The Assemblies of God: A Denomination That May Be Growing
It's rare, but I think I've found one!
What is out of fashion in American Protestant Christianity? Denominations. That’s the easy answer. It’s one of the common themes of this newsletter - almost every major denomination is losing members at an incredibly rapid pace. The Southern Baptist Convention had 16.2 million members in 2006, today that’s under 13 million. The United Methodist Church just navigated what I believe to be the largest Protestant schism in American history - shedding about 25% of their churches and members in a very short window of time. And, they aren’t the only mainline denomination that is shrinking. I’ve written about the Presbyterian Church USA, the Episcopal Church, and the Evangelical Lutheran Church of America. The theme is all the same - down, down, down.
I am regularly asked if there are denominations that are experiencing real and sustained growth. Yeah, there are. The Anglican Church of North America is up a bit - going from 125K in 2022 to 128K in 2023. The Presbyterian Church of America went from 378K to 382K during the same time period. Both denominations combined added less than 8,000 people. The SBC lost 241,000 people last year. I hope I am making my point clear.
But there is a major denomination that was on an upward trajectory for a very long time and stood as an aberration in the larger Protestant landscape: The Assemblies of God. They experienced very consistent growth over the last couple of decades. And, they very much get the gold star from me for producing some of the most detailed and helpful statistical reports of any denomination. You can find that here (PDF).
Like I just noted, the Assemblies of God went on a tremendous run of growth from the mid-1970s all the way through 2019. It’s pretty amazing for me to see a graph of religious adherents actually go up over time, but that’s exactly what happened to the AG for over four decades. They started with 1.25 million members but rose above 2 million by 1980. By the late 1990s, the figure was 2.5 million and it jumped above 3 million in 2010. Impressive stuff, really.
However, the pandemic hurt the Assemblies of God. While the denomination registered some declines in the previous several decades - they were never that alarming. Down just a few thousand year over year, but that wasn’t the case between 2019 and 2021. Their reported membership was 3.26M in 2019, it stuck at 3.26M in 2020, and then a really big decline in 2021 to 2.93M. A total of nearly 328,000 lost in just one year. That brought them to 2009 numbers.
But if you follow the trajectory of the line from the previous graph really closely you may pick up on something - the trend of growth had slowed considerably even before COVID shut down the world.
I calculated the year over year percentage change in total number of AG adherents all the way back to the 1970s. You can see that the denomination was growing at a 2-3 point clip during the 1970s and 1980s. Strong stuff there. But over time, that growth engine began to cool. By 2000, growth was vacillating between 0 and 2 points per year. Then, from 2010 forward, it’s apparent that growth had almost stopped.
Yes, COVID did do a number on the Assemblies of God’s membership. But I'm not sure things were headed in a positive direction even before lockdowns. Just think about this: the average annual growth for AG between 2011 and 2019 was less than 1%. I mean, any growth in this environment is certainly laudable for a religious organization, but that’s even keeping up with the increase in the American population.
What about attendance? The AG is clear in noting that this metric is for “In-Person” attendance, so that means we are comparing apples to apples when looking at the data over the entire time series, which goes all the way back to the late 1970s.
This graph looks a whole lot like the first graph that I showed you - consistent and steady growth for a very long period of time. There are a couple years where attendance does dip just a little bit, but then it quickly rebounds and then continues the upward march. The absolute numeric peak for attendance for the Assemblies of God was in the late 2010s. Between 2016 and 2018, the number of regular attenders in AG churches was above two million. But that wouldn’t last long.
By 2019, attendance was already dropping. It was down to 1,993,370 right before COVID-19. But then it just took a nosedive. It was 1.75M in 2020 and continued to fall in 2021 to 1.68M. It has reversed somewhat since then, though. In the most recent data, about 1.84M folks are attending Assemblies of God churches. That’s at the 2009 level.
Let me give you another look at this, though. I calculated the percentage of AG adherents who were regular attenders over the last several decades.
In the 1979 data, 72% of all AG adherents were also regular attenders. That’s insanely high! For reference about 25% of Catholics are regular attenders, it’s about a third of the mainline. Among evangelicals, it’s 55% now. So - there weren’t a whole of members in the Assemblies of God who were inactive in the late 1970s. But, that percentage did begin to decline very quickly from that high water mark.
