As we approach 2015 and come closer to meeting the Millennium Development Goals (MDGs) targets, the remaining challenges take on increased urgency. One of these is reaching out to the last and most difficult groups to reach—vulnerable populations with tenuous connections to the modern state and its economy.
Indigenous groups tend to have deeper poverty, less access to education, and worse health outcomes than the general population. In some rapidly growing economies, particularly in Asia, indigenous groups have enjoyed reductions in poverty at a pace comparable to general society. In other countries, particularly in Latin America, slow growth has slowed the pace of poverty reduction for indigenous groups, and targeted programs have made only a limited contribution. There is an urgent need for research to improve the availability and reliability of data on human development among indigenous groups—and to determine what kinds of programs have improved their welfare and in what contexts.
This chapter documents the poverty, education, and health of vulnerable and indigenous groups, highlighting their progress—and the remaining gaps. It also reviews countries that are off track in meeting the MDGs mainly because they have large vulnerable populations and countries on track that have off-track population groups. It also touches on the impact of the modern world on small, largely isolated groups, highlighting the need for sensitivity in dealing with them.
Last in heading down the home stretch
The last groups to be helped in countries are invariably in areas difficult to reach, are usually vulnerable, and have a tenuous connection to the modern state and its economy. Coming from minority ethnic groups and indigenous populations (box 4.1), they tend to speak a language different from the majority.1 Whereas data coverage is a perennial challenge, MDG indicators for ethnic minorities and indigenous peoples are (with some exceptions) worse than population averages.
Poverty is higher among indigenous peoples than in the general population
As the global development community looks for ways to meet the MDG of halving (from its 1990 level) the share of people in poverty by 2015, it cannot ignore the plight of the world’s indigenous peoples (box 4.1). They make up 4 percent of the global population—nearly 300 million—but they account for about 10 percent of the world’s poor; and nearly 80 percent of them are in Asia.
Who are indigenous peoples?
“Indigenous peoples” has no widely accepted definition. The World Bank’s position is that “because of the varied and changing contexts in which indigenous peoples live and because there is no universally accepted definition of indigenous peoples, World Bank policy does not define the term. Indigenous peoples may be referred to in different countries by such terms as ‘indigenous ethnic minorities,’ ‘aboriginals,’ ‘hill tribes,’ ‘minority nationalities,’ ‘scheduled tribes,’ or ‘tribal groups’” (World Bank Operational Directive 4.10).
The United Nations has not adopted a definition but has developed a modern understanding of the term based on a variety of characteristics: self-identification at the individual level and accepted by the community as their member; historical continuity with precolonial or presettler societies; a strong link to territories and surrounding natural resources; a distinct social, economic, or political system; a distinct language, culture, and belief system; individuals who form nondominant groups within society; and those who resolve to maintain and reproduce their ancestral environments and systems as distinctive peoples and communities.a
Moreover, evidence is growing that joining together under a common identity as indigenous peoples is fairly new and has accompanied a process among some groups of “reclaiming” identity.
With few exceptions, MDG indicators for indigenous groups across Asia are worse than all-population indicators. Under-five mortality rates for the Nepalese Janajati are distributed around the national level, but as a whole are below (that is, better than) the national level. In India, however, infant mortality among the scheduled tribes is uniformly higher (worse) than the national average, and water deprivation rates both exceed and fall short of their national levels. Among the Hill Tribes in Thailand; the Kammu and Leu in the Lao People’s Democratic Republic; and the Hmong, Muong, and BaNa in Vietnam, these rates are the worst in the region. The lowest female literacy rates are found among the Hmong in Lao PDR and Vietnam. And the existence of vulnerable groups who fare much worse than the general population is hardly unique to developing countries. In Australia and New Zealand, all indicators for Aborigines and Maori are worse than national averages.
Data coverage is far more limited in Africa, making overarching conclusions difficult. In many cases, available data do not cover core groups widely considered to be indigenous because of their small size (for example, the Ogiek in Kenya). The existing data show that under-five mortality rates tend to be highest among West African groups (such as the Fulani and Tuareg) and lowest among the Masai and Ethiopian groups. These last two groups also experience the highest rates of water deprivation. Education indicators are uniformly worse; even in countries with higher literacy (such as Namibia), the male literacy rate for San males is less than half that of the national sample; and for females, it is less than one third.
Source:Hall and Patrinos 2011.a. UN 1981.Turning the situation around will require widespread and sustainable economic growth and poverty reduction, along with strategies to address multiple sources of disadvantage to reach those who need—and are willing to accept—special assistance.
One-third of indigenous people are poor.2 Among the majority of countries with disaggregated data, poverty rates exceed 50 percent (table 4.1). Although the majority of indigenous people come from China and India, the indigenous poor are slightly more evenly distributed, largely owing to their strikingly low poverty rate in China. In other countries, indigenous peoples have disproportionately high poverty rates.
Poverty rates, latest year available
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007). Head count poverty rates are national.
Poverty rates, latest year available
| Poverty head count (% poor) | ||
|---|---|---|
| Country/data year | Indigenous | Nonindigenous |
| Democratic Republic | ||
| of Congo/2005 | 84.8 | 71.7 |
| Mexico/2008 | 80.6 | 45.3 |
| Ecuador/2006 | 78.0 | 46.6 |
| Guatemala/2006 | 74.8 | 36.2 |
| Gabon/2003 | 70.1 | 32.7 |
| Bolivia/2006 | 69.3 | 46.0 |
| Peru/2005 | 62.3 | 35.0 |
| Vietnam/2006 | 52.3 | 10.3 |
| Lao PDR 2002 | 50.6 | 25.0 |
| Brazil/2002 | 48.0a | 23.0b |
| India/2004 | 43.8 | 22.7 |
| Chile/2006 | 15.2 | 9.1 |
| China/2002 | 5.4 | 3.5 |
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007). Head count poverty rates are national.
Poverty rates, latest year available
| Poverty head count (% poor) | ||
|---|---|---|
| Country/data year | Indigenous | Nonindigenous |
| Democratic Republic | ||
| of Congo/2005 | 84.8 | 71.7 |
| Mexico/2008 | 80.6 | 45.3 |
| Ecuador/2006 | 78.0 | 46.6 |
| Guatemala/2006 | 74.8 | 36.2 |
| Gabon/2003 | 70.1 | 32.7 |
| Bolivia/2006 | 69.3 | 46.0 |
| Peru/2005 | 62.3 | 35.0 |
| Vietnam/2006 | 52.3 | 10.3 |
| Lao PDR 2002 | 50.6 | 25.0 |
| Brazil/2002 | 48.0a | 23.0b |
| India/2004 | 43.8 | 22.7 |
| Chile/2006 | 15.2 | 9.1 |
| China/2002 | 5.4 | 3.5 |
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007). Head count poverty rates are national.
A sizable poverty gap remains for indigenous people in global terms, ranging from small in China to significant in Vietnam.3 The gap has even been increasing over time.
Research from Latin American countries with large indigenous populations—Bolivia, Ecuador, Guatemala, and Peru—shows a sticky persistence of (nonimproving) poverty rates for indigenous peoples over time. Progress used to be limited in Mexico as well, but poverty rates there have been falling over the last decade, especially between 2004 and 2008.4 This may be the first period in history that the indigenous population in Mexico has seen an improvement. More important, analysis in Mexico shows that the difference in poverty rates between indigenous and non-indigenous peoples has been declining mostly as a result of explained or observable factors: education and access to services, the policy-driven variables, are important.
