Is the Protestant Work Ethic Real?

Dean Karlan is an academic economist and also—

Dean KARLAN: I am president and founder of Innovations for Poverty Action.

Which is what?

KARLAN: Which is a nonprofit organization which helps do research to figure out what works and what doesn’t, to fight poverty and social problems around the world.

Karlan, who’s at Northwestern, was teaching at Yale back in 2011, when he co-authored a book called More Than Good Intentions.

KARLAN: And one of the reasons I wrote this book was to help philanthropists make better decisions.

Karlan was headed to Hong Kong to give a talk about the book when something really great happened.

KARLAN: I got an email from a guy, from a secretary to a guy, who was the chief financial officer for Morgan Stanley in Asia Pacific.

That sounded like just the kind of philanthropist Karlan was hoping to meet.

David SUTHERLAND: And he was hoping he was going to get a big grant from Morgan Stanley.

And that’s the guy — an American named David Sutherland, who’d been working in Asia since the 1990s. He was in Washington, D.C., before that, including a stint as a tax lawyer with the Treasury Department. So now —

SUTHERLAND: I got a little flyer from the Foreign Correspondents Club that said this guy named Dean Karlan was coming to Hong Kong. So I said, “I want to meet with this guy.”

KARLAN: And I was like, “Perfect, this is why I wrote the book. Sure, happy to have breakfast.”

SUTHERLAND: I met him for breakfast, and 10 minutes into the conversation, he realized he was just talking to somebody who wanted to talk about his own charity.

So no, Sutherland was not going to write Karlan a big check. But was it possible he could offer something even more valuable?

KARLAN: And he got out these reams of folders with tons and tons of spreadsheets and tables. And he said, basically, “My passion is this group called I.C.M. in the Philippines, of which I’m chairman of the board, and here’s all the things we’re doing.”

I.C.M. stands for International Care Ministries; it’s a Christian organization serving what it calls the ultra poor.

SUTHERLAND: The people that we work with, they are living lives in desperate poverty. They don’t have enough food to eat. Thirty-two percent of all the mothers have had one of their own children die, and eight percent have had more than one of their children die.

Sutherland and his wife have three kids of their own. His family first heard about I.C.M. back in 1998.

SUTHERLAND: And it just really moved our hearts. So we started investing a little bit of money, and then just over time, it grew and grew and grew. And we got more and more involved.

I.C.M. today has an annual budget of around $10 million — and Sutherland, by the way, no longer works at Morgan Stanley. I.C.M. in the past decade has served roughly 200,000 Filipinos. Its core program is called Transform. Its curriculum has three components: health and nutrition training; teaching “livelihood,” or economic skills; and religious instruction.

SUTHERLAND: My view is that for sure, there’s a lot of fantastic secular charities in this world. But for me, I wanted to be involved in something that reaches the whole person. And for me, that is both physical poverty and spiritual poverty.

I.C.M. works with a network of some 10,000 pastors.

SUTHERLAND: And these pastors are Protestant pastors. We teach a basic evangelical message that everyone has sinned, they need to put their faith in Christ, and that that is the key to freedom for them.

The Philippines, we should note, is overwhelmingly Catholic; it’s only around 6 percent Protestant.

SUTHERLAND: I often say the Protestants are like anarchy. Some guy wakes up in the morning and he decides he wants to be a pastor. He just wakes up and he puts a sign in his front door and says, “We are now the Church of Jesus Christ of Heaven and Earth.” And there, he’s a pastor. Actually about a third of the pastors we work with are women. Their entire ecosystem is just the hundred families that they can see out front of their door. So they know which husband cheats on his wife, they know who sings good at karaoke, they know which little girl broke her leg. Having that connectivity with the local people that we’re trying to help is a core tenet that allows I.C.M. to reach those very poor people.

So Sutherland had great faith in the reach of his pastor network.

SUTHERLAND: We can find the poorest people, we can reach them very, very efficiently.

So he knew that I.C.M. was good at bringing poor people into the program.

SUTHERLAND: So the only thing that we need to know for sure is whether the program that we run is in fact effective.

