Forecasting Instability: The Case of the Arab Spring and the Limitations of Socioeconomic Data

All stable countries are alike, but all unstable countries become unstable in their own ways. Not surprisingly, the World Bank’s report on the causes of the Arab Spring noted that standard development indicators failed to predict the outburst of popular anger that catalyzed the unrest and revolutions of 2010 and 2011. This paper questions whether economic and socioeconomic trends alone could have held the keys to predicting the instability in the Middle East and North Africa (MENA) region that began in late 2010 and posits that an overreliance on economic data can lead to misguided predictions.

Without question, the trajectory of the unrest and revolutions was fed not just by economic conditions, but by the nature of the governments; the responses of the populations of Egypt, Libya, Tunisia, and Yemen to their real and perceived economic conditions no doubt would have been different had their governments been different. But here, I aim to tackle economic data specifically to determine which indicators proved ineffective at predicting social unrest and what other data might prove more diagnostic for future analysis.

The best analysts are adept at identifying “structural factors” – the long-term, root causes that lead to unrest and revolutions. These factors consist of economic decline, growing inequality, corruption, political repression, and failing infrastructure and social services, among others. The “black swans” or spontaneous occurrences that trigger destabilizing events, however, are by definition virtually unpredictable. The rapidity with which societal forces shook much of the MENA region in 2010 and 2011 surprised most experts, but they were most surprised by the timing and the unexpected catalyst – the self-immolation of a poor fruit vendor in central Tunisia in late 2010 –  and not by the underlying conditions. These were well known and written about for years,[i] much like experts’ assessments of the Soviet Union before its demise. [ii]

Socioeconomic Data Alone Will Disappoint

“Judging by economic data alone, the revolutions of the 2011 Arab Spring should have never happened. The numbers from the decades before had told a glowing story: the region had been making steady progress toward eliminating extreme poverty, boosting shared prosperity, increasing school enrollment, and reducing hunger, child and maternal mortality,” a World Bank article from 2015 concluded.[iii] Most MENA societies had reached Millennium Development Goals related to poverty reduction and access to infrastructure services prior to 2010 and 2011 and had made important strides in reducing hunger and infant mortality and in increasing school enrollment. Separate World Bank research indicates all countries within the MENA region showed significant educational achievement from 1970 to 2010, and collectively, the rate of growth in educational attainment was faster than other developing regions.[iv]

Strong economic growth is a red herring for predicting stability; similarly, low economic growth is not always associated with instability. What these aggregated and topline numbers neglect are nuances and wealth distribution, along with public satisfaction, services, health care, clean water, affordable housing, and political views. For much of the Arab world, economic growth accelerated or was at least steady in the decade prior to the 2011 uprisings, with GDP per capita (as measured by purchasing power parity, PPP) increasing by double digits from the 1990s. Average GDP per capita in Egypt increased by nearly 40 percent from the 1990s to the end of the 2000s and by 47 percent in Tunisia, based on data from the World Bank’s World Development Indicators.[v]   

The “black swans” or spontaneous occurrences that trigger destabilizing events, however, are by definition virtually unpredictable. 

The use of GDP as an indicator of societal welfare has long been debated. In 1959, economist Moses Abramovitz wrote, “We must be highly skeptical of the view that long-term changes in the rate of growth of welfare can be gauged even roughly from changes in the rate of growth of output.” GDP (or GNP) as we know it today was developed in the 1930s to measure changes to national income and economic activity during the Great Depression. It was never meant to be a measure of social welfare, much less public opinion.  

[[{"fid":"108086","view_mode":"default","fields":{"field_file_image_alt_text[und][0][value]":"gcppercapmena","field_file_image_title_text[und][0][value]":"","field_file_source[und][0][value]":"World Bank, World Development Indicators","field_file_caption[und][0][value]":""},"type":"media","attributes":{"alt":"gcppercapmena","class":"media-element file-default"}}]]

Another reason to doubt the effectiveness of using GDP changes as a diagnostic tool for assessing political stability is that the data is often inaccurate, especially for poor countries with few resources for gathering data and statistical analysis. However well trained and professional a country’s statistical authorities may be, aggregating literally thousands of figures from scores of sources (often with varying degrees of thoroughness) can easily lead to misleading results.   