Chile has also seen reductions in indigenous poverty in recent years. But other countries in Latin America have not been so successful. Guatemala reduced poverty rapidly until recently; but even when poverty declined for the population as a whole, progress among indigenous people was much slower. Similar trends were seen in Bolivia, Ecuador, and Peru. Indeed, for most countries in Latin America with a sizable indigenous population, there is almost no progress in poverty reduction for the indigenous population—even in countries where the nonindigenous population is making progress.
By contrast, poverty rates, even among indigenous people, have declined sharply in emerging Asia—notably, China, India, and Vietnam (figure 4.1). China’s progress is remarkable: in 1981, it had one of the world’s highest proportions of people living in poverty—84 percent. By 2005, that had fallen to 16 percent.5 Even better, China’s ethnic minorities’ poverty rates declined faster than did the nonminority rates.
Progress is mixed in reducing the poverty of indigenous groups
poverty rate over time
Source: Hall and Patrinos 2011.Progress is mixed in reducing the poverty of indigenous groups
poverty rate over time
Source: Hall and Patrinos 2011.Progress is mixed in reducing the poverty of indigenous groups
poverty rate over time
Source: Hall and Patrinos 2011.Progress is mixed in reducing the poverty of indigenous groups
poverty rate over time
Source: Hall and Patrinos 2011.Progress is mixed in reducing the poverty of indigenous groups
poverty rate over time
Source: Hall and Patrinos 2011.India and Vietnam also showed significant declines: although not as dramatic as China’s, overall, the declines were much more robust for India’s minorities, scheduled castes, and tribes and for Vietnam’s ethnic minorities. But despite the rapid decline in poverty among indigenous people in Vietnam, their poverty rates remain much higher than those for the nonindigenous population; and the gap has widened slightly since the early 1990s.
Vulnerable groups receive less schooling and underperform in school
As with poverty, indigenous people are also disadvantaged in education, as in access to schooling (table 4.2). To varying degrees in nearly all countries, educational progress has lagged because of indigenous, low-income, gender, or disability status. Leaving such groups at the fringes of the education system will eventually impede a country’s ability to develop. Even in India, despite its success in rapidly reducing poverty for the whole population (including significantly for the scheduled tribes, or Adivasi [literally “original inhabitants”]), this minority population accounts for a fourth of the poorest wealth decile. Education indicators tell a similar story, with improvements but large and persistent differences. Scheduled tribe children lag far behind when it comes to educational attainment above the primary level.6
Years of schooling, latest year available
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007).
Years of schooling, latest year available
| Country | Indigenous | Nonindigenous |
|---|---|---|
| China/2002 | 8.2 | 8.9 |
| Peru/2001 | 6.4 | 8.7 |
| Mexico/2008 | 6.3 | 8.6 |
| Bolivia/2002 | 5.9 | 9.6 |
| Brazil/2002 | 4.6a | 6.6b |
| Ecuador/2006 | 4.3 | 6.9 |
| Guatemala/2006 | 2.5 | 5.7 |
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007).
Years of schooling, latest year available
| Country | Indigenous | Nonindigenous |
|---|---|---|
| China/2002 | 8.2 | 8.9 |
| Peru/2001 | 6.4 | 8.7 |
| Mexico/2008 | 6.3 | 8.6 |
| Bolivia/2002 | 5.9 | 9.6 |
| Brazil/2002 | 4.6a | 6.6b |
| Ecuador/2006 | 4.3 | 6.9 |
| Guatemala/2006 | 2.5 | 5.7 |
Refers to whites and “black/brown” (African origin).
Refers to whites (Telles 2007).
Not only do vulnerable groups receive less schooling; they also tend to achieve less. In addition to poor starting conditions, insufficient inputs, and ineffective teaching, vulnerable groups are likely to experience discrimination that manifests in different ways (including self-perceptions and stigma).
Stigma is sometimes internalized, as an experimental investigation into the impact of caste on test scores in India demonstrated.7 Children ages 11 and 12 were chosen at random from a low caste and three high castes and were given a series of puzzles to solve. When caste was not announced to the children, it had no bearing on the initial score or on the improvement in score registered in subsequent test rounds. But when caste was announced before the test, the scores for low-caste children fell dramatically, by 14-39 percent (figure 4.2). These findings confirm how much social identities are a product of history, culture, and personal experience, creating pronounced educational disadvantages through their effects on individual expectations.
Global evidence also shows how identity can limit development. A fractionalization data set compiled by Alesina and colleagues measures the degree of ethnic, linguistic, and religious heterogeneity in various coun-tries.8 It can be used to test the effects of fractionalization on the quality of institutions and economic growth. The data show just how heterogeneous many developing countries are, but they do not show that a high degree of heterogeneity produces negative outcomes. Fractionalization has been shown to slow the progress on many of the MDGs.9 Although fractionalization is not correlated with schooling progress, there is an indication of a negative effect of heterogeneity on test scores (figure 4.3).
Heterogeneity has a negative effect on test scores
Source: World Bank staff calculations using data from Alesina et al. 2003 and Schlotter 2010.Heterogeneity has a negative effect on test scores
Source: World Bank staff calculations using data from Alesina et al. 2003 and Schlotter 2010.Heterogeneity has a negative effect on test scores
Source: World Bank staff calculations using data from Alesina et al. 2003 and Schlotter 2010.Both the global and Indian examples underscore the need to be careful in developing appropriate learning materials and environments for minorities. Although some of the self-perceptions and stigma will take time to address, these more immediate lessons for promoting rights and accountabilities can be used now.
Indigenous peoples live shorter lives everywhere
The World Health Report 2010 states, “income is not the only factor influencing [health] service coverage. In many settings, migrants, ethnic minorities and indigenous people use [health] services less than other population groups, even though their needs may be greater.”10
Indigenous peoples therefore deserve special attention. They live shorter lives and are in worse health than their nonindigenous compatriots almost everywhere. In seven Latin American countries, the proportion of indigenous women receiving antenatal care or giving birth at health facilities was much lower than for nonindigenous women.11 Such inequality is one of the causes of the disparity in maternal health outcomes between the two populations. African-American women in the same countries also gave birth at health facilities less frequently and had poorer maternal health outcomes than other women.
Such health inequalities are not unique to low-income countries. Australia, Canada, and New Zealand show variations in access to health services between the two populations, frequently linked to distance and transportation costs. Some rich countries also show gaps in life expectancy (figure 4.4).
Both poor and rich countries show gaps in life expectancy
Source:UN 2009, p. 159.Note: The figure shows the difference in years of life expectancy between ndigenous and nonindigenous groups in each country.Both poor and rich countries show gaps in life expectancy
Source:UN 2009, p. 159.Note: The figure shows the difference in years of life expectancy between ndigenous and nonindigenous groups in each country.Both poor and rich countries show gaps in life expectancy
Source:UN 2009, p. 159.Note: The figure shows the difference in years of life expectancy between ndigenous and nonindigenous groups in each country.
Almost nine million children still die each year before they reach their fifth birthday
Source: World Bank staff calculations based on data from the World Development Indicators database.Almost nine million children still die each year before they reach their fifth birthday
Source: World Bank staff calculations based on data from the World Development Indicators database.Almost nine million children still die each year before they reach their fifth birthday
Source: World Bank staff calculations based on data from the World Development Indicators database.Further evidence of insufficient attention to the needs of minority groups is seen in Europe, where the largest and most disadvan-taged group, Roma, lags far behind the continent’s majority population (box 4.2).
