Demographic Turbulence in the Arab World: Implications for Development Policy

Abstract

The Arab Spring's hallmarks of volatility and ‘youthfulness’ thwart sustained development, political cohesion and peace. Demographic ‘youthfulness’, due to recent high fertility, is accompanied by rapid growth. But the backstory is even bleaker. Cohort flows, producing large population waves, batter successive life-cycle stages, generating demands for services, markets, resources and capital. To overcome this, policies directed at youth (15–24) are pivotal; failure to integrate youth into the economy and society results in disadvantaged middle-aged parenting populations. This threatens longer-term, sustainable development and triggers political volatility. Yet, as is shown in this article, the international development community has failed to formulate appropriate strategies.

Demographic Turbulence, Political Change and Development

The recent uprisings and other disturbances across the Arab countries have drawn attention to the importance of demographic factors in development as well as politics. During the ‘Arab Spring’ the media overwhelmingly stressed the youth of Arab countries. A heading in the The Guardian Weekly is typical of many at that time: ‘To be Alive, Young and Arab’ (Porter 2011). This paper aims to show that in terms of factors of population change, the Arab countries are notably turbulent by comparison both with the world as a whole and with non-Arab Muslim countries. Demographic turbulence is due to significant fluctuations in birth cohort sizes.

This article argues that demographic turbulence is a key determinant of systemic under-development and political instability. The term ‘demographic turbulence’ is used in the sense employed in specialist analyses of population changes that are rapid and that have impacts on societal structures and dynamics. They include: age-compositional changes, particularly when accompanied by fluctuations in birth cohort sizes; large, sudden migratory flows; shifts in family structures; and radical trends in fertility or mortality (Pool 2005, 2006a, 2006b). Most social and demographic behaviours are age-specific, and this translates across to economic factors: patterns of production, consumption, supply and demand. Even the fiscal sectors have age-specific drivers, influencing which dependants need tax-based transfers and who pays taxes.

The focus of the article is age-structural changes that successively produce youth-age bulges, proportionately by comparison with other age groups, and then generate deficits as cohort sizes decrease. Patterns of bulges followed by deficits are then seen at successive older ages. For many populations, including Arab nations discussed below, cohort sizes may fluctuate significantly; when severe, they are called ‘disordered flows’ (Waring 1975). Historically, fluctuations came from shifts in mortality, but today, at least nationally, this will typically be due to changing fertility levels resulting in varying birth cohort numbers. Geographic mobility, particularly rural-urban migration, has similar effects both at source regions and destinations; flows of young workers to the Chinese eastern seaboard is an example. Small nation-states subject to massive migratory movements, such as in the Gulf states, may suffer similar impacts.

The passage of cohorts of different sizes across key life-cycle stages has implications for all development policy and planning, and for markets. It produces sudden demands for services and market goods directed to those age groups that are inflated in size by cohort flows and is followed by downturns in demand once they deflate. The passage of disordered cohort flows makes planning even more difficult. An example is the severe pressure on the basic education sector when fertility rates are high; as birth rates drop, pressures are felt successively in middle school, high school and the tertiary sector. But as fertility drops, an inverse scenario unfolds.

Particularly significant for development policies are pressures from servicing post-basic education, skills training and expanding labour markets as large cohorts reach youth ages of 15 to 24 years (sometimes up to 29 years). Youth ages are pivotal; failure to integrate youth – particularly those who belong to cohorts that have inflated sizes – into the entire development process has long-term consequences. As inflated cohorts move into middle age, new demands arise, such as housing and all the other needs of family building. If cohorts are skilled and there has been active employment generation, these needs can be met, but, if not, under-development will ensue. A new economic-demographic paradigm, the ‘demographic dividends’ that are driven by these strategies, is examined in a later section.

Youth ages are pivotal; failure to integrate youth – particularly those who belong to cohorts that have inflated sizes – into the entire development process has long-term consequences.

Cohort flows directly affect development policies, processes and outcomes, having implications for conflict and peace, particularly in divided societies. Population data provide widely available macro-level indicators, now and into the foreseeable future, of trends that will directly affect people-centred development. Yet generally, population factors have not been satisfactorily endogenised into international development initiatives; most egregiously, the international community and nation-states have not addressed so-called ‘youth bulges’ that drive demographic and political turbulence, thus placing severe constraints on development.

