Arab Development Challenges Background Paper 2011/01 The ADCR 2011: Arab Human Development and Deprivation: Phenomenal Progress or Mixed Results?

Transcription

1 Arab Development Challenges Background Paper 11/1 The ADCR 11: Arab Human Development and Deprivation: Phenomenal Progress or Mixed Results? Khalid Abu-Ismail, Abdel-Hameed Nawar, Asmaa Abdel-Nabi and Gihan Abou Taleb

2 United Nations Development Programme Arab Development Challenges Report Background Paper 11/1 The ADCR 11: Arab Human Development and Deprivation: Phenomenal Progress or Mixed Results? Khalid Abu-Ismail, Abdel-Hameed Nawar, Asmaa Abdel-Nabi and Gihan Abou Taleb* Khalid Abu-Ismail is Poverty and Macroeconomics Adviser at United Nations Development Programme- Regional Centre in Cairo (UNDP-RCC). Abdel-Hameed Nawar is Lecturer at Cairo University. Asmaa Abdel-Nabi is Research Assistants at UNDP-RCC. Gihan Abou Taleb is Research Assistants at UNDP-RCC. Comments should be addressed by to the author(s) ii

3 Acronyms and Abbreviations AC DHS DR EAP ECA GCC GDP GNI HDI HDR HIES HPI KSA LAC LDC LEED LIC MD MDGs MIC MICS MPI OPHI OPT PPP SAS SSA UAE UNDP UNICEF WHS Arab countries Demographic and health Surveys Developing region East Asia and Pacific (developing countries only) Europe and Central Asia (developing countries only) Gulf Cooperation Council Gross domestic product Gross national income Human Development Index Human Development Report Household Income Expenditure and Consumption Survey Human poverty index Saudi Arabia Latin America and Caribbean (developing countries only) Least developed countries Leadership in Environment, Energy and Design Low income countries Multi -Dimensional Millennium Development Goals Middle income countries Multiple Indicator Cluster Survey Multi-Dimensional Poverty Index Oxford University Poverty and Human Development Initiative Occupied Palestinian Territories Purchasing power parity South Asia (developing countries only) Sub-Saharan Africa(developing countries only) United Arab Emirates United Nations Development Programme United Nations Children s Fund World Health Survey iii

4 Introduction The debate about the development model followed in the Arab region has become more intense following the recent wave of political contestation in the region, despite a long period of sustained increases in GDP. Clearly the increase in income has not been sufficient to cement social cohesion and enlarge the freedoms and sense of dignity enjoyed by Arabs. This is not unanticipated, as the development model has not followed the path of peoplecentred development that UNDP has been championing since the launch of the first Human Development Report in 199. The economistic vision of development, development by means of increased production of goods and services, can easily go hand in hand with less social cohesion, usurpation of traditional rights and freedoms and undermine human dignity by treating human beings as commodities. Human development, while clearly dependent on material welfare and the provision of the basic necessities of life, is really about people having greater freedom from the elements and from oppressive dependence on other human beings. Human development is about creating a society in which human dignity is respected and human agency is promoted. It is thus inherently in contradiction with situations of mining of environmental assets and denial of basic human dignity. No wonder that many years of unbalanced development led to cries of dignity, freedom and social justice in the Arab region. The human development paradigm postulates a development process that expands human capabilities while increasing material production. In a practical sense it entails/envisions a development model, which spends heavily on improving the human condition by investing in accessible, quality and liberating education that empowers individuals to better understand and respond to developmental opportunities. It also implies general provision of quality health care services to safeguard the sanctity of human life and mitigate and respond to debilitating health conditions that undermine human dignity. The model assigns equal value to all humans and as such is inherently incompatible with large degrees of inequality in income and wealth. In this paper we are also concerned with measuring human development achievements and deprivation from human development, which can also be appropriately defined as human poverty. In general, measuring the level of poverty in a country or society can be a challenge if only because of the subjective element involved in determining who was poor. There was a widely accepted idea that for any given society, poverty exists if an individual (or household) was unable to attain a certain standard of living, or well-being ; that was deemed the minimum acceptable by the standards of that society. 1 Most societies also share, however, a concept of absolute poverty that goes beyond any subjective standards. When people lack the basics of nutrition, health, sanitation and housing, they are poor by the principles of any society. Well-being is thus multidimensional and driven by multiple deprivations the lack of food and shelter, but also the lack of incomes and opportunities, of access to services and of ways to protect themselves against sudden shocks. 2 These are, therefore, not driven purely by monetary factors. More broadly, poverty involves the inability to command resources, such as being healthy and literate. Poverty in this latter sense would constitute a lack of capabilities. The freedom approach offers a broader conceptual framework for understanding poverty and human development compared to other approaches, which identify development with increases in per capita income, industrialization, technological advance, or social modernization. The richness of this approach is best captured in terms of five instrumental freedoms that have immediate policy relevance: political freedoms, including the political entitlements associated with democracies in the broadest sense ; economic facilities, in the 1

