Population and Inequality in the Twenty-First Century: Will Dividends Divide the World?

Abstract

This chapter explores the potential role of demographic change on global inequality, at a time of great concern and uncertainty over future prospects. It fills a gap in a “population-development” literature that has extensively studied the links between population and economic growth but neglected implications for inequality. It extends theories of demographic dividends and uneven transitions to predict that current demographic transitions will reduce inequality between, but widen inequality within, countries. The hypothesis is tested in part with decomposition methods that explore the role of global demographic transitions over the last half century. The approach is then used to extrapolate expected trends over the next half century, based on current demographic projections. Further insights are generated by analyzing socioeconomic differences in fertility across low and middle-fertility countries. The findings support the hypothesis of a mixed effect on global inequality.

1. Introduction

Every new century inspires its share of prophecies, and the twenty-first century was no different. Indeed, as the dawn of a new Millennium, it fed even bolder predictions about the near future: the “Internet of Things” would revolutionize production and consumption processes; it would transform data and work environments, and it would accelerate geographic and spatial exploration, perhaps with the first human colonies settling on Mars. The long list of predictions extended far beyond the economy into the world of culture and politics. Huntington [1] thus anticipated a clash of major civilizations, while Fukuyama [2] envisioned an “end of history” in which liberal democracy would spread far and wide, ultimately emerging as the pinnacle of progress in governance. Many analysts foresaw the end of a unipolar world dominated by the US [3].

In many of these predictions, technology was the driving force, the core engine of a high-tech future that would transform global society. Yet some of the push was also expected to originate from social and demographic forces, as several influential population processes were poised to reach historic milestones. Immigration was emerging as the leading source of population growth in Western Europe, fueling worries about a “great replacement” [4]. The size of world population was swelling to an all-time high that would define the Anthropocene, a time when human activity and wanton consumerism wreak serious damage to the environment [5]. After this peak, world population would begin to decline, in another historic moment that would also mark the century [3]. The aging of world population, coupled with automation would transform the nature of work [6]. Together, these forecasts stoked a larger and emerging debate about population and development in the twenty-first century.

Our chapter advances this debate. It does so in two ways. First, it compares this new debate to past debates on population and development (P-D debates hereafter). Second, it takes a distributional perspective. The idea is to go beyond average outcomes to explore inequality. At issue is how population inequality affects economic inequality. In the process, the chapter addresses pressing questions about global economic inequality: Will the current century bring the world closer together or split it apart? Will it accelerate the “big time divergence” claimed by Pritchett [7] or will it flatten the world as envisioned by Freedman [8]?

We predict a partial convergence in which economic inequality both rises and declines. It will decline between countries but rise within. The resulting involution will gradually concentrate the bulk of global inequality within, rather than between, countries. By the century’s end, nations will have become less important than social class in defining a person’s position on the global totem pole. Both facets of this hybrid trend, we argue, stem in part from demographic forces. In particular, global fertility transitions will turn out to be a major spur, operating both as a unifying and a dividing influence. They unify the world by closing the gap between countries, as the global South catches up demographically with the global North and reaps related dividends. As Figure 1 shows, major world regions have completed their fertility transitions in a staggered fashion, starting with Europe and continuing with East Asia, Latin America, South Asia, the Middle East and North Africa, and ultimately sub-Saharan Africa. On the other hand, these fertility transitions will also divide countries of the global South. Just as fertility transitions unfold in top-down fashion across nations, they also unfold from top-to-bottom income groups within nations. For this reason, the early stages of demographic transitions will widen economic inequality within countries in the global South. In sum, we hypothesize that the demographic dividends in the global South will divide the world in interesting ways, raising inequality within countries but reducing the overall inequality between countries:

Figure 1.

Recent and projected levels of fertility, for all major world regions, excluding the high-income groups (1950–2100). Source: based on 2022 Revision of World Population Prospects (https://population.un.org/wpp/), last consulted in January 2024.

We test this hypothesis of partial convergence with two sets of complementary data. To explore inequality between countries (BCI hereafter), we analyze national statistics compiled by the World Bank [9]. The World Bank [9] curates a comprehensive online database containing over 1500 indicators of development for all the world’s nations and territories. The data needed for our analyses was available for 130 countries covering 92% of the global population. This database now spans over six decades (from 1960 to today), making it possible to assess contemporary trends. Our study period covers the century from 1970 to 2070, i.e., from the completion of fertility transitions in the global North till their projected completion in sub-Saharan Africa (Figure 1). Within this century, the year 2020 stands as a strategic mid-point. From that median line, one can analyze the first half of the study period (1970–2020) and build on findings to project trends in the second half (2020–2070).