The Arab Spring and ‘Youth Bulges’

Most analysts implicitly recognised the significance of ‘youth bulges’ as a cause of the recent ‘Arab Spring’. The political volatility of the youth age group – aged 15 to 24 (or 29)– was particularly invoked, along with high levels of unemployment. This focus has often failed, however, to distinguish youth from the child population, an age group making decreasing contributions to population change. Moreover, as appealing as such diagnoses are, they ignore other complex demographic changes that will have long-term negative impacts on development if not resolved, yet are difficult to address.

Youth bulges are not a new issue in population and development circles, where age-structural factors have been talked about for decades (Coale & Hoover 1958; Keyfitz 1968). In the ‘development decades’ of the 1960s, '70s and '80s, high fertility and thus rapid growth were the issues on which population and development policies focused. But in the 1990s, it was recognised that world population trends had gone through a quantum shift. The emerging issues were twofold: much slower growth but complex structural changes, notably population geography and age composition.

The World Bank Development Report 2007 focused on youth, stressing an issue elaborated in this article for the Arab countries and elsewhere: ‘Such large numbers of young people [12–24 years] living in developing countries present great opportunities, but also risks’.1 Far earlier, the CIA had identified the political implications of demography, particularly youth ‘surpluses’ and ‘deficits’ (CIA 1990).

The media, responding to alarmist reports, emphasise growth, particularly its impact on natural resources and food and water security, and concern centres on simplistic diktats prescribing interventions to reduce growth. But a higher profile for age-structural factors, notably the ‘youthfulness’ of Arab populations, is a new element in the popular media, vying with ‘Islamism’, ‘Jihadism’ and other more commonly asserted explanations of political upheavals. Less acknowledged, however, are the underlying population dynamics and structures of Arab countries that drive youth bulges and their sequelae. Yet these are the most important drivers, or inhibitors, of social and economic development in the medium to longer term.

This article notes first how youth bulges come about as a concomitant of normal demographic transitions, generating birth cohorts of changing sizes (Chesnais 1990). The resulting waves can be regular, but are often turbulent, varying in both magnitude and distances between oscillations. Key life-cycle phases will often be affected by the passage of ‘disordered cohort flows’. These can occur because of policy interventions (e.g. rapid improvements in childhood survival) or other factors (e.g. famines and wars), with major consequences for development.

Secondly, the age-structural changes in the Arab countries as a whole and in four case studies are compared with the ‘World’, and with four countries with large, non-Arab Muslim populations. The Arab countries and territories are shown to be different, an important comparison because some politicians and media view the planet as bifurcated geopolitically into Muslim and non-Muslim; rather, the trends examined here seem to be associated more with developments in the Arab region than with religion. Nevertheless, there are also differences between Arab countries, so demographically at least, they are not homogeneous.

The article looks at the response of the international development community to youth bulges. Finally, it reviews the future of Arab countries and other high-growth populations undergoing rapid and/or turbulent age-structural changes.

Drivers of Demographic Turbulence

According to United Nations 2010 estimates, there are 357 million Arabs in West Asia, North Africa and sub-Saharan African countries.2 In the Appendix Table, key demographic indicators are presented for each territory and for selected non-Arab Islamic comparisons. Eighty percent of Arabs live in only 40% of the 24 countries and territories: Jordan, Libya and Somalia have populations of between five and 10 million. The greatest number of Arab countries, a further 11, have only 28 million between them. The most populous small countries and territories, Lebanon and the Occupied Palestine Territory (OPT), each have four million inhabitants, whereas Bahrain, which has recently had intensive coverage in the media, has merely one million. This article focuses on the Maghreb, Egypt and West Asian Arab countries, touching only in passing on the tropical African outliers, the Comoros and Mauritania.

The key drivers of Arab demographic turbulence are the fertility differentials today and trends over the last few decades. Past trends manifest themselves most forcefully in marked shifts in age and cohort structures, including disordered cohort flows producing the transitory youth bulges this article focuses on. These have immediate multiplier effects – demographic (such as rapid urbanisation and the sectoral redistribution of the labour force), socio-political and economic – that create further destabilisation. But then the same inflated cohorts survive into older life-cycle phases, having longer-term effects on policies for middle-aged and older age groups.

Arab fertility rates are now very uneven, as the Appendix Table shows. In 2005–10 they ranged from a total fertility rate (TFR) of 6.4 births per woman (Somalia), 5.9 (Yemen) and 5.1 (OPT), to below replacement (2.1) in Tunisia, United Arab Emirates (UAE) and Lebanon. Fertility declines have occurred in every Arab country, but these changes have varied in timing and magnitude, from very limited to radical. Today, a factor of even greater significance than differing rates as an underlying determinant of demographic volatility has been variability in the velocity and timing of declines.

