A Few Facts on Religion in America

I have been reading papers that I may use in a Fall class, and one is a survey of the economics of religion. As best I can tell, this field largely consists of sociologists applying rational choice modeling to questions of religion. As subject matter it is very interesting but the modeling is not terribly well-developed or convincing.

In any case, I though I would share the facts of religion, as culled from this paper. Note that this is all from a 1996 paper by Laurence Iannaccone. I should almost put quote marks around it, but it isn’t verbatim so I won’t.

1. American Church membership has risen throughout our history, from 17% in the beginning to 60% today.

2. We have, in the U.S., about 1.2 clergy per thousand people. This number has been about the same for 150 years.

3. Since polling was available in the 30’s, a stable 40% of the population calls itself churchgoing (in a typical week). The only noticeable pattern is a drop among Catholics after the papal announcements in the 60’s.

4. 95% of Americans believe in the existence of God. This number hasn’t moved since it was first surveyed in 1945. Also stable is the number believing in an afterlife (71%), heaven (72%), or hell (60%).

5. Total contributions to Churches are stable at about 1% of GNP. This accounts for half of all charitable giving. Most volunteer work is also religious.

6. Religious belief and activity do not decline with income (though giving rises with income). Most religious indicators rise with education. But income and education do predict the type of religion, with more liberal, less-demanding religions drawing more from the wealthy and educated.

7. College professors are a little bit less religious on average than the population. This is most pronounced in the humanities and social sciences. The bastions of personal antagonism to religion in the academy are the same fields that have pushed the claim that society is secularizing and that there is a tension between science and religion, namely psychology, anthropology, and sociology. Those in the physical sciences, who actually deal with, you know, science, are comparatively much more likely to attend church and/or profess faith.

8. The fastest growing religions require adherence to strict rules. There is some nice, but simple, economic theory about why this occurs.

18 comments for “A Few Facts on Religion in America

  1. Katherine
    August 5, 2004 at 5:06 pm

    Frank, I am interested in your last point–would you elaborate the economic theory that explains why “the fastest growing religions require adherence to strict rules”?

  2. August 5, 2004 at 5:31 pm

    That information is really quite stunning, Frank, given how much it differs from the generally accepted view generally seen in the media: that religion took a big downturn in the sixties; that despite noise from the religious right, religion is in long-term decline; that religious participation falls as income or education rises, and so on.

    I would have to think, however, that the definition of “God” used in the surveys must be pretty broad if 19 out of 20 Americans affirm it.

  3. August 5, 2004 at 6:01 pm

    What is the “nice, but simple, economic theory” of what religions with strict rules doing better?

  4. Frank McIntyre
    August 5, 2004 at 6:18 pm


    I have drafted a couple replies to your query, but was not satisfied with them because I don’t know how much economics you’ve had. So here is the not very elegant version:

    Suppose people join churches to get a certain feeling or to worship, or whatever. Call this end result “R”. Producing R is synergistic, in that doing it with others more than proportionately increases the R produced. In economics this is called “increasing returns to scale”. Further, assume that different people have different ability in producing R, and that high ability producers in turn make other people better producers. The flip side is that low ability producers dampen everyone’s ability to produce R. Think of these groups as the very spiritual, who raise those around them, and the clown who’s cracking jokes during Mass and ruining it for everyone by driving away the Spirit. It doesn’t have to be “ability”. It can be motivation, desire, effort, or whatever; the point is that some people have it and some don’t. We’ll call it ability.

    Given this production form, the key to success for any person is to get together with as many high ability people as one can, while avoiding the low ability people, so that one can enjoy as much R as possible given one’s effort level.

    Of course, this is the goal for the low ability types too. So one has to find a way to keep them out. The way to do it is to impose costs the low-ability can’t or won’t pay but the high ability can or will. So require adherence to strict rules that only the devout are willing to obey. This keeps out those who would drag everybody else down. By keeping those people out, the church is attractive because it produces a lot of R. This causes others to wish to join. The strict rules keep out the low ability, and so the church grows but with consistently high levels of devotion among the members.

