An examination of the Lok Sabha elections from 1951 to 2011 reveals that there are large discrepancies between the seat and vote shares of political parties in each of the 15 elected Lok Sabhas in independent India, especially before liberalisation in 1991. In the post-liberalisation period, the extent of disproportional representation has come down, but is still high, even as the extent of electoral competition has increased.

NOTES

How Representative Has the Lok Sabha Been?

Arun Kaushik, Rupayan Pal

have to choose between two contestants A and B. If A gets 51 votes then he wins. But the people who voted for candidate B are not represented in the political system. In fact, we often observe large discrepancies in the seat share and the vote share of political parties in both the Lok

An examination of the Lok Sabha elections from 1951 to 2011 reveals that there are large discrepancies between the seat and vote shares of political parties in each of the 15 elected Lok Sabhas in independent India, especially before liberalisation in 1991. In the post-liberalisation period, the extent of disproportional representation has come down, but is still high, even as the extent of electoral competition has increased.

Arun Kaushik is pursuing research at the University of Bologna, Italy. Rupayan Pal (rupayan@igidr.ac.in) is at Indira Gandhi Institute of Development Research, Mumbai.

t is often argued that democracy is the best form of governance, since it involves largest representation of general population. However, all democratically elected governments need not necessarily be equally representative of citizens of a nation. It crucially depends on the electoral system in place. More representative government is supposed to be better for any democracy. Therefore, it is important to examine the degree of representativeness of elected bodies. India boasts of being the largest functional democracy in the world. But, the question is, how representative is India’s Parliament?

India is a secular, democratic and sovereign republic with a parliamentary form of government. The government in India is federal in structure. At the national level, India has two main governing bodies – the House of the People (Lok Sabha) and Council of States (Rajya Sabha). Similarly at the state level there are two main bodies – Legislative Assembly (Vidhan Sabha) and Legislative Council (Vidhan Parishad). Members of Lok Sabha are the representatives of people chosen by direct election on the basis of adult suffrage. Normally, general election for Lok Sabha takes place after every ﬁve years throughout India.1 In any constituency generally several candidates contest the election and voters supposedly cast their votes according to their preferences over the candidates and/or the political parties.

Indian electoral system is characterised by the “winner takes it all” or the “ﬁ rst-past-the-post” (FPTP) system. According to this system, the contestant with the highest number of votes wins the election, no matter how many votes the winner got and what is the winning margin. For example, suppose there are 100 people voting for one seat and they

may 12, 2012

Sabha and Vidhan Sabha. However, the issue of disproportional representation in elected houses remains neglected, in spite of the fact that disproportional representation in highest lawmaking bodies may severely distort the formulation and implementation of policies in any nation.

We attempt to examine the extent of disproportional representation in Lok Sabha during the period from 1951 to 2009, using data for 15 general elections collected from the Election Commission of India. We construct the index of disproportional representation for each Lok Sabha of independent India, and analyse the variation in terms of disproportional representation in Lok Sabha over time. Moreover, we also examine whether all political parties have equally beneﬁted (or lost) due to disproportional representation.

We ﬁnd that there are large discrepancies between seat and vote shares of political parties in each of the Lok Sabhas, particularly during the preliberalisation era of the Indian economy. In the post-liberalisation era, disproportionality in Lok Sabha has reduced from that in the earlier period, but has remained sufﬁciently high. Interestingly, we observe an increasing trend in the extent of competition in elections among the political parties during the period of study, which has gathered momentum in the post-liberalisation period. Analysis also reveals that the discrepancy between shares of seats and votes largely favours the ruling party, whereas the opposition parties failed to reap the full beneﬁt of their voters’ base.

The ﬁndings of the article point out the limitations of the proposed electoral law reform in India. The focus of the Core Committee on Electoral Reforms (2010), as documented in the Background Paper on Electoral Reforms, is limited to some selected issues, such as

vol xlviI no 19

Economic & Political Weekly

NOTES

the criminalisation of the political system, ﬁnancing of elections, role of political parties and matters related to conduct and management of elections. There is no doubt about the need to bring appropriate changes in the electoral law in order to address these issues, but that does not appear to be sufﬁcient to pave the way for truly representative governments. The present study highlights the importance to change the FPTP system of representation appropriately. Otherwise, India may experience relatively more fair and free elections in future, but not adequately representative governments.

