Iraq and Gulf Analysis

Where Are the Ministry of Trade Statistics?

Posted by Reidar Visser on Thursday, 3 December 2009 19:22

[Update 3 December 22:50 CET: Several sources say Ayad al-Samarraie has called for a Saturday-morning emergency session in the Iraqi parliament, presumably aimed at avoiding a second veto]

Two serious misperceptions continue to dominate the Iraqi debate about the amendments to the election law and seem to play a role in making the row over a possible second presidential veto unduly heated.

The first is the idea that Iraqi exiles somehow lost a quota of reserved seats as a result of the first veto. This is incorrect. Under the old system, the exiles never had a quota. (Nor would they have got one in the first iteration of the bill or, for that matter, under Tariq al-Hashemi’s proposed alternative.) Rather, whereas domestic voters had their vote counted twice – first at the governorate level and then in the field of compensatory seats – the vote of the exile only counted towards the compensatory seats, along with all of the domestic votes. It would be more correct to think of this as domestic voters voting twice, first in their governorate and afterwards in the 45-seat (2005) national constituency. Exiles voted only in the small national constituency (where their vote of course carried exactly the same weight as that of any other Iraqi citizen) and were thus clearly second-class citizens. Accordingly, the new arrangements should be hailed as a big and unequivocal step forward for the exiles and should not prove an obstacle towards ratification of the bill.

The second problem concerns the distribution of seats between governorates. In the media, one can get the impression that the exact distribution formula of 2005 will be used. This is again incorrect. It cannot be used, because the Iraqi federal supreme court has ruled that it was unconstitutional since it was based on registered voters and not on total population (and therefore covered only around 14 million of Iraq’s 27 million plus inhabitants). In fact, Michael Hanna has argued convincingly that the ruling means the new amendments could be unconstitutional too, since they insist on going back to the 2005 statistics instead of using more recent statistics when these are in fact available. At any rate, as a minimum, the IHEC will now have to use 2005 statistics that reflect total population (and not registered voters) as a basis for distributing the seats between governorates. Crucially, this distribution is widely expected to be different than the 2005 distribution based on registered voters, because voters are thought to have been under-registered in many Sunni governorates back then due to the insurgency.

In times of crisis and when tempers flare bureaucracies can provide safe havens. The Kurds have already decided they did not like the ministry of trade statistics for 2009 as a basis for distributing parliamentary seats because they thought some of the figures were “unreliable”. (It cannot escape mention that they also insisted on using them for the voter rolls in Kirkuk due to their high reliability.) So while we are waiting for the ministry to dig up the 2005 statistics, why not have a look at statistics produced by another Iraqi institution, just for the sake of the argument? The Iraqi statistics bureau happens to be part of the ministry of planning which in turn has been controlled since 2006 by Ali Baban, himself a Kurd. Their numbers are of course only estimates, based on the 1997 census. But by looking at the estimates for 2004 and 2007 and the different growth rates, it seems clear that some attempt has been made to account for regional variations across Iraq caused by factors such as migration. No note on methodology accompanies these statistics but even if the logic may seem impenetrable (and the zero growth in Salahaddin comes across as particularly suspect) there seems to have been an attempt at creating estimates that account for uneven growth in the post-2003 period (“growth % 2004–2007” below).

Governorate Seats








Seats2010 (2) Seats



Change Growth %



Seats 2010


Basra 18 1,797,000 15 20 24 -4 6.2 1,912,000 19 -5
Maysan 7 762,000 6 8 10 -2 8.1 824,000 8 -2
Dhi Qar 12 1,472,000 12 16 18 -2 9.8 1,616,000 16 -2
Muthanna 5 554,000 5 6 7 -1 11 615,000 6 -1
Qadisiyya 8 911,000 8 10 11 -1 8.7 990,000 10 -1
Babel 11 1,493,000 12 17 16 +1 11 1,651,000 17 +1
Najaf 8 978,000 8 11 12 -1 10.5 1,081,000 11 -1
Karbala 6 787,000 7 8 10 -2 12.7 887,000 9 -1
Wasit 8 971,000 8 11 11 = 9.6 1,064,000 11 =
Baghdad 59 6,554,000 56 72 68 +4 9 7,145,000 72 +4
Anbar 9 1,328,000 11 15 14 +1 11.9 1,486,000 15 +1
Salahaddin 8 1,191,000 10 13 12 = 0 1,191,000 12 =
Diyala 10 1,418,000 12 16 13 +3 10 1,560,000 16 +3
Nineveh 19 2,554,000 22 28 31 -3 11 2,811,000 28 -3
Kirkuk 9 854,000 7 9 12 -3 5.6 902,000 9 -3
Sulaymaniyya 15 1,715,000 15 19 15 +4 10.4 1,893,000 19 +4
Arbil 13 1,392,000 12 15 14 +1 10.8 1,542,000 16 +2
Dahuk 7 472,000 4 5 9 -4 7 505,000 5 -4
Subtotals 230 27,140,000 230 299 307 29,682,000 299
Compensation and minority seats 45 45 24 16 24 (16+8) 0
Total 275 275 323 323 323