In the second half of the 20^th century, almost all Arab countries at some stage had high fertility (TFRs > 6.0 births per woman). The sole exception was Lebanon, but its rate was still relatively high, at 5.7, in 1950–55. It has gone through a gradual decline since, but has a TFR trajectory that resembles Indonesia's more than those of its Arab neighbours. Very high rates were still recorded in the 1980s in half the Arab countries and territories, including larger ones – Algeria, Iraq, Saudi Arabia and Sudan – and potentially volatile countries and territories – Yemen, Mauritania, Somalia (2005–10) and the OPT.

In the second half of the 20^th century, almost all Arab countries at some stage had high fertility.

In one or more quinquennia since 1950, 14 of the Arab countries and territories had even higher rates (TFRs > 7 births per woman). Since 1980, such bio-socially ‘hyper-fertility’3 occurred for Saudi Arabia, Libya, Palestine, Somalia and the Comoros, and in the 1990s for Yemen. As recently as the late 1970s, Algeria, Libya, Saudi Arabia and Syria had TFRs above 7.0, yet today their rates have plummeted to 2.4, 2.7, 3.2 and 3.3 respectively. Hyper-fertility levels noted here for Arab states in recent decades are uncommon, historically or post-war. They have been reported in the last few decades mainly for some African countries (e.g. Kenya, Rwanda and Niger), but not for other large Muslim nations (e.g. Iran, Indonesia, Pakistan and Bangladesh).

A further consequence of high fertility is rapid growth, which is still the case in some countries (see Appendix Table). Growth is driven by natural increase – births minus deaths – not migration. But even states with low fertility still suffer momentum effects that drive growth. Large cohorts, born in the past when fertility was higher, pass through the age pyramid, inflating numbers successively at each life-cycle stage. This has a downstream impact on birth numbers. When women in these same large cohorts become parents, their TFRs may be lower, but their birth numbers are higher as there will be more parents. This is illustrated, for example, by Libya. Taking 1950 as a base, its young child population (ages 0–4 years) had grown fourfold by 1985 – the births of which year became the youth of the early 2000s. Following rapid fertility declines, birth numbers decreased to threefold by 1995 compared with 1950, but through momentum will increase to fourfold by 2015.

Thus, even following fertility rate declines, large and fluctuating numbers of children can be born. Changes in cohort numbers because of fertility declines and momentum effects have development implications for all sectors. While researchers focus on rates, for planners the key metric is the numbers to be serviced, who will generate fiscal and familial burdens, and whose taxes and workforce participation will affect fiscal and familial capacities.

While national population dynamics in the Middle East are mainly due to natural increase, added demographic turbulence – changes in population sizes and in age structures – comes from workforce migrations (see the notes with Appendix Table). Within the Arab countries, these migrations are towards oil-rich states such as Libya, Saudi Arabia and the Gulf states. But trends are compounded by inflows of workers from outside the region, plus outflows from Arab nations towards Europe and North America. Labour force and other migratory flows are almost always age-specific.

Migratory flows produce sudden shifts in numbers, having high-profile political implications, but these are often of short duration. Proportionately, the greatest impact of migration has been on the small nations of the Gulf with large foreign worker populations, notably Kuwait, which had fluctuating flows at the time of the Gulf War, Qatar and the UAE – and, conversely, on migration source regions in Asia. This was seen when the economic recession took effect.

It is clear from the foregoing that Arab population dynamics are mainly due to natural increase. This creates far more severe, albeit latent, home-grown instability, as shifts in birth numbers produce cohorts of differing sizes. In turn, this creates significant population and development implications, particularly for some larger Arab states, whose weight of numbers are important globally. Whether from natural increase or migration, the passage of cohorts of fluctuating size across major life-cycle stages, especially if these flows are disordered, affects all aspects of development – cultural, social and economic. Downstream this may generate political turbulence.

If youth have been successfully integrated into the society and economy, a country will benefit from these flows, but if this has been unsuccessful, as is usually the case, under-development may be reinforced or at least maintained.

Long-term effects come from cohort flows through youth ages into middle and older age groups, as illustrated in the graphs below. If youth have been successfully integrated into the society and economy, a country will benefit from these flows, but if this has been unsuccessful, as is usually the case, under-development may be reinforced or at least maintained (Pool et al 2006).

Unfortunately, the international community has failed to recognise the short-term pivotal role of integrating youth in longer-term development. That said, an emerging paradigm, ‘demographic dividends’, points to the critical importance of such strategies of integration. Empirical research shows that, when managed well, demographic dividends have long-term benefits that flow on even when a population subsequently goes through demographic ageing.

