On the ‘Arab Inequality Puzzle’: The Case of Egypt

This article surveys and discusses prominent protagonists of the debate on socio-economic inequality in the Arab region, with a special focus on the World Bank and Egypt. According to ofﬁcial data, the region holds re- markably low Gini coefﬁcients in a context of declining inequality. This contradicts the popular perception of high social inequality as a major cause of regional protests since the Arab Spring; hence the reference to a ‘puzzle’ in mainstream literature. The debate about the reality of social inequality in the region has developed since 2011 — particularly in regard to Egypt, where income and consumption data are periodically collected by means of house- hold surveys. Inequality measures based on this method alone, while income taxation data are inaccessible, are highly questionable and conﬂict with vari- ous observations and calculations based on other indicators such as national accounts, executive income or house prices. Yet, the World Bank upholds ofﬁcial inequality ﬁndings in portraying the Arab upheaval as the revolt of a ‘middle class’ that aspires to greater business freedom, in consonance with the neoliberal worldview.


INTRODUCTION
Interest in income and wealth inequality as a topic of research and discussion was certainly boosted by the international success of the English translation of Thomas Piketty's now very famous book, Capital in the Twenty-first Century (Piketty, 2014). In reality, however, this topic had already been the object of countless studies, related to the sharp increase in socio-economic inequality levels that went along with the global implementation of the neoliberal paradigm from the 1980s onwards. Attention peaked at the turn of the century, after two decades of economic deregulation and curtailment of welfare had started to take their toll in full. The emergence of the Global Justice Movement since the 1999 anti-WTO protests in Seattle was a consequence of this increasing interest, as was the birth of the global Occupy movement 12 years later, in the aftermath of the 2008 Great Recession and under the impact of the 2011 Arab Spring and its most mediatized episode -the occupation of Tahrir Square in Cairo between 25 January and the fall of Egyptian president Hosni Mubarak on 11 February.
Paradoxically, until 2011 the Arabic-speaking region itself stood as an exception to the global research trend in exploring socio-economic inequality, with very few studies dedicated wholly or in part to this issue and its consequences. As could be expected, however, such studies multiplied in the wake of the regional revolutionary shockwave. The sudden surge in attention to regional social inequality that the Arab Spring generated made the assessment and interpretation of available data about Arab countries even more contentious than elsewhere, revealing a range of problems that the region shares to variable degrees with other parts of the world. At the core of this debate stands what is often called the 'Arab inequality puzzle'. According to official data, the region has remarkably low Gini coefficients in what is portrayed by international financial institutions (IFIs) as a context of declining inequality. This, however, contradicts the popular perception, also expressed in some official reports, of high social inequality in the region, constituting a major cause of regional protests.
A central issue in these contradictory perspectives is the credibility of available figures on socio-economic inequality: in the Arab countries, these data are either limited to consumption inequality or based, at best, on household surveys. This points to a major flaw in international comparisons of inequality published by the IFIs: they end up comparing apples and oranges by including in one and the same list figures based on different calculation modes, ranging from household consumption surveys to income taxation data. 1 The latter are available only for a minority of countries in the world, mostly belonging to the global North. In that sense, the debate on the Arab region is symptomatic of general problems encountered in international comparisons of inequality.
This article critically surveys key stages and protagonists of the debate on the assessment of socio-economic inequality in the Arabic-speaking region, involving international institutions such as the United Nations Development Programme (UNDP), the World Bank and UN Economic and Social Commission for Western Asia (ESCWA), and various researchers. It focuses on a few reports selected for their prominence in the debate and the originality of their input. In doing so, it points to the discrepancy between the findings of various researchers, including World Bank researchers, and the World Bank's official stance. The latter stems from an interpretation of the Arab Spring and leads to policy recommendations which are fundamentally consistent with the neoliberal paradigm adopted by the IFIs since the 1980s. 2

THE TERMS OF THE DEBATE IN HISTORICAL PERSPECTIVE
Global debates on inequality revolve in large part around the question of whether inequality is a necessary condition of the initial stage of development as illustrated by the Kuznets curve, or an impediment to it. This disagreement is arguably as much political as it is scientific, since it hinges on ethicalpolitical values influencing the divergent ways in which the same facts are assessed. The debate in the Arab region goes beyond this: it is also, and to a large extent, about the validity and credibility of available inequality data against the backdrop of widespread suspicion that the region's authoritarian regimes distort and manipulate social and economic figures for political purposes.
In both discussions, regarding the validity of data on the one hand and the relation between inequality and development on the other, the focus is predominantly on socio-economic inequality, that is, inequality in consumption, income and wealth. It is also this socio-economic inequality that is the main focus of the argument that has been unfolding with regard to the Arab region, rather than any of the other types of inequality that are reviewed in Challenging Inequalities: Pathways to a Just World, the excellent 2016 edition of the World Social Science Report (ISSC and IDS, 2016).
There is indeed not much factual dispute among researchers on most other dimensions of inequality in the Arab region, be they gender inequality, 'horizontal inequalities' between ethnic and racial groups, inequalities in education, or health, or natural resources, or in access to digital information and communication, or various intersections of all the above. 3 On the most conspicuous type of regional inequality, that is, gender inequalities, the discussion is quite limited compared with the disagreement on vertical social 2. On the 'neoliberal bias' in studying the Arab region, see Bush (2004). 3. For a recent survey of various types of inequality affecting youth in Arab countries, see Arab Human Development Report 2016 (UNDP, 2016). On the Human Development Index (HDI), the report notes that the Arab region 'suffers an average loss of 24.9 percent when the HDI is adjusted for inequalities, which is above the world average loss of 22.9 percent' (ibid.: 23). See also the research paper written by Abdellatif et al. (2019) during the preparation of the Human Development Report 2019 (UNDP, 2019), which is focused on the various dimensions of inequality.
inequality. That is because the facts on gender inequality in the region are hardly disputable and there is an overwhelming consensus among scholars and experts on the desirability of gender equality, which dissuades traditionalist dissenters, rather rare in such research milieus, from raising their voices.
The issue of social inequality is predictably much more divisive. It came to prominence in the Arab region, as in other parts of the 'Third World', when the decolonization era following World War II witnessed a major left turn of nationalist movements during the 1950s under the impact of Communist successes in China and Indochina. As with various other post-colonial national movements, the left-wing radicalization of Arab nationalism peaked in the 1960s when it embarked on building 'Arab socialism' (sub-Saharan African nationalist movements did likewise with 'African socialism'). Those were experiences combining a dominant 'state capitalism' (Cooper, 1983), resulting from sweeping nationalizations, with pro-poor and pro-labour social and welfare reforms, usually under single-party dictatorships. Several countries in the Arab region -Egypt, Syria, Iraq and Algeria (and to a lesser degree, Tunisia) -followed this trend and were later joined by Sudan and Libya, 4 as well as by South Yemen where Arab nationalism mutated into Marxist-Leninist socialism under the Soviet Union's tutelage for two decades.