Even with data in hand, the work of defining what it means and how best to analyze the figures comes to the fore. For instance, the UN’s Economic and Social Commission for Western Asia published an exhaustive analysis of the Arab middle class in 2015 that highlighted the limitations in defining what exactly the middle class is; by some estimates it could range from less than 5 percent of the population to more than three-quarters of the population, depending on how the data was defined. Therefore, consideration of how a policy change will affect the Arab middle class immediately begs the question: what is the middle class?

Qatar showed real GDP growth of nearly 26 percent in 2007 – mostly on the back of high natural gas prices – when the growth average for the rest of the world was 5.5 percent, according to IMF data. But what does such an astonishing growth figure even mean? In 2011, Libya’s GDP fell by an estimated 67 percent when Muammar Qaddafi was deposed and civil war ensued, only to rebound by 125 percent the following year. These figures are meaningless. In her book on GDP, Diane Coyle writes, “It is one of the distasteful aspects of a disaster that the immediate consequence is a boom in GDP growth. GDP does not measure a nation’s assets or balance sheet, only its flow of income, expenditure, and production from year to year.” [vi] Economic activity generated from replacing shattered windows and money spent on cement for destroyed city blocks are not indicators of a healthy economy, but one in shock.

Finding the Right Data

Superb post-mortem analysis has been done to examine the causes behind the Arab revolutions that began in late 2010. Some studies point to specific causes, such as income inequality, unemployment, corruption, or purely political issues such as repressive and unresponsive regimes. One study in 2013 correctly highlighted the multiplicity of factors:

“In the preceding analysis we have shown how structural factors such as deteriorating economies, the uneven distribution of economic resources, the spread of poverty and unemployment, the repressive violent nature of the Arab regimes and corruption coupled with catalytic factors such as the self-immolation of Mohammed Bouazizi (Tunisia) and the arrest and torture of Khaled Said in Egypt and Fathi Tirbal in Libya have been the major causes for the popular uprisings that swept the entire Arab region in 2011.”[vii]

In other words, no single factor predominated. Many structural factors over a long period merged with spontaneous events that prompted the unrest and revolutions. Yet, perhaps no other economic indicator has been more analyzed as being the proximate cause for the Arab Spring revolts than high and persistent rates of youth unemployment. On the eve of the unrest, in 2010, the International Labour Organization (ILO) published data indicating that unemployment among Arab youth was the highest in the world.[viii] The youth unemployment rate in Tunisia was about 30 percent in 2009 and in Egypt, around 25 percent. According to ILO’s analysis, youth unemployment was mostly concentrated among the educated. Somewhat paradoxically, with decreasing levels of absolute poverty and increasing levels of education, youth unemployment in the MENA region generally is higher than in other regions; young people appear less motivated to accept unattractive jobs at the bottom end of labor markets that do not match their skills.[ix]

“Judging by economic data alone, the revolutions of the 2011 Arab Spring should have never happened."

One year before protests took down Tunisia’s autocrat, Paul Rivlin wrote in his book, Arab Economies in the Twenty-first Century, “Tunisia is in many respects a model for other Arab countries to follow. It has a stable economy with growth, good human development indices, a relatively large manufacturing sector plus exports, but it also has high unemployment.”[x] This combination of relatively high levels of education and an inability to find work helped create that potent recipe for instability. The ILO report judged that in Arab countries, some 40 percent of high school and university graduates between ages 15 and 25 could not find work upon entering the job market, fueling a trend of high numbers of unemployed among the educated.

This assessment is incontrovertible, but high youth unemployment had been a regular feature of Arab economies for years. Average Arab youth unemployment rates in 2009 had barely changed from the high rates in 2000 and in some cases were marginally lower. Assessing all Arab economies, Rivlin concluded, “The very high levels of youth unemployment pose a serious threat to social and political stability and have persisted for years.” So what made 2010 and 2011 different?