5 sense of the opportunities that individuals respectively enjoy to utilize economic resources for the purpose of consumption, or production, or exchange ; social opportunities in the sense of the arrangements that society makes for education, healthcare and so on ; transparency guarantees in the sense of the freedom to deal with one another under guarantees of disclosure and lucidity ; and, protective security in the sense of the provision of a social safety net for preventing the [vulnerable sections of society] from being reduced to abject misery and in some cases even starvation and death. These instrumental freedoms tend to the general capability of a person to live more freely, but they also serve to complement one another. 3 Development achievement on the basis of this approach, which is now being measured and regularly reported, by the Human Development Index (HDI), is crucial to grasp the concept of human poverty since "poverty means that opportunities and choices most basic to human development are denied". 4 The Human Poverty Index (HPI) and the more recent Multidimensional Poverty Index (MPI) are therefore mirror images of the HDI that reflect deprivation from instrumental freedoms. Anchored in UNDP s human development approach, this paper paints a broad picture of human development and human poverty in the Arab region by comparing its human development and poverty indicators with those of other developing regions. The paper also evaluates the state of human development and poverty in Arab countries relative to their respective income levels. Finally, the paper makes use of household survey data to present new findings regarding the sub-national disparities in multidimensional poverty, health and education within Arab countries and ends with concluding remarks. Human Development The 1 HDR introduces several reforms to the indicators and methodology used to calculate the Human Development Index. In previous reports the basic dimensions of the HDI were measured by life expectancy at birth; adult literacy rate and combined gross enrolment in education; and GDP per capita in purchasing power parity US dollars (PPP US$) respectively. In the 1 report, the indicators measuring access to knowledge and a decent standard of living have changed. Namely, access to knowledge is measured by: mean years of adult education, which is the average number of years of education received in a life-time by people aged 25 years and older; and expected years of schooling for children of school-entrance age, which is the total number of years of schooling a child of school-entrance age can expect to receive if prevailing patterns of age-specific enrolment rates stay the same throughout the child's life. Additionally, the standard of living is now measured by Gross National Income (GNI) per capita in PPP US$, instead of Gross Domestic Product (GDP) per capita in PPP US$. Moreover, a multiplicative method of aggregation the indicators into an index is used. That is aggregation is made using the geometric mean of the dimension indices composing the HDI. Based on this new methodology, in Chapter 2, the report presents a list of ten top movers countries that have seen the greatest improvements in human development, as measured by the Human Development Index, relative to their 197 starting point. Among these, the presence of five Arab countries (Oman, Saudi Arabia, Tunisia, Algeria and Morocco) is striking. The 1 HDR, however, adopts a specific definition of success the actual rate of change in the HDI relative to a country s starting point, calculated as the residual of a regression of the change in HDI on the country s initial HDI level. It measures how well countries did in comparison to other countries with similar initial HDI levels. The exceptional performance of these countries, however, does not seem to be sensitive to the definition of success adopted: "regardless of how we measure it, the performance of these countries in health and education indicators has been stellar." 5 2

6 For example, in 197, a baby born in Tunisia could expect to live 54 years; one born in China, 62 years. Today, life expectancy in Tunisia has risen to 74 years, a year longer than that of China. So while China s per capita income grew almost three times as fast as Tunisia s, Tunisia s life expectancy grew twice as fast as China s. Since it also significantly outperformed China on the education front, Tunisia gives China a run for its money in the overall development story (as captured by the HDI). Table 1: Ranking of Arab countries in terms of change in HDI, and HDI Improvement Rank Country HDI value Non- Income HDI rank Non- Income HDI value GDP Growth rank GDP Growth HDI Improvement Rank HDI value Non- Income HDI rank Non- Income HDI value GDP Growth rank GDP Growth 1 Oman % % 5 KSA % % 7 Tunisia % % 9 Algeria % % 1 Morocco % % 13 Libya % % 17 Egypt % % 19 UAE % % 34 Bahrain % % 43 Jordan % % 58 Qatar % % 67 Sudan % % 68 Kuwait % % 94 Lebanon % % 122 Djibouti % % Source: Human Development Report, 1. One way to illustrate how striking the progress of these countries was is to look at how well they did in comparison with other countries according to the three components of the HDI. As argued in the report, the comparison between the evolution of life expectancy in these countries and the world average over time, as well as that of a subset of countries with a similar starting point shows that they started well below the world mean in 197 but by 1 had overtaken the rest of the world and significantly outperformed countries that started off in the same place. Similar patterns characterize their performance on the education front. Since almost all Arab countries had very low baselines for health and education in 197, the very high disparity between where they are now compared to what was expected given their initial conditions in the 197 s is not surprising (this is particularly true for Oman). Interestingly, this was not the case for economic growth. The average growth rate of per capita income in Tunisia, Algeria and Morocco, which are among the top-ten movers, was only 2.1%, just slightly above the world average of 1.7%. In terms of growth, the best performer of the lot is Tunisia (3%) but even here 19 countries had higher growth rates for the forty-year period. In contrast, only six countries outperformed Tunisia in terms of improvements in the HDI s combined measure of health and education. As noted in the report, this illustrates a broader point. There is a remarkably weak correlation between economic growth and improvements in health and education. The correlation between these two variables over a forty-year period is.13 and is not statistically significant. Also striking is that the list does not include most of the Arab oil-rich countries on the income improvements. This is explained, however, in part by the fact that the GDP per capita figures are in constant PPP US$ so they measure the increase in real value added. Hence, the higher level of GDP per capita for oil-rich Arab states in 197 may explain, to some extent, the relatively lower growth rates achieved over the entire period. The long-term development record of the region thus demonstrates that human development achievements are possible even without fast economic growth, a story that contrasts sharply 3