The Egyptian paradigm of this specific historical legacy still resonates with large sections of the Arab region's population who remember it as a time of national dignity and an easier life, memories manifested above all in the widespread nostalgia for the champion of 'Arab socialism', Egypt's president Gamal Abdel-Nasser (in office from 1954 until his untimely death in 1970). It feeds powerfully the regional longing for social equality -much more so than Islam, the favoured 'theory of everything' of the 'Orientalist' perspective that Edward Said famously criticized and which continues to inform the IFIs' literature.
As Egyptian economist Mahmoud Abdel-Fadil acknowledged in the conclusion of his critical study of income distribution in Nasser's Egypt (1952-70, prolonged until 1972), a key aspect of Nasser's legacy in Egypt lies indeed in its 'instilling the desire for independent economic development and the ideas of social justice into the consciousness of the masses' (Abdel-Fadil, 1980: 67-68). Abdel-Fadil showed in his study how 'the upper tail of the distribution of personal income has been cut off rather effectively' (ibid.: 67). Nationalizations and land sequestrations under Nasser, along with the capping of top incomes in the public sector, 'have been more effective in 4. Moscow conveniently identified these countries as treading a 'non-capitalist path of development' (Hosseinzadeh, 1989;Solodovnikov and Bogoslovsky, 1975), thus avoiding bestowing on them the 'socialist' label that it reserved for countries run by Communist parties, while justifying its close alliance with their governments. equalising incomes than any system of progressive direct taxation could have achieved under a system of pure "laissez-faire"' (ibid.: 68). 'Arab socialism' began unravelling from the early 1970s onwards under the impact of the decisive blow that Israel inflicted in the June 1967 war on the two most radical Arab nationalist regimes of Egypt and Syria. The decades following Nasser's death and the overthrow of the left-wing faction of the ruling Baath Party in Syria in 1970 saw the region's leadership shift from Egypt and the towering figure of Nasser to his long-time foes, the Saudi kingdom's rulers. The shift was powerfully boosted by the first oil shock following the 1973 Arab-Israeli war. The subsequent surge in oil prices and revenues tremendously increased the Saudi rulers' influence. Egypt's swing under President Anwar el-Sadat (1970-81) from a privileged alliance with the Soviet Union to a no less privileged one with the United States epitomized the regional paradigmatic shift from authoritarian state capitalism and 'Arab socialism' to illiberal private market capitalism devoid of 'socialist' pretensions.
This shift inaugurated a transition period into which several other Arab countries entered at various moments of the 1970s and 1980s, a transition that accelerated in the 1990s and especially after the turn of the century. Under the banner of economic 'opening' (infitah in Arabic) first raised by Sadat in Egypt, gradual measures of economic liberalization were adopted, including privatization of public enterprises and public services, curtailment of the welfare state, liberalization of trade and creation of free zones. These measures were implemented in the framework of structural adjustment programmes sponsored by the International Monetary Fund (IMF) and imposed as conditionalities for the obtainment of loans and good marks from the IMF, both crucially needed to attract foreign aid and investment. 5 These measures were faced with a wave of popular resistance as manifested in the 'bread riots' that swept the region (most prominently: Egypt 1977, Morocco 1981, Tunisia 1983and Jordan 1989, which slowed down the implementation of austerity measures. Consequently, the new policies' impact on the living conditions of the Arab region's populations was less brutal than that of the 'shock therapy' exacted wholesale from ex-Communist states of Eastern Europe in the 1990s. Indeed, there was concurrently less resentment towards the 'socialist' past and fewer hopes of Western-like prosperity in the Arab world than east of the former Iron Curtain. As elsewhere, nonetheless, changes inspired by the global neoliberal paradigm led to a significant increase in social inequality in the Arab region, manifest above all in the rapid and huge enrichment of the upper crust, reversing the trend that had prevailed during three decades following World War II. There are, of course, various and often conflicting sets of inequality 5. For an overview and assessment of IFIs-sponsored economic policies written before the Arab Spring and therefore unaffected by a retrospective view, see Richards and Waterbury (2008: 228-63).
data regarding Arab countries, or the Middle East and North Africa (MENA, which includes Iran in addition to the Arab states). The University of Texas Inequality Project (UTIP) has elaborated a set of global data -the Estimated Household Income Inequality Data Set (EHII) -derived from the economic relation of its own computation of industrial pay inequality measures (Theil index), on the basis of United Nations Industrial Development Organization statistics, with other factors and data, producing more reliable Gini coefficient measures (UTIP, 2017). A 2007 study of the correlation between the EHII findings and two indicators of globalization (foreign direct investment and total trade as percentages of GDP) has found 'empirical evidence that overall income inequality in the MENA region has gone up with the process of globalization in these countries since income inequality increases more through the period of globalization  than through the period pre-globalization ' (Benar, 2007: 201). For Egypt, the country with the longest and most advanced governmental tradition of data gathering in the region, the EHII indicates a clear increase in social inequality between the 1960s and the 2000s, illustrated by the maps on its front page (UTIP, n.d.). Egypt's EHII Gini coefficient increased from 42.7 in 1970, the last year of Nasserite 'socialism', to 53.7 in 2015; during the same period, Egypt's Theil index of industrial wage inequality increased almost 10-fold from 0.015 to 0.14 (UTIP, 2017).
One could surmise that socio-economic inequality would have become a research topic of high interest in the Arab region as in most of the rest of the world -if not more so, given that it remained a prized value in the region. And yet, it didn't -not until the theme of 'social justice' emerged powerfully with the Arab Spring.

PRIOR TO THE SPRING
Social science researchers have regularly complained about the lack of appropriate information for an assessment of social inequality in the Arab region, especially measurements of income and wealth. The UNDP's Arab Human Development Report acknowledged the problem in its first editions: 6 'Analysis of poverty and income-inequality issues in the Arab region is frustrated by lack of comprehensive and comparable data sets as well as by the reluctance of some official sources to share primary survey data with researchers' (UNDP and AFESD, 2002: 90).
