Freakonomics Radio - 266. Trust Me

Episode Date: November 10, 2016

Societies where people trust one another are healthier and wealthier. In the U.S. (and the U.K. and elsewhere), social trust has been falling for decades -- in part because our populations are more diverse. What can we do to fix it?

You may have noticed that we recently held a presidential election. I think you'd agree it was pretty wonderful. According to statisticians, it set all-time highs in civil discourse and social unity. How are we so fortunate? Because America, as we all know and appreciate, is a place where people really trust one another. No, I don't think most people can be trusted because I think everybody's always looking for an angle. Generally speaking, I don't think most people can be trusted. Society seems to have been changing and separating,

and many, many people more than before think about just for themselves. Oh. Apparently I was wrong. Apparently we don't trust one another so much. What's that? Oh. And apparently we didn't set all-time highs in social unity during this election. Sorry.

My mistake. I guess I was thinking of Australia. Australia is the one which looks like it's bucking the trend and moving towards higher social trust in the last 20, 30 years. Or maybe I was thinking of the Netherlands. They're now close to 70% in levels of those who think others can be trusted. In America, meanwhile, if we're being honest, we're not much on trust these days.

Just think of this past election. We mistrusted the candidates and their parties. We mistrusted the police and the FBI. We mistrusted the polls, and some people even mistrusted the election result. But wait, maybe there's a bright side. Maybe it's healthy for society to be untrusting, to be skeptical. Maybe it keeps us on our toes, always looking for ways to improve. What's that? Oh, I'm wrong about that too, Professor? We would be much better off if we were living in a more trustworthy society. Trustworthiness, in short, is a really big deal. All right, then.

Today on Freakonomics Radio, a simple mission to determine why social trust is such a big deal and how to get more of it. Wouldn't you like to know how to do that? I think that might be a trick question. From WNYC Studios, this is Freakonomics Radio, the podcast that explores the hidden side of everything. Here's your host, Stephen Dubner. Social trust is what exactly? It's just one of those things. It's sort of like the dark matter of the economy and society matters very greatly, and yet we don't seem to focus on it very much. That is one of my favorite academic slash policy wonks in the world, David Halpern. I'm the head of the UK's behavioral insight team, often known as the Nudge Unit. The Nudge Unit applies the findings of behavioral science to do things like increase tax payments,

decrease medical errors, and conserve energy. It also looks into broader ideas like social trust. As Halpern was saying... Social trust is an extraordinarily interesting variable and doesn't get anywhere near the attention it deserves. But the basic idea is trying to understand what is it, the kind of the fabric of society that makes economies and indeed just people get along in general. It's clearly so critical for a whole range of outcomes. Outcomes like economic growth. This is a more powerful predictor of future national growth rates than, for example, levels of human capital or skills in the population. Outcomes like individual health. Basically, having someone or feeling that other people can be trusted or people you can rely on in your life is worth a great deal.

It's roughly the same positive effect as giving up smoking. You know, and smoking is really, really bad for you. So, you know, social isolation essentially is incredibly bad for your health. So can people like Halpern reliably measure the level of social trust in a given place? Yes, you can do it in a number of ways. You can ask people, how many names have they got in their file of faxes or in their phones, which should give you some sense of their social networks. You can also measure more subtly with asking a question around social trust. Do you think other people can be trusted? Essentially, that's the question we've been

asking, in fact, for decades. And there's very big national differences in this. Countries range from, you know, many countries like Brazil, where less than 10% of people would say most others can be trusted, to countries like Norway, where more than 70% of people would say most others can be trusted. Countries like the US and the UK are sort of halfway in between, typically in the range 30 to 40% of people say others can be trusted. Places with a lot of social trust also have a lot of what's known as social capital. So most people are familiar with physical capital and financial capital, of course. Can you compare social capital to those in any meaningful way?

The answer is, yeah, in fact, you can have a go at doing it. So you can literally work out how much more do people tend to earn if they have more names in their networks and they know more people or they trust more people in general. You can ask it both at the individual level and you can also ask it really importantly at the community or even national level. So to what extent is it an advantage in terms of your economic growth or your health outcomes to live in a country or a place where people say most others can be trusted? And the answer is it turns out some really quite large numbers indeed. If social trust and social capital are so important, why don't we talk about it all the time, or at least during political campaigns?

