01 August 2014
  · New Issue
  · Archives
  · Subscribe
  · Campaigns
  · Press Room
  · Media Centre
Women's Reservation: Another Approach
Mukesh Dalal

The Women’s Reservation Bill is currently caught in a deadly stalemate. Nevertheless, the idea of affirmative action to enhance the participation of women in our legislatures is finally getting to be debated  in terms of exploring various options and alternatives which will avoid the pitfalls of a lottery based, territorially reserved, rotating quota of seats for women. Over the last four years, MANUSHI has submitted three different options: a) Multi seat constituencies,  b) Dual member constituencies, c) Party based quotas in ticket allocation (MANUSHI  96, 97, 107, 116). Based on the feedback received, we prepared a comprehensive Alternative Women’s Reservation Bill which was introduced in MANUSHI 116. This has been endorsed by numerous women’s organisations, activists and other concerned people, and has also led to widespread debate on the subject all over the country.

Now we present another proposal by Mukesh Dalal which he claims will be far more effective than any other suggested so far. Many of our readers are likely to find it too complicated and problematic in actual implementation. Since we think it is important to keep our minds open and engage seriously with various suggestions being made for improving the Bill, we invite our readers to respond to his proposition. -Editor


Despite several years of national consensus on women’s reservation, we have miserably failed to put it into law. I will now present a novel proposal for women’s reservation that is based on two key ideas. The first key idea is to create some extra seats that are not assigned to any specific constituency. These “quota seats” will be filled only when there is a need to increase women’s representation, that is, whenever the number of women “elected directly” falls below the desired “women’s quota”. The second key idea is that women candidates who were closest to victory (defined shortly) in constituencies not already represented by women will then be “elected by quota” to fill the appropriate quota seats. For all purposes, women elected by quota will be treated on par with men and women who are elected directly.

For example, as explained in Diagram I, suppose it is decided to create 50 quota seats in the Lok Sabha and set the women’s quota at 181, which is one-third of 543, the total number of Lok Sabha seats, we can envisage three different scenarios as illustrated below, depending on the number of women getting directly elected:

Case 1: Only 44 women get directly elected, causing a shortfall of 137 from the women’s quota of 181. In this case, all the 50 quota seats will be filled, by increasing the strength of the house to 595, including two nominated seats.

Case 2: 150 women get directly elected, causing a shortfall of only 31 from the women’s quota. In this case, only 31 quota seats will be filled, increasing the strength of the house to 576.

Case 3: 200 women get directly elected, exceeding the women’s quota. In this case, no quota seat will be filled, keeping the strength of the house to 545.

Although “closest to victory” may be defined in several ways, the most reasonable is to use the fractional margin of loss.  For example, a woman losing by 5,000 votes out of 200,000 valid votes will be chosen over a woman from another constituency losing by 4,000 votes out of 150,000 valid votes. This “normalization” by total votes ensures that all constituencies are treated fairly, irrespective of their size. Women closest to victory are those who are most likely to be directly elected in the next elections.  It may be noted  that using this criterion is now extremely easy because of the advances in information technologies.

Some obvious advantages of my proposal, in contrast to the previous proposals, are as follows:

Ø It will achieve reservation for women without discriminating against any men in any constituency. Men will not be undemocratically and unfairly prohibited to contest from any constituency and every voter will get the democratic right to vote for either a man or a woman. Thus, it should not cause any resentment against women.

Ø Any women candidates in any constituency will be eligible to be elected on quota seats. This will avoid undemocratic discrimination against women contesting from constituencies not reserved for women.

Ø The quota seats will be filled only when the direct elections do not result in adequate representation for women. This adaptive approach should not distort the direct elections, except perhaps by providing an additional motivation to women candidates to campaign harder.

Ø Only women who are closest to victory will be elected on the quota seats. This will improve the possibility of their directly winning the next elections and making the quota seats available to women from other constituencies.

Ø The representation of women will gradually increase over several elections, avoiding unacceptable disruptions caused by a sudden artificial upsurge. This will also improve the quality of women representatives, since they will mature in a natural manner.

Ø The representation of women will eventually increase to the desired level, so that filling seats through quotas will no longer be required.  In the above example, if we assume that the number of directly elected women increases each time by half the number of women elected by quota, then the desired representation of women will be achieved in 5 elections (94, 119, 144, 169 and 194, respectively – quota seats may be increased to get there sooner). Thus, this reservation system will liquidate itself over the long-term, instead of getting permanently entrenched.

Ø It will motivate political parties to nominate 'winnable' women candidates, since this is the best way to also ensure their election by quota. A party that does not nominate 'winnable' women candidates will  lose all quota seats to parties that do nominate 'winnable' women candidates.