[[{"fid":"108091","view_mode":"default","fields":{"field_file_image_alt_text[und][0][value]":"unempgraphmena","field_file_image_title_text[und][0][value]":"","field_file_source[und][0][value]":"International Labour Organization, ILOSTAT database","field_file_caption[und][0][value]":""},"type":"media","attributes":{"alt":"unempgraphmena","class":"media-element file-default"}}]]

Next to youth unemployment, corruption in government and among the elite has been touted as a primary cause of the unrest, especially when combined with real and perceived income and wealth inequality. As in Tunisia, the conditions in Egypt made the country ripe for revolution. In the preceding decade, corruption in the country had essentially given up any pretense of subtlety.[xi] The connected and affluent built gated communities, while most Egyptians lived in "informal" housing and shanties.[xii] Egyptians had enough by 2011.   

High levels of corruption—as defined by Transparency International’s Corruption Perception Index (CPI)—are correlated with weaker standards of living and government services. However, in 2008, Tunisia ranked relatively well in the region on the CPI, on par with Kuwait, and ranked higher than Algeria, Saudi Arabia, and Morocco, countries that experienced large but manageable protests during the region’s unrest. [xiii] Egypt’s CPI score of 3.1 in 2010 was no doubt low, but it was essentially unchanged from 2000. (The higher the score, the lower the perceived level of corruption – Denmark, for example, ranked the highest in 2010, with a score of 9.3.) As with other indicators, government corruption is usually necessary but insufficient for fueling public grievances, as it works in concert with other negative socioeconomic trends.

Unlike income and wealth inequality, however, determining and tracking changes in inequality of opportunity—a more qualitative endeavor—is more difficult.

Similarly, based on multiple analyses of data, there was no strong direct causal influence of income inequality on the Arab uprisings.[xiv] Income inequality, as measured by the Gini coefficient for the distribution of household monthly real per capita total consumption expenditures, ranged between 31 percent in Libya and Egypt to more than 40 percent in Tunisia and Yemen prior to 2010. (The coefficient falls between 0 percent, where everyone earns the same, and 100 percent, where one person owns all the wealth.) The countries at the center of the storms of unrest and revolution showed different patterns of income and inequality: low income and low inequality in Egypt, low income and high inequality in Yemen, medium income and low inequality in Libya, medium income and medium inequality in Syria, and relatively high income and inequality in Tunisia. In short, no pattern of inequality existed among those states with the highest degrees of civil unrest.

Although income inequality as an indicator of potential unrest is not particularly diagnostic, research shows that inequality of opportunity might provide greater insight, as it goes to the issue of fairness.[xv] Although Egypt’s income inequality was moderate by global standards prior to 2011, most Egyptians found that their ability to succeed economically was constrained by a lack of opportunity. For example, Egyptians bristled at inequality caused by gender, ethnicity, family background, and place of birth, which in turn contributed to persistent income inequality. When viewed against the rising perceptions of corruption in government and among the elite, lack of opportunities for the average Egyptian became more glaring. Compared to income and wealth inequality, however, determining and tracking changes in inequality of opportunity – a more qualitative endeavor – is more difficult. 

The Use and Misuse of Aggregated Data

A number of independent institutions and private analytic firms have done comprehensive work on state fragility and instability prospects. Among these are the Fragile States Index (FSI), published by the Fund for Peace, and the Economist Intelligence Unit’s various publications. These databases and scores provide an enormous amount of information relevant to analyzing the stability or fragility of governments and societies, including economic decline, state legitimacy, public services, human rights and rule of law, and demographic pressures, among others. FSI’s country analysis is exhaustive and contains exactly the type of information experts would expect when modelling country stability.

Alas, as with solely relying on socioeconomic data and trends to approximate political stability in a country, aggregated data that include political as well as social and security issues still fall short of predicting widespread unrest or a government’s demise. To be fair, none of these databases and rankings aims to forecast whether or not a country will or will not become unstable. Rather, they aggregate data from multiple sources and disciplines to highlight those factors that would typically contribute to unrest or expand public grievances against the government. In short, they are adept at identifying underlying trends but, not surprisingly, ill-equipped at pinpointing the catalysts.