Youth Bulges

The following graphs show how flows of cohorts of differing sizes affect various age groups over a century, from 1950–2050. The methodology is simple: it graphs the percentage of total population change in any quinquennium due to shifts at younger life-cycle stages. The size of the population in the base year is taken as the denominator.4 Later life-cycle stages are excluded as results are less reliable, due to the dampening effects of the projection assumptions. The index is standardised to life-cycle stages of 15 years: childhood, 0–14; youth, 15–29; young middle ages, 30–34.

The cohort flows in the Arab countries are more disordered than for the ‘World’, as is seen in Figure 1. The contribution of child and youth ages to overall demographic change has been double that for the whole world. More recently, large, fertility-driven decreases at childhood give way to momentum effects at youth ages, and these eventually give way to major impacts at early middle-age life-cycle stages.

Figure 1 Impact of Cohort Flows on Life-cycle Stages, World and Arab Region, 1950 to 2045–2050

Figure 2 shows the timing and magnitude of flows vary between Arab countries. For example, Lebanon's age-structural transition occurred much earlier. At two life-cycle stages, contributions to change became negative; children are making a negative contribution at present and this pattern will flow into older ages. The Libyan shifts to negative contributions across three life-cycle phases are more dramatic.

Figure 2 Impact of Cohort Flows on Life-cycle Stages in Lebanon, Egypt, Libya and Yemen, 1950 to 2045–2050

The volatility in these case-study Arab countries has development implications: policies meeting peak demands must be rolled back and switched to where new demands are emerging, often over short periods, as for Yemen and Libya, yet they may need reformulating later to meet new demands.

This was Lebanon's 1990s situation: peak demands came at young middle ages, when family building, housing, children's education and employment are critical. Lebanon illustrates another important point: that policy makers may simultaneously face rising competing demands from several different life-cycle phases. Moreover, simultaneously its inflated parenting cohorts bore large birth cohorts even though fertility had declined.

Yemen saw radical changes at 0–14 years, but not the disordered flows seen in Libya. Overall, changes are more marked in Libya and Yemen than in the other two Arab countries selected here. For example, they sustained turbulence at youth ages – rapidly increasing or decreasing shares of all changes – around 2005–10. Demographically their flows are starting to dampen down. In contrast, Egypt's fertility decline was more gradual and its path less volatile. For Egypt and Lebanon, demographic turbulence had occurred several quinquennia ago.

Figure 3 presents comparative data on non-Arab Muslim countries. With the notable exception of Iran, they are less turbulent demographically than the Arab examples above. Nevertheless, for Pakistan turbulence at youth ages also has been evident during recent quinquennia. For Iran – a polity sustaining on-again/off-again fertility policies and edicts – the youth bulge was very marked over recent quinquennia. In Iran and Tunisia, spectacular decreases in the contributions of changes at childhood to overall trends are attributable to rapid fertility declines coming from shifts in delayed marriages. A factor is the increasing numbers of young women preferring work to marriage and thereby accepting celibacy and childlessness over childbearing and remaining outside the workforce.

Figure 3 Impact of Cohort Flows on Life-cycle Stages in Bangladesh, Indonesia, Iran and Pakistan, 1950 to 2045–2050

Unfortunately, for policy makers the deflating of youth ages does not signal the end of development problems. The waves that impact on youth life-cycle stages then continue as inflated flows into young middle ages, demanding policy responses. Eventually they continue into older ages, affecting age structures and policy needs for geriatrics, until the cohort's last member dies. But this article focuses on the medium term, not the longer term.

These patterns and trends have major implications for economic, social and political development and thus international affairs. Unless youth are fully integrated into societies and attain meaningful employment to become productive contributors to the economy, they will impose both fiscal and familial burdens. This becomes critical when they assume family responsibilities and look for continuing employment at young middle ages. Swollen unemployed population numbers also have significant implications politically; they form seedbeds for the spread of radical ideologies, political mobilisation, social upheavals and urban violence.

Unless youth are fully integrated into societies and attain meaningful employment to become productive contributors to the economy, they will impose both fiscal and familial burdens.

All countries shown here have youth bulges, albeit more muted for Lebanon and almost over for Egypt and Indonesia where development focuses must shift to the integration into the economy of populations at young middle ages. For several countries deflation will be short-lived as further momentum-driven youth bulges reappear in the 2020s. This is marked for Iran, where policy changes have had impacts, but also for Libya and Yemen, and more modest for Egypt and Bangladesh. But the immediate challenges for development strategies come from bulges at middle ages. The critical underlying factor for development strategies for the next decade and beyond becomes the successful integration into the society and economy of the disordered flows identified here at youth ages. Education, training and labour market entry are paramount policy concerns if they are to be productive in middle age.