Gender is a key dimension
Gender is a key dimension in the social indicators of indigenous or otherwise excluded groups. Although both boys and girls in these groups are significantly underrepresented in school and lag on other important indicators, girls are particularly disadvantaged. Disturbingly, out-of-school girls from indigenous populations make up a large portion of total population of out-of-school girls in developing countries.12
The Roma of Europe
The Roma, or “Gypsies,” are a Romani subgroup of about 7 million to 9 million people living primarily in Central and Eastern Europe. They are poorer than the general population: in some countries, Roma poverty rates are more than 10 times those of non-Roma.a And they are more likely to fall into poverty and stay poor. They also form a large proportion of youth and the potential labor force in these countries. But their low level of education means that European Union economies are losing hundreds of millions of euros annually in production and fiscal contributions.
Lower-bound annual estimates of productivity losses for Bulgaria, the Czech Republic, Romania, and Serbia total more than €2 billion, and there are similar estimates of €860 million in fiscal losses.b Using other Roma population estimates, the economic losses for the four countries combined amount to €5.7 billion annually, and the fiscal losses are €2 billion annually.c The annual fiscal gains from bridging the employment gap are much larger than the total cost of investing in public education for all Roma children—by factors of 7.7 in Bulgaria, 7.4 in the Czech Republic, 3.3 in Serbia, and 2.4 in Romania.
European governments and others have made a political commitment to improve Romani socioeconomic status and social inclusion under the aegis of the Decade of Roma Inclusion 2005-15. The initiative brings together governments and intergovernmental and nongovernmental organizations as well as Romani civil society to help ensure progress toward improving the welfare of Roma and to review such progress transparently and quantifiably. It commits governments to focus on education, employment, health, and housing while taking into account poverty, discrimination, and gender bias.
The Open Society Foundation documents the progress that countries are making at the halfway point for the 4.5 million Roma in participating countries.d Unfortunately, a lack of data remains the biggest obstacle in any assessment, as shown in the table on the next page. But even these patchy data allow one to see, for example, that the infant mortality rate among Roma is two or three times as high as that among the general population.
Roma progress on selected indicators
Percent of Roma living in households below $4.30 per day (purchasing power parity), expenditure based.
Roma progress on selected indicators
| Primary education | Infant mortality rate | |||||
|---|---|---|---|---|---|---|
| completion rate | (per 1,000 live births) | |||||
| Country | Year | Poverty (%)a | General (%) | Among Roma (%) | General | Among Roma |
| Albania | 2005 | Insufficient data | 14.0 | 16.0 | No data | |
| Most recent | 78 | Insufficient data | Insufficient data | 13.0 | No data | |
| Bosnia and Herzegovina | 2005 | Insufficient data | No data | Insufficient data | No data | |
| Most recent | 27 | Insufficient data | No data | 13.0 | No data | |
| Bulgaria | 2005 | Insufficient data | 28.3 | 10.4 | 25.0 | |
| Most recent | 46 | Insufficient data | 31.6 | 8.6 | 25.0 | |
| Croatia | 2005 | 95.0 | No data | 5.7 | 25.1 | |
| Most recent | 10 | 101.0 | No data | 4.5 | 11.5 | |
| Czech Republic | 2005 | 102.0 | No data | 3.4 | No data | |
| Most recent | 45 | 93.2 | No data | 2.7 | No data | |
| Hungary | 2005 | 95.0 | 76.4 | 6.2 | No data | |
| Most recent | 9 | 96.7 | No newer data | 5.6 | No data | |
| Macedonia, FYR | 2005 | 96.0 | 50.8 | 12.8 | Insufficient data | |
| Most recent | 33 | 92.0 | No data | 10.8 | Insufficient data | |
| Montenegro | 2005 | 91.1 | 9.2 | 9.5 | No data | |
| Most recent | 27 | No newer data | 20.0 | 7.5 | No data | |
| Romania | 2005 | No data | 31.8 | 15.0 | Insufficient data | |
| Most recent | 66 | No data | 19.8 | 11.0 | No data | |
| Serbia | 2005 | 95.0 | 22.7 | 8.0 | 25.0 | |
| Most recent | 57 | 99.5 | No data | 6.7 | No newer data | |
| Slovak Republic | 2005 | 94.0 | No data | 7.2 | No data | |
| Most recent | n.a. | No newer data | No data | 5.9 | No data | |
Percent of Roma living in households below $4.30 per day (purchasing power parity), expenditure based.
Roma progress on selected indicators
| Primary education | Infant mortality rate | |||||
|---|---|---|---|---|---|---|
| completion rate | (per 1,000 live births) | |||||
| Country | Year | Poverty (%)a | General (%) | Among Roma (%) | General | Among Roma |
| Albania | 2005 | Insufficient data | 14.0 | 16.0 | No data | |
| Most recent | 78 | Insufficient data | Insufficient data | 13.0 | No data | |
| Bosnia and Herzegovina | 2005 | Insufficient data | No data | Insufficient data | No data | |
| Most recent | 27 | Insufficient data | No data | 13.0 | No data | |
| Bulgaria | 2005 | Insufficient data | 28.3 | 10.4 | 25.0 | |
| Most recent | 46 | Insufficient data | 31.6 | 8.6 | 25.0 | |
| Croatia | 2005 | 95.0 | No data | 5.7 | 25.1 | |
| Most recent | 10 | 101.0 | No data | 4.5 | 11.5 | |
| Czech Republic | 2005 | 102.0 | No data | 3.4 | No data | |
| Most recent | 45 | 93.2 | No data | 2.7 | No data | |
| Hungary | 2005 | 95.0 | 76.4 | 6.2 | No data | |
| Most recent | 9 | 96.7 | No newer data | 5.6 | No data | |
| Macedonia, FYR | 2005 | 96.0 | 50.8 | 12.8 | Insufficient data | |
| Most recent | 33 | 92.0 | No data | 10.8 | Insufficient data | |
| Montenegro | 2005 | 91.1 | 9.2 | 9.5 | No data | |
| Most recent | 27 | No newer data | 20.0 | 7.5 | No data | |
| Romania | 2005 | No data | 31.8 | 15.0 | Insufficient data | |
| Most recent | 66 | No data | 19.8 | 11.0 | No data | |
| Serbia | 2005 | 95.0 | 22.7 | 8.0 | 25.0 | |
| Most recent | 57 | 99.5 | No data | 6.7 | No newer data | |
| Slovak Republic | 2005 | 94.0 | No data | 7.2 | No data | |
| Most recent | n.a. | No newer data | No data | 5.9 | No data | |
Percent of Roma living in households below $4.30 per day (purchasing power parity), expenditure based.
What works for Roma inclusion
Increasing the number of high school graduates would give the current generation of young Roma a better chance in the labor market and would improve school readiness to boost the chances of the next generation entering the classroom with the same skills as non-Roma. International experience suggests that investments in early childhood development and improvements in school attendance and completion are the most promising interventions to break the intergenerational transmission of social exclusion.e
Such interventions can be for supply (such as explicit desegregation efforts, teacher training, and school grants) and for demand (such as Roma mediators and conditional cash transfer programs).
Sub-Saharan Africa has about 23 million girls out of school, 75 percent from excluded groups; South Asia has 23 million and 67 percent; the Middle East and North Africa have almost 5 million and 90 percent; Eastern Europe and Central Asia have 1.5 million and 90 percent (including rural populations in Turkey); and Latin America has 1.5 million—almost all from excluded groups. In aggregate, 71 percent of out-of-school girls in developing countries are from excluded groups.13 In India, tribal women fare the worst: even among younger age cohorts (ages 15-21 in 2005), they attain an average of just four years of education, three years less than nontribal women.14 In Lao PDR, non-Lao-Tai women receive significantly less education than non-Lao-Tai men or Lao-Tai of both sexes. For example, 34 percent of rural non-Lao-Tai girls had never attended school in 2002-03, compared with 6 percent of Lao-Tai girls.15
Indigenous groups and the MDGS
Some countries that are off track to reach one or more of the MDGs have large indigenous populations, although their presence may not be the only reason for this shortfall. Even some on-track countries have these populations; thus, when data are disaggregated, they are off track in the sense that although the country may reach all (or most) of the MDGs, but has significant segments of the population that are off track (or seriously off track). Indigenous people get little attention in country reports designed to monitor progress in meeting the goals (box 4.3).
