A new study published in the American Journal of Sociology argues that conservative religious belief is a contributor to the substantially lower median income among conservative Protestants have compared to the rest of the American population. I will discuss the key findings of the study below, but I want to use the report from Religion News Service as a conversation partner. The report has now appeared on both Christianity Today and Beliefnet (Update: Add Pew to this list). The report misrepresents not only Keister's observations in the study, but the intent and possible outcomes of the study as well. An issue that this paper raises is absolutely vital for churches of all socioeconomic brackets to hear: Your theology may be justifying and contributing to socioeconomic inequality rather than helping to solve it.
Religion News Service reports:
Keister says a typical "conservative Protestant" might be a member of the Assemblies of God, Churches of Christ, Nazarene and Pentecostal churches.
Keister's new article in the American Journal of Sociology, "Conservative Protestants and Wealth: How Religion Perpetuates Asset Poverty," argues that traditional views of money — it's God's, not ours — keep many Protestants from building a financial safety net.
While some struggle to build up their bank accounts, others, like Anne Thompson of Louisville, Ky., choose to give it all away in order to pursue what they see as God's plan for their lives.
The report goes further to say:
But theology may be the biggest factor, especially conservative views on the inerrancy of the Bible. Steve Diggs, a stewardship minister at the Antioch Church of Christ in Nashville, Tenn., said many Christians have misunderstood what the Bible says about accumulating money.
The Religion News Service report indicates that theology "may be the biggest factor" in the relationship between wealth accumulation. It sounds as if more conservative theology is a cause of more conservative financial distribution among believers that is biblically based. Not only is this not a correct reading of the paper, neither does the notion that theology is the biggest factor accurately report the results. Moreover, the use of the word "theology" by Religion News Service may be misleading as well. (Of side interest, Christianity Today's headline is, "Study: Conservative Theology Means Smaller Bank Accounts" and the title for Both the RNS and Beliefnet is, "Study: Theology impacts Protestant bank accounts". So Christianity Today actually adds their own spin on the article perhaps for their subscriber base?) The abstract of the actual publication which is found here reads rather differently:
The results demonstrate that religion affects wealth indirectly through educational attainment, fertility, and female labor force participation. The results also provide evidence of a direct effect of religion on wealth. Low rates of asset accumulation and unique economic values combine to reduce CP wealth beyond the effects of demographics. The findings improve understanding of the relationship between religious beliefs and inequality.
The primary effects of religious belief are found in the notion that all property is God's which is reinforced by inerrancy in the religious groups that Keister studied. "Some of the larger and more widely-recognized CP denominations are Assembly of God, Baptist, Churches of Christ, Church of God in Christ, Nazarene, and Pentecostal." At least 25% of the U.S. population belongs to some conservative Christian denomination.
Keister proposes that "education, age at first birth, number of children, and female labor force participation mediate the relationship between religion and wealth" contribute to lower income status in conservative protestant households. These connections are rather obvious as have been observed before as Keister argues. But they are necessary to include in order to control for her novel thesis which is the connection of religious belief and relative wealth.
When controlling for class, the results are interesting. "Because class is an amorphous concept, it is difficult to conclude with certainty that religion is independent of class. However, these results suggest that religion is an important and unique correlate of wealth ownership." Thus, one's religious belief does contribute to the effect of wealth even if this effect cannot be argued as one that is independent of one's class background. In other words, if one tends to be of a certain socioeconomic status, this will be correlated with one's religious belief, but we cannot say that the religious belief is an independent cause of that socioeconomic status.
The continued exposure and inculcation of conservative religious values over time plays out as well. In terms of adults, "those who either (1) were CP in both childhood and adulthood, (2) left CP churches after childhood (apostates), or (3) became CPs as adults" all show lower wealth numbers. This is a suggestion that the religion and social structures of the particular churches in question have long-lasting effects over time and affect wealth behaviors. But again, this result is a correlation between religion and wealth and not evidence for a causal relationship.
