The Moorish Wanderer

Elections – Dewey Defeats Truman. (And a Mea Culpa)

Posted in Flash News, Intikhabates-Elections, Morocco by Zouhair ABH on December 26, 2012

I should first start by apologising for some shortcomings in my earlier predictions about results in the couple of by-elections that took place last week, as well as those scheduled in the near future. I have realised I was manipulating the wrong codes to generate my results, and so results from the two latest posts on predictions about electoral outcomes may have been false. It is even more of an unforgivable error that I got mixed up in computations; Mea Culpa, as they say.

How clumsy it was too. the line codes I was using read:

# Sample Size
# Data generation on the basis of moments (Median + Standard Deviation)
RNIND<-rlnorm(n, meanlog =7.69802917027281 , sdlog =1.04882476079341)
# This generates a distribution for RNI historical voting performance per district
# Computes the probability of performing a particular score (a margina)
plnorm(8569,meanlog = mean(RNIND), sdlog = sd(RNIND, lower.tail = FALSE, log.p = FALSE)

whereas it should read:

# Sample Size
# Data generation on the basis of moments (Median + Standard Deviation)
RNIND<-rlnorm(n, meanlog =7.69802917027281 , sdlog =1.04882476079341)
# This generates a distribution for RNI historical voting performance per district
# Computes the probability of performing a particular score (a margina)
plnorm(8569,meanlog = log(mean(RNIND)), sdlog = log(sd(RNIND)), lower.tail = FALSE, log.p = FALSE)
# Probabilities are computed on the basis of logged, not level moments.
# Dummkopf.

So this is the brand new method with which I crunch the numbers is simpler and hopefully, more understandable. I should say it has done a good job in “explaining” PJD’s victory in Inzegane (an obvious result as a matter of fact) and why the same party may have lost to RNI in Chichaoua. Since elections in Azilal, Moulay Yacoub, Settat, Sidi Kacem and Youssoufia have not taken place, I would like to redeem myself by offering a more thorough assessment of electoral probabilities in shifting away, or holding in contested seats.

Enrolled voters:                                 197,679
Votes cast:                                       71,608  36.2
Invalid votes:                                    13,580  19.0
Valid votes:                                      58,028  81.0
Party                                         Votes       %        Seats
Popular Movement (MP)                                230  00.4       -
Party of Authenticity and Modernity (PAM)          2,411  04.2       -
Independence Party (PI, Istiqlal)                  4,958  08.5       -
Party of Justice and Development (PJD)            29,541  50.9       2
Party of Progress and Socialism (PPS)              1,545  02.7       -
National Rally of Independents (RNI)               5,257  09.1       -
Constitutional Union (UC)                            135  00.2       -
Socialist Union of Popular Forces (USFP)           7,807  13.5       1
Others                                             6,144  10.6       -
Total                                             58,028             3

In essence, I try to provide a probability for a party to win a seat in a particular district, given its past electoral history there, conditioned also on the fact they may or may not have a seat there already. Inzegane was as expected an easy win for PJD because not only they got 2 seats there, but also because they had in 2011 a 21,000 lead over they nearest USFP competitor, whose seat was up for election last week. It would have been enough for the PJD candidate to turn out a fraction of their electoral to carry the seat, most likely with the same votes they got last year. According to the party’s own communiqué, they managed to carry 15,000 votes, which was an overkill, to say the least. It matters little they managed to mobilise only half of their 2011 turnout, it was more than enough to complete its control of all three seats at Inzegane.

Moreover, The probability for PJD of getting at least as many votes as USFP in 2011 was the highest among all competing parties: there was a 46.82% chance of getting at least 7,800 votes, and anything between 7,800 and 21,000 was likely to happen at 44.07%. These results show a very strong lead for PJD compared to other parties: RNI had virtually no chance of getting more than the required 7,800 votes – because its own national and local made it so: the probability of improving its 2011 electoral performance.

And so were USFP’s chances, especially so when its marginal seat in Inzegane ranked in the top 1% districts for the 2011 elections. Istiqlal was the only real contender whose electoral performance allowed it some significant chance to improve its score to 7,807 votes – a probability of 9.4%. In absolute terms, the likelihood of getting more than their 2011 votes was 13.54%, the closest to PJD in this district.

I have some results on Chichaoua as well, which I will be posting later on, with predictions for the remaining seats following shortly.