So it's an issue which has got long roots, but it doesn't mean that governments had done very much about it until very recently. I should explain, I myself wrote a book a number of years ago on social capital, specifically working with, at the time, with Bob Putnam at Harvard. My name is Bob Putnam. I teach public policy at Harvard University. Putnam is also the author of the landmark book, Bowling Alone, which was published in 2000. But he started thinking about social capital decades earlier. It began with a question about Italian politics.

Some parts of Italy are way more efficient than any state in America. And other parts of Italy are way more corrupt than any place in America. And the question is, why? Why are some places better governed than others? The answer, Putnam concluded, didn't have to do with economic development or education level or politics. It was the degree to which there was a dense civic network in a community. If there was a dense civic network so that people in those places behaved with respect to one another in a trustworthy way,

their governments worked better. And I dubbed that concept social capital. Meaning? The core idea of social capital is so simple that I'm almost embarrassed to say it. It is that social networks have value. Social networks have value first of all to the people who are in the networks. For example, there's a huge amount of work on how social networks help us find jobs. Social networks also exert other more indirect leverage. They have effects on bystanders and not just effects on the people in them. Communities that have high levels of social capital benefit in many ways.

Their kids do better in school. They have lower crime rates. They have other things being equal, higher economic growth rates. Many, many benefits, both personally and collectively. Having made these observations about the power of social capital in Italy, Putnam returned to the U.S. And I was worried, just as a citizen, not as a scholar, I was worried that ever since I personally began to vote, which was way back in the 1960s, America seemed to be going to hell in a handbasket.

And I said to myself, I wonder whether social capital might have something to do with this collapse of American civilization. So he began looking for measures of connectivity in American civic life. Membership in parent-teacher organizations, for instance. That is, what fraction of all parents in any given year belong to the PTA? And I discovered, to my shock, that actually PTA membership had been declining a lot. He looked at membership numbers among rotary clubs, among scout troops, among bowling leagues. Participation was falling in all these groups.

We were becoming more and more isolated. Or, as a friend suggested to me once, you mean we're bowling alone? That's a really enormous effect. David Halpern again from the British government's behavioral insights team. As obvious as the benefits of social capital might seem. We almost hardly seem to notice that it's there. So it's incredibly consequential. And we see it in lots of areas of policy that we touch on.

So you write this about low trust. Low trust implies a society where you have to keep an eye over your shoulder, where deals need lawyers instead of handshakes, where you don't see the point of paying your tax or recycling your rubbish since you doubt your neighbor will do so, and where you employ your cousin or your brother-in-law to work for you rather than a stranger who would probably be much better at the job. So that has all kinds of business and ultimately economic implications. However, when you talk about high trust being good for us on a personal level, whether it's health or individual income, do the two necessarily go in hand? In other words, can we have a society that has a business climate where there isn't a lot of trust,

and therefore you do need all those lawyers instead of the handshakes, but where you have good social trust among neighbors, family and friends, communities, and so on? Or are they really the same thing that you're talking about? Well, there is a key distinction, and Bob Putnam has often made this too, between what's sometimes called bonding social capital and bridging social capital. Social capital is about social networks, but not all social networks are identical. And one important distinction is between ties that link us to other people like us. That's called bonding social capital. Bonding social capital often refers to your closeness to your friends, your relatives,

those immediately around you. It's particularly important, it turns out, for things such as health outcomes. Because empirically, if you get sick, the people who are likely to bring you chicken soup are likely to represent your bonding social capital. So bonding social capital is plainly important, but it's primarily about our ties to people we're close with. When it comes to how our broader society gets along, that's where bridging social capital comes in. Do you trust not just your immediate neighbor, but those people in your community more generally, or indeed even relative strangers who you meet in everyday life in your country or your society? So my ties, my friendships to people of a different religion or a different race or a different economic class or a different occupation or a different age, that's bridging social capital.