Off-track countries—minority groups are lagging severely
Many of the off-track countries are in Africa.16 And some of them are off track because of the severe lack of progress of their minority groups. Nigeria, for example, has three major tribes—the Igbo, Hausa, and Yoruba—more than 200 ethnic groups, and more than 500 indigenous languages and dialects. It would be important to use disaggregated data on these subgroups to investigate the link between minority groups and the failure to achieve the MDGs—not just for Nigeria, but more generally.
Vulnerable groups deserve more attention in MDG country reports
A review of 50 MDG country reports found that ethnic or linguistic minorities are mentioned in only 19, and the inequalities experienced by religious minorities are mentioned in only 2. An additional 10 reports mention only indigenous peoples, without identifying any other minority groups.
The mentions of minorities vary widely, with some reports providing a good range of information and disaggregated data under several MDGs. In other cases, minorities are mentioned only in the background section describing the national population and without particular attention to their situation in relation to the MDGs. Minorities are mentioned most frequently in connection with Goal 2 on universal primary education.
Attention to indigenous peoples, in general, is significantly higher than attention to nonindigenous marginalized minorities. Although attention to gender issues in many MDG country reports is positive, there is very little consideration of discrimination experienced by minority women or of targeted policies for them.
Minorities are virtually absent from the MDG country reports from donor countries. None of those reports considers minorities under any of the eight goals, and only the reports for Nepal and Vietnam give sufficient attention to indigenous peoples. Nor do any reports indicate any consultation with indigenous peoples or consistently provide disaggregated data for them.
Sources:UN 2010a; Hartley 2008.Demographic and health surveys show the frequent considerable disadvantage in belonging to an indigenous group in some low-income African countries (table 4.3). In Cameroon, the Peulh and the Fulfulde (a large minority) are doing poorly on adult literacy and net primary enrollment. Niger’s Peul and Tuareg face inequalities, even though the national population is far off track. Mali is nationally so far off track that there is seeming equality, with all groups performing poorly.
Status of indigenous groups relative to the MDGs, selected African countries
Status of indigenous groups relative to the MDGs, selected African countries
| Under-five mortality | Water deprivation | Nutrition deprivation | Literacy rate | Net primary enrollment | |||||
|---|---|---|---|---|---|---|---|---|---|
| Country/data year/group | (per 1,000 live births) | (% of households) | (% of children) | (% of males) | (% of females) | (% of boys) | (% of girls) | Households in sample (n) | |
| Cameroon/2004 | 148 | 29 | 13 | 80 | 65 | 81 | 78 | 10,462 | |
| Peulh | 170 | 21 | 11 | 43 | 19 | 60 | 55 | 387 | |
| Pygmy | 160 | 54 | 54 | 17 | 10 | 65 | 100 | 18 | |
| Fulfulde | 192 | 26 | 18 | 27 | 13 | 57 | 49 | 1,632 | |
| Niger/2006 | 218 | 28 | 29 | 28 | 12 | 44 | 29 | 7,660 | |
| Peul | 204 | 38 | 33 | 5 | 5 | 17 | 19 | 350 | |
| Tuareg | 204 | 52 | 30 | 17 | 5 | 30 | 29 | 635 | |
| Mali/2006 | 215 | 6 | 16 | 37 | 17 | 46 | 39 | 12,998 | |
| Peulh | 229 | 4 | 16 | 31 | 16 | 38 | 35 | 1,400 | |
| Tanachek | 194 | 20 | 17 | 28 | 11 | 41 | 28 | 815 | |
| Guinea/2005 | 188 | 59 | 15 | 44 | 16 | 58 | 49 | 6,282 | |
| Peulh | 171 | 61 | 12 | 40 | 14 | 56 | 45 | 2,228 | |
| Senegal/2005 | 135 | 11 | 5 | 54 | 35 | 58 | 58 | 7,407 | |
| Poular | 137 | 13 | 8 | 44 | 29 | 56 | 56 | 1,990 | |
Status of indigenous groups relative to the MDGs, selected African countries
| Under-five mortality | Water deprivation | Nutrition deprivation | Literacy rate | Net primary enrollment | |||||
|---|---|---|---|---|---|---|---|---|---|
| Country/data year/group | (per 1,000 live births) | (% of households) | (% of children) | (% of males) | (% of females) | (% of boys) | (% of girls) | Households in sample (n) | |
| Cameroon/2004 | 148 | 29 | 13 | 80 | 65 | 81 | 78 | 10,462 | |
| Peulh | 170 | 21 | 11 | 43 | 19 | 60 | 55 | 387 | |
| Pygmy | 160 | 54 | 54 | 17 | 10 | 65 | 100 | 18 | |
| Fulfulde | 192 | 26 | 18 | 27 | 13 | 57 | 49 | 1,632 | |
| Niger/2006 | 218 | 28 | 29 | 28 | 12 | 44 | 29 | 7,660 | |
| Peul | 204 | 38 | 33 | 5 | 5 | 17 | 19 | 350 | |
| Tuareg | 204 | 52 | 30 | 17 | 5 | 30 | 29 | 635 | |
| Mali/2006 | 215 | 6 | 16 | 37 | 17 | 46 | 39 | 12,998 | |
| Peulh | 229 | 4 | 16 | 31 | 16 | 38 | 35 | 1,400 | |
| Tanachek | 194 | 20 | 17 | 28 | 11 | 41 | 28 | 815 | |
| Guinea/2005 | 188 | 59 | 15 | 44 | 16 | 58 | 49 | 6,282 | |
| Peulh | 171 | 61 | 12 | 40 | 14 | 56 | 45 | 2,228 | |
| Senegal/2005 | 135 | 11 | 5 | 54 | 35 | 58 | 58 | 7,407 | |
| Poular | 137 | 13 | 8 | 44 | 29 | 56 | 56 | 1,990 | |
Groups left behind in countries otherwise performing well
Indigenous people in on-track middle-income countries are also being left behind. A few examples will suffice. Mexico, a middle-income country and an Organisation for Economic Co-operation and Development member, is on track to meet most of the MDGs. But under-five mortality for indigenous groups is at 52 per 1,000 live births—almost twice the 27 for nonindigenous groups. In addition, the poorest municipalities, heavily populated by indigenous peoples, have poor indicators for poverty and access to education and health. The worst-performing municipalities have a Human Development Index similar to those in poorer African countries: the community with the lowest index in Mexico, the Metlatonoc, is at about the same level as Angola and Malawi. Perhaps it is not surprising that 98 percent of the population there is indigenous.17
India is also on track to meet the MDGs, but the scheduled tribes are largely off track. Although educational enrollment rates are fairly high for all groups, scheduled tribes suffer from high under-five mortality, poor access to water and sanitation, nutrition deprivation, and low levels of adult literacy.18 Ecuador’s under-five mortality is almost twice as high among indigenous people, at 138 per 1,000 live births versus 77 per 1,000 for nonindigenous groups.19 Vietnam, a star performer in economic growth and poverty reduction, does very well on the MDGs overall and its ethnic minorities are generally progressing, although some are seriously off track for adult literacy.20 Although indigenous groups in Vietnam have seen rapid declines in poverty, their poverty rates remain some 40 percentage points higher than those for the rest of the population.