Aha! Enter Dean Karlan. That was exactly the kind of research he tried to do with Innovations for Poverty Action. And he was well-acquainted with the religious-based charity model: roughly 60 percent of American non-profits that work in international charity are religious.

KARLAN: And when I talked with some of these groups I would often hear claims made that they’re not doing the preaching just because they’re trying to spread their beliefs, but they actually believe that it’s changing economic outcomes.

So Karlan was as interested in answering Sutherland’s question as Sutherland was in having it answered.

KARLAN: So we do see a lot of correlations out there. People, where religion is a bigger part of their life, tend to drink less, do fewer drugs, live longer, report higher levels of happiness, do less crime, things of this nature. So lots of correlations.

Correlations yes, but, as we’ve been preaching on this show for years, correlation does not equal causality. People like to use umbrellas when it’s raining. Do umbrellas cause rain? Science says no. What about religion? The economics literature has long noted that religiosity often coincides with attributes like diligence, thriftiness and trust. But again, does religion cause any of this?

KARLAN: That’s the basic problem with why you can’t go out into a cross-section of the world and just compare religious people to non-religious people and say, “Oh la la, look at these differences. Religion must have caused that to happen.

Karlan had long dreamed of running a randomized controlled trial, or R.C.T., to tease out the effect of religion itself. Where you’d have a control group and one or more treatment groups. Most religious charities, however, were not interested. David Sutherland could understand why:

SUTHERLAND: Too many of the faith-based organizations weren’t willing to set up a control group because they didn’t want to set aside a group of people they weren’t going to preach to.

But Sutherland himself didn’t think like that.

SUTHERLAND: If an R.C.T. showed that our current program was not effective, I want to know that as soon as possible so that we can pivot to a program that is even more effective.

And so it was that during that breakfast in Hong Kong, Dean Karlan realized that David Sutherland would be a willing partner in his long-standing desire to measure the economic effects of religious belief.

KARLAN: “Would you ever consider removing the religion from what you’re doing so that we can understand how that influences the outcomes of your program?”

SUTHERLAND: So, we ticked through all the boxes and in 10 minutes I said, “Yeah, all those things work for us, we’re happy to do that.”

After a few years of fund-raising and logistical planning, they began running this R.C.T. in the Philippines.

*      *      *

Dean Karlan is hardly the only American economist interested in how religion shapes our economic lives. There’s the Association of Christian Economists; the Christian Finance Faculty Association; the Catholic Research Economists Discussion Organization, or CREDO.

James CHOI: We know that religion is one of the most important and powerful forces in society, and we just want to know, what effect does that have on all sorts of things, including economic outcomes?

That’s James Choi, a finance professor at Yale. He works with Karlan on Innovations for Poverty Action. The problem with doing economics-of-religion research is the lack of convincing evidence.

CHOI: There’s been a lot of suggestive evidence. What you really want to do for the purpose of social science, to identify an effect, is to randomly assign certain people to be a certain religion and other people to be a different religion, and then just follow them forward and see, how do their economic outcomes differ? And that just doesn’t happen in the real world.

Choi did once try to get at this question in a lab experiment, using research subjects with different religious backgrounds.

CHOI: So, they came into the lab and they did a series of sentence scrambles where they had to rearrange words in order to form English sentences that made sense. And half the subjects got sentences that had religious content in them, the other half didn’t. And so the theory was that those who got the religious sentences would have their religious identity made more salient to them, and the religious effects would be stronger for them.

Then Choi and his colleagues had the subjects play a game that measured how much they’d contribute to a public good.

CHOI: So, this is a problem that society has to solve: How do we get people to contribute to public goods?

Public goods being things like the environment or democracy or safe streets.

CHOI: And what we did find was that when Protestants had their religious identity made salient to them, they contributed more to a laboratory public good. Whereas Catholics contributed less to the laboratory public good, and they expected others in the group to contribute less, so it seemed like they were less trusting.

Interesting, maybe, but Choi acknowledges the limits of this kind of research.

CHOI: Yeah, laboratory tasks, no matter how cleverly designed they are, no matter how compelling they are, at the end of the day, that’s not real life. So the ultimate test really is, how do these interventions affect people’s real lives outside the laboratory?

