Gendercide and marginalisation – An initial review of the knowledge base

1. Introduction

Purpose

According to the political platform of the current government of Norway (Granavolden-erklæringen), the government will:

• Develop a strategy to ensure that marginalised groups, such a religious minorities, LHBTIQ-persons, and people with disabilities, are duly considered in relevant development programmes.
  
  
• Take an international initiative to prevent gendercide, the systematic selection of and violence against girls that has led to a large deficit of women in several countries.

The purpose of this report is to provide an initial review of the knowledge base for the development of strategies and policies in these areas.

Scope

Thematically, the report focuses on the deficit of girls and women (Section 2) and on the marginalisation of the three groups explicitly mentioned in the government platform (Sections 4-6). Section 3 discusses marginalisation in the broader perspective of Agenda 2030 and its vision to “leave no one behind”.

The report uses existing data and research to shed light on the magnitude and nature of four main issues, their causes and consequences, and possible remedies.

The main geographic focus is on countries and regions that are large recipients of Norwegian development assistance. We adopt a broader geographical perspective in the case of gendercide, as this is a phenomenon that largely takes place in China and India, where Norway directs little of its international aid.

Limitations

This assignment was carried out within a very short time frame that did not allow for systematic literature reviews and proper quality assurance of all the evidence reviewed.

2. Preventing gendercide

The Norwegian government has a new focus on gendercide, defined as systematic selection of and violence against girls.^[1] In this report we will use the term interchangeably with missing women, or missing females. The term missing women refers back to Amartya Sen's seminal work.^[2] In Norwegian language a useful term will be kvinneunderskudd (female shortage). The shortage is measured by a comparison of the ratio of the number of males to females in different age groups to what can be considered a normal ratio in societies that invest sufficiently in the health of both males and females. We will first discuss the male to female ratio at birth and then the ratio for different age groups. The first will reflect sex-selective abortion, the latter deviations from normal mortality rates for males and females.

The number of missing females after birth is an old phenomenon found in most poor countries, although it has been particularly high (but now declining) in India.^[3] The number of missing females at birth has increased since the 1980s. The rise in sex-selective abortion has occurred primarily in China and India. Among Norway's development partners, there is some evidence of sex-selection in Nepal. Since the trends differ over time, the relative importance of missing females at and after birth will vary over time. But in 2010, a year when many countries conducted their censuses, the number of women missing at birth was about the same as the number missing after birth.^[4] Adding up the numbers there were in total 125 million missing females in 2010. About 50% of the missing females are missing from China, 1/3 from India, 7% from Pakistan and Bangladesh combined, and the remaining 10% from the rest of the world.

Sex-selection at birth and higher mortality for women after birth both reflect an underlying son-preference. This fact suggests that both problems may be counteracted by advocacy campaigns, law reform, and policy change in relevant areas. Policies that can increase the relative value of girls, such as investments in education and health, are also likely to help counteract both problems. On top of this, investments in female nutrition and health will directly help in bringing down the female mortality rate.

2.1 The extent of gendercide

In this section we discuss the extent of the gendercide problem, while we discuss underlying causes in section 2.2, and potential remedies in section 2.3.

2.1.1 Sex ratio at birth

From nature's side more boys are born than girls (leading to a male to female ratio larger than one), but the natural sex ratio at birth seems to vary between ethnic groups and with the average age of the mother.^[5] The actual sex ratio is, for many countries, reported as constant over decades, and may thus reflect an estimate of the average natural sex ratio of the population. For other countries there are small fluctuations from year to year, but still the reported ratio seems to reflect the underlying natural ratio, as shown in Table 2.1.^[6]

Table 2.1: Male/female sex ratio at birth

Country/region	1967	1977	1987		1997	2007	2016	Increase 1967-2016
Regions
Euro area	1.059	1.059	1.058		1.058	1.058	1.057	0.00
Sub-Saharan Africa	1.038	1.038	1.038		1.038	1.038	1.039	0.00
South Asia	1.059	1.059	1.066		1.090	1.098	1.096	0.04
High male/female sex ratio
China	1.070	1.070	1.080		1.140	1.170	1.152	0.08
India	1.060	1.060	1.070		1.099	1.109	1.107	0.05
Armenia	1.064	1.052	1.058		1.151	1.151	1.131	0.07
Azerbaijan	1.062	1.062	1.062		1.130	1.166	1.132	0.07
Pakistan	1.064	1.064	1.064		1.080	1.086	1.087	0.02
Maldives	1.058	1.062	1.053		1.062	1.064	1.074	0.02
Core countries of Norwegian development cooperation
Nepal	1.045	1.045	1.045		1.071	1.054	1.065	0.02
Afghanistan	1.060	1.060	1.060		1.060	1.060	1.060	0
Myanmar	1.030	1.030	1.030		1.030	1.030	1.030	0
Indonesia	1.050	1.050	1.050		1.050	1.050	1.050	0
Colombia	1.050	1.050	1.050		1.050	1.050	1.050	0
Palestine					1.050	1.050	1.050	0
South Sudan	1.040	1.040	1.040		1.040	1.040	1.040	0
Somalia	1.030	1.030	1.030		1.030	1.030	1.030	0
Ghana	1.053	1.053	1.053		1.053	1.053	1.053	0
Mali	1.050	1.050	1.050		1.050	1.050	1.050	0
Niger	1.050	1.050	1.050		1.050	1.050	1.050	0
Ethiopia	1.040	1.040	1.040		1.040	1.040	1.040	0
Malawi	1.020	1.020	1.020		1.020	1.025	1.030	0.01
Mozambique	1.030	1.030	1.030		1.030	1.030	1.030	0
Tanzania	1.030	1.030	1.030		1.030	1.030	1.030	0
Uganda	1.030	1.030	1.030		1.030	1.030	1.030	0

Source: United Nations Population Division (downloaded from World Development Indicators)

As we can see, the natural ratio for most African countries Africa is 1.03,^[7] while it appears to be close to 1.06 for the rest of the world. Not counting the small fluctuation around the natural ratio, only a few countries have seen an increase in the sex ratio at birth over time. These include a few countries in the Caucasus and Central Asia (Armenia, Azerbaijan and Uzbekistan), and possibly some Asian countries (Vietnam, Pakistan, Nepal).^[8] But primarily there are two countries, China and India, with extensive sex-selection before birth. The size of these countries contributes to the large numbers, but they are also among the countries with the highest ratios of missing girls at birth, with China having the highest ratio. In China the sex ratio at birth is directly linked to the government’s one-child policy implemented in the 1980s.^[9]

In India, some states have numbers similar to those of China, with the highest number of boys born relative to girls being found in the most developed states (Haryana, Gujarat, Uttarakhand, Rajastan, Delhi) where presumably people have easy access to ultrasound technology. The relative value of boys may also be higher in these states, as the income potential is higher.^[10] The ratio goes up to 1.2 in these states, which can be compared to for example Odisha, which is at the normal level of 1.05. For most states the sex ratio at birth is higher in urban than in rural areas. The largest rural-urban difference is found in Chattisgarh, where the urban-rural ratio is 1.2.

The sex-selection in India is clearly driven by a son-preference, as the sex ratio is normal for the first-born. If the first-born turns out to be a girl, though, then sex selection happens for the second pregnancy.^[11] A supplementary strategy is just to wait for a boy, and thus have additional children if the first-born is a girl, which also has support in the data.^[12]

2.1.2 Sex ratio after birth

As mentioned in the introduction, the annual number of missing females due to premature mortality is of the same magnitude as the number of missing females due to sex-selective abortion. While sex-selective abortion is a relatively new phenomenon, a gender bias in premature mortality has been relatively constant over time and seems to reflect that families invest more in sons and in particular in the first-born son. This is particularly so in India.^[13] While the sex ratio at birth is higher in China, the under-five-mortality (U5M) rate is higher in India, and in particular so for girls.^[14] Thus while boys in most countries are more likely to die during their first years of life, in India this is reversed.

Underinvestment in girls starts even before they are born.^[15] The high mortality rate below the age of five for girls in India is still, however, not enough to bring the sex ratio for this age group in India (1.11) up at the level of China (1.16).^[16] Table 2.2 reports the sex ratio for different age groups around 2010 (when many countries conducted a census) for selected countries and aggregate regions.^[17]

Table 2.2: Male/female sex ratio by age group in 2010

Country	Total	0-4	5-14	15-24	25-49	50+
Groups of countries
World	101.7	107.1	106.9	105.3	102.6	90.9
Northern Europe	96.4	105.0	105.0	103.4	99.8	87.1
Asia	104.8	109.9	109.6	107.0	104.8	96.6
Low-income countries	98.5	102.4	101.8	99.5	96.9	86.6
Africa	99.5	102.9	102.2	101.0	98.3	89.0
Sub-Saharan Africa	99.2	102.5	101.7	100.4	98.1	87.5
Selected countries
Armenia	87.9	114.6	114.7	93.1	83.0	75.5
Azerbaijan	98.2	115.9	113.4	100.7	95.6	83.6
China	106.1	116.7	114.5	108.9	104.7	100.2
India	107.8	111.4	111.5	109.6	107.6	99.4
Viet Nam	97.5	109.7	106.1	103.8	99.9	76.4
Pakistan	105.8	107.6	107.9	105.1	104.2	105.3
Bangladesh	102.3	104.3	104.2	102.2	99.6	105.3
Core countries of Norwegian development cooperation
Nepal	95.2	104.8	104.7	88.9	87.3	97.7
Afghanistan	105.7	104.9	105.7	106.6	108.4	97.2
Myanmar	95.3	101.7	100.9	99.0	94.8	83.0
Indonesia	101.7	104.3	106.1	103.5	100.7	95.4
Colombia	97.3	104.3	104.2	102.9	95.8	85.7
Palestine	102.9	104.4	104.3	103.9	102.2	96.0
South Sudan	99.9	102.7	102.1	101.0	99.0	90.7
Somalia	99.7	101.5	100.9	99.0	97.4	100.1
Ghana	98.0	104.4	104.3	101.3	92.4	86.1
Mali	99.8	103.5	102.8	102.4	98.1	82.6
Niger	99.9	104.4	103.9	91.9	95.9	104.2
Ethiopia	99.6	102.7	102.2	101.4	97.5	89.8
Malawi	97.8	101.6	100.5	99.3	98.4	77.0
Mozambique	94.4	101.3	100.1	98.6	87.2	78.9
Tanzania	97.4	101.9	99.9	96.0	97.0	86.7
Uganda	98.6	101.9	101.1	100.4	98.0	78.0

Source: https://population.un.org/wpp/Download/Standard/Population/

In many countries, rich and poor, women tend to live longer than men, which is reflected in low sex ratios for age 50+. In some countries, the male surplus stays high till the age of 50, that is, particularly for India, but also for Afghanistan, Pakistan and China. In Africa the trend is quite similar to that in Northern Europe.

With India as the extreme case, we go in some more detail on intra-state variation. The estimates for U5M vary, but one estimate indicates that excess female U5M is high in Delhi and Rajastan, both among the states with high sex ratios at birth, but also in the poorer states of Uttar Pradesh and Bihar.^[18] Still, the census data indicates that the male to female sex ratio for children of age 0-6 is at the highest in the richer states of Haryana, Delhi and Punjab.^[19] For all India, the highest sex ratio is found for the age group 15-19, for which the sex ratio is similar to the high sex ratio at birth in China.

Table 2.3: Male/female sex ratio (census: China-2010, India-2011)

Age group	China			India
	Rural	Urban	Total	Rural	Urban	Total
All	1.05	1.05	1.05	1.05	1.08	1.06
0	1.19	1.16	1.18	1.10	1.11	1.10
0-4	1.20	1.18	1.19	1.08	1.10	1.08
5 - 9	1.19	1.19	1.19	1.09	1.12	1.09
10 - 14	1.16	1.17	1.16	1.09	1.12	1.10
15 - 19	1.10	1.06	1.08	1.13	1.13	1.13
20 - 24	0.99	1.03	1.01	1.07	1.07	1.07
25 - 29	1.02	1.01	1.01	1.02	1.03	1.03
30 - 34	1.05	1.03	1.04	1.00	1.05	1.02
35 - 39	1.04	1.05	1.05	1.01	1.04	1.02
40 - 44	1.02	1.06	1.04	1.07	1.09	1.08
45 - 49	1.01	1.07	1.04	1.05	1.09	1.06
50 - 54	1.05	1.06	1.05	1.10	1.13	1.11
55 - 59	1.04	1.00	1.02	0.94	1.09	0.99
60 - 64	1.06	1.01	1.03	0.97	1.03	0.99
65 - 69	1.05	0.98	1.02	0.95	0.98	0.96
70 - 74	1.01	0.96	0.99	1.02	0.98	1.01
75 - 79	0.88	0.92	0.90	0.95	0.94	0.95
80 - 84	0.75	0.86	0.79	0.92	0.81	0.89
85 - 89	0.60	0.70	0.64	0.93	0.80	0.89
90 - 94	0.47	0.56	0.51	0.85	0.75	0.82
95 - 99	0.41	0.53	0.47	0.90	0.81	0.87
100 +	0.30	0.36	0.33	0.90	0.94	0.91

Source: http://data.un.org/Data.aspx?d=POP&f=tableCode%3A22

With China and India having the highest numbers and ratios of missing women, we report the sex ratio for these two countries in detail in Table 2.3. We find a drop in the sex ratio after age 14 in China, and a drop after age 19 in India. This may be explained by a slightly earlier expansion of ultrasound in India. In urban China and India there is a surplus of women from age 65 onwards, as men tend to die earlier than women. In rural India the female surplus starts already at age 55, while for rural China it starts only at age 75. Indian men in rural areas thus die relatively early, possibly due to a combination of hard work and more recent lifestyle changes.^[20]

Turning to Africa, some have argued that even though most missing women are found in Asia, adult women in Africa tend to die early, and measured in percentages there are thus many missing women there as well.^[21] The argument is based on an analysis of burden of disease and estimates that compare deaths in different age groups with more developed countries. The estimates can, however, not be reconciled with aggregate numbers.^[22] For a discussion of methodology, and what we considered to be balanced estimates, see Bongaarts and Guilmoto (2015).^[23]

Bongaarts and Guilmoto report (in their Table 2) in total 125 million missing women in 2010. The estimate includes both girls missing at birth and adult women missing. There were 62 million missing in China, 43 million in India, 7 million in Pakistan and Bangladesh together, and 13 million in the rest of the world. Sex-selection at birth is a relatively new phenomenon, which, as discussed, basically is found in China and India. Excess mortality for girls has, however, been relatively constant over time in all parts of the developing world. These are obviously two very different problems that to some extent require different solutions.

2.2 Causes

Son preference is an underlying cause for both prenatal and postnatal gender bias and discrimination. The availability of prenatal sex diagnosis from the 1980s and access to abortion methods allowed for sex selection before birth by (some of) those with strong son preferences.^[24] Postnatal discrimination and violence in various forms due to son preference also led to excess female mortality, and may have manifested for instance as infanticide, abandonment, deliberate neglect of care, including access to health care, immunization, nutrition, and educational investment.^[25]

Broadly, son preference is associated with the lesser economic value of girls to parents and cultural practices that increase the perceived value of boys and correspondingly undermine the value of girls. Traditional patrilineal kinship systems that give rise to strong son preference are typically linked to sex selection.^[26] In these kinship systems, group membership, descent, and inheritance are typically traced through male members. Upon marriage a woman resides with her husband’s family and cares for her in-laws.^[27] Other marriage practices such as dowry also play into this. The costs of dowry in India has for instance been found to motivate the elimination of female births.^[28]

We find the same traditional patrilineal kinship systems throughout South-Asia, which explains the extent of son-preference found also in Pakistan and Nepal.^[29] Based on data from India, there is some evidence that religion is a factor. The evidence suggests lower acceptance of sex-selective abortion among Muslims than Hindus.^[30] The data is, however, far from clear. Muslims in India tend to get more children than Hindus,^[31] and are thus more likely to get the wanted son, so it is not straightforward to identify differences in son-preference in population data. Data on school attendance and labour market participation indicates that Muslim women are disadvantaged as compared to Hindu women,^[32] but again the data is not straightforward to interpret as there may be other factors that are correlated with religion that may explain these findings. In general, one shall be very careful in interpreting religion as an explanation for son-preference. On the contrary, the cross-country evidence indicates that son-preference is universal and manifests itself in different forms under different contexts.

When it comes to sex-selective abortion, son-preference is a problem only in combination with low birth-rates and ultrasound technology that can identify girls. (Lower birth rates reduce the chance of having the wanted son: with two children there is a 25% chance of no boy while the chance is only 3% with five children). As family planning is considered an objective in itself in many countries and given that ultrasound is increasingly available and affordable, it appears that the only way forward will be to tilt the value parents put on girls, for instance by strengthening women’s ability to compete with men in the labour market or in home-production in emerging economies such as China and India. With an increasing lack of women, we shall expect a self-regulating effect with regard to marriage and home-production.^[33]

2.3 Policy responses^[34]

Numerous measures have been taken to prevent missing females before and after birth, including measures aimed at regulating the access to prenatal sex selection technology and measures targeting the root causes of son preference.^[35] Yet there is still scant knowledge about what works in terms of policies to resolve the problem. Thus, taking steps to bolster efforts to monitor and evaluate relevant interventions should make up a key component of future endeavours to resolve the problem of missing females. 