7 HDI Rank Hybrid HDI with the conventional stories of development success based on economic growth (such as the diversified, export-lead economies of the Asian Tigers). Despite what may be regarded as phenomenal progress, however, if 197 is instead taken as the base year, Arab countries still lag behind on human development. This is clear in Figure 1.A; which shows the HDI and GNI ranks in 1 for 18 developing countries. All Arab countries (with the exception of Tunisia, Jordan and Comoros) lie above the regression line that separates countries that have higher GNI relative to their HDI from countries that have lower GNI relative to their HDI. Thus, with the exception of Tunisia, all Arab countries, which have been identified by the global report as top-movers, still lag behind on human development relative to their income levels. The same picture emerges from the location of Arab countries on the regression line in Figure 1.B, which depicts the cross-sectional relationship between HDI and GDP per capita for the same year. Figure 1: GNI per capita rank versus HDI rank (A) and HDI and GDP per capita (B) for Arab countries and developing regions, 1 (A) (B) Sudan Djibouti Mauritania Yemen Comoros Morocco Syria Egypt Algeria Jordan KSA Tunisia Libya Kuwait Bahrain Qatar UAE GNI per capita Rank Source: Based on data from Human Development Report, 1. UAE BHR KWT LBY QAT LAC KSA LBN JOR OMN DZA TUN EAP EGY AR ECA MAR DR SAS SDN DJI SSA Ln GDP per capita, PPP Further compelling evidence, demonstrating the sensitivity of the measure of human development achievement to the choice of the period, is also provided in Table 1. Taking 199 as the base year, instead, changes the picture dramatically. First, HDI rakings drop for all Arab countries with the exception of Sudan, Kuwait, Djibouti and Lebanon, all of which lag behind in the ranking. Second, none of the top-ten Arab countries retain their position and Oman, the highest ranking Arab country for the period, regresses to the 15 th place and is overtaken by Morocco and Tunisia (ranked 12 th and 14 th respectively). Rankings for the non-income HDI also decline for the majority of Arab countries and only Oman and Saudi Arabia retain a leading position among the top-ten countries. Results for GDP, however, are more ambiguous, with countries equally divided among winners and losers. To sum, the region's more recent human development record indicates that the pace of progress has declined significantly since 199. This conclusion is also consistent with the findings of the Arab MDGR (1), which points to a mixed performance with respect to the attainment of MDGs when 199 is taken as the base year. The report classifies Arab countries into three main groups: countries that are likely to achieve MDGs targets (mainly GCC countries); countries with mixed progress and mixed opportunities, which constitute the bulk of Arab middle-income countries (MICs); and countries that are unlikely to achieve MDGs targets (mainly LDCs and MICs where security issues pose a serious threat to the achievement of MDGs, namely Iraq and the Occupied Palestinian Territory). The report also shows, however, that progress has been heterogeneous even across the various goals, with many Arab countries lagging significantly behind on employment, hunger, poverty and maternal health targets in particular. None of these indicators are directly captured by the HDI

8 To further explain how the choice of measurement methodology adopted by the HDR has an upward bias on Arab development achievements, we may, for the sake of illustration, propose an alternative methodology that measures progress in closing the HDI deficit between Arab countries and countries with the highest HDI score. Although this is a very different question to ask, it is nonetheless of direct relevance to any assessment of real progress in human development. Table 2: Evolution of HDI and real GDP per capita gaps between Arab countries and developing regions and highly developed economies and rank of convergence, for example, shows the results of one attempt to undertake such an exercise. The first column shows the ratio of the HDI of the country to the simple average for the countries with highest ten HDIs in 1. This ratio is then deducted from the same ratio in 197. Hence, countries that were most successful in closing the HDI gap will record the highest positive percentage change between both ratios. The third column ranks developing countries accordingly. The same procedure is repeated for the GDP in the second and fourth columns. Table 2: Evolution of HDI and real GDP per capita gaps between Arab countries and developing regions and highly developed economies and rank of convergence, 1 Country/Region HDI Gap GDP Gap HDI Rank GDP Rank Bahrain 14% -33% Kuwait 3% -75% Oman 88% 63% 4 14 Qatar 6% -57% 7 83 Saudi Arabia 49% -39% United Arab Emirates 23% 26% Djibouti % -8% Sudan 4% -3% Algeria 44% -29% Libya 29% -77% Morocco 54% % Tunisia 47% 62% Egypt 41% 25% Jordan 21% -17% Lebanon 3% -23% Arab countries 43% -1% 3 5 East Asia & Pacific 69% 526% 1 1 Europe & Central Asia -2% 1% 6 3 Latin America & Caribbean 14% -8% 5 4 South Asia 5% 78% 2 2 Sub-Saharan Africa 35% -31% 4 6 Developing region 38% 58% Source: Authors estimates based on data from Human Development Report, 1. Although all Arab countries were successful in closing the HDI gap, the results in Table 2 confirm our doubts regarding any phenomenal progress. Only Oman is ranked within the top ten HDI achievers using this methodology, followed by Morocco and Saudi Arabia as a distant second and third (ranked fourteen and nineteen, respectively). The evolution of the GDP per capita gap, however, is strikingly unimpressive by comparison with the HDI where a divergence from the average per capita of the richest twenty countries, rather than a convergence, was recorded for Arab countries. Thus, compared to other developing regions, Arab countries ranked third and fifth, in terms of their ability to close the HDI and GDP per capita gaps, respectively. As one would expect, China (and hence the East Asia & Pacific region) ranks first on both indicators according to this methodology. 5

