1 Introduction: Poor Man’s Religion?

There is a significant risk that social and political conflicts arising from the changing religious composition of states could increase over time, especially when overlapping with high inter-community levels of social inequality. This is particularly important in an era of high migration and significant fertility gaps, which results in differences in age structures and different population growth. In this chapter, we focus on social inequality, poverty, and one central dimension of cultural identity, religion. We therefore study social inequality in a different context than the geographic entities often considered—such as countries or subnational areas. Understanding the interlinkages between religion and poverty is relevant to several domains, including to assess how (a) poverty levels across religions coincide with religious tensions, conflicts and societal divisions around the world, (b) policies, which may be most effective at poverty alleviation, could depend on the religious composition of the economically deprived, (c) poverty can both draw some to religion when financial, social or moral support and a sense of belonging could be found, however, how it could also alienate others from religious beliefs, and (d) behaviours that can be grounded in religious convictions may alleviate or perpetuate or even raise poverty.

To describe and understand the relationships between economic deprivation and religious belief, one needs to assess faith and economic deprivation in tandem. Until now, there has not been any worldwide assessment on the relationship between religion and poverty. This chapter seeks to do so by combining unique global data on religious affiliation and economic inequality—and by assessing trends over time. We present new data and estimates on poverty grouped by religious belief worldwide, and we combine religious and income distribution data to produce a distribution of poverty by religion and country. Our sample comprises 152 countries and encompasses 95% of world population. Apart from small (less than 500,000 inhabitants) states, we provide estimates for all major countries. See Table A.4.1 in the online appendix for a list of countries that we include.

Demography can directly or indirectly affect political processes, particularly when different groups have opposing social and economic interests: it can serve as a catalyst or a conditioning factor for political decisions. Population change needs to be considered as a political force in its own right. The unprecedented global demographic turbulence created by the accelerated rise in world population of the last 200 years is likely to crest around the middle of this century. The next few decades will present substantial changes in long-term demographic trends, including the shrinkage of Europe’s labour force, the extreme ageing of advanced industrial societies, as well as the advanced ageing of China, a global shift from rural to urban habitation and a substantial turn in global economic growth towards the developing world, where 9 out of every 10 of the world’s children now live, 22% of those in India alone (UN, 2019).

Inequality can be a powerful driver of political action. In contexts where economic inequality levels are high (and perceived as unfair), those at the bottom levels of the income distribution who are (or feel) disadvantaged may try to alter the system, while others who benefit from the current system may try to maintain the status-quo. If large enough, groups are constrained with few economic opportunities, low levels of wealth and lack of social mobility, it can act as a mobilizing force for political action that could lead to political instability and possibly societal change and reordering. In an era with public data on income distributions across the world and social media and television, it is increasingly clear that people compare their own wealth and well-being to others not only within their own regional community and country, but also within and across comparison groups with others, including people of the same or different religious denomination.

A key marker of group identity is one’s religious affiliation (Castells 2003; Goujon et al., 2006). Religion and religious denominations have been suggested to be key markers of conflict lines in terms of, for instance, attitudes towards same-sex marriage or attitudes towards abortion (Halman & van Ingen, 2015; Norris & Inglehart, 2011; van den Akker et al., 2013). Religion and the degree of religiosity are often also strong markers of political behaviour, influencing whether one votes, which political allegiances one has and which parties one chooses to vote for (Gayte et al., 2017; Gerber et al., 2016).

Population increases are fastest among those who are more religious and at the lower end of the income distribution globally. Demographic growth in the world is mainly taking place among those with less education and lower income, while those with higher education and greater income tend to have close to replacement fertility (Madsen et al., 2018; Skirbekk, 2008; Vogl, 2015). Demographic growth is mainly taking place among certain religious groups such as Muslims and Hindus, while it is negative among those with no religious affiliation (Stonawski et al., 2015). The relations between demography and politics are complex: demography can drive political and religious behaviours (e.g. high fertility can lead to more conservative views), and population groups can drive political processes (greater degree of religious populations can vote more conservative parties). One of the drivers of demographic behaviour is religion, a cultural institution pervasive globally and with a very high impact on political processes across societies.