Bans on prenatal sex determination and sex-selective abortions are among the most frequently used measures to prevent the problem of missing females.^[36] This reflects the recent scholarship’s narrow focus on prenatal sex selection that has come at the cost of postnatal discrimination and excess female mortality.^[37] The emphasis in the literature has been ascribed to the fact that prenatal sex selection is a quite new phenomenon.^[38] Other possible explanations of the focus on bans are that the linkage between remedies and practice has been relatively obvious, and short-term effects of policies and programs could be expected. Still, there are a lack of viable solutions to curb the practice. There have been multiple challenges with the implementation of bans, including difficulties for instance in proving cases of sex selection, financial and logistical challenges related to enforcement monitoring, and unintended consequences of bans.^[39] On top of this, there are critical ethical concerns related to women’s reproductive rights that must be considered.^[40] 

In addition to bans on prenatal sex determination and sex-selective abortions (frequently enforced by penalties for breaking regulations), measures include financial incentives in the form of conditional cash transfers to parents with daughters. The rationale behind these programs is to deter preferential treatment of sons by subsidizing investments in girls. So far evidence suggests that conditional cash transfer programs overall have not been very effective, due to severe implementation challenges, mostly short-term gains, and inability to change son-preference.^[41] There are also enormous expenses related to compensating parents for the extra economic costs of having a daughter. Also, it might be counterproductive to implement programs that directly signal, and in this way also reinforce preconceived notions, that girls are less valuable than boys.

Measures targeting the main causes of son preference are less apparent in the scholarly literature on missing females. There is, however, a large literature on associated issues, including on gender inequality, although studies may not be directly focused on excess female mortality. Some of these measures are long-term as it will take time to improve the status of women and girls in society. Some have indicated that indirect measures may be more promising than direct measures when it comes to reducing son preference.^[42] Potential remedies include advocacy and legislative measures to advance girls’ and women’s rights.

Advocacy campaigns to sensitize and create awareness about various aspects of missing females (both pre-and post-natal) may contribute to reducing son preference.^[43] Media exposure and messaging have been found to have an effect in India,^[44] and large-scale public awareness campaigns that seek to increase the value of girls are commonly used in South and East Asia. Studies further demonstrate that infotainment shows on radio and television may be used to affect family planning, and thus may also have the potential to effect change in attitudes and behaviour that increase gender equality. A study of an entertainment-education radio soap opera in Tanzania found that the soap opera had strong behavioural effects on family planning adoption, and evidence from Brazil suggests that soap operas that portray small happy families lowered the fertility rate among the relevant audience.^[45] The potential of media messaging to affect norms and behaviours are consistent with numerous studies in various fields, including public health studies and studies of voting behaviour in the US.^[46] However, care must be taken to make sure that relevant message – and messenger – effects, including the potential for backlash, are considered when developing campaigns. Rights concerning gender equality, sexuality, and reproduction are policy areas that are sensitive to prior beliefs and backfire effects, and reform promotion applying the “wrong” messenger could impact the support of the reform significantly. This underscores the need to gather context-specific information and to develop an effective strategy for engaging citizens positively with different leaders when conducting campaigns.^[47]

A number of measures may be taken to reduce the problem of missing females by way of increasing gender equality. Relevant interventions include reform of family law (marriage, divorce, inheritance, and property rights), access to health, education, and paid work. Policies that may increase the relative value of girls in the marriage market, as well as in the labour market, are related to their human capital, that is, investments in health and education. Improvements in female health conditions are related to broad health measures including improved nutrition, vaccinations, sanitary facilities and practices, sexual and reproductive health, including maternal health and family planning, primary health services, and affordable hospital treatments. Thus, policies for improved health of women, which includes better education, may also increase the expected value of girls, and thus counteract both sex-selective abortion and the chances of survival after birth.

Some reforms have been found to be effective. Interventions to improve girls’ inheritance rights in India in 2005, for instance, resulted in a significant increase in likelihood to inherit land, educational attainment, and a recent study also identifies gains in health status for daughters of women who benefited from the intervention.^[48] South Korea has been put forward as a success story since in recent times it has managed to reverse its male-skewed sex ratio at birth as well as its child sex ratio. The development has been attributed to normative shifts brought about by urbanization that in turn led to reduced son preference. Legislative changes addressing family law coincided with or followed the normative change, making it difficult to properly isolate the impact of the reforms.^[49] A study finds that pension reform altered son preference by reducing the reliance on sons during old age, but the study has limitations due to data weaknesses.^[50] Taking an alternative approach to the study of sex selection in South Korea, Edlund and Lee argue in a 2013 study that the return to normal sex ratios does not reflect a decline in son preference, but rather an increased value of women as the economy has developed.^[51]

More education and employment opportunities for women have been documented to positively affect outcomes for girls, for instance in terms of increasing schooling, educational attainment, and health.^[52] In a recent review of policies to tackle son preference, Kumar and Sinha (2018) argue that a role model effect may also emerge by way of increasing the number of women entering the formal labour force, possibly effecting change in girls’ ambitions and parents’ will to invest in daughters.

In the gender and politics scholarship, a growing strand of research focuses on female politicians as role models (symbolic representation), and whether increased presence of women in politics is associated with enhanced female empowerment. Several studies find positive symbolic representation effects.^[53] Experimental and survey evidence from India provide evidence on how female politicians positively affect voter attitudes about female leaders’ effectiveness, weaken stereotypes about gender roles, and influence girls’ career aspirations and educational attainment.^[54] There is a large body of research on whether women politicians act for women. Some of this literature does not find any evidence of this, but multiple studies find that women in many instances promote pro-women policy concerns more than men.^[55] Thus women’s presence in politics may ensure that girls’ and women’s interests and needs are prioritized to a larger extent. Evidence from a randomized policy experiment in India shows that village councils with reserved seats for women in West Bengal and Rajasthan invest more in infrastructure relevant to the needs of women.^[56]  Thus measures to increase the number of women in politics, such as electoral gender quotas and gendered electoral financing, could work to ensure positive policy outcomes for girls and women.

3. Marginalisation and “leave no one behind”

This chapter provides a short introduction to the three next chapters. We first define what we mean by a marginalised group and how this relates to the UN pledge to leave no one behind. We then discuss what it takes to identify marginalised groups, including some practical challenges. Finally, we briefly highlight the importance of politics in addressing marginalisation.

3.1 What is a marginalised group?

The government has committed to developing a strategy for the inclusion of marginalised groups in relevant development programs. Three such groups are explicitly mentioned in the government platform – religious minorities, LGBTQI+ persons, and disabled persons – but since these are presented as examples, our interpretation is that the strategy will also encompass other marginalised groups.

In the research literature, marginalisation is commonly defined as a social process whereby people are denied opportunities and rights normally available to members of the society. A marginalised group is a group that is denied such opportunities/rights based on a shared characteristic (e.g., gender, religion, sexual preference, class, origin, race, ethnicity). A marginalised group is thus a group that is discriminated against.

Broader definitions may emerge if we approach the issue of marginalisation from the perspective of social justice. Consider people with disabilities. Even without any denial of opportunities/rights available to others, theories of justice are concerned with the fact that physical or intellectual impairments may put disabled persons in a disadvantaged position. For example, Sen’s capability approach is concerned not only with formal equality of opportunity (i.e., avoiding discrimination), but also with substantive equality of opportunity, addressing the inequalities that arise because people bring different capabilities to the “starting line.” ^[57] Within such a framework, a marginalised group can be defined as a group that suffers social injustice, formally and/or substantively.

These definitions do not distinguish between more or less severe discrimination/injustice, which may imply that large portions of populations are classified as marginalised. This is not necessarily a useful approach if the policy goal is to improve the condition of the worst off, as is often be the case with policies targeting marginalised groups. A better approach will then be to reserve the concept for those who are subject to severe discrimination/injustice.

In this report, we therefore define marginalised groups as groups that share one or more characteristics that make them severely deprived, formally and/or substantively, of equal opportunities to participate in economic, political and/or social life.

3.2 Leave no one behind

The objective to include marginalised groups in development programs resonates well with the vision and goals of UN Agenda 2030. A key element of the agenda is the pledge to leave no one behind:

As we embark on this great collective journey, we pledge that no one will be left behind. Recognizing that the dignity of the human person is fundamental, we wish to see the Goals and targets met for all nations and peoples and for all segments of society. And we will endeavour to reach the furthest behind first (UNGA Resolution, 2015).

Leaving no one behind means not only that development should benefit those furthest behind, but that those furthest behind should benefit more than those who are better off. Leaving no one behind is thus a pledge for reduced inequality.

The goal of reduced inequality is further elaborated in SDG 10. Goal 10.2 is to “empower and promote the social, economic and political development of all, irrespective of age, sex, disability, race, ethnicity, origin, religion or economic or other status”, and goal 10.3 is to “ensure equal opportunity and reduce inequalities of outcome, including by eliminating discriminatory laws, policies and practices and promoting appropriate legislation, policies and action in this regard”.

Thus, Agenda 2030 goes further than aiming for equality of opportunity by explicitly also aiming for reduced inequalities of outcome. In this light, it is natural to define the furthest behind as those with the worst outcomes.

Against this background, it can be argued that policies for including marginalised groups in line with Agenda 2030 should start with identifying the worst off before defining categories of marginalised groups that will be targeted. Starting at the opposite end, as the government has done in its political platform, involves a risk of leaving out individuals and groups that are even more marginalised.

We notice that sexual minorities are explicitly mentioned in the government platform but not in Agenda 2030. This is probably not a coincidence but a reflection of political realities. Norway needs to take these realities into account in its policy approach.

3.3 Identifying those who are left behind

In this section, we will assume that a strategy for including marginalised groups in Norwegian development programs will aim to start from a precise understanding of who are furthest behind.

This requires first a measure of deprivation. Marginalised groups are commonly understood as groups that are deprived along multiple dimensions. This may be an argument for using a multidimensional measure, such a multi-dimensional poverty. However, in the context of specific development programs it may be more appropriate to adopt a narrower focus. For instance, in the case of Norwegian support to education, the furthest behind can be defined as those with poorest access to quality education.

A further step will be to characterize those who are left behind. This may serve two purposes: (1) identifying the underlying causes of deprivation, and (2) effectively targeting policy interventions to deprived groups (though this may also be possible without any further characterization). We may for instance ask whether the deprived belong to particular ethnic groups, whether they live in urban or rural areas, etc. Lenhardt and Samman^[58] conducted an analysis along these lines to identify characteristics that are associated with inequalities in education and health. Their work provides several findings of general interest:

1. Data is scarce. Since marginalised groups often are minorities, there may be too few relevant observations in nationally representative datasets to draw conclusions about their status. Another challenge is that the number of relevant characteristics captured in available datasets may be few. Some datasets do not contain information about religion, ethnicity, sexual orientation and other relevant variables. Lenhardt and Samman were able to analyse health and education outcomes and their association with (only) three characteristics (wealth, ethnicity and rural/urban residence) in 16 countries.      
   
   
2. The associations between background characteristics and outcomes differ across outcomes. Wealth was for instance strongly associated with education outcomes (explains between 20% and 40% of the inequality), while there was a much weaker association between wealth and health. Therefore, such analyses need to be conducted separately for development programs that target different outcomes.
   
   
3. The associations between background characteristics and health/education differed across countries. While wealth explained around 20% of inequality in education in Mali and Zimbabwe, it explained 40% in Bolivia. Therefore, separate analyses need to be conducted in different contexts.
   
   
4. Multiple characteristics may predict inequalities better than single characteristics. The literature on “intersectionality” or “intersecting inequalities” emphasizes that marginalised groups often are subject to multiple causes of inequality.^[59] This implies that marginalised groups are best identified by combining several characteristics associated with inequality (e.g., race, gender, ethnicity, class). Lenhardt and Samman conduct a preliminary analysis along these lines and find that the combination of living in rural areas and belong to an ethnic minority explains a much larger share of total inequality than any of the factors separately. Therefore, several dimensions need to be combined to identify the truly marginalised.
   
   
5. Even a comprehensive analysis based on intersecting inequalities is unlikely to capture everyone who is left behind. A large share of the inequalities was left unexplained in Lenhardt and Samman’s analysis. This suggests that important drivers are left out of the picture and illustrates the danger of identifying those left behind through the lens of preconceived categories. It may be more useful to start from another angle by simply asking “who are left behind, and what do they have in common?”

These perspectives should be kept in mind when we now turn to a discussion of specific groups that may be considered marginalised. While the perspective in this report is on how these groups can be included in relevant development programmes, it may be equally important to ask the more open question of which groups are marginalised in the context of Norwegian development programmes, and how can they be included. Other groups may be more vulnerable than the ones we discuss below.

3.3 Politics and policy response

Marginalisation is rooted in economic and political power struggles and in cultural, religious, and ideological structures (e.g., beliefs and attitudes). International actors must be aware of how their interventions may influence the underlying dynamics that contribute to marginalisation. While international aid aimed at marginalised groups creates avenues for recognition and mobilisation, it may also mobilise counterforces, as will be discussed in Chapter 4 on sexual minorities.  

Interventions aiming at reducing marginalisation can directly address the root causes, or may focus on enhancing the wellbeing of marginalised groups by alleviating symptoms. Sustainable solutions can only be achieved through the former, but root causes may be very difficult to change. Practical policies must therefore strike a balance between alleviation and structural change, and between the short-term and the long-term. Such considerations are of relevance for a discussion about mainstreaming vs. more targeted efforts. While mainstreaming of marginalised groups in development programmes may be important for the wellbeing of the marginalised, it is unlikely to address root causes. For example, it will be beyond the scope of programmes for universal access to health care to address the root causes of discrimination against sexual minorities. 

As more policies aim to include marginalised groups, being labelled as marginalised becomes more important for attracting attention and resources.^[60] The limited availability of good data on who are truly left behind makes it easier for relatively strong groups (perhaps with international allies) to attract disproportionate attention. Better data is therefore important for evidence-based policy-making in this area. Note, however, that the group in question may not always embrace the marginalisation label, as the label may strengthen the underlying dynamics that contribute to marginalisation. Such mechanisms may be at play in the case of religious minorities (see Chapter 6).

These few examples illustrate the importance of taking politics into account. A deeper analysis is needed for any strategy to include marginalised groups.

4. Inclusion of sexual and gender minorities

4.1 Definition of concepts

There are mainly two concepts employed both in the scholarly literature and in policy fora. These are: a) LGBTQI+, which is an acronym for Lesbian, Gay, Bisexual, Transgender, Queer and Intersex, and b) Sexual Orientation and Gender Identities (SOGI)/ Sexual orientation and Gender Identities and Expressions (SOGIE). The prevailing definitions of sexual orientation and gender identity employed internationally stem from the Yogyakarta principles on the application of international human rights law to sexual orientation and gender identity.^[61]  Sexual orientation is defined as an identity based on a person’s sexual attraction to others and how that person’s own gender corresponds to the gender of the people to whom that person is attracted (Lesbian, gay, and bisexual). The term homosexuality is a term used to describe people of any gender who are sexually and romantically attracted to people of the same gender.

Gender identity is defined as an individual’s internal, deeply felt sense of being a man, a woman, both, neither, or in-between. This may or may not match the individual’s biological or legal sex. Transgender is here an umbrella term for people whose gender identity and/or expression differ from cultural expectations based on the sex they were assigned at birth. This term includes those who are transsexual, transvestite, and gender queer. Transsexual refers to a person who does not identify with his or her sex assigned at birth and typically has or desires to alter sex characteristics, often through hormones and/or surgeries. Transvestite (also called cross-dresser) describes a person who regularly, although part-time, wears clothes mostly associated with the opposite gender to her or his birth gender, while gender queer challenges the gender norms as such.

Intersex people are persons who are born with chromosomic, hormonal levels, or genital characteristics that do not correspond to the given standard of “male” or “female” categories as for sexual or reproductive anatomy.

Queer is a term laden with various meanings and one that has a long history. In some contexts, it is used as a catch-all term for LGBTI+ people, but currently often denotes persons who do not wish to be identified with reference to traditional notions of gender and sexual orientation and eschew heterosexual, heteronormative, and gender-binary categorizations. It also often carries with it a critical perspective on heteronormativity.

4.2 Causes/drivers and consequences

Sexual and gender minorities do not constitute a homogenous group and experiences of discrimination and vulnerability vary considerably across contexts. This group is marginalised/vulnerable along at least two (interrelated) dimensions: legal discrimination/criminalisation and homophobia, which have several negative consequences for LGBTQI+ persons, including discrimination in the public and private sector, homophobic or anti-queer violence, social ostracism, poverty, and adverse health risks, including to mental health.

