There is something called the GSS, the General Social Survey. It is a large sample survey of the US population regarding several interesting data-sets like religious identity. It starts roughly in the 1970’s and has continued through today. Here is wikipedia on the GSS.
Here is a post by much better social scientists than I, looking at protestant affiliation across time through the GSS. Their point is the dramatic decline of denominations. And this is true, but I also think it covers up something else. What I think it is covering up is actually a great sorting out. Inside each one of those grouping that they have put together is a constant core. What you are seeing is the collapse of a specific type of American Christianity and a sorting out along the way.
A note on the methodology. The list of denominations that the GSS records might be called “pre-schism”. Lutherans are pretty easy as we schism-ed early and often. What I’ve done is taken the ELCA bodies and the “Lutheran don’t know specifics” and put them in the same bucket. I’ve put the LCMS and the other specific bodies together. My assumption is that if you know enough to answer a specific denomination you have some tie to its theology. If you don’t there is an accomodationist grouping that doesn’t care about doctrine in any serious way. What I’m trying to construct is what I call a “proto-denomination”. The Episcopal Church, the PC-USA, the ELCA and the United Methodists are all in pulpit sharing arrangements with each other. What that means is that a minister who is ordained from one of those bodies is able to take a call or be a pastor of a congregation in any of those bodies. They have declared that any doctrinal divisions do not stand in the way of spiritually leading a congregation. So while church politics might keep them separate, and by church politics I mean that bishops in each body don’t want to risk losing a chair as long as the money holds out, those bodies are effectively one denomination or moving that way. I’ve kept all Baptists together. There are portions of presbyterians (PCA), Episcopal (Anglicans) and United Methodist that I’d have loved to split out, but this is the pre-schism problem. The GSS just doesn’t have those bodies. But when you look at the graph, I don’t think you need to see that to get the point. I’ve included the historically black denominations in the other line. And the last methodology note is that for the graph I’ve taken a 5 sample moving average to smooth the graph. What that means is that any one survey tends to jump around. For example, from 1986 survey to 1987 survey the Baptists went from 32% to 41% of the sample. That is the nature of a sample. Averaging over 5 years, each point having the preceding 5 survey points, helps to smooth such single survey effects.
This is the resulting graph:
Here is what I see – 3 flat lines and 2 converging lines. Baptists moving around 32% of protestants with little movement 80’s – today. Other/non-denom’s hovering around 23%. Lutherans hovering around 5%. The accomodationist proto-denomination hovering around 30% until roughly 2000 where it entered a steep decline. Around exactly the same inflection point “none” goes from 7% on a steep upward slope.
So, what does this mean? Well, other than nothing because I messed up the data which is a possibility, this is what I think. First, none of this says anything about how these groups do in comparison to the population. If the share of the population that gets lumped none-none (i.e. atheist/agnostic) is growing (and other surveys say it is) that is a separate point. This is within the religious world. Second, those groups that maintain differentiation are holding their own. Maybe not in reported membership. The LCMS has been declining in reported membership, but in survey response which is more akin to attendance than membership, it holds its own. Likewise with baptists and non-denoms. There might be an argument that just wait, the kink down is coming, but the fact that there are denominational groups that haven’t would seem to say not all denominations are the same. Third, the decline and growth is really a story of the accomodationist grouping. The great sort is happening or has happened. If you “believe, teach and confess” a specific doctrinal body you probably know it and find it significant enough to maintain. You might have troubles keeping some congregations going because the “don’t knows” have drifted away, but it should be possible given some political grace to stand. (What I mean by that is some congregational consolidation is probably necessary, and that might mean driving a little further for some, but the results should be larger and more vibrant/healthy congregations.) But the accomodationist grouping doesn’t have that core body of doctrine. They are not a theological group but a sociological grouping. And looking at the data, sociology is not enough to maintain them. There is a remnant in there – a portion of the UM, the PCA, the ACNA. My guess is that if we could split them out, the accomodationist group would look worse.
Spiritually, what I would say is what Christ says to the church of Philadelphia in Rev 3:10-11. “Because you have kept my word about patient endurance, I will keep you from the hour of trial that is coming on the whole world, to try those who dwell on the earth. 11 I am coming soon. Hold fast what you have, so that no one may seize your crown.”