Political Swings at the Centre

The ﬁ rst ﬁve Lok Sabhas in India were dominated by the then single largest party, Indian National Congress (INC). The major change came only in 1977 elections when a non-Congress government was formed at the centre. However, the non-INC government could not continue for long. The INC came back to power in the next elections itself. According to Yadav (1999), the postindependence period can be broadly divided into three electoral regimes on the basis of electoral outcomes. When there is replacement of the old pattern of an electoral outcome with a new one, we may call that as regime change. For example, if a single party rules for a long period of time and suddenly the party loses in elections, it is termed as a change in electoral regime. In this sense we may call the period from 1952 to 1967 as ﬁrst electoral regime, since INC was the main political party which ruled most of the states and had the highest representation in Lok Sabha during that period. The 1967 Lok Sabha elections signalled the end of domination of INC. From a large representation of 361 seats of INC in Lok Sabha in 1962 elections, the representation fell to 281 seats in the Lok Sabha 1967 elections. Furthermore, there is a fall of 4% in the vote share of INC, from 44.72% in 1962 elections to 40.78% in 1967 elections. In the Lok Sabha elections of 1971 with the resurgence of Congress, winning 352 seats, it seemed that the INC regained its control. But in the subsequent elections in 1977 Bharatiya Lok Dal (BLD) came to power with a majority in seat terms in the Lok Sabha. Thus, it became clear that India’s political system was no more characterised by a single party dominance. The electoral system became multipolar. Soon after the assassination of Indira Gandhi, INC came into power with a very high majority, winning 414 (more than 76%) seats in Lok Sabha. Once again INC became the single most powerful party. But, similar to the previous rollback, the INC was voted out as the United Front won in the 1989 elections, which marked the beginning of the third electoral regime in India. The effect of fragmentation of larger political parties and emergence of various regional political parties in India started playing a crucial role in formation of government during the third regime, which left no other option but to have potentially unstable coalition governments in each of the Lok Sabhas starting from 1989 till date.

The Indicators

In this section, we describe the methodologies to quantify the extent of disproportional representativeness in elected bodies and intensity of competition faced by political parties in elections.

(1) Disproportional Representativeness: A political system may be said to be perfectly representative, if each party’s seat share in the elected body is equal to its vote share. On the other hand, if only one party gets all the seats without any voter’s support, the political system is not at all representative, i e, dictatorial. Following this, Gallagher (1991) proposes the following index of disproportional representativeness where sj denotes the number of seats of political party j in the elected body, vj denotes the number of votes received by the j-th political party and P denotes the set of political parties contested.

2

1 sv

 jj 

DISPR = 100

()  ()

 

2 jP  sk vk 

 kPkP 

Clearly, DISPR takes the value zero, if the political system is perfectly representative. On the other extreme, if there is dictatorship, DISPR takes value 100. Therefore, we have DISPR . It is easy to observe that, higher is the extent of disproportional representativeness, higher is DISPR. Therefore, DISPR can be considered as a measure of disproportional representativeness.

(2) Electoral Competition: One may argue that, if the number of political parties increases, competition would be higher. However, note that all parties are not necessarily equally strong. It is widely observed that there are variations across political parties both in terms of

(a) number of seats won, and (b) their voters’ support base. It indicates that increase in number of political parties contested does not necessarily mean increase in electoral competition. Relative sizes of the political parties should be taken into account while measuring electoral competition. The percentage of seats won or the percentage of votes received by a political party can be viewed as the size of that political party. Thus, the effective number of political parties (ENP), which is deﬁned as follows, can be considered as an appropriate measure of electoral competition (Laakso and Taagepera 1979).