Lessons from progress—and from stagnation
As we approach 2015, the daunting challenge of reaching the last of the excluded groups becomes apparent. They have a tenuous connection to the modern economy, likely speak a language different from the majority, and typically are an ethnic minority. Where such groups make up a sizable proportion of the population, the solution requires rapid, broad-based economic growth and poverty reduction. But even that may not be enough to address multiple sources of disadvantage from centuries of political and social exclusion. Instead, a direct effort at empowerment is required, using targeted approaches and enforcing political and social rights.
Growth is essential for poverty reduction
In some Asian countries, rapid poverty reduction and the relative gains of minority groups were generated by rapid growth and fairly low inequality in access to productive inputs such as land and human capital.21
From the 1970s’ reforms to 1985, China’s main strategy to reduce poverty was to focus on rural economic growth. After 1985, however, the government, recognizing that growth was necessary but insufficient to reduce poverty, adopted a development-oriented strategy, providing assistance through various programs in poor regions.22 Nonetheless, direct redistributive interventions have not been prominent in China’s efforts to reduce poverty.23
India’s strong growth—focused since the 1980s on moving from a state-controlled, inward-looking economy to an outward-oriented, market-led economy—has not been damped by its poorly performing targeted programs.24 Similarly, Vietnam’s remarkable gains in economic growth and poverty reduction were the product of market reforms and a loosening of state control.25 It was only after growth was established that policy makers turned their attention to poverty reduction strategies and ethnic minority development;26 and, even then, such programs focused on developing local economic activities rather than targeting the poor, as in Latin America.
Latin American countries with large indigenous populations also have had some success in equalizing human capital and expanding access to basic services, starting from very unequal conditions. Indigenous people often have political representation, and their rights are generally well established. All Latin American countries but one have national bilingual education programs,27 some of which may benefit indigenous populations (box 4.4). And 17 have conditional cash transfer (CCT) programs.28
Despite these efforts, slow growth has hampered progress in overall poverty reduction and more equitable income distribu-tion.29 The Latin American countries with the largest indigenous populations experienced per capita growth rates of no more than 1 percent in the 1990s and not much more since then. High inequality, slow growth, and little participation by indigenous peoples in the economic progress that did occur meant that indigenous peoples made little progress in poverty reduction. Most distributive programs, as currently designed and implemented, have not reduced poverty among indigenous people. One significant exception is Mexico’s CCT program Oportunidades, an important part of the country’s antipoverty program that has benefited indigenous people disproportionately (box 4.5). It is characterized as well targeted (to the poor, among whom the indigenous in Mexico are overrepresented), transparent, and evidence based—with the results of impact evaluations used to adjust and improve it.
Bilingual education in Guatemala
Bilingual education is an efficient public investment. According to a crude cost-benefit exercise, a shift to bilingual schooling for indigenous people in Guatemala would result in considerable cost savings because of reduced grade repetition. The higher quality of education would probably help students complete the primary education cycle and substantially increase total education levels at lower cost.
The cost savings from bilingual education are estimated at $5 million, equal to the cost of providing primary schooling to 100,000 students a year. A reduction in the number of dropouts and the effect on personal earnings would thus be significant.a
a.Patrinos and Velez 2009.
For girls in some African countries, secondary education remains elusive
Source: World Bank staff calculations based on data from the World Development Indicators database.For girls in some African countries, secondary education remains elusive
Source: World Bank staff calculations based on data from the World Development Indicators database.For girls in some African countries, secondary education remains elusive
Source: World Bank staff calculations based on data from the World Development Indicators database.Examples of successful targeted programs
CCT programs have become very popular for attacking poverty. They provide assistance conditioned on the beneficiary’s actions—that is, the government provides money only to beneficiaries who fulfill certain criteria, such as enrolling children in school, ensuring that they attend regularly and complete grades, and receiving regular medical attention. CCTs have become the largest social assistance program in some countries, covering millions of households—for example, in Brazil (Bolsa Familia) and Mexico.a
The program in Mexico—PROGRESA (now Oportunidades)—started early (1997) and evolved thoughtfully. The program produced successive waves of data to evaluate its impact and put the data in the public domain. Hundreds of studies attest to its success in confronting poverty and disadvantage. Although targeted generally to the poor, the program has improved indigenous children’s schooling attainment relative to Spanish-speaking or bilingual children (see figure at right). Before the program, indigenous children had lower school attainment than did Spanish-speaking or bilingual children. After three years of implementation, school attainment among indigenous children increased, reducing the gap. Ecuador’s CCT—Bono de Desarrollo Humano—also recorded success in reaching indigenous populations.b Cambodia’s Education Sector Support Project has a CCT that targets ethnic minorities. Exported from the developing world to the developed, CCTs have also been used in New York City and for indigenous communities in Australia.
average years of schooling, 8- to 12-year-olds
Sources:Bando, Lopez-Calva, and Patrinos 2005; Godoy et al. 2005, 2006.average years of schooling, 8- to 12-year-olds
Sources:Bando, Lopez-Calva, and Patrinos 2005; Godoy et al. 2005, 2006.average years of schooling, 8- to 12-year-olds
Sources:Bando, Lopez-Calva, and Patrinos 2005; Godoy et al. 2005, 2006.The differing experiences in China and Latin America underline the importance of growth for poverty reduction.30 Considerable reductions in poverty can be achieved through rapid growth, at least at first. But given the typical widening of inequality with rapid development (as in China and India), there is a desperate need to design effective programs that target underprivileged groups. This is one area where Asia can learn from the successes of such programs in Latin America (including Oportunidades and a similar program in Brazil)31 that combine careful identification of beneficiary groups, transparent program delivery, and learning from experience.
Empowering indigenous people can improve their development prospects
To properly empower vulnerable groups so that broad-based growth and targeted poverty reduction programs work in their favor, the groups need greater political representation and greater voice. Political representation in legislatures or parliaments, for instance, could significantly increase their say in the design of programs and delivery of services. It could also ensure that they are referred to in plans to achieve the MDGs and that they are consulted in the development and design of national programs.
Although beneficial, effecting change through the political process takes time, as made evident in Latin America. Empowering stakeholders directly—for example, enabling parents to influence school policies or patients to choose health providers—holds more immediate promise. In Mexico, parents are given the right to exercise voice and decision-making authority over the use of small sums of money for improving schooling in rural areas. These programs target disadvantaged and poorly performing schools, including a substantial number that provide services to indigenous groups. A randomized evaluation of one such program shows that parental pressure produced efficiency gains, increasing access and improving test scores.32
To work, such empowerment must be accompanied by real efforts to generate, use, and disseminate disaggregated data and information on vulnerable groups—something that can be done well only to the extent that those groups play a role in conceptualizing and implementing the data gathering, and in formulating policy.
Although much of the economic disadvantage of indigenous and excluded groups is the result of lower human capital, there is also evidence of labor market discrimi-nation.33 In decompositions of the earnings gap, the portion of the difference between indigenous and nonindigenous peoples that is “unexplained”—perhaps because of discrimination or other unidentified factors—represents one quarter to more than one half of the total differential. This means that although about half the earnings differential can be influenced by improvements in human capital (education, skills, and abilities that an indigenous person brings to the labor market), another half may result from discriminatory labor market practices or other factors over which the indigenous person has little control. So efforts to empower excluded groups and increase their human capital need to look at how the labor market can be made more fair for them.