    The low ability producers can then go form their own “social club in drag disguise” as a lax rules church.

    One can expand this model in many ways, but the key feature is some sort of synergy across members.

    And 10 points to whoever can name the author and source of “social clubs in drag disguise”.

  5. Frank McIntyre
    August 5, 2004 at 6:22 pm


    I agree. Of course, there are denominational correlations to income and education, which may be where some of the media claims come from.

    The belief survey is conducted by Gallup and they ask: “Do you believe in the existence of God or a universal spirit?” So it is as broad as one could possibly get!

  6. August 5, 2004 at 6:38 pm

    Interesting Frank. How do you respond to those who say many rules are actually for those who don’t like to think too much as it provides structure in the lives of those who lack initiative? (You see that claim even in LDS circles as the justification for why the law of the gospel is so much harder than the law of Moses)

    The other problem is that the production of “R” doesn’t seem intrinsically related to the rules religions have.

  7. Scott
    August 5, 2004 at 6:41 pm

    Dylan, right?

  8. Frank McIntyre
    August 5, 2004 at 6:59 pm

    Dylan. “It’s alright Ma, I’m Only Bleedin”. Here’s the context:

    “For them that must obey authority
    That they do not respect in any degree
    Who despise their jobs, their destinies
    Speak jealously of them that are free
    Cultivate their flowers to be
    Nothing more than something they invest in.

    While some on principals baptize
    To strict part platform ties
    Social clubs in drag disguise
    Outsiders they can freely criticize
    Tell nothing except who to idolize
    And then say God bless him.

    While one who sings with his tongue on fire
    Gargles in the rat race choir
    Bent out of shape by society’s pliers
    Cares not to come up any higher
    But rather get you down in the hole
    That he’s in.”



    So the argument these people are advancing is that people flock to rules because those people
    like rules. One can explain any behavior by saying that “people like it”. In economics, that approach is not looked upon favorably as it isn’t very interesting or clever. That doesn’t mean it isn’t true. I suppose the idea would be to see if synergies do seem to exist in churches. But that would only be the first step.

    As for the relation between the rules and R, the theory dictates that one find rules such that the ease of obeying them is as correlated with R productivity as possible, given social norms and such. Ideally one would just observe R productivity and use that. But if that is difficult to codify and monitor, other rules can work. Lots of meetings, social stigma of membership, dietary restrictions, dress and grooming standards, etc. all can serve to classify those who prefer the worldly things and those who are more interested in producing R.

  9. Frank McIntyre
    August 5, 2004 at 7:01 pm

    Oh, I almost forgot; Scott gets 10 points.

  10. Katherine
    August 5, 2004 at 9:10 pm

    Frank, thanks for the explanation–you must have made it simple because I understood it. Very interesting indeed.

  11. August 5, 2004 at 11:29 pm

    “So the argument these people are advancing is that people flock to rules because those people
    like rules. One can explain any behavior by saying that “people like it”. In economics, that approach is not looked upon favorably as it isn’t very interesting or clever.”

    One would hope it would be looked upon on the basis of accurate predictions. Aren’t economists supposed to be big Instrumentalists in their theory? (grin)

    The fact is that some people are simply not inclined to be go-getters. I think we all recognize this. So I think the theory (which I don’t think has been tested – but perhaps we ought ask the sociologists amongst us? Brayden?)

  12. Sheldon
    August 6, 2004 at 11:17 am

    “1. American Church membership has risen throughout our history, from 17% in the beginning to 60% today.”

    So what about all this whining from the religious right that America is slouching toward Gomorrah, becoming dangerously secular, etc… So the golden age of religion that the founding fathers supposedly belonged to didn’t exist after all? I think we often forget that the reason there were great religious revivals is Joseph’s days was because America was NOT a religious country back then, and preachers were worried about that fact. Things like Ten Commandments monuments in courthouses are a recent invention.