PJD’s “Pocket Landslide”

Posted in Flash News, Intikhabates-Elections, Morocco, Read & Heard, Tiny bit of Politics by Zouhair ABH on December 8, 2012

A little over one year after PJD‘s victory, it would be interesting to look at what is already their electoral legacy, or indeed the lack of real appreciation of how important it is. Not matter how past electoral results come under criticism, they have been de facto the law of the land – whatever the real results following the 2002 elections, all parties agreed to the official results, and these have been validated as such. And it would be better, I think, if this criticism was laid aside, especially since the 2002 election did give PJD a clear win, were it not for the diluting proportional ballot.

Majorities are 'easier' to form when the ballot system weighs in pluralities in districts

Majorities are ‘easier’ to form when the ballot system weighs in pluralities in districts

In general terms, I describe a method that points to majority-based ballot system as a good indicator of how political parties can improve their probabilities of forming a government by themselves, thus delivering stable governments and even more stable parliamentary majorities.

PJD’s victory in 2011 was a pocket landslide because the party was 65 seats short of an absolute majority – even if it was well ahead of its nearest competitor. A majority-based ballot system could have delivered the absolute majority they needed. Their feat was only matched by the joint USFP-PI 1993 campaign. Unfortunately for PJD, they are in a coalition with parties directly (and adversely) affected by any re-districting, or majority-based ballot system. (read here the theoretical argument against coordinated effort among political parties)

This is how I compute these majorities: for a particular district, all of the seats are allocated to the party with a plurality of votes. The simplest, crudest rule of politics – and poker: winner takes all. On the basis of this principle, the electoral map since 1963 is radically changed. I further consolidated party performances by aggregating split-offs – which leads to 13 big ‘partisan conglomerates’ – and these results tell a story: a consolidated political competition over parliamentary control allows for larger probabilities of reaching an absolute majority (in the cases of seats open for local ballot) and these contradict the final outcomes observed over the past couple of elections; for one, the 1997 Alternance would have been led by Istiqlal instead of USFP, and 2002, not 2011 would have been PJD’s coronation. But then again, efficiency is not Morocco’s forte.

These results can then be compared against the probabilities of each party to get a majority of the votes. These probabilities are computed on the basis of past electoral results – with increasing weightings for more recent electoral campaigns. And I am pleased that up to 92% of the historical results are explained by the following, rather simple linear model: V(\max_{i,j})=\sum\limits_{p_{k,i,j}}^n\alpha_k \mathbb{E} \left[V(P_{k,i,j})\right]+\epsilon_{k,i,j}

where \alpha_k the estimated probability for a party k to get the plurality in a district i. These parameters need not sum over 1, because there are a lot of cross-party historical votes. This confirm my earlier claim about PJD’s robust position on its 2016 prospects, as well as the need to go for a majority-based system – some parties have clearly more chances to get the majority, while others do not (those have been taken out of the fitting because of the statistical insignificant results)

      Source |       SS       df       MS              Number of obs =     153
-------------+------------------------------           F(  6,   147) =  292.75
       Model |  5.1894e+10     6  8.6491e+09           Prob > F      =  0.0000
    Residual |  4.3430e+09   147  29544400.6           R-squared     =  0.9228
-------------+------------------------------           Adj R-squared =  0.9196
       Total |  5.6237e+10   153   367564886           Root MSE      =  5435.5
         max |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
     pnd_pam |   .2806256   .0863479     3.25   0.001      .109982    .4512693
     rni_ind |   .4540341   .1007729     4.51   0.000     .2548834    .6531848
    mpdc_pjd |   .5029201   .0445715    11.28   0.000     .4148364    .5910038
          pi |   .3938989   .0974869     4.04   0.000      .201242    .5865558
     fdic_mp |   .2368222   .0761314     3.11   0.002     .0863687    .3872757
   unfp_usfp |   .3958337   .0742675     5.33   0.000     .2490639    .5426035

As one can see, the next elections tend to favour PJD (with an estimated 50.2% chance of getting a vote majority on all seats – quite different from getting a majority of seats) although there are a couple of contenders, and an even stiffer competition between say RNI vs USFP, RNI vs PI, and finally, USFP vs PI. As for PAM, my estimate is they are not likely to get anywhere close to a serious contender for governing party.

In essence, PJD’s electoral legacy would be that of ‘breaking the mould’ of opposition parties: strong enough to have a large caucus, but too weak to force censure motions, and definitely unable to form a government on their own right. It would be a breakthrough legacy if PJD could force through an electoral reform that seeks to improve the chances of a one-party government. True, this would mean PJD is most likely to stay in office for the next decade, but this is about representative democracy.