Ø It will motivate people to vote for women candidates. A vote for a woman will be more “effective”, since it may get two shots at electing a winner, instead of the standard one shot given to a vote for a male candidate. Note that this discrepancy will vanish as soon as women get adequate representation.

Ø It will motivate all sitting MPs (elected directly or by quota) and potential candidates to nurture their constituencies on a long-term basis, since they will never be prohibited from contesting there.

Ø It will not lead to women contesting and winning only the constituencies reserved for them, as has happened in the SC/ST case. Moreover, women will not be artificially restricted to contesting against women only, stunting the growth of their leadership qualities.

Ø It will easily get the support of the current MPs, since their seats will not be threatened.

Some potential drawbacks of my proposal are as follows:

  • q It will increase the strength of the Lok Sabha, unless the number of constituencies is reduced. Recall that Justice Rajinder Sachar has already recommended increasing the strength of the Lok Sabha to 750. Although reducing the number of constituencies may be opposed by current MPs, they will prefer it to the government’s bill for two reasons – this reduction will be much less than a third of the total strength and MPs will be eligible to contest from their own, but possibly larger, constituencies. Note that the 2001 census will provide us with a fresh opportunity to revise constituencies.

  • The actual strength of the Lok Sabha will not be fixed in advance – it will depend on how many quota seats are filled based on the election results. This is not without precedence – the President is currently allowed to nominate up to two Anglo-Indians if their representation is deemed inadequate. This variability will be caused only after women’s representation gets close to adequate, that is, only after the problem we are trying to address gets solved!

  • A man-woman pair contesting a constituency may attempt to engineer the outcome so that the man gets directly elected and the woman gets elected by quota. This will not happen unless the man is willing to risk losing, a very unlikely choice.

  • Some constituencies will have two representatives, potentially causing some conflict between them. Since their competition may actually benefit the constituency, this potential advantage may even encourage people to vote for women candidates. Moreover, like the above case, the risk of losing even the quota seat will prevent a deliberate strategy to create a runner-up out of the more popular woman candidate.

  • A man-woman pair may keep getting elected from the same constituency. This is not likely to happen very often and may not even be undesirable if it happens.

  • It is completely untested. I am not aware of any election anywhere in the world where such an approach has been tried. This proposal is currently just a product of my imagination.

My proposal naturally extends women’s reservation to constituencies reserved for SCs and STs, by requiring that the quota seats and women’s quota be distributed proportionately. In the above example, suppose 16 per cent constituencies (86 of 543) are reserved for SCs and 8 per cent (43 of 543) are reserved for STs. The 55 quota seats will get distributed in the same proportion: eight for SCs and 4 for STs. The women’s quota of 181 will also get distributed similarly: 28 for SCs and 14 for STs. Now suppose the 44 directly elected women include 12 from SCs and two from STs, it follows that out of 50 women elected by quota, eight will be SCs and four will be STs (see Diagram II).

In general, the constituencies may be grouped in any manner, before or after the elections, as long as the quota seats and the women’s quota are also grouped proportionately. This principle enables us to also handle the situation when results get delayed, for whatever reasons, in a significant number of constituencies. The delayed constituencies can then be grouped together so that women’s reservation may be applied to them as described above for SC and ST groups.

This technique may not work well, if a very small number of results are delayed, for example, fewer than 10 (this cutoff needs to be carefully selected).  For example, if only three results are delayed, then a better approach will be to delay filling three quota seats until those three results are also declared. A description of some other natural extensions of my proposal, for example, allowing multiple quotas for overlapping groups, is available in a longer document that also argues for its use in most kinds of reservations.

I believe that there are good arguments against the very concept of reservations for disadvantaged groups – they are inherently discriminatory, they encourage resentment against these groups leading to more disadvantages, and they get perpetually entrenched.

However, I do not know of any other effective way to nullify the effects of centuries of injustice suffered by some of these groups. I think of reservation as a necessary evil, which has been accepted by an overwhelming majority of Indians. Our continuing challenge is to either find a better way to ameliorate these disadvantages or to find the least unpalatable forms of reservations. My proposal aims to do the latter over the short term, while leading to the former over the long term.

In conclusion, I propose the following course of action. The Election Commission, in consultation with the government and the political parties, should come up with the acceptable numbers for women’s quota, quota seats and constituencies, for the Lok Sabha and all the Vidhan Sabhas.

These numbers should be used to draft a new women’s reservation bill based on this proposal, which should then be tabled in the parliament for debate and approval. In parallel, I hope to see a constructive public debate for removing the remaining wrinkles from this proposal.

Manushi content is reproduced on India Together with permission. Click here to visit the Manushi home page