9 Algeria Bahrain Comoros Djibouti Egypt Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Qatar Saudi Arabia Sudan Syria Tunisia UAE Yemen Human Poverty As noted above, the HPI concentrates on deprivation in three essential areas of human life already reflected in the HDI: longevity, knowledge and decent living. Three indices are chosen to represent human poverty in a composite index: (i) an index of deprivation relating to survival meaning the vulnerability to death at a relatively early age; and is measured by the percentage of people expected to die before age 4;(ii) an index of deprivation with respect to knowledge, meaning exclusion from the world of reading and communication; and is measured by the percentage of adults who are illiterate; and (iii) an index of deprivation from a decent standard of living (overall economic provisioning), which is composed of two sub-indicators for measurement purposes: This is measured by a composite of two variables: the percentage of people without sustainable access to a safe water source (with weight.5) and the percentage of underweight children under age five(with a weight of.5). The three indices of deprivation (i.e. (i) (iii)) are combined into a composite index allowing for possible substitution between them. Figure 2: HPI for Arab countries and developing regions, shows that the HPI for the Arab region in 7 is slightly above the average for developing regions and is more than double that of Latin America& Caribbean and East Asia& Pacific. This thus renders Arab human poverty rates closer to the vicinity of South Asia. Notwithstanding its relatively high level, the region achieved an impressive reduction in the HPI since 1997 (from 31 to 23%), at par with the average for developing regions (from 27 to %) but below that achieved by Latin America& Caribbean and East Asia & Pacific (from 13 to 9% and from 18 to 1%, respectively). Figure 2: HPI for Arab countries and developing regions, HPI % Change AC EAP LAC SAS SSA DR % Change 5% 4% 3% % 1% % Source: UNDP Human Development Reports Note: All Arab countries are included except Iraq, OPT and Somalia Figure 3: HPI for Arab sub-regions (A) and Arab countries (B), (A) (B) HPI % Change Source: ibid. Note: Arab countries included in figure 3.A are same as figure 2 19 Change 5% 11 Mashreq LDCs Maghreb GCC 4% 3% % 1% % HPI % Change Change 7% 6% 5% 4% 3% % 1% % 6

10 AC EAP LAC SAS SSA DR GCC LDCs Maghreb Mashreq Algeria Bahrain Comoros Djibouti Egypt Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Qatar Saudi Arabia Sudan Syria Tunisia UAE Yemen At the sub-regional level, Figure 3.A indicates that the HPI for the Arab LDCs was significantly above the average for the region in 7 and nearly two thirds above the HPI for the Mashreq and Maghreb. Furthermore, the LDCs achieved a considerably lower reduction in the HPI since 1997 (from 39 to 34%) compared with the reduction rates achieved by the Maghreb and Mashreq. It is also worthy to note that, as opposed to the case of consumption poverty, both the HPI levels and their trends are quite similar for the latter two sub-regions. The highest rate of reduction in the HPI was achieved by the GCC (43%). It is not surprising, therefore, that the three highest rates of reduction in human poverty were achieved by Kuwait, Qatar, Oman and UAE, while the lowest reductions were achieved by Sudan, Yemen, Morocco and Mauritania respectively. Among the more diversified economies of the Mashreq and Maghreb, Syria and Algeria were able to reduce human poverty at a stronger pace. It is worthy to note however that the HPI for Morocco is significantly higher than the average for middle income Arab countries, making it closer to the ranks of LDCs. The reasons for this will become clear in the following analysis of HPI components. We begin our decomposition of the HPI with the indicator for deprivation from leading a long life as measured by the percentage of population who are not expected to reach age 4. Error! Not a valid bookmark self-reference..a-c present this variable and its percentage change over the period from 1997 to 7 at the same levels of aggregation: global, regional, sub-regional and country. The figure clearly shows that Arab countries do not suffer from high levels of deprivation from survival to age 4 compared to other developing regions. Moreover, Arab countries scored an impressive reduction in this component of the HPI bringing the region closer in terms of levels and trends to East Asia& Pacific and Latin America& Caribbean than to South Asia or Sub-Saharan Africa. Figure 4: People not expected to reach age 4 in developing regions (A), Arab sub-regions (B) and Arab countries (C), (A) (B) (C) Change % Change % HPI Change % Change % 25% % 15% 1% 5% % 3 1 5% 4% 3% % 1% % % 5% 4% 3% % 1% % Source: ibid. Note: Arab countries included are same as figure 2 Arab countries are however more deprived from literacy as Figure 5 illustrates. The figure shows that illiteracy rates for Arab countries are significantly above the average for developing regions, making it almost at par with South Asia and Sub-Saharan Africa in 7. It is also interesting to note that the disparities between Arab sub-regions are less sharp with regards to this indicator, particularly the disparity between LDCs and the Mashreq and Maghreb, which implies that illiteracy is a cross-cutting problem for the vast majority of Arab populations. It is also a particularly concerning problem for Morocco, which retains the highest illiteracy rates among Arab countries, explaining its relatively high HPI. 7