Planners and policy-makers in these countries, especially those with disordered and/or inflated cohort waves, face major challenges in a number of key social and economic sectors. For example, education is particularly affected by age-structural changes. However, as Wolfgang Lutz notes:

Much of the … literature that tried to empirically assess the returns to education at the societal level in terms of higher economic growth, better health and better governance and democracy has been seriously hampered … because the authors did not analyse age-specific data on educational attainment but typically studied the mean years of schooling of the entire adult population (15+ or 25+) as one presumably homogeneous group (2010: 14).

When large cohorts are at childhood, problems revolve around meeting objectives, as in the Millennium Development Goals (MDGs) prescribing universal school participation. As Alaka Basu notes, ‘mass universal schooling … has become the mantra in development planning during the last quarter century or so’ (2010: 25). This goal has been criticised because of the demographic and financial difficulties of realising it in order to enhance human capital development, rather than teaching in poorly equipped, overcrowded classrooms. There is also concern that mass education may not deepen human capital. As these cohorts then reach secondary and tertiary educational ages, there are even greater challenges, such as generating meaningful employment for high school and tertiary graduates. A majority of jobs even in technical sectors may be less skilled: ‘jobs in the lowest quartile of earnings will account for about 40% of growth in the top 30 occupations’ (Graff 2010: 19). Does national-level planning in the international development paradigms proposed by global civil society and by development agencies address such issues?

International Development Community Responses

Almost universally, development experts see the integration of youth as an important objective. Yet beyond virtual platitudes and hand wringing, little that is of concrete value has been posited or implemented. Prevailing attitudes seem to be seen – for instance, youth on television during riots – but not heard. Certainly, the real needs of youth are not seriously addressed in major development paradigms.

A first problem is generic: to see integration strategies (for any sub-population, not just youth) as relevant only for the social sectors, the economy's demand-side. Youth are thus seen as ‘problems’ for social welfare policy, but are not identified as central concerns for the productive sectors of the economy. In part this is because classical economics tends to restrict labour force analyses to ‘entries and exits’ to the employed population, rather than looking at the supply side: cohort flows across the workforce ages. This grossly downplays the role of the young, the producers of the immediate future and its taxpayers, as well as providers of services for children and older populations.5 The integration of youth and their subsequent contributions to the society and economy at family-building ages are thus pivotal for every sector, even the fiscal and financial. Moreover, the flows through the various life-cycle stages that fall into the working years are critical to this. Entries and exits set external parameters of the workforce, but its internal architecture determines the active workforce's productive functioning. This is due to cohort flows: their relative sizes; sectoral composition by cohort; cohort differentials in training and education; and thus job and geographic mobility. To respond to all these requires policies that address ‘widening’ as well as ‘deepening’ of human capital.6

Secondly, and again generic, is that recently the financial and fiscal sectors have played disproportionately important roles in setting overall development agendas. Typically, under pressure from agencies of the ‘Washington Consensus’,7 they have focused on spending cuts and debt repayment rather than formulating real development strategies, implemented in the structural adjustment programmes enforced on poorer Third World countries. The net effect is that cost-cutting, and concomitant shifts to user-pays strategies, have fallen disproportionately on those sectors and programmes most likely to generate ‘widening’ as well as ‘deepening’ of human capital; frequently cited are health, education and training. These interventions have often disadvantaged girls. Where there has been investment in higher education, as in some Arab states, this has not always been linked to training and pro-active labour market policies, or investment in plants and other enterprises. The numbers of unemployed graduates cited by media are a witness to this. Moreover, marketisation has been seen as the key instrument for development, to the detriment of human capital generation; this fits neo-liberal economic theory, yet may really be anti-statist ideology.

A focus on finance and ‘responsible’ fiscal policies overlooks the widening and deepening of human capital in youth and then the capacities of the young middle-aged. To paraphrase Robert Cassen's pithy critique (2001), economic development has focused on ‘the progress of growth in per capita income, but very much more about the income than the capita’. Critiques of these focuses have come from across the political spectrum and from a wide range of agencies in the development field. Critics include authoritative figures, notably Joseph Stiglitz (2003) and David Bloom, David Canning and Jaypee Sevilla (2003).