The fact that economics doesn’t really have an answer about the effects of religion is pretty remarkable when you consider that roughly 8 out of 10 people on the planet identify themselves as religious. The three largest denominations are Christianity, with around 31 percent; Islam, 23 percent; and Hinduism, 15 percent. Christianity is most concentrated in the Americas, Europe and sub-Saharan Africa; the Middle East and North Africa, meanwhile, are 93 percent Muslim, while the Asia-Pacific region is the most variegated: just 7 percent Christian, with 24 percent Muslim, 25 percent Hindu, 12 percent Buddhist, 9 percent belonging to “folk religions,” and 21 percent unaffiliated. Jews, in case you’re wondering, make up 0.2 percent of the global population. And what about James Choi?

CHOI: Yes, I do consider myself religious. I identify as an evangelical Protestant.

Academia, as you likely know, is not exactly a hotbed of religious sentiment. Even so, Choi has never been shy about his own beliefs.

CHOI: If you think that what you believe matters for where you’re going to go for the rest of eternity, then why wouldn’t you want to share that with other people?

On his Yale faculty website, Choi has a page that reads, “Why I am a Christian (Even Though I Don’t Believe in Santa Claus, the Easter Bunny, or the Tooth Fairy).”

KARLAN: And actually, that post is part of the reason why we ended up working together.

Dean Karlan again. He was raised Jewish and now considers himself agnostic. Back when he was on the Yale faculty with James Choi, he’d seen that Easter Bunny page.

KARLAN: One night at dinner, I brought it up.

CHOI: What we were talking about was how might we identify the effect of religion.

KARLAN: So we started brainstorming about how to actually test this.

CHOI: And we had the crazy idea that we would find some missionary organization that went out and evangelized to rural villages that were all isolated from each other. And we would try to convince this missionary organization to randomize which villages they went to and which villages they didn’t go to. Of course, this is crazy talk, and this idea went nowhere.

It went nowhere until a few years later, when Karlan had that breakfast in Hong Kong with David Sutherland.

CHOI: And then Dean excitedly contacted me. He said, “I found just the organization that can do this, and this is International Care Ministries. And they are a very data-driven organization, and they care about whether their programs are actually helping people.” And so they were willing to randomly assign which villages ended up getting different versions of their programs.

Okay, here’s how I.C.M.’s Transform program typically works. The participants — most of them women and self-identified as Catholic, and all of them excruciatingly poor — they’d meet once a week for roughly four months. The meetings were usually held in a church building — but, again, even the I.C.M. pastors have very little money; the building might have a dirt floor, a scrap roof and no electricity.

SUTHERLAND: So, every meeting is 90 minutes long: 30 minutes of values, 30 minutes of health, 30 minutes of livelihood: V.H.L.

David Sutherland again:

SUTHERLAND: So, we train the pastor to teach the values curriculum, the spiritual part of the curriculum, and they get 30 minutes each week. We teach them about the love of God, our respect for biblical principles, how Christ can help you to solve your day-to-day problems.

And then there are the two non-religious components of the program: health instruction—

SUTHERLAND: We will have our own person, an I.C.M. employee, that we train as a health trainer.

KARLAN: They’re taught ideas about hygiene, nutrition, provided nutritional supplements for children.

And the “livelihood” portion.

SUTHERLAND: And we have our own livelihood person trained by I.C.M.

KARLAN: The term livelihood in this context refers to economic activity.

SUTHERLAND: We show them how to do organic gardening, which is very low-cash. And then a few weeks later, you can have vegetables to feed your kids.

KARLAN: There’s lessons about small-business enterprises, how to get that started. A small grant often goes with that. They also form a savings group, so there’s a financial-inclusion aspect to the program.

So that’s the typical 90-minute weekly meeting: one-third economic-livelihood, one-third health and nutrition, one-third religious instruction. David Sutherland and I.C.M. suspected the program was helpful, long-term; but was it helpful because of the religious values? That is what Dean Karlan, James Choi and a third economist — Gharad Bryan — set out to learn. The first step in a randomized controlled trial is to randomize. They started by recruiting 160 pastors affiliated with I.C.M., and having each pastor identify two Philippine villages where they had never worked. And in each of those two villages—

CHOI: We had the pastors go out and pick out what they believed to be the 30 poorest families in each of these communities. And then we randomly assigned which community the pastor would be ministering to.