In 2009, a special report by the Economist Intelligence Unit looking at the sociopolitical ramifications of the global financial crisis ranked countries from 1 (least likely to experience political unrest) to 10 (most likely to experience unrest), using generally agreed-upon inputs. For MENA countries, the EIU understandably ranked Iraq among those countries most likely to experience unrest (a score of 7.9), with Lebanon at 7 and Algeria at 6.6.[xvi] For sure, these countries hardly showed the attributes generally associated with stability, but they did not undergo wholesale revolutions. Next in line were Yemen (6.1) and Syria (5.8), clearly at risk of instability but also ranked more stable than stable Latvia and Paraguay, among others. Egypt’s score of 5.4 and Tunisia’s 4.6 were the same as the scores for Israel and the UK, respectively, and both ranked higher than Saudi Arabia.

The point here is not to impugn the usefulness of these types of indicators and rankings; the FSI’s analysis, for example, is comprehensive and can provide a general framework for examining the ability of a country’s society and government to withstand shocks. The hazard for policymakers and analysts in relying on such indicators is that they can provide either a false sense of security or underestimate a country’s fragility. These scores are somewhat analogous to the AAA ratings that credit rating agencies gave to hundreds of mortgage-backed securities and collateralized debt obligations prior to the global financial crisis. As U.S. housing prices began to tumble in 2007, Moody’s downgraded 83 percent of the $869 billion in mortgage securities it had rated AAA just the year before.[xvii] In other words, investors looked at a AAA rating for a mortgage-backed security, compared that to the AAA rating of U.S. sovereign debt, and (wrongfully) assumed the risk of default to be the same. Similarly, an analyst comparing the EIU’s scores for Tunisia would see it ranked with the UK and conclude Tunisia a safe bet.

Other Indicators Might Show Predictive Promise

By now, it is evident that official statistics and data analysis by international institutions and NGOs can provide insight, but by themselves are incapable of accurately portraying a society’s well-being or predicting the timing of a crisis. Figures such as GDP growth, GDP per capita, measures of inequality, and development indicators provide a structure for viewing a country’s macro trends, which manifest over years and decades. What is needed, therefore, are short-term indicators that analysts and policymakers can view to get a better sense of pending unrest and instability. No single indicator will answer those questions, [xviii]  but two in particular can at least begin to approximate rising short-term vulnerability: food prices and opinion surveys. These indicators are, of course, imperfect, but they can provide a level of diagnosticity, given their timeliness (as in the case of food prices), or relatively up-to-date public perceptions that provide insight not to be gleaned from crude data.

Data from the UN’s Food and Agriculture Organization (FAO) shows this in stark terms. The FAO’s Food Price Index, which measures monthly changes in international prices of a basket of food commodities, rose 41 percent (in inflation-adjusted terms) from 2005 to 2010, and continued to rise through 2011. Nominal prices, of course, rose even more. The increase in food prices far outpaced the level of overall inflation in MENA countries and endangered the livelihoods of citizens. [xix] Although the prices of individual agricultural commodities moved up and down in the years prior, global food prices in the aggregate showed remarkable stability for more than two decades. Populations, therefore, had become accustomed to stable prices and then suddenly saw their food expenditures rise rapidly over a short period of time.[xx]

Such was the large-scale subsidization of wheat prices that by 2010, the Egyptian government’s bread subsidy bill topped $3 billion annually, much of it in the form of subsidized flour to local bakeries. 

More than 50 percent of the food consumed in the MENA region is imported, making it the largest food-importing region in the world, according to a World Bank study,[xxi] and Egypt is the world’s single largest wheat importer. This dependence makes many in the MENA region highly vulnerable to changes in global food prices. Moreover, severe restrictions on water and land-use in the region, exacerbated by climate change, suggests there is little likelihood these countries can expand domestic agricultural production to reduce this dependence. Wheat is the main source of daily calorie intake in Egypt and any uptick in prices or availability of wheat and bread directly impacts virtually every family.

