Attend Spring Meetings on Development topics from Apr 17-21. Comment and engage with experts. Calendar of Events

Syndicate content

Facts vs. Perceptions: understanding inequality in Egypt

Paolo Verme's picture
Also available in: العربية
Press coverage of the Egyptian revolution, both local and international, made frequent use of the word “inequality” to describe one of the factors that generated discontent. And during the most recent phase of the political transition, two of the themes driving popular debates are the questions of social justice and equality. The general perception inside and outside Egypt before the revolution was that social injustice and a somehow unequal distribution of resources were deep rooted phenomena, simply part of the social landscape. That has changed with the revolution, and the expectation now is that the new leaders have to deliver on the popular demands for equality and social justice.

One of the puzzling aspects of this preoccupation with inequality is that the measurement of monetary inequality in Egypt by means of household surveys does not seem to match perceptions. According to both official government figures and the World Bank, the Gini coefficient (a measure of the average distance between incomes) has been declining throughout the last decade from 36.1% in 2000 to 30.7% in 2009 (World Bank 2007 and 2011). This last figure is low by regional and international standards and is also atypical for low income countries that experience rapid growth periods. Egypt has enjoyed a very prolonged growth phase since the late 1980s. Between 1996 and 2010 the country saw growth rates above 5 percent, including peaks of over 7% between 2006 and 2008 (

How can the apparent mismatch between the facts and perceptions of inequality be explained? A recent study[1] addresses this question by looking thoroughly into the facts of inequality using three rounds of the Egyptian Household Income, Expenditure and Consumption Surveys (HIECSs) and by reviewing people’s perceptions of inequality using two rounds of the World Values Surveys (WVSs) during the decade that preceded the revolution. The results of the study point to a possible explanation for the discrepancy between facts and perceptions.  

What do the data say?

In addition to confirming the HIECS as a valid measure of inequality among observed Egyptian households, the study also revealed that income inequality did not increase between 2000 and 2009. This is consistent with previous studies and with other findings related to household welfare. Between 2000 and 2009 poorer households performed relatively better than richer households, which is consistent with a decline in inequality. These results clearly do not support the accepted narrative about the origins of the Egyptian revolution. It is possible that the HIECS does not measure top incomes well, which would explain the low rate of observed inequality. However, the study did not find any evidence that top incomes were not fully captured, and furthermore, the distribution of incomes in Egypt is not atypical when compared to the distribution of incomes in other similar countries. In other words, the problem is not one of measurement, as the HIECS does an effective job not only in measuring top incomes but  is also, overall, a good survey to measure inequality.

On the other hand, the HIECS do reveal two important clues that could potentially help explain the mismatch between facts and perceptions. The first clue is that, on the whole, household welfare did not improve during the growth period. This would imply that growth did not trickle down to the general population. Therefore, while households may have been reading about growth in newspapers, they were not experiencing it first-hand. Reading and hearing about growth evidently increases expectations for a better future while experiencing none of the benefits of growth crushes these same expectations.

The second clue is that the mismatch between GDP growth and household welfare has increased over time. This can be clearly seen in Figure 1 where we plotted GDP growth against growth in household income and expenditure as measured by the HIECS. The wedge between GDP growth on the one hand and household income growth on the other increased sharply during the period that preceded the revolution. It is also evident that this wedge increases not only between GDP and household income measured with the HIECS but also between GDP and household consumption measured with national accounts. Rather than growth in inequality, the data does indicate a significant gap between the anticipated rise in income (represented by GDP growth) and their actual rise (represented by take-home income) – a gap that increased over the decade that preceded the revolution.

                                                    (Click to Enlarge)

What do Egyptians think?

