It is not impossible that a price for support would be
Sonia's assignment of the prime ministership to someone like Manmohan Singh, who would be able to work well with
allies, as an honest and competent leader, but who would be unlikely to challenge Sonia's ultimate power. The Congress's
focus in international affairs has always been Europe (the links with the Soviet Union were pragmatic rather than
emotional, for most of those who have remained in the Congress). It is also likely to avoid close links with the
United States, and will be particularly sensitive to perceived attacks on India's autonomy, as an after effect,
as it were, of nonalignment. Domestically, the Congress mid-1980s' flirtation with Hindu nationalism (and the "benign
neglect" policy of Narasimha Rao's government) would be discarded, particularly since the Muslim vote would
have been a major contributor to its electoral success.
A repeat version of the last BJP-led government would
be unlikely to change its policies very much. This
is underlined by the fact that the BJP is not issuing a separate manifesto this time; the manifesto is in the name
of the National Democratic Alliance. The party agenda that attracts the core BJP following and enthuses the workers
in other organizations of the "Sangh Parivar" (a "family" of Hindu nationalist organizations)
will be postponed: no change in Kashmir's constitutional status, no universal civil code, etc. The one item of
the 1998 manifesto that could be implemented without dealing with the coalition partners–the testing of nuclear
weapons–has been done.
Since the economy seems set for a spell of solid and deep
growth, it is unlikely that the approach to the economy will change. The bills on the insurance sector, and others
designed to push liberalization forward, will be revived. The impact of the Kargil crisis on the government's policy
toward Pakistan is unclear, and may well depend on how voters react. On the one hand, the policy of vigorous negotiations
(symbolized as the Delhi-Lahore "bus trip") is being attacked as a policy of weakness that tempted Pakistan
to take advantage in the heights of Kashmir. On the other, the success of the army in pushing the intruders back,
and Pakistan's diplomatic isolation, put the government in a position to resume negotiating from a stronger position.
The BJP is the one party that seems to value closer ties with the United States, but that position is colored by
its dislike of "westernization" in the cultural sphere, and marred by its perceived commonality of interest
vis-‡-vis the Muslim world.
A stronger BJP-led coalition–depending on where the additional strength came from–would probably not force
through the BJP core agenda, nor change its economic and foreign policy very much. It might be even more vigorous
in pursuing minor threads of its program–in education, for example, where the education minister has spent some
effort to reverse what the BJP sees as decades of "leftist" influence on the school curriculum. A BJP
less vulnerable to the threat of withdrawal of support by its partners might give BJP-led state governments freer
rein in implementing core BJP goals, particularly its Hindu nationalist agenda. It is not difficult to imagine
the BJP government of Uttar Pradesh turning a blind eye as "volunteers" assemble in Ayodhya the new temple
whose stone-carved parts are being prepared. Much would depend on which BJP candidates had been successful: those
who have been long-time supporters, particularly of the BJP ideology, or recently promoted leaders of the "backward"
castes or tribals, or pragmatic (or corrupt), machine politicians of the small towns and cities.
It is remotely possible that there will be a sharp
change in voters' perceptions of who should govern in Delhi, a rejection of local-interest—based, coalition politics,
and a precipitous decline in votes cast for the non-BJP, non-Congress parties. But since both the BJP and the Congress
are themselves coalitions of social and economic interest groups, with quite a range of ideologies, even a single-party
government would not change the policies suggested above very much. Even then, this election will mark an incremental
change in the multistranded political dynamics of India.
1. For an exact chronology of political
developments in 1999, see Appendix
1.
2. First names are frequently used in the
press and in academic writings, as well as in conversation. But so are surnames for other leaders; there are no
clear rules.
3. Yogendra Yadav, "The Third Electoral
System," Seminar 480: Cruel Choices (August 1999), p. 18.
4. Chosen by 29 percent of the respondents
(with only 3 percent selecting "religion/caste" as the basis for their choice). Outlook—Centre
for Media Studies poll, Outlook, August 16, 1999, p. 34.
5. See Yogendra Yadav, "Politics,"
in Marshall M. Bouton and Philip Oldenburg, India
Briefing: A Transformative Fifty Years (Armonk: M.E.
Sharpe, 1999) and Yogendra Yadav, "Understanding the Second Democratic Upsurge: Patterns of Bahujan Participation
in Electoral Politics in the 1990s," in Francine Frankel et al., Transforming
India: Social Dynamics of Democracy (Delhi: Oxford
University Press, in press).
6. "Survey data" throughout will
generally refer to polls conducted by the Centre for the Study of Developing Studies in Delhi, in 1967, 1971, 1996,
and 1998. The results were published at the time of the election in the newsmagazine India
Today in 1998, with analysis done by Yogendra Yadav
and his colleagues, and in Subrata K. Mitra and V. B. Singh, Democracy
and Social Change in India: A Cross-sectional Analysis of the National Electorate
(New Delhi, Thousand Oaks, London: Sage Publications, 1999). Page numbers noted in the text are from the latter
source.
7. See also Yogendra Yadav, "Politics,"
in Marshall M. Bouton and Philip Oldenburg, eds., India
Briefing: A Transformative Fifty Years (Armonk: M.