Poverty does not take place in a vacuum, but rather within a set of defined group identities, which help delineate identity within a subset of individuals across a variety of domains (Banks, 2008; Huddy, 2015; Sachdeva, 2016). Such group identities could lead to motivation for collective action, help motivate and coordinate political action that could help to reduce poverty. Group memberships could, however, also lead to apathy and disenchantment, create a victim-based explanation for why development has led to one being poor and lead to fatalistic views that entail a belief that one cannot escape poverty and thereby demotivate actions that could have helped one escape poverty (Akerlof & Kranton, 2000; Jenkins, 2014; Lewis, 2017). The combination of high levels of social inequality and rapid demographic change along religious lines can have important ramifications. Global demographic religious change implies that more individuals will have a religion in the near future, and that these individuals will be increasingly likely to be situated in regions with high levels of economic deprivation (Stonawski et al., 2015).

2 Religion and Poverty

As Figure A.4.1 (online appendix) indicates, religion is powerfully intertwined along a number of causal pathways with the production of economic, environmental and social outcomes in societies. Poverty and riches, like other social outcomes, are not likely to be randomly distributed across religious groups. Modern organized monotheistic religions such as Christianity and Islam have at their core a very strong redistributive message and foster behaviours conducive to reciprocity and equality between its members. Indeed, the world’s four major religions—Hinduism, Buddhism, Christianity, and Islam—all include doctrines of salvation and damnation that provide sets of different incentives for performing economic activities and charitable deeds. Comparative analysis of the four religions has shown that each promotes accumulation of wealth and hard work, while discouraging idleness, debt and poverty. The primary difference across these religions is with respect to charity. Religions that allow believers to contribute to their own salvation tend to emphasize varieties of selective charity. Religions that hold salvation is only possible through divine selection stress universal charity (McCleary & Barro, 2006).

Religions and religious institutions and movements both have played major roles in supporting established social order, but also in advocating change, rebellion and revolutionary movements, from Ancient Rome to the English Civil War, the Islamic revolution in Iran and many more (Lincoln, 1981). Finding effective policies to deal with poverty can hinge on the type of religious convictions and beliefs of those who are disproportionally poor. For instance, religion may relate to the likelihood of attaining higher female education and prioritizing a career ahead of alternative life goals, such as marrying at a young age and having a large family. Some religious groups have a fatalistic view of life, where one’s economic position is predetermined at birth—including the Hindu caste system, which prescribes occupational opportunities based on caste. According to their religion, Hindu populations may respond differently to incentives advocating change through own actions than other religions.

Poverty has a subjective and an objective dimension. Individuals may compare themselves to others within a specific region, but they may also compare themselves to others with a similar group identity (Bhuiyan, 2018; Clark & Oswald, 1996). One typically compares one’s economic standing not with a global mean, but with a reference group closer to one’s own social context. For instance, some may consider their own relative income rank in relation to others with similar ethnic background, from the same linguistic group or the same type of profession. One important dimension that has received very little attention is religion and income; the aim of the current study is to fill that gap.

Religious groups belonging can act as a strong determinant of group belonging and may help define who one’s compatriots, competitors and adversaries are. Religious communities, in which many are economically deprived and which at the same time are growing demographically in absolute and relative terms, can represent a growing political and demographic challenge. It could potentially lead to more conflict, as has been the case in recent decades. More inequality across religious groups, where some religious groups are at different ends of the income distribution, can lead to greater likelihood of social upheaval, political instability, tensions and violent conflict (Bartusevičius, 2014; Nordås, 2014).

Economic growth and educational expansions have been found to correlate with religious and cultural groups, and inference models suggest that these relations are of causal nature (Becker & Woessmann, 2013; Bonsang et al., 2017; Cantoni, 2011; Iyer, 2016). Diversity may affect the willingness to support political institutions or to pay taxes (Alesina et al., 1999; Lee et al., 2015). Religions represent a new line of tensions in the global order and may be a key dimension of regional and global disputes and conflicts (Burstein, 2018; Huntington, 1993; Philpott, 2018). This has also lead to widespread beliefs and tensions, where majorities of Europeans seek bans on Muslim immigration (Goodwin et al., 2017), and the majority of adolescents in the Norwegian capital of Oslo believe that the West and Islam are in a state of war (Vestel & Bakken, 2016).