Thirdly, global society has been slow to the point of being negligent in recognising the pivotal importance of youth for development. The emphasis instead has been on what are perceived to be negative aspects of youth behaviour, not the positive potential of youth to further future development once the energy released by cohort flows has been harnessed. Even the World Development Report 2007: Development and the Next Generation (World Bank 2007) is long on risks, although it also discusses the potential of youth. A further weakness in the report is that it puts the onus on youth, rather than on the society (see ‘Introductory Remarks, Part II’). Its underlying message is that youth need our interventions, whereas we need them: without successful integration of youth into society there will be no future development. Stress is also placed on the risk-taking (e.g. unprotected sex) and deviant behaviours (e.g. drugs) among youth. The biases reappear in declarations by the international community on development. These statements either ignore youth altogether or lump them with children, or highlight the negative risks their behaviours might impose on the wider society, notably the possible implications for political stability, and peace or violence. In tone, they prescribe to youth rather than facilitate general development.

As an example, the Programme of Action emanating from the 1994 United Nations' International Conference on Population and Development (ICPD) was particularly remiss. Almost 20% of the entire world population at that date was at 15–25 years, yet youth received no special treatment in the report, except for repeated, almost unhealthy, concern over their sexuality. They were lumped in with children in a section of the report that covered around 50% of the entire global population. This contrasted markedly with the chapter devoted to the 6% of the global population aged 65+ years (Pool 2006b). Overlooked was the fact that youth are characterised by what American demographer Ronald Rindfuss (1991) calls ‘high demographic density’: they are completing formal education, seeking first jobs; they constitute a high proportion of the world's ‘floating populations’ and workforce migrants; they are maturing biologically, leaving home, even starting their own families. All these are normative behaviours for their life-stage.

But the ICPD was not alone in this regard. The UN World Plan of Action for Youth discusses numerous youth issues, but most of the ‘priorities’ are directed at youth behaviours as ‘problems’. Only three or four look at youth in terms of development: their contribution to the economy; the burdens they impose in terms of education, employment and globalisation; and, negatively, hunger and poverty. The remainder deal with pathologies and risks. Some of the recommendations for action are too platitudinous or vague to provide meaningful guidelines.

To take another case, the MDGs represent arguably the broadest consensus yet of the international community about development. But they are a-demographic: they do not address youth per se, although, by contrast, children receive specific and appropriate mention in terms of basic education and infant mortality. Poverty, the poster child of MDGs, actually represents failed development, and in any case its solutions are phrased as ‘one size fits all’: ‘a dollar, or two dollars, a day’. Moreover, the different goals are almost seen as silos, not as interrelated problems. Poverty is undoubtedly an underlying determinant of all the problems the MDGs address, yet with malnutrition it is in a different silo from the three health goals, infant and maternal health and the great epidemic diseases. The determinants of all of these are strongly conditioned by poverty.

Demographic Dividends: A New Perspective

Fortunately, there is not a total absence of paradigms for development that might harness youth capacities. In the late 1990s a body of knowledge emerged, entitled ‘demographic dividends’. The clearest statement of the underlying theory and substantive results is in a monograph by Bloom, Canning and Sevilla (Bloom et al 2003). The underlying theory, built on a classical study by Ansley Coale and Edgar M. Hoover (1958), has become the baseline for ‘new’ perspectives in economic demography that have far-reaching implications for development (Higgins & Williamson 1997; Carvalho & Wong 1998; Bloom et al 2003; Jones 2005). This approach has been tested empirically in many countries, particularly in Asia (Higgins & Williamson 1997) and on European historical data (Malmberg 1994; Malmberg & Sommestad 2000).

The idea of the dividend is simple. As fertility falls and child dependency drops, age-dependency ratios remain low since population ageing has not yet set in. A ‘window of opportunity’ opens when current social sector expenditure, notably on education and health, can be diverted to productive sectors, savings (especially intra-household), investment and infrastructure and institution building. Consequently, many use the less deterministic phrase ‘window of opportunity’ (WO) rather than ‘dividends’, which implies a definite, positive return.

In previous studies, the impacts of population on development were weak, except for extreme situations where there was very rapid growth yet limited resources. In contrast, the demographic dividend analyses to date show that age-structural transitions are strongly related statistically to trends in economic growth (Bloom et al 2003)8, as demonstrated by improved national income. Some authors argue that the dividend is the most powerful explanation of development (Malmberg & Lindh 2006; Fink & Finlay 2007; Pool 2007b). But, there is a major caveat: development relies on good management of the demographic dividend, which is a transitory phenomenon.

