I’ve written many times about the increasing evidence that religiosity is negatively correlated with the well being of a society and its inhabitants. That is, those countries (and U.S. states) that have higher indices of well being are those that are the least religious. Of course, this is a correlation and doesn’t prove causation, but there are some data that the improvement of social welfare is causal in eroding religion. In the U.S., for instance, a fall in the GINI Index, which means more income equality, is followed in subsequent years by an decrease in religiosity, suggesting that people become less religious when they feel they’re on a level playing field of life.
A new paper by Miron Zuckerman, Chen Li, and Ed Diener in Personality and Social Psychology Bulletin (free access and pdf with the legal Unpaywall application; full reference below) adds more evidence to not only the negative relationship between government care of its people and their religiosity, but also suggests that the former causes changes in the latter. The former relationship holds not just for countries of the world (155 were assessed) but also the 50 United States, and data from the U.S. suggests that increased government welfare actually reduces religiosity.
I’ll try to be brief. The authors compiled data from 155 countries and the 50 U.S. states on
a. Religiosity (the proportion of people who answered “yes” to the question “Is religion an important part of your life?)
b. Quality of life. For the countries this was a composite index that incorporated educational attainment, infant mortality, percentage of population in urban areas, life expectancy, proportion of physicians among all inhabitants, and percentage of the population below the poverty level. Similar data, but using eight factors, was the index of quality of life in the U.S.
c. Index of government services. This used two statistics: health expenditures and education expenditures as percentages of the gross domestic product (GDP)
d. The Gini Index. Data from the World Bank: a value of zero means that all incomes are equal, while 1 means that one family has all the income and everyone else has none. The value thus ranges from zero to one, with higher values meaning more income inequality.
e. Subjective well being (SWB). These data were taken from a Gallup World Poll, and includes the components of life satisfaction, positive emotions, and negative emotions.
The authors tested several hypotheses in both countries of the world and U.S. states.
1.) The government can replace God as a service provider. That is, the hypothesis is that, as the authors say, “the government provides an extra layer of security and can replace God as an agency of last resort.” As Marx noted, religion may well be the last resort of people who have no hope from other people or their society. This would predict that, holding other variables equal, including SWB and Quality of Life, higher government services would be associated with lower religiosity.
2.) They tested the further hypothesis that religiosity would be lower ONLY with a combination of government services and higher quality of life. Do these factors need to be present simultaneously, or do they act in an additive way, having pretty independent effects?
3.) Finally, does higher SWB also correlate with reduced religiosity?
Along the way, they were able to use the U.S. longitudinal data to study temporal relationships between these variables, suggesting that it is social services that erode religion rather than the other way around. ]
In all cases, the relationships were controlled for other variables using hierarchical regression analyses, so that each factor’s effects could be studied independently.
The main results (read the paper for more).
For countries
a. Higher Gini scores (more inequality) were, across countries, associated with higher religiosity. This itself shows no causality; in fact, you could posit that more religious countries somehow increase income inequality, though other data (below) suggests that if there is a causality, it is between inequality and belief.
b. Better government services were associated with lower religiosity. Higher quality of life was also associated with lower religiosity (remember, each variable is controlled by holding the others constant).
c. There is an interaction between government services and quality of life: if you have higher levels of both, your country is even less religious.
d. Subjective Well Being was negatively associated with religiosity when government services were low, but not when they were high, implying that high government services are by themselves enough to reduce religiosity and efface any relationship between SWB and religiosity.
For U.S. states
These results are for the U.S. by itself, looking at the 50 states, and they also did a time series of measurements over six years. Quality of life included eight variables, which you can see in the paper.
a. Holding other factors constant, higher quality of life was associated with lower religiosity in a state, as was higher average government services. The interaction between higher quality of life and government services was of borderline significance (p = 0.069), but the lowest religiosity was still found in states having both higher quality of life and better government services. This mirrors the results above for different countries.
b. Interestingly, the level of government services in a given year predicted the level of religiosity (a negative relationship) one and two years in the future, implying that if there is a causality, it’s an erosion of religion by government services rather than an erosion of government services by religion. This temporal relationship did not hold for quality of life, nor did it hold in the reverse direction except for one-year data (but not two-year data) with a p value of 0.02 (religiosity predicting government services in the future)
The upshot. In general, the authors’ hypotheses, which are also those advanced by Norris and Inglehart in their excellent book Sacred and Secular: Religion and Politics Worldwide, were confirmed: there is a negative correlation between government care of its people, as well as social well being, and religiosity. States and countries that take better care of their people are less religious, and the temporal analysis suggests that an improvement in government welfare erodes religiosity, as predicted by many sociologists—as well as by Karl Marx. Further, as we already know, there’s a negative correlation between quality of life (and happiness, too!) and religiosity. The most religious countries are the least well off and have the unhappiest people.
So much for the claim that religion makes things better, and for the other claim that religion is essential for people’s well being. In fact, religion is a substitute for material well being, and goes away when people’s well being improves. The happiest countries in the world, those with the most well being (material, medical, and so on), as well as the best government services, are the least religious countries—countries like Denmark, Norway, Sweden, Switzerland, and so on. This is a message that we can’t shout too loudly. Religion doesn’t improve people’s well being. (I suppose the goddies will argue, “yes, but they’re spiritually better off.” But if that’s the case, why aren’t they also happier?)
Finally, as the world improves and people increase their well being (see Pinker’s talk that I’ll post later today), religiosity will decrease. Of that I am sure. You don’t need god in a world where people feel secure and taken care of by their society. Eventually, as the rising tide of well being lifts all boats, religion will become a vestigial belief.
________
Zuckerman, M., C. Li, and E. Diener. 2018. Religion as an Exchange System: The Interchangeability of God and Government in a Provider Role. Personality and Social Psychology Bulletin: First published April 12, 2018; https://doi.org/10.1177/0146167218764656