6. Five editions were published before the Arab Spring, in 2002Spring, in , 2003Spring, in , 2004Spring, in , 2005Spring, in and 2009 The four initial instalments of the report were co-sponsored by the Arab Fund for Economic and Social Development (AFESD), which is a Kuwait-based development fund involving all members of the League of Arab States. After the AFESD ended its co-sponsoring, the report went into a three-year eclipse.
Indeed, social inequality data in the Arab region, when available, are mostly based on consumption/expenditure rather than income, let alone wealth. Official data, used by international institutions, have been indicating levels of inequality in the region that rank among the lowest in the developing world, if not the whole world. Observers could not fail to note, however, the striking discrepancy between the picture offered by these data and the actual experience of inequality in the awareness of the region's inhabitants. As the second edition of the UNDP report explained: Even when field surveys of income and expenditure (which constitute the basic source for estimates of income distribution) exist, such surveys suffer from defects that diminish their credibility, particularly with regard to the parameters of income distribution, as a result of bias in the collected data. In Egypt, for instance, relying on the results of income and expenditure surveys in the first part of the 1990s leads to an improvement of the Gini coefficienti.e., income distribution becomes more equal. But this does not correspond to the overall economic situation, particularly unemployment and poverty criteria and the observations made of wealth distribution during the same period. . . . Labour's share of the value added declined [in Egypt] from nearly 40 per cent in 1975 to nearly 25 per cent in 1994, which indicates a deterioration of GNP distribution in favour of wealth returns. (UNDP and AFESD, 2003: 139-40) The 2009 UNDP report moderated somewhat the critical tone of the four earlier instalments. It seemed to accept the view that income inequality is lower in the Arab region than in other parts of the developing world, tentatively attributing this to 'the cumulative achievement of social contracts in the Arab countries since independence' (UNDP, 2009: 116). However, its assessment remained quite critical, particularly regarding the issue of wealth inequality about which no official data are available: [T]here is evidence to suggest that the inequality in wealth has worsened significantly. In many Arab countries, for example, land and asset concentration is conspicuous and provokes a sense of exclusion among other groups, even if absolute poverty does not increase. Furthermore, the crowding of the poor in slums without sanitation, safe water, recreational facilities, reliable electricity and other services aggravates such exclusion. These trends, combined with high unemployment rates, result in the ominous dynamics of marginalization, visible in the high rates of urban slum dwellers in Arab cities and towns: 42 per cent in 2001. (ibid.: 116) The most thorough examination of the research record on regional social inequality that came out before 2011 is a study authored by Sami Bibi and Mustafa Nabli, published in 2010 by the Cairo-based Economic Research Forum (ERF), a regional non-profit network. It starts by acknowledging the record's paucity: It is surprising to see how little attention the issue of inequality has received in research given its critical role in the economic development of Arab countries. . . . It is perhaps because of the limited access to micro-data from household surveys and the political sensitivity surrounding the issues of inequality. And that explains why governments have not been keen to see this activity develop and may have even prevented collection of authentic data and research. (Bibi and Nabli, 2010: 4) The authors then draw a rather conventional picture of the region's experience in development, while acknowledging the positive aspects of the 'socialist' past in contrast with most contemporary assessments. 7 Thus, characterizing the pre-1980s model of development as 'based on strong governments, central planning, and wide-scale policies for redistribution and equity', they recognize that it bore what they call 'strong payoffs': [L]arge oil and oil-related revenues, in addition to directing large resources toward public infrastructure . . . and substantial investment in education . . . led to an increase in the average years of schooling of the adult population from less than a year in 1960 to more than 3 years in 1980. Impressive gains in basic education coupled with heavy investments in healthcare resulted in substantial reduction in poverty and improvement in health indicators. . . . Although there were very limited data on poverty in Arab countries before the 1980s, few estimates that are available for Egypt and Tunisia show dramatic declines. (ibid.: 10) On the negative side, the authors stress the strong inhibitions that the model created for the growth of a dynamic private sector and the fact that the bulk of the private sector that came out of this experience was 'living off the state' (ibid.: 10-12). They summarize the reasons why the paradigm shift that unfolded starting from the 1980s led to massive unemployment in the context of the protracted depression of oil prices after the Second Oil Shock of 1979: 'Growth collapsed under the burden of declining public spending and unattractive private investment climate, and continuing losses attributed to inefficiency. . . . This, for the first time, led to the emergence of a serious unemployment problem in the Arab states' (ibid.: 12).
The IMF-sponsored structural reform failed to resolve this problem: in the Arab region, it was 'cautious, selective, and often subject to pause and reversals', avoided 'most governance reforms including opening up of political space', and did not manage overall to provide 'a substantially improved climate for growth and investment' (ibid.: 14). 8 The authors highlight the sorry state of the regional labour market, describing how youth entering the labour force 'with substantially higher levels of education than the generation before them' were prevented from realizing their economic potential (ibid.: 17).

The World Bank acknowledged in 1995 that '[d]uring 1960-85 the MENA region outper-
formed all other regions of the world except East Asia, not only in income growth per capita, but also in the equality of income distribution. . . . Improvements in social indicators were dramatic' (World Bank, 1995: 33). However, the credit for this achievement, which was rightly linked to 'a massive build-up of public investment', was characteristically attributed to the international context: The era of statism coincided with a far more accommodating international context for MENA countries. Oil prices were high, the world economy was buoyant, industrialization was still in the 'easy' stages, and the world was a less competitive place. Times have changed, and many of the policies and institutions that seemed to serve MENA countries well in the past have become stumbling blocks to the future. (ibid.: 34) Note that oil prices were 'high' only during the last 10 years of the 1960-85 period. 8. For an explanation of this developmental failure that relates it to the socio-political nature of an Arab state system dominated by rentierism and (neo)patrimonialism, see Achcar (2013).
The gigantic revolutionary shockwave of the 2011 Arab Spring triggered a vast amount of research about its causes, leading to countless books and articles in various languages. 9 In contrast with the pre-2011 paucity of research on social inequality in the Arab region, this topic became the focus of an intense debate, most of which centred on Egypt.

THE ARAB SPRING: CONFLICTING VIEWS ON EGYPT
Most scholarly investigations of the Arab regional upheaval recognized socio-economic factors as fundamental to the explosion of popular anger, underlying the political issues that come inevitably to the fore in any mass uprising. The sharp rise in social struggles during the years preceding the Arab Spring in Tunisia and Egypt, the two countries that spearheaded the movement, as well as the centrality of social issues in the demands put forward by demonstrators all over the region, pointed to the socio-economic underpinning of the political turmoil. The prominence of the issue of 'social justice' as a key demand in the protests suggested that the issue of social inequality provides an important clue to the regional explosion.
