Professor Elias Papaioannou Deliveres Open Lecture ‘Religion, Ethnicity and Social Mobility in Africa’ at NES


Professor Elias Papaioannou from London Business School delivered a closing lecture within the Honorary Lectures on Political Economy in Memory of Alberto Alesina organized by NES. His presentation ‘Religion, Ethnicity and Social Mobility in Africa’ was based on a joint paper with Alberto Alesina, Stelios Michalopoulos, and Stefan Hohmann.

Why Africa?

Religious fundamentalism appears globally on the rise: the world sees more and more terrorists, violence, conflicts attributed to the religion. Africa was chosen as a focus of the research because of the two important points. The first is that many social scientists have coined Christianity in Africa as the biggest social experiment. Only a century ago Africans were adhering to local indigenous religions, but Christian missions, colonial powers, and many others tried to christianize Africans in an effort towards modernizing them. The other one is that nowadays Islam is the fastest religious group in Africa and coexistence is very intense. It is projected that by 2050 there will be a sizable portion of Muslims Africans in the continent (670 million people) while the Christians will count 1 billion people.

How Does Religion Influence Educational Mobility?

The research looks at Muslims, Christians, and Africans adhering to local indigenous religions and tries to explain the drivers of religious intergenerational mobility (IM) gaps. According to the research the Africans born in the 40s attended primary education and secondary education just before imbuing independence, among them Christians have twice as much the use of schooling as Muslims, and after 40 years those differences have persisted. For example, in Nigeria, the likelihood that a Christian boy or a girl, whose parents have not completed primary school, has completed primary school is 80%. So, they do better than the previous generation. But it is less than one out of two Muslim boys and girls, whose parents have not completed primary education, have managed to complete primary education. The corresponding likelihood for African boys and girls, whose parents adhere to traditional religions, is a third—considerable gaps in upward mobility across religious affiliation. Likewise, for the downward mobility, which is the likelihood that kids would not manage to complete primary schooling although their parents had managed to do so, the gaps are more negative for Muslims and animists. That correlates with income negatively, so high religious gaps in education mobility in relatively poorer African countries and in countries with high levels of religious segregation suggest that the way religious communities reside in countries has a first-order impact on educational mobility.

Family, Location and Other Factors

Regional features tend to explain half of the considerable gap in upward mobility between Muslims and Christians and between animists. Nonetheless, this suggests that a Christian girl and a Muslim girl from the same region and from families with similar observable characteristics have different chances to get educated: the likelihood of the Christian girl to attend school is something like 7.5 percentage points higher. There are similar patterns with the downward mobility; there is a much higher chance that Muslim boys and girls and animist boys and girls would not manage to reach the educational level of their parents. It seems that family features matter somewhat more for girls, especially with the downward mobility. Household and family features overall explain something10% of the observed differences in upward or downward mobility between religious groups. The role of regional factors appears to be the largest in Nigeria, Benin, Ghana, and Senegal.

The Earlier, the Better

The final part of the research examines the regional childhood exposure effects and how much regions matter for mobility from special sorting which is the fact that parents or grandparents or household members with higher latent ability may decide to relocate to a good area.

The evidence shows that a Christian boy that moves at the age of 1 has a much higher likelihood to complete primary schooling than the Christian boy that moves at the age of 15, because this kid is unlikely to attend a school or benefit from the better region. The moves after the age of 15 also match with higher levels of upward mobility. The families that move tend to have better characteristics, but there is an almost linear effect with larger regional effects; exposure affects the earlier, the more. The Muslims who move from worse to better places tend to gain in a causal matter from the move to the better region. Similar patterns are animists; the reason is that animists are very heterogeneous recalled it is just 3% of the population, so those estimates are quite noisy. Nonetheless, for those Muslims who move or for those animistic that move, the kids benefit a lot in terms of upward mobility, in terms of likelihood of acquiring education, especially when the move takes place early in life.

Some Conclusions

The results of the research show that there are considerable gaps in upward and downward mobility between Christians and Muslims which is mostly explained by regional factors, although there are considerable intergenerational mobility gaps between Christians, Muslims, and animists, even if one looks within the same African region. The correlation of religious intergeneration mobility and religious IM gaps shows that such features as early development, colonial investments, favorable location, and geography matter equally for all religions. As a consequence, those features do not correlate at all with the religious IM gap. But Muslims and animists reside in areas of lower economic development and with older cohort children. What is further interesting is that Muslims are doing relatively worse not in regions where there are minorities, but in regions where they are either large minorities or majorities. This is not the case for animists, and it is slightly positive for the Christian population. The childhood exposure effects clearly show that if Christians, Muslims, and animists move from lower to higher mobility places, they all gain, and Muslims, especially Muslim girls, gain even more, which suggest that spatial shorting, which is considerable for all regions, it is especially the case for animists and a bit more for Muslims.

Video

Slides 

Thu, 26 November 2020
Elias Papaioannou
140 persons read this article, 3 liked it. Did you enjoy this post?