4.2.1 Legal discrimination/criminalization

Former British colonies and Sharia-ruled states are more likely to criminalise consensual same sex relations. Explicit criminalisation of non-heteronormative gender identities is less common, and in states where transgender persons are prosecuted, it is public morality and/or prostitution laws that are most commonly employed. Even in countries where consensual same-sex relations are de-criminalised in both the global north and the global south, LGBTQI+ persons lack legal protection against discrimination.

In 2018, 70 countries continue to criminalise consensual same-sex sexual acts, according to the 2019 ILGA report on state-sponsored homophobia.^[62] The majority of criminalising states are found in Asia and Africa. 11 Muslim majority countries in Asia and Africa, two of which are Norway’s partner countries, have imposed the death penalty on consensual same-sex sexual acts (Brunei, Yemen, Sudan, Iran, Northern Nigeria, Qatar, Somalia, Afghanistan, Pakistan, Mauritania, Saudi Arabia). These Sharia laws criminalize sodomy specifically. Sodomy is a crime under the hudud (singular, hadd, meaning limit, restriction, or prohibition), which are regarded as the ordinances of Allah, and they have fixed punishments derived from the Islamic sources. The death penalty for sodomy is seldom implemented in many of these countries, but some countries use gender-neutral “morality” laws to persecute people for consensual same-sex conduct as well as gender non-conformity such as cross-dressing. Which sexual and gender minorities are at risk depends on the context.^[63] Other Muslim-majority countries explicitly criminalise homosexuality, and to a lesser extent gender non-conformity, but with less severe penalties. These include Norway’s partner countries Palestinian Territories (Gaza strip), and Indonesia, in which some provinces have seen an increased crackdown against LGBTQI+ persons.

Consensual same-sex sexual acts are legal in the majority of Latin American states. In fact, in some Latin American countries (Argentina, Uruguay, Colombia, and Brazil) the legal status of LGBTQI+ protections are ahead of some of the most advanced and democratic nations of the world. In contrast, consensual same-sex sexual acts are illegal in the majority of African states, including in key partner countries to Norway (South Sudan, Tanzania, Uganda, Ethiopia, and Malawi). In the literature on LGBTQI+ rights and homosexuality in Africa, a central argument holds that the criminalisation of homosexual acts is a colonial import, especially from Britain, criminalising “sodomy” or “unnatural offences” with long prison sentences.^[64] Han and Mahoney (2014) investigate the influence of British colonialism on the spread of laws criminalising homosexuality and find that British ex-colonies are far more likely to have such laws than countries that were colonised by the Spanish and the French.^[65] However, sodomy laws were in most cases dormant, and only in the 2000s revived and enforced.^[66] Most public attention has been on proposals to impose harsher penalties for homosexual offences – including the death penalty in countries such as Uganda, South Sudan, Gambia, Nigeria  and Tanzania. Such laws also commonly violate rights of expression and organization for LGBTQI+ organizations, including prohibiting them from receiving foreign financial support. Many countries have also introduced constitutional provisions to prevent the introduction of same-sex marriage.^[67]

Although the majority of criminalising countries are in the global south, specifically in Asia and Africa, countries on these two continents have made great strides towards de-criminalisation and legal protection of LGBTQI+ persons. Among those countries are Norway’s partner countries, Mozambique and Nepal. South African law is among the most protective of LGBTQI+ rights in the world. The 1994 and 1996 post-apartheid constitutions ban discrimination based on gender and sexuality. In addition, same-sex marriage is recognised, and adoption rights and diverse protective rules against discrimination have been put in place. Mozambique (in 2015) has also de-criminalized same sex relations and provided some protections against discrimination. Nepal’s Supreme Court ruled in 2008 that LGBTQI+ persons would be regarded as “natural persons” under the law. The country’s constitution specifically says that the state shall not discriminate against “sexual minorities.” Citizens can designate themselves under a third gender category. Several other countries have in the past couple of years taken concrete steps to de-criminalise homosexuality and advance LGBTIQ+ rights including India and Angola. Other countries are discussing similar steps, including Tunisia, Kenya, and Botswana. Litigation and courts have played a central role in recent attempt to de-criminalise homosexuality.^[68]

There has been increased politicisation of homosexuality especially in African politics the last decade.^[69] While political governments on the continent claim homosexuality is un-African, against religion, and an import from the West, LGBTQI+ activists and scholars argue that it is homophobia that is an import from colonialism. Here they also point to the influence of American Christian fundamentalist groups. But we have also seen popular mobilisation against including sexual orientation and gender identity in sex education in several Latin American countries (Brazil, Colombia, Uruguay, and Peru)^[70] and in Europe (Poland, U.K). There is an emerging literature on politicisation of homosexuality in Africa that has been attributed to a number of factors.  (1) Religion: the growth of Evangelical Pentecostalism, and in particular international religious actors as framers and agenda setters, exacerbated by increased local competition between churches  (2) Political competition: political actors aim to divert attention from economic or governance crises.^[71] (3) International aid: direct external pressure on African states from western governments or international organizations have fuelled politicisation by strengthening the argument that homosexuality is an import from the West.^[72]

4.2.2 Homophobia

Prevailing research suggests that there are more homophobic attitudes in countries that criminalise same-sex relations, countries with a higher share of Islam and Evangelical Pentecostalism and economic underdevelopment. Much of this research is based on Africa.

A Pew Research Center survey from 2013 finds huge variance by region on the question of whether homosexuality should be accepted or rejected by society. The survey of publics in 39 countries finds broad acceptance of homosexuality in North America, the European Union, and much of Latin America, but equally widespread rejection in predominantly Muslim nations and in Africa, as well as in parts of Asia and in Russia.^[73]

Although widespread homophobia in Africa is often assumed, research demonstrates enormous differences in levels of homophobia between countries.^[74] There are important country-level differences. In countries that have de-criminalised same-sex relations (Cape Verde, South Africa, Mozambique, and Namibia), the majority of each country’s citizens express acceptance of homosexuality. In three other countries (Mauritius, Sao Tome and Principe, and Botswana), more than 40% say they are not opposed to having homosexual neighbours. Still intolerance toward homosexuals remains widespread, reaching near unanimity in Senegal at 97% and in Uganda and Niger at 95%. Younger and more educated Africans are generally more tolerant.^[75] Interestingly, while de-criminalisation has a clear effect, the countries that have never criminalised homosexuality are almost as homophobic as the countries that have the harshest laws.^[76]

What is driving homophobia in Africa? Drivers of homophobia are identified in the literature and overlap with the drivers of politicisation described in the previous section related to religion, economic development, and colonialism. Religion is perhaps the most cited driver of explanation for homophobia, irrespective of geographical location. Again, Islam and the role of the rapidly growing Evangelical Pentecostal churches are emphasised. It is documented that countries with a higher share of adherents to Islam and Evangelical Pentecostalism have more homophobic attitudes. Economic development is strongly associated with less homophobia. But the effect is indirect. It is mediated through the general standard of living, educational level, and size of the service sector in a country.^[77] The third key factor identified in explaining the root causes of homophobia is the sodomy laws that prohibit same-sex intimacy, largely imposed by the colonial powers.^[78]

4.2.3 Consequence of legal discrimination/criminalisation and homophobia

The negative consequences of legal discrimination/criminalisation are documented in the literature and can be summarised as heightened risk of homophobic violence or anti-queer violence, adverse health risks, and discrimination in the public sector.

Violence

There is a literature on homophobic or anti-queer violence, especially in the U.S and other de-criminalised contexts. Much of the focus has been on anti-queer violence in the public sphere from state authorities and the community, and much less attention has been paid to family violence. For many LGBTQI+ persons, the family is the primary arena where they experience rejection, violence, and ostracism. The literature puts emphasis on the importance of an intersectional understanding of such violence. A focus on homophobia might obscure race, class, and gender in anti-queer violence. For example, police violence against black LGBTQI+in the U.S cannot only be understood through homophobia.^[79] There is a literature specifically on “corrective rape” in Africa, mostly in South Africa, which refers to the rape of lesbian women with the intention to “cure” such raped women by force to change their sexual orientation. The term has been criticised, as it implies that rape has some inherent “rehabilitative focus.” Although most people condemn this assault on lesbian women, the term itself creates some idea of correction and should not be used at all. The term “homophobic rape” has been proposed in its place.^[80]

Although much of the literature on homophobic or anti-queer violence has focused primarily on countries where consensual same-sex relations are de-criminalised (as the topic is extremely difficult to research in criminalising contexts), there is some literature documenting violence and harassment against LGBTQI+ persons in Uganda and Nigeria, two countries where homosexuality has been politicised and renewed criminalisation legislation proposed. For example, in Uganda, politicisation and tabling of the “Kill the Gays Bill” resulted in heightened violence from the community with the implicit consent of the police, arbitrary arrest and detention, evictions, and loss of jobs for LGBTQI+ individuals. HIV clinics were also raided.^[81] Arrests of alleged gays and lesbians have become increasingly common across the African continent. In the 2014-2017 period, arrests were reported in Cameroon, Egypt, Gambia, Guinea, Kenya, Libya, Malawi, Mauritania, Morocco, Nigeria, Senegal, Somalia, Swaziland, Tanzania, Tunisia, Uganda, Zambia, and Zimbabwe.^[82]

Health

There is a consensus in the scholarship that marginalisation/discrimination is associated with higher rates of mental distress among LGBTQI+ persons. The terms gay-related stress and minority stress have been used to describe a range of stressors resulting from individual and institutional discrimination against LGBTQI+ people. Studies have found lifetime suicide attempt rates to be higher among LGBTQI+ persons than in the general population, especially among young gay and bisexual men, whose lifetime prevalence of suicide attempt is four times that of comparable heterosexual males.^[83] Other health disparities that have been noted in the literature is increased risk of depression, substance abuse, physical or sexual abuse, and self-destructive behaviors like engaging in unprotected sexual behavior, increasing the risk of STDs and HIV/AIDS. Unprotected and receptive anal sex increase the risk of HIV considerably.^[84] This literature in this area is almost exclusively based on Western countries.^[85] Epidemics of HIV in men who have sex with men continue to expand in most low, middle, and upper income countries and rates of new infection have been consistently high among young men who have sex with men ^[86]

Discrimination

Regarding discrimination in access to public institutions, the literature is for the most part based on de-criminalised Western countries. Research on discrimination in the health sector is the most developed. Evidence from around the world highlights that sexual and gender minority patients experience discrimination, stigmatisation, and even denial of care in the health system due to their sexual orientation and gender identity.

Criminalisation or perceived criminalisation of homosexuality plays a role. A study on sexual and gender minority adolescents in Southern African countries (Malawi, Mozambique, Namibia, Zambia, and Zimbabwe) found that LGBTQI+ persons experience double-marginalisation in pursuit of sexual and reproductive health services: as adolescents, they experience barriers to accessing LGBTQI+ organizations, who fear being painted as “homosexuality recruiters,” whilst they are simultaneously excluded from heteronormative adolescent sexual and reproductive health services. Such barriers to services are equally attributable to the real and perceived criminalisation of consensual sexual behaviors between partners of the same sex/gender, regardless of their age.^[87] A study from Nigeria reported negative attitudes toward provision of healthcare services to men who have sex with men among undergraduate students in Nigeria.^[88]

A hostile political environment can have negative effects on access to health for LGBTQI+ persons. In Senegal, where same-sex behavior between consenting adults is illegal, considerable advances in provision of HIV services for men who have sex with men  had been underway until 2008. In December 2008, just after Senegal had hosted the International Conference on AIDS in Africa (ICASA), the country saw a police crackdown. The impact of these arrests was immediate and marked declines in access to health care, fear of using services among MSM, and reports of men going into hiding. NGOs working with MSM suspended their activities, and providers reported sharp declines in MSM uptake and use of services ^[89]

In South Africa which is a de-criminalized context, an emerging body of literature documents health system bias against sexual and gender minorities: for example, Lane and colleagues interviewed men who have sex with men in Soweto, and revealed that all men who disclosed their sexual orientation at public health facilities had experienced some form of discrimination. Such discrimination, and also the anticipation thereof, leads to delays when seeking sexual health services such as HIV counselling and testing. Gender minority individuals, who are recognized as a key “at risk” group due to socio-economic marginalisation and exclusion, and who experience high levels of violence because of such marginalisation and gender non-conformity, encounter multiple layers of discrimination in South African health care facilities, ranging from verbal abuse to denial of care.^[90]

Within political contexts where homosexuality is politicised and homophobia widespread, there are also severe limitations of LGBTQI+ persons’ possibilities for activism, obstructing their struggle for their interests and political recognition.

4.3 Policy responses

4.3.1 De-criminalisation, but not a magic bullet

There are more homophobic attitudes in countries that have criminalised consensual same-sex relations/have never criminalised compared to those who have de-criminalised. However, we do not have rigorous research that suggest whether de-criminalisation of LGBTQI+ has come as a result of more tolerant attitudes or whether/to what extent legal reform (as opposed to other factors) has changed attitudes in South Africa and other de-criminalising African countries. Research from Mozambique suggests that de-criminalisation in 2015 reflected social practice in the country.^[91] The South African case (for which we have comparatively robust public opinion data over time) is interesting, as they show a liberalising trend over time, and particularly after the legal changes from 1994 onwards.^[92]

The literature on law and social transformation suggests that new legislation protecting vulnerable groups can change attitudes as well as create an enabling environment for civil society to both advocate for social as well as political changes for the group. It is assumed, but not documented, that de-criminalisation and legal protection of LGBTQI+ persons will have similar effects, but that it is not a magic bullet. For example, although Brazil provides strong legal protection to LGBTQI+ individuals, it is also one of the world’s murder capitals of LGBTQI+ individuals. At the same time, there are no reliable statistics on hate crimes towards LGBTQI+ persons in states where it is not considered a crime. In those states it is unlikely that a hate crime would be reported as such. In that light, it is a positive sign that hate crimes are recognized as such and reported in Brazil.

Although the literature largely sees de-criminalisation and legal protection as necessary, it needs to be combined with efforts to change homophobic attitudes and behaviours. Although South Africa is one of the most progressive countries in the world in terms of constitutional and legislative rights for LGBTQI+ individuals, several studies show health system bias against sexual and gender minorities and that South Africa’s progressive rights for LGBTQI+ individuals is not always reflected in teaching practices.^[93]

4.3.2 Aid to LGBTQI+ rights/health important, but prone to politicization

Aid to LGBTQI+ rights (and especially conditional aid) has made the topic prone to politicisation, but both drivers and effects need careful contextual analysis. We know little about under which conditions aid causes backlash effects and when it opens up space for LGBTQI+ organizations to claim rights/legal reform. Following pressure from international aid donors, Malawi began to liberalise its anti-LGBTQI+ laws, while in Zambia, Zimbabwe, and Uganda, pressure from international donors fuelled politicisation and criminalisation.^[94]

The literature on aid and politicisation is in its nascent stage, but there are some scholarly writings suggesting that public condemnation of human rights violations concerning the LGBTQI+ community and conditional aid to African states from Western governments or international organizations could actually be harmful to the cause.^[95] This is because such pressure feeds into a growing perception on the continent that the West is engaged in “cultural imperialism” and seeks to spread homosexuality, a “Western construct,” to Africa. Such external pressure may work in the short term depending on the degree to which African governments are dependent on aid and thereby vulnerable to Western pressure,^[96] but it can in the longer term have adverse effects, particularly on public opinion.^[97] Demone (2016) argues that aid conditionality ”seems to be a blunt instrument that does little to change embedded views.” Conditional aid has also distracted attention from intersectional oppressions,^[98] and it has shifted the debate to legal recognition and away from a focus on substantive social change.^[99]

Despite the pitfalls, scholars recommend continued aid to/support for LGBTQI+ rights and movements, but “under the radar” rather through public condemnation and conditional aid. Some recommends an indirect focus on health rights and non-discrimination. Epprecht (2012) suggests that strategies primarily focused on health concerns that simultaneously yet discreetly promote sexual rights are having some success in challenging prevalent homophobic or “silencing” cultures and discourses in Africa.^[100] Kojoué (2017) exploring links between the LGBTQI+ rights movement and the fight against HIV/AIDS in Cameroon, which is among the countries that criminalizes sodomy, argues that the state’s recognition of the LGBT population within the HIV/AIDS framework constitutes a “therapeutic citizenship.”^[101] In the contexts with the harshest penalties for homosexuality, the health approach represents perhaps the only opportunity for the LGBT movement to further its cause. Others have expressed doubt about the health approach, because a) the focus on men who have sex with men in HIV/AIDS policies have privileged men who have sex with men and thereby made other sexual and gender minorities less visible i and b) it has re-enforced a focus on the sexual aspect, which in itself has fuelled politicisation in certain contexts.^[102]

5. Inclusion of disabled persons

5.1 Measuring disability

According to the UN Convention on the Rights of Persons with Disabilities, “persons with disabilities include those who have long-term physical, mental, intellectual, or sensory impairments which in interaction with various barriers may hinder their full and effective participation in society on an equal basis with others.”^[103] This definition recognizes that disability is not caused by a medical condition alone but arises in the interplay between an underlying health problem and the environment.

