Freakonomics Radio - 288. Are the Rich Really Less Generous Than the Poor?

Episode Date: May 25, 2017

A series of academic studies suggest that the wealthy are, to put it bluntly, selfish jerks. It's an easy narrative to swallow — but is it true? A trio of economists set out to test the theory. All it took was a Dutch postal worker's uniform, some envelopes stuffed with cash, and a slight sense of the absurd.

What inspired this is a discussion that has come up in the last four or five years about the growing income disparity. The rich growing richer, the poorer poorer. That's Jim Andreoni. I'm a professor of economics at the University of California in San Diego. The discussion he's talking about, now I'm guessing you've had this discussion yourself, the rich, as the data have shown, are getting richer. So it's important to know whether the rich are going to work in the best interest of the whole society. So one obvious question to ask, how does wealth affect how a given person treats other people?

The scientific evidence to date has been not very encouraging. So no, it's not just you. Science also agrees that the more money a person has, the more likely she is to be an inconsiderate, rude jerk. We took to the streets to see how widely held this view is. We asked a simple question. Who do you think are more selfish? Rich people or poor people? That's a tough question because I've known both.

Rich people are more selfish because if you live uncomfortable and you see the next man out here not living and you can't help him, that's selfish. That is selfish. I would say the poor person is more willing to give because they know what it's like to not have. I don't think it depends on your wealth i think it depends on the type of person you are and the way you were raised maybe rich people i think just because of maybe more of the stigma i don't know if they actually are but i

think rich people i don't know the answer to that i know a number of rich people who are extremely generous but at the same time i have seen homeless people be very generous with each other. I think it depends on the individual. I don't necessarily think it's an economic determination. We've been told, however, that it is an economic determination. In a country more and more polarized by inequality, Paul Piff led a series of startling studies. Wealthier participants took two times as much candy from children as did poor participants. There are a lot of studies that came out saying that in some domains, some ways, a lot of ways,

the rich people seem to be less pro-social is the term that psychologists use for that. Yeah, there's all this view, right, in society that the rich are super selfish. That's Jan Stoop, another economist. Even in the Bible, right, there's this quote that it is easier for a camel to crawl through the eye of a needle than it is for a rich man to enter the kingdom of heaven. Even in pop songs, you hear about how selfish the rich are. Everybody knows the fight was fixed. In pop songs, you hear about how selfish the rich are. But here's something important to keep in mind. A lot of the scientific evidence for the rich being selfish came from lab experiments.

But there are differences between the lab, our lab setting, and the field. That's Nikos Nikiforakis, another economist. So it's important to go and check your intuition in the field. And that's exactly what Nikiforakis, Stope, and Andreoni did. They ran a field experiment. Well, why don't we just go and throw these letters to houses of the rich and the poor people and see, you know, who is nicer, who is more pro-social.

All it took was a Dutch postal worker's uniform, some envelopes stuffed with cash, and a slight sense of the absurd. Today on Freakonomics Radio, what did this experiment show? Yeah, so for us, the results were quite shocking. From WNYC Studios, this is Freakonomics Radio, the podcast that explores the hidden side of everything.

Here's your host, Stephen Dubner. So, would you consider yourself an altruistic person, not altruistic? Well, you know, when you study altruism, people always presume that's because you are altruistic. So that opens you up to lots of people hanging around you in your cocktail hour time, hoping you're going to pay for the drinks. I would think you could make the opposite argument, though, which is, plainly, I can't buy you a drink, because that would establish a set of priors that would show a bias in my research. Yes, that's what I say. If I understood this altruism stuff, I wouldn't have to study it. Jim Andreoni was a pioneer in developing what's known as the economics of altruism. The idea that studying people who are governed by morals, are governed by care and concern for others, and who work hard to give their money away, all seem like outside of economics. So that was something that was for the sociologists to understand,

maybe some psychologists, but it's really not economic. Why didn't you leave it to the sociologists? What was it that led you to become interested? There was a paper that I read when I was in graduate school by a guy named Russell Roberts. He drew some conclusions that I thought were, didn't quite make sense to me. And that was that he assumed that people care about the total supply of charity,

irrespective of where it comes from. So if the government comes and takes money from your pocket, gives that to the charity, then you should dollar for dollar withdraw donations to the charity. So was this about the notion of crowding out and how one kind of dollar would crowd out another? Yes. So that's what we've come to call the crowding out hypothesis in economics. That you don't care where the dollars come from or how they get to the charity. You just care what the final consumption of the charity and the final consumption of yourself.