Egypt, with the region's largest population and a central contribution to the Arab Spring, received the largest attention in terms of research on inequality after 2011. This was facilitated by the country's long tradition of official periodical data gathering performed by its statistical and census agency, the Central Agency for Public Mobilization and Statistics (CAPMAS), created in 1964 in the heyday of Soviet-inspired 'central planning'. 10 What follows is a critical examination of four key contributions to the debate on the relation between social inequality and Egypt's involvement in the Arab Spring that was prompted by World Bank researchers.

Hlasny and Verme
The post-2011 debate on inequality in Egypt was triggered in 2013 by a World Bank working paper on inequality in Egypt authored by Vladimir Hlasny and Paolo Verme (2013). The authors start by dwelling on the main problem that casts doubts on the accuracy of the CAPMAS data: the exclusive reliance of their periodical Income, Expenditure and Consumption Survey (IECS) on visits to a sampling of households. 11 They acknowledge euphemistically what is 'rather well known', namely 'that household 9. There are already several bibliographies on the Arab uprisings, such as the one prepared by the Project on Middle East Political Science (POMEPS, 2015). 10. The first large-scale survey in Egypt was conducted in 1958/59 before the creation of CAPMAS. 11. CAPMAS used to conduct an IECS every five years until 2008/09. Since that fiscal year, the IECS has been conducted every other year.
surveys are not particularly accurate at measuring top incomes because richer households tend to either underreport income or expenditure, or are less likely to participate in household surveys altogether' (the same goes, of course, for the poorest of the poor who do not have stable housing). Under such conditions, the authors concede, 'measures based on incomes such as the Gini index for income inequality are biased and do not reflect the actual extent of inequality in a country' (Hlasny and Verme, 2013: 3). And yet, Hlasny and Verme devote the bulk of their paper to the purpose of establishing the credibility of the low inequality figures reported by CAPMAS for Egypt, by way of statistical and econometric demonstration. Their core arguments are, on the one hand, the conformity of official figures to the Pareto distribution after a statistical correction based on an estimate of unit non-response by top income households and, on the other hand, a comparison of Egyptian data with household surveys in several countries, showing that 'the Gini coefficient in Egypt is significantly below median values for other countries, while the distribution of top incomes is around the median' (ibid: 29). In sum, as is often the case with econometrics, the authors take a statistical plausibility for an economic one, thus asserting the credibility of Egypt's particularly low Gini coefficient without confronting it with what is empirically known of the gap between the country's social strata. 12 As a result, Hlasny and Verme are able to dismiss the talk about high social inequality in Egypt as merely a matter of 'popular perceptions'. They thus rescue from ridicule the good marks attributed to Egypt by the World Bank and the IMF during the years that preceded 2011. Notwithstanding the huge social explosion of that year, it would be possible indeed to argue -if the official statistics were accurate -that Egypt had truly deserved to be hailed as a success story by the IFIs in the preceding years, a feat that they naturally attributed to the implementation of their neoliberal recommendations. This is still reasserted by the IFIs in disregard of alternative assessments of the same period that drew wholly different conclusions, predicting a popular explosion. 13

The 2014 World Bank Study
Hlasny's and Verme's (2013) paper provided the statistical ammunition for a long study on inequality in Egypt that circulated in draft the same year and came out the year after, in 2014. Verme is the lead author of this latter study, along with Branko Milanovic, the well-known head economist of the unit studying poverty and income inequality at the World Bank Research Department, and four Egyptian contributors. 14 Entitled Inside Inequality in the Arab Republic of Egypt, the study by Verme et al. (2014) opens with a foreword by the World Bank chief economist for Africa candidly describing the 'puzzle' that the study was meant to solve: [W]hen the Arab Spring took sudden and dramatic hold of North Africa in 2011, many pointed to high and rising inequality as one of the causes. There was only one problem: Income inequality, as conventionally measured, was not high in the region. And in Egypt, its pivotal country, it actually declined in the decade preceding the 2011 revolution -even as the public concern with inequality and injustice [was] growing. (Verme et al., 2014: xi, emphasis in the original) Indeed, the study is not apologetic in the least about the clash between the official findings that the IFIs endorsed and the widespread reference to 'high and rising inequality' as one of the core causes of the uprising in Egypt. It accepts unreservedly the view that the country's Gini coefficient is 'similar in level to that of the developed OECD members' and 'lower than inequality in most countries that are close to Egypt in terms of GDP per capita' (ibid.: 4). Mobilizing academic references in support of the wellknown view that periods of economic growth engender frustrations due to the increased expectations that they provoke, 15 the main chapter entitled 'Facts and Perceptions of Inequality' and authored by , explores the 'mismatch' between facts and perceptions that is to blame for the frustration that exploded in 2011.
The argument can be summarized as follows: in 2000-09, Egypt experienced sustained GDP growth and increased volatility in prices, favouring a heightened level of social and economic awareness. Since the growth in GDP did not trickle down to households while absolute household welfare was declining, the 'perceived decline in welfare was greater than the actual decline' while the 'expected welfare' increased a frustration aggravated by the lack of democratic institutions. This engendered 'an evident sharp rise in inequality aversion for almost all income groups and all social groups' (ibid.: 96-97). The perception of higher inequality is also attributed to a change in the 'reference group' of social self-comparisons, resulting from the expansion of the new means of information and communication: The expansion of Internet-based social networks has clearly . . . expanded the reference group to encompass a much larger number of people and has broken the national boundaries of the reference group. Through social networks, people gained more peers and peers abroad, Increased aversion to inequality in Egypt is reported by Verme from two opinion polls realized in that country in 2000 and 2008 by the World Value Survey (WVS). They reveal a strong rise in inequality aversion in Egypt over those years, matching a similar trend of what Verme characteristically calls a 'reduction in the appreciation for inequality' at the MENA regional level (ibid.: 90). This, and not a rise in real socio-economic hardship, is the cause of the explosion that shook the whole region in 2011. Thus, for Egypt, 'an attentive look into these data might have provided a different picture from the one portrayed by GDP growth alone' (ibid.); that is, the picture conveyed by the IFIs in the years preceding the uprising.
It did not occur to Verme to explore the possibility that the rise of inequality aversion during the first decade of the 21st century, correlated with a perception of rising inequality, might have been a reason for the decline in inequality reported in household surveys during the same period. In other words, the World Bank study did not consider whether the 'evident sharp rise in inequality aversion' that it noted could have increased the top strata's propensity to either understate their income expenditure or dodge the survey altogether, especially in a country and a part of the world characterized by a high degree of governmental unpredictability, political instability and high levels of corruption in the view of the business milieus themselves, as the World Bank knows well (Gatti et al., 2013, pp. 128-37;World Bank, 2009; see also EBRD et al., 2016).