Data description: seats are allocated to each district per each electoral districting. All of the seats are allocated to the party with a plurality of votes. Parties are merged afterwards when applicable.


Predictions for the 2016 Elections, Part.3

Posted in Flash News, Intikhabates-Elections, Moroccan ‘Current’ News, Morocco, Read & Heard by Zouhair ABH on November 10, 2012

Marginal seats per large party turned out to be quite a fascinating subject: each party has its strongholds – even marginals. It also shows that out of the 83 marginal seats, only 36 are actually in play. So here’s my first set of predictions as to the 2016 configuration for parliamentary caucuses:

1/ Barring exceptional surge in favourable turnout, PJD cannot hope to achieve anything near 153 seats on local ballot, for two basic reasons: first, its electoral distribution does not allow systematic strong probabilities to carry at least 2 seats per large district, and second, other parties have stronger probabilities to hold districts quite important for PJD to reach 153. Its best case scenario is that of an increased plurality to 121 – that is, a surge of additional 394,000 voters, either pinched from the competitors in these seats, or new electors.

PJD can expect some pick ups next election, but needs to take a close look at some endangered seats they could lose by less than 2% of the votes.

Either way, PJD cannot perform 30% more in very competitive districts when they find it difficult to deliver a 7.3% nationwide swing. So ix-nay on the Absolute Majority, Morocco carries on with a PJD-dominated coalition government on 2016. This is simple arithmetic: an absolute majority in parliament means 198 seats, or 153 on the local ballot. In order to achieve it, there is the need for each contender party so secure a plurality of votes from the 92 districts. It so happens the present system requires the winner, in order to govern by themselves, to carry at least 47 of the 92 districts to have a chance to get 153 seats on local ballot, on the basis of past electoral performance in turnout (hence controlled for party ‘leadership’) some districts are an absolute winner – because they have a lot of seats, or because registered population and turnout is high enough to make them essential to the ‘winning district coalition’. Alternatively, First Past the Post provides more favourable conditions for strong parliamentary majorities – but then again, this would shake up a consensus across partisan and political lines (Makhzen and many of its dissidents oppose it vigorously)

The methodology is simple enough: start by listing districts with decreasing highest turnout, and sum their corresponding seats until 153 is reached (by the current number of seats, the absolute majority is 155, but that’s close enough) then look at the last district to make up for the absolute majority: Essaouira. Bad news for PJD or PI – the most likely (albeit a small one) to have a certain electoral distribution able to deliver a super-majority.each of these essential districts is controlled by strong majorities of different parties. A simple look at the map opposite shows it: it is even worse news for PJD because some of its regional strongholds (like Tangier) does not account much in the path to 153. This is why I predict, on the basis of its past electoral performance, it cannot acquire the required 65 additional seats.

Essaouira is the last district needed to get the super-majority. A truly stable party government would value its district as much as anything else.

2/ There is a considerable chance PJD, PI and RNI might lose more seats than gain some, because most of their marginals are very competitive – i.e. challengers are close enough to change the electoral outcome in 2016 with only a dozen of votes per district. On the other hand, PAM, UC and to a lesser extent, USFP and MP could gain moderately; Overall, and bearing in mind the initial assumption of a stable turnout, the 2016 parliament ins unlikely to change much, other than transfer 25 seats one way or the other.

Souad Boulaich El Hajraoui: 55 votes majority (0.31%)

Abdellatif Rachid: 722 votes majority (2.02%)

Abdellah Benhammou: 942 votes majority (2.32%)

Mohamed Slimani: 601 votes majority (2.41%)

It is worth pointing out that these predictions do not take into account PPS’ own chances, as well as the expectations for the smaller parties in Parliament. It is clear however these will have slimmer chances to be represented again in parliament.

3/ The majority party with a vote plurality, and leader of the government coalition, PJD, will have to look very closely to 18 districts with 66 seats to achieve the 153-absolute majority threshold, that is, gathering between 217,000 and 268,000 additional votes, while maintaining or expanding their majorities. And again, it is quite unlikely, unless some surge in favourable voter registration and turnout allows it. In absolute numbers, the 2011 election brought to the polls about the same number of voters in 1984, respectively 4.5 Mln and 4.32 Mln (even as registered and likely voters population surged 60% and 67% respectively) Based on weighted averages of past electoral performances (weighted so as take into account time lags) turnout for 2016 is likely to stabilise itself around 4.5 and 5.1 Million votes – that is 38.3% and 42.7% (this does not take into account invalidated ballots, so official turnout could well be a little above 50%) but this turnout sustains itself only through an increased share of disenfranchised Moroccans, where registration rate would drop from 72% in 2007, to 60% in 2011, and a little under 51%.

