Religion is For Older People, Right?

How age impacts religion at the county level

I was talking to someone who works in the nursing home industry a couple of weeks ago and she said a term that I had never heard, “the silver tsunami.” It’s the demographic reality facing the United States (and most other industrialized countries), that the population is getting a heck of a lot older and most of these countries are just not ready to handle the hundreds of thousands of people who will need some type of assisted living/nursing facility in the near future.

In the United States, that’s being driven by the Baby Boomers, the oldest of which were born in the late 1940s. That means that they are entering their seventies now and in 10-15 years almost all Boomers will be retired, with many entering their twilight years.

Well, I got tipped off that the Census Bureau had just released some interesting data about the average age of every county in the United States over the last three years (2020-2022). When you visualize that data, the crisis that is facing the United States comes into sharper focus.

I just looked at the share of each county that was at least sixty-five years old in 2020 and then that same share just two years later in 2022. Holy cow, we are an aging country. I have a total of 3,168 counties in my dataset. Of that, just 211 have a smaller share of 65+ folks in 2022 compared to 2020. Put another way, 93% of all counties in the United States have a larger share of retired people in 2022 than they did two years earlier.

Here are the states with the highest percentage of counties with a decreasing share of old people:

• Oklahoma - 27%

• Texas - 25%

• South Dakota - 20%

• Montana - 13%

• New Mexico - 12%

There’s only one county in Minnesota, Mississippi, and Alabama were the share of 65+ went down.

But having that data in hand means that I combine it with other data from the Religion Census to try and understand what religion looks like in older counties versus younger ones.

Here are the ten largest religious traditions in the oldest 300 counties in the United States. The youngest of these counties has a median age of 49 years old, the oldest had a median that was north of 60 years old. For comparison, the median in the overall sample was 41.5 years old.

Catholics dominate in old counties. Sixteen percent of all religious adherents in those 300 counties are members of a Catholic Church. Two evangelical groups follow: non-denominationals at six percent and Southern Baptists at five percent. The United Methodists are next at three percent. Then, there are a whole bunch at one percent or less.

Three of these groups are from the mainline tradition: the United Methodist Church, the Evangelical Lutheran Church of America, and the Episcopal Church. But notice that all of these are basically Christian traditions. (Let’s not get into a fight about whether Jehovah’s Witnesses or LDS are Christians, please).

How do things change when I restrict the sample to the 300 youngest counties in the United States? For the record, there are 46 counties where the median age is less than 30 years old. None in this sample have a median age that is beyond 36 years old.

Again, the Catholic Church is number one. But that should come as no surprise, really. There are sixty-million Roman Catholics in the United States. That’s three times the size of non-denominationals, who are the second largest. So, it’s just how the math works out. The top three are exactly the same, by the way, and the percentages don’t shift that much.

Look at the LDS share. It was only 1% in the oldest counties. It’s 5% in the youngest counties. There are also new groups that show up here: Muslims are much more prevalent in younger counties and so are those in the Assemblies of God. There are a lot of evangelicals in the youngest counties, without a doubt. Also, while there were three mainline churches in the oldest counties, it’s just one in the youngest: United Methodists.

Given that we have just seen how religion looks different in the youngest versus oldest counties, it seems appropriate to take a look at the oldest religious traditions in the Cooperative Election Study. One caveat - it’s only a sample of adults, which means that the youngest person who takes the survey is 18 years old. So, you aren’t getting the overall mean age, just the mean of the adults in the sample.

Of the ten oldest religious traditions in the sample, six of them are from the mainline: UMC, TEC, ELCA, DoC, UCC, and PCUSA. There are only one that doesn’t show up here: American Baptists. There are also Lutherans from the Missouri Synod here and National Baptist Convention, an historically Black denomination. Again, note that all ten are Christian groups here.

Compare that to the ten youngest religious traditions. Just one of them is a clearly Christian tradition. Instead, it’s a whole lot of nones - atheists, agnostics, and nothing in particulars. But also, smaller religious groups like Orthodox Jews, Hindus, and Muslims. You can see how these two sets of analysis related to each other: Christians are found in older counties and non-Christian groups (and nones) are doing well in places with a lower mean age.

I think I’ve been circling around this for a while now, so let’s get to it: are older counties more likely to be religious than younger counties? The Religion Census tells us what percentage of each county is attached to a religious tradition, so I can use that a measure of how religious each one is (knowing that it’s not a perfect measure). That’s what is on the y-axis in the graph below. The x-axis is the median age of each county from the Census bureau.

And look at that linear trend line! It’s headed downwards, which means that the older a county is, the less religious. That’s not what the prevailing wisdom would say about this topic, I would guess. How big is the shift? Among counties where the median age is thirty years old, about 55% are attached to a religious tradition. Among those counties where the median age is 50 years old, about 45% of folks are attached to a religious tradition.

So, not a huge shift, but it’s nothing. Obviously one thing that makes this difficult is how concentrated this data around the median, which is 41.5. The outliers can kind of throw off this type of analysis. But even when I trimmed up the data a bit and kicked out the really old and really young counties, the trend line is still in a negative direction. Older counties are just less religious.

Let’s finish this with a nice little table. I wanted to look at the twenty counties that shifted the most toward religion between 2010 and 2020. (I only looked at counties with at least 25,000 folks, by the way. Don’t want any weirdness from some random place in Alaska with 225 people.) The thing I was really looking for was the median age of those counties. Recall that the median overall was 41.5.

In the five counties that saw the biggest move toward religion, the median age is really low! Note, that four of them are in Arizona, Texas, and California, by the way. Immigration is a big part of the story, obviously. But among these twenty counties, twelve of them have a median age that is lower than the national average.

This is just another little bit of evidence that older counties don’t really mean more religious counties. There’s a ton of dynamic religious growth in very youthful pockets of the United States. Now, this growth may look like a lot of Hispanic Catholics, or Muslims, or Latter-day Saints. But that’s how the American religious landscape is shifting.

There’s always this tension when I write about religious growth or decline. I think most people still think of religion in terms of Christianity (typically white Christianity). There are still a ton of white Christians in America, and they will still be the plurality of the country for decades to come. But the United States is becoming more racially and religiously diverse as each year passes.

Sometimes I will give a talk and say religion is growing in the United States. Lots of nods of approval from the audience. Then I will describe just what I laid out here: growth of Hispanic Catholics, Muslims, etc. and their faces turn a little bit. They want religion to grow in the United States, but only their kind of faith.

Code for this post can be found here.