A close look at the available data, however, shows not only that electoral irregularities can’t explain the ruling Justice and Development Party’s (AKP) defeat, but, in fact, the irregularities that took place predominantly benefited the ruling party.
Were there electoral irregularities?
The YSK explained the decision as primarily based on the composition of the local election councils in Istanbul. Yet, it canceled only the metropolitan mayoral election despite the fact that the same councils oversaw the district mayoral elections and municipal council elections, in which the governing parties managed to win a majority.
The main opposition candidate for Istanbul, Ekrem İmamoğlu, complained that by the YSK’s logic, “the 2017 constitutional referendum and the 2018 presidential election should also be canceled as the same local election councils worked and the same procedures were followed during those elections.” As with this year’s election, the ruling party had won controversial victories in both cases.
The main remaining case for redoing the election, then, is the claim of electoral irregularities. To find out whether the irregularities overall benefited the pro-government parties or the opposition, I analyzed the results of the latest two local elections, 2019 and 2014.
My findings confirm irregularities with the extraordinary increase in the number of registered voters, high variation in the percentage of invalid votes, and geographical distribution of voter support for the nationalist pro-government parties. But the district-level distribution of the electoral irregularities strongly coincide with the distribution of the increase in the vote share of the governing parties, not the opposition.
Unusual shifts in the number of registered voters
During the election campaign, opposition parties reported numerous cases of illegal voter registrations. One example cites a total of 1,108 citizens registered as voters in a single apartment unit, yet governing parties consistently disregarded these claims.
On average, the number of registered voters increased from 2014 by 8 percent per district. Also, the vote share of the two pro-government parties decreased by seven percentage points on average per district compared to 2014.
When it comes to the districts with more than a 30 percent increase in the number of registered voters, the share of governing parties increased by seven percentage points. In the rest of the districts, the same parties lost about eight percentage points. The main opposition Republican People’s Party (CHP) and its alliance performed significantly worse in such districts when compared with their performance in remaining districts.
Variation in the invalid vote ratios
At the core of electoral fraud discussions in the former elections is the distribution of invalid votes. Looking at the Istanbul districts, there is a strong positive correlation between the percentage of invalid votes and the percentage of the ruling AKP’s votes as well as a strong negative relationship between the percentage of invalid votes and the percentage of CHP votes.
One possible explanation for this may be the variation among the socio-demographic characteristics of districts. Studies found that lower education levels may lead to higher invalid votes. In this respect, if CHP is stronger in districts with a higher level of education, one might expect an increase in the CHP votes in districts with lower invalid vote rates.
However, this reasoning fails to explain why the change in the invalid vote percentages since 2014 is negatively associated with the CHP’s increased popular support since 2014. In fact, CHP managed to increase its votes predominantly in the districts where the percentage of invalid votes were lower, compared with 2014.
Also, there is a significant positive correlation between the change in percent of invalid votes and the governing coalition’s voter support across the nation. The popular support of the pro-government parties on average increased by 9 percent in districts where the number of registered voters increased by more than 30 percent since 2014 and the invalid votes were more than 4 percent in 2019. This is an unprecedented result, given, on average, the seven-percentage-point nationwide decline in popular support.
Vote share changes in the pro-Kurdish east
The pro-government parties, which have been pursuing aggressive nationalist and anti-Kurdish policies over the last four years, increased their vote share by 3 percent on average in the pro-Kurdish regions in the east and southeast of Turkey while decreasing their share in the rest of the country by 10 percent.
Even more striking, among those eastern districts with more than 30 percent increase in the number of registered voters, the pro-government nationalist parties increased their vote share by 21 percent, while the pro-Kurdish party (People’s Democratic Party) decreased its vote share by 17 percent.
What to watch for in the coming weeks
CHP’s candidate İmamoğlu made a historic speech on Monday night, mobilizing millions against the YSK’s decision and the governing parties. The opposition parties have, nevertheless, accepted the challenge of the June 23 election rerun for Istanbul.
The opposition may try to attract pro-AKP voters and politicians. Former president Abdullah Gül’s tweet making an analogy between the Constitutional Court’s decision that created a huge crisis in 2007 and the YSK’s latest decision is an important step. This was followed by former prime minister Ahmet Davutoğlu. If more former AKP figures join them, the opposition may gain a huge advantage.
While the CHP won in Istanbul even with irregularities correlated with increased vote share for the ruling AKP, there is a possibility of increased fraud strategies. There are reports from the March 31 election in which security personnel illegally voted as a group in some districts. Current laws enable security personnel to vote in whichever polling station is convenient for them as long as they hold the relevant paperwork. Yet, there is no serious oversight mechanism. AKP will have a chance to deploy thousands of security personnel to Istanbul from other cities to provide security during the elections. The government has been recently hiring tens of thousands of new police officers. There is a serious possibility that these officers would vote for the AKP to help close the narrow (14,000) margin of loss.
The timing of the election is also against the opposition’s prospects, with wealthier residents and hundreds of thousands of university students probably leaving Istanbul for the summer. Considering the AKP’s relatively lower popularity among these groups, this would be in the ruling AKP’s favor. If, however, CHP effectively mobilizes its base, fights against fraud, and attracts voters from other parties, it may manage to keep its margin of victory.
Abdullah Aydogan is a research scholar at Rice University’s Baker Institute for Public Policy and a lecturer of political methodology at Columbia University. Follow him on Twitter at @abdaydgn.