Several approaches have been used to operationally define and measure disability. The commonly accepted method is the one developed by the Washington Group on Disability Statistics (Box 1). The method measures functional limitations, in line with the definition in the Convention.

A number of studies have used other approaches, including asking only about medical conditions, or simply asking whether a person has a disability. The latter method is believed to underestimate disability because disability is often associated with shame. Moreover, compared to the Washington Group method, this method is considered to capture less well the disabilities of older people (which often are perceived not as a disability but rather as a natural consequence of aging) and not to capture moderate disabilities to the same extent.^[104]

The standard estimate used by WHO is that 15% of the world population, or more than one billion people, is disabled.^[105] This estimate is highly uncertain. It is based on data from the World Health Survey (2002-4), conducted among the adult population in 59 countries. One reason for serious concern with the quality of the data is the large variability in disability prevalence across countries (for example; Bangladesh 31.9%, Malaysia 4.5%, Norway 4.3%, Sweden 19.3%).^[106]

Lower estimates are obtained in more recent studies using the Washington Group methodology.  Studies from 12 developing countries estimate a disability prevalence between 1% and 9%.^[107] Another set of studies from Tanzania, Uganda, Malawi, and Ethiopia find prevalence rates between 1% and 4%.^[108] Potential reasons why these estimates are lower than in the World Health Survey (WHS), apart from low quality of the WHS data, include: (1) The WHS data are based on a broader set of questions than the Washington Group approach. It includes 16 items in eight domains, including items such as bodily discomfort, difficulties in dealing with conflicts, and difficulties with falling asleep.^[109] (2) In the WHS studies, respondents were asked to report their experiences over the last 30 days, and not only long-term conditions, which increases prevalence rates. (3) Poor implementation of the Washington Group method in more recent surveys may have biased the estimates downwards. In studies that have received technical support from the Washington Group itself, disability prevalence typically ranges between 6% and 12%.^[110]

To identify individuals who are truly marginalised because of disability, we need to distinguish severe from moderate disability. The WHO estimates that 2.2% of the world population has very significant difficulties in their everyday lives.^[111] Roughly speaking, a very significant difficulty involves having extreme functional limitations in at least three of 16 items measured.^[112] There are also many measures of severe disability using the Washington Group approach, not very different from the WHO estimate. There seems however to be some confusion about how to define severe disability within the Washington Group approach. Several studies use the term “severe disability” about the standard threshold recommended by the Washington Group and use a lower threshold (requiring only “some difficulty” in one dimension) to classify a person as disabled.^[113]

Disability prevalence varies strongly with age. The WHS data suggest that disability prevalence increases four- to ten-fold from age group 18-49 to those above 60 (Table 5.1).^[114] Some of the health problems that cause functional limitations for older people are often perceived as a natural aspect of aging, and these people may therefore not consider themselves as disabled.

Table 5.1: Disability by age group

Age group	Disabled	Disabled with very significant difficulties
18-49	8.9	0.7
50-59	20.6	2.4
60+	38.1	7.4
Total	15.6	2.2

Source: World Disability Report, 2011

Disabled persons are found in all socio-economic groups of society, from the richest to the poorest. According to the World Disability Report, the WHS data show that disability is more prevalent among the poorest, ranging from 21% in the lowest wealth quintile to 11% in the highest.

However, a more recent analysis of the same data, but with a measure of disability more in line with the Washington Group definition, was not able to identify any association between socio-economic status and disability^[115] (see more below). 

There is no support in the data for higher disability prevalence in low income countries than in high income countries. The WHS prevalence estimate is highest in middle-income countries and lowest in high-income countries, but differences are not statistically significant.^[116]

5.2 Causes and consequences of disability

5.2.1 Causes and types of impairments

Very little data is available for understanding the causes of disability at an aggregate level. Below we report self-reported causes of disability in studies from Malawi, Nepal, and Zambia (henceforth referred to as the Sintef studies).^[117] The pattern is quite consistent across these countries.

The most commonly reported cause of disability is disease/illness, followed by congenital causes and accidents/falls (Table 5.3).^[118]  

Table 5.3: Self-reported causes of disability.

	Malawi (%) (n=4,154)	Nepal (%) (n=2,123)	Zambia (n=2,331)
From birth/ congenital	27.2	33.6	12.3
Accident	6.8	8.3	6.3
Fall	0.9	10.5	1.4
Burns	0.5	1.3	0.4
Disease/illness	57.9	38.3	32.7
Beaten by member of family	0.3	0.3	0.4
Violence outside the house	0.2	0.6	0.4
War related	0.0	0.6	0.1
Animal related	0.3	0.6	2.0
Stress related	1.0	0.9	2.7
Witchcraft	2.7	0.4	3.1
Others	2.2	1.4	16.1
Don’t know / refused	n.a.	3.6	24.0

Source: Sintef reports

The prevalence of various impairments is reported in selected country surveys. Figure 5.1 shows the prevalence rates in Zambia. Difficulties in walking and seeing dominate. This pattern seems to be consistent across African countries.^[119] In Nepal, however, difficulties in seeing appear less prevalent than other impairments.^[120]

Figure 5.1: Impairment prevalence among the disabled, Zambia (%

 Source: Sintef report, Zambia

5.2.2 Consequences: Marginalisation?

To which degree are the disabled marginalised? And in what areas of life does marginalisation of disabled persons typically occur? These are key questions for developing a strategy for inclusion of marginalised groups.

One way of answering both questions, and the one applied below, is to compare living conditions for the disabled with those of the non-disabled across various dimensions, such as socio-economic status, access to public services, work life, political participation, and stigma and abuse. 

There are several challenges with this approach. First, as the disabled make up a very heterogenous group, the broader the definition of disability, the smaller the differences between the disabled and the non-disabled are likely to be. As some of the definitions applied in the literature are quite broad, there is a risk that we conclude that there is no marginalisation, while the fact is that the more severely disabled are strongly marginalised. When possible, we should therefore conduct the analysis both for the disabled and for the severely disabled.

A second reason why small differences between disabled and non-disabled are not sufficient to conclude that the disabled are not marginalised is that the disabled often need more than others to obtain equal opportunities. Being enrolled in school is not of much help if the deaf do not receive deaf language instruction.

With these caveats in mind, the rest of this section explores available data on how disability affects the living conditions of the disabled. We focus on education, health, work and employment, and stigma. A more comprehensive analysis ought to look at political inclusion and the justice sector as well. 

Education

The disabled are severely disadvantaged in terms of access to education. The most recent comprehensive analysis is conducted by Mizunoya et al. (2018).^[121] They estimate the disability shortfall in education in 15 low- and middle-income countries. The disability shortfall is the difference, measured in percentage points, between the share of children who are out of school among the disabled and the non-disabled. The median disability shortfall is estimated to be 30 percentage points among primary-school-aged children and 38 percentage points among secondary-school-aged children. The disability shortfalls in school attendance in selected Norwegian partner countries are illustrated in Figure 5.2.

Figure 5.2: Disability shortfall in school attendance. Percentage points

 

Source: Mizunoya et al. (2018)

The great majority (85%) of disabled children who are out of school have never attended school at all. Moreover, there do not appear to be large systematic differences between boys and girls across the 15 countries. Finally, there are limited effects of socio-economic status on the probability for a disabled child to be out of school. Mizunoya et al. uses the Washington Group standard definition of disability, and the median disability prevalence in the relevant age group is 0.8% and 1% for primary-age and secondary-age children. This relatively low estimate suggests that this study is capturing primarily the severely disabled.   

Similar results were obtained by Filmer (2008) on older data from 13 countries.^[122] These datasets used different measures of disability with prevalence rates of 1-2% of the entire population, suggesting that only the severely disabled are included. The disability shortfall varies considerably across countries – from 10 to 60 percentage points – with a median shortfall of 21% for young children and 25% for older children. This is less than the estimates referred to above, suggesting that the disability shortfall may have increased over time as school enrolment has increased. Importantly, Filmer demonstrates that the schooling deficit associated with disability in most cases is larger than the deficit associated with other sources of inequality: gender, rural residence, and low economics status.

The Sintef studies also find very large schooling deficits in some countries. In Nepal, only 40.5% of the disabled ever attended school, compared to 71.1% among the non-disabled. These data cover all age groups older than 5 years. Interestingly, the differences between disabled and non-disabled is larger in the younger cohorts; the disabled have not been able to benefit as much as the non-disabled from the large increase in school enrolment during the last decades and have been left further behind. 

School deficits for the disabled are smaller in Zambia and Malawi, around 10 percentages points. Note, however, that the Sintef studies use a broader definition of disability than the studies referred to above. The Sintef studies use the Washington Group definition, but they define the cut-off point lower than the recommended level so that they also include people with “some difficulties” in functioning. This is likely to reduce the disability shortfall.

Further analysis of the Sintef data show that the shortfall is larger for the severely disabled. Severely disabled girls in rural areas are at a particular disadvantage, with school attendance of only 46.1% compared to the average of 80.8% for all disabled and 89.9% for the non-disabled. These findings suggest that discrimination against girls may be stronger among the disabled than among non-disabled and that this pattern is more pronounced in rural than in urban areas. These results underscore that an effective strategy for including marginalised groups needs to capture the intersectionality of factors that contribute to inequality and marginalisation. 

Figure 5.3: School attendance Zambia (%)

 

Source: Sintef report Zambia, p. 71

The most important reason for not going to school, both among disabled and non-disabled, is lack of economic resources. For the disabled, their disability is the second most important reason. Limited access to schools is not reported as a more frequent problem for disabled than for non-disabled.

Children with sensory impairments (vision, hearing and/or communication) and mental impairments run a much larger risk of being excluded from education than those who suffer a physical disability.^{[123] [124]}

Differences in the quality of education are also important. We have not looked further into this issue.

Health services, medical rehabilitation and assistive devices

The disabled do not seem to be at a severe disadvantage when it comes to access to health services. Based on the WHS data, the share of the population who needed but did not get health services is estimated at 5.8% among disabled and 3.9% among non-disabled.^[125]

This pattern is confirmed in later country studies by Sintef in Malawi, Zambia and Nepal.^[126] Note that all these studies use a broad definition of disability and do not provide any analysis for access to health services for the severely disabled. A study from Afghanistan and Zambia found that the severely disabled and the very severely disabled spend on average 1.3 times more on health services than the non-disabled.^[127]

Between 10% and 20% of the disabled report that inadequate equipment, inadequate provider skills, and previous bad treatment are among the reasons for lack of health care.^[128] These figures are higher than for non-disabled persons and may indicate that disabled persons receive lower quality care than the non-disabled.^[129] Maternal and reproductive health services have been identified as one area where the disabled receive poorer services, partly related to the fact that the disabled may be perceived as sexually inactive.^{[130] [131]}

Figure 5.4: Unmet need for services ((1-received/needed)*100)

Source: Sintef studies

Unmet needs for services directly targeting the disabled, such as assistive devices and medical rehabilitation, seem to be much larger than unmet needs for health services (Figure 5.4). This is supported by a study that finds that a very small share of the severely disabled in Tanzania, Malawi, and Uganda uses any assistive devices.^[132]  

This suggests that targeting those with large, special needs should be a main priority within the health sector.

Work and employment 

Available evidence suggests that disabled persons are disadvantaged in work life in low-income countries. They seem, however, to be less disadvantaged relative to the non-disabled than in high-income countries.

Employment rates^[133] are typically lower for disabled persons than for non-disabled, but the differences vary substantially between countries.^[134] In low income countries, the difference in employment rates is often small and not statistically significant, while the difference often seems to be larger in middle- and high-income countries.^[135] This may be explained by the extensive informal markets in low income countries and high rates of self-employment. In a number of countries, a higher share of persons with disabilities are self-employed compared to people without disabilities.^[136] 

Differences in employment rates are more pronounced for males than for females; for women the barriers to employment may well be primarily gender-related.^[137]

Figure 5.5: Share of population currently working (%)

Source: Sintef studies

More recent surveys from Malawi, Zambia and Nepal all show that a smaller proportion of the disabled are “currently working” compared to the non-disabled (Figure 5.5). The gaps between employment rates are however relatively small compared to those found in many high-income countries.^[138]

Difference in employment rates can be explained by lower levels of education, lower productivity, or discrimination. A study among men in rural India (Tamil Nadu) found that lower productivity and discrimination both were contributing factors.^[139] Discrimination was not confined to employers: heads of household displayed strong negative attitudes to the employability of people with disabilities. Interestingly, the study did not find any wage differences between disabled and non-disabled persons. This contrasts with the evidence of disability-based wage difference in most studies in developed countries.^[140]

Socio-economic status and poverty

Disabled persons may have lower socio-economic outcomes and be overrepresented among the poor both because disability may cause poverty (through lower educational attainment and lower earning potential) and because poverty may cause disability (through less nutrition, poorer sanitation, and less access to quality health care).^[141]

The evidence suggests that persons with disabilities are more likely to be income poor in developed countries. However, in developing countries the differences between disabled and nondisabled are small and in most cases not statistically significant. A study across 15 developing countries, using a number of different poverty measures as a well as a measure of asset ownership, finds that the disabled perform slightly worse than the non-disabled, but the difference is statistically significant in only a few countries.^[142] This study is based on data from the World Health Survey but uses a definition of disability more in line with the Washington Group approach, implying that it focuses on the more severely disabled.^[143]

Relatively small differences in socio-economic status between disabled and non-disabled were also found in the more recent living standard surveys in Malawi, Zambia, and Nepal. Figure 5.6 shows the wealth index (asset possessions) of Malawian households. The small differences between the disabled and the non-disabled are not the only striking finding here: it is also interesting that the differences between urban and rural areas are much bigger than the differences between the disabled and the non-disabled. The same pattern is found in Zambia and in Nepal.^[144] Based on this evidence it is not possible to conclude that disabled persons are marginalised economically.  

In interpreting these findings, it is important to note that socio-economic status and poverty are measured at the household level. In developing countries, disabled persons typically live in households with many members and thus rely on quite a number of others for their living. The Sintef studies suggest that disabled typically live in slightly larger households than the non-disabled.

Figure 5.6: Wealth index Malawi

Source: Sintef report

Moreover, when persons with disabilities are living in poor households, the resources available to them are not significantly different from the limited resources shared by other members of their households.^[145]

An analysis of the economic status of the most severely disabled might have revealed larger deprivations. Filmer (2008) finds that disabled adults are more likely to live in households in the poorer quintiles (while there is no such pattern for disabled children). His study differs from the ones discussed above in using disability measures that lead to a disability prevalence of 1-2%, which is considerably lower than in the studies discussed above. 

Stigma, discrimination and abuse^[146]

Prejudice, stereotypes and stigma contribute to the discrimination and exclusion of disabled persons across the world.

Literature from selected countries in Sub-Saharan Africa suggests that limited understanding of the causes of disabilities is a key driver of stigma. Various cultural and religious beliefs tend to explain disability by the misdeeds of ancestors, parents, or the person with disability, or by supernatural forces such as demons/spirits, witchcraft, or God. Consequently, disabled persons may be thought not to be quite human and to be a source of shame. Wrong beliefs about the nature of the disability and its consequences (e.g., that it is contagious, brings bad luck, or that the disabled are witches) further contribute to stigma and exclusion. 

Attitudes towards disabled persons vary across countries, communities, and families. People with intellectual and sensory disabilities are often more stigmatized than the physically disabled. Severely disabled persons are more stigmatized than those who can participate more in their communities, and stigmatized people are more exposed to harmful practices in rural areas.

For an overview of the role of disability stigma in reducing access to education, see Azalde and Braathen (2018).^[147]

The Afrobarometer reports self-reported incidence of discrimination and harassment due to disability during the past 12 months. Across all countries in the sample, the incidence is 16.3%. In Norwegian partner countries the reported incidence varies from 4.6% in Mali to 31.3% in Tanzania. Note that the implied disability prevalence in many countries is very high, implying that people have used a very wide definition of disability. With a narrower definition of disability, discrimination and harassment are likely to be more prevalent.

Table 5.4: Have been discriminated against or harassed because of disability during the past 12 months

	All countries	Ghana	Malawi	Mali	Mozambique	Tanzania	Uganda
	(n=9,547)	(n=498)	(n=142)	(n=283)	(n=1,008)	(n=51)	(n=167)
Never	82.6%	86.8%	81.9%	94.9%	82.4%	68.7%	77.5%
Once or twice	8.8%	5.1%	6.8%	3.4%	9.3%	17.3%	12.7%
Several times	4.9%	4.8%	6.0%	0.3%	4.3%	1.9%	5.4%
Many times	2.6%	2.4%	4.8%	0.9%	1.4%	12.1%	4.4%
Don´t know	1.1%	1.0%	0.5%	0.5%	2.7%	-	-
Implied disability prevalence	21%	12%	24%	43%	2%	14%	21%

Source: Afrobarometer 2016-18

The Sintef surveys measure whether the disabled have been abused or discriminated against. In Malawi and Zambia, 18% and 10% of the disabled report that they have been beaten or scolded because of their disability, and 8% report that they have been discriminated against by public service. In Nepal, the incidence of beating and scolding is much higher: 45%. In all countries, around 50% of the beating/scolding takes place within the family. Abuse is also considerably more common against the severely disabled.

