Sunday, February 22, 2009

Electoral reform


A recurring feature of recent Israeli governments is the presence of many small parties in the coalition. No single party ever has enough Knesset seats to form a government on its own. The largest party must always recruit an array of smaller parties, trying to balance their divergent agendas and distribute authority in a way that pleases all of them. Much of the prime minister's effort is spent trying to preserve the coalition rather than actually running the country. Inevitably, after a couple years this effort fails and the country is forced to premature elections. With all the effort spent on struggles for power, not enough time is left to fully confront serious problems which the country faces.

These difficulties have been the subject of much discussion, and occasional legislation, in recent years. Three basic proposals have been suggested to deal with the problem. In this post I plan to explain how each of those proposals would actually cause more harm than good. Then I will suggest a new approach, which would likely possess the advantages of the other proposals but not their disadvantages.


The first proposal is to create a two-party system where one party, or one prime minister/president, runs the government at a single time. Unfortunately, Israeli society is too fragmented to be effectively represented by two parties. In particular, Arabs and charedim would end up having to vote for parties which, respectively, launch wars in Gaza and try to institute civil marriage. Neither group would have a meaningful vote, since neither of the two parties could be broad enough to accomodate their views. We like to talk about how Arabs have more democratic rights in Israel than in any Arab country. But if they can only vote for candidates they do not agree with, in practice they would be disenfranchised. The 1990s violence of the US extreme right wing (Timothy McVeigh, Ruby Ridge) shows what happens when people feel that nobody represents them. In Israel, dealing with Muslims rather than Christians, the response would likely be much more extreme.

The second proposal is to raise the threshold for admittance to the Knesset, so that only large parties received seats and coalition negotiations would be simplified. Depending on the level to which it was raised, this would be either completely ineffective, or else catastrophic for democracy. If the level was low, the small parties would simply unite with each other to avoid missing the threshold (as two charedi parties did the last time it was raised). If the level was high, then even the united parties would fall below the threshold. Groups such as Arabs, charedim, and the national-religious would be mathematically prevented from electing a party of their own to the Knesset. Thus the second proposal would lead to outcomes just as dangerous as the first proposal.

The third proposal is to divide Israel into districts and elect one Knesset member from each district. Thus, it is hoped, parties which receive a few percent of the vote would be eliminated. However, practically speaking, such a system would eliminate meaningful voting for much of the country. In every election Jerusalem would vote for the right wing, Tel Aviv for the left wing, Bnei Brak for a haredi party, Um El Fahm for an Arab party, and so on. Only a handful of districts would ever be competitive. It would be like the electoral college system in the US, where nobody's vote counts except in a few swing states.

In Israel, the social consequences of such a system could be even more serious than the moral consequences. In the US, social trends span the whole country, so most people identify with the election results even if they didn't contribute to them. But Israeli society is more fragmented and each locality more homogenous. Predictable local elections would lead to indifference and alienation among those who feel little connection to voters in other districts. Which brings me to another crucial objection: gerrymandering is inevitable when the geographic boundaries between population groups are so clear.

Besides eliminating single-interest parties, the other predicted advantage of district elections is that politicians will be held responsible for the particular needs of their constituents. Such accountability is certainly an advantage (though with Knesset districts, it would come with the disadvantage of pork-barrel spending). But in fact such accountability already exists. Each city already has mayoral and city council elections which often are bitterly contested over local issues. All that is needed is to delegate a few more local issues from Knesset to municipal control, and the benefits of the district system can be achieved without its disadvantages.


If each of these three proposals would have side-effects that are likely worse than the current difficulties, then what alternatives are available? I think the following approach can likely solve the problem of unstable governments, while avoiding the side effects of the other proposals.

Imagine that 110 Knesset seats were elected using the current proportional method, and the remaining 10 were given to whichever party received the most votes overall. The fringe parties would receive almost exactly the same representation they have now. But nearly all Zionist democratic votes would go to two or three parties with a shot at winning overall. Those parties would be larger since they would include current small parties, and with the 10 extra seats, the winner would have no trouble forming a stable government.

Of course, the choice of exactly 10 bonus seats is debatable - I guessed at it, and a different number might be preferable. If the number is too small, the benefits of the change will be minimal. If the number is too big, then one party could gain overwhelming control of the Knesset and do all sorts of things which the country as a whole does not support. The number might need to be tweaked after an election or two, when it is clear how exactly voters and politicians react to the changes. Unlike in the systems discussed above, in this system the amount of change to be instituted is easily adjusted.

The fundamental question we face is how to strengthen the "top" of the system - the party forming the government - without undermining the "bottom" of the system - the current existence of equitable representation for the whole spectrum of political opinion. My proposal is to give a focused gift to the "top" while keeping the "bottom" unchanged as much as possible. Unlike other government reform proposals commonly suggested these days, this proposal should preserve the advantages of the current system, while hopefully eliminating the inefficiencies which have generated so much debate.

1 comment:

Ben said...

Your critique of the alternatives is on target. However, you might be underestimating the strategic effect of your proposal. Since ten bonus seats is a huge payoff, there is a big incentive to vote for a party that has a chance to be largest. So it's very likely that many voters will abandon the smaller parties for the large ones. In that respect, this proposal is similar to (though, much better than) the one being pushed by the IDI in which the head of the top party is automatically Prime Minister. (They have an additional incentive, besides killing the small parties: Since the right is split between secular and religious, the largest leftist party has a good chance to be the largest overall party, even though the right has more overall support (for example, the current situation)).

I hope to post on this topic tomorrow.