With the approval of the President’s Cabinet Paper, it has been decided to adopt an entirely new Constitution, not to tinker any more with the 1978 Constitution as we have been doing so far through 19 Amendments.
Those engaged in the process of electoral reform, which began in the 1990s but gathered considerable momentum in the failed attempt to enact the 20th Amendment earlier this year, must now step back and situate their efforts within the larger context of Constitutional reform.The President and the Prime Minister have indicated that the executive presidency will not be part of the new Constitution. This, in itself, will require a complete rethinking of the electoral reform options.
The Proportional Representation (PR) system of elections introduced to Sri Lanka by the 1978 Constitution is far superior to the First-Past-the-Post (FPP) system in reflecting the people’s will. It does not yield results like those at the recent British election (conducted under FPP), where a party that gained 3.8 million votes obtained only one seat in Parliament while another which attracted 1.5 million votes obtained 56 seats.
What we were trying to do through the 20th Amendment was to remedy the problems caused by the Preferential Voting segment introduced in 1988, not to get rid of PR. But the proposed abolition of the executive presidency requires us to re-examine the basic features of the PR system.
PR without a strong presidency
PR tends to proliferate political parties and produce coalition governments. From Independence to 1978, there were no Muslim parties as such. In addition to the two main parties, all we had were one or two parties seeking to represent the geographically concentrated Tamil populations in the North and the East and two or three parties seeking to represent the geographically concentrated ideological minorities in the Wet Zone.
The principal parties had within them a significant minority representation and put forward candidates of different ethnicities. The FPP system created incentives for geographically dispersed minorities such as the Muslims to seek representation through the principal parties and for no-contest pacts and coalitions. Since we started electing legislators through the PR system, and especially since the threshold was lowered to five per cent, parties have proliferated. Few incentives exist for minorities, ethnic as well as ideological, to seek to work through the principal parties. At most, they have an incentive to form small parties and then negotiate terms for joining coalitions before or after elections.
PR-based elections have yielded a winning party/coalition with a majority only twice, in the atypical years of 1989 and 2010. At all other national elections, it has been necessary to cobble together ruling coalitions after the elections. The President/Prime Minister has had to provide various incentives to the coalition partners including, but not limited to, lucrative Ministries.
The Sarath Silva decision on permitting crossovers of MPs elected through party lists weakened the control of party leaders over their members.Without an independently elected executive President, constituent parties within coalitions and factions within political parties will gain even greater leverage.
Local government illustrates the problem. After the recent amendments to the local government legislation, it was found that members were making certain local-government authorities ungovernable. If a Mayor or Chairman took action against some illegal action such as the filling of wetlands, members would gang up to defeat the budget.
Giving vetoes to the small groups and factions over the conduct of government is not conducive to effective governance. It breeds corruption.
PR with a high qualifying threshold
This problem has been addressed in many PR systems by setting a high qualifying threshold. The 1978 Constitution set the minimum threshold at 12.5 per cent. That meant that any party that failed to win at least 12.5 per cent of the votes in an electoral district would not be entitled to any seats. The parties that had votes above the threshold would, as a result, gain a proportion of seats that was greater than their share of votes.
While this appears“unfair” to small parties, it must be remembered that it is still better than what prevailed under FPP. Under FPP all votes cast for candidates other than the one who came first were completely devalued. Under PR with a high minimum threshold, it is true that parties such as the SLMC and the JVP may not be able to elect as many MPs as they do now. But a high threshold will create incentives for different behavior by both the ideological/ethnic minority parties and by the principal parties. The former will have more reason to integrate themselves with the principal parties. The national parties will have incentive to accommodate more minorities within their “big tents.”
PR with a high qualifying threshold will not only address the problem of governability and reduce the structural incentives for corruption. It will also encourage the emergence of national parties that transcend ethnic and narrow ideological identities. The need to find common ground could also contribute to the desperately needed inner-party democratization of political parties.
The chances of raising the threshold are slim. Parties that have emerged as a direct result of the low-threshold PR system are certain to oppose changes that would lead to their demise.
The manifestoes of both the President and the UNP specify the solution will be a hybrid one. We should look at the experience of Germany, Japan, New Zealand and Taiwan, among others. But they differ from us in terms of political culture. What works there will not necessarily work here.
A principal rationale for electoral reform is the removal of structural factors giving rise to corruption, by for example reducing the need to campaign over large areas and by bringing back election petitions and by-elections. The revival of by-elections also provides a solution to the cross-over-for-money problem. MPs will be free to vote according to their conscience and not be puppets of party leaders, which would be the case if cross-overs are completely prohibited as in Bangladesh. Under a hybrid system, a sizable number of MPs would have been directly elected. Therefore, they can always ask the people who elected them to ratify their crossover decision through a by-election.
But in a hybrid system, there would be MPs elected through PR for whom by-elections are not feasible. There is no alternative to prohibiting crossovers for them.
But the trick is to manage the ratio. During the discussions in the early part of 2015, we found that majority governments were more likely if significantly more MPs were elected by FPP than by PR, e.g., 125 through FPP versus 100 through an additional person’s list based on PR of some form. The clearly articulated desire by the UNP and several other parties to keep the number of MPs at or below 225 provides the opportunity to evolve a uniquely Sri Lankan solution.
Earlier, the problem was that of ensuring an outcome similar to that of the current system which yields minority governments mostly. But if the requirement is for a system capable of yielding majorities, it is possible to have a reasonable number of FPP electorates in a hybrid system that caps the number of MPs at 225. Such a system could accommodate additional persons nominated by the political parties, including women and also candidates who did well in the FPP contests despite not coming first.The concerns of the geographically dispersed ethnic minorities can be addressed through the judicious design of multi-member constituencies that can be periodically recalibrated by the Delimitation Commission. Singapore makes extensive use of multi-member constituencies. We ourselves have considerable experience with them.
In the context of the abolition of the executive presidency, we need to come up with an electoral system grounded in our political realities without importing solutions from other countries where political conditions are different. The elements of such a solution may be found in the report of the Dinesh Gunawardene select committee.