5.3 Policy responses

A strategy for including marginalised groups in development programs should reflect the evidence discussed above, e.g.:

• Some disabled persons are marginalised, others not. While disability correlates with disadvantage in several dimensions in developing countries, large groups of disabled can live close to normal lives. This reflects the broad definition of disability commonly applied by the WHO and others. The severely disabled are more likely to be marginalised.

• Disabled persons are more disadvantaged in some dimensions than in others. Access to education seems to be a bigger challenge than access to health services and income-generating activities. The association between disability and socio-economic status and poverty is weak.

• Certain types of impairments are more strongly associated with disadvantage than others. Access to education is weaker for children with intellectual or sensory impairments than for children with physical impairments. Stigma is also typically more pronounced with intellectual impairment.

A strategy for including marginalised groups must address the specific challenges faced by marginalised persons with disabilities, including (1) disability-related beliefs and misconceptions leading to stigma and discrimination, (2) needs for medical rehabilitation, assistive technologies, and adaptation of public infrastructure, and (3) specialised knowledge by service providers, including teachers and health workers, on how to provide adequate services to the disabled.

Very few rigorous evaluations have been conducted to assess the effectiveness of interventions for stronger inclusion of persons with disabilities in developing countries.

Stigma

Intellectually disabled are most severely affected by stigma, but there is limited effort to combat such stigma in developing countries.^[148] Interventions have been implemented at family, community, and government/structural levels, and most interventions have focused on awareness-raising rather than actively aiming at changing behaviour.^[149] Unfortunately, few interventions have been evaluated, and even fewer have been evaluated with rigorous methods.^[150] Research in other areas show, however, that similar interventions often fail to meet their aims and at best result in increased knowledge, with little attitude or behaviour change.^[151]

Researchers have advocated for interpersonal contact with members of stigmatized groups as the most promising stigma reduction strategy, especially if it starts at an early age.^[152] Schools and other community settings are important arenas. Parents play a central role, but they need allies as it may require vast amounts of physical and emotional energy to sustain such contacts over time.^[153]  Contact may be combined with education to correct wrong beliefs, but pure educational efforts are perceived as less promising.^[154]

Education

Policies to include disabled children in education need to address the specific causes why disabled are out of school (see above). Costs are reported to be the most important reason, but this is the case also for non-disabled out-of-school children. The cost issue therefore needs to be dealt with across all children, not just specifically for the disabled. This conclusion is also supported by the limited association between socio-economic status and the disability shortfall in school attendance.

The second most important reason is “disability.” Behind this term could be a number of different causes, including (1) low priority in the family on education for the disabled; (2) stigma in the family, among peers, or among teachers; (3) lack of adaptation of the school environment to the needs of the disabled, including lack of assistive technologies such as hearing and seeing aids. Limited physical access may also play a role, although available evidence suggests that such barriers are not more important for the disabled than for the non-disabled (see above).

Regarding the quality of education, there is an ongoing debate on whether this is promoted most effectively through general policies for individualized teaching or through specific programs targeting disabled children. Croft (2013) argues that group-specific programs are needed in low-income contexts; although teaching adapted to individual characteristics may theoretically be preferred, such a degree of personalisation of learning is hard to achieve even in high-income countries, and even less so in low-income countries individualism is not traditionally valued and the education systems do not take much account even of the most common individual differences.^[155]

Employment

Policies to strengthen the opportunities for employment and income for the disabled include (1) laws and regulations; (2) quotas or incentives for employers, supported and sheltered employment; (3) vocational rehabilitation and training; (4) micro-finance; and (5) interventions to change attitudes.

The evidence on the effectiveness of such interventions is scarce. A systematic review of the effectiveness of interventions to improve labour market outcomes for people with physical and/or sensory disabilities identified only 14 relevant studies in the period 1990-2013 that passed the (low) quality threshold.^[156] The review points out that the methods used in the studies are open to a high degree of bias and that it is impossible to say with any certainty whether disabled in low- and middle-income countries can improve their labour market outcomes as a result of these interventions.

The effects of employment interventions may even be negative, as has been argued in the case of disability discrimination acts in the US and UK.^{[157] [158]} 

On the positive side, “supported employment” has been shown to be effective for people with severe disabilities.^[159] Supported employment includes employment coach­ing, specialized job training, individually tai­lored supervision, transportation, and assistive technology to enable disabled persons to function in the labour market.

Finally, a recent randomized, controlled trial of a targeted savings- and loans-program in rural Uganda found a strong, positive effect on income, consumption, and welfare of the disabled participants.^[160]

6. Inclusion of religious minorities

6.1 Definition of concepts

The question of what constitutes a “religious minority” is less straightforward than commonly assumed. There is no internationally agreed definition as to which groups constitute minorities. According to Capotorti, a former Special Rapporteur at the United Nations Sub-Commission on Prevention of Discrimination and Protection of Minorities, a minority is: “A group numerically inferior to the rest of the population of a state, in a non-dominant position, whose members - being nationals of the state - possess ethnic, religious or linguistic characteristics differing from the rest of the population and show, if only implicitly, a sense of solidarity, directed towards preserving their culture, traditions, religion or language.”^[161] This definition combines numerical inferiority with being in a non-dominant position in relation to other groups within a given state.^[162] While numerical minority status is often associated with some form of discrimination or lack of power, the correlation between minority status and subjugation is a matter to be investigated, not a given fact on which to build an analysis. Just as there are plenty of examples of powerful small groups, a numerical majority may be marginalised and discriminated, as is the case in Bahrain, where a Sunni monarchy rules over a 65% Shi’i majority. The latter group can be considered a “sociological minority,” which refers to groups that are systematically discriminated against, irrespective of their number.

In both popular usage and social theory, the term “minority” is often understood as synonymous with inferiority and subordination.^[163] Hence numerical majorities that face marginalisation and discrimination are often said to be “minoritized.” Today, the term “religious minority” is increasingly applied to groups whose religious practice differs in some respect from the majority religion practiced within a given nation-state. This development is a distinctly modern phenomenon that reflects both the salience of the nation-state and the fact that the “minority issue” has become a key preoccupation within the international human rights regime.

For the purposes of this report, “religious minorities” are understood as groups that are subject to some form of discrimination, violence or persecution based on their social identity, religious practice or belief.  More specifically, this report mainly centers on groups that face violations of “freedom of religion or belief” (FoRB), an important, albeit contested, right within the international human rights system.^[164] This right is outlined in Article 18 of the Universal Declaration of Human Rights (1948) and holds that: Everyone has the right to freedom of thought, conscience, and religion; this right includes freedom to change his religion or belief, and freedom, either alone or in community with others and in public or private, to manifest his religion or belief in teaching, practice, worship and observance.^[165]

The right to FoRB includes the right not to have a religion. The inclusion of “belief” enshrines the right to profess nonreligious or atheistic convictions. Hence atheists and agnostics are considered religious or belief minorities. FoRB is often presented as a neutral, universal right. Yet scholarship charting the genealogy of this concept and the conceptual work HR-conventions perform in defining “religion,” “freedom,” and “minorities” illustrates that such concepts are imbued with ideological content.^[166] Far from being matters of purely academic interest, greater awareness of these complexities is important as the struggles of ethnic and/or religious groups around the world increasingly unfold around the twin concepts of religious liberty and minority rights.

6.2 Causes and consequences

Despite wide adherence to international and regional conventions in support of “freedom of religion or belief,” there is growing consensus that violations of FoRB are widespread and that such violations have probably increased in recent years. Religious nationalism, xenophobic attitudes, and the rise of resurgent faiths are often cited as key drivers. Many analysts argue that globalisation is enabling more competition among religious communities, and that such competition has an adverse impact on religious minorities.^[167]

This section outlines some of the more severe HR-violations and issues affecting members of religious minorities, drawing on Pew Research Center reports, the International Religious Freedom Reports produced by the US Department of State, HR-reports, and academic sources.^[168] A few caveats on sources are in order: First, this report draws extensively on two reports issued by the U.S.-based Pew Research Center, a nonpartisan “fact tank” that publishes overviews on global trends relating to religion, specifically restrictions on religion. The reliance on this source reflects the formidable challenge of gathering and assessing comparative data on “religious minorities,” an extremely broad category as already indicated.

Second; while this report hints at the type of problems affecting certain religious minorities, it has not been possible to analyse the wide variety of consequences persecution and discrimination have for particular groups, such as job discrimination, poverty, inequitable government services, lack of documents, etc.  Nor do we discuss how groups facing discrimination seek to improve their situation, strategies that could be analysed in terms of “exit,” “voice,” and “loyalty.”^[169] Moreover, such strategies ought to consider transnational dynamics involving religious minorities, including the role of diaspora networks.  Exit has long been a strategy for groups fleeing conflict and discrimination. Just as many Christians are leaving parts of the Middle East,^[170] some Muslim minorities, such as the Ahmadiyah, have gained asylum in the West. The same goes for Bahai’is escaping persecution in Iran, a development contributing to turn Bahai’i into a global religious movement.

According to Pew Research, the number of countries with “high” or “very high” levels of restriction has increased from 20% in 2007 to 28 % in 2016. Pew estimates that 27 % of the world’s countries have “high” or “very high” hostilities relating to religion (2018).^[171] Since 2013, Pew has also identified the prevalence of restrictions and hostilities that tend to target religious minorities (understood as groups that constitute less than 50% of a country’s population), including subgroups within major religious traditions, such as Orthodox Christians, Sunni, Shia Muslims, etc.^[172] A key finding is that religious minorities are often the disproportionate targets of governmental restrictions in countries that score high on restrictions on religion and religious behaviour. That is, the types of restrictions and hostilities that tend to target religious minorities do not exist independently of general restrictions on religion or social hostilities involving religion. Instead, restrictions and hostilities targeting religious minorities often correspond with higher levels of restrictions and social hostilities.                        

Figure 6.1: Number of countries where religious groups experienced government harassment

Source: Pew report on restrictions on religion (2018), p. 29

6.2.1 Persecution by state actors^[173]

The military-led violence targeting Rohingya Muslims in Rakhine state in Myanmar has been described by the UN as a “possible genocide” and “crimes against humanity” that includes the killings of thousands of people, widespread rape, destruction of entire villages, resulting in a regional refugee crisis, with over 700,000 Rohingya still housed in camps in Bangladesh.^[174] The discrimination against the Rohingya is multifaceted  (ethnic, racial, economic, political). Hence it is far too simplistic to identify their religious identity as the main cause of their persecution. Growing Islamophobia is contributing to a worsening situation for other Muslim minorities, including Kaman, the only Muslim group recognised as one of the country’s 135 ethnic groups, many of whom have become internally displaced (IDPs), but also several ethnic Christian groups, notably Chin and Karen, have long faced repression.^[175]

In China, the UN estimates that up to one million Uighurs and other Muslim minorities may be held in detention centres in Xinjiang province, under the pretext of curbing “terrorism and religious extremism.”^[176] Falun Gong followers are also subjected to mass arrests and detention in re-education centres. China tops Pew’s list of countries ranked as having the highest score of “government restrictions involving religion.” It is widely recognized that conflict-displaced refugees and IDPs are vulnerable groups with special humanitarian needs.

6.2.2 Violence perpetrated by non-state actors

According to Pew’s “Social Hostilities Index,” which measures acts of “religious hostility” by individuals and religion-related armed conflict, including acts of terrorism, in 198 countries, the countries ranked as “very high” were India, Syria, Iraq, Nigeria, Egypt, Palestinian territories, Israel, Bangladesh and Russia.^[177] India and Nigeria have both ranked among the top five since 2008, while the only newcomer is Bangladesh, where there have been a spate of attacks on Hindu temples and growing contention around so-called atheist bloggers. Other countries with “high” scores in terms of “social hostilities” are Afghanistan, Central African Republic, and Pakistan. Pew analysts found that among the world’s 25 most populous countries, Egypt, Russia, India, Indonesia, and Turkey stand out as having the highest overall levels on “government restrictions” and “social hostilities,” a measure that appears to be relatively stable.                                                                                      

Militant Islamist group in Nigera, such Boko Haram and its affiliates, continue to stage abductions, suicide bombings targeting both Christians and Sufi-oriented Muslim groups in the northern states, with many killed and some 200,000 people forcibly displaced in 2018.^[178] In Egypt, recent years have seen a string of lethal violence targeting members of the Coptic Orthodox Church, such as the Palm Sunday twin church suicide bombings in Alexandria and Tanta in 2017, for which ISIS claimed responsibility. Sufi mosques have also been targets of numerous ISIS attacks, a particularly deadly attack occurred in Northern Sinai in late 2017.  The fact that India consistently scores “very high” on the Social Hostilities Index is related to a violent conflicts in several parts of the country, a rising trend of Hindu nationalism, whose latest incarnations includes “cow protection” vigilantes who have targeted Muslims and Dalits and Adivasi (indigenous) communities and others who allegedly violate bans on bovine slaughter.^[179] While federal Indian law provides minority community status to six religious groups (Muslims, Sikhs, Christian, Parsis, Jains, and Buddhists), many analysts argue that the ascendancy of the Hindu Right in the political area and the associated targeting of religious minorities posits a challenge to India’s legacy of religious pluralism.^[180]                                                           

6.2.3 Legal restrictions and criminalization

Blasphemy/defamation

As of 2017, 71 countries have laws prohibiting blasphemy or defamation and such laws are found in most regions of the world. In Europe such laws are nowadays rarely enforced, which likely reflects both awareness that such laws have historically served to protect the sensibilities of dominant religious groups and growing secularism. Insofar as FoRB is conceived as “a form of equality right,”^[181] it follows that the state should not involve itself in internal theological disputes and should strive to treat different faith groups in an even-handed manner. The majority of states that enforce blasphemy provisions are Muslim-majority countries in the Middle East, Africa, South Asia, and Southeast Asia, with penalties ranging from imprisonment to death.^[182] Blasphemy-like provisions, such as contempt for Buddhism, are also actively enforced in many Buddhist-majority states, such as Sri Lanka, Myanmar, and to a lesser extent in Thailand.^[183]                              

A common feature of blasphemy laws is vague wording, which gives wide scope for criminalizing “deviance,” as illustrated in Indonesia, where the upsurge of blasphemy convictions in the past decade has mainly affected non-mainstream Muslim groups. Recent cases targeting atheists, Christians, and Buddhists suggest that the scope for criticising Islam and Islamic authorities is narrowing, but there is also great regional variation in enforcement.^[184] Malaysia and Brunei, both of which are confessional states, have very strict blasphemy laws and apostasy laws. Brunei’s new Syariah Penal Code (2013) extends the blasphemy laws, which may carry the death penalty, also to non-Muslims. In the process of implementing new Islamic laws, both countries appear to have looked to countries located outside the region, namely to Pakistan and Saudi Arabia, where blasphemy is a capital offence. In both countries, minorities appear to be the main targets, particularly Shia Muslims and atheists. Saudi Arabia’s counter-terrorism law (2014) which criminalizes the “calling for atheist thought in any form, or calling into the fundamentals of the Islamic religion” is an extreme case of the broader phenomenon of using counter-terrorism laws and policies for repressive purposes.

Formal bans on particular religious groups

In 42 of the 198 countries surveyed by Pew in 2016, governments banned certain religious groups, compared to 35 countries in 2007. Examples include Baha’i in Iran; Jehova’s Witnesses in Russia, Ahmadiyah Muslims in Malaysia, Pakistan, and parts of Indonesia. The rationale for such bans includes both security-related and non-security reasons, such as “inter-religious harmony” and protecting social order.                      

Restrictions on proselytisation

According to Pew, proselytisation is restricted in 77 countries. Activities of foreign missionaries are restricted in 66 countries; 10 countries have bans on foreign missionaries. Restrictions on proselytisation are found throughout Asia, including countries like Singapore, Laos, Cambodia, Nepal, China, and across the MENA region. Questions about the limits of acceptable proselytising behaviour have long been thorny issues in international law.^[185] Such issues have taken on new urgency as globalisation poses opportunities for groups to propagate their beliefs, even as such activities may be perceived as threats to social order and as a form of competition by other religious authorities. In Sri Lanka and Myanmar, controversies over so-called “unethical conversions” have contributed to the introduction of new laws and legal initiatives to protect Buddhism and the monastic order. In both countries, legal activism on the part of Buddhist actors is informed by the human rights paradigm of religious freedom, albeit in ambiguous ways.^[186] Similar dynamics have been noted in Indonesia, where coalitions of Islamic groups have elaborated their argument for the need to maintain the Law on Blasphemy against the backdrop of a perceived threat of ‘Christianization’ on the one hand, and the prospect of losing nominal Muslims to “deviant teachings” on the other.^[187]

6.3 Critical reflections and policy responses

Given the wide variety of religious and belief minorities, the variety of religious governance arrangements and the complexity of causes that contribute to make certain minorities vulnerable to various form of discrimination, violence and persecution, it almost goes without saying that there are no simple solutions.