To explore inequality within countries (WCI hereafter), we use survey data from the United States Agency for International Development (USAID). The USAID’s Demographic and Health Surveys program [10] likewise offers a rich online data platform to support international studies of fertility. It stores and compiles data from household surveys fielded across high and medium fertility nations over the last three decades. We use these data to examine fertility inequality among countries that fielded a DHS survey since 2015.¹ Specifically, we analyze the differentials in total fertility rates between families in the top versus bottom quintiles. These survey data support micro-analyses that complement the macro-analyses based on World Bank statistics.

The rest of the chapter flows as follows: first is a background review of the recent evolution in P-D debates. The next three sections focus each on a separate aspect of inequality, whether dynamic, cross-national, or internal aspects. Section III thus examines how the GDP rankings of various countries evolved in the half century from 1970 to 2020. Section IV explores BCI and the contribution of demographic transitions. Section V explores how fertility transitions are shaping WCI. The chapter closes with a summary of findings and speculation about the future.

2. Population and development: classic vs twenty-first-century debates

The recent P-D debate unfolded into three historical phases that differ in substance and scale (Table 1). Its first and classic phase evinced a strong focus on aggregate outcomes, mostly national rates of population and economic growth [11, 12]. Ideologically, it pitted free-market advocates against Malthusians worried about the “population bomb” and its fallout on scarcity, conflicts, and epidemics [13]. Empirically, it relied on evidence from historical or cross-country regressions, despite caveats on the reliability of these methods [14].

	Phase 1	Phase II	Phase III
Time period	Pre 1990s	1990–2000	>2000
Label	Classic	Post Cairo	Millennium
Analytical level	Macro	Micro	Meso
Key population variables	Population growth	Demographic events and reproductive health	Age structure/cohort
Key economic variables	GDP growth	Individual and family wellbeing	Demographic dividends
Tone	Worry	Hands off	Middle ground
Analytical methods	Cross-country regression	Microregression, causal analysis	Decomposition methods

Table 1.

Three historical phases in the recent debate on population and development.

Source: author.

The second phase in this debate grew out of the 1994 Cairo’s International Conference on Population and Development. This post-Cairo phase moved the debate from a macro- to a micro-level scale, with the policy paradigm shifting away from national targets and to individual reproductive experiences. It spawned a wave of micro-level research that was empirically more detailed and rigorous, thanks in part to large data and statistical tools that were becoming increasingly accessible [15, 16, 17]. Yet this new P-D equation had two weaknesses of its own: its left-hand side downplayed structural influences and its deep interaction with individual factors. Such individual focus runs counter to current social science theory [18, 19]; and its right-hand side downplayed national impacts, running counter to the United Nations’ agendas of Millennium and Sustainable Development Goals [20].

A third perspective was thus essential to stay relevant in the era of Millenium Development while keeping the spirit of Cairo. The challenge was to reconcile micro and national perspectives, i.e., to craft a people-centered approach that also remains attentive to national goals. The solution was found in meso-level studies using sub-regions, social classes, cohorts, or age groups as units of analysis. Studies in this “Millennium” phase of the P-D debate meet both criteria of macro-relevance and micro-detail. They do so with analytical strategies that explain national trends by aggregating the more detailed information from constituent subpopulations [21].

Beyond scale and substance, the P-D debate evolved in tone, with the strident opposition between Malthusians and Cornucopians softening to concede that “demography is not destiny [22, 23]:” in other words, innate demographic features such as sex or race do not seal one’s fate, as they leave room for individual agency and structural effects. Yet this rejection of determinism does not imply irrelevance. Rather, it merely accepts the reality of overdetermination: no single factor, including demographics, can fully account for the entire variation in human experience. Population does matter, but in nuanced ways that consider contextual variation, intersectionality, cumulation, and momentum.

The principle of contextual variation means that demographic influences vary over time and space.² The intersectionality principle acknowledges the complex interactions between multiple facets of individual identity. The principle of cumulation means that small but persistent annual differences between groups will end up building large gaps over time. Finally, the momentum principle means that the effects of demography gather strength over time, and they can persist even after the initial push wanes. Compared to economic and social phenomena, demographic influences are less volatile. An economic bust can quickly follow a boom, but demographic processes are less prone to wild swings and rollbacks; they are more likely to show momentum and remanence.