KARLAN: So, they would flip a coin about which village they actually went and worked in, and then we used the other one as the comparison. So we had four different treatment arms to the study. We had one treatment arm that received the full package.

The “full package” meaning the standard Transform curriculum: equal parts religious values, health and nutrition, and economic livelihood.

KARLAN: For another quarter of the villages, just the pastor went, not the I.C.M. employees.

So those participants got only the “values,” or religious instruction.

KARLAN: For another quarter of the villages, just the I.C.M. employees went and the pastor never showed up.

CHOI: In those villages, people were receiving just a secular curriculum and did not receive a Christian curriculum.

In these cases, the program was not held in a church, just to make sure the religious component was totally removed.

KARLAN: And then a quarter of the villagers, there’s nothing. They’re left as is.

CHOI: So that was the control group.

Okay, so those were the four conditions from a set of villages that had been chosen randomly, with each condition including hundreds of households. Pretty impressive for a real-world R.C.T.! Six month later, the economists went back to measure the results. Now, how do you do that?

CHOI: Yeah, so we wanted the study to be about the effect of religiosity on economic outcomes.

KARLAN: We ended up working with 6,000 households.

CHOI: So in order to identify the effect of religiosity, we were comparing those who got values, health and livelihoods against those who got just health and livelihoods. And then we also compared those who got just values against those who got nothing.

This would allow them to isolate the economic and others effects of the religious instruction. But first, they needed to know whether the religious instruction increased religiosity.

CHOI: Our study would go nowhere if the religious curriculum didn’t actually change the religiosity of the individuals attending these classes.

They surveyed people to see if the program made them more likely to read the Bible, pray or attend religious services.

CHOI: And we found that it did.

KARLAN: So we found big increases in religious behavior.

CHOI: So then we can go to the second stage, which is, given that the religious programs increased religiosity, what does the effect downstream on other, economic outcomes seem to be?

They looked at everything from food security to life satisfaction to, of course, income. Because the whole purpose of this project — the whole purpose of the I.C.M. charity — was to alleviate poverty. So what was the biggest “downstream effect,” as they call it, of just the religious instruction?

CHOI: What we found was that just being exposed to the religious curriculum increased income by 9 percent, relative to the control group.

KARLAN: And that’s actually pretty noticeable. That’s food. That’s food on the table, and this is a household that has children going to bed hungry.

So that’s pretty interesting: a 9 percent increase in household income that seems to be caused by religiosity. Okay, so that’s the what, but what’s the why? In this kind of research, the what is usually far easier to answer than the why. But why do we think religiosity might drive income? Dean Karlan, James Choi and David Sutherland each have their own ideas — which, to my ear at least, seem to reflect their own perspectives on religion. First, we’ll hear from Sutherland, the man behind this religious charity:

SUTHERLAND: Those people live such depressing, fatalistic lives that they don’t take advantage of opportunities that are available to them. If we can inject hope into those people’s lives, then life can change. And we think that that happens by the secular things that we do, but we also teach them that God loves them and God cares for them, and that aspect of reorienting your whole life around the love of your Creator and therefore the love of people around you — I think reorients their whole life, and it changes their approach to how they decide what they’re going to do the next day.

And James Choi, the unabashedly religious economist:

CHOI: I think that it really probably has to do with the fact that they’re told that they have worth in God’s eyes and their suffering has meaning, and therefore they can continue on, and they can pick themselves up after they experience setbacks. And that’s tremendously encouraging, and it allows them to engage in more productive economic behavior.

And finally, the agnostic, Dean Karlan:

KARLAN: I am very interested in understanding more about issues on hope and aspiration. And drive and ambition, and what is it that makes it so that some ultra-poor households later are doing better and others are not. And some of that is bad luck. But some of that is about taking opportunities and seizing them and running with them.

Karlan and Choi did turn up some data that addresses the why question, to understand why religiosity might lead to higher income.

CHOI: Yes, I think that there were two potential channels that stood out.

KARLAN: One is that we see optimism increase.

CHOI: And second, we found that their grit increased.