On the whole, these statistics lend some credence to the view that the 2009 trade ministry statistics were in fact a neutral estimate of the current Iraqi population based on ration cards. They clearly show that if the 2004/2007 estimated statistics of total population had been applied to the 2005 elections – i.e. on the pattern of what  was later demanded by the federal supreme court – the Kurdistan federal region would have had 4 seats less than it actually got and Nineveh would have had 3 more seats (“seats 2005 revised”). With respect to the apportionment for 2010 according to the amended bill (“seats 2010 (2)”), the picture would have been somewhat more even, but not in any way uniformly disadvantageous to the Sunni-majority areas. The Kurdistan region would win one (2004) or two (2007) seats net and Nineveh would lose three compared to the 2009 ministry of trade statistics (“seats 2010 (1)”), but Kirkuk would also lose three seats – possibly a reflection of Kurds switching their ration cards from Sulaymaniyya to Kirkuk between 2007 and 2009? Diyala and Anbar would gain; in the perennial Mosul–KRG struggle, Nineveh would still do a lot better proportionally speaking than in 2005.*

It should be stressed over and again that these are not the coveted ministry of trade statistics (and the above is a back-of-an-envelope calculation). But still, these are numbers produced by yet another Iraqi ministry that just like the trade ministry cannot plausibly be suspected of harbouring any hidden Sunni Arab conspiracy against the Kurds. And yet just like the MOT statistics they seem to suggest that the Kurdistan share of seats in 2005 was higher than it should have been if the federal supreme court principle of using total population statistics had been adhered to.

The question today is why the MOT statistics cannot be released into the public domain. Were the Kurds unaware of the federal supreme court ruling and did they think the 2005 distribution key was guaranteed? Are there complications between Sunni leaders and especially between Tariq al-Hashemi (who now is aligned with the nationalist Allawi/Mutlak/Isawi/Nujayfi bloc) and the Tawafuq alliance (which authored the original complaint to the federal supreme court but has been critical of the veto)? In the absence of the statistics we have backroom deals that risk producing arrangements that are unconstitutional from day one, and a shouting match about constitutionality by people who clearly don’t know what’s in the Iraqi constitution. Today some even claimed the veto deadline had expired, even though Hashemi said publicly on 25 November that he had not formally received the amended bill. In February 2008, 13 days lapsed from the adoption of the provincial powers bill to the veto by Adil Abd al-Mahdi. It now seems everyone is looking towards Sunday (which is a day off in parts of Iraq due to Eid al-Ghadir, a Shiite holiday). Regardless of what these statistics may turn out to actually say about the populations of Nineveh and Kurdistan in 2005, it would have been a great advantage if they could have formed part of the public discussion about the Iraqi election law.

*Compared with the ministry of trade statistics for 2009 one does get the impression that the internal balance between the three Kurdish governorates should be adjusted, but the totals make sense.

One Response to “Where Are the Ministry of Trade Statistics?”

  1. Thanks for publishing these figures Reidar and in injecting a little sense into the debate. Given the degree of uncertainty in any population figures without a full census, the importance of the changes brought about by the amended law are being overblow, particularly when compared to the major changes resulting from the switch from registered voters to total population (although the downside is growing the parliament by 48 – which means an additional $120m+ cost over the parliament’s term given MPs ridiculous $25,000/month in salary and expenses).

    The change in the out-of-country voting is much more significant, and I agree that it is a move towards more equitable treatment of those Iraqis, although the ability of many Iraqi embassies to implement it effectively and fairly is dubious and it does introduce another avenue for fraud (e.g. I suspect there will be an excess of votes in the diaspora directed to Kirkuk and Ninewa). I assume that the IHEC will issue guidelines which allocated out-of-country votes to the governorate of birth listed in passports, or former residency from some other ID, but this could easily be distorted and will anyway add an administrative burden and delay the finalisation of election results in each governorate until the out-of-country votes are in and verified, which may probably take weeks.

    A final point, I think the allegations (elsewhere) of sectarianism behind Hashemi’s first veto are misguided, as it is far from clear that the diaspora is disproportionally Sunni – my impression is that, including the whole diaspora, not just refugees in recent years (who may be majority Sunni), is much more representative of the whole Iraqi population. The only clear difference is that Christians and Mandeans are massively overrepresented in the diaspora, and so they will benefit most from the voting amendment.

Sorry, the comment form is closed at this time.