Instead, the inquiry strives to locate the reason for the rise in inequality aversion in political-cultural features presumably characteristic of the region. From the WVS findings, Verme concludes that the roots of inequality aversion are to be found in the lack of democracy and freedom, but also in a lack of trust in other people, a low taste for democracy, a high level of religious practice, and conservative and religious views about gender roles (Verme et al., 2014: 91-95), all usual 'Orientalist' tropes. Neoliberal clichés inspired by 'methodological individualism' are also part of the analysis where 'People who think that work is very important and that cheating on taxes is never justifiable are clearly pro-inequality' (ibid.: 94).
Verme overlooked an important essay by Malte Lübker (2004) on the relation between globalization and perceptions of social inequality, 16 which shows how differences in perception between countries and regions at similar levels of inequality (similar Gini coefficients) are related to locally prevailing notions of social justice. This is illustrated by the case of 'the transition economies of eastern Europe' where several studies observed a degree of inequality aversion and prevalence of egalitarian principles higher than in 16. See also Lübker's (2007) study on inequality and the demand for redistribution, refuting the rational choice theory linking redistribution to reduced growth. other countries. This bears a direct relation to the fact that 'people grew up under systems whose ideology advocated the elimination of class differences, and they experienced a reality that, though not living up to this goal, achieved extremely low levels of inequality by any historical or international standards' (Lübker, 2004: 100). Lübker's findings provide a clue to high inequality aversion in the Arab region, and particularly in Egypt where the legacy of 'Arab socialism' is strongest. They are consistent with Amartya Sen's (2000) argument that the perception of inequality is a function of both the actual degree of inequality and prevailing norms of social justice, and that it varies therefore according to both factors. The perception of high inequality is never a fantasy, even though its acuteness at equivalent levels of real inequality may differ according to ideological mindsets. Indeed, as Lübker put it, 'the greater the inequality, the more it is generally condemned ' (2004: 125).
But in the view of the 2014 World Bank study, it is inequality aversion, not inequality per se, that should be deplored, since inequality must inevitably rise with development from a Kuznetsian perspective. 17 Had GDP growth been accompanied by a trickle-down effect, the Egyptians would have had a more positive view of inequality, as 'people can hardly appreciate inequality if their own status and the status of their peers do not improve' (Verme et al., 2014: 97). Following the same logic, in order to conform to the Kuznets curve, it is more inequality rather than less that Egypt needs, since its purported low level of inequality is due to overall poverty: Reducing poverty among the poorest people of Egypt may necessarily entail an increase in inequality and the low level of inequality is explained by the widespread level of poverty rather than by an economy that distributes increasing resources equitably across the population. This is in line with the idea that, at very low levels of incomes, an increase in inequality may signal an improvement in overall living conditions while very low levels of inequality may simply signal widespread poverty. (ibid.: 11) From the same neoliberal perspective, the government -in tackling poverty, which is narrowly related to the lack of employment -should not provide jobs, but only training and education that would allow the private sector to recruit, in the spirit of the World Bank's 'poverty reduction strategies': Providing work opportunities in the government or the public sector would reduce inequality; however, it would also have a negative impact on household poverty status and lead to increased overall poverty in the light of the very low wages provided for government employees. The results suggest that in the context of Egypt's poorest villages, some of the most significant factors conducive to increased inequality -such as higher employment and higher education -are the same standard objectives of pro-poor poverty reduction strategies. Hence, pro-poor growth accompanied by increased inequality is a legitimate goal that should be pursued. (ibid.: 120) This assumption is clearly contradicted by CAPMAS's survey. The IECS for 2015 showed that the average annual income from work (wages) of rural Egypt households whose head works in the governmental or public sectors is significantly higher than that of households whose head works in any other sector, including private business, cooperatives and non-governmental organizations (CAPMAS, 2016: Tables 3-42).

Alvaredo and Piketty
Shortly after the publication of the 2014 World Bank study, the Cairo-based ERF published a working paper, authored by Facundo Alvaredo and Thomas Piketty (2014). This paper provides a pertinent counterpoint to the World Bank study, openly stating its disagreement with the latter: 'Of course, we agree that there are potentially many sources of dissatisfaction other than the value of the Gini coefficient or the top decile income share. . . . However, we disagree about the claim that income inequalities in Egypt or the Middle East are quantitatively small by international standards' (Alvaredo and Piketty, 2014: 4). The two authors criticize the World Bank team's acceptance of the reliability of household surveys because of statistical plausibility: [H]ousehold income and expenditure surveys that are generally used by economists and international organizations almost certainly underestimate the level of inequality, possibly by a very large margin. One would need reliable fiscal sources in order to make a precise comparison between the top decile or percentile income shares prevailing in Egypt (or other Middle East countries) and the top shares prevailing in other emerging or developed countries. Unfortunately, such sources are lacking in the region, so that no satisfactory comparison is possible at this stage. . . . While the lack of transparency on income and wealth is an important issue in many (if not most) areas of the world, it appears to be particularly extreme in the Middle East, and arguably raises in itself a problem of democratic accountability, quite independently from the actual level of inequality. (ibid.: 2) There indeed lies the major flaw of Hlasny's and Verme's (2013) working paper according to Alvaredo and Piketty, who base their counterargument on statistical calculation. They make the following argument: Hlasny and Verme use household income surveys for Egypt between 1999 and 2010, and find relatively small Gini coefficients (below 0.35). They then argue that the inverted Pareto coefficient b is about 1.5-1.7, and is in line with other countries. The problem is that they compare the Egyptian b to coefficients that also come from household surveys, which are always artificially small. If we compare their 1.5-1.7 coefficient to the more reliable inverted Pareto coefficients estimated using tax data, then the Egyptian b coefficient is actually extremely small by international and historical standards. Of course, it is possible that Egypt is currently as egalitarian as the most egalitarian countries in history (such as Scandinavian countries in the 1980s). However, this does not seem overly plausible. (Alvaredo and Piketty, 2014: 5) For Alvaredo and Piketty, the only truthful way to estimate Pareto coefficients and top income shares in the Middle East (defined as including the Arab countries of West Asia and Egypt), and make proper comparisons with other countries, is to use income tax data or similar data, such as inheritance tax or wealth tax. But these kinds of data are not available for most of the region, as the authors observe. In fact, the condition that they rightly set for an accurate calculation of socio-economic inequalities renders it simply impossible in the Arab region. For, even where taxation data are available, they do not guarantee much higher accuracy in the estimation of inequality. This is because taxation of income and profit is eminently flawed and unreliable in a region where only wage-earners in the formal sector, which is estimated to account for only one third of total regional employment, are duly subject to tax, usually withheld from their pay. Alvaredo and Piketty (ibid.: 1) make a second important point, which connects with the issue of the 'reference group' discussed in the World Bank study: Next, and irrespective of these uncertainties on within-country inequalities, there is no doubt that income inequality is extremely large at the level of the Middle East taken as wholesimply because regional inequality in per capita GNP is particularly large. According to our benchmark estimates, the share of total Middle East income accruing to the top 10 per cent income receivers is currently 55 per cent (versus 48 per cent in the United States, 36 per cent in Western Europe, and 54 per cent in South Africa). Under plausible assumptions, the top 10 per cent income share could be well over 60 per cent, and the top 1 per cent share might exceed 25 per cent (versus 20 per cent in the United States, 11 per cent in Western Europe, and 17 per cent in South Africa). Popular discontent might reflect the fact that perceptions about inequality and the (un)fairness of the distribution are determined by regional (and/or global) inequality, and not only by national inequality. This goes beyond the common emphasis on the extremely high inequality between countries, and is a point well taken, which was further developed by Alvaredo and Piketty, along with Lydia Assouad (Alvaredo et al., 2017). It is true indeed that the sense of frustration in the Arab region is not restricted to intra-country inequality but is also exacerbated by the huge level of social inequality in the region as a whole. In this regard, it is not only the expansion in the means of information and communication that is at work, but also, and to a high degree, the direct experience of millions of Arab migrant workers -the majority of them Egyptian -who witnessed the conspicuous, often ostentatious, lavishness of the super-rich in the oil and gas monarchies of the Gulf Cooperation Council.