Despite declines in financial hardship in recent decades, high shares of the world population still live in poverty (Inklaar & Rao, 2017; Kraay & van der Weide, 2017). Great progress has been made in reducing poverty, but as rates decrease, reducing poverty any further seems to be getting harder (Crespo Cuaresma et al., 2018). If one believes that one’s fate (in terms of, for instance, becoming wealthy or maintaining good health) is an outcome of God’s will, then one may be less likely to rationally plan for or adapt one’s behaviour to improve one’s conditions and prospects. Religion could influence important human behaviours that have a direct impact on economic outcomes such as whether one chooses to pursue higher education, whether one seeks career-oriented positions or whether one wants to have a family and higher or lower fertility levels (Buber-Ennser & Skirbekk, 2016; Lehrer, 2004; McQuillan, 2004). Economic inequalities and time trends can differ by country and religion around the world. Although part of this variation could be ascribed to demographic and human capital differences across religions, religion may well have a bearing on key aspects of human behaviour that influence poverty risks as well as socio-economic outcomes among the poor. This could, for instance, include how religion may affect education, labour force participation and career choice by gender. Religion may also relate to health risk behaviours including diet, exercise patterns, alcohol use, vaccinations and choice of health treatment. Importantly, religious views may exert influence on financial behaviours such as savings and investment behaviours.

Economic inequality and religion have substantial links and a traditionally interwoven relationship. Religion has been found to relate to poverty-relevant outcomes, including economic aspirations and family size (Becker et al., 2016; Heaton, 2011). Religion also affects intermediaries of poverty, including the likelihood and timing of getting married and the age of having children as well as cohort fertilityFootnote 1 (Bakibinga et al., 2016). Those who are more religious tend to have a greater number of children—and to have them at earlier ages, which may conflict with educational attainment and lower capacity to attain high earnings. Intergenerational transmission of wealth could imply that poverty is more likely to be concentrated in religious groups as long as the more religious tend to have on average lower education and lower earnings. Religious influences may affect human capital formation and occupational choice (Abbasi-Shavazi & Torabi, 2012; Barnes & Brownell, 2016; Glaeser & Sacerdote, 2008). Fertility timing and the resulting sibship size can also be related to individual impoverishment and national economic growth (Berman, 2000; Canning & Schultz, 2012; Teller, 2011).

Moreover, of particular importance for the poor and in poorer countries that lack developed welfare systems is the provision of welfare goods through religious entities (Göçmen, 2013; Kagawa et al., 2012; Olivier et al., 2015). Frequently, there is a lack of old-age care systems, and universality and generosity of public systems may be inadequate. Also, in many nations, religious organizations manage key aspects of care, including old-age care, hospitals, childcare and free education. Access to such services may depend on membership and participation in activities in religious beliefs and organizations. If poverty is concentrated in certain religious groups, this may motivate others to carry out services for these groups.Footnote 2

3 Data and Methods

The poverty assessment data are taken from the World Bank´s PovCalNet initiative (World Bank, 2018), which tracks and measures poverty through a variety of measures of absolute and relative poverty of income. We use two measures of poverty to enable a global comparison from the World Bank’s PovCalNet database:

  1. 1.

    Absolute poverty. It is measured as the percentage of population living on less than 1.90 USD per capita 2011 (International USD at Purchase Power Parity as defined by the World Bank). Under this level of income per person and per day, it is deemed that on a global level the basic needs of a person cannot be fully met, irrespective of the country he or she is living in. This would be the lowest common denominator of poverty globally.

  2. 2.

    Relative Poverty. This is a measure of poverty that takes into account the distribution of income. It is based on national poverty lines that vary considerably across countries. It is measured as the share of the population under a certain percentage of disposable national income that, compared to the rest of the population of the country, counts as poor. While basic needs may be satisfied, the level of income does not enable an individual to live an inclusive and fulfilling life within its society. For the OECD, the standard is to take the share of the population earning less than 60% of median income. For countries where the measure is not available, we have constructed the measure by using that approach in our sample. The PovCalNet database contains income/consumption distributional data from about 1200 household surveys spanning 1979–2015 and 138 developing countries. More than 2 million randomly sampled households were interviewed in these surveys, making them representative for 96% of the population of developing countries. More information and a brief discussion on the comparability of poverty measures used in PovCalNet in the online Appendix, A.4.1