2 –1 Sj

ENP = Σ

jP k

P

[( )]

In the above formula, sj denotes the number of seats won by the j-th party and P is the set of political parties. Clearly, ENP takes into account the relative sizes of the political parties. ENP can be interpreted as the number of hypothetical equal sized parties that would have the same total effect on electoral competition as have the actual parties of unequal size. It is evident that the lowest possible value of ENP is one, when only one party wins all the seats. On the other extreme, if all the parties win equal number of seats, ENP is equal to the number of parties. Clearly a higher value of ENP indicates more intense competition among political parties in election. A number of papers have used ENP as the measure of electoral competition (see for example, Chhibber and Nooruddin 2004; Bortolotti and Pinotti 2008).

Note that ENP can also be measured by considering share of votes received by a party as its size. In that case, we need

may 12, 2012 vol xlviI no 19

NOTES

to replace sj by vj, in the above formula, Figure 1: Trends of ENP, ENPv and DISPR: 1951-2009 where vj, denotes the number of votes 8

v

Needless to mention here that the inter-6

		ENPv





ENP	DISPR

pretation of ENP and its lower bound

v

and upper bound remains the same as

that of ENP based of seat share. However,

ENP & ENPv

ENP and ENPv

4

DISPR

seat share of all parties are the same. 2

This is because, in democracies like 1 India, distribution of votes among the political parties is generally quite differ-0

1951 1957 1962 1967 1971 1977 1980 1984

ent from the distribution of seats among the political parties. Prior to 1989, the total number of parties

contested in Lok Sabha election was Lok Sabha: 1951-2011 close to 35 on an average, thereafter the Let us now turn to examine the extent of average increased to as high as about disproportional representation in Lok 200 during 1989-2009. In fact, the total Sabha and the extent of competition number of parties contested in the 2009 among political parties. Table 1 reports Lok Sabha elections was as large as 364. the constructed indices – ENP, ENP and Based on this observation, one might be

v

DISPR – along with the total number of tempted to conclude that the average political parties and the number of na-intensity of competition in Lok Sabha tional level political parties2 contested elections during 1989-2009 became 462% elections in each of the 15 Lok Sabhas. It higher than that in pre-1989 period. is noteworthy that the number of nation-However, such a conclusion is incorrect, al political parties in India has remained since the average of ENP (ENP ) has

v

Table 1: Competition among Political Parties, Disproportional Representation and Number of Political Parties: 1951-2009

Year of Election	Lok Sabha	ENP	ENPv	DISPR	Total Number	Number of
					of Parties	National Parties
1951	1	1.78	4.06	22.61	54	14
1957	2	1.73	3.46	21.76	16	4
1962	3	1.84	4.18	21.54	28	6
1967	4	3.12	4.73	11.19	26	7
1971	5	2.12	4.48	18.67	54	8
1977	6	2.63	3.36	10.75	35	5
1980	7	2.16	4.17	19.86	37	6
1984	8	1.69	3.84	21.99	34	7
1989	9	4.11	4.74	8.40	114	8
1991	10	3.60	5.10	7.80	146	9
1996	11	5.82	6.92	8.09	210	8
1998	12	5.27	6.88	6.81	177	7
1999	13	5.86	6.70	9.16	170	7
2004	14	6.52	7.46	5.35	216	6
2009	15	5.00	7.59	8.23	364	7
Average in pre-1989 period	2.13	4.04	18.55	35.50	7.13
Average during 1989-2009	5.17	6.48	7.69	199.57	7.43
p value for t test for the difference	0.000	0.000	0.000	0.000	0.805
Percentage increase of average
for the period 1989-2009 from that
in pre-1989 period	142.23	60.70	-58.53	462.17	4.26

