Can elections remove an autocrat like Turkish President Recep Tayyip Erdogan from power? If you pose this question to Turkey watchers in Western capitals to get their take on the country’s upcoming election, you will get a resounding “no” from a significant number of them. Some will say Erdogan is still very popular—or at least adept at mobilizing his followers. Others will argue that elections do not matter in the entrenched autocracy he has built; one way or another, he will find a way to stay in power. Take the Western conventional wisdom about this Sunday’s election with a grain of salt, and here’s why.

Can elections remove an autocrat like Turkish President Recep Tayyip Erdogan from power? If you pose this question to Turkey watchers in Western capitals to get their take on the country’s upcoming election, you will get a resounding “no” from a significant number of them. Some will say Erdogan is still very popular—or at least adept at mobilizing his followers. Others will argue that elections do not matter in the entrenched autocracy he has built; one way or another, he will find a way to stay in power. Take the Western conventional wisdom about this Sunday’s election with a grain of salt, and here’s why.

Read more of FP’s coverage of Turkey’s pivotal elections.

Erdogan is indeed a popular leader. He commands somewhere between 40 percent and 45 percent support, no small feat after 20 years in power. But he is not nearly as popular as he once was. In the 2018 presidential election, Erdogan captured 52 percent of the vote, or some 26 million votes. Several factors worked in his favor then. The elections were held just two years after the failed 2016 coup and its “rally-around-the-leader” effect. Erdogan was riding high on a wave of nationalism after the Turkish military intervened in the Syrian civil war to fight the Syrian Kurds. The country was not suffering from a major economic crisis like today. The opposition was fractured: The popular Peoples’ Democratic Party (HDP) co-chair Selahattin Demirtas, Iyi Party leader Meral Aksener, Republican People’s Party (CHP) candidate Muharrem Ince, and Felicity Party leader Temel Karamollaoglu were each on the ballot running separately against Erdogan. The nationalist base was more unified, with the majority still backing Erdogan; the nationalist breakaway Iyi Party had been established too recently to draw away much of the vote.

Fast-forward to 2023. To win the election, Erdogan has to capture more than the 26 million votes he secured in 2018 because Turkey’s voting population has grown. His problem is that he faces a dramatically different political context that makes that task very difficult. The failed coup’s rally-around-the-leader effect is long gone. The wave of nationalism that Erdogan once rode has come back to haunt him: There is now a growing nationalist opposition to Erdogan, with several nationalist parties peeling away votes from his far-right ally, the Nationalist Movement Party (MHP). The Turkish economy has plunged into a major crisis, with runaway double-digit inflation and soaring food prices. Most importantly, the opposition is more united than it has ever been: Six parties have come together under the Nation Alliance banner and a single presidential candidate, CHP leader Kemal Kilicdaroglu, with additional backing from the pro-Kurdish HDP. Altogether, Kilicdaroglu commands 50.9 percent of the vote, according to the latest poll.

Skeptics might say that these arguments and poll numbers would only be relevant if Turkey were a democracy and add that Erdogan has so much to lose that he would do anything to secure victory. They have a point. It is easy to be cynical about elections in a country run by an entrenched autocrat who has demonstrably manipulated previous votes and refused to accept the results when they haven’t gone his way. In the June 2015 parliamentary elections, Erdogan’s ruling Justice and Development Party (AKP) lost its parliamentary majority. Erdogan stalled talks between the AKP and the CHP about forming a coalition government and forced new elections. He renewed the fight against the outlawed Kurdistan Workers’ Party to reverse his party’s defeat in new elections held that November. In 2017, Turkey held a controversial constitutional referendum to switch to an executive presidency that would grant Erdogan unprecedented powers. The referendum, which Erdogan won by a narrow margin, was marred by widespread allegations of fraud. In both the parliamentary elections and referendum, the opposition was not organized enough to protect the ballot or challenge Erdogan’s efforts to create a fait accompli.

In 2019, however, things changed. Erdogan’s party lost almost all of Turkey’s major cities in municipal elections. Particularly frustrating for Erdogan was the loss of Istanbul, the financial capital where he had launched his political career. Erdogan did not accept the opposition’s narrow win in Istanbul and called for a rerun. When the election was run again, the ruling party lost by a much bigger margin. Erdogan abusing his power to deny the election result had the effect of mobilizing the opposition.

What does this tell us about elections in Turkey? That they are popular and fraud is not, making heavy-handed election fraud risky for Erdogan. The 2019 elections made something else clear, too. When the opposition parties get their act together, they can beat Erdogan at the ballot box. Skeptics might point out that the stakes are much higher for Erdogan in the upcoming vote than they were in the 2019 municipal elections and that he will not accept defeat gracefully. They are not entirely wrong. In personalist autocracies like Turkey, rulers who lose power are likely to end up in jail or exile, so they risk everything to cling to power.

What are Erdogan’s options in a scenario where he loses the vote by a small margin? He might declare that the election was stolen and ask the Turkish bureaucracy to back him up. That Turkey’s top electoral body and security bureaucracy will heed his call is not a foregone conclusion, however. A recent decision by the electoral watchdog to turn down a request for voter data from the Erdogan government, part of an effort to create a new online election monitoring system, is a case in point. In March, the Constitutional Court ignored Erdogan’s objections when it reversed a previous decision to block the HDP from receiving allocated state funding to finance its electoral campaign over its alleged ties to militant groups. These and other decisions by key state institutions suggest that Turkish bureaucrats are hedging their bets. They are unlikely to back Erdogan after an election loss and risk legal repercussions under the new government.

Similarly, opposition supporters optimistic about finally beating Erdogan would be more likely to take to the streets if they think the election was rigged or its outcome denied.

A smarter option for Erdogan would be to accept the result and wait for the new government to fail. He still has a strong following he can mobilize for this purpose. Given the enormous economic problems an inexperienced new government would have to address, surging back through democratic means is not impossible—especially if the current opposition makes good on its pledge of switching to a reformed parliamentary system, which would open a path for Erdogan to return to power as prime minister.

Finally, one might expect Erdogan to fight tooth and nail to stay in power in order to avoid facing trial. But according to Turkish law, an indictment would have to be approved by two-thirds of parliament, a supermajority that would be very difficult to achieve—not least because the opposition includes former key Erdogan allies who might get sucked into any investigation, an outcome the opposition will likely want to avoid. The fact that a trial and potential jail time are unlikely makes it easier for Erdogan to accept defeat.

All of this is to say that not all autocracies are created equal; Turkey is neither Russia nor China. In some, elections matter more than in others, and strongmen are weaker than they seem.