That tends to be really important. Especially in a modern, diverse democracy like ours. And therefore, what worries me most about trends in America is the decline in bridging social capital. It's something we saw plenty of during the presidential election. So much distrust among so many separate constituencies, a constant splintering and re-splintering instead of the drawing together that America likes to be known for. So coming up on Freakonomics Radio, we look at an experiment designed to measure trust along racial and ethnic lines. The news is not good. A lot of the cheating was across racial and ethnic lines.

But there is some good news. For instance, people who go to university end up trusting much more than those who don't. And when we went out to the streets of New York and asked people the standard survey question, generally speaking, would you say most people can be trusted? We got plenty of affirmatives. I think so, generally. There's more good people in the world than there are bad people, so that leads me to believe that you can trust people.

Yeah, I would say they could. I think that's maybe like a naive assumption on my part, like to believe that there's good in everyone. I've had great experiences with strangers. I think in the majority of cases, people are good. I trust people, yeah. So what have we learned thus far? We've learned that social trust seems to be quite powerful and desirable.

But how do you get more of it? You can't just magically turn up social trust in a given place. You can't force everyone to join the PTA or the church group or whatever. So rather than solving for X, where X is how to increase social trust, let's first solve for Y, where Y is when social trust and social capital are low, why are they low? The short version is that in the short run, increases in diversity seem to be correlated with decreases in social capital. That's Bob Putnam again. He and others have observed over and over that diversity, racial, ethnic, religious, and so on, make trust more elusive. Consider some research done by the Harvard economist Ed Glazer. Trust is everywhere in economic transactions.

So we wanted to contribute to this literature. And one of the things that seemed very important to us was measuring trust, was measuring social capital. Rather than relying solely on survey data, Glazer and his colleagues set up an experiment. Not that experiments are perfect either. So we took a bunch of Harvard undergraduates, because what could possibly be more representative than that? They tried to measure trust in a variety of ways, including a game where students were paired with each other, with one sending money to the other without being sure whether they'd get the money back.

It's basically meant to mimic the idea of an investor giving money to a firm, and the firm then chooses whether or not to cheat the investor or not. Some students treated their partners fairly. Others, however, essentially cheated, keeping most or all the money for themselves. When did that happen? It happened when the two players didn't look alike. A lot of the cheating was across racial and ethnic lines. And this was primarily white on Asian, meaning the whites were cheating the Asians. And I think there are lots of cases in the world in which we've seen racial fractionalization be related to less than perfectly functioning social relations.

Do you think that more ethnically homogenous societies, you know, one argument behind Scandinavian economic and social successes is that they tend to produce better social outcomes? And do you think there's evidence for that? I do. It is true, for example, that the welfare states are both more generous in ethnically homogeneous places. And it's almost assuredly true that they're also more functional. They function better in more ethnically homogeneous places. It's just easier in lots of ways. There are downsides to that. I mean, I happen to love Stockholm. It's a great city. But certainly one can argue that small homogeneous places are not necessarily as creative as they might be. So I think there are lots of benefits for being in an ethnically and culturally mixed society.

David Halpern again. Look, I'm speaking to you today from London, one of the most diverse cities in the world, and it has a deep vibrancy that follows from that. I mean, the real challenge for us is can we have our cake and eat it? What is it that drives and enables diverse and interesting sort of varied populations, cities, countries to be able to live together well? That's the real challenge. Every place in the world, including us and Canada, are all going to be more ethnically diverse 25 years from now. And that's Bob Putnam. Diversity in the long run is a big advantage. But, he warns, it's not easy to do diversity. Diversity brings out the turtle in us that in a more diverse setting, everybody kind of pulls in and disconnects from their neighbors.

But look, if the world is becoming more diverse, and if diversity tends to lower social trust, and if social trust is so important, shouldn't we be looking at ways to handle this problem? I think so. Although, let's face it, there isn't that much work really doing intervention studies to figure it out. What strategies I would want to emphasize for moving in a positive direction would be more context in which people connect with one another across lines of race or economics or gender or age. Some classic examples of that, sports teams, the military, and university. People who go to university end up trusting much more than those who don't, particularly when they go away residentially. It doesn't look like it's explained by income alone. So there's something about the experience of going off as a young person in an environment where you were lots of other young people from different backgrounds and so on, hopefully, and different ethnicities. You learn the habits of trust because you're in an environment where you can trust other people. They are trustworthy. And you internalize these habits and you take them with you the rest of your life.