E. Sharpe, 1999).
8. See his article, "The Fractionalisation
of Indian Parties," Seminar 480: Cruel Choices (August 1999); Pradeep Chhibber notes the same phenomenon (Democracy without Associations: Transformation of the Party System
and Social Cleavages in India [Ann Arbor: University
of Michigan Press, 1999], pp. 80—81).
9. See my "Pundits, Pollsters, and a
Mandate to Rule," Journal of Commonwealth and
Comparative Politics (1988).
10. Outlook, August 16, 1999, p. 26.
11. See Table 2
in the Appendix for the strength of the various alliances in each state.
Specialists
Jagdish Bhagwati
Arthur Lehman Professor of Economics
Professor of Political Science, Columbia University
828 International Affairs Building
New York, NY 10027
tel: 212-854-6297
fax: 212-854-8059
e-mail:jb38@columbia.edu
Marshall M. Bouton
Executive Vice President
Asia Society
725 Park Avenue
New York, NY 10021
tel: 212-288-6400
fax: 212-517-8315
Pradeep K. Chhibber
Associate Professor
Political Science Department
University of Michigan
6611 Haven Hall
Ann Arbor, MI 48109
tel:734-763-2220
e-mail:pradeepc@umich.edu
Stephen P. Cohen
Brookings Institution
1775 Massachusetts Avenue NW
Washington, DC 20036
tel: 202-797-6015
fax: 202-797-6003
e-mail:scohen@brook.edu
John Echeverri-Gent
Associate Chair, Associate Professor
Department of Government and Foreign Affairs
University of Virginia, Cabell Hall, Room 257
Charlottesville, VA 23903
tel: 804-924-3968
fax: 804-924-3359
e-mail: jee8p@Virginia.edu
Ainslie T. Embree
10450 Lottsford Road # 1008
Mitchelville, MD 20721-7208
tel: 301-925-7208
fax: 301-925-7689
Francine R. Frankel
Professor
Department of Political Science
232 Stiteler Hall
University of Pennsylvania
Philadelphia, PA 19104
tel: 215-898-7648
e-mail: ffrankel@sas.upenn.edu
Sumit Ganguly
Professor of Political Science
Hunter College of the City University of New York
695 Park Avenue, Room W1705
New York, NY 10021
tel:212-772-5666
fax:212-650-3669
e-mail:sganguly@shiva.hunter.cuny.edu
Robert L. Hardgrave
Professor of Government and Asian Studies
University of Texas-Austin
Austin, TX 78712
tel: 512-471-5121
fax: 512-417-1061
e-mail:rlh@uts.cc.utexas.edu
Selig S. Harrison
Senior Scholar
Woodrow Wilson International Center for Scholars
1000 Jefferson Drive, SW
Washington, DC 20560
tel:202-287-3000 ext. 337
fax:202-357-4439
Atul Kohli
Professor of Politics and International Affairs
Woodrow Wilson School of Public and International Affairs
Princeton University
316 Bendheim Hall
Princeton, NJ 08544
tel:609-258-6408
fax:609-258-5349
e-mail:kohli@wws.princeton.edu
Philip K. Oldenburg
Southern Asian Institute
420 West 118th
Street, Room 1128
New York, NY 10027
tel:212-854-4565
212-927-0422
fax:212-854-6987
e-mail:pko1@columbia.edu
Lloyd I. Rudolph
Professor of Political Science
University of Chicago
5828 South University Avenue
P422
Chicago, IL 60637
tel:773-702-8056
fax:773-702-3854
e-mail:lrudolph@midway.uchicago.edu
Susanne Hoeber Rudolph
Professor of Political Science
University of Chicago
5828 South University Avenue
Chicago, IL 60637
tel:773-702-8055
fax:773-702-3854
e-mail:srudolph@midway.uchicago.edu
Howard B. Schaffer
Director of Studies
Institute for the Study of Diplomacy
School of Foreign Service
Georgetown University
Washington, DC 20057
tel:202-687-8974
fax:202-687-8312
e-mail:schaffeh@gunet.georgetown.edu
Shashi Tharoor
Executive Assistant to the Secretary General
United Nations
Room S-3802D
New York, NY 10017
tel:212-963-2912
fax:212-963-3511
e-mail:tharoor@un.org
Thomas P. Thornton
Adjunct Professor of Asian Studies
Georgetown and Johns Hopkins Universities
4214 Wickford Road
Baltimore, MD 21210
tel: 410-467-3429
Ashutosh Varshney
Associate Professor of Government
University of Notre Dame
217 O'Shaughnessy Hall
Notre Dame, IN 46556
tel: 212-631-7312
e-mail: Ashutosh.Varshney.1@nd.edu
Robert G. Wirsing
Professor of Government and
International Studies
University of South Carolina
Columbia, SC 29208
tel:803-777-2707
fax:803-777-8255
e-mail:rwirsing@scarolina.edu
Frank G. Wisner
Vice Chairman
American International Group, Inc.
70 Pine Street, 18th Floor
New York, NY 10270
tel: 212-770-5262
fax: 212-480-5400
|
|
