Predicting the next uprising

February 24, 2011

One casualty of the uprisings in the Middle East has been the professionals who didn’t see them coming. The International Monetary Fund has taken a hit for its April 2010 report on Egypt, which praised the country’s ‘‘sustained and wide-ranging reforms since 2004,’’ noting they had made the economy more durable and less vulnerable to external shocks. Ditto the C.I.A., whose director, Leon Panetta, endured the very personal ignominy of seeing his public predictions to Congress proven wrong within hours of making them.

For anyone who watched the collapse of the Soviet Union or the 2008 financial crisis, there is something very familiar about this failure of the experts. There seems to be something about swift, massive paradigm shifts — whether they are the bursting of a financial bubble that has been years in the making, or a popular revolt against a political regime that had been stable for decades — that we find hard to anticipate.

Research by behavioral economists like Dan Ariely of Duke University has suggested that part of the problem may be that when we have a vested interest in the status quo our brains are wired to view it as good and stable. Dr. Ariely’s work has focused on the cognitive blinders our financial self-interest imposes. But a similar bias may shape the views of political experts, who can end up developing a sense of ‘‘ownership’’ of the national elites they study that seems to be nearly as powerful as the proprietary feeling bankers had for the credit derivatives they created.

In a prescient book about democracy and authoritarianism written before he went to work at the White House, the political scientist Michael McFaul argued that assumptions of regime stability are always dominant, and that, when those regimes are authoritarian, these assumptions are always wrong. Dr. McFaul strenuously disagreed with that default view, arguing: ‘‘assuming that the current configuration of autocratic regimes in play today will persist 50 years from now is much more naïve than believing that some of these regimes might succeed in making the transition to democracy.’’

Dr. McFaul’s conviction is looking pretty good today. But even if we are able to overcome our psychological resistance to the very notion of regime change, anticipating precisely when dictators will be toppled may not be possible. ‘‘By their very nature, these tipping points are not predictable,’’ said Daron Acemoglu, an economics professor at the Massachusetts Institute of Technology.

A better way of thinking about whether regimes will endure, he suggested, might be to try to understand the potential for rebellion, given the right catalyst. ‘‘Most of the time it’s dormant and hence there is no predictability of uprisings,’’ he argued. ‘‘But once we enter into a critical period like the current one, this latent factor has some predictive power.’’

In that spirit, my colleague Peter Rudegeair and I have done a back-of- the-envelope calculation to identify countries with a high latent potential for uprisings. We considered four factors — political freedom (on the grounds that democracies don’t usually require popular rebellions to achieve regime change), corruption, vulnerability to food price shocks and Internet penetration. Our spreadsheet used publicly available measures of the four factors and came up with a list of 25 most vulnerable countries.* You can see the spreadsheet explaining the publicly available measures of the four factors we used and the top 25 countries we came up with here. Libya, Algeria and Egypt made it into the top 10. Perhaps more surprisingly, so did Russia, Uzbekistan, Azerbaijan and Venezuela.

According to, the U.S. military and intelligence agencies have spent more than $125 million over the past three years on computer models that try to forecast unrest. Bearing that in humbly mind, this fast and dirty calculation is meant to provoke discussion, not to pinpoint the next hot spot.

Dr. Acemoglu suggested that one way to refine this sort of calculation would be to consider ways in which the different factors that make a regime vulnerable to revolution interact: ‘‘For example, a lot of corruption without any Internet penetration or a lot of Internet penetration without corruption may create no pressure for uprisings, but when both of them are present that might be a whole different ballgame.’’

Lucan Way, a professor of political science at the University of Toronto, said another contributor to regime fragility that it would be worth factoring in to a more sophisticated analysis (you can try this at home!) is whether the authoritarian government is itself the product of recent revolutionary struggle. Dictatorships run by an ideologically united revolutionary party — Iran, for instance, and to a lesser extent China — are, Dr. Way argued, more durable than those whose rulers rely purely on guns and patronage.

Food-price shocks are often the catalyst that tips a regime with a latent vulnerability to an uprising into one facing people in the streets: that was the case in Tunisia, and has been true as far back as the Bolshevik Revolution. Something else that can propel a society with a latent potential for rebellion into action is the demonstration effect, or what Dr. Acemoglu calls ‘‘contagion,’’ a phenomenon also familiar to anyone who was caught in the wildfire global spread of the financial crisis in 2008.

