What happens when ailing dictators die? The world has 55 dictatorships — and 11 autocrats age 69 or older.

Political observers see a dictator’s death in office as a catalyst for political change — either creating an opening for political liberalization or triggering elite infighting and instability. So what happened in Uzbekistan, where President Islam Karimov, age 78, died on Sept. 2?

Karimov had led the country with a “ruthlessly authoritarian approach” since independence in 1991. Despite attempts to guard information about his health like a “state secret,” rumors that Karimov was gravely ill had surfaced several weeks before his death. These rumors sparked a flurry of commentary on the country’s future.

But two months after Karimov’s death, Uzbekistan’s political system remains intact. Power passed seamlessly to Prime Minister Shavkat Mirziyoyev, who is serving as interim president until elections on Dec. 4 make his position official. Although the transition bucked the constitutionally established guidelines (power should have passed to the Senate chairman), Karimov’s passing has been remarkably unremarkable.

Is Uzbekistan’s experience typical?

We looked at what happens when an autocrat dies in office

In a new piece in the Journal of Democracy, we find that Uzbekistan’s experience is broadly representative of what happens when dictators die in office. We looked at data on all 79 autocratic leaders who died in office of natural causes from 1946 to 2012 (using data from here and here).

Like Karimov, most autocrats resist identifying a successor out of fear that doing so might enable a competitor to establish a base of support that could be mobilized to unseat them early. Moreover, autocrats who die in office tend to be longtime occupiers of their position. They have tenures lasting an average of 16 years, compared to just seven years for leaders who leave power through means other than death. This longevity in office enables these leaders to portray themselves as indispensable to the political system.

For these reasons, a high degree of uncertainty about a country’s future trajectory is a common feature in countries with aging or ailing leaders, particularly in the days and weeks surrounding a leader’s passing.

However, we find that such concern is often misplaced. Death in office seldom leads to near-term liberalization. And only rarely does it precipitate coups or protests or the end of a regime.

We found that 87 percent of the time that leaders died in office, the regime — or group in power and rules for governing — remained intact the following year. And in 76 percent of cases, it was still in power five years later.

Compare these statistics to regime survival after other forms of leadership transition, like coups, revolts or term limits. In these other transitions, only 43 percent of regimes remain intact the following year, and only 32 percent five years later. The figure below shows regime survival rates in the one- and five-year period following a leader transition, broken down by the type of exit.

In the rare cases that regimes do collapse in the wake of a leader’s death, it’s likely a new dictator will emerge. Of those cases in which a leader’s death prompted regime change, 70 percent spurred the establishment of a new autocracy by the following year; only 30 percent led to democratization. Other forms of leadership transition resulted in a somewhat more balanced outcome: New dictatorships arose in 54 percent of cases and democracies in 46 percent.

Death in office infrequently creates instability

We also find that coups and protests are rare following a dictator’s death. During the year of a leader’s death in office, coups occurred in only 6 percent of cases, compared with 32 percent when autocrats have left power via other means. Similarly, mass public protests are far less likely to break out following a dictator’s death than after other forms of authoritarian leader exit. This pattern persists even when we adjust our time frame and look at the five-year period following a leadership transition.

So what is it about situations where leaders die in office that predict regime stability? Most forms of leadership transition have political motivations and are based on the decisions or actions of political actors. Leadership transitions resulting from coups and protests, for example, signal a disgruntled elite and citizenry.

When a leader dies in office, in contrast, they leave behind a set of regime actors who, up until that point, have opted to support the leader’s tenure rather than challenge it. When a dictator dies, therefore, the same elites have incentives to coalesce around a new successor rather than engage in political bickering and infighting. To act otherwise would risk destabilizing the system and endanger their privileged access to power.

In a small number of cases, a dictator’s death did precipitate regime instability. Our study found that recent experience with protests or coups, and the absence of a strong political party or other institutionalized succession mechanism (like a politically active military or royal family), increased a country’s risk of instability following a leader’s death.

We found, for example, that 22 percent of highly personalized dictatorships (those regimes lacking strong parties or a military) collapsed when the leader died — compared to 6 percent of institutionalized dictatorships. Although instability risk in highly personalized settings is comparatively higher, the actual prospects remain low. Even when institutional channels for handling succession are weak, the elite have strong incentives to rally around a new leader.

What does this mean for the roughly 20 percent of the world’s dictatorships that may see their leaders die in the next few years? The most likely outcome for most of these regimes will be a persistence of the status quo. However, these factors — concentration of power and recent experience with protests — suggest which countries are most more likely to deviate from the smooth transition that normally follows a leader’s death in office. At present, these factors suggest that Zimbabwe is at elevated risk of instability if the 92-year-old Robert Mugabe dies in office.

Although Mugabe’s political party, ZANU-PF, provides a focal point to coordinate the succession process, the #thisflag protest movement and other “hashtag pop-up protests” have been challenging the regime for months. These movements create the networks, experience and knowledge that facilitate mobilization should discontent over a leadership transition arise.

Though concerns — or optimism — that a longtime dictator’s death in office will prompt political change are common, the historical record shows that change rarely transpires. Dictators who die in office tend to leave behind a set of elites who are highly motivated to get on with business as usual. Though most dictator leadership transitions create openings for political change, death in office is not among them.

Erica Frantz is assistant professor of political science at Michigan State University. She specializes in the politics of dictatorship and is a collaborator on the NSF-funded Autocratic Regimes Data Set.

Andrea Kendall-Taylor is deputy national intelligence officer for Russia and Eurasia at the National Intelligence Council and nonresident senior associate in the Human Rights Initiative at the Center for Strategic and International Studies. Some of the data used in this research was funded by the National Science Foundation (BCS-0904478 and BCS-090463).