Small and largely untouched groups
Indigenous peoples across the globe have very different exposures to nonindigenous peoples and varying degrees of cooperation with governments and government programs. It is hard to design programs that would benefit those who are least exposed to a modern national economy and allow them to maintain their identity and choice of livelihood. The prime consideration of such programs is the need to tread lightly and be well informed by on-the-ground research.
Indigenous peoples have dealt with the expansion of western influence and market systems in at least two broad ways: some have confronted the expansion and either adapted to it or been deracinated, and others have retreated farther into the backlands to avoid exposure and deal with the modern world at arm’s length.34 We know a great deal about the vulnerability and well-being of the first group, but little about the second.35 Nor is it clear what role, if any, the government should play in enhancing the well-being of people who eschew the modern economy. Two examples illuminate the issues.
Persistent poverty in Central Africa
Among the oldest inhabitants of Africa, the Pygmy peoples lead a nomadic or semino-madic lifestyle (depending on the country) that has persisted largely unchanged for thousands of years, with livelihoods based on hunting, fishing, and gathering wild fruits and nuts. Because of urbanization and deforestation, however, many Pygmies have become sedentary, leaving them disadvantaged and vulnerable. In the Democratic Republic of Congo, for example, Pygmies have traditionally been closely attached to the rain forest—the source of their spirituality, livelihood, and protection. They once lived in camps of 30-40 families, maintaining regular links with each other. But today their lifestyle is in danger, as they become more sedentary, lose access to the forest, and face a deteriorating relationship with Bantu farmers. Census data for the Central African Republic and Gabon show that Pygmies lag behind the general population in wealth, education, and access to basic infrastructure.
Their poverty rates are extremely high. In the Democratic Republic of Congo, 85 percent of the Pygmy population is estimated to be poor, compared with 72 percent of the general population.36 Unemployment is typically lower among Pygmies, perhaps because they are so poor that they cannot afford not to work. But the share of workers who are not paid cash for their work is much higher, suggesting low-productivity jobs.
School enrollment rates and average years of schooling among Pygmies in the Central African Republic and Gabon are dramatically lower than among other groups. The Central African Republic’s gross primary enrollment is only 21 percent for Pygmies, compared with 73 percent for nonindigenous peoples; Pygmy men average 0.3 years of schooling and women average 0.1 years, compared with 2.8 years for non-Pygmy men and 1.4 years for non-Pygmy women. Gabon’s gross secondary enrollment is only 4 percent for Pygmies, but 80 percent for non-Pygmies; and the average years of schooling are 3.0 for men and 2.8 for women among Pygmies, compared with 6.5 years for both non-Pygmy men and women. Lower school enrollment among Pygmies may result from high cost, distance to schools, weaker cultural emphasis, and the need for children to work.
Pygmies in the Central African Republic and Gabon are less likely to have access to safe water and electricity; more likely to live in a dwelling with walls, roof, or floor made of temporary materials; and less likely to have adequate sanitation. Even compared with other rural households, Pygmies in Gabon fare worse in all housing dimensions.
In the Central African Republic, the share of households living in a village or area with a health facility is smaller for Pygmies than for the population as a whole, although the differences in distances to health facilities are small. Access to condoms is much lower for Pygmies; and the share of households that lost a member to a long-term illness is higher among Pygmies than among the overall population (although the share of households that have lost a member to HIV/AIDS is smaller).
The Pygmies embody a valuable cultural heritage that should be protected and preserved, one of the most original forms of human adaptation to the ecology of the rain forest. They have a sophisticated knowledge of their environment and the possibilities for humans to adapt to it sustainably. They are increasing their integration into broader society as they become more sedentary, and this process—unmanaged and with little input from the Pygmies—is linked to their impoverishment, exploitation, and poor health and education outcomes. Although these findings do not point to specific policies that the government could implement to improve Pygmies’ living conditions, they do suggest the need for more qualitative and institutional analysis of measures to target this especially vulnerable group.
Lack of interaction with wider society: the case of the Tsimane’ of Bolivia
The Tsimane’, a native Amazonian society of foragers and farmers in Bolivia, provide an intriguing example of the benefits and costs when an indigenous group exercises significant autonomy in its exposure to the outside world. Active withdrawal into the backlands and collaboration with Protestant missionaries since the late 1940s have enabled Tsimane’ to retain a great deal of their traditional culture and territory.37 They show no change in core cultural values38 or in ethnobotanical knowledge,39 although a more recent unpublished study by Reyes-García et al.40 suggests they may be losing that knowledge at the rate of 2 percent a decade.
Close to 75 percent of the population follows the traditional preferential system of cross-cousin marriage (that is, a man marries his mother’s brother’s daughter or father’s sister’s son),41 which produces a tight endogamic group. With ample access to farmlands, Tsimane’ daily personal income (adjusted for inflation and in purchasing power parity) reaches $9 a day per person, much higher than the international poverty line of $1-$2 a day.42 And, over the period 2002-06, many indicators of individual well-being improved, including real income, real wealth, consumption, body-mass index, social capital, and happiness.43 Little wonder that permanent outmigration is rare; but they are being encroached on by highland colonist farmers, cattle ranchers, and logging and oil firms.44
Tsimane’ experience little damage in physical stature,45 widespread parasitism46 and child growth stunting, and vulnerability to adverse weather conditions (such as excessive rainfall). Such events during gestation, infancy, and early childhood suppress adult physical stature, particularly among women.47 And when adverse shocks strike households (illness, theft, crop loss, floods), people weather them on their own, without help from kith, kin, or formal institutions. That suggests the Tsimane’ remain poorly insured against mishaps that strike either the individual or the village.
Conclusions
Indigenous groups make up a sizable share of the world’s people and even a greater share of the world’s poor. But their isolation reduces their gains from overall growth and makes them difficult to reach through targeted poverty programs. As we come closer to 2015, reaching even more isolated people will become progressively more difficult. Addressing their needs will require widespread and sustainable economic growth and poverty reduction, along with strategies to reach those who need a special lift. Helping indigenous groups will require better understanding of their needs and of how they interact with the modern economy.
Nevertheless, some programs have improved their lot. For example, well-implemented bilingual education programs promote school completion and subsequent earnings gains, thus contributing to multiple goals; and these programs can be cost effective. CCT programs can promote schooling, health, and poverty reduction goals; and in countries with large vulnerable populations, CCTs show disproportionate gains for the minority population. Empirical research suggests that investments in early childhood development and improving school attendance and completion are the most promising means of breaking the intergenerational transmission of social exclusion.
The international community needs a knowledge base of what specific programs and policies work best, and for which vulnerable populations. It is widely held that outright discrimination may explain some of the observed differential in poverty outcomes among minority groups,48 and there is evidence consistent with labor market discrimination against indigenous peoples.49 But overt tests for discrimination against indigenous peoples (such as tests in the United States comparing call-back rates for blacks and whites with otherwise similar profiles)50 are distinctly lacking.
To properly address the needs of the world’s excluded populations, more and better data are needed. That is, one needs to analyze disaggregated data that are collected regularly and consistently. This data collection and analysis effort can only be done successfully to the extent that the minority populations play a role, thus ensuring coverage and relevance.
Extending the gains from development to indigenous people will not be easy. It will be necessary to continue progress in overall poverty reduction; to overcome physical, institutional, and social obstacles to their participation in broader society; and to empower them by supporting their political and social rights.
References
Alesina, A., A. Devleeschauwer, W. Easterly, S. Kurlat, and R. Wacziarg. 2003. “Fractionalization.” Journal of Economic Growth 8: 155–94.