In this regard too, the legacy of 1960s 'Arab socialism' -in particular its denunciation of the plundering by the West of the region's natural resources, seen as belonging to the whole 'Arab nation' -remains crucial in determining the deep sense of social injustice that prevails in the region. This legacy has been revitalized and energized by the shocking spectacle of an ever-increasing social gap, both at the country level and at the regional one.

The House Prices Test
The challenge posed by Alvaredo and Piketty to Hlasny's and Verme's analysis, and to the 2014 World Bank report based on it, prompted a team of three World Bank researchers, Roy van der Weide, Christoph Lakner and Elena Ianchovichina, to probe the conflicting assessments against the evidence provided by house prices as an indicator of the top tier of income distribution. The starting point of the authors' working paper is that household surveys fail to capture the top of the income pyramid (van der Weide et al., 2016). To confirm this in Egypt's case, they compare between the top 1 per cent of the household survey and senior Egyptian executives, using data on executive pay from Payscale, an online company providing information on salary, benefits and compensation. The authors calculated the total compensation of senior executives, who represent 2 per cent of survey participants. They found that the median senior executive income is closer to the maximum income in the household survey than to the median income of the richest 1 per cent, and that the maximum income earned by senior executives is much higher than the survey's maximum income, thus confirming that 'the household survey under-represents the top earning households, particularly the top earning senior executive households ' (ibid.: 15-16).
The authors also recognize that Hlasny's and Verme's attempt to offset the unavailability of income tax data for Egypt by modelling and re-weighting existing household survey data deemed to be flawed, was not convincing since 'no adjustment that relies solely on the survey will resolve the downward bias in estimates of inequality' (ibid.: 3). They instead use data on house prices compiled from real estate listings available in the public domain in 2014/15 to assess the top tier of the income distribution, by estimating the relation between house price and household income as reported in the household survey. This, the authors note, demands 'two methodological innovations' to the study of top incomes: Firstly, we will not be observing actual household income or expenditure (as is the case with tax record data), but rather a predictor of income. Secondly, a database with house price listings is generally not obtained using a particular sampling design. Therefore, the data are not guaranteed to provide a nationally representative sample, they will arguably be biased towards large urban centers. (ibid.) The conclusion reached by the authors confirms the widespread suspicion that inequality in Egypt is considerably underestimated by official figures: they found that when house prices are factored in, the Gini coefficient for urban Egypt increases considerably from 36.4 to 47 (ibid.: 24). This implies a still higher Gini coefficient for the whole country, given the appalling level of abject poverty found in rural Egypt. 18 On the basis of their calculations 18. 'The multidimensional poverty index (MPI) . . . shows that the Arab region has the highest ratio of rural to urban poverty (3.5) among all developing regions except Latin America related to house prices, the authors estimate that 'there are approximately 170 households in Cairo whose household expenditure exceeds 1 million USD per year', believing their estimate to be 'rather conservative ' (ibid.: 25).
In a revised version of their 2016 paper published in October 2018 in a supplement issue of the Review of Income and Wealth on 'Poverty, Inequality, and the Arab Spring', the authors use a substantially higher number of observations taken from the 2008/09 household survey (10,763) than they did in the working paper (6,935). They revise their figures accordingly: whereas the Gini coefficient for urban Egypt calculated on the basis of the 2008/09 household survey stands at 38.5, it increases by over one third when household prices are factored in, reaching 51.8 (van der Weide et al., 2018: 76-77). The authors' revised estimate is that 'there are approximately 300 households in Cairo whose household income exceeds 1 million USD per year', while they still regard this estimate as 'rather conservative' (ibid.: 77).