3.1 Religious Affiliation Data

PEW-ACC religious projections were the first ever global estimates of religion that considered global religious demographic detail. It incorporated age-sex variation and took into account more than 2500 demographic databases.Footnote 3 The 2015 PEW-ACC estimates of religious population have opened a wide avenue for research on religious development of the world population: ACC and PEW produced the first global data set on affiliation by age and sex, covering 199 nations and more than 99% of the global population. Based on a large database of more than 2,500 surveys, registers and censuses (Pew, 2015; Skirbekk et al., 2016; Stonawski et al., 2010, 2015), data on religious affiliation were taken from a separate unique project that estimated global religious affiliation by age and sex globally. Religious affiliation information from more than 2,500 data sources, including censuses, demographic surveys , general population surveys and other studies, was analysed—the largest project of its kind to date (Stonawski et al., 2015). These data were used to obtain country-specific estimates on religious affiliation by 5-year age groups separately for men and women. Census, survey , focus group and other demographic data collection methods were used to identify belief systems as they relate to the demographic make-up of society, age and sex distribution, and geographic factors. 83% of the world had a religion in 2010 according to our estimates, and this proportion is projected to increase to 87% by 2050, with changes taking place more rapidly among the young groups (Hackett et al., 2015; Skirbekk et al., 2016; Stonawski et al., 2015).

We study religious affiliation data based on self-identification of religious beliefs from surveys and censuses. We do not use registries from religious communities or proxy interviews, as these may be biased or inflated due to overcounting, skewed due to political or economic reasons and many may not fully register religious conversions.

4 Findings and Discussion

Extreme or absolute poverty is defined as living on less than USD 1.90 a day, measured in 2011 Purchasing Power Parity prices. According to the most recent estimates (World Bank, 2017), in 2013, 10.7% of the world’s population lived at or below USD 1.90 a day. That is down from 35% in 1990 and 44% in 1981. Many countries have experienced dramatic shifts in poverty in this time span, but the most dramatic has been the decline in absolute poverty experienced by countries in Asia, where the combined absolute poverty went from 81% in the 1970s to less than 13% in 2013.

In this timespan, the world population doubled from 3.8 billion to over 7 billion people. If we look at the distribution by religions in Fig. 1a, b, we can see the distribution of the world’s 1.4 billion poor in 1970, and how the numbers evolve until 2010. At 968 million people, the number of poor broke the billion people barrier around 2010. In terms of religion, there has been a marked increase in the number of Christians and a stagnation of Hindus. In addition, the shifts have been dramatic for the unaffiliated and folk religionists, driven by China’s growth that has accounted for lifting 200 million unaffiliated and 130 million folk religionists out of poverty.

Fig. 1
figure 1

(Source Own calculations and databases on global religion and poverty [Stonawski et al., 2015; World Bank, 2018])

a, b Absolute poverty by religion, 1970 and 2010

Full size image

From Fig. 2a, b, we can see the sharp drop in rates of poverty for all religious groups globally, which is particularly marked for the unaffiliated and folk religionists. The unaffiliated, spread in numbers equally between China and nine other countries with high levels of development (Germany, France, Japan, Korea, US, Canada, Italy, Brazil), show the lowest levels of poverty together with the Buddhists and Jews, who are concentrated in the US and Israel.

Fig. 2
figure 2

(Source Own calculations and databases on global religion and poverty [Stonawski et al., 2015; World Bank, 2018])

a, b Absolute poverty by religion in 1970 and 2010 (in %)

Full size image

In Figs. 3 and 4, we show the respective values for relative poverty distributions according to national poverty lines. The orders of magnitude and trends are comparable and similar to the absolute poverty numbers. This points to the fact that while there has been a marked growth of inequality in the distribution of income globally, at national levels for most countries income inequality has been driven by a relative stagnation of the middle sections of society, while the poor have grown at or faster than the global average.

Fig. 3
figure 3

(Source Own calculations and databases on global religion and poverty [Stonawski et al., 2015; World Bank, 2018])

Relative Poverty by religion in 2010

Full size image
Fig. 4
figure 4

(Source Own calculations and databases on global religion and poverty [Stonawski et al., 2015; World Bank, 2018])

Relative poverty by religion in 2010 (in %)

Full size image

We find that poverty is disproportionately concentrated in the Hindu populations (between 24 and 29.3%), mostly driven by poverty in India—followed by Christianity (12.9 and 19.7%) and Islam (14 and 22%). Folk religionists (12.4 and 13%) as well as other religions (14 and 18%) tend to rank mid-ways when it comes to poverty incidence. Finally, the unaffiliated (8.5 and 9.7%) and Buddhists (6.8 and 10%) have low levels of poverty, and the Jews (0.3 and 1.2%) have the lowest of all shares living in poverty. This holds true for both relative and absolute levels of poverty.