A ‘window of opportunity’ opens when current social sector expenditure, notably on education and health, can be diverted to productive sectors, savings (especially intra-household), investment and infrastructure and institution building.

Mechanisms for the dividend are labour force, human capital and savings (Bloom et al 2003: 39–42). The first are essentially demographic and in turn determine the third. The notion of growth, especially at young labour-force ages, corresponds to the concept of human capital widening discussed earlier (see endnote 4); Bloom et al use ‘human capital’ in its conventional sense.

The transitory nature of the dividend means its management must not only be efficient and effective, but as the conditions for it will often be present or only potential, management strategies must be implemented quickly. To measure its onset, economic demographers use support ratios (employed persons gaining labour incomes/dependent populations), but that ratio is confounded by definitional problems, for instance because it excludes non-wage work. A good approximation can be obtained by a simple, conventional demographic dependency ratio: children plus elderly/working-age persons.9 As a rule of thumb, once the proportion of the population who are children drops below 30%, the WO commences, and it finishes when the elderly constitute 15%. Appendix Table data show that, by these criteria, the WO is some time off for many Arab countries. As is also clear, the data for Middle Eastern countries vary significantly. Some, including larger ones, are more advanced in their WO and urgently need investment in human capital if they are to realise their dividend before it ends.

That is also true for the non-Arab Muslim comparators. Iran is interesting because the wider age groupings conventionally used for demographic dividend analyses statistically dampen down the effects of disordered cohort flows shown in Figure 3. Its WO is already under way, yet the demographic turbulence produced by disordered flows will have impacts on all aspects of policy – tertiary education and training are good examples – necessary to capitalise on the dividend.

Finally, the dividend paradigm has also spawned new national accounts methodologies (national transfer accounts, or NTAs), which incorporate cohort effects and also pay attention to intra-family as well as inter-family transfers: the latter come from the social sectors – education, employment, health, housing and social welfare. The NTAs for Asian and Latin American countries and the United States provide a different perspective on development issues. For example, intra-family transfers significantly exceed those for inter-family that are mainly public policy instruments (Mason & Lee 2007).

Future Development in the Arab Countries

No one can forecast the duration and outcomes of the current Arab political uncertainties, but the demography of these countries is rather clearer. Because the data cover people already born and passing through life-cycle stages, one can chart cohort changes and project their human capital implications. Demographic turbulence will continue as disordered cohort flows pass through the age pyramid. For the political structures emerging from the present turmoil to succeed as coherent nation-states, this demographic energy must be harnessed positively. This costs money, but so too does the economic, social and political dislocation that stalled development brings. Development planning should therefore focus on human capital deepening and widening, exploiting available labour forces.

No one can forecast the duration and outcomes of the current Arab political uncertainties, but the demography of these countries is rather clearer.

The scenarios here indicate blocked development in Arab countries. The ‘Arab Spring’ may or may not result in democratic societies that allow religious, ethnic and other forms of diversity. But unless the effects of demographic changes are also mitigated, economic and social development will be thwarted. To reiterate, some of the extreme levels of growth driven by high fertility occur in the politically more volatile states: Yemen, for example. These are often also states facing the effects of highly disordered cohort flows: again Yemen and Libya (Figure 2). It is also worrying to look at the severe fluctuations faced by Iran, a non-Arab state but currently being identified as a threat to regional security. This article has focused on social and economic development. But what if the demographic turbulence spills over into political realms, in turn causing destabilisation that threatens peace? Somalia, which is demographically turbulent, is a sobering example.

One can extrapolate this argument beyond the Arab states to other poorer regions with less educated youth populations, such as sub-Saharan Africa. Will the lack of youth policies highlighted for Arab countries have even more severe consequences elsewhere than in the Middle East, where demographic turbulence has been associated with economic problems and political turmoil, even in oil-rich states?

Notes

¹ Press release issued at the launch of the report on 16 September 2006.

² The raw data come from United Nations' Estimates and Projections, published by its Population Division (2009). Ben Amey computed the indices used here. The UN series is subject to critiques, but is the most widely accepted set of projections. The methodology was developed for work the author did internationally (see Pool in References).

³ This is a term used to describe the fertility of European-descent New Zealanders in the later 19^th century (Pool et al 2007). It implies that women of all marital statuses have a rate in excess of seven births per woman, and married women have almost nine births per woman, the level of American Hutterites, whose rates approach bio-socially feasible maxima.

⁴ The denominator cannot be total change in any period, as some populations have negative growth. This methodology was used in Pool 2005, 2006a, 2007a and 2010.