3. Population trends and the global order

3.1 Theory

This section explores the dynamic dimension of inequality. Will there be a major turnover in world order during this century, or will the current economic rankings of countries persist? The theory is unclear. According to dependency theory, no reshuffling is forthcoming precisely because the “development project” is set to maintain the exploitation of poor countries by rich ones [18]. Rich countries stay on top, and inequalities can in fact widen. Modernization theories paint a less gloomy picture: the global order may not change, but inequalities can shrink. Today’s poor countries slowly catch up, thanks to modern technology, improved governance, and international aid but also thanks to demographic processes. “Opportunistic” theories of development are even bolder in their outlook. Unlike dependency or modernization theories, they see turnovers as possible: drastic changes in the global resource environment can create unique openings that lower-income countries can exploit to overtake leading nations. The key question here is whether fertility transitions can support such a quantum leap in the development of poor nations.

3.2 Extent of global mobility

To assess economic mobility, we rank all countries by GDP per capita, grouping them into five quintiles. By comparing each country’s rankings/quintile positions between 1970 and 2020, one can thus monitor jumps in rankings, distinguishing between nominal, real, and major jumps. Nominal jumps are small but sufficient to move a country to the next quintile, simply because the country stood near the cutoff line in 1970. Such jumps are artifactual and not statistically meaningful. Jumps become more meaningful when a country overtakes at least 26 nations, i.e., the number of countries in a quintile. They are even more meaningful and labeled “major” if they move a country across more than one quintile.

Global mobility is gauged by the frequency of real and major jumps. If these jumps turn out to be frequent, a case can be made for opportunistic theory. If they are rare, then the evidence is consistent with Marxist theories that posit a continued subjugation of countries from the global South [18].

The evidence on mobility is summarized in Table 2. The diagonal on this matrix indicates immobility, i.e., the countries staying within the same quintile between 1970 and 2020. Cells above the diagonal capture upward mobility, and those below the diagonal capture downward mobility. Cells close to the diagonal capture single-quintile jumps; those farther away capture multiple-quintile jumps. Off-diagonal countries are marked with an asterisk, the number of asterisks reflecting the real number of quintiles jumped. In theory, the number of countries above the diagonal should be the same as the percentage below, since we assume a zero-sum hierarchy where ascending countries replace other countries.

Country’s global position		As of 1970
		Top quintile (Q1)	Second quintile (Q2)	Third quintile (Q3)	Fourth quintile (Q4)	Bottom quintile (Q5)	Total
As of 2020	Q1	20 (76.9%)	Japan, Hong Kong, Austria, Greenland, Singapore^*, Ireland^*				26
	Q2	Fr. Polynesia, Bahamas, Kuwait^*, Italy, N. Caledonia, Andorra	13 (50.0%)	Malaysia, Seychelles, Costa Rica, Oman, St Kitts and Nevis^*	Korea^**	China^***	26
	Q3		Suriname, Jamaica, South Africa^*, Peru, Argentina, Mexico, Cuba	14 (53.8%)	Paraguay, Equatorial Guinea^*, Thailand^*, St Vincent^*	Botswana^**	26
	Q4			Senegal^*, Nicaragua, Iran^*, Algeria, Tunisia	15 (57.7%)	Myanmar^*, Cambodia, India, Kenya, Bangladesh, Indonesia^*	26
	Q5			Zambia^** Zimbabwe^**	Madagascar, Afghanistan, DRC^*, Syria^*, Pakistan, Kiribati	18 (69.2%)	26

Table 2.

Global mobility matrix for world countries, 1970 to 2000.

Source: author’s calculations, based on GDP statistics from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last accessed in January 2024.

* denotes countries that jumped one quintile by overtaking at least 26 nations.

**, and *** denote countries that jumped two or three quintiles, respectively.

The matrix suggests three main findings. First, the turnover in world rankings was substantial, at least in nominal terms. Belying expectations from dependency theory, cross-quintile mobility was common, as it was experienced by more than one-third (39%) of all nations. Yet many of these jumps were only nominal, i.e., they occurred only because the country was very close to the cutoff line. Real jumps—countries moving at least 26 spots—represent only about 15% (19/130) of all cases. Major jumps were even less common, with only 4% of countries moving by more than one quintile. Countries making major leaps forward included China (from 5th to 2nd quintile), South Korea (from 4th to 2nd quintile), and Botswana (from 5th to 3rd quintile). The two countries that fell back the most were Zambia and Zimbabwe (both from 3rd to 5th quintile).