In both cases, the sudden belief that a previously stable status quo could change had the power to alter reality. This interplay between perception and fact is what George Soros, an expert in paradigm shifts in both markets and countries, calls reflexivity.

Even some of the world’s most powerful authoritarian regimes seem to be getting concerned about the danger of contagion and the power of perception. Hence China’s efforts to block electronic information and discussion of the uprisings in the Middle East. The Kremlin may have even more reason to worry: A Russian opinion poll found that one-third of respondents thought the ‘‘Egyptian scenario’’ of mass protests was possible in Russia. That is the kind of thinking that can tip a latent potential for rebellion into a revolution.

*Update: For ease and simplicity, we used Nomura’s Food Vulnerability Index to calculate how rising food prices would affect a country’s domestic economy.  Because Nomura limited their Food Vulnerability Index to 80 countries, our uprising  index is also limited to 80 countries.  This explains why some countries that seem like they would be prime candidates for having a high latent potential for rebellion — like Iran, Cuba, and Jordan — do not make our list.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

Thank you for another excellent article.

A couple of things I thought might be worth considering as other factors for rebellions would be the population demographics of a country, and the regularity of state sanctioned violence.

The populations of Egypt and Tunisia are very young, and lopsidedly male. My guess is that the young males would have a higher propensity to engage in aggressive tactics as part of a protest, and be more willing to put up with violent countermeasures by the state. While hardly objective evidence, it seems that in any pictures/footage of violent protests worldwide, the crowd is made up of primarily young males. ationGateway.php

Making a guess again here, but I’d imagine from an individual’s perspective, most people would try to keep their heads down to protect themselves and their families, even when a regime is immensely unpopular; nobody wants to be in the firing line. I’d imagine the tipping point for regime change occurs when the individual’s situation becomes so desperate, that personal survival seems more likely through an upheaval than in continuing day to day life. Here, high food prices create a situation where many people find themselves facing the risk of starvation, however I think a background of state violence is needed to create the feeling among individuals that they have no other options. This would also foster an “us vs them” attitude among the protestors; as seen in the middle east uprisings where many different groups with differing views on an eventual state have banded together against a common enemy.

I’m not nearly well travelled enough to know how the high scoring countries in your spreadsheet are affected by the above factors, so I’d be curious to know anyone else’s thoughts on this.

Posted by Kyle-M | Report as abusive

How come Zimbabwe is not on this list?
We have seen extreme poverty, dictatorial tendencies, mayhem, highest food price rise (even non availability), aver 20% population forced to run away from the country and still no revolution?
any explanation for this dichotomy?

Posted by Chisha | Report as abusive

I agree with you Kyle; however, I would offer a slightly more concise description: cultural pre-dispensation (historical propensity to rebel).
For evidence supporting cultural pre-dispensation I would offer Afghanistan: though not in Dr. Acemoglu and Peter Rudegeair’s spread, it’s life expectancy of 46 years old (lowest in the world), corruption (one of the highest in the world) and extreme vulnerability to food prices would combine for a very high score in their system.
Couple those factors with the opposite durability situation of Lucan Way (multi-party democracy in a tribal honor system instead of dictatorship run by an ideologically-united revolutionary party), and cultural pre-dispensation appears to provide good evidence supporting the ‘propensity to act’ at the catalyst point (a food-price shock).
In other words, I’m saying cultural pre-dispensation is the dominant factor of the ‘latent vulnerability’ described by Acemoglu and Rudegeair.

Posted by steve.thompson | Report as abusive

The failure to be able to predict is because its raw frustration not fully tainted by outside influences. Its human reaction not stupid cows being prodded by thier masters. Theres no shadow group propelling the people. No clandistine operation to muster the malice. Its just sheer simple people who want to live with dignity and live a simple ordinary peaceful life. And not owe that life to anyone. They themselves are standing on thier own feet for themselves by themselves and demanding decent government. You can turn this into a rocket science but you will miss the essence in your quest to intellectualize something so fundamental in us all.