A further interesting conclusion Keister draws is that conservatives appear to "seek divine guidance in making financial and work decisions and that they avoid accumulation of wealth". The wording here is important. Conservative protestants are more likely to avoid accumulation of wealth based on religious beliefs. This does not, however, mean that religious belief causes someone to be more spendthrift especially if they are already wealthy. Again, this is not to be confused with a causal effect. That religious belief alone can contribute to asset accumulation behaviors is the primary issue here. The author does not, however offer conclusive evidence that it causes new sets of wealth behaviors apart from other factors. That is to say, that if you are wealthy, you will not suddenly become poor by virtue of your participation in a given tradition that specifies wealth habits consistent with their understanding of the Bible.
Finally, the results "identify the process by which one large group remains asset-poor, but the findings and the approach are useful for understanding other distinctive groups and other elements of cultural orientation as well." Therefore, religion can be a reinforcing factor to persist in a given socioeconomic status bracket.
While this alone is an important finding as Keister notes, it is not exactly how Religion News Service couched the results of the paper. In fact, Religion News Service uses an example that the study does not indicate as conclusive or even argued at all!
Thompson, 43, gave up a six-figure salary in 2002 and quit her job to follow God's call to impact popular culture with a Christian message.
She said good-bye to plans for an early retirement and an $800,000 house and has not received a paycheck since.
To be sure, her above-average means put her in a higher-than-average income bracket, but her motivations aren't much different from many other conservative Protestants.
Thompson believes using her money to answer God's call is more important than using it for herself — even if her current financial situation is a "mere shadow" of what it once was.
"Words can hardly describe the life that comes in return for whatever it is that you're sacrificing," said Thompson, who describes herself as a conservative Protestant. "In my case it happened to be financial."
That wealthy people become poor due to their religious beliefs is clearly not argued in the paper and it is clearly not evident. Religion News Service, therefore, poorly misrepresents the data and misrepresents the very relationship between religious belief and wealth that the paper actually argues. The argument of the paper is that religious teaching has a significant effect on how ideas of wealth and actual wealth are correlated and then maintained. The also missed the very purpose of the study which seems to contain the most important Christian message of all. While this is not Religion News Service's place, one would think Christianity Today would have picked up on it while noting enough to alter the title of the report to market it towards a conservative audience. This purpose is an important one to which Christians need to hear loud and clear and it is worth quoting at length:
Finally, and perhaps most important, these findings have significant implications for understanding inequality in the United States and inequality in comparative perspective. The starting point for this study was the observation that wealth inequality is extreme in the United States and the supposition that religion may have some influence on levels of inequality. While I did not study inequality directly, my findings do suggest that if CPs—a large segment of the American population—began to accumulate wealth more like other groups, we might find that wealth inequality would be reduced. Future research might explore this issue more directly using simulation models to understand how changing saving behavior for CPs affects the wealth distribution. In addition, the United States has distinctively high levels of inequality and high levels of religiosity. It is possible that inequality is at least partially explained by religiosity among both CPs and other religious groups that are chronically asset poor. Future research might also explore this question directly by comparing the United States first to other developed countries and then to developing countries in order to identify the degree to which inequality varies with religion.
Religion News Service published an report that clearly misrepresents the findings and the purpose of the study. The study is a very good one and I commend everyone to read it. I will read it again to post other thoughts at another time for what this means to churches and how we behave as organizations devoted to the Gospel. Finally, the message of this study needs to be heard clearly and here it is again: Your theology may be justifying and contributing to socioeconomic inequality rather than helping to solve it!
No related posts.
Very nice analysis of press release botch-jobs on the actual findings. Composing scalars for theological convictions can be notoriously difficult, and to the extent that the study relied on self-report, I wonder whether respondents were sufficiently sophisticated (or primed) to distinguish non-theological motives for not saving, like ambient feelings of altruism that otherwise would prompt self-giving (non-saving) behaviors no matter which theology tickled the system. Since this study is preliminary and ungrounded in a corpus of collateral studies, I’d like to see whether prosperity-doctrine teachings repeated at Pavlovian increments (sorry for the sinister sarcasm) result in larger investment or savings patterns. And whether emotive or putative altruistic rewards by God to either side (poor or prosperous) really vary in substance, beyond simple ecological heuristics like, “well, gee, God said.” How stable are these traits across spectrums of ecological pressure?