So we tend to not think of going away to university as being the reason why you're doing it is to build social capital and social trust. We think about learning skills and so on. But it may well be that it has as much or even more value in terms of culturing social trust going forward. The question is, do you have to do it in university? Can you do it another way? So in the UK, following partly an American lead, the government has championed a national citizen service. And what this means is for every young person, essentially a 17-year-old increasingly,

it starts off with a, not everyone does it at the moment, but it's more and more every single year, goes and does voluntary experience, community service. This deliberately includes a couple of weeks which are residential, and deliberately includes mixing with people from all different walks of life. Look, it's only limited data, but in terms of before-after data, we see significant impacts in terms of higher levels of trust between groups and individuals, as well as, incidentally, higher levels of life satisfaction and well-being too. So it looks like we can do something about it. It's also helpful, David Halpern says, simply to look at the countries where social trust has been rising and see what they're up to.

The Netherlands, for instance. In the most recent data, it looks like it's one of the biggest rises. So the Netherlands had pretty similar levels of social trust in the 1980s to America and the UK. But whereas we've now drifted down towards sort of 30 odd percent, they're now up close to 70 percent in levels of those you think others can be trusted. What would you say it's, you know, you have to be a bit careful when you pick off one country, is it has wrestled quite hard with the issues of not just inequality, but social differences. They've really tried to do a lot in relation to making people essentially build cohesion, particularly Amsterdam is a very famous area for its long been an extremely multicultural city. It's had issues over that over time, but they've really, in a sort of succession of governments, have tried to quite actively make groups get along with each other in quite an

active way. So that may, of course, itself root in the Netherlands has quite a deep culture of a strong sense of the law being trustworthy and that contracts will be honoured and so on. It's what helped to power its economic success in previous centuries. So it does have that tradition also to draw on. And what role does technology play in social capital, especially the bridging social capital that people like Halpern and Putnam say is so important? In his book Bowling Alone, Putnam found that social capital was relatively low in the U.S. in the early 1900s and rose fairly steadily through the 1960s. But that's when the decline began.

I looked hard to find explanations, and television, I argued, is really bad for social connectivity for many reasons. More television watching, Putnam wrote, means less of virtually every form of civic participation and social involvement. As Bob sometimes put it, I think rather elegantly, when we were looking forward in terms of technology or the internet and of course even pre-Facebook and so on, would it be, in his words, a fancy television? In other words, it will isolate us more and more. Or would it be a fancy telephone and would connect us more and more? Because technology has both those capabilities. So when I played video games when I was a kid, you basically did them mostly by yourself or with a friend. When I look at my teenage kids playing videos, they're actually talking to each other all the time. To some extent, it looks like to me that

we get the technology that we want. And even this is true at a sort of societal level. So one of the arguments you can make, and in my view is true anyway, about explaining some of these differences in the trajectories across countries is in Anglo-Saxon countries, we've often used our wealth to buy technology and other experiences. That means we don't have to deal with other people. The inconveniences of having to go to a concert where I have to listen to music I really like. I can just stay at home and I just watch what I want and so on and choose it.

And even in the level of, if I think about my kids versus me growing up, I mean, when I was growing up, we had one TV and there were five kids in the household. We really had to negotiate pretty hard about what we were gonna watch. My kids don't have to do that and probably not yours either. There are more screens in the house than there are people. They can all go off and do their own thing. To some extent, that's us using our wealth to escape from

having to negotiate with other people. But that isn't necessarily the case. Some people and some countries seem to use their wealth more to find ways of connecting more with other people. And the technology has both these capabilities. And we can't just blame it. It's the choices we're making and how we use it and the technology which we're kind of asking and bringing forth. It reminds me a bit of, we once looked into the kind of global decline of hitchhiking, for instance, one of the central reasons being that people no longer trusted strangers to not kill each other, really, is what it boiled down to, even though there was apparently very little killing involved, but just the fear of one. And yet now, Uber is a 60-some billion