1989 1991 1996 1998 1999 2004 2009

increased from about two during the pre-1989 period to about ﬁve (six) during 1989-2009. The increase has been from four in the pre-1989 period to about six during 1989-2009 in vote share terms. It implies that the extent of electoral competition among political parties has increased at a much lower rate than the rate of increase in number of political parties. It seems to support the view that there has been signiﬁ cant “proliferation of non-serious parties”, which justiﬁ es the reformulation of criterion for registration and deregistration of political parties as proposed in by the Core Committee on Electoral Reforms (2010) in the Background Paper on Electoral Reforms. Nonetheless, note that the extent of competition among political parties has increased over time (Figure 1), which is supposed to be good for democracy compared to a single party being the pole against which all other political formations are deﬁned. But, increased competition led by substantial fragmentation of political parties, as observed during last two decades, can make it difﬁcult to form and run a stable government, as experienced during the third electoral regime in India. Therefore, it seems to be necessary to look at the mechanisms to regulate political parties and their conducts taking in to account both the positive and negative effects of multiparty political system.

Now, from column 4 of Table 1, it is evident that in each of the Lok Sabha in independent India there has been disproportional representation. In other words, India has never had a truly representative Lok Sabha. Needless to more or less stagnant over time, while the total number of political parties contested elections has an increasing trend over time with a major shift in 1989 (see the last two columns in Table 1).

increased by a much lower amount – from 2.13 (4.04) during pre-1989 period to 5.17 (6.48) during 1989-2009. In other words, effectively the number of equal sized parties based on seat share has

may 12, 2012 vol xlviI no 19

Economic & Political Weekly

NOTES

mention here, ideally DISPR should take the value zero. That is, in the ideal situation, the relative strengths of political parties in Lok Sabha should be at par with their support base reﬂ ected through the percentage of votes received in election. A positive value of DISPR implies that there is a disproportionality between the seat and vote shares of parties in Lok Sabha; i e, some parties have got proportionately more seats than the proportion of votes in their favour, while the percentage of votes was greater than the percentage of Lok Sabha seats for some other political parties. In other words, a positive value of DISPR indicates that the strength of some political parties in Lok Sabha was greater than their actual strength determined by the share of votes polled in their favour. That is, supporters of some political parties have democracy or not. This is because, as discussed before, increased competition among political parties may lead to unstable governments. Therefore, there is a need to address the problem of disproportional representation more directly, possibly by deviating away from the FPTP system of representation.

When there is disproportional representation in Lok Sabha, it is natural to ask which political party gains due to disproportional representation. Note that, if a set of political parties (or any main ruling party that has always beneﬁted from disproportional representation in Lok Sabha, starting from the very beginning. See Table 2 for the main ruling party’s percentage of seats in Lok Sabhas, percentage of votes in favour of the ruling party and the difference between the two.

From the above discussion it appears that, given the FPTP system of representation, it is not sufﬁcient for a political party to enhance its support base in order to form the government. Political parties need to take into account the

Table 2: The Main Ruling Political Party’s Seat Share, Vote Share and the Difference

Year of Election	Main Ruling Party	Seat Share (%)	Vote Share (%)	Difference between
				Seat Share and Vote Share
1951	Indian National Congress	74.28	44.99	29.29
1957	Indian National Congress	75.10	47.78	27.32
1962	Indian National Congress	73.08	44.72	28.36
1967	Indian National Congress	54.42	40.78	13.64
1971	Indian National Congress	67.95	43.68	24.27
1977	Bharatiya Lok Dal	54.43	41.32	13.11

been more represented, by getting dis-1980 Indian National Congress(I) 66.73 42.69 24.04

proportionately more number of Mem-1984 Indian National Congress 76.52 48.12 28.40

bers of Parliament. in the Lok Sabhas at 1989 Janata Dal 0.270 0.178 0.092

the expense of the supporters of other political parties, which is undesirable in a representative democracy.

Further, we observe that, though there are some variations in DISPR across the Lok Sabhas, it was close to 20 in most of the ﬁrst eight Lok Sabhas, with 18.55 as the average during this period. In the latter period, i e, in the third electo ral regime 1989-2009, there was a slightly declining trend of DISPR (Figure 1). Nonetheless, it remained sufﬁciently high, with the average DISPR close to eight, in the third electoral regime as well.