Jim
Very nice analysis of press release botch-jobs on the actual findings. Composing scalars for theological convictions can be notoriously difficult, and to the extent that the study relied on self-report, I wonder whether respondents were sufficiently sophisticated (or primed) to distinguish non-theological motives for not saving, like ambient feelings of altruism that otherwise would prompt self-giving (non-saving) behaviors no matter which theology tickled the system. Since this study is preliminary and ungrounded in a corpus of collateral studies, I’d like to see whether prosperity-doctrine teachings repeated at Pavlovian increments (sorry for the sinister sarcasm) result in larger investment or savings patterns. And whether emotive or putative altruistic rewards by God to either side (poor or prosperous) really vary in substance, beyond simple ecological heuristics like, “well, gee, God said.” How stable are these traits across spectrums of ecological pressure?
Jim
Thanks, Drew. This is so interesting. I look forward to reading your further thoughts on it.
I write about the inequities a little bit in my book. I concluded that it's partly from our mission/evangelism efforts.
There's certainly an attitude in the mainline church that I didn't sense in my conservative background–we're expected to have our financial portfolio in good order before we enter the doors. And our evangelism efforts are often focused on people who look like us.
In mainline churches, we don't proselytize when we help people. We don't pressure them to convert so they can eat.
But, in conservative churches, they see feeding the body as a means to feeding the soul. In my opinion, this is deplorable. But, in effect, CPs invite the poor to be in their midst. They welcome the opportunity worship alongside the people they feed.
Thanks, Drew. This is so interesting. I look forward to reading your further thoughts on it.
I write about the inequities a little bit in my book. I concluded that it's partly from our mission/evangelism efforts.
There's certainly an attitude in the mainline church that I didn't sense in my conservative background–we're expected to have our financial portfolio in good order before we enter the doors. And our evangelism efforts are often focused on people who look like us.
In mainline churches, we don't proselytize when we help people. We don't pressure them to convert so they can eat.
But, in conservative churches, they see feeding the body as a means to feeding the soul. In my opinion, this is deplorable. But, in effect, CPs invite the poor to be in their midst. They welcome the opportunity worship alongside the people they feed.
I will need to take some time and read the study. It seems a bit peculiar at the moment, but I have no doubt that there are interesting insights. I have always understood that the number one factor in poverty is family instability. The US as a whole also has one of the worst savings rates in the world, but this has usually been attributed to a taxation system which punishes savings and enthusiasm for debt. Perhaps I will need to adjust my stereotypes …
I will need to take some time and read the study. It seems a bit peculiar at the moment, but I have no doubt that there are interesting insights. I have always understood that the number one factor in poverty is family instability. The US as a whole also has one of the worst savings rates in the world, but this has usually been attributed to a taxation system which punishes savings and enthusiasm for debt. Perhaps I will need to adjust my stereotypes …
I echo the appreciation of the post — very interesting. My first reaction to reading this is bent toward the causation point that you bring up. Perhaps it is reversed and socio-economic status causes more conservative religious belief? This might be indicated by the fact that people who were born CP's and people who are apostates both show similar patterns…
I echo the appreciation of the post — very interesting. My first reaction to reading this is bent toward the causation point that you bring up. Perhaps it is reversed and socio-economic status causes more conservative religious belief? This might be indicated by the fact that people who were born CP's and people who are apostates both show similar patterns…
I need to re-read it again, but the author did perform a regression to see if religion is a predictor of wealth. It can predict a given socioeconomic status, but any regressive analysis can indicate different sources of causation. What is clear is that religion does not cause wealth, but maintains a given wealth status.
As you indicate Paul, this cause can work the other way around. I think that this is one area that the study points for more research. But it is true that socioeconomic status or class does have an effect. The paper indicates that this was controlled for, but I am not clear yet if that was tested in a model that looked at the regression from both directions. I have to read the modeling again (always the driest part of a good study right?)!
And as Jim indicates, now that there is evidence of a clear relationship and a clear type of relationship between religious belief and wealth status, there is now great opportunity to run more tests on both these data sets and to reproduce the study controlling for different variances. I think the author points us in some very fruitful directions in the discussion at the end of the paper.
I need to re-read it again, but the author did perform a regression to see if religion is a predictor of wealth. It can predict a given socioeconomic status, but any regressive analysis can indicate different sources of causation. What is clear is that religion does not cause wealth, but maintains a given wealth status.