As a second observation, the jumps were least frequent at the tails of the distribution, i.e., countries that were either very rich or very poor in 1970: most of the reshuffling took place around the middle, not at the extremes of the distribution. The percentages of countries staying in place were 77% at the top and 69% at the bottom, against smaller percentages (58%, 54%, and 50%, respectively) in the middle quintiles. This relative immobility of the poorest and richest nations should nuance the optimistic conclusions derived from overall mobility data. While the overall mobility data support the optimism from opportunistic theory, a focus at the top or bottom of the GDP distribution shows less mobility. It thus supports dependency theory, especially since very few jumps exceeded one quintile.

3.3 Role of population in global mobility

Given the above reshuffling of GDP rankings between 1970 and 2020, the question is whether population was a factor. To answer this question, we first looked at the statistical correlation between fertility declines and economic mobility, allowing for a 5-year lag. We found only a very modest correlation, with the changes in fertility rankings explaining only 1% of the variance in the change in GDP rankings (Figure 2). Even if this correlation had been large, it still would be insufficient proof of a causal effect. Given this small and possibly spurious correlation, a causal or single story may not explain the economic trajectories of all nations during that period. A case-by-case analysis is warranted. The chart shows four types of countries, including countries that (a) progressed on both the fertility and the economic fronts (China), (b) progressed on neither front (the DRC), (c) progressed on the economic but not the fertility front (Equatorial Guinea), and (d) regressed economically despite progress on the fertility front (Syria). Again, countries are best studied on a case-by-case basis.

Figure 2.

Correlation between national gains in fertility rankings and gains in GDP rankings (1970–2020). Source: author’s calculations, based on GDP data from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last access, January 2024.

We thus complement the correlation analysis with decomposition methods. The specific method used estimates how much a country’s GDP gains between 1970 and 2020 is explainable by changes in the age dependency of the national population. The general idea is to express a country’s GDP per capita as a product of its labor productivity, employment levels, and age structure.³ Any change in per capita GDP can therefore be traced to changes in these three components. We pay special attention to the contribution of changing age structure, which captures the so-called dividend from fertility transitions. We can then assert whether these dividends played a major role in fostering the mobility of nations.

The results show substantial boosts among the Asian Tigers, with South Korea in particular growing its GDP nearly sixteen-fold from 1965 to 1981 [9]. In the process, the country vaulted over many countries, moving from the bottom to the second quintile by the turn of the century. Past studies to estimate the contribution of demographic dividends to this economic growth have generated estimates ranging anywhere from 1/3 to 1/5 [25, 26]. Dividends also fueled Botswana’s remarkable economic progress during that period. The country’s GDP grew by 140% between 1990 and 2020, and 22% of this growth reflected favorable changes in the country’s age dependency [26].

4. Population and “between-country” inequality

The previous sections analyzed trends in rankings, i.e., the relation positions of world countries and how much these positions changed over time. As a complement, this section explores absolute gaps. More than rankings, absolute gaps capture the full extent of inequality and its trends. In theory, world economies could change in isomorphic ways that preserve the rank order of nations but increase or reduce their absolute gaps. For this reason, absolute inequalities are also worth monitoring. The question in this section is about contemporary trends in inequality, a topic of spirited debate. So spirited, ideological, and partisan the debate that prominent scholars deemed it “silly,” with one side steadfast in claiming “big time divergence,” and another just as unequivocally seeing a “flattening world” [27]. The two sides eventually reconciled under a consensus view that inequality was both waning and widening, depending on whether one looks between or within countries [20]. This section focuses on BCI, and the role of demographic forces in the process.

4.1 Theory

Dividend theory can explain why fertility transitions would reduce BCI. According to this theory, sustained declines in national fertility tend to lower rates of age dependency in ways that create a temporary window of opportunity for saving and investing in economic growth [28]. If fertility declines occur in poor countries, as is the case now, the corollary effect at the global level will be an economic convergence: as today’s low-income nations finally join the global fertility transition, the resulting economic dividends will narrow the income gap with richer nations. In other words, demographic convergence begets economic convergence.

This corollary theory remains untested for conceptual and methodological reasons: conceptually, as long as the P-D debate remained limited to its “growth-growth” dimension, concerns about inequality were obscured. This framing has fortunately broadened on both the population and the development sides of the equation. Analyses increasingly extend beyond population growth to include age structure [28]. They likewise cover multiple goals of sustainable development, such as schooling, gender, insecurity, or inequality. Our chapter embraces this broader framing. Rather than linking population and economic growth, it studies the link between demographic and economic inequality.