No size fits all – avoiding simplistic policy narratives

The project of promoting “freedom of religion or belief” and “minority rights” currently enjoys great traction in American and Western European policy circles. Much as the “return” of religion in the public sphere has been enabled by processes of globalization and economic liberalization, secular discourses of human rights are rapidly becoming globalised.^[188] As advocacy for religious freedom and minority rights has gathered momentum over the past two decades, it is important to reflect on the potentially unintended effects of such initiatives.^[189] Scholars caution that the powerful policy narrative that attributes the wide range of conflicts and injustices affecting religious minorities around the world to “religious” factors is too simplistic. Similarly with the proscription of “freedom of religion or belief” as the solution to resolve these complex challenges.

Governing social difference through the lens of religious rights – hardening boundaries?

The depiction of a group as a religious minority, as opposed to an ethnic, cultural or indigenous minority, is a form of categorisation that is fraught with both political and legal implications. Consider the Alevis, a group that Turkish authorities consider to be a heterodox variant of Sunni Islam and the European Court of Human Rights regards as a non-Sunni religious minority. According to one analyst, the Court’s recognition of Alevis as a non-Sunni Muslim minority in need of legal protection, erases the cross-cutting links between Alevi and non-Alevi communities. This may reinforce the exclusionary connections between Sunni Islam and Turkish nationalism.^[190] A different strategy has been adopted by the Shi’is of Bahrain, who have claimed to be an “indigenous group,” a strategy also adopted by some Shi’is in Saudia Arabia. Both examples illustrate that categories are fluid and that minority issues today are negotiated in a transnational context.

Several implications follow from this and similar examples.^[191] One, it is problematic to single out “religion” or “belief” as the most salient identity marker, overriding other aspects of identity. When states or outside actors seek to guarantee the rights of groups designated as religious minorities, this involves a process of “fixing” their position vis-a-vis other groups. Second, the promotion of religious minority rights in societies that lack solid democratic institutions and a firm tradition of inclusive tolerance amplifies a politics of difference in which sectarian identities may take on added social and political significance. By advocating the right to FoRB, and thus accentuating the advantages that come with identifying as a member of a particular religious group, the possibility for resentment and conflict over fundamental social norms may be enhanced. Such processes may engender a confessionalisation of public life, whereby individuals and groups are incentivised to conceive themselves in religious terms in order to obtain both tangible and less tangible benefits. In short, this raises the uncomfortable question of whether this form of HR-promotion helps to create the very circumstances that it is most keen to avoid?^[192]

The politics of categorization – who speaks for marginalised groups?

In many contexts the category “minority” is controversial. With regard to the Coptic Orthodox Christians in Egypt, Saba Mahmood (2016) notes that “the Coptic Orthodox Church in Egypt regularly proclaims that Copts do not constitute a minority, even as there is widespread agreement among Copts that they suffer from various kinds of systematic discrimination.”^[193] Copts and other long established Christian minorities in the Middle East are often reluctant to self-identify as minorities, as this status is associated with inferiority. Today, diaspora communities, NGOs and Western governments encourage groups to redefine themselves as “religious minorities.”^[194] Efforts to safeguard marginalised groups should be cognizant of such politicised dynamics. Some of this thinking is reflected in the Guidelines on the “Protection and promotion of the Rights and Freedoms of Persons belonging to Religious Minorities” (2012), which states: “The work of the Foreign Service to safeguard minorities’ religious freedom should not be confined to whether or not the group in question defines itself as a minority, but include any population group that is in practice prevented from freely practicing its religion or belief in the same way as other population groups.”  The question is how this stance is translated into concrete activities. As the promotion of FoRB tends to prioritize legal reform, a few comments about liberal, rights-based HR promotion are warranted.

Promoting inclusion along multiple axis - questioning overly legalistic approaches                   

The promoting of FoRB typically emphasises efforts to promote legal reform that involves transforming existing religious governance arrangements in line with liberal Euro-American models of legal multiculturalism and rights-based approaches.^[195] The basic idea is that the rights-based use of legal tools is the best way to accommodate cultural and religious difference.^[196] Given the strong push for legal change, it is not surprising if these efforts provoke legal counter-measures by governments and majoritarian faith groups, as indicated by recent Pew Research reports. In a globalised, media-saturated world, groups that advocate for legal counter-measures are prone to highlight “double-standards” and “hypocrisy” with respect to the treatment of Muslims in Europe. Not only are Muslims increasingly victims of Islamophobia and hate crimes, many countries have issued building regulations on mosques, ban the use Islamic of headgear in public places and in certain occupations, etc.^[197] As the legal realm has become a battleground of “lawfare,”^[198] scholars and HR-institutions are increasingly debating the “backlash” against the human rights field in parts of the Global South, Russia, Eastern Europe and what this means for human rights promotion going forward.^[199] Against this background, it is imperative to promote inclusion of religious minorities along multiple fronts and to avoid overly legalistic approaches that risk a further polarization,  both among different religious communities and within majorities and minorities within specific religious traditions.  It is also important to broaden the set of actors and institutions that are involved in the day-to-day promotion of FoRB, many of which are faith-based organizations that historically are associated with mission activities. Unless broader coalitions are built, there is a risk that efforts to promote FoRB could be seen as having a Christian bias, especially in the Middle East and in parts of Southeast Asia.

A multi-pronged strategy to counter religious radicalization

Attacks on churches, synagogues and mosques illustrate the transnational nature of sectarian conflicts and terror-networks. During the 2019 Easter Sunday attacks, the victims were Catholic worshippers and patrons of luxury hotels in Sri Lanka. A month earlier, the victims were Muslims in New Zealand, who were shot by a white supremacist as they prayed. Such atrocities are stark reminders that extremist ideologies and religious radicalism claim victims in surprising places. The main conflict lines in Sri Lanka are not among Christian and Muslims, but terror attacks and similar atrocities may help to generate such conflict, erode trust and cause spatial segregation.                    

Ecumenical initiatives to build bridges across faith-communities are important and can probably help to stem incipient radicalism. Catholic congregations on Sri Lanka have long worked to safeguard inter-religious relations, for instance by developing codes of ethics for proselytization among Christian congregations in order to de-escalate tensions between evangelical churches and Buddhist nationalists. Similarly, the major Sunni Muslim mass-organization in Indonesia, Nahdlatul Ulama (NU), has taken initiatives to discredit anti-pluralistic and violent jihadist currents. These efforts have been supported by international actors, but Nahdlatul Ulama has consistently presented their pluralistic vision of Islam Nusatenggara as a local yet cosmopolitan movement.

Local ownership is critical. Religious engagement efforts whereby donors and foreign governments enlist religious actors/groups to stem radicalization and seed democracy-friendly forms of religiosity harbours potential for creating backlash. Given the significant support to such programmes, including inter-faith dialogue, since 9/11, there is as yet surprisingly little empirical research on the success and failure of such interventions. Finally, donor support for counter-terrorism efforts should also make sure that such efforts respect the human rights of those who are enrolled in such programmes.

Acknowledgements

We are grateful for inputs and comments from the Ministry of Foreign Affairs, Norad, Arne Eide, and Siri Gloppen.

The authors are responsible for any remaining errors or omissions.

 

Notes

^[1] https://www.regjeringen.no/no/aktuelt/siv_sam/id2641121/

^[2] Sen, A. (1992). Missing women. BMJ: British Medical Journal, 304(6827), 587.

^[3] All numbers in this paragraph are from Bongaarts, J., and Guilmoto, C. Z. (2015). How many more missing women? Excess female mortality and prenatal sex selection, 1970–2050. Population and Development Review, 41(2), 241-269.

^[4] In 2010, a year when many countries did a census, the number of new missing women was about 1.7 million of both types, prenatal and postnatal.

^[5] Garenne, M. (2004). Sex ratios at birth in populations of Eastern and Southern Africa. Southern African Journal of Demography, 91-96; Bongaarts, J. and Guilmoto, C. Z. (2015). How many more missing women? Excess female mortality and prenatal sex selection, 1970–2050. Population and Development Review, 41(2), 241-269; Anderson, S. and Ray, D. (2017). Excess female mortality in Africa. Wider-WP 2017/116.

^[6] Data are from the United Nations Population Division, as reported in WDI. The report of a constant ratio in some countries, versus small fluctuations over time in others, is likely to reflect different census methodologies for estimating the population size.

^[7] Garenne (2004) demonstrates the large variation between ethnic groups in Africa and concludes that the 1.03 ratio is an average over many different ethnic groups, which is consistent with the African-American populations of the USA and the UK.

^[8] We note that Muslim Pakistan has a higher sex-ratio at birth than Hindu-dominated Nepal, indicating that religion is not a factor. Within India there is some evidence that sex-selective abortion is lower among Muslims: Bhalotra, S., Clots-Figueras, I., and Iyer, L. (2018). Religion and Abortion: The Role of Politician Identity. IZA DP No. 11292.

^[9] Ebenstein, A. (2010) The “Missing Girls” of China and the Unintended Consequences of the One Child Policy. The Journal of Human Resources, 45 (1), 87-115; Li, H., Yi, J., Zhang, J. (2011) Estimating the Effect of the One-Child Policy on the Sex Ratio Imbalance in China: Identification Based on the Difference-in Differences. Demography, 48(4), 1535-1557.

^[10] Data are from the 2014-2016 Sample Registration System, as reported in: www.censusindia.gov.in/vital_statistics/SRS_Report_2016/7.Chap_3-Fertility_Indicators-2016.pdf

^[11] Hu, L., and Schlosser, A. (2015). Prenatal Sex Selection and Girls’ Well‐Being: Evidence from India. Economic Journal, 125(587), 1227-1261.

^[12] Hu and Schlosser (2015), also see Hatlebakk (2017). For an analysis of the combined strategy, see: Anukriti, S., Bhalotra, S. R., and Tam, H. (2016). On the Quantity and Quality of Girls: New Evidence on Abortion, Fertility, and Parental Investments. IZA- DP No. 10271.

^[13] See Jayachandran, S., and Pande, R. (2017). Why are Indian children so short? The role of birth order and son preference. American Economic Review, 107(9), 2600-2629. Also see Anukriti, Bhalotra and Tam (2016), and references therein.

^[14] In 2017 the U5M rate was 3.9% for boys and 4% for girls, while in China it was 1% for boys and 0.9% for girls. For all low- and middle-income countries it was 4.5% for boys and 4% for girls. Source: World Development Indicators (WDI).

^[15] Prenatal decisions affect health conditions even when they do not lead to abortion: Bharadwaj, P., and Lakdawala, L. K. (2013). Discrimination begins in the womb: Evidence of sex-selective prenatal investments. Journal of Human Resources, 48(1), 71-113.

^[16] See Figure 1 of: Kumar, S., and Sinha, N. (2018). Preventing More “Missing Girls”: A Review of Policies To Tackle Son Preference. Policy Research Working Paper 8635. The World Bank.

^[17] https://population.un.org/wpp/Download/Standard/Population/

^[18] Guilmoto, C. Z., Saikia, N., Tamrakar, V., and Bora, J. K. (2018). Excess under-5 female mortality across India: a spatial analysis using 2011 census data. Lancet Global Health, 6(6), e650-e658.

^[19] http://censusindia.gov.in/2011-prov-results/data_files/india/Final_PPT_2011_chapter5.pdf

^[20] The table indicates an apparent surplus of men for age 40-54 in India, and a deficit for age 55-69, but for the combined range of 40-69 the sex ratio is the same as for China, at 1.04, thus possibly reflecting under-reporting of age among middle-age women in India.

^[21] Anderson, S., and Ray, D. (2010). Missing women: age and disease. Review of Economic Studies, 77(4), 1262-1300.

^[22] For a critique, see: Klasen, S., and Vollmer, S. (2013). Missing women: Age and Disease: A correction (No. 133). Courant Research Centre: Poverty, Equity and Growth-Discussion Papers.

^[23] Bongaarts, J., and Guilmoto, C. Z. (2015). How many more missing women? Excess female mortality and prenatal sex selection, 1970–2050. Population and Development Review, 41(2), 241-269.

^[24] Bongaarts, J., and Guilmoto, C. Z. (2015); Guilmoto, C. Z.  (2012); Das Gupta, M. (2015). “Missing Girls” in the South Caucasus Countries: Trends, Possible Causes, and Policy Options”. World Bank Policy Research Paper 7236.

^[25] Das Gupta, M. et al. (2017) Son Preference, Sex Ratios and ‘Missing Girls’ in Asia. In Z. Zhao & A.C. Hayes, eds. Routledge Handbook of Asian demography.  London: Routledge; Pande, R., (2003). Nutrition and Immunization in Rural India: The Role of Siblings. Demography, 40(3), 395–418; Pokhrel, S. et al., (2005). Gender Role and Child Health Care Utilization in Nepal. Health Policy, 74(1), pp.100–109.

^[26] For a perspective on how kinship systems and sex selection are linked, see Das Gupta, M. (2015), Das Gupta, M. (2010), Family systems, political systems and Asia’s ‘missing girls’: The construction of son preference and its unravelling, Asian Population Studies 6 (2), 123–152. Das Gupta, M. et al. (2003). Why is son preference so persistent in East and South Asia? A cross-country study of China, India and the Republic of Korea, Journal of Development Studies 40(2), 153–18; and Guilmoto, C.Z. (2012).

^[27] See for instance Das Gupta et al. (2003). “Why Is Son Preference So Persistent in East and South Asia? A Cross-Country Study of China, India, and the Republic of Korea.” Journal of Development Studies 40 (2), 153–87; Dudwick, N., (2015). “Missing Women" in the South Caucasus: Local Perceptions and Proposed Solutions. World Bank Policy Research Working Paper 94705, (February), 1–46; Jayachandran, S. and Kuziemko, I. (2011) Why Do Mothers Breastfeed Girls Less Than Boys: Evidence and Implications for Child Health in India, Quarterly Journal of Economics,126, 1485–1538.

^[28] Bhalotra, S., A. Chakravarty, and S. Gulesci (2016). “The Price of Gold: Dowry and Death in India,” IZA Discussion Paper 9679

^[29] Hatlebakk, M. (2017). Son preference, number of children, education and occupational choice in rural Nepal. Review of Development Economics, 21(1), 1-20.

^[30] See Bhalotra, S., Clots-Figueras, I., and Iyer, L. (2018), and references therein.

^[31] Dharmalingam, A., and Morgan, S. P. (2004). Pervasive Muslim-Hindu fertility differences in India. Demography, 41(3), 529-545.

^[32] Srivastava, N., and Srivastava, R. (2010). Women, work, and employment outcomes in rural India. Economic and political weekly, 49-63.

^[33] For a discussion of some of the complex impacts of lack of females see: Anukriti, Bhalotra and Tam (2016), and references therein.

^[34] This section is partly based on: Kumar, S., and Sinha, N. (2018). Preventing More “Missing Girls”: A Review of Policies To Tackle Son Preference. Policy Research Working Paper 8635. The World Bank.

^[35] For an overview see Kumar, S., and Sinha, N. (2018).

^[36] See Kumar, S. and Sinha, N. (2018).                                                                                                                                                  

^[37] Bongaarts, J., and Guilmoto, C. Z. (2015).

^[38] Bongaarts, J., and Guilmoto, C. Z. (2015).

^[39] See for instance Guilmoto, C. Z. (2012). Skewed Sex Ratios at Birth and Future Marriage Squeeze in China and India, 2005-2100. Demography, 49(1), 77-100; Das Gupta., M. (2016). Is Banning Sex-Selection the Best Approach for Reducing Prenatal Discrimination? Paper presented at the Population Association of America meeting, Washington DC, March 31-April 2; Li, S. (2007). Imbalanced Sex Ratio at Birth and Comprehensive Intervention in China, paper prepared for the UNFPA 4th Asia Pacific Conference on Reproductive and Sexual Health and Rights, Hyderabad, India.

^[40] For an elaboration of some of these concerns, see Kumar, S. and Sinha, N. (2018).                                                                                                                                                   

^[41] See for instance Sinha, N. & Yoong, J., (2009). Long-Term Financial Incentives and Investment in Daughters: Evidence from Conditional Cash Transfers in North India. The World Bank Policy Research Working Paper 4860: Washington, DC., (March), 1–37; Sekher, T.V. & Ram, F., (2015). Conditional Cash Transfers for Girls in India: Assessment of a Girl Child Promotion Scheme from Beneficiary Perspective. Report. International Institute for Population Sciences. For an overview of studies see Kumar, S., and Sinha, N. (2018).

^[42] Das Gupta, M. (2016), Kumar, S., and Sinha, N. (2018).

^[43] Das Gupta, M. (2015), Das Gupta, M. (2016).

^[44] Jensen and Oster (2009) Jensen, R. and Oster, E. The Power of TV: Cable Television and Women’s Status in India, Quarterly Journal of Economics, 124 (3), 1057-1094; Pande, R. P., and Astone, N. M. (2007). Explaining Son Preference in Rural India: The Independent Role of Structural versus Individual Factors. Population Research and Policy Review 26(3), 1-29; Ting, H-L., et al. (2014) “Television on Women’s Empowerment in India”, The Journal of Development Studies, 50(11), 1523-1537.