By the mid-1990s, about a third of members were not regular attenders and it continued to inch downward in the intervening years. Right before COVID hit, about 63% of members were regular attenders, down nearly ten points from the high in 1979. But, you can see that the pandemic created a huge aberration in the data. In 2020, only 54% of members were regular attenders, but that rebounded quite quickly. It’s noteworthy that it’s back to 62%, which is about the same as it was five years prior.
Let’s do a little data experiment now. I am always interested in trying to figure out what impact the pandemic had on denominational attendance. I’ve done that for The Episcopal Church and the Evangelical Lutheran Church of America, so let me try that using the data from the Assemblies of God. Here’s what I did - using the attendance data from 2009 through 2019, I generated a projection of what attendance should have been in 2023 based on that decade worth of information.
As you can see, the projection model says AG would have enjoyed continued growth between 2020 and 2023. It estimated that in 2023 there would be an average weekly attendance of 2.11M. What was the actual number? 1.84 million. That means attendance is down about 13% from the projection model. This is largely in line with what I found with the ELCA - which was 20%. In The Episcopal Church it was 14%. I am pleased to see that the data is coming to the same conclusion - COVID hurt denominations by 15% or so when it comes to weekly attendance figures.
But let me get back to a bit more analysis of the sources of growth for the Assemblies of God over time. They separate out inflows into three categories: water baptisms, spirit baptisms, and then all other additions. This is what that looks like over time.
In the 1980s and early 1990s, the AG was adding between 250K and 300K through these three avenues. But it really began to spike by the late 1990s. For instance, in 1998, the Assemblies of God reported the addition of 560,000 new people on their membership rolls during this time period. What I notice in terms of composition of inflows is that baptisms used to make up a bigger portion than they currently do. For instance, there were periods in the 1980s, when they were 40% of the total. Now, it’s much closer to 30%. Spirit baptisms are down about ten points, too. It looks like a lot more people are moving their letter over from another denomination than getting baptized in the Assemblies of God.
Now, I have to admit my ignorance in this situation. Because the AG only reported a total membership of about 2.5 million during this time period, too. That means 20% of all members had converted that year. The only thing I can guess is that the AG is including adherents who hadn’t been baptized into their totals. Thus, getting baptized doesn’t really represent an addition, just a change of status. I might be way off on this, but that’s the only way that these numbers make sense to me. I mean, according to their records, the AG had 3 million converts in just the past decade. They don’t even have 3 million total members. I know that there are losses through transfer and death, but that still doesn’t explain the whole thing.
I wanted to end this by pointing to a reason that I believe that the AG has recorded long term growth while most other larger denominations have been going the other direction - the AG has continued to move in the direction of racial diversification.
In 2001, the Assemblies of God’s records indicate that 71% of their rank and file membership was white and another 16% were Hispanic. African Americans were just 6% of members and Asians were only 3%.
For reference, the Southern Baptist Convention is currently 71% white, 3% Hispanic, and 20% Black. However, the pews of the average AG church today look a whole lot different. Now, only 55% of those members are white, down 16 points in just 22 years. Meanwhile, the Black share has nearly doubled to 11% and the Hispanic portion has risen to 23%. That’s pretty impressive given the inability of many other denominations to become less white to reflect the changing demographics of the country.
When I say that no major denomination has reported membership growth, I have to include the Assemblies of God in that conversation now. They were doing just fine for decades and then things turned south in the last five years or so. Some of it was the pandemic, but one can’t place all the blame on COVID. However, there are some green shoots in this data. If current trends continue, it seems possible that the AG will move above its 2018 membership levels in the next decade or so. If it does then it will stand as the only larger Protestant denomination that is bigger in 2035 than it was in 2020.
Code for this post can be found here.
Ordained AG minister here. Loved the article. Thanks for writing it.
One thought about the stats that puzzle you: we wouldn't consider baptisms or spirit baptisms inflows. They're more things that happen to people who are in our orbit. In an ideal scenario, someone would begin attending one of our churches and become a Christian, they would then be baptized in water, they would also at some point be baptized in the Holy Spirit, and they would keep coming to church and eventually become a member. If that all happened in a single year, then that same person would appear in the data you're pulling from in multiple categories: they would count as a conversion, a baptism, a spirit baptism, an adherent, and also a member.
Also, the AG statistician's office is very easy to contact if you (or any researchers reading this) want to do a deeper dive on the data. You can reach them at numbers@ag.org
I think the weird growth statistics are because AG allows and counts rebaptisms.