The Bible Says ‘Don’t Work, Don’t Eat.’ But Does That Hold Up in the Data?

A closer look at how religion, gender, and employment really intersect.

I am running a holiday special - 20% off an annual subscription to this newsletter. That brings the cost down to just $48 per year. Take advantage by clicking this button:

---

So, there’s this verse you sometimes hear in an evangelical worship service. It’s from 2 Thessalonians 3:10: “For even when we were with you, we gave you this rule: The one who is unwilling to work shall not eat.” From my experience, it’s not one you hear all that often, but it definitely comes up when discussions about government welfare arise. It’s usually cited in defense of free-market capitalism and a smaller government.

That’s not to say there aren’t plenty of biblical passages that discourage laziness. For instance, Proverbs 21:25 says, “The craving of a sluggard will be the death of him, because his hands refuse to work.” So, there’s a clear message in both the Old and New Testaments that people should work if they’re able.

Well, if you’ve been around here for a while, you can probably guess where this is going: is it true that religious people are more likely to have jobs than non-religious folks? The variable in question comes from the 2024 Cooperative Election Study: “How many jobs do you have?” There are four possible response options — zero, one, two, or three or more.

I’ve never really played around with data like that before, so I figured I’d start by giving you a sense of what it looks like before diving into the religious part of the equation. For example, here’s the share of respondents who say they have zero jobs, broken down by their year of birth.

It should come as little surprise that almost everyone born before 1940 has no job — those folks are in their eighties, for goodness’ sake. But among people born around 1950 (which puts them in their mid-seventies), about 10% still have a job. Also, notice the point estimates for people born in 1959 vs. 1960? Yeah, that’s when folks start turning 65. For people born in 1959, about 72% have no job, while only 60% of those born a year or two later are not employed.

You can clearly see the years of peak employment, too. A majority of people born between 1965 and 1998 have at least one job. The most likely to be employed are those born in the 1980s — at least two-thirds of them are working. You can also see how younger adults are less likely to have a job, with less than 50% of respondents born after 2000 currently employed.

Now, let’s zoom in on those key employment years by focusing on respondents born in 1960 or later. This time, I’ve broken the sample into two groups: those who identify with any religious tradition and those who describe themselves as atheist, agnostic, or having no religion in particular.

That’s a pretty clear conclusion, right? When you compare religious folks to non-religious folks of the same age, the religious respondents are more likely to report being employed. Now, to be clear, the differences aren’t huge. In most years, the gap between the two lines is just two to four percentage points. So, it’s not a chasm or anything — but it’s remarkably consistent across the entire range of data.

That’s what strikes me, honestly. From 1970 onward, the gap is almost perfectly stable — the lines run in parallel. There’s no evidence that the difference ever narrows, let alone reverses. From this first bit of analysis, I think it’s fair to say, at least tentatively, that there is an employment gap based on religion.

But let’s dig a little deeper. I’m now going to focus on people born between 1970 and 1990 — the “peak employment” years — and show the full breakdown of responses to the jobs question across a range of different religious traditions.

The group most likely to be unemployed are folks who describe their religion as “something else.” I’ve written about these people before, and it’s fair to say they’re generally an odd group. Many of them seem to struggle to answer even basic questions about religion. The next group in line is “nothing in particular.” If you’ve read The Nones, you know this group consistently sits at the bottom of the socioeconomic spectrum, so it tracks that they have relatively low employment rates.

Here’s a real surprise, though — those who are “nothing in particular” are just as likely to have no job as Muslims and Orthodox Christians. That’s pretty wild. But what’s really odd about Orthodox Christians is that they’re also more likely to have multiple jobs than almost any other religious group (aside from Hindus). That could just be a small-sample-size issue, but it’s still worth keeping an eye on.

Who’s most likely to be employed? Atheists, agnostics, and Jews. In each case, less than 30% report having no job. Jews are also the most likely to have exactly one job — that describes 69% of them. It’s also worth noting that the big Christian groups, Protestants and Catholics, look pretty “normal” on these measures. I wouldn’t say they’re more or less likely to be employed than other groups.

So here’s an angle I really wanted to explore: is there a gender gap among religious traditions that tend to take more “traditional” views of gender roles compared to groups like atheists and agnostics? I used the same slice of data — people between the ages of 35 and 55 — and looked at the share of each gender who reported having zero jobs.

Here’s what really blew me away — there’s a gender gap in employment across nearly every subgroup I looked at in the sample. In some cases, the differences aren’t statistically significant, like for Orthodox Christians, Buddhists, or Hindus. But there isn’t a single instance in this data where men are clearly more likely to be unemployed than women. The smallest gap is among atheists — 31% of women have no job compared to 25% of men. Among agnostics, it’s an 11-point difference.

In many traditions, though, that gap gets really big. For instance, 43% of white evangelical women don’t have a job compared to 23% of men. It’s a 17-point difference for mainline Protestants and a 20-point gap for white Catholics. Among Latter-day Saints, the divide is even larger — 25 percentage points. The only group with a bigger gender divide are Muslims: just 26% of men aren’t working compared to a whopping 63% of women.

So, I think two things can be true at once here: men are more likely to work than women — that’s nearly universal. But it’s also empirically accurate to say that, in some religious traditions, there’s clear evidence of a significant gender gap.

I did start to wonder, though, whether those differences among Christian groups are being driven mostly by the most devout. For instance, is a female Catholic who attends Mass weekly more likely to be unemployed compared to a female Catholic who goes less than once a year?

This was a key finding for me — and something I’m going to be thinking about a lot: the types of Christians most likely to be unemployed are the ones who don’t go to church at all. You see that pattern for both men and women. Among men who never attend, about 30% have no job. For men who go to church multiple times per week, that number drops to around 17%. For women, a majority of those who never attend also don’t have a job, compared to about 35% of women who attend a Christian church weekly.

This is something that I write all the time - dropping out begets dropping out. In this instance it’s dropping out of church attendance and dropping out of the workforce. Of course, the causal arrow could go either way.

But that gender gap still persists. When you compare men and women at the same attendance level, women are roughly twice as likely to have no job as men. And I don’t think we can say there’s a major difference between Protestants and Catholics on this either — in almost every case, the differences between the two groups aren’t statistically significant.

Let me close this post with a more rigorous test. I really wanted to get to the bottom of whether religious people are more likely to be employed than non-religious folks. So, I ran a regression where I held education constant at its mean — that way, we can’t chalk up any differences in employment to different levels of educational attainment.

What I found was that non-religious folks are slightly more likely to report having no job compared to religious respondents. Among men in the sample, about 22% of religious respondents reported being unemployed, compared to 24% of non-religious respondents. For women, the percentages were 40% vs. 42%. Once again, the gender gap in employment is huge, but the religion gap is basically non-existent.

In this simple analysis, that two-point difference isn’t statistically significant. And even if it were, it certainly wouldn’t be substantively significant. In other words, whether someone has a job or not has very little to do with whether they identify as a Christian, Muslim, Jew, or atheist. It’s driven by a whole range of other factors that extend far beyond religiosity.

Code for this post can be found here.

---

Ryan P. Burge is a professor of practice at the Danforth Center on Religion and Politics at Washington University.