And any path to that is just the same. And if you have that as your assumption, you're going to be able to draw conclusions like the government should just get out of the business of helping poor people. Because if we're altruistic, then the private sector will take over that responsibility all on its own. And what made you care enough about that to want to do something about it, to want to challenge it? then the private sector will take over that responsibility all on its own. And what made you care enough about that to want to do something about it, to want to challenge it? Because I'm stubborn, basically. And I see something that I think isn't right, I have to kind of go figure it out and figure out what's not sitting with me until I produced a paper about it. The paper that Andreoni produced in 1988 showed that government contributions did not completely crowd out private gifts. That is,

in contrast to the direct dollar-for-dollar reduction model, a $1 increase in government contributions decreased private giving by only 5 to 28 cents. This led to the next question. Why do so many people give away so much money when they're not obligated? Andreoni called this phenomenon warm glow altruism. The idea that people are motivated to give to charities for reasons other than the output or the production of the charity itself that i'm happier when it's coming out of my wallet but what makes me happier when it comes out of my wallet that's boy there's a just a huge list of things to think about that could possibly be be at stake here name some oh so my personal impact i can have some guilt that I'm relieving. I have some pride.

I have sympathy for the person that's being helped. I get a reputational boost that helps me in other domains. All those things are motivations. Some of those stories are much more complicated than others to tell. I think a lot of people would hear that and say, but wait a minute. Altruism is meant to be, you know, pure, giving, selfless, and you're saying there's utility derived from giving. So, how do you square that circle? We call it impure altruism. So, you know, I'm not going to give to an organization that I don't

think is doing good work and the work that I believe in that I want to support. Just like I'm not going to eat food that I'm not hungry for. But the actual food that I choose, I choose because it tastes good. Andreoni was arguing that pure altruism is very rare. Much more common are impure or warm glow altruism, which reward the giver along with the recipient. Now, you might argue this presents humanity in a bad light, that even when we give to charity, we want something for ourselves. But that's what the data seem to indicate. And moreover, if you really care about raising money for charity, wouldn't you like to understand a giver's true motivations? And, even more valuable, what specifically are the invisible forces lying behind altruistic behavior? And that's what, basically, in the last 25 years or so since I published that paper, there's been a huge, rich literature trying to answer that question. Andreoni and other economists have explored all kinds of angles on altruism.

The power of matching funds and prizes, the role played by guilt and the herd mentality, and how blonde women raise more money than anyone else by a long shot. If you want to hear more about this, you can check out an earlier Freakonomics Radio episode. It's called How to Raise Money Without Killing a Kitten. Now, in the last several years, as income inequality has become a hot topic, there's been more and more work on the relationship between income and altruism and other forms of what's called pro-social behavior. You may have heard about this research. Experimental evidence that rich people are more likely to break the law while driving, cheat in a game of chance, also to lie during negotiations, and endorse unethical behavior, including stealing at work. The academic paper that resulted made headlines everywhere.

The paper making headlines was by Paul Piff. I'm an assistant professor of psychology and social behavior at the University of California, Irvine. The evidence he presented was disturbing, to say the least. People all the way at the top who made $150,000, $200,000 a year were actually cheating four times as much as someone all the way at the bottom who made under $15,000 a year just to win credits for a $50 cash prize. But how compelling was the evidence? There's a difference between what we then call behavior or preferences. That, again, is Jan Stoop.