It is interesting to note that competition among political parties measured by ENP and ENPv has increased steeply during the third electoral regime, along with the steady increase in total number of political parties. The correlation coefﬁ cient between ENP and DISPR is also found to be large and negative (-0.88).3 It seems to suggest that increase in competition among political parties and disproportional representation are negatively associated. However, it remains ambiguous whether enhanced competition among political parties through appro priately designed mechanism to regulate political parties, in the hope of reducing disproportional representation in elected bodies, can be beneﬁ cial for

1991 Indian National Congress 45.69 36.40 9.29

1996 Bharatiya Janata Party 29.65 20.29 9.36

1998 Bharatiya Janata Party 33.52 25.59 7.93

1999 Bharatiya Janata Party 33.52 23.75 9.77

2004 Indian National Congress 26.70 26.53 0.17

2009 Indian National Congress 37.94 28.55 9.39

one political party) gain(s) due to disproportional representation, the other set of political parties (or at least one other political party) are (is) bound to lose. Therefore, the question is as follows. Does the (main) ruling party beneﬁ t from disproportional representation? Or, is it just the other way round? Analyses reveal that it is the political party with the largest seat share in Lok Sabha that gains from the disproportional representation at the expense of the political party with the second largest seat share. Moreover, as expected, the political party with the largest seat share was also the main ruling party in each of the Lok Sabhas, except in the ninth Lok Sabha. In the ninth Lok Sabha, which lasted for merely one and half years (December 1989-June 1990), INC was the opposition party in spite of being the party with the largest seat share (37.24%). Interestingly, in 1989 elections, INC received 39.53% of votes, which is higher than INC’s percentage of seats in the ninth Lok Sabha. Therefore, it so turned out that, it is the mechanism of government formation appropriately and attempt to reap the beneﬁt from disproportional representation. In other words, political parties can manipulate their election strategy to reap the undue beneﬁt of disproportional representation due to the existing FPTP representation system. It is worth mentioning here again that disproportional representation in elected bodies is detrimental for democracy. This is particularly so when there is disproportional representation in the highest law making body, Lok Sabha, of the country.

Concluding Remarks

This note addressed the following questions in order to examine the implications of the existing FPTP system of representation in India. How representative is Indian democracy? Is there any disproportionality between seat share and vote share of political parties in Lok Sabha? If yes, to what extent? Is there any relation between the extent of disproportional representation and the

may 12, 2012 vol xlviI no 19

NOTES

extent of electoral competition among political parties? Analysing data on the 15 Lok Sabha elections, which independent India so far had, we ﬁnd that India never had any truly representative Lok Sabha. The extent of disproportional representation, measured by DISPR, was quite large in each of the Lok Sabhas, though DISPR has registered a decreasing trend since the ninth Lok Sabha election in 1989. Moreover, it is found that the main ruling party in any Lok Sabha has always managed to get disproportionately higher percentage of seats in Lok Sabha compared to the percentage of votes polled in their favour in general elections. Therefore, in India it has always been the case that a preference from a minor section of the voters is dominant in the highest lawmaking body – the Lok Sabha – which is essentially due to the existing FPTP system of representation. Needless to mention here that such disproportional representation in elected bodies is undesirable in any democracy. Thus, there is need to reform the system of electoral representation in India.

This note also demonstrates that consideration of total number of political parties, which contested elections, as the proxy for electoral competition among political parties would lead to misleading results. It is observed that there is phenomenal upsurge in total number of political parties during last two decades. To be speciﬁ c, the average of total number of political parties contested Lok Sabha election has increased from 35 in pre-1989 period to as large as 199 during the period 1989-2009. However, the average of the number of hypothetically equal sized parties on the basis of seat share, i e, ENP, has increased from 2.13 in pre-1989 period to 5.17 in the post-1989 period. It justiﬁ es the reformulation of criterion for registration and deregistration of political parties as proposed in the Background Paper on Electoral Reforms by the Core Committee on Electoral Reforms (2010).