11 AC EAP LAC SAS SSA DR GCC LDCs Maghreb Mashreq Algeria Bahrain Comoros Djibouti Egypt Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Qatar Saudi Sudan Syria Tunisia UAE Yemen AC EAP LAC SAS SSA DR GCC LDCs Maghreb Mashreq Algeria Bahrain Comoros Djibouti Egypt Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Qatar Saudi Arabia Sudan Syria Tunisia UAE Yemen Figure 5: Illiteracy rate in developing regions (A), Arab sub-regions (B) and Arab countries (C), (A) (B) (C) Change % Change % HPI Change % Change % 5% 4% 3% % 1% % 5% 4% 3% % 1% % 7% 6% 5% 4% 3% % 1% % % % Source: ibid. Note: Arab countries included are same as figure 2 The third dimension of the human poverty index is calculated as the simple average of the population not using an improved water source and underweight children under age five. Access to an improved water source is a major challenge for most Arab countries given the high level of aridity in the region. Nonetheless as Figure 6.A shows, the population without access to an improved water source stood at 13.8% in 7, lower than the average for developing regions and almost the same as in 1 (14%). Figure 6: Percentage of people without access to an improved water source in developing regions (A), Arab sub-regions (B) and Arab countries (C), 1-7 (A) (B) (C) 1 7 Change % 1 7 Change % HPI 1 7 Change % Change % 5% 4% 3% % 1% % -1% % 5% 4% 3% % 1% % -1% -% % 5% % -5% -1% -15% -% -25% -3% Source: ibid. Note: Arab countries included are same as figure 2 Despite this region-wide stagnation, sub-regional trends varied distinctively. Reductions in deprivation achieved by the Mashreq were thus equally subdued by increases in deprivation in the Maghreb and LDCs. At the country level, Algeria scored the highest rate of increase in deprivation (from 6 to 17%) followed by Saudi Arabia and Sudan whereas the highest rates of decline were achieved by Tunisia, Oman and Egypt. It should be emphasized that the two largest LDCs, Sudan and Yemen, recorded setbacks as the share of population without access to an improved water source increased by 72 and 23%, respectively. 8

12 AC EAP LAC SAS SSA DR GCC LDCs Maghreb Mashreq Algeria Bahrain Comoros Djibouti Egypt Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Qatar Saudi Arabia Sudan Syria Tunisia UAE Yemen The second component of this third dimension of the HPI is based on the ratio of underweight children under the age of five to the total population of children, which is summarized in Figure 7. The figure shows a lower level of deprivation for this component compared to the average for developing regions, significantly below South Asia and Sub- Saharan Africa and slightly above East Asia& Pacific and Latin America& Caribbean. On the other hand, Arab countries scored the lowest rate of reduction in deprivation over the period from 1997 to 7. As in the case of other human deprivations, Arab LDCs recorded a significantly higher percentage of underweight children under age five. Moreover, it is the only sub-region which witnessed an increase in deprivation (by %). However, this deprivation is curiously higher than expected for the GCCs, which had nearly twice the deprivation rates for the Mashreq and Maghreb in 7. This is likely due to factors related to nutritional habits, rather than food insecurity. Within the Mashreq and Maghreb, Algeria and Egypt succeeded in achieving significant reductions (by 7 and 6%, respectively). Figure 7: Underweight children under age five in developing regions (A), Arab sub-regions (B) and Arab countries (C) (Percent), (A) (B) (C) Change % Change % HPI Change % Change % 4% 3% % 1% % % 4% % % -% -4% % 6% 4% % % -% -4% -6% -8% Source: ibid Note: Arab countries included are same as figure 2 Multi-Dimensional Poverty The method used for constructing the HPI does not preclude including other aspects of human deprivation. Such an expanded measure of human poverty is found in the recent Multi-Dimensional Poverty Index (MPI) produced by the Oxford University Poverty and Human Development Initiative (OPHI) and UNDP in 1, which covers 14 countries (including 13 Arab countries). According to this methodology, a household is identified as multidimensional-poor if it is deprived in 3 of 1 indicators (also called dimensions) which, as in the HPI, are divided across three areas of deprivation (health, education and standard of living). The dimensions and their pertinent weights in the MPI are: health (child mortality and nutrition, each having 1/6 of the weight), education (years of schooling and child enrolment, each having 1/6 of the weight) and standard of living (electricity, drinking water, sanitation, flooring, cooking fuel and possession of assets, each having 1/18 of the weight).the MPI is then calculated as the product of two numbers: the headcount ratio or proportion of people who are multidimensional-poor and the average intensity of multidimensional-deprivation which reflects the proportion of dimensions in which households are deprived. Despite the large degree of subjectivity in determining the cut-off thresholds for each dimension (and in determining the minimum number of dimensions required to identify whether or not a household is multi-dimensionally poor), the MPI has an important advantage compared with the HPI as it captures many more dimensions of human poverty and includes both the level of human deprivation and a measure of the intensity of poverty. 9