I am a behavioral economist at the Erasmus School of Economics in the Netherlands. And the reason this matters, the difference between a behavior and preferences? The reason why this matters is that situations give different incentives to behave differently. We call it the endogeneity problem. Jim Andreoni again. Are the rich fundamentally different people than the poor? Or do the rich face fundamentally different choices than the poor? For example, there are studies that show that the rich tax evade more. Does that make them more selfish? A rich person is more likely to have self-employment income, going to itemize their taxes, be able to overstate some deductions here,

hide some things there, and more easily get away with cheating. But if you're poor, you probably have a job that pays a wage and your employer reports your income and your taxes to the IRS. And there's really nothing, no choices for you to make, no opportunity for you to cheat on your taxes. So also the previous literature, for example, about the rich maybe speeding more in traffic or cutting off other pedestrians. A rich person might say, well, $200 is not a pittance,

but it might materially affect my life. Whereas a poor person would say, if I get that $200 traffic ticket, I'm not going to be able to pay my daycare worker this month. So to conclude from this behavior that the rich have more selfish preferences, that seems like a bit of a stretch. So that made us wonder. They wondered, in part, because many of the findings about the rich being more selfish were derived from surveys or lab experiments.

But it can be tricky to measure something like altruism in the lab, for several reasons. Yeah, the term is the experiment or demand effect. So if subjects know that they participate in an experiment, they may behave a little bit differently than they would otherwise. If they think you're studying generosity, they don't want to be seen as not being generous. Also, many of the studies, for example, by Paul Piff, use student subjects. And they then use a trick where they prime these subjects to either think that they're poor or to think that they're rich. And they do that by having them compare themselves to the richest or the poorest people in the country, something like that. But from an objective standpoint, they're sort of an homogenous group, right?

They're about the same age and about the same wealth. What we found is that the highest earners, they were actually substantially less likely to give to help others. Nikos Nikiforakis again. I'm a professor of economics at New York University in Abu Dhabi. He had run his own lab experiments. We actually had individuals come into the lab, perform a task, and earn a non-trivial amount of money. And then we paid them according to their relative performance and asked them at the end whether they would like to share some of their earnings with other people who may have earned less in the lab. His findings were similar to Paul Piff's. People with more money were less likely to share. But of course the problem here is that people outside the laboratory earn their wealth in very different ways.

Some people just inherit their large fortunes, or they just are born with a gift and they have a brilliant idea that makes them very wealthy. Why else might lab findings not reflect reality? Another reason is, in our lab, the most competitive people would inevitably earn more. But in reality, we also have vetting processes, typically in organizations, such that we reward competitiveness, but you also want to make sure that the person you put in a key position is not some kind of mean-spirited, selfish person who just wants to extract resources from the organization. And the third reason? And of course, the third reason is that these laboratory studies are typically done with students

who earn, in our case, a considerable amount of money. But when we're talking generalizing these results, the rich people are at least millionaires. There are studies that use panels. Jan Stoop again. And panels being a group of people who will sign up to participate in a study. So with this trick, you can get rich or poor households to get involved in your experiment. But here then we have another problem. And that's what we call selection bias. So it could be a certain type of household that signs up to participate in such panels. And those could be the more pro-social, scientific, do-gooders households.

So it seems it's always a problem, right? No matter what kind of experimental method you use, there is a problem. The actual experiment you'd like to do is get Eddie Murphy in here and Dan Aykroyd and trade places, right? As in the movie Trading Places. Make the rich one poor and the poor one rich and see if they adopt each other's behaviors. I'll wait till you get to about 64, then I'd buy. You'll have cleared out all the suckers by then.

I was poor and no one liked me. I lost my job, I lost my house. Penelope hated me. But, you know, we can't do that experiment in reality. So we have to see if we can measure things about the environment or about the choices people can make that would allow us to run this experiment in our minds and through our data. So if we can put something in the field,

and we can do it in a way where people don't actually know they're being studied that's the best because that's the actual behavior that we're trying to study coming up on Freakonomics Radio how Andreoni, Stope and Nika Farrakis pulled off this field experiment yeah I was actually I was when when I did this I I was super nervous. And, of course, we'll tell you the results. When I first saw the results, I kind of thought we failed. That's right after this break. The economists Jim Andreoni, Nikos Nikiforakis and Jan Stoop wanted to run a field experiment to find out if the rich are more selfish than the poor. Here's Stoop. The background is there is this laboratory experiment, which is pretty famous in our community, called the dictator game. For what it's worth, we wrote about the dictator game at length in Super Freakonomics and a related game called Ultimatum.