As you indicate Paul, this cause can work the other way around. I think that this is one area that the study points for more research. But it is true that socioeconomic status or class does have an effect. The paper indicates that this was controlled for, but I am not clear yet if that was tested in a model that looked at the regression from both directions. I have to read the modeling again (always the driest part of a good study right?)!
And as Jim indicates, now that there is evidence of a clear relationship and a clear type of relationship between religious belief and wealth status, there is now great opportunity to run more tests on both these data sets and to reproduce the study controlling for different variances. I think the author points us in some very fruitful directions in the discussion at the end of the paper.
Responses to the body of the study seriatim.
The title – “Conservative Protestants and Wealth: How Religion Perpetuates Asset Poverty”
The title is profoundly misleading and inappropriately sensational without understanding the author's methodological caveats qualifying her “findings” as hardly more than framing heuristics for further, more closely correlated studies. I’d certainly temper my judgment of sensationalism to make it more benign if titles like these generated more grant funding from the National Institutes on Health (National Institute on Aging and National Institute of Child Health and Human Development) to promote sociological studies in the US toward a closer match to European social medicine oriented models of “empirical” theology, where theology can be correlated to health or quality of life questions. The sensational component of the title to general readers untrained in sociological metrics is the suggestion that religion “perpetuates” asset “poverty.” For, why this title is misleadingly sensational except possibly as terms of art to sociologists, see passim below.
To dispatch the title summarily;
1. “Perpetuation”
While one underlying data set (National Longitudinal Survey of Youth) is sufficiently longitudinal to identify a “perpetuation” of poverty, the second data set (Economic Values Survey) seems at best to support secondary product-moment predictions of relative values (Wuthnow survey of religious scholars), using survey-feedback interviews by Gallup directed to the adult labor force identified by the 1990 census data. The correlation between these two data sets is heuristic and not scientific in finding shared religious “values” between the two sets. This approach is fine for framing further studies. But, the key problem with the notion of “perpetuation” (with religion “perpetuating” poverty) is that since the correlations between the data sets is heuristic, the author admits: “these data do not allow me to distinguish is whether CPs (conservative Protestants) have unique values because they are religiously conservative or because they accept the principles that CP churches promote.”
Thud.
What’s the difference between “religiously conservative” versus “the principles that CP churches promote”?
The author first draws a near tautology between the two (religious conservatives, and, principles in CP churches) via her heuristic association of the two data sets, but then, she re-introduces this distinction between them. This is either extremely poor science writing, nearly disingenuous, or a way to signal the National Institutes on Health of the need for further grants to study this question.
The real question is whether adherents affiliate as CP’s because the totality of their class factors makes them feel at home with the “the principles that CP churches promote”?
The author concedes that “class” is an amorphous concept and that religious conservatism is derivative of class. There’s nothing wrong with framing “religious conservatism” or ““the principles that CP churches promote” by articulating scales to identify these values, and then to do longitudinal studies to see if religious values really do “perpetuate,” that is, propagate vectors of effects that cause poverty.
But, that’s not done here.
One economical approach would be to examine existing data sets on behaviors like denomination-switching, or data sets identifying relational values in different religious families (e.g., “Readiness for Ministry” sociological data set on differential relational values that various church families expect of clergy) – to learn whether other “class” factors outweigh religious convictions, and whether religious convictions are anything more than trivial (vanishingly small) considerations in affiliating in CP families in the first place.
Lastly, on the “perpetuates” (religion perpetuates poverty) thematic, the author says: “While I did not study inequality directly, my findings do suggest that if CPs—a large segment of the American population—began to accumulate wealth more like other groups, we might find that wealth inequality would be reduced.”
Nonsense.
If CP's accumulated wealth like other groups, then there’s no reason why they wouldn’t switch out of CP's affiliations, and all become Episcopalians (okay, this is low-brow humor; but, get the point). Or go to work for the agnostic Popperian (Karl Popper) George Soros, doing international currency trading (okay, this is high-brow humor: it teases the border between religious faith and agnosticism).
2. “Poverty”
The notion of “wealth inequality” is loaded term and tied to the sociological definition of “poverty,” which I hope to address in a later post. Suffice it to say that “poverty” here is not what many popular readers might think, nor is it tied to federal clearinghouse standards of poverty (like those used to define indigent status for poverty law and social services).