Popular perceptions of inequality were measured in Egypt by the World Values Surveys in 2000 and 2008, covering the same period as the HIECS. The World Values Survey is a global survey of views and opinions across a wide range of issues. These surveys contain a question about people’s attitudes toward inequality, measured on a scale from one to ten, where one indicates a desire for more equality and ten indicates a tolerance for higher levels of inequality. Figure 2 depicts the average results divided up by income groups for the same question in 2000 and 2008. The first group represents the poorest 10% of the population while the tenth group represents the richest 10%. The figure clearly shows two trends. The first trend is that all people became more averse to inequality between 2000 and 2008. The red line for 2008 is almost uniformly lower than the blue line for 2000 meaning that the tolerance for-inequality declined among almost all income groups. The second trend is that the difference in perceptions between the rich and the poor has increased. In 2000, this difference was modest with the poor and the rich being divided by only one point on the scale. In 2008, the difference between rich and poor doubled to two points.  Therefore, Egyptians have clearly become more averse to inequality and this increase in aversion has been greater for the poorest in society, two factors that would point to increased discontent with the distribution of incomes. 

                                                   (Click to Enlarge)

Looking more broadly at changes in people’s sentiments as observed by the WVSs, the period from 2000 to 2008 also saw a remarkable change in people’s perceptions on a vast range of issues. People’s priorities changed from general concerns about freedom and the environment to very concrete aspirations about GDP growth and stable food prices. There is also a clear decline in the levels of self-reported incomes and social status which is consistent with the HIECS data. In 2008, households felt poorer than in 2000 and they believed that they belonged to a lower social stratum. Concerns about social class have risen even when compared to concerns about income, suggesting that people became more aware of their social status and more concerned about relative status as opposed to absolute status. In fact, the concern for social status seems to explain the aversion to inequality better than income itself. When compared to other countries we also find that the increase in aversion to inequality in Egypt has been greater than elsewhere in the world and only comparable to other countries in the Middle East and North Africa  region. Theories of revolutions and social justice tell us that during extended periods of deep economic changes - such as the one experienced by Egypt during the past two decades – people’s   expectations change very quickly. This is effectively what we find when we compare the 2000 and 2008 WVS.

Are what the data say and what people say really different?

The truth is that by comparing what the data say with what people say, we find few inconsistencies. The HIECS and macroeconomic data tell us of a mismatch between GDP growth and household income growth, suggesting that people may have grown more frustrated with their situation. The WVS in turn confirms that people have indeed grown more frustrated and more averse to inequality. There is a mismatch between income inequality as measured with the HIECS and perceptions of income inequality measured with the WVS. But this is explained by the fact that people judge inequality not only by the perceived gap between their own income and the income of others but also, equally critical, by the perceived gap between their actual income and their expected income. The real question for Egypt today is not the mismatch between facts and perceptions of inequality but the mismatch between facts and expectations of income. A measure of inequality such as the Gini coefficient is not designed to measure this last discrepancy and cannot convey a complete picture of how people perceive inequality. This is perhaps an aspect of social inequality that we failed to see before the revolution and something to consider more closely in the future.

[1] World Bank (2012) Inside Inequality in Egypt: Historical trends, recent facts, perceptions and the spatial dimension. Mimeo.


Submitted by Professor Feisal Yunis on
I think the nature of the samples differ widely between the two WV surveys. The second survey is biased towards women, mainly housewives. Will this make the comparisons between the two rather difficult? As I remember, in some questions of the first survey, there were some errors in data recording, which I pointed to the central data office in Spain and they corrected it. Did you check this possibility concerning the questions at hand in the 2000 survey. best

Submitted by Paolo on
Thank you, this is a good point.

The two WVSs are probably not comparable across all dimensions and they show different distributions for some of the variables (you mention gender for example). We found that responses to the question of inequality which was our main interest were quite consistent in showing an increase in inequality aversion between 2000 and 2008 and this trend was also in line with the trend observed in other comparable countries. We also discussed the issue of data comparability with the IDSC.

However, we all agree that comparisons across the two surveys should not be given for granted but considered on a case by case basis.


Paolo Verme

Add new comment