4.2 Trends in “between-country” inequality

Our empirical analyses begin by reviewing trends in GDP inequality between countries. We explore four complementary indices of inequality, including the Gini, the Theil index, the Mean Logarithmic Deviation (MLD), and the coefficient of variation. We focus on the MLD index which is more easily decomposable. The MLD computes as

$${\mathit{MLD}}_t=\sum p_{\mathit{jt}}\ast \mathit{ln}\left(\frac{1}{i_{\mathit{jt}}}\right)$$E1

Where j and t index countries and time respectively; (p_j) is a country’s share of the world population, and i_j is the national income ratio, i.e., its income per capita relative to the world’s average.

Using data from the World Bank Development Indicator database [9], we extract country information on population size, labor force participation, age structure, and gross national income per capita (GDP). We chose the Atlas variant of GDP because of data availability. The data for this variant covered 121 countries, including 44 in Asia, 52 in Latin America, and 25 in Africa over the five decades. The results in Figure 3 clearly show a historical decline in GDP inequality between countries. To be sure, one can see qualitative differences in the initial levels and the magnitude of the decline: depending on metric, the decline ranged from a low of 17% (Gini coefficient) to a high of 44% (Mean Log Deviation). Yet all metrics show a clear decline, which becomes steeper and steadier after the turn of the century. While the trend oscillated before the 1990s, the decline became monotonic after the year 2000.

Figure 3.

Global GDP inequality between countries (1970–2019). Source: author’s calculations, based on GDP data from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last access, January 2024.

5. Population and “between-country” inequality

We used a decomposition analysis to explore how population trends affected between-country inequality. Building on Eq. (1), and further noting that a country’s GDP per capita is a function of labor force productivity (π_j) and population age structure (α_j), one can decompose the change in MLD between two time periods as

E2

where barred values represent averages, and Δ marks change between two time periods. For instance, when studying change in global inequality between 1980 and 1990, ${\underset{\_}{i}}_j$ =(i_j(1980) + i _j(1990))/2, and Δp =p₁₉₉₀ − p₁₉₈₀.

Table 3 shows the findings from this decomposition. It shows what percentage of the 1970–2020 decline in BCI comes from the countries’ relative changes in the population size, age structure, and productivity. The table lists the results for the entire period (last column) but also for each decade, from 1970 to 2020.

	Time period
Inequality outcomes	1970–1980	1980–1990	1990–2000	2000–2010	2010–2019	1970–2019
Total change in MLD	0.019179	−0.055	−0.06042	−0.16469	−0.06403	−0.32496
Percent change in MLD due to changes in
Population size	−20%	12%	−1%	−9%	−18%	−7%
Age structure	−5%	24%	33%	12%	31%	19%
Productivity	128%	64%	68%	95%	88%	86%

Table 3.

Relative contributions of economic and demographic factors to the global convergence in GDP per capita observed between 1970 and 2019 [29].

Source: author’s calculations, based on GDP statistics from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last accessed in January 2024.

*Some of the percentages do not add up to 100%, due to rounding error.

For the full period, the relative changes in productivity accounted for 86% of the total convergence: countries grew economically similar because their labor productivity became more similar, perhaps the result of a worldwide expansion of education and diffusion of technology. However, demography was also a factor. Countries became more similar in age structure, and this demographic convergence accounted for 19% of the GDP convergence. Of the two population factors, age structure was the most influential. Changes in the relative size of population did in fact exacerbate rather than reduce inequality. This is because lower-income nations were growing faster, thus steadily accounting for a growing share of the world population. The fact that age structure, not population size, is the more influential factor underscores the limitations of the classic P-D debate and its narrow focus on population growth.

A detailed analysis by decade shows that the global trend in age dependency has consistently worked to reduce inequality. In the first decade, when global inequality was rising, they slowed that rise. In subsequent periods, when global inequality was declining, they accelerated that trend, especially during the 1990s and the most recent decade (2010–2019). The relative growth in population size did worsen inequality throughout the study period. However, the first two decades were an exception, with the results for 1980–1990 at a time when China was implementing its one-child policy.