Back to work for now (I don’t really have time for this: it’s just so close to home)
Jim
Responses to the body of the study seriatim.
The title – “Conservative Protestants and Wealth: How Religion Perpetuates Asset Poverty”
The title is profoundly misleading and inappropriately sensational without understanding the author's methodological caveats qualifying her “findings” as hardly more than framing heuristics for further, more closely correlated studies. I’d certainly temper my judgment of sensationalism to make it more benign if titles like these generated more grant funding from the National Institutes on Health (National Institute on Aging and National Institute of Child Health and Human Development) to promote sociological studies in the US toward a closer match to European social medicine oriented models of “empirical” theology, where theology can be correlated to health or quality of life questions. The sensational component of the title to general readers untrained in sociological metrics is the suggestion that religion “perpetuates” asset “poverty.” For, why this title is misleadingly sensational except possibly as terms of art to sociologists, see passim below.
To dispatch the title summarily;
1. “Perpetuation”
While one underlying data set (National Longitudinal Survey of Youth) is sufficiently longitudinal to identify a “perpetuation” of poverty, the second data set (Economic Values Survey) seems at best to support secondary product-moment predictions of relative values (Wuthnow survey of religious scholars), using survey-feedback interviews by Gallup directed to the adult labor force identified by the 1990 census data. The correlation between these two data sets is heuristic and not scientific in finding shared religious “values” between the two sets. This approach is fine for framing further studies. But, the key problem with the notion of “perpetuation” (with religion “perpetuating” poverty) is that since the correlations between the data sets is heuristic, the author admits: “these data do not allow me to distinguish is whether CPs (conservative Protestants) have unique values because they are religiously conservative or because they accept the principles that CP churches promote.”
Thud.
What’s the difference between “religiously conservative” versus “the principles that CP churches promote”?
The author first draws a near tautology between the two (religious conservatives, and, principles in CP churches) via her heuristic association of the two data sets, but then, she re-introduces this distinction between them. This is either extremely poor science writing, nearly disingenuous, or a way to signal the National Institutes on Health of the need for further grants to study this question.
The real question is whether adherents affiliate as CP’s because the totality of their class factors makes them feel at home with the “the principles that CP churches promote”?
The author concedes that “class” is an amorphous concept and that religious conservatism is derivative of class. There’s nothing wrong with framing “religious conservatism” or ““the principles that CP churches promote” by articulating scales to identify these values, and then to do longitudinal studies to see if religious values really do “perpetuate,” that is, propagate vectors of effects that cause poverty.
But, that’s not done here.
One economical approach would be to examine existing data sets on behaviors like denomination-switching, or data sets identifying relational values in different religious families (e.g., “Readiness for Ministry” sociological data set on differential relational values that various church families expect of clergy) – to learn whether other “class” factors outweigh religious convictions, and whether religious convictions are anything more than trivial (vanishingly small) considerations in affiliating in CP families in the first place.
Lastly, on the “perpetuates” (religion perpetuates poverty) thematic, the author says: “While I did not study inequality directly, my findings do suggest that if CPs—a large segment of the American population—began to accumulate wealth more like other groups, we might find that wealth inequality would be reduced.”
Nonsense.
If CP's accumulated wealth like other groups, then there’s no reason why they wouldn’t switch out of CP's affiliations, and all become Episcopalians (okay, this is low-brow humor; but, get the point). Or go to work for the agnostic Popperian (Karl Popper) George Soros, doing international currency trading (okay, this is high-brow humor: it teases the border between religious faith and agnosticism).
2. “Poverty”
The notion of “wealth inequality” is loaded term and tied to the sociological definition of “poverty,” which I hope to address in a later post. Suffice it to say that “poverty” here is not what many popular readers might think, nor is it tied to federal clearinghouse standards of poverty (like those used to define indigent status for poverty law and social services).
Back to work for now (I don’t really have time for this: it’s just so close to home)
Jim
[...] couple of days ago I wrote a piece on a recent article published in the the American Journal of Sociology which argues that [...]
[...] post follows quite circuitously starting here, followed by a direct reference to the post [...]
[...] but it then goes for two more posts as my idea got a bit bigger. The first one is called "The Effect of Religion on Wealth Inequality" followed by my development of this into an idea of the church as a sacrament here and [...]