Posted by MrEz | Report as abusive

I hope that people of communist nations will do the same to have their human rights back.

Posted by BillNG | Report as abusive

This seems awfully close to the “Shoe-Thrower’s Index” that was in The Economist on 2/9. See t/2011/02/daily_chart_arab_unrest_index

Posted by Alex02129 | Report as abusive

I think the next uprising will be in Illinois!

Posted by MikeChicago | Report as abusive

I think the most telling statement in this great article is that policy makers tend to believe what they want to believe. Penetta made the comments he did because of two things: 1) a mindset in Washington that our dictator buddies were stable and had things under control; 2) not enough case officers on the ground with agents gathering human intelligence and then the folks in Washington listening to them rather than ignoring them. We can’t know what’s going on without humans on the ground. We can’t keep have out political leaders telling the intelligence people what truth is. This not only goes for the United States but all countries.

Posted by neahkahnie | Report as abusive

The Media factor is missing from my point of view, Aljazeera news channel was crucial in three uprisings till now, once Aljazeera starts digging it gives the courage to people to go to streets, you will see that the next regime candidate will top the news once Libya is done.

Although, the mobile phone with camera makes an every person a journalist …

Posted by ebrhim | Report as abusive

Maybe the Arab administrations have neglected domestic affairs. Maybe also some of those in power may have sat on the power for far too long. This may have bred boredom coupled with the hardship due to waning Western opportunities as a result of increasing Western interest in Asia proper like China and India, following the financial crisis, people may just want to vent their frustration some how, some way. What better way and target than on those who had long been in power? Maybe this is the reason. Persians in contrast is less susceptible to such problems because their population is not huge, which means manageable and their people are more concerned with their branch of religion as they are wary of certain threats from other religions as well as other branches of their religion which creates a critical need for them to stick togather no matter what. Besides their people can easily slip out to make a living. As for China and India, both are equally and more susceptible to protests and it has nothing to do with democracy or communism. Whether it is a one party or a multiple party state, at the end of the day it is the quality of administration and the importance that the people/team at helm shouldn’t over stay even though they might have done well. Corruption, for example is an element that may spark protests in India and China and it can be contagious for example if a mass violent protest happens in India, this may affect China and vice versa.

Posted by vision966 | Report as abusive

Policy makers have yet to take into account the impact that rapid communication and the internet play.

These days it is harder for Authoritarian regimes to control what their people hear or read. Text messages, social media, real time overage from news sources not controlled by the government all let people respond faster and coordinate better.

This is why Egypt tried turning off their cell phone network and disconnecting the internet, and why North Korea is so upset with fliers and CDs being flown over their border – as long as the government controls the news (like the Soviet Union did and Venezuela is trying to do) they think that they can control their people, and that is no longer the case.

Posted by Eric.Klein | Report as abusive

I really think you should add in average age of the population. Or percentage of people under 25. Not only do people grow more cynical and conservative with age, people with no kids are a lot less easy to scare.

Posted by Dafydd | Report as abusive

How frustrating this conversation is! Please by all means, quantify all the factors of rebellion and figure out the right function. But what will you get? An after the fact post-diction of factors that correlated with regime change in the past. I have no problems with these efforts to determine when the field for civic movements is fertile, but please don’t forget what everyone here is missing: The creative, strategic planning and action of brave civic actors who can do marvelous unexpected things to undermine the legitimacy and support of authoritarian orders. These people will continue to tirelessly work for change–even where the models above show the fields for action are barren. And quite often, as they have in the past, they will prove us all to be no better than historical ambulance chasers masquerading as behavioral scientists. (-: Okay, back to my large-n data set, sniff.

Posted by jagou | Report as abusive

It would be interesting to consider how this idea would apply to the US, but I think it would also be quite difficult to determine this. What would make it an interesting study: Poverty in the US has increased and deepened considerably over the past 20 years or so, while much of the social safety net has been torn out, but these issues are pretty much ignored by the media (mainstream and alternative). Job security is now largely a thing of the past. Lose your job today, and you might lose everything – even your family, possibly your life. The longer
these issues — ranging from exporting US jobs at taxpayer expense to our lack of a social safety net — continue to be ignored, the higher the stress level gets.

Posted by DHFabian | Report as abusive