One can probe deeper into the drivers of this GDP convergence by looking at the contributions of different world regions (Table 4). As in Table 3, one can review the full study period or focus on individual decades. The results suggest the following: first, the largest contributions (61%) came from South Asia, followed by East Asia (16%) and the Global North and the Middle East and North Africa (MENA) region (14% each). Over time, East Asia and South Asia stood out as the world leaders in reducing global inequality in GDP, with the first of these regions dominating in the first three decades and the second taking over since. Asia’s contribution comes from a mix of slowing population growth and rising productivity (results not shown). So far, sub-Saharan Africa is the only world region that has not contributed to this global economic convergence. The next half century will be interesting in that regard. As Africa completes its demographic transition and if the region reaps substantial dividends, it can become a leading driver of global convergence. Already, over the last three decades, several African countries have advanced their transition, reaping substantial dividends. During the 1990–2020 period, a dozen of African countries at least doubled their GDP, based in part on harnessing dividends. Leading countries on that list (Table 5) include Kenya, Morocco, Namibia, Tunisia, Eswatini, Botswana, Rwanda, Cabo Verde, and Lesotho. In all of these countries, changes in age structure accounted for anywhere from 14% to 28% of the entire economic growth registered.

	Time period
Inequality outcomes	1970–1980	1980–1990	1990–2000	2000–2010	2010–2019	1970–2019
Total change in MLD	0.019179	−0.055	−0.06042	−0.16469	−0.06403	−0.32496
Percent change in MLD due to changes from
Global North	268%	−72%	−86%	−7%	−34%	14%
East Asia	−299%	124%	163%	55%	38%	16%
South East Asia	−70%	18%	25%	13%	22%	2%
Latin America and Carribbean	49%	−4%	−7%	0%	−2%	7%
South Asia	32%	61%	63%	42%	107%	61%
Middle East and North Africa	79%	10%	−9%	−2%	−9%	14%
Sub-Saharan Africa	45%	−38%	−50%	−3%	−23%	−14%

Table 4.

Decomposition results for the relative contributions of various world regions to the global convergence in GDP per capita observed between 1970 and 2019.

Source: author’s calculations, based on GDP statistics from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last accessed in January 2024.

	GNI per person		Period change		Decomposition results: % of growth linked to change in
	1990	2020	Nominal	Relative	Labor productivity	Employment	Labor force participation	Age dependency
Sudan	999	642	−357	−36%	101%	12%	9%	−21%
Niger	336	549	213	63%	123%	2%	−18%	−7%
Somalia	133	424	291	219%	104%	−1%	1%	−4%
The Gambia	306	743	437	143%	104%	−3%	−1%	−1%
Mali	288	833	545	189%	105%	−5%	−3%	2%
Nigeria	556	2003	1446	260%	115%	−5%	−12%	3%
Tanzania	192	1052	860	448%	99%	1%	−3%	3%
Angola	777	1776	998	128%	104%	−6%	−2%	4%
Malawi	183	586	402	220%	100%	−1%	−3%	4%
Chad	267	634	367	138%	125%	−1%	−28%	4%
Mauritius	2429	10,228	7799	321%	97%	2%	−4%	5%
Guinea	407	963	556	136%	101%	−2%	−4%	6%
Uganda	325	799	474	146%	100%	−1%	−4%	6%
Cote d’Ivoire	776	2280	1504	194%	110%	1%	−18%	6%
Benin	367	1278	912	249%	95%	0%	−2%	7%
Equatorial Guinea	263	5803	5540	2106%	96%	−1%	−3%	7%
Zambia	434	1163	729	168%	96%	5%	−9%	8%
Ghana	396	2313	1917	484%	99%	1%	−9%	8%
Sierra Leone	186	509	323	174%	109%	−2%	−16%	8%
Guinea-Bissau	224	761	537	240%	94%	0%	−2%	8%
Egypt, Arab Rep.	756	3004	2249	298%	100%	2%	−11%	9%
Burkina Faso	332	773	441	133%	128%	−3%	−34%	9%
Ethiopia	257	893	636	248%	91%	0%	0%	10%
Congo, Rep.	927	1815	889	96%	102%	−6%	−7%	10%
Mauritania	729	1669	940	129%	107%	−2%	−17%	12%
Togo	402	921	520	129%	92%	0%	−4%	13%
Kenya	373	1835	1462	392%	88%	−2%	0%	14%
Morocco	1167	3069	1902	163%	99%	3%	−20%	16%
Namibia	1993	4555	2561	128%	77%	−1%	7%	17%
Tunisia	1413	3305	1891	134%	90%	−1%	−6%	17%
Senegal	923	1430	507	55%	110%	5%	−33%	18%
Eswatini	1294	3393	2099	162%	96%	−6%	−11%	20%
Cameroon	966	1520	554	57%	91%	10%	−23%	21%
Botswana	2724	6504	3780	139%	81%	−6%	3%	22%
Rwanda	348	774	426	123%	87%	−1%	−8%	22%
Zimbabwe	864	1140	276	32%	75%	1%	0%	24%
Madagascar	308	469	161	52%	77%	7%	−8%	24%
South Africa	3130	6012	2883	92%	97%	2%	−25%	25%
Cabo Verde	898	3058	2160	241%	90%	−4%	−13%	26%
Lesotho	546	1211	665	122%	69%	24%	−22%	28%
Gabon	4820	7030	2209	46%	93%	−15%	−10%	31%
Comoros	979	1410	431	44%	72%	−13%	6%	35%
Algeria	2366	3571	1205	51%	24%	60%	−24%	39%
Central African Republic	481	508	27	6%	84%	−13%	−19%	49%
Burundi	218	231	12	6%	257%	−2%	−264%	109%
Congo, Dem. Rep.	na	557	na	na	na	na	na	na
Eritrea	na	na	na	na	na	na	na	na
Liberia	na	602	na	na	na	na	na	na
Libya	na	7739	na	na	na	na	na	na
Mozambique	na	467	na	na	na	na	na	na
Sao Tome and Principe	na	2094	na	na	na	na	na	na
South Sudan	na	na	na	na	na	na	na	na
Seychelles	5033	13,767	8734	174%	na	na	na	na

