Looking at the returning results from General Elections in the United Kingdom, I was interested in the idea of ‘University Constituencies‘ where seats are not particularly allocated on a geographical basis, and what is more important to my mind, targeting a particularly homogeneous but ultimately illusive electorate.
As numbers stand now, first-time voters are at dangerously low levels, both as a percentage of total registered voters and in their respective cohort. The danger being, short of a profound reform of electoral rolls, an increasing trend in disenfranchising voters, thus subverting the electoral process and the very idea of representative democracy. The fact the ruling party in the current coalition government has seemingly dropped their support for electoral reform has pretty much precluded any push for renewal in the electoral rolls. Recent comments from the Interior Ministry suggest little or no change should be expected on that front.
As it stands now, the Moroccan parliamentary system exhibits a two-tiers system in its upper house: 305 members of parliament are elected on the local ballot, while 90 are selected on the basis of national results, 60 seats of which are allocated to female-only lists, and 30 for de facto young (under 40) males. Yet these members do not have a constituency, which is both a blessing and a curse: on the one hand, they are free to pursue whatever cause they fancy with no fear of backlash from their hypothetical voters, but on the other hand, they are beholden to their party leadership, for fear of being deselected, or worse still, put at the bottom of the party list, where there is little hope of taking up a seat.
There is also the apropos argument these elected representatives are not “full” members has they do not have a mandate: a recent opinion handed down by the designate constitutional court, striking down the provision of a Women’s Caucus suggests otherwise, since their argument was based on the fungibility of members of parliament, i.e. they are not subject to community allegiances whatsoever. The same line motivated another opinion on the continuity of government as well.
Members of the Court have honoured the French tradition of their curriculae, since by denying the individual qualities of each of the 395 members of parliament, they assert the idea of a homogeneous nation, whose representatives are no longer bound to the local constituencies that got them there. I fear this view is widely shared across the political spectrum, and does weaken the reformist claim many (including political opinions I am partial to) herald as their own.
Let us go back to the idea of university constituencies: obviously the first order of business in parliamentary reforms is to reduce the number of seats, so as to flatten regional discrepancies: sparsely populated areas tend to be allocated more seats per capita than, say metropolitan regions, which gives undue advantage to some parties and candidates, as well as produce counter-intuitive results, for instance during the 2007 elections, where PJD had a slight advantage over the Istiqlal on the popular vote, but was eventually a good dozen seats behind.
Seat allocation does not seem to obey a specific, let alone transparent rule: on the eve of each general election, the same ballet is performed by the Interior Ministry, in charge of rewriting electoral regulations, and political parties, each with grievances that often translate in patchy compromises the current ballot system does nothing to alleviate: as a result, there is virtually no chance one party could get hold of an absolute majority and form a government on their own.
Nonetheless, a simple rule can be adopted for all future distributions of seats, the statistical distribution of voter registration allocates seats per province, with a minimal number of two per constituency, historically close to 43.000 per seat. This system has the benefit of reducing the local-ballot seats by 51 seats, distributed as follows:
|Region||2011 Seats||Reform||Uni Seats||Net Change|
Densely populated regions lose comparatively fewer seats compared to Southern and rural regions, but the results greatly reduces discrepancies of past elections. Furthermore, in the context of a first-past-the-post system list, the requirement of having at least a female candidate for all competing candidate list ensures a minimum 92-strong female caucus, a 36% female representation, double the current 17%, and in line with female labour participation.
The younger generation also need not be granted a quota: university constituencies can serve the double purpose of expanding the electoral roll, as well as provide voting incentives to an otherwise disaffected population: there are about 600.000 Moroccans registered at universities, vocational/occupational schools, institutes and other high-education facilities, many of whom far from their home towns. Also, instead of having 30-odd members with no fixed constituency to answer for, the Moroccan youth will have a chance to elected their representatives on their own terms and rights. As a result, Parliament would look radically different:
the number of sitting members would be cut from 395 to 269, with 15 University Seats, and at least 92 Female seats, since nothing precludes female university candidates, or winning lists with more than one female candidate.
This is the case for most liberal democracies: there is no actual need for an absolute majority of popular vote, and depending on ballot system, a relative minority can actually allow for a political organization to get into office. This is due to no impediment to free vote, but that’s how it work with representative democracies: whatever the ballot system, some votes are bound to be lost in the process, be it because of turnout, how majorities are required to be formed, or the structure of political systems (bi-partisan or multi-party system)
Consider the 1997 General Elections in the United Kingdom, widely considered to be a New Labour landslide -and it was, in every sense of the term: Labour managed to command a super-majority of 179 (i.e. 63% of seats) in the House of Commons with 43.2% of the votes. In fact, under the First Past The Post system in the UK, the winner needs not to carry 50% plus one vote in their constituency; all they have to do is carry one additional vote to top up their second.
So goes another landslide election in the UK :1983 saw Margaret Thatcher lead the Conservative Party to a large victory, with a majority of 144 (61% of the seats) but only 42% of popular votes. Consider another Parliamentary monarchy: the Spanish Elections of 2004, considered by many observers as a stunning upset for PSOE, were won with 42% of popular votes.