13 DR SAS ECA EAP SSA LAC AC UAE Kazakhstan OPT South Africa Georgia Russian Fed. Tunisia Thailand Albania Uzbekistan Serbia Bosnia and Herz. Egypt Ecuador Sri Lanka Hungary Ukraine Belize Moldova Jordan Syria Azerbaijan Uruguay Montenegro Armenia Macedonia Kyrgyzstan Mexico Morocco Iraq Argentina Trin.&Tobago Croatia Tajikistan Myanmar Suriname Using survey data from 14 countries with a combined population of 5.2 billion (78 per cent of the world population), the OPHI estimated that 1.7 billion people -one third of the world population- live in multidimensional poverty. Of this population living in multidimensional poverty, only 2.5% belong to Arab States (41.2 million people), approximately half live in South Asia (844 million people) and nearly one third are in Africa (458 million). In Figure 8.A, it is clear that the MPI is below the HPI for Europe & Central Asia, Arab countries, Latin America& Caribbean and East Asia& Pacific. The ratio of the HPI to the MPI is second highest for Europe & Central Asia and Arab countries, which implies a significantly lower MPI compared to the HPI. The reason for this is made clear in the country-level data of Figure 8.B which shows developing countries with highest ratios. Of the highest thirty six ratios, eight belonged to the Arab region and six of them among highest twenty two ratios, including the more populous Egypt and Syria. Figure 8: Ratio of HPI to MPI for developing regions (A) and thirty six highest ratios (B), 7 (A) (B) HPI/ HPI/MPI HPI MPI HPI & MPI MPI 6% 4% % 4% 3% % 1% % % Source: UNDP Human Development Reports and OPHI Note: For HPI, Arab countries included are same as figure 2. For MPI, Arab countries included are Comoros, Djibouti, Egypt, Iraq, Jordan, Mauritania, Morocco, OPT, Somalia, Syria, Tunisia and Yemen Figure 9 shows the poverty headcount and intensity components of the MPI for developing regions and Arab sub-regions. The average intensity of deprivation for developing regions in the sample is 53.1%. Sub-Saharan Africa, has the highest intensity (58.2%) occurs while the lowest (42.1%) intensity occurs in Europe & Central Asia. The Arab Region s average intensity of 49.3% is below world average but both Latin America & Caribbean and East Asia& Pacific have significantly lower intensities (46.2 and 46.5%, respectively). As observed from the figures, whether for developing regions or Arab sub-regions, the intensity of deprivation is clearly the less significant source for the variation in the MPI. Figure 9: MPI components for developing regions (A) and Arab sub-regions (B), -8 (A) (B) Headcount Intensity 7% 6% 5% 4% 3% % 1% % DR SAS EAP SSA LAC ECA AC 7% 6% 5% 4% 3% % 1% % Source: Authors estimates for Arab countries and sub-regions and OPHI, 1 Note: Arab countries included for LDCs are Comoros, Djibouti, Mauritania, Somalia and Yemen. For Maghreb: Morocco and Tunisia. For Mashreq: Egypt, Iraq, Jordan, OPT and Syria 7% 6% 5% 4% 3% % 1% % Headcount Intensity Linear (Intensity) 7% LDCs Maghreb Mashreq 6% 5% 4% 3% % 1% % 1

14 Table 3sheds further light on the data sources and MPI headcount and intensity components for Arab countries. It is important to note that the ratio of the HPI to MPI for both sub-regions improves significantly vis a vis that for developing regions (6.5 for the Mashreq and 7.3 for the Maghreb versus 2.9 for developing regions). Conversely the ratio for Arab LDCs, which takes a value of one, suggests a significant deterioration in their relative position versus developing countries in general and Arab sub-regions in particular. Thus, with the glaring exception of LDCs, the region is less poor from the multidimensional poverty perspective. Table 3: MPI components and HPI to MPI ratio for Arab countries and sub-regions, -8 Multidimensional Poverty HPI 9 Country Survey Year HPI/MPI MPI Value MPI Rank H A Value Comoros MICS Djibouti MICS Mauritania MICS Somalia MICS N/A N/A Yemen MICS Morocco LSMS Tunisia WHS Egypt DHS Iraq MICS Jordan DHS OPT MICS Syria MICS Mashreq Maghreb LDCs Arab countries East Asia & Pacific Europe & Central Asia Latin America & Caribbean South Asia Sub-Saharan Africa Developing region Source: Authors estimates based on data from Alkire& Santos, June 1 and LSMS for Morocco. The percentage contribution of each dimension to MD headcount poverty is important to discover the areas of deprivation. With the exception of South Asia and Sub-Saharan Africa regions, where standard of living contributes more to deprivation, lack of education is the highest contributor to global headcount poverty (Figure 1.A). Arab countries also follow this global pattern as the percentage contribution of education to overall MPI in the Arab Region is 45% followed by Health, 34% (in contrast to the rest of the world where living standards is the second largest contributor to MD poverty). Figure 1: Contribution of living standard, health and education to MD headcount deprivation for developing regions (A) and Arab sub-regions (B), -8 (A) (B) Mashreq Maghreb LDC DR AC SSA SAS LAC EAP ECA Living Standard 17 Health Source: Alkire& Santos, June 1. Note: Arab countries included are same as Figure Education Living Standard Health Education