Table 5.

Decomposition of recent economic growth among African nations (1990–2020).

Source: author’s calculations, based on GDP data from World Development Indicators (https://databank.worldbank.org/source/world-development-indicators), last access, January 2024.

Numbers in these columns represent, respectively, the numerical change (rounded values) in the GDP per capita and the percent change in GDP per capita between 1990 and 2020.

6. Population and “within-country” inequality

Global inequality combines the inequalities found between and within nations. While BCIs may reduce inequality across nations, it is also essential to pay attention to inequalities within countries because these WCI are essential to fair and inclusive societies, and to individual sense of relative deprivation. From an individual perspective, and despite the reach of globalization, the ability to keep up with one’s immediate neighbors remains the ultimate yardstick of accomplishment. We specifically focus attention on the possible role of fertility transitions in fostering economic inequality within nations.

6.1 Theory

Why would fertility transitions affect WCI? One answer comes from dilution theory. According to this theory, parents with large progenies must split their time and material resources among multiple siblings, and this reduces the endowments per child [30]. For this reason, a society in which poor families bear more children than richer families do will breed resource inequality among children. This resource inequality in childhood later fuels income inequality when children become adults. For the same reason, fertility transitions that unfold in top-down fashion should widen economic inequality, especially if these fertility changes occur alongside adverse changes in family structure [31]. Altogether, if higher-income groups are first to adopt new reproductive behavior that deepens the per capita endowments of children, the early phases of fertility transitions will foster economic inequality. This divergence could wane during the late phases of fertility transitions when fertility begins to fall among low-income groups. Yet it could also persist if rich families actively leverage their past position to restrict the poor’s social mobility.

6.2 Trends in within-country inequality

Recent studies have shown a sharp rise in WCI in most industrial nations [32]. These internal inequalities have been less studied in the global South where poverty remained the primary worry. Yet this region may be facing similar or even steeper rises in inequality. A recent review of national statistics on inequality shows that Africa is now home to 8 of the top 10 most inequal (and 11 of the top 20) countries in the world [9]. Given this rising inequality, given concurrent fertility transitions, and given expectations from dilution theory, it is reasonable to consider how these transitions might affect WCI.

6.3 Role of population in within-country inequality

Dilution is compelling as a theory, but its empirical evidence is elusive. To prove dilution, it is not enough to show a negative correlation between sib size and children’s outcomes because such correlation may merely reflect endogeneity in the reproductive choices of families. For this reason, it is useful to complement correlation with a decomposition analysis to show how changes in the resource inequality of children derive from a mix of economic, social, and demographic factors. Key factors are likely to include inequalities in (a) families’ incomes, (b) the share of income devoted to children, (c) the number of children, and (d) access to public subsidies. Unfortunately, the detailed data needed for this analysis was not readily available. Nonetheless, we draw tentative inferences by monitoring fertility inequality. Economic inequality will widen in the course of a fertility transition if it is accompanied by rising levels of fertility inequality, especially if the country’s income inequality is also rising.