^[45] Rogers, E. M., Vaughan, P. W. Swalehe, R. M.A. Rao, N., Svenkerud, P., and Sood, S. (1999). Effects of an entertainment-education radio soap opera on family planning behavior in Tanzania. Studies in Family Planning 30(3), 193–211; La Ferrara, E., Chong, A. and Duryea, S. (2012). Soap operas and fertility: evidence from Brazil. American Economic Journal: Applied Economics 4(4), 1-31.

^[46] Keefer, P., and Khemani, S. (2012). Do Informed Citizens Receive More…or Pay More?: The Impact of Radio on the Government Distribution of Public Health Benefits, Washington DC: The World Bank Policy Research Working Paper 5952; McVey, D., and Stapleton, J. (2000). Can anti-smoking television advertising affect smoking behaviour? controlled trial of the Health Education Authority for England's anti-smoking TV campaign, Tobacco Control 9(3), 273-82; DellaVigna, S., and Kaplan, E. 2007. The Fox News Effect: Media Bias and Voting, Quarterly Journal of Economics, 122 (3), 1187–1234.

^[47] For an experimental study of endorsement effects and the need to tailor programs, see Muriaas, R., Wang, V., Benstead, L., Dulani, B., and Rakner, L. (2018) When the Gender of Traditional Authorities Matters: Intersectionality and Women’s Rights Advocacy in Malawi. Comparative Political Studies, First published online May 28, 2018.

^[48] Deininger, K. et al. (2010) Inheritance Law Reform and Women’s Access to Capital: Evidence from India’s Hindu Succession Act, Washington DC: The World Bank Policy Research Working Paper 5338; Roy, S. (2011) Empowering Women: Inheritance Rights and Female Education in India, Department of Economics, University of Warwick, Working Paper No.46; Deininger, K. et al. (2014) Inheritance Law Reform, Empowerment, and Human Capital Accumulation : Second-Generation Effects from India”, Washington DC: The World Bank Policy Research Working Paper 7086.

^[49] For more on the Republic of Korea and lessons learned see Kumar, S. and Sinha, N. (2018).

^[50] Ebenstein, A. (2014). Patrilocality and Missing Women. Patrilocality and Missing Women (April 7, 2014). Available at SSRN: https://ssrn.com/abstract=2422090 or http://dx.doi.org/10.2139/ssrn.2422090

^[51] Edlund, L. and Lee, C. Son Preference, Sex Selection and Economic Development: The Case of South Korea. NBER Working Paper No. 18679.

^[52] Pande, R. P., and Astone, N. M. (2007); Duflo, E., (2000). Child Health and Household Resources in South Africa: Evidence from the Old Age Pension Program. American Economic Review, 90(2), 393–398.; Luke, N. and Munshi, K., (2011). Women as Agents of Change: Female Income and Mobility in India. Journal of Development Economics, 94(1), 1–17; Qian, N., (2008). Missing Women and the Price of Tea in China: The Effect of Sex-Specific Earnings on Sex Imbalance. The Quarterly Journal of Economics, 123(3), 1251–1285.

^[53] For positive symbolic representation effects, see e.g. Alexander, A.C. and Jalalzai, F. (2018). Symbolic empowerment and female heads of states and government: a global, multilevel analysis. Politics, Groups, and Identities. Published online 25 Feb 2018; Schwindt-Bayer, L.A. and Mishler, W. (2005). An Integrated Model of Women’s Representation. The Journal of Politics, 67(2), 407-428; Schwindt-Bayer, L.S. (2010). Political Power and Women’s Representation in Latin America. Oxford: Oxford University Press; Burnet, J. (2011). Women Have Found Respect: Gender Quotas, Symbolic Representation, and Female Empowerment in Rwanda. Politics & Gender, 7(3), 303-334.

^[54] Beaman, L., Chattopadhyay, R. et al. (2009). Powerful Women: Does Exposure Reduce Bias? Quarterly Journal of Economics, 124(4), 1497-1540; Beaman, L. et al., (2012). Female Leadership Raises Aspiration and Educational Attainment for Girls: A Policy Experiment in India. Science, 335 (6068), 582-586. See also Kalsi, P. (2017) Seeing is Believing – can increasing the number of female leaders reduce sex selection in rural India? Journal of Development Economics, 126, 1-18.

^[55] For studies finding a positive link, see e.g. Clayton, A. et al. (2018) In Whose Interest? Gender and Mass-Elite Priority Congruence in Sub-Saharan Africa, Comparative Political Studies. First published online March 19, 2018; Clayton, Josefsson and Wang (2017) Quotas and women’s substantive representation: Evidence from a content analysis of Ugandan plenary debates. Politics & Gender, 13, 276-304; Chattopadhyay, R. and Duflo, E. (2004) Women as Policy Makers: Evidence from a Randomized Policy Experiment in India, Econometrica, 72(5), 1409-1443. For overviews of the literature see Franceschet, S. and Piscopo, J.M. (2008). Gender Quotas and Women’s Substantive Representation: Lessons from Argentina. Politics and Gender 4, 393-425; Wängnerud, L. (2009). Women in Parliaments: Descriptive and Substantive Representation. Annu. Rev. Political Sci. 12, 51-69.

^[56] Chattopadhyay, R. and Duflo, E. (2004).

^[57] Sen, A. (1985). Commodities and capabilities. Amsterdam, New York: North-Holland.

^[58] Lenhardt, A. and Samman, E. (2015). In quest of inclusive progress Exploring intersecting inequalities in human development, Development Progress Research Report 4. London: Overseas Development Institute.

^[59] “Intersectionality” was first discussed in the context of race and gender in the US (Crenshaw, K. (1991). Mapping the Margins: Intersectionality, Identity Politics, and Violence against Women of Color, Stanford Law Review, 43(6), 1241-1299). For a discussion in the context of poverty, see N. Kabeer (2010). Can the MDGs provide a pathway to social justice? The challenge of intersecting inequalities. New York: United Nations Development Programme.
 

^[60] The new constitution in Nepal is illustrative: “The economically, socially or educationally backward women, Dalit, indigenous nationalities, Madhesi, Tharu, Muslims, backward classes, minorities, marginalised communities, persons with disabilities, gender and sexual minorities, farmers, labourers, oppressed or citizens of backward regions and indigent Khas Arya shall have the right to participate in the State bodies on the basis of principle of proportional inclusion” (Article 42).

^[61] The “Yogyakarta Principles,” or “Principles,” is a statement concerning the “application of international human rights law in relation to sexual orientation and gender identity” adopted by representatives from various non-governmental organizations and United Nations treaty monitoring committee members following a November 2006 conference held in Yogyakarta, Indonesia. https://www.pgaction.org/inclusion/pdf/handbook/annex-1.pdf

^[62] Lucas Ramon Mendos (2019). State sponsored Homophobia. ILGA. Available at: https://ilga.org/downloads/ILGA_State_Sponsored_Homophobia_2019.pdf

^[63] For example, although it is only sodomy which is explicitly criminalized in Sudan, it is lesbian and bisexual women and transgender women who are most vulnerable to prosecution. See L. Tønnessen and S. Al-Nagar (2019). “Activism from the Closet: Fear of Double Backlash towards a Nascent LGBT Movement in Sudan”; S. Gloppen, A. Msosa and F. Viljoen eds. (forthcoming) Queer Lawfare in Africa.  Pretoria, PULP press. The publication is part of the CMI project Political determinants of sexual and reproductive health: Criminalisation, health impacts and game changers (2016-2020) funded by GLOBVAC, Research Council of Norway.

^[64] See for example P. R. Ireland.2013. A Macro-Level Analysis of the Scope, Causes, and Consequences of Homophobia in Africa. African Studies Review, 56(02), 47-66.

^[65] E. Han  and J. O’Mahoney (2014). British colonialism and the criminalization of homosexuality. Cambridge Review of International Affairs 27(2), 268-288. There are of course exceptions to this rule as former French colonies like Senegal and Cameroon have criminalised homosexuality.

^[66] S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford C, & Maine A (eds) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar.

^[67] Ibid.

^[68] A. Jjuuko (2018). Beyond court victories: Using strategic litigation to stimulate social change in favour of lesbian, gay and bisexual persons in Common Law Africa (PhD Dissertation University of Pretoria); V. Kolmannskog (2018). Love in Law – The Indian Supreme Court decides in favour of LGBT persons. Bergen: Chr. Michelsen Institute (CMI Brief no. 2018:06); B. M. Wilson and C. Gianella-Malca (2019). Overcoming the Limits of Legal Opportunity Structures: LGBT Rights’ Divergent Paths in Costa Rica and Colombia, Latin American Politics and Society, 61(2), 138-163; Gianella, C., and B. M. Wilson (2016). 4 LGBTI rights, The Latin American Casebook: Courts, Constitutions, and Rights, 60.
These publications are all part of the UiB-CMI project Sexual and reproductive rights lawfare: Global battles (2014-2018) funded by the Research Council of Norway. https://www.cmi.no/projects/1836-sexual-and-reproductive-rights-lawfare.

^[69] For a summary of the literature on politicisation in Africa, see S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford, C., & Maine, A. (eds) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar. This is written as part of two CMI-UiB projects funded by the research council of Norway Sexual and reproductive rights lawfare and Political Determinants of Sexual and Reproductive Health.

^[70] C. Gianella, M. Rodriguez de Assis Machado, and S. Gloppen (2017). Political determinants of Sustainable Development Goals, The Lancet 390.10112, 2545-2546; C. Gianela, M. Rodriguez de Assis Machado, A. Peñas Defago (2017). “What causes Latin America’s high incidence of adolescent pregnancy?” Bergen: Chr. Michelsen Institute (CMI Brief vol. 16 no. 9).

^[71] See for example G. Grossman (2015). Renewalist Christianity and the Political Saliency of LGBTs: Theory and Evidence from Sub-Saharan Africa, The Journal of Politics, 77(2), 337-351.

^[72] See for example Bradley Demone (2016). LGBT Rights in Malawi: One Step Back, Two Steps Forward? The Case of R v Steven Monjeza Soko and Tiwonge Chimbalanga Kachepa,  Journal of African Law, 60(3), 365-387.

^[73] Pew Research Center (2013). The Global Divide on Homosexuality: Greater Acceptance in More Secular and Affluent Countries. Available online:  https://www.pewresearch.org/wp-content/uploads/sites/2/2014/05/Pew-Global-Attitudes-Homosexuality-Report-REVISED-MAY-27-2014.pdf

^[74] P. Awondo, P. Geschiere, and G. Reid (2013). Homophobic Africa? Toward More Nuanced View, African Studies Review, 55(3), 145-168.

^[75] This is based on Afrobarometer data from 2016. A summary of findings can be found here: http://afrobarometer.org/sites/default/files/publications/Dispatches/ab_r6_dispatchno74_tolerance_in_africa_eng1.pdf

^[76] R. B. Mathisen (2018). A Postmaterialist Explanation of Homophobia in Africa: Multilevel Analysis of Attitudes Towards Homosexuals in 33 African Countries. Master thesis. Available online: https://www.lawtransform.no/wp-content/uploads/2018/08/Master-2.0-version-3-3-min.pdf
The master thesis is part of the CMI-UiB project Political determinants of sexual and reproductive health

^[77] Ibid.

^[78] See for example S. Tamale (2013). Confronting the Politics of Nonconforming Sexualities in Africa, African Studies Review, 56(2), 31-45.

^[79] See a summary of the literature and its blind spots in D. Meyer.  Violence Against Queer People: Race, Class, Gender, and the Persistence of anti-LGBT discrimination. Rutgers University Press.

^[80] For a summary of this literature, see C. R. Cornelis and O. R. Chabalala (2015). Phenomenological description of corrective rape and a new terminological perspective of the phenomenon, Acta Criminologica: Southern African Journal of Criminology, 28(3), 50-62

^[81] Makofane, K., Beck, J., Lubensky, M., & Ayala, G. (2014). Homophobic legislation and its impact on human security, African Security Review, 23(2), 186-195.

^[82] S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford C, & Maine A. (eds.) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar.

^[83] King, M., Semlyen, J., Tai, S.S., Killaspy, H., Osborn, D., Popelyuk, D., Nazareth, I. (2008). A systematic review of mental disorder, suicide, and deliberate self-harm in lesbian, gay and bisexual people. BMC Psychiatry doi: 10.1186/1471-244X-8-70

^[84] H. Hafeez, M. Zeshan, M. A. Tahir, N. Jahan, and S. Naveed, A systematic review of mental disorder, suicide, and deliberate self-harm in lesbian, gay and bisexual people, BMC Psychiatry,  doi: 10.1186/1471-244X-8-70

^[85] S. H. Cook, T. G. M. Sandfort, J. A. Nel, and E. P. Rich (2013). Exploring the Relationship Between Gender Nonconformity and Mental Health Among Black South African Gay and Bisexual Men, Sexual Behavior, 42(3), 327–330.

^[86] C. Beyer, P. Sullivan, J. Sanches, S. Baral, C. Collins, A. Wirtz, D. Altman, G. Trapence, and K. Mayer (2013). The increase in global HIV epidemics in MSM, AIDS, 27(17), 2665–2678. doi: 10.1097/01.aids.0000432449.30239.fe

^[87] A. Müller, S. Spencer, T. Meer, and K. Daskilewicz (2018). The no-go zone: a qualitative study of access to sexual and reproductive health services for sexual and gender minority adolescents in Southern Africa, Reproductive Health, 15(12). Published online 2018, Jan 25. doi: 10.1186/s12978-018-0462-2

^[88] Sekoni A. O., Jolly K., Gale N.K., Ifaniyi O.A., Somefun E.O., Agaba EI, Fakayode V. A. (2016). Provision of Healthcare Services to Men Who Have Sex with Men in Nigeria: Students' Attitudes Following the Passage of the Same-Sex Marriage Prohibition Law,  LGBT Health, 3(4), 300-7. doi: 10.1089/lgbt.2015.0061.

^[89] Poteat T., Diouf D., Drame F.M., Ndaw M., Traore C., et al. (2011). HIV risk among MSM in Senegal: a qualitative rapid assessment of the impact of enforcing laws that criminalize same sex practices, PLoS ONE, 6: e28760. https://doi.org/10.1371/journal.pone.0028760

^[90] A. Müller (2016). Health for All? Sexual Orientation, Gender Identity, and the Implementation of the Right to Access to Health Care in South Africa, Health and Human Rights Journal, 18(2), 195–208.

^[91] C. Rosario and C. Gianella (forthcoming) “Xhi esse mundo djon, ja n está prestar" – progressive legislation in the context of generalised conservative public opinion. The case of LGBT Rights in Mozambique, in S. Gloppen, A. Msosa, and F. Viljoen (eds.) Queer Lawfare in Africa. Pretoria: PULP press.

^[92] S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford C., & Maine A. (eds.) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar.

^[93] See for example,  R. Depalma. South African Life Orientation Teachers: (Not) Teaching About Sexuality Diversity, Journal of Homosexuality, 61(12), 1687-1711. doi: 10.1080/00918369.2014.951256; R. Zahn, A. Grosso, A. Scheibe, L.-G. Bekker, S. Ketende, F. Dausab, S. Iipinge, C. Beyrer, G. Trapance, S. Baral (2016). Human Rights Violations among Men Who Have Sex with Men in Southern Africa: Comparisons between Legal Contexts, PLoS ONE, 11(1), e0147156. https://doi.org/10.1371/journal.pone.0147156; A. Müller (2016). Health for All? Sexual Orientation, Gender Identity, and the Implementation of the Right to Access to Health Care in South Africa, Health and Human Rights Journal, 18(2), 195–208.

^[94] S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford C, & Maine A. (eds.) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar.

^[95] A. Kretz (2017). Aid Conditionality as (Partial) Answer to Antigay Legislation: An Analysis of British and American Foreign Aid Policies Designed to Protect Sexual Minorities, Vienna Journal on International Constitutional Law, 7(4) DOI: https://doi.org/10.1515/icl-2013-0404

^[96] M. Wahman and A. C. Drury (2018). Leverage, diplomacy, and African lesbian, gay, bisexual, transgendered, and intersex rights: Malawi and Zambia compared, Journal of Human Rights, 17(5), 622-641. University of Missouri
View further author information Michigan State University; Correspondencewahmanmi@msu.edu
View further author information

^[97] P. Awondo, P. Geschiere, and G. Reid (2013). Homophobic Africa? Toward More Nuanced View, African Studies Review, 55 (3), 145-168.

^[98] See for example, C. Biruk (2014). Aid for gays’: the moral and the material in ‘African homophobia’ in post-2009 Malawi, The Journal of Modern African Studies, 52(3), 447-473.