BRINGING EGYPT BACK IN REGIONAL CONTEXT: ESCWA
A similar conclusion about the implausibility of Egypt's official low and declining inequality figures was drawn in 2015 by two researchers of the UN Economic and Social Commission for Western Asia (ESCWA), Khalid Abu-Ismail and Niranjan Sarangi. 19 From the examination of changes in average real expenditure for different tiers of the income pyramid in Egypt between 2000 and 2011 on the basis of CAPMAS household surveys, they find that, over that period, 'Egyptian poor households witnessed a decline in average real expenditure per capita by 1.54 per cent as against an 8.96 per cent increase of the same for the affluent groups' while 'vulnerable and middle-class groups witnessed an increase in expenditure per capita by 1.75 and 1.92 per cent, respectively'. They conclude that these results indicate a 'growing disparity between population classes' (Abu-Ismail and Sarangi, 2015: 16). 20 In the same year, the same authors, along with two other ESCWA researchers, published a study on poverty and inequality in Arab countries and the Caribbean' (UNDP, 2016: 24). For an insightful discussion of poverty and social exclusion in Egypt, see Bush and Ayeb (2012). 19. The UN ESCWA is composed of 18 MENA states, all members of the League of Arab States. Yet, like other UN research bodies, it has accommodated a few critical researchers over the years. 20. The two researchers had produced an initial version of their working paper two years earlier, quoting a surprising 23 per cent decline in average real expenditure per capita for all households, ranging from 24 per cent for the poor to 16 per cent for the affluent (Abu-Ismail and Sarangi, 2013: 17). Like their World Bank peers discussed in the previous section of this article, the ESCWA researchers provided no explanation for the change in their new version. . Challenging explicitly the official World Bank perspective represented by Hlasny and Verme, Sarangi et al. compare what they call 'the official growth narrative' (private household final consumption expenditure in national accounts) and the narrative based on household surveys of consumption expenditure. They find a 'large and growing divergence' between national household final expenditure per capita from national accounts and household consumption expenditure per capita from surveys for a sample of Arab countries, in contrast to the official 'picture of low and stagnant inequality' in Arab countries (ibid.: 6-7). They set out their findings as follows: The difference between the two indicators should, theoretically, be minimal and similar over time . . . . However, for all countries in the sample, the results reveal considerable disparity. For example, in Egypt, the per capita household final expenditure is 2.6 times higher than that reported by the survey based on per capita consumption expenditure. Similarly high levels of divergence are noted in Oman, Jordan and Tunisia. Importantly, the gap between the two measures is increasing over time for all countries. Empirical exercises from several countries and conceptual analysis provide a basis for arguing that the widening divergence indicates increased inequality over time, even taking into account the fact that household expenditure surveys may miss some items of consumption, and that national household final expenditure includes some components that household consumption surveys do not cover. (ibid.: 6) Taking the analysis further, ESCWA issued a major report in 2018 on inequality in the Arab region that starts by examining the historical roots of the Arab uprising under the heading of 'The Build-up of Inequality', followed by two equally telling subheadings: 'The rise and rollback of the State' and 'From State-led growth to an unbalanced form of capitalism' (ESCWA, 2018: 11-21). The report describes what it calls the 'extreme retrenchment of the state' during the 1980s and 1990s due to the fiscal adjustment programmes -stopping short of mentioning that these programmes were sponsored, if not enforced, by the IFIs.
Egypt provides a stark illustration of the retrenchment: between its highest point in the 1970s and 1998, state expenditure went down in Egypt from 62 per cent of GDP to 25 per cent. 'Subsidies on consumer goods were slashed most during the adjustment period. In Egypt, they fell from a whopping 23 per cent of GDP at their height in the 1970s, to less than 2 per cent of GDP at the end of the adjustment period of the 1990s' (ibid.: 16). Public investment declined dramatically over the same period in Egypt, from a high of 19 per cent of GDP to 8 per cent. As a result, economic growth rates fell after the 1980s (ibid.: 11-14). The natural consequence has been a steady rise in unemployment.
The report goes on, describing the rise of crony capitalism and corruption in the Arab region, as well as the political factors that led to the 2011 explosion against that background. This general picture refutes the report's assessment of the impact of this evolution on inequality. It omits in this respect the important findings of the working paper by van der Weide et al. (2016) and, more surprisingly, the equally important findings by ESCWA researchers Sarangi et al. (2015). Referring to the 'Arab inequality puzzle', the 2018 report notes that 'seasoned regional observers agree that inequality must be rising in the region, even if this has not been visible in the data' (ESCWA, 2018: 29). In that regard, it approves the reservations on household surveys expressed by several researchers, referring to Alvaredo et al. (2017) in particular. However, since 'it is plausible to assume that the underestimation of inequality due to missing top incomes does not vary much over time and across countries' (ESCWA, 2018: 31), the report uses the available data for a comparison across time and countries that is its main focus. It ends up displaying what resembles a conciliatory stance, accepting at face value the existence of a 'trend of decreasing income inequality' (ibid.: 34) in some countries, including Egypt and Tunisia, in striking contradiction with the narrative of the report's first chapter, while acknowledging that: There is reason to believe that the oil boom of 2004-2014, which brought incredible wealth into the region, may have increased the share of top incomes beyond what extrapolations of the survey data can project. . . . The reason for the increase in inequality tied to the inflow of oil money is simple: the transfer of oil rents to private individuals follows paths set by family and clientelistic networks which, by nature, increases only top incomes and consequently overall inequality. (ibid.: 37) The aftermath of the Arab Spring saw an intense debate about inequality in Egypt. While research during this period saw a clear tilt towards a critique of the low and declining inequality thesis, this body of research was not enough for the IFIs to depart from their long-held views on inequality in Egypt.

THE ARAB SPRING IN THE WORLD BANK'S PERSPECTIVE
In an another remarkable and still more striking illustration of the discrepancy that may exist between working papers by individual researchers at intergovernmental institutions and these institutions' official output, the World Bank published in the same year (2018) a major study on the economics of the Arab Spring and its aftermath. Its main author is Elena Ianchovichina, one of the three authors of the papers on house prices by van der Weide et al. (2016van der Weide et al. ( , 2018. She is presented as having worked 'under the general guidance' of Shantayanan Devarajan, then Chief Economist of the World Bank's MENA region before being appointed as the Bank's Senior Director for Development Economics. Her two co-authors of the papers by van der Weide et al. are named as having contributed inputs on economic inequality.
One could therefore expect that the new study would move away from trying to reduce the view of actual and rising social inequality in the Arab Spring to that of a mere misperception of an actual pattern of low and declining inequality -the World Bank's solution to the 'Arab inequality puzzle'. And yet, the study fully espouses the views expressed by Hlasny and Verme in 2013 and reiterated in the 2014 World Bank report by Verme et al. on inequality in Egypt. As clearly apparent in its own summary of findings, it asserts this view while acknowledging the possibility of its opposite: [It] rules out high and rising inequality as a reason for the Arab Spring uprisings. It finds that expenditure inequality in most Arab countries was low to moderate in the years before the uprisings, although it may have been substantially underestimated in the data because of missing top incomes, as shown for Egypt and other countries around the world. (Ianchovichina, 2018: 11) The study could not possibly ignore the findings of van der Weide at al. about house prices. It summarizes them, affirming that they 'provide evidence that the level of inequality in urban Egypt has been significantly underestimated' (ibid.: 32). However, it then brushes them aside by asserting peremptorily a 'declining trend in expenditure inequality in Tunisia and Egypt' with reference to studies based on official data and published two years before the first paper by van der Weide et al. came out in 2016 (ibid.: 33). In line with the IFIs' mainstream analysis of the Arab Spring, the study reduces the regional upheaval to a revolt of disgruntled well-off categories: 'mostly middle-class and relatively affluent people rose to protest the status quo', affirms the study (ibid.: 117) -a description that would come as a surprise to those who took part in the regional uprisings.