To assess this theory, we first explore how fertility inequality evolves during a fertility transition. One can explore this link from a historical perspective, by following individual countries over time. One can also explore it from a comparative perspective, by comparing countries at successive stages of their transitions. Results from a historical perspective (results not shown but see [33]) confirm the inverse U pattern expected in theory. Fertility inequality grows after the onset of a fertility transition, and it gradually recedes as the country completes its transition.

Figure 4 shows the results from a comparative perspective, based on recent DHS data. Specifically, we use information for all 61 countries that fielded DHS surveys in/after 2015. For each country, we compute the differences in the birth rate of the bottom versus the top SES quintile. The numbers thus indicate the difference in average birth rates between families in the bottom versus top SES quintiles. The results show that all the numbers are positive: on average, poorer families always have a higher fertility rate than richer families do; the difference averages almost 2.5, and it ranges from a low of 0.1 (Armenia 2016) to a high of 4.5 children (Angola 2015). Importantly, the data show an inverse U pattern; inequality peaks in the initial stages of fertility transitions when the top SES group begins to innovate demographically and to separate itself from the rest. There is of course some variation around this inverse U curve, but the average pattern is strong: taken alone, the stage in fertility transition explains a quarter of all variation in the fertility inequalities across all study countries. This initial divergence in birth rates is expected to fuel the resource inequality among children and, ultimately, the economic inequality in the next generation. Resource inequality among children will widen most severely if inequality in parental income also grows during that period and if public support is weak or weakening.

Figure 4.

Fertility gap between top and bottom SES quintile over the course of fertility transitions. Source: author’s calculations, based on DHS data (https://www.statcompiler.com/fr/), last access, January 2024.

7. Conclusion

The central question raised in this chapter is whether demographic dividends will unite or divide the world, i.e., whether they will reduce or raise global income inequality. Our analysis of recent data shows mixed influences. Current fertility transitions in the global South are pushing global income inequality in two different directions: they narrow the inequality between countries but raise the inequalities within nations.

Looking across nations, fertility transitions have typically unfolded from the top down, beginning in higher-income regions before spreading to other regions. By 1970, countries of the global North had completed their transitions. Since then, lower-income nations have begun to catch up demographically. As they reap the dividends from these demographic transitions, they catch up economically as well, closing the GDP gap between rich and poor nations. Over the last half century, the between-country inequality in GDP per capita fell anywhere from 17% to 44%, depending on the inequality metric used. A decomposition analysis shows that much of this cross-national convergence in GPD (86%) was due to convergence in labor productivity. Yet, population was a notable factor. Further, changes in age structure, rather than population size, were the bigger population factor.

Looking within nations, fertility transitions also unfold from the top down. Such a pattern is expected to raise economic inequality within countries during the early stages of the transition, insofar as large sib size dilutes the resources available to individual children. The effects of this dilution on economic inequality also depend on the patterns of parental and public investment in children. Because we lack detailed information on these patterns of families’ resource allocation, we did not directly estimate the contribution of fertility transitions to resource inequality. Nonetheless, the partial evidence mustered is consistent with the expectation of rising income inequality during the preliminary stages of fertility transitions.

In sum, contemporary fertility transitions have both “dividend” and “divider” effects. The “dividend” effect is fueling an economic convergence of world countries driven in part by fertility transitions in the global South. The “divider” effect also derives from the top-down pattern in which transitions begin, a pattern that initially raises economic inequalities among children. This combination of ‘dividend’ and ‘divider’ effects account in part for the double movement in inequality observed during the last half century.

Looking forward, the world is projected to complete its fertility transition by 2070. If these transitions continue to induce the same mix of “dividend” and “divider” effects, the current involution of inequality will continue. The end result would be to flatten the world from a cross-national view but also to fracture it from a national view. This mix of flattening and fracturing will turn the world into a global village in which nations become less important than social class in defining people’s standing on the global totem pole. Neither dividends nor divider effects are automatic. Countries harness dividends only if they work to take advantage of their periods of low age dependency by creating employment, encouraging savings, and channeling investments in the most productive sectors. Likewise, divider effects are most severe when fertility transitions occur alongside other demographic and socio-cultural transformations that foster ruthless competition and inequality. Adverse transformations in family structure –notably assortative marriage and greater prevalence of singlehood and divorce among poorer mothers—have occurred in the global North [31]. If these transformations extend to the global South, they will accentuate the fertility transitions and magnify inequality.