^[99] H. Seckinelgin (2018). Same-sex lives between the language of international LGBT rights, international aid, and anti-homosexuality, Global Social Policy, 18(3), 284-30. to clipboardShare URL copied to clipboardShow all authors

^[100]   M. Epprecht (2012). Sexual minorities, human rights and public health strategies in Africa, African Affairs, 111(443), 223–243, https://doi.org/10.1093/afraf/ads019

^[101] L. Kojoué (2017). LGBT rights in a Republic of Therapy. HIV/AIDS policies and the redefinition of citizenship in Cameroon, Critical African Studies, 9(1), 91-105, DOI: 10.1080/21681392.2017.1284010

^[102] S. Gloppen and L. Rakner (2019). LGBT-rights in Africa, in Ashford C, & Maine A. (eds.) Research Handbook on Gender, Sexuality and the Law. London: Edward Elgar.

^[103] https://www.un.org/disabilities/documents/convention/convoptprot-e.pdf. (Article 1.)

^[104] Disability Data Review in 40 countries (2018) (p. 32). Leonard Cheshire. https://www.disabilitydataportal.com/fileadmin/uploads/lcdp/Documents/Disability_Data_Review_Extended_Summary_-_A_collation_and_analysis_of_disability_data_from_40_countries.pdf. Accessed 11 April.

^[105] World Disability Report (2011). The WHO and the World Bank.

^[106] World Disability Report (2011), Appendix A.

^[107] Disability Data Review (2018), p. 36.

^[108] S. Mitra (2018). Disability, Health and Human Development. Palgrave Macmillan. When lowering the threshold of disability to include everyone with “some difficulty” in at least one dimension, the prevalence is 11%-15%.

^[109] World Disability Report (2011), p. 289.

^[110] B. Altman (ed.) (2016) International Measurement of Disability; Purpose, Method and Applications. Springer: Social Indicators Research Series 61. Referenced in Disability Data Review.

^[111] World Disability Report (2011), p. 27.

^[112] Ibid., Appendix C.

^[113] Mitra (2018); the Sintef studies discussed below.

^[114] See Mitra (2018), p. 69., for estimates of onset of severe disability in Tanzania and Uganda. 

^[115] Mitra, S., Posarac, A., & Vick, B. (2013). Disability and poverty in developing countries: A multidimensional study. World Development, 41, 1–18.

^[116] World Disability Report (2011), Figure 2.1.

^[117] A. H. Eide and A. Munthali (2018). Living conditions among persons with disabilities in Malawi. A National, representative survey 2017. Sintef report 2018:00695.
Zambia National Statistics Survey 2015. Central Statistical Office, Lusaka, Zambia. December 2018.
A. H. Eide, S. Neupane, K.‐G. Hem (2006). Living conditions among people with disability in Nepal. Sintef report A27656.

^[118] There is clearly a potential for preventing certain types of disabilities, but we did not look further into this issue.

^[119] See also Mitra (2018), p. 68., using data from Ethiopia, Uganda, Tanzania, and Malawi. 

^[120] Sintef reports: Malawi p. 82, Nepal p. 110.

^[121] S. Mizunoya, S. Mitra, I. Yamasaki (2018). Disability and school attendance in 15 low- and middle-income countries. World Development, 104:388-403.

^[122] Filmer, D. (2008). Disability, poverty, and schooling in developing countries: Results from 14 household surveys, The World Bank Economic Review, 22(1), 141–163.

^[123] Arne H. Eide, et al. (2011). Living conditions among people with disabilities in developing countries, in A. H. Eide and B. Ingstad (eds.) Disability and poverty. A global challenge Bristol, UK: Policy Press.

^[124] World Disability Report (2011), p. 207.

^[125] Ibid., p. 62.

^[126] A higher level of unmet need is reported in Nepal (19.2%), but comparable figures for non-disabled are not available.

^[127] Trani, J., & Loeb, M. (2010). Poverty and disability: a vicious circle? Evidence from Afghanistan and Zambia. Journal of International Development, 24(1), S19–S52.

^[128] World Disability Report (2011), Table 3.2.

^[129] Ibid. p. 63.

^[130] Ahumuza S.A., Matovu K.B.J., Ddamulira J.B., & Muhanguzi K.F. (2014). Challenges in accessing sexual and
reproductive health services by people with physical disabilities in Kampala, Uganda, Reproductive
Health, 11:59. doi: 10.1186/1742-4755-11-59 PMID: 25086444

^[131] Ganle J.K., Otupiri E., Obeng B., Edusie A.K., Ankomah A., Adanu R. (2016). Challenges Women with Disability Face in Accessing and Using Maternal Healthcare Services in Ghana: A Qualitative Study, PLoS ONE, 11(6): e0158361

^[132] Mitra (2018), p. 71.

^[133] Employment includes both paid work and self-employment.

^[134] Mizunoya, S., and Mitra, S. (2013). Is there a disability gap in employment rates in developing countries? World Development, 42:28–43. A more recent study in four low-income countries finds somewhat bigger disability gaps in employment for the severely disabled (Mitra (2018), p. 104.)

^[135] Ibid. & World Disability Report (2011), Table 8.1.

^[136] Mitra (2013).

^[137] Ibid. 

^[138] World Disability Report (2011), Table 8.1.

^[139] S. Mitra, U. Sambamoorthi (2008). Disability and the Rural Labor Market in India: Evidence for Males in Tamil Nadu, World Development, 36(5), 934-952.

^[140] Baldwin, M. L. & Johnson, W. G. (2005). A Critical Review of Studies of Discrimination against Workers with Disabilities. In W. M. Rodgers III (Ed.), Handbook on the Economics of Discrimination (pp.119-160). Cheltenham, UK: Edward-Elgar Publishing.

^[141] See overview in Mitra (2013).

^[142] Mitra (2013).

^[143] Another study based on the WHS data, using the broad WHS measure of disability, finds significant differences in disability prevalence between the poorest and the richest quintile in around 50% of the 49 countries studied (after controlling for confounders). See A. R. Hosseinpoor, et al. (2013). Socioeconomic Inequality in Disability Among Adults: A Multicountry Study Using the World Health Survey, American Journal of Public Health, 103, 1278-1286.
Theoretically, it is possible that the socio-economic gradient in disability prevalence is larger with a broad definition of disability; it seems likely that many of the disabilities the poor acquire more easily than the rich are mild and moderate disabilities.

^[144] In Nepal, the difference between the urban and the rural is smaller than in the African countries but still bigger than the difference between the disabled and the non-disabled.

^[145] J.-F. Trani and M. Loeb. Poverty and disability: a vicious circle? Evidence from Afghanistan and Zambia,
Journal of International Development. Referenced, in N. Groce et al. (2011). Disability and Poverty: the need for a more nuanced understanding of implications for development policy and practice, Third World Quarterly, 32(8), 1493-1513.

^[146] The first part of section draws heavily on the literature review by B. Rohwerder (2018). Disability Stigma in Developing Countries. K4D Helpdesk Report, Brighton, UK: Institute of Development Studies. 

^[147] G. Azalde and S. H. Braathen (2018). The role of stigma in accessing education for people with disabilities in low and middle-income countries: a review of the evidence. SINTEF Technology and Society.

^[148] K. Scior et al. (2015). Intellectual Disabilities: Raising Awareness and Combating Stigma – Global Review. University College London.

^[149] B. Rohwerder (2018). Disability Stigma in Developing Countries. K4D Helpdesk Report, Brighton, UK: Institute of Development Studies

^[150] Werner, S. and Scior, K. (2016). Interventions Aimed at Tackling Intellectual Disability Stigma: What Works and What Still Needs to Be Done. In K. Scior and S. Werner (eds.). Intellectual Disability and Stigma: Stepping Out from the Margins. Palgrave Macmillan.

^[151] K. Scior et al. (2015).

^[152] See references in Rohwerder (2018), p. 19.

^[153] McConkey, R., Kahonde, C., and McKenzie, J. (2016). Tackling Stigma in Developing Countries: The Key Role of Families. In K. Scior and S. Werner (eds.). Intellectual Disability and Stigma: Stepping Out from the Margins. Palgrave Macmillan.

^[154] See Rohwerder (2018).

^[155] A. Croft (2013). Promoting access to education for disabled children in low-income countries: Do we need to know how many disabled children there are? International Journal of Educational Development, 33(3), 233-243.

^[156] J. Tripney et al. (2017). Employment Supports for Adults with Disabilities in Low- and Middle-Income Countries: A Campbell Systematic Review, Research and Social Work Practice, 1-13.

^[157] Acemoglu D., Angrist J. (2001). Consequences of employment protection? The case of the Americans with Disabilities Act, The Journal of Political Economy, 109, 915-957.

^[158] Bell D., Heitmueller A. (2009). The Disability Discrimination Act in the UK: helping or hindering employment among the disabled? Journal of Health Economics, 28, 465-480.

^[159] Crowther R.E. et al. (2001). Helping people with severe mental illness to obtain work: systematic review, BMJ (Clinical Research Ed.), 322, 204-208.

^[160] Bjorvatn, K. and Tungodden, B., Empowering the Disabled Through Savings Groups: Experimental Evidence from Uganda. (March 14, 2018). NHH Dept. of Economics Discussion Paper No. 5/2018. http://dx.doi.org/10.2139/ssrn.3143674
 

^[161] Minority Rights: International Standards and Guidance for Implementation, p. 2, United Nations, 2010.

^[162] The nationality criterion in the above definition has often been challenged. This is relevant in connection to statelessness. This status that can occur for several reasons, including discrimination of ethnic or religious groups. The UN Refugee Agency (UNHCR) has stepped up efforts to end statelessness by 2024.

^[163] See for instance A. Longva (2012). Introduction: Domination, Self-Empowerment, Accomodation. Religious Minorities in the Middle East, A. Longa and A. Roald (eds). Pp.1-23. Leiden and Boston: Brill.

^[164] That this report mainly considers minorities in relation to violation of FoRB also reflects that this focus informs current policies, as spelled out in the Guidelines on the ‘Protection and Promotion of the Rights and Freedoms of Persons belonging to Religious Minorities’ (2013) issued by the MFA.

^[165] The 1966 International Covenant on Civil and Political Rights (ICCPR), which is binding for the 172 countries that have ratified it, reaffirms these rights, adding the right of religious minorities to practice their own religion.

^[166] During the drafting of the UDHR the inclusion of “belief” was a contentious matter. As Lerner notes, “A major politically motivated confrontation between Western countries and Communist countries was avoided by inserting ‘belief’ after ‘religion.” Lerner N. (2004). The Nature and Minimum Standards of Freedom of Religion or Belief. Facilitating Freedom of Religion and Belief: A Deskbook. T. Lindholm et al. (eds). Leiden: Martinus Nijhoff Publishers.

^[167] See for instance, T. Brekke (2011), Fundamentalism: Prophecy and Protest in an Age of Globalization. Cambridge: Cambridge University Press.

^[168] The UN Special Rapporteur on Freedom of Religion and Belief, an independent expert appointed by the Human Rights Council, produces annual reports, but these reports have not been consulted.

^[169] A. Hirschman (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge: Cambridge University Press.

^[170] According to media reports about the interim report commissioned by the UK Foreign Secretary to document religious persecution, the report used the language of “near genocide” when speaking of the situation of Christians in the Middle East, see Guardian Weekly, May 2, 2010.

^[171] Pew Research Center, (2018) Global Uptick in Restrictions on Religion in 2016.

^[172] Pew Research Center, (2015). Latest Trends in Religious Restrictions and Hostilities.

^[173] Persecution refers (to risk) of physical attacks threatening the life and freedom of individuals and groups. For acts to constitute persecution, they should be systematic and ongoing.

^[174] Independent International Fact-Finding Mission on Myanmar, (2018).

^[175] Keeping the Faith: A Study of Freedom of Thought, Conscience, and Religion in ASEAN. Human Rights Resource Centre (2015); NN Kyaw (2018). Myanmar’s Other Muslims: The Case of the Kaman. Citizenship in Myanmar: Ways of Being In and From Burma, A. South and M. Lall (eds). Singapore: Institute of Southeast Asian Studies (ISEAS).

^[176] UN Committee on the Elimination of Racial Discrimination, August 20. (2018).

^[177] Pew’s Social Hostilities Index uses a 10-point scale based on 13 measures involving religion.

^[178] World Report 2019: Nigeria. Human Rights Watch.

^[179] Violent Cow Protection in India, Human Rights Watch Report (2019).

^[180] See for instance, R. Kapur (2014), A Leap of Faith: The Construction of Hindu Majoritarianism through Secular Law. The South Atlantic Quarterly,113(1), Winter 2014.

^[181] R. Moon, Putting Faith in Hate: When Religion is the Source and Target of Hate Speech (2018). Cambridge University Press.

^[182] The bi-partisan US Commission on International Freedom (USCIRF) identifies Iran, Pakistan, Yemen, Somalia, Qatar, Egypt as the countries “where blasphemy provisions deviate the most from international human rights principles maintain an official state religion.” Respecting Rights? Measuring the World’s Blasphemy Laws (2017) p.1). The report is, however, of questionable quality and is clearly intended to advocate for the repeal of blasphemy laws. 

^[183] See Keeping the Faith: A Study of Freedom of Thought, Conscience, and religion in ASEAN (2015) for in-depth country analysis of all ASEAN countries. Sponsored by the Norwegian Embassy, this report by the Human Rights Resource Centre, is an extremely valuable resource produced by researchers from within and outside the region.

^[184] K. Telle (2018), Faith on Trial: Blasphemy and ‘Lawfare’ in Indonesia, Ethnos, 83 (2):371-391; https://www.tandfonline.com/doi/full/10.1080/00141844.2017.1282973; K. Telle (2018b), Blasphemy and ‘hate spin’ campaigns in Indonesia. https://tif.ssrc.org/2018/12/14/blasphemy-and-hate-spin-campaigns-in-indonesia/

^[185] The adoption of the 1981 Declaration on the Elimination of All Forms of Intolerance and Discrimination on Religion or Belief was delayed by controversies over proselytism, see Lerner (2004).

^[186] I. Frydenlund (2017), Religious Liberty for Whom? The Buddhist Politics of Religious Freedom during Myanmar’s Transition to Democracy, Nordic Journal of Human Rights.35(1):55-73; B. Schontal, (2014), Constitutionalizing  Religion:The Pyrrhic Success of Religious Rights in Post-Colonial Sri Lanka. Journal of Law and Religion, 29(3):470-90.

^[187] Finucane, J. and M. Feener, (2014) (eds). Introduction. Prozelytizing and the Limits of Religious Pluralism in Contemporary Asia. ARI -Springer Asia Series 4. Springer.

^[188] Contemporary Religiosities: Emergent Socialities and the Post-Nation-State (2010). Kapferer, Telle, Eriksen (eds). Berghahn Books: New York and London.

^[189] For more elaboration, see K.Telle, (2018). Promoting ‘Freedom of religion or belief’ and minority rights: Policy, practice and research. CMI Report. Commissioned by the MFA, (37 pp).

^[190] Hurd, E. S. (2015). Beyond Religious Freedom: The New Global Politics of Religion. Princeton: Princeton U Press.

^[191] In 2016, segments of the Syrian Alawite community distanced themselves from the Assad regime by redefining themselves as belonging to the Abrahamic fait and stating that they are not a branch of Shia Islam. BBC, April 2016.

^[192] S. Hopgood, J. Snyder, L.Vinjamuri (eds), (2017). Human Rights Futures.  Cambridge: Cambridge University Press.

^[193] Mahmood, S. (2016). Religious Difference in a Secular Age: A Minority Report. Princeton: Princeton U. Press.

^[194]   A. Longva and A. Roald (eds.) (2012). Christian Minorities in the Middle East. Brill.

^[195]   José Casanova notes that “religious freedom is becoming a universal aspiration,” but also caution that such policies may “conflict with other cultures’ understanding of the free exercise of religion, and they will be resisted accordingly.” Globalization and the Free Exercise of Religion Worldwide. In Challenges to Religious Freedom in the Twenty-First Century (2012), G. Bradly (ed).Cambridge: Cambridge University Press.

^[196]   On legal multiculturalism, see B. J. Berger (2010), The Cultural Limits of Legal Tolerance. After Pluralism: Reimaging Religious Engagement. C. Bender and P. Classen (eds). New York: Colombia University Press.

^[197] The European Court of Human Rights has, in fact, ruled that bans on Muslim head coverings do not violate the religious freedom of Muslim women. See M. Fernando (2012), ‘Belief and/in the Law.’ Method and Theory in the Study of Religion, vol 24(71):71-80. For a more quantitative overview, see J. Fox and Y. Akbaba (2015), Securitization of Islam and Religious discrimination: Religious Minorities in Western Democracies, 1990- 2008.  Comparative European Politics, vol 13 (2),175-197.

^[198] See J. Comaroff (2010), ‘Reflections on the rise of Legal Theology’ for an analysis of how faith communities in Africa and the Middle East resort to ‘lawfare’ to assert their sovereignty. Contemporary Religiosities (2010), B.Kapferer, K.Telle, A. Eriksen (eds). Oxford: Berghahn Books.

^[199] S. Hopgood, J. Snyder, L.Vinjamuri (eds), (2017). Human Rights Futures.  Cambridge: Cambridge University Press.