The cost of Egypt’s food imports jumped from $3.7 billion in 2005 to $9.7 billion in 2010 – an increase of more than 160 percent. Even with heavily subsidized prices, the country’s dependence on imports during a time of sharply rising prices wreaked havoc on Egypt’s poor and vulnerable population. Such was the large-scale subsidization of wheat prices that by 2010, the Egyptian government’s bread subsidy bill topped $3 billion annually, much of it in the form of subsidized flour to local bakeries. The size of the subsidies, at a time of high world wheat prices, led to widespread corruption, where bakers could resell the subsidized flour and bread on the black market. That, in turn, pushed up the prices consumers paid by as much as five times.[xxii]

Widespread protests and violence in Iran at the end of 2017 and into early 2018 demonstrate the visceral impact of rising food prices. In a short period of time, the price of eggs had risen by some 40 percent and prices for poultry had gone up sharply as well.[xxiii] Despite heavy-handed regime tactics, lack of representation, and poor economic conditions for most Iranians, the country in recent years had been seen as the proverbial stable island in a sea of regional unrest. The protests partially threw into question the hypothesis that the “best indicator of a country’s future stability is not past stability but moderate volatility in the relatively recent past.”[xxiv] After all, Iran experienced ferocious country-wide unrest and brutally persecuted protesters as recently as 2009, after a disputed election. Iranians have many reasons to protest, but the quickness with which protests erupted over the sharp rise in food prices is further evidence that analysts and policymakers need to carefully watch these price movements.  

Public opinion surveys can also prove prescient. Those that show rising levels of dissatisfaction with the quality of life can be important indicators. Throughout much of the Middle East, the deterioration in life satisfaction was not captured in macroeconomic data, household financial surveys, or in standard indicators of inequality, but was evident in perceptions data from value surveys. There was a notable rise in the incidence of dissatisfaction in a number of areas considered crucial to quality of life. Absolute poverty was low and the level of income inequality was, in the aggregate, moderate during the 2000s. Instead, the unhappiness was associated with deteriorating government services, widespread corruption, and lack of fairness.

According to Gallup, 48 percent of Egyptians in 2010 said they were satisfied with public transportation systems – a dramatic decline from 2009, when 78 percent expressed satisfaction with the country’s public network of buses and trains. [xxv]  This precipitous fall in the public’s estimation of government services suggests dissatisfaction on the deeper level of the state’s basic duties. Public opinion surveys are in many ways similar to macroeconomic statistics; they develop over time, and many responses are consistent from one year to the next. However, the speed of change in public opinion can prove diagnostic in tracking looming instability, particularly if opinions had changed little over a period of years.

[[{"fid":"107571","view_mode":"default","fields":{"field_file_image_alt_text[und][0][value]":"Groups of Egyptian anti-government protesters on their way to Cairo's central Tahrir Square. Cairo, Feb 1, 2011","field_file_image_title_text[und][0][value]":"","field_file_source[und][0][value]":"Groups of Egyptian anti-government protesters on their way to Cairo's central Tahrir Square. Cairo, Feb 1, 2011; Shutterstock","field_file_caption[und][0][value]":""},"type":"media","attributes":{"alt":"Groups of Egyptian anti-government protesters on their way to Cairo's central Tahrir Square. Cairo, Feb 1, 2011","class":"media-element file-default"}}]]

Of course, not all surveys are created equally. Methodologies can vary greatly, and to be useful, they need to be timely, in order to accurately track sentiments on issues dear to the hearts of the public. But with the right focus on key socioeconomic concerns and if conducted well, surveys can peer behind official economic statistics to get a sense of public perception. As in many elections, it’s often less about the data a politician or ruler can point to and rather, how the public perceives its lot.