Interestingly, ENP has registered an increasing trend during 1989-2009, as opposed to the decreasing trend of DISPR during the same period. In other words, it seems that the extent of electoral competition is negatively associated with the extent of disproportional representation. However, it is likely to invite serious problems, if political parties are regulated in order to achieve higher ENP with the ultimate objective of reducing DISPR. The reason is, higher ENP may imply greater instability and/or more restricted functioning of the government, as observed during the last two decades in India. Therefore, it seems to be essential to address the problem of disproportional representation directly. To this effect, it seems to be necessary to shift away from the FPTP system of representation and adopt a carefully designed Proportional Representation system,4 which is also likely to induce the political parties to be more accountable to citizens of the country as well as strengthen Indian democracy by making it more ﬁrmly consociational (Lijphart 1996).

Notes

1 On the other hand, members of Rajya Sabha are elected by the method of indirect election, by the elected members of Vidhan Sabhas of states, in accordance with the system of proportional representation.

2 A political party shall be treated as a recognised National party, if, and only if – either

(A) (i) the candidates set up by it, in any four or more states, as the last general election to the House of the People, or to the Legislative Assembly of the State concerned, have secured not less than 6% of the total valid votes polled in their respective states at that general election; and (ii) in addition, it has returned at least four members to the House of the People at the aforesaid last general election from any State or States; or (B) (i) its candidates have been elected to the House of the People, at the last general election to that House, from at least 2% of the total number of parliamentary constituencies in India, any fraction exceeding one-half being counted as one; and (ii) the said candidates have been elected to that House from not less than three states. (Source: The Registration of Electors Rules, 1960 published by Ministry of Law, Government of India, New Delhi.

3 The correlation coefﬁcient between ENP and DISPR is -0.75.

4 See Zimmerman (1994) for a lucid discussion on implications of alternative system of representation in democracies from theoretical point of view.

References

Bortolotti, B and P Pinotti (2008): “Delayed Privatisation”, Public Choice, 136 (3-4): 331-35.

Chhibber, P and I Nooruddin (2004): “Do Party Systems Count? The Number of Parties and Government Performance in the Indian States”, Comparative Political Studies, 37(2): 152-87.

Gallagher, M (1991): “Proportionality, Disproportionality and Electoral Systems”, Electoral Studies, 10(1): 33-51.

Laakso, M and R Taagepera (1979): “The Effective Number of Parties: A Measure with Application to West Europe”, Comparative Political Studies, 12(1): 3-27.

Lijphart, A (1996): “The Puzzle of Indian Democracy: A Consociational Interpretation”, American Political Science Review, 90(2): 258-68.

The Core Committee on Electoral Reforms (2010): Background Paper on Electoral Reforms (Legislative Department Ministry of Law and Justice, Government of India, New Delhi).

Yadav, Y (1999): “Electoral Politics in the Time of Change: India’s Third Electoral System, 1989-99”, Economic & Political Weekly, 34(34/35): 2393-99.

Zimmerman, J F (1994): “Alternative Voting Systems for Representative Democracy”, PS: Political Science and Politics, 27(4): 674-77.

UNIVERSAL HEALTH COVERAGE

February 25, 2012
Universal Health Coverage in India: A Long and Winding Road	– Gita Sen
Thailand’s Universal Health Coverage Scheme	– Viroj Tangcharoensathien,
	Rapeepong Suphanchaimat,
	Noppakun Thammatacharee,
	Walaiporn Patcharanarumol
Medicines for All: Unexceptionable Recommendations	– S Srinivasan
Political Challenges to Universal Access to Healthcare	– R Srivatsan, Veena Shatrugna
A Limiting Perspective on Universal Coverage	– Rama V Baru
Human Resources in Health: Timely Recommendations,
Some Lacunae and What about Implementation?	– George Thomas
Gender in the HLEG Report: Missed Opportunity	– T K Sundari Ravindran,
	Manju R Nair
In Pursuit of an Effective UHC: Perspectives Lacking Innovation	– Padmanabh M Reddy
For copies write to:
Circulation Manager,