Bando, R., L. F. Lopez-Calva, and H. A. Patrinos. 2005. “Child Labor, School Attendance, and Indigenous Households: Evidence from Mexico.” Policy Research Working Paper 3487, World Bank, Washington DC.
Baulch, B., T.T.K. Chuyen, D. Haughton, and J. Haughton. 2007. “Ethnic Minority Development in Vietnam.” Journal of DevelopmentStudies 43 (7): 1151–76.
Becker, G. S. 1971. The Economics of Discrimination. 2nd edition. Chicago: University of Chicago Press.
Ben-Achour, A., P. Backiny-Yetna, and Q. Wodon. 2011. “Socioeconomic Status of the Pygmies in the Democratic Republic of Congo.” Unpublished manuscript, Human Development Network, World Bank, Washington, DC.
Bertrand, M., and S. Mallainathan. 2003. “Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” Working Paper 9873, National Bureau of Economic Research, Cambridge, MA.
Borja-Vega, C., T. Lunde, and V. Garcia-Moreno. 2007. “Economic Opportunities for Indigenous People in Mexico.” Background paper for the study Economic Opportunities for Indigenous Peoples in Latin America, Conference Edition. World Bank, Washington, DC.
Brosnan, P. 1984. “Age, Education and Maori-Pakeha Income Differences.” New ZealandEconomic Papers 18: 49–61.
Daley, A. E., and J. Lui. 1995. “Estimating the Private Rate of Return to Education for Indigenous Australians.” Discussion Paper 97, Centre for Aboriginal Economic Policy Research, Australian National University, Canberra.
Dang, H.-A. H. 2010. “Growth in Vietnam Is Strong But Not Shared Equitably across Ethnic Groups.” Indigenous Peoples Country Brief (Vietnam) No. 1, World Bank, Washington, DC.
Das, M. B., G. Hall, S. Kapoor, and D. Nikitin. 2010. “India’s Adivasis.” Indigenous Peoples Country Brief (India) No. 4, World Bank, Washington, DC.
Das, M. B., G. Hall, S. Kapoor, and D. Nikitin. Forthcoming. “India: The Scheduled Tribes” In Indigenous Peoples, Poverty and Development, ed. G. Hall and H. A. Patrinos. Cambridge, UK: Cambridge University Press.
de Laat, J. 2010. “Economic Costs of Roma Exclusion.” Europe and Central Asia Topic Brief, Human Development Department, World Bank, Washington DC.
Dollar, D., P. Glewwe, and J. Litvack, eds. 1998. Household Welfare and Vietnam’s Transition. Washington, DC: World Bank.
Dollar, D., and A. Kraay. 2002. “Growth Is Good for the Poor.” Journal of Economic Growth 7 (3): 195–225.
Fiszbein, A., and N. Schady. 2009. ConditionalCash Transfers: Reducing Present and FuturePoverty. Washington DC: World Bank.
Garcia-Moreno, V., and H. A. Patrinos. Forthcoming. “Education and Poverty in Mexico: Indigenous Peoples.” Teachers College, Columbia University, New York; and Human Development Department, World Bank, Washington, DC.
Gertler, P., H. A. Patrinos, and E. Rodríguez-Oreggia. 2010. “Parental Empowerment in Mexico: Randomized Experiment of the Apoyos a la Gestión Escolar (AGE) in Rural Primary Schools in Mexico: Preliminary Findings.” Unpublished manuscript, Human Development Network, World Bank, Washington, DC.
Godoy, R., D.T.A. Eisenberg, V. Reyes-García, T. Huanca, W. R. Leonard, T. W. McDade, and S. Tanner. 2008. “Assortative Mating and Offspring Well-Being: Theory and Empirical Findings from a Native Amazonian Society in Bolivia.” Evolution and Human Behavior 29: 201–10.
Godoy, R., W. R. Leonard, V. Reyes-Garcia, E. Goodman, T. McDade, T. Huanca, S. Tanner, and V. Vadez. 2006. “Physical Stature of Adult Tsimane’ Amerindians, Bolivian Amazon, in the 20th Century.” Economics and HumanBiology 4: 184–205.
Godoy, R., V. Reyes-García, J. Broesch, I. C. Fitzpatrick, P. Giovannini, M. R. M. Rodríguez, T. Huanca, W. R. Leonard, T. W. McDade, and S. Tanner. 2009. “Long-Term (Secular) Change of Ethnobotanical Knowledge of Useful Plants: Separating Cohort and Age Effects.” Journalof Anthropological Research 65 (1): 51–67.
Godoy, R., V. Reyes-García, C. C. Gravlee, T. Huanca, W. R. Leonard, T. W. McDade, and S. Tanner. 2009. “Moving Beyond a Snapshot to Understand Changes in the Well-Being of Native Amazonians.” Current Anthropology 50 (4): 563–73.
Godoy, R., V. Reyes-García, T. Huanca, W. R. Leonard, V. Vadez, C. Valdés-Galicia, and D. Zhao. 2005. “Why Do Subsistence-Level People Join the Market Economy? Testing Hypotheses of Push and Pull Determinants in Bolivian Amazonia.” Journal of AnthropologicalResearch 61 (2): 157–78.
Godoy, R., S. Tanner, V. Reyes-García, W. R. Leonard, T. W. McDade, M. Vento, J. Broesch, I. C. Fitzpatrick, P. Giovannini, T. Huanca, and N. Jha. 2008. “The Effect of Rainfall during Gestation and Early Childhood on Adult Height in a Foraging and Horticultural Society of the Bolivian Amazon.” American Journal ofHuman Biology 20 (1): 23–34.
Hall, G., and H. A. Patrinos, eds. 2006. IndigenousPeoples, Poverty and Human Development inLatin America. London: Palgrave Macmillan.
Hall, G., and H. A. Patrinos, 2011. Indigenous Peoples, Poverty and Development. Cambridge, UK: Cambridge University Press.
Hartley, B. 2008. “MDG Reports and Indigenous Peoples: A Desk Review.” Report prepared for the Secretariat of the United Nations Permanent Forum on Indigenous Issues, New York.
Heckman, J. J. 2006. “Skill Formation and the Economics of Investing in Disadvantaged Children.” Science 312: 1900–02.
Hoff, K., and P. Pandey. 2004. “Belief Systems and Durable Inequalities: An Experimental Investigation of Indian Caste.” Policy Research Working Papers 3351, World Bank, Washington DC.
Huanca, T. 2008. Tsimane’ Oral Tradition, Landscape, and Identity in Tropical Forest. La Paz, Bolivia: Imprenta Wagui.
Kimmel, J. 1997. “Rural Wages and Returns to Education: Differences between Whites, blacks, and American Indians.” Economics ofEducation Review 16 (1): 81–96.
King, E. M., and D Van de Walle. Forthcoming. “Laos: Ethno-Linguistic Diversity and Disadvantage.” in Indigenous Peoples, Poverty and Development, ed. G. Hall and H. A. Patrinos. Cambridge, UK: Cambridge University Press.
Kremer, M., E. Miguel, and R. Thornton. 2009. “Incentives to Learn.” Review of Economicsand Statistics 91 (3): 437–56.
Lanjouw, P., and M. Ravallion. 1999. “Benefit Incidence, Public Spending Reforms, and the Timing of Program Capture.” World BankEconomic Review 13 (2): 257–73.
Lewis, M. A., and M. E. Lockheed. 2006. InexcusableAbsence: Why 60 Million Girls Still Aren’tin School and What to Do about It. Washington DC: Center for Global Development.
MacDonald, K. Forthcoming. “Indigenous Peoples and Development Goals: A Global Snapshot.” in Indigenous Peoples, Poverty andDevelopment, ed. G. Hall and H. A. Patrinos. Cambridge, UK: Cambridge University Press.