dollar company that's basically all about using technology to lure a complete stranger into your car, which I guess argues, if nothing else, the fact that technology can be harnessed very much in either direction. Indeed. So, I mean, as you say, there's actually two points here, and there's a really important behavioral one, which I think we've only figured out in recent years to bring together these different literatures. How does it relate to behavioral scientists versus those people studying social capital? We look like we have certain systematic biases about how we estimate whether we think other people can be trusted. And in essence, we overestimate quite systematically the prevalence of bad behaviour. We overestimate the number of people who are cheating on their taxes, or take a sickie off

work, or do other kinds of bad things. This doesn't seem to be just the media, although that may reinforce it. It seems to be a bit how we're wired as human beings. So why is that relevant? Why is this to do with technology? Actually, technology can help you solve some of those issues. So when you're buying something on eBay or you're trying to decide where to go using a trip advisor, it's actually you're getting some much better information from the experiences of other people, as opposed to your guesstimate, which is often systematically biased. So it turns out it's a way we can sometimes use technology to solve some of these trust issues, not just in relation to specific products and shall I buy this thing from this person, but

potentially more generally in relation to how do we trust other people. Because ultimately, the social trust question must rest on something. It must be a measure of actual trustworthiness. The United States, for all the factionalism and bitterness we've seen during the presidential campaign, and let's be honest, for years and years preceding, is actually well positioned to regenerate social trust, even as the country becomes more diverse. Why? Because, as Bob Putnam argues, we've done it before. A lot. If we were talking in America in the 1920s or 1930s, the difference between Irish people

and Italian people would have been enormous. I have some friends who got married in the 1960s. He was from an Italian background. She was from an Irish background. And when they got married, everybody called it a mixed marriage. Parents on both sides all said mixed marriages never work. And now that seems like a joke because what's happened in the ensuing years is that the line, the sharp line between Italians and Irish has just disappeared in America. It's not that they don't know that they're from Irish or Italian backgrounds, but it no longer has that same salience. We've done this repeatedly over our own history. This current wave of immigration is not the first time that we have had a big wave of immigration

that causes turbulence, and then when you come out the other side, we're all better off. I mean, look, it happens that my ancestors came to this country in 1640. So we've been here forever. And we were doing just fine. And then the Dutch arrived. Now, don't get me started on the Dutch. It was really hard for us to get along with the Dutch. But then eventually we got along with the Dutch. And then we forgot they were Dutch. And then there was just us. And then the Germans arrived. And they were really difficult. We had a lot of trouble assimilating the Germans. And then after a while, we got adjusted to them and we sort of didn't even notice that the Germans were Germans. And then we invented at that point a term called Anglo-Saxon

to refer to the Dutch and the Germans and us. And then we had a lot of trouble when the Irish arrived. I hope you see that there's a smile on my face. We've done this a lot. Coming up next week on Freakonomics Radio, have you ever heard of the gambler's fallacy? So the gambler's fallacy is expecting that if you've had a streak of a couple of outcomes in a row, that the next outcome is just much more likely to go the other way. And that's just simply not true. Well, it turns out that gamblers aren't the only people

who get fooled by the gambler's fallacy. We look at Major League Baseball umpires. We look at loan officers. And we even look at federal court asylum judges. What do the umpires think of this? One of the biggest things you have to do when you're an umpire is be honest with yourself. That's next time on Freakonomics Radio.

Freakonomics Radio is produced by WNYC Studios and Dubner Productions. This episode was produced by Greg Rosalski. Our staff also includes Shelley Lewis, Christopher Wirth, Jay Cowett, Merritt Jacob, Noah Kernis, Allison Hockenberry, Emma Morgenstern, and Harry Huggins. You can subscribe to this podcast on iTunes or wherever you get your podcasts. And you should really come visit Freakonomics.com because there you will find our entire podcast archive, as well as a complete transcript of every episode we've ever made, along with the music credits and lots more. Thanks for listening.