And in the dictator game, there are two players. One is the dictator and he gets an amount of money, typically $10, and he can split that between himself and the recipient. Split it by, say, putting some of the money in an envelope and giving it to the other person. And any amount of money that's given is typically interpreted as a measure for altruism. But this kind of lab experiment, as we discussed earlier, has a lot of limitations, especially the subject's awareness that they're in an experiment which might increase their desire to appear generous. So I was thinking, how can I have such a game in the real world where money comes falling from the sky, let's say, and people can choose to divide it between

themselves and someone else? And then I thought of this envelope trick. Not the envelope trick from the dictator game, a different kind of envelope trick. So there's this household and all of a sudden they receive an envelope with cash. It's not theirs, but they have the power to give it back, yes or no. So it is a little bit like they play a dictator game, but then in real life and without people knowing that they participate in an experiment. So when Jan phoned me up and he said, you know, I have this technique. I said, well, why don't we just go and throw these letters to houses of the rich and the poor people and see, you know, who is nicer, who is more pro-social.

We found poor households and rich households in the Netherlands. And we did it there because we can get a lot more data about the households so we can know how rich they are and how poor they are. That's because European governments, like the Netherlands government, have much more open policies about collecting data for research purposes. Plus which, Jan Stoop happened to live there. I said, look guys, let me conduct all the treatments here in Holland, because I just, I know that you're probably too busy to do it, and I know that I would love it too much

to do it. So I think it's a win-win for everyone. The idea was to intentionally misdeliver a letter that was addressed to a real person with real money in it and see whether people kept the letter or send it on to the rightful recipient by dropping it into a mailbox on the street. We picked a little greeting card that was signed to be coming from somebody's grandpa. The front side was a picture of a windmill, just an old fashioned, something typically that a grandfather could choose, right? And then on the other side of the envelope, we had a message, it was, Dear Joost, here is 20 euros for you, grandfather. And then there was a note of 20 euros. Joost, the addressee, was a real friend of Jan's

And the envelope listed his real street address So that it could be easily forwarded It also had a real-looking postmarked stamp Courtesy of some skillful photoshopping There was only one weird detail The envelope was semi-transparent So you could clearly read the card and see the cash Because then our subject in the experiment, the rich household or

the poor household, they can see that grandfather intends to send 20 euros or 5 euros to Joost. For experimental purposes, there would be two cash treatments, either 5 euros or 20 euros. And so I get this question a lot, right? Who sends out money in a transparent envelope? That is kind of weird, right? Yeah. But from a scientific standpoint, we are interested in treatment differences. We always use this same envelope. So that means that the weirdness of it is the same in all treatments.

Okay. So the recipient can clearly see the cash inside the envelope. Now this is something that's a benefit to the proper owner, but it's also a benefit to the accidental recipient. They also loaded some envelopes not with cash, but with bank transfer cards, something like a check, which therefore

can only be cashed by the person it's made out to, again, in denominations of both 5 and 20 euros. Which is worthless to the accidental recipient, but it's still worth 20 euros to the intended recipient. So if you're altruistic, you still have the same incentive because you're going to help the intended recipient just as much in both cases. If you're selfish, you're going to benefit much more when it's got cash in it.

So we try to map where all the rich people live and where all the poor people live, and then we randomize them into treatment. So our observations are filled with the people that we want, but there's also, there is no selection bias. They identified 360 households, 180 rich, 180 poor. The poor houses were identified by public assistance and subsidized housing records. In rich households, we identified those by the market value of the houses that were for sale in that neighborhood. And after the envelope stunt was over, they'd be able to get hold of government data to confirm each household's level of wealth. So we know how rich and how poor they are. The average wealth of the rich is about $2.5 million. And the average wealth of the rich is about 2.5 million.