Nopo, H., A. Chong, and A. Moro, eds. 2010. Discrimination in Latin America: An EconomicPerspective. Washington DC: World Bank.
Ohenjo, N., R. Willis, D. Jackson, C. Nettleton, K. Good, and B. Mugarura. 2006. “Health of Indigenous People in Africa.” Lancet 367: 1937–46.
OSI (Open Society Foundation). 2010. No Data—No Progress: Country Findings. Data Collectionin Countries Participating in the Decadeof Roma Inclusion 2005–2015. New York.
Parodi C. T., S. Muñoz, and A. Sanhueza. 2010. “Acceso y gasto de salud para grupos étnicos/ raciales en la región de las Américas.” WorldHealth Report 2010 background paper, no. 46. http://www.who.int/healthsystems/topics/fi nancing/healthreport/whr_background/en.
Patrinos, H. A. 2007. “Living Conditions of Children.” In Solutions for the World’s Biggest Problems: Costs and Benefits, ed. B. Lomborg, 358–75. Cambridge, UK: Cambridge University Press.
Patrinos, H. A., and C. N. Sakellariou. 1992. “North American Indians in the Canadian Labour Market: A Decomposition of Wage Differentials.” Economics of Education Review 2(3): 257–66.
Patrinos, H. A., E. Skoufias, and T. Lunde. 2010. “Social Networks among Indigenous Peoples in Mexico.” Latin American Research Review 45 (2): 49–67.
Patrinos, H. A., and E. Velez. 2009. “Costs and Benefi ts of Bilingual Education in Guatemala: A Partial Analysis.” International Journal ofEducational Development 29 (6): 594–98.
Psacharopoulos, G., and H. A. Patrinos, eds. 1994. Indigenous People and Poverty in LatinAmerica: An Empirical Analysis. Washington, DC: World Bank.
Ravallion, M. 2010. “A Comparative Perspective on Poverty Reduction in Brazil, China and India.” World Bank Research Observer 26 (1): 71–104.
Reyes-García, V., E. Kightley, I. Ruiz-Mallén, N. Fuentes-Peláez, K. Demps, T. Huanca, and M. Martínez-Rodríguez. 2009. “Schooling and Local Environmental Knowledge: Do They Complement or Substitute Each Other?” Unpublished manuscript, Tsimane’ Amazonian Panel Study, Brandeis University, Waltham, MA.
Reyes-García, V., M. Orta-Martínez, M. Gueze, A. C. Luz, J. Paneque-Gálvez, M. J. Macía, J. Pino, J. Cristobal, and TAPS Bolivian Study Team. 2010. “Does Participatory Mapping Increase Conflict? A Randomized Experimental Evaluation in the Bolivian Amazon.” TAPS working paper 59. http://tsimane.org/research/pgs/workingpapers.html.
Ringhofer, L. 2009. Fishing, Foraging and Farmingin the Bolivian Amazon: On a Local Societyin Transition. New York: Springer.
Ringold, D., M. A. Orenstein, and E. Wilkens. 2005. Roma in an Expanding Europe: Breakingthe Poverty Cycle. Washington DC: World Bank.
Rubio, K., E. Undurraga, O. Magvanjav, C. Gravlee, T. Huanca, W. R. Leonard, T. W. McDade, V. Reyes-García, S. Tanner, R. Godoy, and TAPS Team. 2009. “Modernization and Culture Loss: A Natural Experiment among Native Amazonians in Bolivia.” Tsimane’ Amazonian Panel Study Working Paper. http://tsimane.org/research/pgs/workingpapers.html.
Schlotter, M. 2010. “The Relationship between Human Capital Quality, Governance and Educational Institutions: First Steps towards a Panel Approach.” Unpublished manuscript, Human Development Network, World Bank, Washington, DC.
Scott, J. 2009. The Art of Not Being Governed:An Anarchist History of Upland SoutheastAsia. Yale Agrarian Studies Series. New Haven, CT: Yale University Press.
Tanner, S., W. R. Leonard, T. W. McDade, V. Reyes-García, R. Godoy, and T. Huanca. 2009. “Influence of Helminth Infections on Childhood Nutritional Status in Lowland Bolivia.” American Journal of Human Biology 21: 651–56.
Telles, E. E. 2007. “Race and Ethnicity and Latin America’s United Nations Millennium Development Goals.” Latin American and CaribbeanEthnic Studies 2 (2): 185–200.
UN (United Nations). 1981. “Study of the Problem of Discrimination against Indigenous Populations.” Final Report submitted by the Special Rapporteur, Jose R. Martinez Cobo, United Nations Economic and Social Council, Commission on Human Rights, Sub-Commission on Prevention of Discrimination and Protection of Minorities, 34th session, E/CN.4/ Sub.2/476, New York.
UN (United Nations). 2009. State of the World’s IndigenousPeoples. New York.
UN (United Nations). 2010a. Human Rights and the MillenniumDevelopment Goals in Practice: AReview of Country Strategies and Reporting. New York.
UN (United Nations). 2010b. The Millennium DevelopmentGoals Report. New York.
UNESCO (United Nations Educational, Scientific, and Cultural Organization). 2010. Educationfor All Global Monitoring Report 2010:Reaching the Marginalized. Oxford, UK: Oxford University Press.
WHO (World Health Organization). 2010. WorldHealth Report 2010: Health Systems Financing:The Path to Universal Coverage. Geneva, Switzerland.
World Bank. 2009. “From Poor Areas to Poor People: China’s Evolving Poverty Reduction Agenda. An Assessment of Poverty and Inequality in China. Washington, DC.
World Bank. 2010. “Roma Inclusion: An Economic Opportunity for Bulgaria, Czech Republic, Romania and Serbia.” Policy Note, Human Development Sector Unit, Europe and Central Asia Region, Washington, DC.
In formal terms, “indigenous” is more suitable for describing socially excluded groups in the western hemisphere and other areas of mass migration than in areas where majority groups have been in place for many centuries. However, “indigenous peoples” has been used generally to refer to socially excluded groups, and we will continue that usage here.
Hall and Patrinos 2011.
Ibid 2011. This gap is expressed as the money required to raise the poor from their present income to the poverty line, as a proportion of the poverty line, and averaged over the total population.
Garcia-Moreno and Patrinos forthcoming.
Ravallion 2010.
Das et al. forthcoming.
Hoff and Pandey 2004; UNESCO 2010.
For example, Alesina et al. 2003.
Lewis and Lockheed 2006, p. 8.
Das et al. forthcoming.
King and Van de Walle forthcoming.
figure 1.3 and table 1.1 in Chapter 1 and table A1.1 in the appendix show the countries by MDG status.
MacDonald forthcoming.
Dang 2010.
Ravallion 2010.
Ravallion 2010.
Ravallion 2010; Lanjouw and Ravallion 1999.
Hall and Patrinos 2006.
Fiszbein and Schady 2009.
Hall and Patrinos 2011.
See, for example, Dollar and Kraay (2002).
Ravallion 2010.
Nopo, Chong, and Moro 2010; Patrinos, Skoufias, and Lunde 2010; Hall and Patrinos 2006; Psacharopoulos and Patrinos 1994.
Godoy et al. 2005; Rubio et al. 2009.
But see Scott (2009), who provides a comprehensive overview of various Southeast Asia groups who have avoided contact with modern society.
Huanca 2008; Ringhofer 2009.
Rubio et al. 2009.
Reyes-García et al. 2009.
Godoy et al. 2005.
Reyes-García and others 2010.
Godoy, Tanner et al. 2008.