Understanding Trends Is Not Predicting Events

Dissatisfaction can endure for years, or even decades. Economic decline and poor social services persisted in the Soviet Union years before rumblings of unrest could be heard. Declassified documents show that analysts at CIA[xxvi] from the mid-1970s to Gorbachev’s assumption of party leadership in 1985 portrayed the Soviet Union as plagued by a deteriorating economy and intensifying social problems. Numerous papers and testimonies highlighted problems arising from growing consumer discontent, ethnic divisions, and a corrupt and incompetent political system. As the decade of the 1980s progressed and Gorbachev’s reforms began to confront economic reality, multiple intelligence products accurately addressed the situation. One assessment from July 1988 judged that regardless of Gorbachev’s policy direction, Soviet citizens would need to see improvement in living standards for the country to avoid widespread discontent. As it was, the Eastern bloc fragmented in 1989, and by 1991, the Soviet Union as a political entity had dissolved.

Muhammad Bouazizi, the poor Tunisian fruit vendor who set himself on fire to protest his treatment at the hands of a broken bureaucracy, unknowingly launched regional unrest and revolutions. But he was not the first Tunisian to do so.

According to one study, the probability of a state’s falling into instability is a function of “trends” (which measure broad patterns in authority, resilience, and legitimacy over time) and “triggers” (events likely to precipitate a state’s instability). Analysts are adept at identifying the trends, such as increases in youth unemployment, failing public sector services and infrastructure, rising household debt and income inequality, but the spark – such as an ill-timed (or misguided) policy decision or an individual’s protest, as in Tunisia – remains elusive. Data and expertise can take predictions far, but rarely all the way.  

Yet, not every seemingly salient event is stability’s proverbial coup de grace. Not wanting to be the next analyst or pundit to miss a revolution, every minor event comes to be viewed as a possible trigger, but innumerable would-be catalytic events come and go without a revolution. In March 2002, more than a dozen girls died in Saudi Arabia trying to escape a burning school, only to be cruelly prevented by the kingdom’s Committee for the Prevention of Vice and the Promotion of Virtue, the mutaween, because they were not properly covered. Rare but vocal public criticism of the mutaween ensued, leading some in the West to judge that this barbarity would undoubtedly lead to a revolution in social norms. Then, the episode faded away.

Muhammad Bouazizi, the poor Tunisian fruit vendor who set himself on fire to protest his treatment at the hands of a broken bureaucracy, unknowingly launched regional unrest and revolutions. But he was not the first Tunisian to do so. Only nine months before the protest-suicide, a Tunisian street vendor in the coastal town of Monastir, Abdesslem Trimech, did the same. No doubt tragic, his suicide prompted little public reaction and was ineffective at nudging Ben Ali and his regime.[xxvii] Copycat self-immolations in Algeria followed in early 2011 and led to minor protests, raising eyebrows within the regime, but failing to prompt widespread unrest.[xxviii]

And then, some of history’s landmark events were spawned by the haphazard decisions of individuals. On November 9, 1989, Lt Colonel Harald Jager, an East German border guard officer in charge of a crossing point to West Berlin, made the snap judgement to open his section of the Berlin Wall, painting one of the most iconic images of the 20th century.[xxix] But his decision was not in any way predictable. Thousands of East Berliners arrived at the gate after hearing a senior Politburo member announce – mistakenly –that East Germans would be allowed to cross into West Germany. The rest, of course, is history. That East Germans and other Warsaw Pact populations had suffered under dismal economic conditions and oppressive regimes for decades was hardly news to intelligence analysts and the general public, alike. However, no one – not even Lt Col Jager himself – could have anticipated his decision to open the gate and effectively mark the end of the Cold War. All the data indicated that the end was inevitable, but the timing and immediate circumstances of the end could not be foreseen.

The beauty of data is that it provides definitive signposts. Rather than relying on anecdotes (who hasn’t heard the story of the taxi driver in Cairo?), socioeconomic data can highlight broad trends in stark relief and frame an analyst’s approach to a problem. Accurately forecasting events based on aggregating reams of data will consistently fall short, however. It is perhaps too soon to tell whether big data exploitation can provide the final ingredient for forecasters, but the complexity of societies and markets suggests that this, too, will have its limitations. A fortune teller can always predict your mortality, just not the date and time.

The author is an employee of the United States Government (USG), which is funding his fellowship at the Wilson Center. All statements of fact, opinion, or analysis are those of the author and do not reflect the official position or views of the USG.