And the average wealth of the poor is about 25,000. So that's a factor of almost 100. From October through December of 2013, Stope delivered the envelopes at regular intervals. Again, there were 360 envelopes, all addressed to little Joost, all containing either cash or transfer cards worth either five or 20 euros, and all intentionally misdelivered

to specifically identified rich or poor households. Yeah, so through friends of mine, I was able to acquire an official outfit of the mail company that we have in the Netherlands. So that means I had a polo and I had a bag and also even I had a cap. So I would be dressed up and cycle through the city on my way to a rich household or a poor household. For the record, Freakonomics Radio does not endorse impersonating a postal worker, even in the name of scientific research. Yeah, I was actually, when I did this, I was super nervous. So whenever I misdelivered an envelope at a mailbox, I tried to get out as fast as possible.

The rich, they tend to have huge driveways, right? And therefore, most of them have a mailbox at the side of the street. So I loved those villas because that's a hit and run, right? And also, obviously, I got chased by a dog once. And I got away. No one saw me. But yeah, I thought that was pretty cool as well. And then we just waited to see what would show up.

So the big question then is, what did you find? And what did you learn about rich people versus poor people in this kind of creative, interesting, albeit very unusual form of what we might call altruism? Given the research that we had been reading from Paul Piff and others like that, we kind of expected the rich people to be less likely to return these envelopes. What we found was in fact the opposite. Yeah, so for us, the results were quite shocking. The rich returned way, way more than the poor.

In fact, they returned twice as much. So return rates of the rich were roughly 80% and return rates of the rich were roughly 80% and return rates of the poor were roughly 40%. And the rich didn't care whether there was money in the envelope or not. Cash versus the check, they returned them at about the same rate. We find that roughly 25% of the cash came back from poor families, whereas roughly 75% of cash came back from the rich families. But for us, the biggest shock was in observing that the non-cash envelopes were also not returned as much by the poor families. The poor also, for these envelopes, returned roughly half. So it's looking to us when we first get the results that the rich people are actually much more altruistic than the poor.

I kind of thought we failed. We found the wrong result, which in the sense I did expect that the poor would greatly outperform in terms of kindness the rich. Since rich people and poor people may differ on dimensions other than income, what were some of the potential confounding factors that might influence or pollute your findings? Yeah, so when the experiment was over, we got data from Statistics Netherlands that provided us with details on all our subject houses, their education level, their age, for all people in each household. We include all of these in this regression analysis and basically what we find. So then we can see statistically if all these factors matter. And quite to our surprise, none of them seemed to have an effect.

But then we noticed a couple other things. First of all, the envelopes that the rich people got were being returned much faster than the envelopes of the poor folks. And it is here, actually, where Steve Levitt played a big role. Steve Levitt is my Freakonomics friend and co-author. He's an economist at the University of Chicago. Hey, Dubner. How are you?

Hey, Levitt. You can go back to what you're doing. Thanks. So the first time I presented this paper was at a seminar at the University of Chicago. And I was telling this story about how the rich and poor differ in selfishness. So this slide came up where I present the figure of the return rates. And Steve Levitt said, well, Jan, you're mistaken. So you tell us the story that you're measuring pro-social behavior of the rich and the poor, but this is not what you're actually measuring. You're measuring the incapability of the poor to return an envelope.

And for me, that was such a new, fresh insight. And this then reminded us of a discussion that's been very popular lately about poverty actually being a causal factor in people not being able to get stuff done as efficiently. I guess there are other factors like, you know, a poor person is more likely, I'm assuming, to have a longer commute time and therefore less just time on the clock, right? A lot of things like that. Yeah, so that might well be true as well. They have more strict hours at work, perhaps, probably more likely to be single parents, all kinds of things that are correlated with being poor that add to the stress that people have, preoccupy their minds, make it harder for them to keep their priorities. There's this famous study where a bunch of researchers from Harvard

go to India to poor farmers. And these farmers have a harvest once per year. So that means that once per year, they have a big bag of money because they sell everything. But as the year progresses, their bag of money becomes less and less. Hence, their financial stress increases and increases. So what they did is they did an IQ test at the moment when the poor were sort of at their richest and when they were at their poorest. And they found that in the IQ test, they had a lower score when the poor were at their poorest. So this story inspired us to look at our data through a different lens. So if we look at the distance from the payday, we can maybe see a difference between the behavior of the poor and the behavior of the rich. Expecting that as we get farther from payday, we're going to get more stress in the poor