15 It would seem therefore that the character of multidimensional poverty in Arab states (and particularly the Maghreb) is closer to that in more developed regions of East Asia& Pacific and Latin America& Caribbean while for LDCs there is higher similarity with the poorer regions of Sub-Saharan Africa and South Asia in having higher deprivation in living standards. Indeed, as shown in Table 4: Components of the headcount deprivation component of the MPI for Arab countries (Percent), -8, the LDCs are far more much deprived in clean drinking water, improved sanitation, clean cooking fuel, electricity, flooring material and assets ownership compared to other Arab countries. Table 4: Components of the headcount deprivation component of the MPI for Arab countries (Percent), -8 Country Education Health Living Standard Years of Sch. Child Enrol. Under 5 Mort. Nutrition Electricity Sanitation Drinking Water Floor Cooking Fuel Asset Owned UAE OPT Jordan Tunisia Syria Egypt Iraq Djibouti Morocco Yemen Somalia Mauritania Comoros Source: Alkire& Santos, June 1 and LSMS (7) for Morocco. These results yield a mixed bag of conclusions. Arab countries as a whole enjoy low multidimensional poverty compared to developing regions. However, the gap between Arab LDCs and the Mashreq and Maghreb countries is conspicuously higher than expected based on the more conventional HPI. Thus and despite having less than one quarter of the population, Arab LDCs contain half of the Arab States MD poor people with Somalia being the poorest country in that group (Figure 11.A). In fact, Somalia s MPI ranked 99 out of the 13 countries in the sample. Figure 11: Distribution of the MP Population (millions, percent) in Arab sub-regions (A) MD and poverty headcount and intensity in Arab countries (B), 7 (A) (B) LDCs Maghrib Mashreq.7 Intensity R² =.8932 Somalia 9.177, 22% 1.662, 26% , 52% Arab countries Egypt OPT Iraq Syria Jordan Tunisia UAE Moroco Djibouti Mauritania Comoros Yemen Headcount Source: Authors estimates based on data from Alkire& Santos, June 1. Note: Arab countries included are same as figure 9 12

16 As shown in Figure 11.B, the MPI headcounts are strongly related to average intensity of MD poverty across the Arab countries. The intensity of poverty is therefore commensurate with the headcount poverty leading Mashreq and Maghreb countries to a favourable position with respect to both components as revealed by their location in the lower left quadrant of Figure11.B Conversely, LDCs are situated at a clear distance in the upper right quadrant of the figure. The exceptions however is Djibouti which are located in the lower right quadrant indicating a lower than average intensity and higher than average headcount poverty ratio compared to the average for Arab countries. Human Deprivation and Income: Summary of Stylized Facts In order to give more informed human poverty stylized facts, the progress (or lack thereof) of poverty indicies should be set against an appropriate measure of the capacity or means available to the state (and its individuals) to reduce such deprivations. Although it is by far not the sole determinant of human poverty, we have chosen per capita GDP (in 5 PPP) as the appropriate proxy for this capacity. Accordingly, Figure 12.A and B plot the HPI and MPI against per capita GDP, respectively, for the relevant years. The high correlation coefficients displayed on the top left part of the scatter diagrams is expected due to the well established strong cross-country relation between human poverty and income pe capita. Figure 12: GDP per capita (PPP) and HPI (A) and MPI (B), 5 (A) HPI MPI R² = Log GDP per capita Source: Authors estimates based on data from World Bank for GDP per capita, Human Development Reports for the HPI and OPHI for the MPI Note: Arab countries included are same as figure 8 except Iraq, OPT and Somalia for both figures 12.A and B Using the downward-sloping regression line, we can estimate the expected value of the human poverty index for any given GDP per capita. It should be noted that the regression line is non linear, which implies that the nature of the relationship between both variables changes as GDP per capita increases. This result is not difficult to understand as it would be far more difficult for an incremental rise in GDP per capita to reduce human poverty by the same rate in a country with a very low HPI or MPI than in a country with a very high HPI or MPI. The expected or estimated values (HPI* and MPI*) are then compared with the actual HPI and MPI values. If the ratio of the latter to the former is higher than one, this suggests the country or region has a higher than average level of human poverty and vice versa. It follows that the change in the value of this ratio over time represents the change in position of the country or region over time relative to its estimated position on the regression line. A summary of this data is shown in Table (B) R² = Log GDP per capita 13

17 Table 5: Actual and estimated HPI for Developing regions and Arab countries, Country/Region HPI HPI* HPI* HPI/HPI* HPI/HPI* Change in HPI (7) (1997) (1997) (7) (1997) (7) HPI/HPI* Egypt Jordan Syria Lebanon Mashreq Libya Algeria Morocco Tunisia Maghreb Mauritania Sudan Yemen Comoros Djibouti LDCs Bahrain Oman Qatar Saudi Arabia Kuwait United Arab Emirates GCC Arab countries East Asia & Pacific Latin America & Caribbean South Asia Sub-Saharan Africa Developing region Source: Authors estimates based on data from Human Development Reports and OPHI Three important stylized facts can be deduced from the table. First, the observed HPI for Arab countries is significantly higher than their expected HPI (HPI*) compared to any other developing region. The oppostie is true for East Asia& Pacific and Latin America & Caribbean (their ratios were.6 and.7 in 7, respectively). Second, within the Arab region, the actual to expected HPI is signifantly larger for Maghreb countries. Third, the Arab region did not witnessed any change in the gap between actual and expected HPI over the period from 1997 to 7 as indicated by the zero percentage change in the ratio of actual to expected HPI shown in the last column. Fourth, the stagnation of the HPI to HPI* ratio was not uniform within the Arab region. The ratio declined significantly for GCCs which suggests that these countries were able to match the rapid growth in GDP per capita with a more rapid reduction in the HPI. The ratio also decline in Maghreb and Mashreq countries. But the former still sustained a large gap between their actual and expected HPI in 7. The Arab LDC group, however, was the only Arab sub-region to witness an increase in the gap between actual and expected HPI indicating a retrenchment in human poverty reduction efforts. This retrenchment is even more alarming when we factor in the rapid growth in GDP per capita (in 5 PPP) for the sub-region which took place over the period from 1997 to 7 (from approximately $13 to $21). The main conclusion to drive from the cross-country comparisons between GDP per capita and the HPI is that Arab countries have a significantly higher rate of human poverty than what would be expected given their level of income. However, cross-country estimates based on the MPI yield a different conclusion. As shown in Figure 13.A, the actual and expected MPI for the Arab region are identical giving it a MPI/MPI* ratio of one. This ratio places the Arab region in a far more advantageous position compared with the HPI and 14

