The outlook 10 days ahead of Turkey’s referendum

The outlook 10 days ahead of Turkey’s referendum

Yesterday on April 5 I talked to the heads of three of Turkey’s leading polling companies about the possible outcome of the April 16 referendum on whether to shift to an executive presidential system, as targeted by President Tayyip Erdoğan and the ruling Justice and Development Party (AK Parti).

I will not give you any figures for the “Yes” and “No” sides, as none of the companies have completed their final fieldwork. But their preliminary estimates show the vote as being so close that it falls into the limits of standard deviation, with a margin of error of 2 to 4 points. Giving figures now in such a volatile political atmosphere could mislead the reader.

So let me just try to give the impressions I got from my off-the-record interviews with leading pollsters.

The race is basically neck-and-neck between the two campaigns, despite obvious support given to the “Yes” campaign by the president and the government. This support extends to the massive use of the media and state resources, which has prompted the “No” camp to complain about a highly unfair race as Turkey heads to the ballot box under the ongoing state of emergency.

* The AK Parti won 40.9 percent in the June 2015 election but managed to increase its votes to 49.5 percent in the snap election of November later the same year. That increase could be explained by the drop in the votes for the Nationalist Movement Party (MHP) from 16.3 percent to 11.9 percent, and the drop in the votes for the Kurdish problem-focused Peoples’ Democratic Party (HDP) from 13.1 percent to 10.8 percent. There was also a drop in the votes for the religious conservative Felicity Party (SP) from 2.1 percent to 0.7 percent in the same period. This flow from all three parties went to the AK Parti due to the ambitious campaign of President Erdoğan in support of the AK Parti government. Votes for the social democratic main opposition Republican People’s Party (CHP), meanwhile, remained around the same level of 25.0 percent and 25.3 percent respectively.

It should also be noted that the support Erdoğan got when he was elected president in August 2014 was 52 percent.

* MHP leader Devlet Bahçeli has given his support to the “Yes” camp in the referendum, and on paper the total vote share of the AK Parti and the MHP should leave no space to breath for the “No” camp: 57.2 percent according to the June 2015 election and 61.4 percent according to the November 2015 election. However, the MHP has been suffering from internal turmoil. A number of popular names in the party have objected to Bahçeli’s decision to team up with the government, declaring their support for the “No” campaign and subsequently being expelled from the MHP by Bahçeli. According to pollsters, it is mostly the more urban and better educated members of the MHP who are tending to say “No,” though they refrain from giving a specific percentage regarding how many MHP voters will vote “Yes,” “No,” or simply opt to abstain. 

* The AK Parti is not happy with the performance so far of the MHP, which they believe puts at risk the “Yes” campaign. What’s more, some AK Parti figures are critical of Bahçeli’s holding of “Yes” rallies in places where the AK Parti is already strong, such as Kayseri, Sakarya and Konya. Instead, they want the MHP to rally in places like Adana, Mersin and İzmir, where they feel they could possibly convert more “No” votes to “Yes.”
* The MHP’s lackluster performance is a reason why Erdoğan and Yıldırım recently decided to turn their faces to Kurdish voters in the middle of the campaign. The invitation of Iraqi Kurdish leader Massoud Barzani to Ankara was thought to help attract conservative and tribal Kurdish votes into the “Yes” camp. But this strategy partly backfired when Barzani - as soon as he returned to Iraq - raised the Kurdish flag in Kirkuk, (which has a high Turkmen and Arab population), and launched a campaign for an independent Kurdistan.

 Erdoğan was forced into condemning these moves.

* All pollsters that I have spoken to agree on the estimate that the possible loss of “No” votes to “Yes” within the CHP is less than 1 percent of the party’s potential vote. But there are conflicting opinions regarding the HDP’s potential vote. A number of HDP executives, including its co-chairs Selahattin Demirtaş and Figen Yüksekdağ, are in jail on charges of “aiding terrorism” under the state of emergency, which negatively affects the party’s campaign. In addition, during the massive clashes in 2015 and 2016 between the outlawed Kurdistan Workers’ Party (PKK) and the security forces in southeastern Turkey, bordering Syria and Iraq, a number of towns and districts saw huge destruction, displacing some 500,000 people. Many of these have not registered themselves as voters in the places where they have settled, and that may cost the “No” camp around 1 to 1.5 percent, thus feeding the “Yes” campaign indirectly.

* Another advantage forecasted for the “Yes” camp may be the votes from citizens abroad, especially from Europe. Last month’s antagonism with Germany and the Netherlands over the banning of Turkish ministers campaigning there, and the polemic that broke out when Erdoğan accused them of acting like “Nazis,” may have encouraged some Turkish voters in Europe to unite more behind Erdoğan, possibly adding another 1 to 1.5 percent to the “Yes” votes.

* But there is also another disadvantage for the “Yes” campaign. The AK Parti executives are not sure that 100 percent of the party will approve the consolidation of all executive power in the president’s hands. As with the case of the MHP, skeptical AK Parti voters are among the more urban and well-educated parts of its base, including professionals below the age of 35. There are no reliable figures, but estimates show that around 2 to 4 percent of the AK Parti’s voters have not been convinced to say “Yes,” which could mean a lot in a neck-and-neck race. That is one of the reasons why Erdoğan and Yıldırım have recently been focusing their campaign on wooing more educated and urban-based voters.