When it comes to coups, timing is a curious pattern.
This post is a guest contribution by OEF/University of Denver post-doctoral fellow William Akoto. William’s work covers the intersection between foreign investment, trade and aid on international relations and coups in the developing world.
On June 8 2003, a group of army officers in Mauritania staged a coup attempt to depose incumbent president Maaouya Ould Sid Ahmed Taya.
The debate surrounding the timing of coups and their success or failure raises an interesting question — is there ever really a good time for a coup?
After a 36 hour gun battle, 15 people were dead and the rebellion was over. One hundred and twenty-nine officers suspected of involvement in the coup were swiftly rounded up and put on trial months later on charges of high treason, assassination and sabotage.
More recently, the Turkish government arrested 223 serving military officers across the country in its continuing efforts to round up persons suspected of involvement in the failed July 2016 coup.
In the three years since the failed coup, more than 77,000 people have been jailed pending trial and over 150,000 military personnel and civil servants have been suspended from their jobs or sacked for their suspected support or involvement in the coup attempt.
Planning a coup? Aim for a Wednesday in November.
Debate about why coups fail still rages on but there is some suggestion that coups may fail because they are poorly timed.
In the case of the Turkish coup attempt, perhaps the conspirators should have aborted the coup once word leaked to state intelligence agencies. Perhaps they should have delayed it until they had sufficient broad support within the military and among the domestic political elite.
The debate surrounding the timing of coups and their success or failure raises an interesting question — is there ever really a good time for a coup?
Assuming coup conspirators have broad support within the military and civil society and manage to keep the plot sufficiently hidden from state security agencies, is the coup guaranteed to succeed or does the specific timing of the coup matter?
The academic literature provides little guidance on this issue, but historical coup data provides interesting clues.
Coup data from 1950 to present, compiled by Political Scientists Jonathan Powell and Clayton Thyne, suggest that July is the second most popular month for coups, tied with April and October (March and November are joint first place).
The failed 2016 Turkish coup attempt was in July.
Picking July for their coup attempt might not have been a bad choice — 55% of all coups attempted in July succeed, the third highest success rate for any month.
For the Mauritanian coup conspirators, June was a terrible month for a coup. Only 42% of all June coups succeed, the second lowest success rate (behind December) for any month of the year.
November or August, with success rates of 57% and 56% respectively, may have been better choices based on a reading of historical patterns.
But what about the specific day a coup takes place?
The magic appears to happen on the 24th of the month. A whopping 86% of all coups attempted on this date succeed.
The Turkish conspirators chose the 15th of the month for their coup and the data further suggests this might have been bad timing. Historically, coups staged on the 15th of the month have a relatively high failure rate of 48%.
It also certainly didn’t help that the 15th of July was a Friday. According to the historical coup data, 49% of all Friday coups fail.
In the Mauritanian case, June 8 was a Sunday — even worse than Friday for a coup attempt. Fifty-six percent of all Sunday coups fail, the highest of any day.
In general, coup conspirators appear to actively avoid staging coups on weekends. The weekend is for sports, meeting up with friends or catching up on some yard work.
As a result, adding a coup to the weekend to-do-list may be a stretch too far for most putschists.
Conspirators also appear to prefer Thursday coups. However the data suggests Wednesdays actually offer the better chance for success — 55% of all Wednesday coups succeed (although Thursday is not too far behind at 54%).
In reality, coup attempts are highly complex and the structural conditions permitting coup plotters to go forward involve a multitude of factors. But historical patterns in timing regarding success and failure remain curious.
We may never know why a particular coup happens on any specific day but future coup plotters may want to think carefully about when to schedule D-day.
A Wednesday coup on the 24th of November might be a good bet. The next November 24th that falls on a Wednesday is in 2021 and one may want to keep an eye on such an auspicious date.
*Thanks to Clayton Besaw and Matt Scott Frank for comments on an earlier version of this post.
