Pollster: Turkey’s referendum race is still close and will lead to polarization

By Barçın Yinanç for Hürriyet Daily News

The race ahead of the April 16 referendum on constitutional changes shifting Turkey to an executive presidential system is still close, with the ruling Justice and Development Party (AKP) needing to convince the undecideds among its constituency, according to Bekir Ağırdır, the general director of polling company Konda.

“We see some undecided even among the core loyalist constituency of the AKP,” Ağırdır told the Hürriyet Daily News.

Tell us about the general picture of opinion over the last two years, leading up to the latest situation ahead of the April 16 referendum.

First, in this time we have had four elections based on identity politics, with identities confined to four political parties. Second, there is a hardening polarization based on being pro or anti-AKP. Third, there has been a consolidation in the four parties; the chance of another party making its weight felt does not look possible. Forth, there is a lack of political rivalry, as there is no second party that could challenge the AKP.
When it comes to the referendum, a significant segment of society lacks information and interest. And they are right, as this referendum has not been brought about due to society’s demands. People are not talking about it – not because they are ignorant but because it does not touch their daily lives. That’s why people have not been so enthusiastic from the beginning. The AKP has seen this problem and therefore has based its campaign not on the substance of the changes but rather on polarization.

In addition, there is a process of deepening polarization. We can observe a change in the accusations of each camp against the other, with more hostile terms used by each camp to identify the other. For example, the ruling party’s argument equating of “no” voters with terrorists seems to have gained ground among its own constituency. This demonization of the other is something that will be hard to carry in the future. Whether “yes” or “no” wins, we are going to have to live with this burden. One of the most important problems in Turkey is that the feeling of “us” is becoming more and more shattered.

Regarding the referendum, today we are not at a point where one side or the other can say confidently: “This is it, we are way ahead.” The race is still neck and neck.

But if the whole political spectrum is confined to four parties, the result should be pretty clear in advance simply by adding up the votes of the parties that favor “yes.”

There is no hesitation in the “no” camp, where you have the voters of the [main opposition Republican People’s Party] CHP and [the Peoples’ Democratic Party] HDP. The majority of the undecided are among the constituency of the ruling AKP and the [Nationalist Movement Party] MHP: One fifth of the AKP voters remain unconvinced of the changes. A quarter of the MHP’s voters are undecided and more than half of the MHP’s voters are in the “no” camp.

If “yes” is to win, it will not depend on the “no” vote getting smaller or “no” voters not going to the ballot box, it will depend on whether the ruling party and the president are able to succeed in persuading their supporters. Currently, what we see in the campaign is basically the propaganda of the ruling AKP. The result will be “yes” if this campaign to convince succeeds.

How do you identify the undecided voters among the “yes” camp?

When you look at Turkey’s approximately 56 million voters from a political party’s perspective, you first see a core constituency that is either ideologically loyal to the party or emotionally loyal to the leader. Then there are the sympathizers in the second circle. They vote for the party but there is a possibility that they could switch to another. Then there is a group that is neither in the core nor in the sympathizer circle. They are equidistant from all parties and vote according to a campaign issue or candidate, etc.

Looking from that perspective, it is easy for the AKP to convince its core group, though we believe there is hesitation – however little – even in that group. In the sympathizer circle, there is an unconvinced group who say they will vote for the AKP in the general election but are not convinced of the necessity of this referendum. In the MHP, there are fractures both in the core and in the sympathizer group, while in the CHP and the HDP both the core and the sympathizer group are clear on the “no” vote.

There is also a group that is in the grey area in terms of the referendum and in terms of politics. Around 65 percent of Turkish society is polarized, while around 35 percent remains outside of this polarization, and we have to consider whether this group – which is roughly a third of voters – will be persuaded to vote “yes.”

You see HDP voters in the “no” camp. Some people argue that conservative Kurdish voters will lean toward the “yes” camp.

You can traditionally divide Kurdish votes into those who are for or against the system or the dominant power. It’s roughly a 50-50 divide. You can also divide Kurdish voters in two in terms of whether they are religiously conservative or more secular. That’s also around 50-50.

Until 2010/2011, the AKP got around 60 percent of Kurdish votes, while the HDP’s precursors got around 35 or 40 percent. Then the events surrounding the Syrian Kurdish town of Kobane in 2014 strengthened Kurdish identity. This is a deep change. Those voting for the HDP did not give up their conservatism or get less pious. They still criticized the [outlawed Kurdistan Workers’ Party] PKK, but they are not going to give up on their Kurdish identity. I expect around 70 percent of Kurdish votes to be for the “no” camp and around 30 percent for the “yes” camp.

But with the ongoing state of emergency, to what degree can genuinely measure the pulse of the public? Would you say the margin of error of polling companies could be larger this time?

In addition to the state of emergency, this is the first time we have seen such social tension ahead of a referendum. A particular climate has developed over the years due to polarization, so our job is difficult.

Some claim that the initial number of undecided voters has decreased.

We think it is still around 15/17 percent, while those who do not want to reveal their decision could be around 3 to 5 percent.

What about the effect of participation on the outcome?

It is not possible to forecast an outcome based on the level of participation. You can’t claim “no” votes or “yes” votes will be higher or lower due to low or high participation. We have found that among both the “yes” and the “no” camp, the number of those who say they are sure about their decision is around 95 percent, while among both groups around 85 to 90 percent say they will definitely go to the ballot box. We expect a high participation, as in the general elections in June and November 2015, which was 87 percent.

Does time play in favor of the “yes” camp?

All I can say is that we have been having the debate on the referendum for a long time but there is still a significant number of undecided voters among those who say they would vote for the AKP or the MHP if there were a general election tomorrow.

What would happen if there were equal conditions for both sides to campaign freely?

If there were equal conditions and if the debate was focused on the substance of the changes, the “no” votes would be ahead of the “yes” votes. Half of the public is not convinced that the change in the system will solve the country’s problems. The AKP has seen this, which is why it is basing its campaign on polarization axis.

Do you think any development in the final days of campaigning could radically affect the votes.

It is possible. We saw this between the June and November elections in 2015. Conditions that create the perception of an [existential] threat to the country, or the need for a strong force, serve the government.

What about the aftermath?

Around 48 to 49 million valid votes will be cast, with a division of around 23 million to 25 million. The rational mind says that you cannot do something if 23 million is against it. But when you look at the rhetoric of the dominant political actors, the government will go ahead even without taking into consideration what the other 23 million thinks. On the other hand, if there are 25 million “no” votes, that camp will celebrate without taking into consideration why the other 23 million voted “yes.” This is not sustainable, but polarization and tension will continue.