Overall, it seems unlikely the present parliamentary configuration would change significantly, as far as the big parties (those controlling 95% of the seats) are concerned: there are few competitive seats (of theoretical 92 seats, only 84 could are actual marginals, and only 36 could actually change majorities)

Predictions for the 2016 Elections, Part.2

So a uniform swing across all 84 marginal constituencies does not change the picture that much: of the 305 seats, the ranking of the 8 largest parties that concentrate 94% of parliamentary, local ballot selected seats did not change but for two: Because it exhibits a larger than usual number of marginal seats, Istiqlal is bound to lose a net 6 seats, thus leaving RNI as second-in-command. PJD on the other hand, can expect to win a net additional 3 seats, still 65 seats shy of an absolute majority.

2011 and 2002 exhibit similar and almost coinciding density curves.

But this is a highly unlikely outcome, precisely because these swings are assumed to be uniformed: indeed, the discrepancies in voting results do point to heterogeneous outcomes, essentially due to the discrepancies in carried votes between the incumbent marginal and the potential challenger. Let me use an example to illustrate my point: the Speaker of the First Chamber, Karim Ghellab, represents a marginal – voted in with 4,789 votes, a long way behind PJD and UC, respectively 14,853 and 8,925 votes. The immediate competitor to Mr Ghellab is the MP list, with 2,735 votes. What is the probability, ceteris paribus, Mr Ghellab would lose his seat to the MP challenger in 2016?

Consider the party’s past electoral performance. Amazingly enough, Istiqlal’s vote distribution did not vary much if not at all between 2002 and 2011 (the density curves for 2002 and 2007 shows it on the graph). Therefore, it makes sense to assume Mr Ghellab’s margin can be computed in terms of probabilities. And this shows us he would keep his seat with a probability of 46.7%, with a vote tally fluctuating most likely between 3,976 and 5,841 votes, regardless of Mr Ghellab’s or the nationwide electoral performance, he needs to ensure a a marginal majority of 813 votes.

(incidentally, voter distribution per party or per seat exhibits very common forms, as various papers from academia[pdf] testify to that)

How about PJD’s six marginal constituencies? According to the averaged PJD past electoral performances since 2002, three of these are more likely to go over the edge to PJD’s local challenger, and one is a pure, statistical toss-up. This is moderate good news for PJD, because they are assured to keep a comfortable margin, provided 2016 Elections come with no exceptional events (the “Black Swan”) In fact, all three marginals have strictly positive likelihoods to swing against PJD incumbents, but three only of these lean toward PJD’s local challengers, and Laayun remains in a tie (a feat for maverick PJD considering the voting pattern down South)

Of the 6 marginals, PJD has a significant chance to lose three, and a fourth is nearly a toss-up

On the other hand, one could also look at the Head of Government’s own seat in Salé, and while it is by no means a marginal (M. Benkirane got 27,000 votes to a second with a little over 8,000) yet there is a positive (very low) likelihood of losing his seat: less than 3% – precisely because it would entail a huge swing, with a third of Salé voters going over to any challenger.

The Istiqlal party is a peculiar contender: though it ranks a distant second in electoral results, it has maintained its initial strength even as the party led a relatively unpopular government, and has been associated with all government coalitions at least since 1997. It is undoubtedly an establishment party. Yet for its formidable 60-members strong caucus, many of those have been elected on razor-thin margins in very competitive districts, and the Istiqlali challengers have not been numerous enough to make up for the endangered seats. All in all, Istiqlal could well expect a net loss of 6 seats.

The party with the most interesting marginal record is undoubtedly Istiqlal, first, because they have the largest number of marginals, and second, because their electoral performance per district did not change over the years (a remarkable stability given its Establishment status) the odds for and against voter swing in PI-held seats is the perfect case study for the proposed method. As one can see, the initial assumption of a uniform swing across all marginals was unrealistic. By that account, Istiqlal was to expect to net a loss of 6 seats. But now, and according to the computed probabilities, Istiqlal can expect – again, ceteris paribus, to lose only 4 seats, with one is a small district where lopsided swings are expected and observed, and four others with moderate loss expectations. Assouerd is a peculiar seat, because a 200 voter switch is a dead heat, 600 is a landslide (respectively 4.3% and 13% swing)

Aousserd is very competitive, the high likelihood of part swing is due to the small constituency to start with.