households and fewer things added to their to-do list. If you cut up a month in four weeks, then in Holland at least, typically people get paid their salaries or their unemployment benefits or pensions in the last week of the month. We then looked at how much is returned one week later, two weeks later, and three weeks later. And for the rich, we found no pattern at all. So they don't care. They return envelopes equally spread out over the month, but not so much for the poor. They were returned at a very high rate at the week of the paycheck arriving. And then the probability of being returned went down, down, down, down, practically to zero. And it was the week right before payday. It's a very striking pattern. So that's consistent with this idea that as the month goes on and your budget gets tighter and

your constraints get harder and your stress builds, you have difficulty accomplishing small chores or prioritizing things. But this holds for the non-cash envelopes. It did not hold for the cash envelopes, which is also sort of a puzzle. Now, to us, it tells an interesting story, though. So if you're poor and you get an envelope that has a 20-euro note in it by mistake, that 20 euros is very valuable to you. And you have to ask yourself, am I a responsible parent, for instance, if I don't take this 20 euros? And that's a real moral dilemma there in

that. Yes, exactly. The marginal utility of money. Yeah. So an extra 100 euros for a rich family does not give as much pleasure as for someone who's poor. So we call that the diminishing marginal utility of cash. So those are our three variables. One we call our basic propensity for altruism. One we call our basic need for cash. And the other is the stress of being poor. And then the different environmental factors between the rich and the poor, which try to factor out. So we get at the true value of that basic underlying altruism. Right. So we then need a model, a theoretical model to map behavior to preferences. We model it as follows. So when a household returns an envelope, if alpha, it's altruism towards Joost, the intended recipient of the card, minus N, the neediness of the contents of the envelope, minus P, so the financial pressure, the stress costs.

When this is greater than zero, so when altruism outweighs the neediness and the stress. Then the household returns the envelope. So with the data that we have, we can actually estimate the alpha, the N, and the P. And when we account for those, what we find is that the basic tendency to want to do the right thing is the same for the rich and the poor. But it's the fact of being rich and poor that affects these other aspects of the decision and affects the outcome. What we find is not surprisingly that the M differs between the rich and the poor. So meaning that the poor need the money harder than the rich. Also in line with this relatively new literature on financial stresses of the poor, we find that the P, the financial pressure, is greater for the poor than for the rich. So then we have left alpha, altruism, and we find

that these are the same between the rich and the poor. I consider this to be a really hardcore economic insight, right? So as economists, we always say that incentives shape behavior. And this is another example of that. So there are many other studies that look mainly, well, actually, that look only at behavior. And so far, it seems as if our study is the only study that has disentangled behavior from preferences. So what does this study, for all its cleverness and novelty and thoughtfulness, what does it actually have to teach us? We know that poverty has lots of social costs. Our study actually suggests there is one more and potentially an important one.

It means that when someone loses, let's say, some of their income, this doesn't only affect them personally, but it also affects people around them who otherwise may have benefited from prosocial actions, altruistic actions of that person who's now poor. So when we're thinking about the benefits and the cost of poverty programs, we need to take that factor into account, that the financial pressure may make the poor behave more selfishly than they would have in different circumstances. It's true true we don't know how this would translate to different countries what we did is a study in a medium-sized city in holland in 2014 so the question is how does this translate to behavior in japan in 2027 for example that's hard to say there's this

hypothesis that income inequality matters a lot. So the more income inequality that there is, the more selfish the rich behave. And that thought it'd be worth hearing from Paul Piff, the psychologist whose research has helped build that conventional wisdom. Okay. My name is Paul Piff, and what I study are the origins of human kindness and how inequality, and in particular economic inequality, shape relations between individuals and within groups in society. So what does Piff make of the argument that a field experiment like this one with real rich and poor people and real money is more robust than a lab experiment? So the first thing I would say is the field experiment that Andreoni and colleagues ran, it is a really compelling and well thought out experiment. So, I mean, I think that this study and other studies that are emerging are all a piece of the complex mosaic that's emerging, which is to say that how wealth and poverty shape the mind is complex. It's multifaceted. Those relationships aren't categorical or essential, but are nuanced. Piff does have a couple of qualifications. I think the first