18 AC EAP LAC SAS SSA DR Egypt Jordan Syria Mashreq Morocco Tunisia Maghreb Comoros Djibouti Mauritania Yemen LDCs AC EAP LAC SAS SSA DR Egypt Jordan Syria Mashreq Morocco Tunisia Maghreb Comoros Djibouti Mauritania Yemen LDCs second to East Asia& Pacific. The improved rank of Arab countries is also due to the sharp slide in the ratios for South Asia and Latin America& Caribbean in particular as confirmed in Figure 13.B. The figure also shows the East Asia & Pacific and Sub-Saharan Africa regions are the only regions where MPI and HPI ratios were consistent. Figure 13: Actual and expected MPI (A) and the ratio of actual to expected HPI and MPI (B), 7 (A) (B) MPI MPI* MPI/MPI* HPI/HPI* Source: ibid. Note: Arab countries included for HPI and MPI are same as figure 8 The discrepancy between the HPI and MPI ratios is not the same across Arab sub-regions. All Arab countries with the exception of Comoros, Mauritania and Yemen had lower HPI ratios. With the striking excpetion of Egypt and Tunisia, the values of HPI and MPI ratios were of commensurate order of magnitude. In fact, given its demoghraphic weight, it can be argued that the bulk of the region-wide discrepancy between HPI and MPI ratios can be mainly attributed to Egypt. The results for Yemen and Mauritania on the other hand are more consistent with the stylized facts observed for LDCs, namely a higher than expected actual to expected ratio for both the HPI and MPI. Facets of Inequality in Human Development and Deprivation Human development Unlike the HDI which only measures human achievements in health, education and income - regardless of how they are distributed, inequality-adjusted HDI captures the inequality that HDI does not measure in each dimension, thus one can think of the difference between them as the loss in potential human development due to the unaccounted inequality. The IHDI will be equal to the HDI when there is no inequality across people but falls further below the HDI as inequality rises, in other words, the lower the ratio of IHDI to HDI, the higher inequality in human development. 6 As shown in Figure 14.A,Arab countries alongside Sub-Saharan Africa and South Asia have the lowest ratios of IHDI to HDI (72, 67 and 7% respectively), which means they lose the most from inequality in human development, while Europe & Central Asia has the lowest loss due to this inequality. The implementation of the IHDI displayed that losses in health and education were greater than losses in income globally, where as in the Arab countries, inequalities in education contribute to the loss in human development more than inequalities in health and income. 15

19 Djibouti Comoros Mauritania Yemen Morocco Syria Egypt Jordan Tunisia It also seems that countries that have less human development are more likely to have higher inequality and thus larger losses in human development. As shown in Figure 14.B, LDCs (Comoros, Djibouti, Mauritania and Yemen) have the highest inequalities in human development, while the lowest inequality is found in Jordan. Although Egypt is higher than the Arab countries in human development, it loses more in human development due to inequality. Figure 14: Inequality adjusted HDI for developing regions (A) and Arab countries (B), 1 (A) (B) IHDI & HDI 72% Human Development Index (HDI) Inequality-adjusted HDIa IHDI/H IHDI/HDI DI 78% 86% 1% 75% 7% 67% 8% AC EAP EAC LAC SAS SSA 6% 4% % % IHDI & HDI % 56% Human Development Index (HDI) Inequality-adjusted HDIa IHDI/HDI 65% 66% 72% 79% 72% 81% IHDI/ HDI 1% 75% 8% 6% 4% % % Source: Authors estimates based on data from Human Development Report, 1. Note: All Arab countries are included except Iraq, Lebanon, Oman,OPT and Somalia Multi-dimensional poverty Inequality in human deprivation has generally received far less attention than inequality in the distribution of expenditure. This is perhaps due to the causal link between money-metric poverty and distribution of expenditure which makes it necessary to report the latter in order to explain overall money-metric poverty dynamics. In this section we make amends by reporting multi-dimensional poverty at the sub-national level from two sources. The first source is based on data reported by OPHI which yields rural versus urban inequalities. The second source is household expenditure surveys for the following Arab countries: Egypt (8/9), Lebanon (5) and Syria (7) from the Mashreq; Morocco (7) from Maghreb; and Yemen (5/6) from LDCs. Collectively, these countries comprise of more than half of the Arab population and an even higher share of the Arab poor population. Figure 15: Rural and Urban MPI (A) and the Rural to Urban MPI ratio versus GDP per capita (B), -7 (A) (B) Rural MPI SOM.5 MRT SSA DJI LDC COM.4 SAS.3 YEM MAR Maghreb DR.2 LAC AC.1 IRQ EGY Mashreq EAP SYR Urban MPI TUN ECA JOR Source: Authors estimates based on data from Human Development Reports and OPHI MAR Maghreb From Figure15.A, we can clearly see a close relationship between urban and rural MPI. In other words, countries which have a low degree of human deprivation in rural areas are also likely to have a low degree of deprivation in urban areas and vice versa. However, as Figure Rural/Urban MPI LAC 4 AC TUN SAS DJI EGY IRQ EAP DR SSA MRT LDC YEM Mashreq ECA 2 SYR COM JOR Log GDP per capita