Monday, July 04, 2005

Want to amend the B.C. Elections Act? Re-think these three assumptions

Now that the dust has settled from the referendum debate on electoral reform, politicians in Victoria will soon have to decide whether or not to sweep it under the rug. And that is a real option, despite the strong 58% vote in favour of BC-STV last May. The reason is that, although most people will prefer change if you offer it to them, the intensity of this preference is weak. The issue is simply not a burning one for most British Columbians, and it has been further dampened by the result of the May general election in B.C., which yielded a "normal" legislature with both a strong government and a strong opposition.

So, as long as Premier Campbell is willing to cede his place in history as a pioneer of electoral reform (as a majority of his caucus colleagues and not a few opposition members are hoping he will), there is little chance that the momentum for change will be sustained. But, if we were to act, how should we proceed? It should be remembered that the debate thus far has been conducted mostly within the ground rules of the B.C. Citizens' Assembly on Electoral Reform, on the basis of three assumptions:

1) The B.C. Legislature shall continue to have 79 constituencies (ridings).
This assumption was agreed upon by the Citizens' Assembly for a very simple, and very sensible political reason: it was felt that if the Assembly recommended more than 79 ridings, the referendum would be decided by an antipathy toward having more politicians working in Victoria. In order to give its recommendation a good chance of succeeding, the number of constituencies had to be frozen. The difficulty with this assumption is that the population of British Columbia has grown by about half a million souls since the last census in 1996. Simple math tells us that in order to maintain existing levels of per-capita representation (e.g. constituencies of roughly 50,000 people), we need to expand the Legislature by at least ten seats. Since most of the population growth has been in the Lower Mainland, constituencies in the more sparsely populated north and interior will have to be expanded and stretched just to stay within legal limits (which state that constituencies should not deviate more than 25% from the average.) Put simply, we need to add a number of seats just to maintain absolute levels of per capita representation, especially outside of Vancouver.

2) The voters shall only be offered one recommendation by the Citizens' Assembly.
This assumption was built into the terms of reference for the Citizens' Assembly. Understandably enough, it helps to make the referendum decision a manageable one, between the status quo and the one best system as decided upon by the Assembly. (There was inadequate explanation and understanding of these two options in the spring election; this difficulty would have been compounded by a multiple-choice question.)

Nevertheless, it is possible to design a two-part referendum in which the first question asks voters whether they would like a new electoral system that offers them greater choice and greater proportionality of result (briefly explaining what that means); and then asks them to choose between two models or rank three models in order of preference.

3) A 60% threshold is required for the ratification of the recommended proposal.

This requirement was added by the government, along with the regional threshold of at least a majority of voters in 60% of ridings, probably at the urging of a Liberal caucus who feared radical change. It has engendered some curious debate among political commentators. Those critical of the super-majority have wondered why 50% plus one has been sufficient to decide prohibititon, conscription, Newfoundland joining Confederation, and Quebec sovereignty, but not B.C.'s choice of electoral system. Others (most prominently Norman Spector) have stressed that BC-STV is "constitutional" in that it is a radical proposal that fundamentally alters the ground rules of political life; and that, legally, The Secession Reference decided that a simple majority vote was not sufficient to determine Quebec sovereignty.

In retrospect, Dr. Spector was probably right to regard BC-STV as radical enough to justify a higher threshold, although the second part of his argument (that secession of Quebec would require a consitutional amendment by 2/3 of theprovinces representing a majority of the population) is a tad too legalistic--we all know that if 58% of Quebeckers ever vote for sovereignty, they will be gone! In any case, the precedent has been set in BC that for changes as radical as BC-STV, an over-all super-majority of 60% is required. This leads to an interesting question for reformers, and for the Citizens' Assembly, if they are ever reconstituted: would they be willing to accept a less radical proposal than BC-STV, in exchange for a simple majority threshold?

Where We Are Now?

In the past three years, British Columbians have:
  • rejected the existing First-Past-the Post system, by an almost 58% margin;
  • rejected BC-STV , by failing to ratify it by 60%;
  • rejected the kind of strong party-list PR proposed by Adriane Carr and the Green Party, both when Carr attempted a Citizens' Initiative in 2002 and when The Citizens' Assembly explicitly considered and rejected this kind of proposal.

I have suggested in previous columns that these results point us toward a manageable number of general options for meaningful electoral reform. One would be a moderate Mixed-Member Plurality system that the Citizens' Assembly felt constrained not to choose, because of the Assembly's anti-party populism, the greater voter choice embodied in STV, and because the Assembly felt that it could not make MMP work satisfactorily within a total limit of 79 seats. A legislature of 90 seats, with between 1/4 and 1/3 of the seats allocated to 4-6 regional, open-list multi-member consituencies, would provide a modest injection of proportionality that would both raise the threshold for the achievement of majority government and lower the threshold for third parties, without throwing the door wide open to a factionalized parliament.

Another option would be a more moderate version of STV that is less proportional than BC-STV but which would still make "every vote count". In a 90-seat legislature, one could simply have the northern half of the province represented by 6 dual-member constituencies and the southern half of the province represented by 26 three-member districts. This 90-seat "STV-Lite" proposal would reduce the problems of "monster ridings" in the north and "monster ballots" in the south under BC-STV. Majority governments would be harder to come by under STV-lite than they are now, but they would not necessarily be rare. They would certainly have to be more inclusive and consensual than they are now, however.

If "regional open list" MMP-lite and "STV-lite" were the only reform options on the ballot, the argument could be made that the requirement of a 60% over-all vote is not necessary. A two-part ballot could be constructed asking voters first, if they favour a form of proportional representation, and, in the second part of the ballot, which of these two forms that they prefer. (BC-STV could be added to this list, but precedent would make it impossible to argue for a 50% threshold, and a clear majority decision will be less likely to result. )

Some have argued that such talk of abandoning BC-STV is simply throwing the debate wide open again; it is to open Pandora's box. Asking the Legislature to enact these or similar recommendations is to invite either the reality or the appearance of conflict of interest that the Citizens' Assembly was set up to avoid. I have tried to argue that the road we have travelled over the past three years has led us to this point; we only need to take the next step.