KARLAN: Grit is a direct measure of work ethic. Do you always finish tasks? And this is exactly what we’re talking about when we talk about the Protestant work ethic.

Ah, the old Protestant work ethic! That idea has been around for a long time. What is it, exactly? Where’d it come from? And how strong is the evidence that it’s real?

Jorg SPENKUCH: So, how can we attribute this difference in prosperity to religion rather than some other perhaps unmeasured factor?

*     *     *

The randomized controlled trial of a missionary project in the Philippines found that very poor people earned more money as a result of receiving religious instruction. Why? The researchers suspect there were two primary drivers: optimism and grit.

KARLAN: And this is exactly what we’re talking about when we talk about the Protestant work ethic.

The idea goes back to the early 20th century, when the pioneering social scientist Max Weber wrote a long essay called “The Protestant Ethic and the Spirit of Capitalism.”

SPENKUCH: And this was one of the most influential theses in the social sciences, but also one of the most controversial ones.

That’s Jorg Spenkuch.

SPENKUCH: And I’m an assistant professor at the Kellogg School of Management at Northwestern University.

Spenkuch, like Weber, was born and raised in Germany.

SPENKUCH: I would describe myself as agnostic, though I did grow up in a Catholic household. My parents go to church regularly.

As a grad student in the U.S., Spenkuch got interested in Weber’s famous theory.

SPENKUCH: Basically he argued that the Protestant doctrine of self-determination would make people work harder. My question was whether Weber was right, whether there is a causal effect of Protestantism.

To answer this question, we need to go back to 16th-century Germany. A young Catholic priest named Martin Luther had grown disgusted by many Church practices, especially the sale of indulgences — that is, the forgiving of sins in exchange for large donations.

SPENKUCH: And initially he wanted to reform the church from within. But when he met resistance and was persecuted by Church authorities, he decided to break off. And at the time the printing press had just been invented. So he essentially launched what we would call today a large-scale media campaign.

Luther also launched what came to be known as the Protestant Reformation. It would play out all across Europe, for decades, with brutal wars between reformers and Church loyalists, with millions of deaths. On the religious front, meanwhile:

SPENKUCH: His teachings became what’s now the Lutheran faith.

Luther’s teachings departed from Catholicism in several important ways, including a new conception of work.

Tim KELLER: So the medieval church, Catholicism, said, “If you really want to be religious and really please God, you need to go into the monastery.”

That’s Tim Keller, a theologian and founding pastor of Redeemer Presbyterian Church in New York.

KELLER: The Reformation changed that completely. Martin Luther comes along and says, “Look at all the places in the Bible where it says that God feeds every living thing.” He says, “Well, how is God feeding every living thing? Is food just appearing on their table? No.” He says, “Obviously, God is feeding every living thing through human labor.” So even the humblest farmer’s daughter who is milking a cow is actually doing God’s work, and therefore he said, “All work is a calling. All work is doing God’s work.”

And so it came to be that centuries later, Max Weber would argue that Protestantism helped fuel the rise of capitalism, in part by reframing secular work as a religious expression. Weber also argued that discipline played a strong role: that Protestantism tried to instill the discipline of the Catholic clergy into the Protestant laity. Keller again:

KELLER: Catholicism basically said the religious orders — the monks and the nuns — they had to live a very disciplined life. But the hoi polloi, the masses, can be out there and doing what they want to do. Protestants came along and put a lot of emphasis on: all Christians have got to live virtuous, disciplined lives. All Christians. And that created a basis for thrift and self-control.

That’s the theory, at least. There was also the influence of the French theologian John Calvin.

SPENKUCH: Calvin was another reformer. I think of Calvin as Luther on steroids.

KELLER: Where Calvin added his own wrinkle was the idea that your salvation is strictly by grace, without any contribution on your part.

CHOI: If salvation is by grace then how do you know that you are actually saved?

James Choi again.

CHOI: It’s in that sense, people are predestined to either be saved or not.

Max Weber had a very particular belief about Calvinist doctrine. He believed that the doctrine of predestination meant …

KELLER: The doctrine of predestination meant that all Calvinists were radically insecure and that they desperately wanted to prove to themselves that they were saved. But they didn’t have any way of proving that. So how can I be sure I’m predestined? And he said the anxiety of that lack of assurance was channeled into hard work.