India and China are two countries that have elevated more citizens out of poverty in the past 30 years, and which will be crucial in bringing down poverty further if the world wants to meet the SDGs set for 2030 to eradicate absolute poverty. We therefore hone in on the religious characteristics of the least well-off citizens in these two countries in the context of political demography developments in the last few decades analysed in further context in the respective chapters by Noesselt as well as Balachandran/James (this volume). Table 1 highlights the respective numbers of people living in absolute poverty in India and China in 2010.

Table 1 India and China, population living in absolute poverty by religion, 2010
Full size table

In 2010, India had the largest number of people living in absolute poverty worldwide, with more than 300 million citizens. Since 2010, the trend towards poverty reduction on the back of overall strong economic growth has accelerated. According to the latest estimates (Crespo Cuaresma et al., 2018), the ratio of India’s population living in absolute poverty since 2010 has more than halved from 30% in 2000 over 21% in 2010 to less than 8% in 2018. By contrast, barely 1% of China’s population had been living in absolute poverty, which according to the latest estimates has also more than halved to less than 0.5% in 2018.

It is important to highlight that both countries had very similar levels of absolute poverty, close to 80% of their populations, just a generation ago (World Bank, 1990). With different but broadly similar development plans and insertion into the global economy, these two countries have achieved significant progress in poverty reduction with very different political approaches and institutional constellations, but that can be characterized as having had very similar outcomes in their development trajectories. One of the factors that may explain the differential development paths of these two large countries is the relationship between identity and inequality.

To explore the relationship between identity and redistribution of resources, Shayo (2009) proposes an analytical framework that finds that (1) national identification is more common among the poor than among the rich, (2) national identification tends to reduce support for redistribution, and (3) across democracies, there is a strong negative relationship between the prevalence of national identification and the level of redistribution. The longer persistence of poverty in India in relative terms to China could therefore perhaps be due to the democratic processes in India and the differential national and ideological identification of the respective citizens, and their relative tolerance of absolute poverty.

Religion in Indian politics has played an increasing role since the 2000s, with political formations playing national and religious cards. Hinduism’s role in politics is marked and pervasive since the ongoing electoral successes by the BJP in the mid-2000s. Of relevance to the redistribution debates is that by religious affiliation, India’s Muslims have a higher relative poverty headcount than the majority of Hindus. Whereas absolute poverty among Hindus affected almost 246 million people (about one in five of the overall Hindu population), Muslim poor numbered close to 45 million (more than one in three).

By comparison, China’s absolute poor population numbered 45 million unaffiliated and almost 16 million Buddhists—all living in rural regions far from the economic centres of activity. As highlighted in the chapter on China by Noesselt (this volume), social care, which has relied in China mostly on families, will come under increased stress because of demographic pressures. To establish an all-encompassing social insurance system, China would have to increase its tax-revenue, which at 9.6% of GDP is comparatively low for its level of development and is far below the world average of 14.6%.

The difficulty to further raise the tax base in these conditions is also a worry for the least well-off citizens, or those in danger of sliding back into poverty in old age. China’s successful fight against poverty may be at risk of regressing as older generations face uncertain retirement and social security conditions. Therefore, religious institutions may see increased demand for social charity and step into fill the demand.

By contrast, in India, differentials in income among religious groups (as highlighted above) and in fertility (as highlighted by James & Balachandran, this volume) have created intense political discussions around demography. According to Indian survey findings, total fertility rate (TFR) for Hindus is 2.1, while it is 2.6 for Muslims (Ranjan et al., 2017). The finding that differential fertility could result in differential population growth gradually shifting population composition has led to intense political debates and several instances of regional conflicts. Some Hindus fear that they will be ‘outbred’ by growing demographic weight of the Muslims—and that they will lose political and cultural influence. An important aspect of whether some groups will feel challenged by other demographic group is the economic level and growth among the different population groups and their relative age structure. As noted by James and Balachandran (this chapter), the Indian sixty-plus population is expected to triple from 92 million today to 316 million by mid-century. With ageing, there has been a growing demand for introducing social security schemes. Possibly, the introduction of higher coverage and more generous social security measures could lead to a greater feeling of economic security, reducing fertility differences between groups and lowering group level opposition.

5 Conclusion

Religion can offer several ways of coping with poverty, including offering meaning and hope for people in need, particularly those who are impoverished or lack resources. It could also strongly affect the level of financial wealth flows. The religious composition of a country or a region may also affect the level of social support and level of financial transfers, as welfare and social welfare systems can be organized through religious groups and indirectly affected by the degree of social welfare (Gruneau Brulin et al., 2018; Feldman & Scherz, 2017; Herbert, 2018). In poorer countries, religious organizations often offer basic health and poverty relief systems and tend to be more prominent in the provision of education, psychological and social services compared to in more developed countries, where these services are more likely to be provided by branches of the government (Chiswick, 2010; Young, 2009).