[i] Richards, Alan. “Economic Reform in the Middle East: The Challenge to Governance,” The Future Security Environment in the Middle East: Conflict, Stability, and Political Change, edited by Nora Bensahel and Daniel L. Byman, 1st ed., RAND Corporation.

[ii] Richards, Alan. “Economic Reform in the Middle East: The Challenge to Governance,” The Future Security Environment in the Middle East: Conflict, Stability, and Political Change, edited by Nora Bensahel and Daniel L. Byman, 1st ed., RAND Corporation.

[iii] http://www.worldbank.org/en/news/feature/2015/10/21/middle-class-frustration-that-fueled-the-arab-spring

[iv] Iqbal, Farrukh and Youssouf Kiendrebeogo, “Education Attainment in the Middle East and North Africa: Success at a Cost,” World Bank Policy Research Working Paper 7127, December 2014.

[v] Hassine, Nadia Belhaj, “Economic Inequality in the Arab Region,” Policy Research Working Paper 6911, World Bank, June 2014.

[vi] Coyle, Diane, GDP: A Brief But Affectionate History, 2014.

[vii] Salih, Kamal Eldin Osman, “The Root Causes of the 2011 Arab Uprisings,” Arab Studies Quarterly, Vol. 35, No. 2 (Spring 2013).

[viii] Kawar, Mary, “Impact of the Economic and Financial Crisis on the Skills and Employability of Young People in the Arab Region,” International Labour Organization, 2010.

[ix]Global Employment Trends for Youth 2015, International Labour Organization, 2015.

[x] Rivlin, Paul, Arab Economies in the Twenty-First Century, 2009.

[xi] https://www.reuters.com/article/us-egypt-corruption/hopes-for-a-new-egypt-marred-by-pervasive-corruption-idUSBRE83Q0W220120427

[xii] Lesche, Ann M, “Egypt’s Spring: Causes of the Revolution,” Middle East Policy Council, http://www.mepc.org/egypts-spring-causes-revolution

[xiii] Transparency International, 2008.

[xiv] Hassine, Nadia Belhaj, “Economic Inequality in the Arab Region,” Policy Research Working Paper 6911, World Bank, June 2014.

[xv] Hassine, Nadia Belhaj, “Inequality of Opportunity in Egypt,” World Bank Economic Review, 2011.

[xvi] Economist Intelligence Unit, “Manning the Barricades: Who’s at risk as deepening economic distress foments social unrest,” 2009.

[xvii] https://www.cfr.org/backgrounder/credit-rating-controversy

[xviii] ElGindi, Tamer, “Well-Being Before the Arab Spring: Objective Vs. Subjective Measurements,” Middle East Policy, Vol. XXIV, No. 2, Summer 2017.

[xix] http://www.fao.org/worldfoodsituation/foodpricesindex/en/

[xx]http://www.slate.com/articles/health_and_science/feed_the_world/2014/04/food_riots_and_revolution_grain_prices_predict_political_instability.html

[xxi] World Bank, “Agriculture and Rural Development in MENA,” September 2008.

[xxii] https://www.newstatesman.com/business/2013/07/what-price-bread-egypt

[xxiii] https://www.nytimes.com/2017/12/28/world/middleeast/iranians-protest-rising-food-prices.html

[xxiv] Taleb, Nassim Nicholas and Gregory F. Treverton, “The Calm Before the Storm: Why Volatility Signals Stability and Vice Versa,” Foreign Affairs, Vol. 94, Issue 1 (January/February 2015).

[xxv] http://news.gallup.com/poll/157043/egypt-arithmetic-revolution.aspx

[xxvi] https://www.cia.gov/library/center-for-the-study-of-intelligence/csi-publications/csi-studies/studies/97unclass/soviet.html

[xxvii] http://www.aljazeera.com/indepth/features/2011/01/2011126121815985483.html

[xxviii] http://www.aljazeera.com/news/africa/2011/01/20111162363063915.html

[xxix] https://www.npr.org/sections/parallels/2014/11/06/361785478/the-man-who-disobeyed-his-boss-and-opened-the-berlin-wall