Max Weber’s thesis — especially the “insecure Calvinist” angle — was not universally accepted. Jorg Spenkuch again:

SPENKUCH: There are all kinds of people saying this is B.S. It wasn’t just limited to Catholics.

And: Keller.

KELLER: Yeah. I am a Calvinist — I’m a Presbyterian. And all of us — I mean, every Calvinist I know has always laughed at that.

But what about the broader thesis: that Protestantism encouraged a work ethic that drove capitalism and leads to better economic outcomes? Some scholars have attacked Weber’s thesis on empirical grounds. For instance: the very Catholic northern Italy saw the inklings of capitalist enterprise well before the Protestant Reformation. And while it’s true that, even today, Protestant countries tend to be more prosperous than Catholic ones, there’s the old correlation-does-not-mean-causality problem:

SPENKUCH: Yes, the Netherlands and Great Britain are more Protestant than say Spain or Italy, but they also speak different languages, they have different cultures apart from religion, and there are many other things that vary. So how can we attribute this difference in prosperity to religion rather than some other perhaps unmeasured factor?

That was exactly the question that Jorg Spenkuch wanted to answer:

SPENKUCH: Whether there is a causal effect of Protestantism.

But if you think about how hard it was for Dean Karlan and James Choi to learn whether religious instruction helped Filipino villagers earn more money, how was Spenkuch supposed to learn whether a 500-year-old religious transformation had an economic impact that could still be felt today? Well, it started with a simple idea. Remember, the Protestant Reformation triggered wars throughout Europe. Germany, at the time, was made up of more than 1,000 independent territories, each ruled by its own lord or prince.

SPENKUCH: And what settled these wars was the Peace of Augsburg in 1555. And that peace treaty gave local rulers and lords the right to determine the religion of their territory and therefore the religion of all people who lived on that territory. And you have to realize that these were times of indentured servitude. So if my lord converted to Protestantism, that meant I also converted to Protestantism.

And if your lord stayed Catholic, so did you. But if a Catholic family lived in a territory where the local prince chose Protestantism, they could freely migrate to a Catholic area, and vice versa. This created throughout Germany a checkerboard of Protestant and Catholic micro-populations. And Jorg Spenkuch noticed something.

SPENKUCH: What I noticed was that if you overlay a map from 1555 with a map from modern-day Germany and compare the geographic distribution of Protestants and Catholics, they almost match exactly. Meaning that whichever way a prince decided in the aftermath of the Peace has a huge impact on how many Protestants live in the same area today.

Spenkuch realized he could exploit this demographic accident of history.

SPENKUCH: So I used that to determine the effect of Protestantism on modern-day economic outcomes.

Modern Germany is about one-third Protestant, one-third Catholic and one-third non-religious. The Protestant and Catholic populations, beyond religion itself, are quite similar. And Germany being Germany, there was a lot of data available for Spenkuch to drill down further.

SPENKUCH: So we can do better than just looking at, are Protestants richer than Catholics? We can ask, are they more educated than Catholics? Or do they work longer hours than Catholics? Do they want to work longer hours than Catholics? Etc.

With these data in hand, Spenkuch could now control for any differences beyond religious affiliation. Which let him compare apples to apples — or, really, to compare observationally equivalent modern German Protestants and German Catholics to see who earns more money, and perhaps why. So what did he find?

SPENKUCH: I find three things. One: yes, Protestantism increases labor income in modern-day Germany. Two: Protestants work longer hours than Catholics. And three: they don’t earn higher wages. Meaning, yes, Protestants are a little bit more prosperous, but because they work more.

In other words, the Protestant work ethic does seem to be real. So, again, that’s the what; as for the why?

SPENKUCH: So, this is where the data gets very thin. So, what I do find in the data that when you ask people, “How many hours a week would you want to work, conditional on your income adjusting accordingly?” Protestants would want to work longer hours. But what drives these differences? Sort of if you dig a layer deeper, I don’t know.

Some have argued that institutional differences between Protestantism and Catholicism may trickle down to the individual level.