⁵ There is another epistemological problem with major development implications. For many economists, ‘If the demand for labour equals supply, there is never any involuntary unemployment. Someone who is not working has evidently chosen not to work’ (Stiglitz 2003: 35).

⁶ Conventional economics refers only to deepening, but emerging uses have introduced the concept of widening (Akgün et al 2010: 3). I thank Jacques Poot, a co-author of Akgün, for bringing the concept ‘widening’ to my attention.

⁷ A phrase coined by Jeffrey Williamson and cited by Bloom et al (2003). Current use goes beyond Williamson's technical idea to allude to the policy environment of the 1980s, 1990s and 2000s driven by the IMF and World Bank, and the neoliberal thrust towards marketisation etc. (see critiques such as Stiglitz 2003). Recent (2011) IMF statements recognise that global inequality is a major issue. Agencies of the Washington Consensus undervalued the role of population in development. This is seen even in the recent World Development Report on agglomerations (World Bank 2008). Their growth was not seen as due to demographic changes, but attributed uncritically to ‘market forces’, which were never really defined. It tended to dismiss concerns about the negative aspects of agglomeration, both on sources and destinations of rural-urban migration.

⁸ This link was not identified when grosser indices, GDP and population growth, were correlated. ‘Window of opportunity’ was first coined by Jose de Carvalho (see Pool & Wong 2006: 7).

⁹ Personal communication by Tim Merrill of CELADE, the Population Division of of the UN Economic Commission for Latin America, Santiago, who compared the two methodologies and found almost no significant differences.

Appendix Table Key Demographic Indicators for Arab and Non-Arab Muslim Territories

Large (10 million+)	Window of opportunity ^1	Population (millions)	Growth (percentage)	Fertility rate
Algeria	2005–2045	35	1.5	2.4
Egypt	2020-	84	1.8	2.9
Iraq	2035-	31	2.2	4.1
Morocco	2010–2050	32	1.2	2.4
Saudi Arabia	2015-	26	2.1	3.2
Sudan North & South	2035-	43	2.2	4.2
Syria	2025-	23	3.3	3.3
Tunisia	2005–2040	10	1.0	1.9
Yemen	2040-	24	2.9	5.9
Total: nine		308
Medium (5 – 10 million)
Jordan	2025-	6	3.0	3.1
Libya	2015–2050	6	2.0	2.7
Somalia	Not in period	9	2.3	6.4
UAE^2	1975-	5	2.8*	2.0
Total: five		26
Small (less than 5 million)
Bahrain	2000–2045	1	2.1	2.3
Comoros	2030-	1	2.3	4.0
Djibouti	2030-	1	1.8	4.0
Kuwait^3	1995–2045	3	2.4*	2.3
Lebanon	2005–2040	4	0.8	1.9
Mauritania	2035-	3	2.4	4.5
Occupied Palestine	2045-	4	3.2	5.1
Oman	2015-	3	2.1	3.1
Qatar^4	1985-	2	10.7*	2.4
Western Sahara	2010–2050	1	3.7**	2.7
Total: 10		23
All Arab territories: 24			357
Non-Arab Muslim Countries
Bangladesh	2015-	153	1.4	2.4
Indonesia	2005–2045	219	1.2	2.2
Iran^5	2005–2045	71	1.2	1.8
Pakistan	2035-	166	2.2	4.0
Source: Population Division (2009).
Notes: 1 A dash indicates if the window of opportunity comes after the projection series finishes. ‘Not in period’ means the window of opportunity starts after projection series finishes.
2 In UN data, presented as one figure. UAE's dynamics are complex. Until 1995 the proportion 0–14 years remained under 30%; but the percentage of people older than 65 was low because workforce migration sometimes affected growth extremely (especially in the 1970s). But fertility levels were high. Thus, during 1960–75, the child population grew by 273%; 1975–90 by 274%; but 1990–2005 only 44%. Population (0–14) less than 30% of total by 1975; population (65+) still less than 15% in 2050.
3 Workforce inflows were particularly marked in the 1960s.
4 Inflows were heavy 1960–75, and since 2000. Like UAE, the dynamics are complex. Consequently, the child population grew very rapidly: 211% (1960–75), 130% (1975–90), but only 23% (1990–2005). Population (0–14) less than 30% of total by 1985; population (65+) still less than 15% in 2050.
5 See text on disordered cohort flows affecting development planning, which accounts for fluctuations in the size of the child population.
* Workforce migration affects figures.
** Partly affected by migration through conflict.