  13. John David Payne
    August 6, 2004 at 11:18 am

    Frank, I heard a few years ago that the reason for very low church affiliation in the post-Revolution days was that most Americans were Anglicans before the Revolution. Membership in a church headed by King George III seemed somehow less appropriate afterwards.

    The sudden, widespread disaffiliation worried many Americans, who were by and large a religious bunch. Thus the Revolution is followed the second great awakening, setting the stage for Joseph Smith and others.


    But I have just been looking around and haven’t been able to confirm this. Has your study shed any light on the low rates of church membership in the early days of the republic?

  14. Frank McIntyre
    August 6, 2004 at 11:20 am


    “One would hope it would be looked upon on the basis of accurate predictions.”

    2 points:

    1. That is why I said, “That doesn’t mean it isn’t true” right after the part you quoted. Even if something isn’t clever it still may be right.

    2. Assuming people do things because they like to is essentially tautalogical, thus it does not yield any testable predictions because it is not falsifiable. There have to be some restrictions on the way they like it or how that preference changes over time in order to have something to test. Thus one explanation for any behavior is always “because they want to.” That’s fine, but the interesting thing is when you can find reasons why people do things based on a more concrete and well-defined set of preferences. The theory I outlined above does that, so it is sort of neat. But like I said, the modeling in this sub-field is not very well-developed, so the theories may be clever but not much more testable than just assuming people do it because they like to.

    Further, if one wishes to take stated preferences as people’s real preferences (always dicey), the evidence for people liking arbitrary rules is not great. I doubt it is strong enough to explain why strict religions do better. So in that case, a theory like the one above can give some reason why people join these sorts of religions. Ultimately, though, the increasing returns/barriers to entry theory is very hard to test unless R is measurable. That is another reason why this sub-field is not going to explode into a major discipline anytime soon.

  15. Frank McIntyre
    August 6, 2004 at 11:34 am


    Church membership rose to about 34% in the mid-1800’s. That is as much as the survey mentions. Note that although Church affiliation has risen, clergy per capita has been stable since at least 1850…


    Low church affiliation should not be confused with atheism. The non-church goers may well have been like Joseph Smith Sr., with a faith in Christianity but no formal Church affiliation. Church affiliation is not the same thing as belief in the Bible or God or the Ten Commandments; so the revivals may have been attempts to stimulate Church affiliation among unaffiliated Christians.

    Of course, without polling data on beliefs in the colonial era, it is hard to say for sure.

  16. Last_lemming
    August 6, 2004 at 3:28 pm

    Among the Puritans, everybody attended services, but actually becoming a member of the church was a big step that most did not take. Kind of like if we only counted temple recommend holders as real members. I don’t know about the other religions, though.

  17. August 6, 2004 at 6:56 pm

    “Assuming people do things because they like to is essentially tautalogical, thus it does not yield any testable predictions because it is not falsifiable.”

    Except that people do things they don’t like and don’t do things they do like. The problem with “what they like” is that “like” is too vague and is often generalized so much as to render it meaningless. i.e. those who say we are all hedonists because whatever we do is by definition what we like at that time.

    On your other point, yeah, I saw you made a similar point. However I often like to distinguish between “truth” and “apparent truth.” Some things may be difficult to determine empirically but still be true. (Yeah, yeah, philosophical nitpicking. But some people don’t make those distinctions)

    I agree on the problem of “stated preferences” and I always cringe when sociologists do that. I tend to think that most people are the worst judges of their real preferences. Further I tend to think people have lots of reasons not to be honest in their answers when they do answer.

  18. Frank McIntyre
    August 12, 2004 at 1:09 pm

    Actually, Clark, in economics “like” is well defined. It is that which one does. I like A at least as much as B if I do A instead of B. There are some niceties around this that I won’t go into, but in most empirical work, we use observed choices as the basis of analysis. Unless one is really into welfare analysis, the actual underlying preferences can be left to the psychologists.

Comments are closed.