Can this set-up deliver reliable results for marginals as well as ‘safe’ districts? The Tangier by-election provides good case study (given the fact polls do not curry favour with our esteemed representatives[pdf]) the model tells PJD was strongly favoured to retain all its Tangier seats, since the probability of losing its majority is so small it is not even statistically significant (larger than 99%) but on the other hand, the model predicts it may lose 12,500 votes (but this is probably due to the fact that Tangier district is a PJD stronghold, and their maximum voter reserve has been reached) so even if it retains its majority almost certainly (and it did) it was expected to lose some votes. The actual results show PJD lost 16,200 votes – a figure statistically close (well within the margin of error) to the projected 12,500 synthetic potential loss.

Howe about the other competitors for the Tangier district representatives? According to their respective results, PAM was the marginal representative, with a potential risk of losing it to UC, with a probability of 17%. So actually, PAM’s marginal Tangier seat was relatively safe, with expected losses of no more than 1,730 votes – it turns out PAM lost only 300-odd votes. The actual loser in this Tangier recall election was RNI, even though it was not a marginal (yet exhibited a loss probability of 9.7%) yet it failed to carry any vote. The model predicted PAM could lose to UC, but since RNI did not get any votes, UC candidate list simply filled in.

Predictions for the 2016 Elections, Part.1

The proportional vote ballot with a minimum threshold adopted in Morocco has proven to be of weakening effect to representative democracy. Sure, it has allowed more parties to get on the game, and even get a seat or two, but it has definitely weakened government party coalitions, as these grew larger, more heterogeneous and political weaker, beholden to the non-elected part of Moroccan government. Consider the last 2011 elections: PJD came first with 1.08 Million votes with Istiqlal following behind with little less than half the votes – in fact, PJD needed an additional 268,700 votes to get an absolute majority, a 7.33% nationwide swing.

Istiqlal, on the other hand, needs an 18.9% swing to overtake PJD. Where did that figure come from? The proportional ballot system allocates different levels of majorities to each district. In 2011 for instance, the corresponding majority votes would have been 1.35 Million votes, or 28% of the popular vote (I have posted on the reasons behind discrepancies between absolute majorities, plurality and parliamentary seats’ majorities)

The operating principle behind the politics of government coalition is simple enough: since no political party can form a government on their own, they need to build a coalition. In fact, it does not have to be the party with a plurality of votes, as long as the ‘majority swing‘ i.e. the percentage of popular votes needed to attain absolute majority, is significantly low. In 1997, the 40,000 votes difference between USFP and Istiqlal could have put either in charge of the Alternance government (Istiqlal leaders did protest as they felt they had a slight advantage in the popular vote) because these votes made up only .68% of the electoral turnout.

The same goes for 2002, while USFP was again ahead in the returns, the differences with the second ranking party – Istiqlal – were statistically insignificant (less than 2% of the votes) and there was good chance either M. Abderrahmane Youssoufi or M. Abbas El Fassi could have led the 2002 government. History of course shows it was otherwise, the internal squabbles in the Koutla prompted the King to dismiss both and instead appoint non-partisan M. Driss Jettou. In 2007, the re-match was between PJD and Istiqlal, with an even lower margin of .15%, so in truth, the 2007 Abbas El Fassi government could have been the Saad Ed-Dine Othamni government.

2011 shook-up the ambiant 1997 consensus because it was the first time since 1984 one party managed to carry more than one million vote (in 1993, USFP and Istqlal contested elections on joint lists, carrying 1.5 Million votes) and at the same time minimize the majority swing, from 15.72% in 1984 for Union Constitutionnelle, to 7.33% for PJD.

The table reads: “PAM can lose 5 seats to MP” and “MP loses no seat to MP”

This is precisely what is at stake: out of 305 seats on local ballot, 84 of these are marginals, i.e. due to the proportional ballot system, a member of parliament might lose his seat (yes, there is only 10.4% chance to get a female representative elected on local ballot) if only a couple of votes switch to the next party. Incidentally, the ‘real’ number of marginals should have been 92 (for 92 district) the last slot for each one being the ‘marginal’. This proves some parties (PJD, Istiqlal, PAM and RNI) tend to have strong constituencies, and that displacements are not as frequent as one might think they are. On the other hand, as shown on the table above, these displacements are not evenly spread – some marginals are more marginal than others, with various probabilities of vote swing.