qualification I would make is that any single study, like any single stroke of the brush on a canvas, won't give you a full sense of the picture. Pro-sociality is something really, really broad. It broadly refers to times, instances, actions that prioritize the welfare of someone else or the well-being of other people at sometimes a cost to yourself. when waiting at a coffee shop, or stopping for a pedestrian who's waiting to cross at a crosswalk, or volunteering your time to help someone else, or giving to charity. In each of these different situations, you can imagine that a specific behavior or decision could be influenced by any number of factors. There are all sorts of other incentives that play into human decision-making, And one of those things is social incentives. And whereas people from less advantaged backgrounds have less money by definition, for them, their ability to say, rely on a friend to get by when times are tough,

that is ever more salient. And that's a primary coping mechanism for people who are poor or who are relatively disadvantaged. And so in contexts where there's an opportunity to connect with someone, when there's an opportunity to kind of invest in a relationship, it's in those instances that I would say or that we would predict you're most likely to find these rich-poor differences in pro-sociality that align with what we've been finding. The scenario we've been talking about today, meanwhile, the Dutch field experiment is less a social context and more one where those social incentives have been kind of removed from the picture. It also involves a kind of pro-social behavior or a kind of measure, if you will, a dependent variable that I think people rarely encounter in their daily lives. In fact, I think if you were to ask a friend of yours or someone off the street, when is the last time you received an envelope that was see-through with cash in it that was mistakenly put in your inbox, but that was actually addressed for someone else,

most people would say, that's never happened to me. So that's not to say that you can dismiss the results of Andreoni and his colleague's paper at all, but I think it needs to be interpreted within the limitations of the measure. That said, PIF seems to appreciate the economist's contribution. And what it prioritizes in my mind is that it's important to get outside the lab and complement your laboratory work with field experiments, but also to complement your field experiments with laboratory work. And so there's an approach of complementarity that I want to stress that I think is really important. Both kinds of approaches are really important. And this being among the first really careful field experiments that's been run in a clear way, a clear contribution. In the end, perhaps the most salient lesson from this Dutch field experiment is just how hard it is to generalize about any group of people,

male or female, liberal or conservative, rich or poor, because we humans are plainly far more than the sum of our biological parts. We're a dynamic bundle of preferences, decisions, and behaviors. Some of them observable, others not. Jim Andreoni again. And so the moral of this story is, if you try to think a little bit deeper about how the very fact of being rich or being poor affects the kind of choices that you're able to make and the incentives you have to

change your behavior. Before you draw the conclusions that rich people are either better or worse than poor people, you need to ask whether you've accounted for all the ways in which being rich or poor itself affects your behavior. Science requires you to do that. Coming up next time on Freakonomics Radio, they are an American tradition. I know you want it. They're incredibly abundant. That's about 14.5 million acres of turf. They're also incredibly labor and resource intensive.

Every square foot requires 28 gallons of water. We love our lawns but are they worth the trouble? And the cost? Financial, environmental, and otherwise? And what would we be doing with our yards if we didn't turn them into manicured parkland? Like, it could be, you know, a hundred different leafy greens. As part of our occasional stupid stuff series, we take a long, hard look at lawns. That's next time on Freakonomics Radio. Freakonomics Radio was produced by WNYC Studios and Dubner Productions.

This episode was produced by Stephanie Tam. Our staff also includes Shelley Lewis, Merritt Jacob, Greg Rosalski, Christopher Wirth, Eliza Lamber, Allison Hockenberry, Emma Morgenstern, Harry Huggins, and Brian Gutierrez, with engineering help from Rick Kwan. Special thanks to Scott Vincent Andrews and Maria Rosa Lunati at OECD. You can subscribe to Freakonomics Radio on Apple Podcasts, Stitcher, or wherever you get your podcasts. You should also check out our archive at Freakonomics.com. You can stream or download every episode we've ever made.

You can also read the transcripts and look up the underlying research. We can also be found on Twitter, Facebook, blah, blah, blah, or via email at radio at Freakonomics.com. Thanks for listening.