But then again, it is always possible to look for a smaller minority to win – as a matter of fact, there is an interesting quantitative measure that could help to rank ballot systems likely to deliver big majorities in representative institutions, but ultimately won on small relative majorities (and independent of any turnout figures): the Gallagher Index (a variant of the Chi-Squared statistic) is indeed very useful to illustrate my point; First Past The Post (FPTP) as the index points out, tends to magnify majorities – thus widening the gap between parliamentary majority and popular votes. And yet, FPTP is perhaps the most indicated ballot system for Morocco to make sure the smallest possible coalition; benefits from the ballot are obvious and would, in my opinion, offset the discrepancies in “representativeness”: a strong government accountable to the people, with a popular mandate that enables them to confront the unelected tier of power in Morocco.
And yet, even under the current system, it is very possible to have one party marshal enough votes to gain by themselves alone an absolute majority of seats. I’ve been twisting and torturing numbers long enough to make them tell the truth about the current ballot system: Proportional Ballot at a 6% threshold allows for a well-organized, well-financed and well-led political organization to get a majority of seats with only 17% of total registered votes, and only 11% of total potentially registered voters.
Let us tell the story as simply as possible: There are, according to HCP figures, 22 Million adult Moroccans, and 13.6 Million of them are registered. Accounting for local ballot seats therefore, there is one representative per 44,600 registered voters, or 34,000 for all 395 seats. There are of course big discrepancies between, say voters in Sahara districts and those in Casablanca, for instance, but these do not affect significantly the final outcome.
A political organization contesting elections needs to carry at least 198 seats; since it seems national ballot replicates very closely national local ballot results, they need to carry 153 local seats of the 92 districts. In absolute numbers, it means 5.2 Million votes are needed to carry an absolute majority of seats. And there goes the first indicator of ‘small coalition’ winning lax requirement: there is a need for only 38.2% of registered popular votes to win a majority in local ballot, then magnified by the national ballots-slots.
But hang on: there is no need to carry all 34,000 votes per seats. There is the 6% limit embedded in the ballot system; so in reality, a candidate list needs only 32,000 votes; or, to be more precise, only half of those: the total number of votes per seat needs not to be carried in its entirety; in facts, the most stringent requirement for a winning coalition is to actually get exactly half the number of votes, i.e. 16,000 votes, so as to capture all slots per district.
And there it goes: on average, only 16,000 votes are needed to win seats; that means 2.4 Million are enough to capture 153 seats with 50% votes, and then 198 seats per national ballot replication. What turns out to be an outright majority in Parliament House carried at most 17.6% of popular vote. The more people abstain or cast a blank/invalidated ballot, the less votes a party needs to get a majority; a 50% turnout would require only 1 Million votes to command an absolute majority.
Does this sound fishy? Yes. And here’s why:
1/ is it 34,000 or 44,600 registered votes per representative? Both. In terms of overall parliamentary seats, it is 395 seats for 13.6 Million registered voters. But voters don’t get to vote on all 395 seats. They only have to choose for 305 seats; and so, the actual ratio per seat is 1:34,000. However, when a party carries a certain number of votes, they get a boost slightly more than proportional to the number of seats they get on national ballot, this is why the most significant number of votes is relative to the 305 local ballot seats; as a result, the number of votes needed to secure a majority (153) on local ballot is more than enough to get the 198 seats on the full 395 seats.
2/what about competition? turnout? the computations assume no particular configuration for political competition; meaning that the assumption of a maximum required majority per seat of 16,000 holds whatever the distribution of votes among rival candidates; it takes out insignificant votes (less than 6%) as per legislative description, and then just looks at the most stringent case, whereby the second candidate has fewer votes. In the standard Moroccan case, with a heterogeneous distribution of votes, the winning party needs a little less than an absolute average majority of 16,000. It is function of the smallest party, the closest to the 6% to compute the electorate coefficient conditions the required majority to capture all opened seats on a particular district.
Turnout is assumed to be 100% for a party to carry 16,000 votes. Obviously, the lower turnout is, the less constrained a party is in terms of absolute majority per seat. In fact, the required majority decreases at a higher rate compared to the observed turnout;
3/ how come PJD did not capture a majority of seats in November? They did get 1 Million votes with a 45% turnout, didn’t they? there goes the contradicting fact: PJD carried about 1 Million votes (according to a cited PJD source) and yet captured only 27% of total and local parliamentary seats.
First, the votes are not uniformly distributed across constituencies: that is, deviation from the theoretical 50% majority is too high to allow the popular vote significant impact on total carried seats. Some have been won with big majorities (I suppose the Head Of Government’s own seat at Salé was handsomely won) and others could qualify as marginals – meaning, with just enough votes above the local electoral coefficient to get the lead candidate a seat.
Second, while it is true 41% of all 92 districts offer 3-seats slots, PJD’s victory has been concentrated in large metropolitan areas (Casablanca, Tangiers and Marrakesh make up for 44% of their local-seats parliamentary caucus) where the ratio of votes per representative is actually higher; there is therefore a concentration of the Million votes in difficult areas; it is as though PJD candidates had a good reserve of votes, good enough to get an absolute majority, but these have been concentrated in difficult, vote-consuming constituencies, thus leaving little or no remaining votes for easier constituencies, hence the shortfall in absolute majority.