The 'middle class' is a very extensible concept, of course, and the study acknowledges the debate on this issue by referring to the older version of Abu-Ismail's and Sarangi's (2013) paper. According to the criteria retained by the study, there is a huge gap between regional countries: 'The middle class was smallest in Egypt and Syria, where it accounted for between 10 and 15 percent of the population; and it was largest in Tunisia and the West Bank and Gaza, where the majority of the population belonged to the middle class' (Ianchovichina, 2018: 45). And yet, despite a plausible estimate for Egypt, the study regards the uprising there as having been 'middle class' and 'relatively affluent' all the same. For this, the only evidence it produces is a graph based on findings of the Arab Barometer, a research and polling network involving the universities of Princeton and Michigan with Arab partners. 21 For Egypt, the Arab Barometer surveyed 1,219 people in June 2011 (Arab Barometer, 2011: 3), selected out of the CAPMAS household survey's sample. The survey was not originally meant to study the uprising (Soltan et al., 2011). Thus, only 8 per cent of the surveyed admitted taking part in the revolutionary protests, that is, less than 100 people. Among this tiny number of declared participants, the proportion of those belonging to the fifth (richest) 21. The Arab Barometer epitomizes what critics of the industry of opinion polls have long been warning against (see Bourdieu, 1993). and third income quintiles was higher than 20 per cent while it was less than 20 per cent for those who belonged to the first (poorest), second and fourth quintiles. This very limited sample of protesters was enough though for Princeton scholars involved in the Arab Barometer research to assert that 'Egyptian Revolution participants were overwhelmingly recruited from the middle class' (Beissinger et al., 2012: 17), a conclusion happily endorsed by the 2018 World Bank study. Such views neatly fit the neoliberal thesis that the 2011 Arab Spring was essentially a revolt of a 'squeezed' middle class, frustrated at the impediments hindering its private enrichment. Hernando de Soto (2011) crudely expressed a similar thesis in his assessment of the Arab Spring, referring to the poor young Tunisian street vendor who triggered the shockwave by setting himself on fire on 17 December 2010: 'The forces of the market have come to the Arab world -even if governments didn't invite them in. Political leaders must realise that, since Bouazizi went up in flames and his peers rose in protest, poor Arabs are no longer outside but inside, in the market, right next to them'.
Proposed solutions to this situation focus on reduction in public employment and curtailment of public subsidies of basic food staples and energy prices. As usual, the World Bank locates the region's problem in the constraints imposed on the private sector and recommends that: [C]itizens must be empowered to become active participants in the private economy. To empower private entrepreneurship, governments must demonstrate their commitment to business-friendly reforms that lower the costs of doing business, reduce complex regulations protecting the rents of connected firms, and ensure unbiased application of the laws; they must also make investments that complement and do not crowd out private investment. (Ianchovichina, 2018: 18) In sharp contrast with the democratic and rule-of-law aspects of what the IFIs call 'good governance', neoliberal austerity and deregulation measures are systematically forced on Arab governments when the latter are reluctant to implement them for fear of arousing popular anger. Egypt provides the most striking illustration of the natural logic of this policy: it took the most authoritarian and brutally repressive regime that the country has known in decades to implement, from November 2016 onwards, the full range of austerity and pro-business measures that the IMF had been demanding for several years and that no previous president was able to impose on the Egyptian population. 22 As was to be expected, this led to further and brutal decline in the standard of living of most Egyptians and a sharp increase in poverty. This is confirmed by the main findings of the IECS conducted by CAP-MAS in 2017/18, which reveal a conspicuous increase in poverty: the proportion of Egyptians living under the national lower poverty line soared from 22. On the political context in Egypt prior to the November 2016 measures, see Achcar (2016). 27.8 per cent in 2015 to 32.5 in 2017/18 (CAPMAS, 2019: 77). The publication of the new findings had been postponed from January to July 2019 'due to the objections of "sovereign entities" to the survey results', according to the Egyptian media (Mada Masr, 2019). And yet, according to the findings of the same survey, Egypt's Gini coefficient has remained remarkably low at 0.30 since 2015 (CAPMAS, 2019: 81). In an interesting interview, Heba al-Laithy, a Professor of Statistics at Cairo University who supervised the 2017/18 survey, recognized that the way the Gini coefficient is calculated by CAPMAS does not reflect the real degree of inequality in the country (Kassab, 2019).
Between 1999/2000 and 2017/18, the proportion of Egyptians living under the national lower poverty line almost doubled from 16.7 to 32.5 per cent, according to the new official survey (CAPMAS, 2019: 77). The proportion of those living in extreme poverty, unable to satisfy their basic human needs, more than doubled during the same period, jumping from 2.9 to 6.2 per cent (ibid.). These are the most striking results of an economic liberalization that has mostly profited cronyism and corruption, hence provoking a steep rise in social inequality with the tip of the social pyramid getting ever richer while the lower tiers are increasingly unable to satisfy their basic needs.
Acknowledging this causality would require a recognition of the need for radically different policies combining a reinstatement of the social and developmental role of the state, 23 along with conditionalities centred around concrete steps towards achieving 'good governance' (in IFI speak) in the sense of democratization. This, however, would require a sharp break with the neoliberal paradigm of which the IFIs have been the zealous guardians since the build-up of the Washington Consensus in the 1980s.

CONCLUSION
Dogmatism is indeed a key feature of the neoliberal paradigm. Impervious to blatant empirical evidence of its failure, not least in the Arab region, a perspective purporting to be the only holder of the economic truth is adamant in carrying on with its policy recommendations against all odds. It thus forgets that it subverted the Keynesian paradigm by blaming on it the 1970s' global recession, a crisis that was indisputably milder than the Great Recession that started unfolding in 2007.
In the Arab region, the IFIs, acting as guardians of the neoliberal credo, have deliberately ignored the fact that their recipes were responsible, to a 23. By the IMF's own admission, 'Raising public investment in the Arab Countries in Transition by 5 percent of GDP cumulatively over five years would require $24 billion in financing, while significantly raising growth and employment ' (Finger and Gressani, 2014: 12). For an alternative view of social and economic policies in the Arab region based on social justice, see the report edited by Said (2017). large extent, for the formidable socio-political explosion of the Arab Spring and the protracted destabilization of the region that it inaugurated. There is no better illustration of this wilful blindness than the insistence on endorsing the highly implausible view that the Arab region ranks among the world's least socially unequal.