That some religious groups have a relatively large share living in poverty can have a number of implications. These religions may motivate different political behaviour if a large share of their compatriots is poor. For instance, this could motivate a stronger preference for transfers within the particular religions—but may also lower the ability to implement an effective, universal and sufficiently generous social security scheme as the economic burden would be too high on the rest of the community.

Moreover, it may also be that a greater religious share of the poor could lower income inequality (Elgin et al., 2013; Filipova & Bednarik, 2009). The very high levels of economic inequality and poverty within Hinduism may, for instance, be linked with a greater worship of Lakshmi, the goddess of luck, economic prosperity and wealth (Dwyer, 2013; Young et al., 2011). If one believes poverty is the outcome of an act of God, destiny or outcomes too difficult to decipher, then one may be opposed to policies that may effectively reduce this.

Several religions, including the largest ones, Christianity, Islam and Hinduism, encourage transfers. For instance, Muslims are encouraged to give a certain percentage of their income (2.5%) to the Zakaat as one of the five pillars of Islam. Similar encouragements exist both within Hinduism and within Christianity. Such systems may be easier to effectuate if the share of poor within any religion is smaller or the contributing part of the population is rich enough to be able to pay for transfer systems without suffering from too strong reductions in own living standards.

Our finding that large shares among the Hindu, Muslim and Christian communities are living in difficult economic conditions can influence the likelihood that conservative interpretations of religious traditions will be more common. Poorer religious adherents may benefit less from adjusting to more liberal views of, for instance, marriage, gender roles, family forms and fertility choices. When sufficiently high shares are conservative due to poverty, this may imply the entire religion is practised more conservatively.

In effect, a greater share of poor also implies that the behaviour of the religion as a whole may be different—they may be more likely to emphasize the aspects of religion that relate to how to live with dire conditions, in distress, how to cope with uncertainty and low levels of stability. These findings on poverty coincide with other factors, including low or lacking educational attainment, which is more prevalent among certain religions compared to others, and long-term determinants of wealth in the world areas where the particular religion is dominant, including schooling investments, capital investments and investments in family planning and stabilization or reductions of population growth.

Finally, the growing religious discordance between richer and poorer communities in many nations could lead to growing challenges in terms of willingness to pay, such as the case in many Western countries where poverty is concentrated in minority groups with different religions. One should then consider, when possible, ways of dealing with these and try to find policies that may at least in part alleviate some of these challenges associated with religious variation in poverty. Further studies are needed to try to better understand trends and variations in poverty across countries and how this may influence challenges related to poverty alleviation.

In terms of policy recommendations, for countries dominated by religions that exhibit high inequality, there may be a case for governments to collaborate with religious groups to reduce inequality by increasing transfers to its poorest members. There could be a high tolerance of poverty abetted by religious practices that could be detrimental to the affected groups and to society overall. Also, by understanding that poverty is not only randomly distributed across religious groups, we can better elucidate religions’ potential roles in affecting poverty risks and how different religions may offer different pathways to escape poverty. This includes welfare/transfer/social security systems, both of organized religion and of religion as a guidance system. We need to better understand why different religions have different economic outcomes, and under which circumstance the poor are more likely to belong to certain religions.

Conversely, religious actors in countries with high levels of economic inequality should be persuaded to reconsider their approach to their poorest members and to consider aligning poverty alleviating measures to be sensitive to the needs of particular religious groups. We also find that religion-based poverty support may be stronger in less secularized societies. There may be scope for more secular societies to interact with religious organizations that take care of poorer members. Or it could be that more religious organizations should concentrate on shifting its redistributive systems towards those of non-religious entities, such as has been the transition in several Western countries.

In increasingly less labour-intensive societies, religions may come to provide increasingly social activities, help to minimize loneliness and feelings of worthlessness, for instance through spiritual work that enhances the level of well-being in society. Providing more funding for religious organizations that have an explicit focus on not only material necessities but also providing spiritual well-being could represent one way of dealing with new measures of poverty within religions and act as a way of channelling workers towards activities with positive social influence. Clearly, more research and better data on why large sections of some religious groups end up in poverty are a matter of importance to public policy and political order in the twenty-first century.