KELLER: The Catholic Church is very, very hierarchical and top-down.

The theologian Tim Keller again.

KELLER: Protestantism is very entrepreneurial, you start your own denomination and you get things started.

CHOI: Horizontally organized religions like Protestantism are argued to be more friendly towards the building of social capital.

The economist James Choi.

CHOI: And indeed, what you do find as you look across countries, and even within regions, is that Protestants are more trusting than Catholics are.

Another idea comes from the fact that Protestantism has historically encouraged its adherents to regularly read scripture.

CHOI: There’s an interesting theory that it was the literacy that was promoted by Protestantism that really led to this economic success.

Indeed, one study that controlled for the literacy rate in Catholic and Protestant countries found no difference in their economic success. In any case, Jorg Spenkuch’s evidence for the Protestant work ethic, at least in the context of modern Germany, is pretty persuasive. We should say, this is not necessarily an economic argument for conversion. Reverend Keller again:

KELLER: If a person is Catholic living in America today and becomes a Protestant, does that mean that person is going to make more money and work harder? I doubt it.

And several downsides of Protestantism have been measured. One recent study found that Protestants in 19th-century Germany were substantially more likely than Catholics to die by suicide. Jorg Spenkuch recently published a paper about another troubling association with Protestantism in Germany.

SPENKUCH: The 10-second summary is that religion is the single-most important determinant of Nazi vote shares at the end of the Weimar Republic. Protestants at the time were two-to-four times as likely as equivalent Catholics to vote for the Nazi party. Now, why was that? In the paper, we argued that that’s due to the Catholic Church’s influence over parishioners.

It’s worth noting that what Spenkuch observed about incomes in Germany being driven by working more hours, does not seem to account for the income gains that Dean Karlan and James Choi observed in the Philippines:

KARLAN: We don’t see an increase in the number of hours that people work, but we do see important changes in what they’re doing.

CHOI: They seem to shift away from agricultural work towards other kinds of work, which may be more high-paying, so that may be just the accounting channel through which their income is increasing.

There are, of course, a lot of differences between the Germans that Spenkuch studied and the Filipinos that Choi and Karlan studied — especially the fact that the Filipinos were punishingly poor and perhaps had quicker gains to make by trying a different kind of work than what they’d been doing. It’s also possible they were already working about as many hours as humanly possible. David Sutherland, the ex-Morgan Stanley C.F.O. who runs the Philippines charity, thinks the religious instruction works for a very basic reason.

SUTHERLAND: I think it reorients their whole life, and it changes their approach to how they decide what they’re going to do the next day.

As devout as Sutherland is personally, he’s not sure the Protestant work ethic is really the key.

SUTHERLAND: Hey, I’m a true believer, right? So, for sure, I would like to say that the Protestant relationship with God is a unique differentiator. Now, could someone with a Buddhist background or a Hindu background or a Muslim background that engaged with God the way they understood it — and if that really injected hope in their life, that clearly would have an important change for them. But whether it’s the same as the Protestant, that’s really hard for me to say. I don’t know. I think that additional study would be needed to know whether that’s true.

This is something that everyone we’ve spoken with today has in common. It may be useful to have learned a few things about the relationship between religion and income; there is much, much more that isn’t yet known.

KARLAN: At the end the day, we have not answered the question of what is the impact of religion. No one study is ever going to do that.

CHOI: We don’t have the data to answer that definitively. So basically, we’re left to speculate.

SPENKUCH: I wish I knew the answer.

So, for the many of you listening today who are entering the season of a religious holiday — and for those who aren’t — here’s to questions asked, questions answered, and even those that remain unanswerable. It’s nice to live in a world that’s still got a little mystery, don’t you think?

*      *      *

Freakonomics Radio is produced by Stitcher and Dubner Productions. This episode was produced by Greg Rosalsky, with help from Stephanie Tam and Zack Lapinski. Our staff also includes Alison Craiglow, Greg Rippin, Harry Huggins and Alvin Melathe. Our theme song is “Mr. Fortune,” by the Hitchhikers; all the other music was composed by Luis Guerra. You can subscribe to Freakonomics Radio on Apple Podcasts, Stitcher, or wherever you get your podcasts.