Steven LEVITT: My guest today, Maya Shanker, is a great example of how with the right mix of talent, confidence, and doggedness, anything is possible.
Welcome to People I (Mostly) Admire, with Steve Levitt.
LEVITT: While only in her twenties, Maya built from scratch the Obama administration’s Social and Behavioral Sciences Team. That’s the U.S. government’s first systematic attempt to integrate behavioral economics into its policies. It’s a U.S. version of the highly successful British Nudge Unit, which was launched by Prime Minister David Cameron in 2010 and named after Richard Thaler and Cass Sunstein‘s hugely impactful book, Nudge. She also served as the first behavioral science advisor to the United Nations. And more recently as a global director of behavioral sciences at Google. Still only in her early thirties, Maya might be the youngest guest I’ve had so far on this podcast and I’m guessing she will bring a big dose of youthful enthusiasm.
* * *
LEVITT: I first became aware of you when you were put in charge of the social and behavioral sciences team in the Obama administration. How old were you when that team was created with you as the leader?
Maya SHANKAR: I think I was either 26 or 27. I had no public policy experience at all. I was an academic. I was doing a postdoc in cognitive neuroscience at Stanford at the time.
LEVITT: Had you ever had a real job?
SHANKAR: Never had a real job.
LEVITT: Okay, so your first job was running an agency in Washington?
SHANKAR: Yes it was. What happened is I pitched the White House on creating a new position for a behavioral scientist, and then I just made it my mission to create a team. Because I don’t think I was totally qualified for the gig.
LEVITT: So, you just wrote an email to whitehouse.gov and said, “My name is Maya, and I would like to be your advisor.” And they said, “Sounds great.” Or how did that work?
SHANKAR: I still remember this one moment where I was in the basement of an fMRI laboratory. It was probably my fourth or fifth hour scanning people’s brains. And this guy came in and within minutes I’m peering into this person’s brain. And I’m thinking to myself, given my personality, I feel like the order of operations is wrong here. I don’t know whether this person has children, what his job is, what his favorite ice cream flavor is. I quickly realized that maybe scanning people’s brains was not exactly the right choice for me and that I wanted to do something that was a bit more social, working on teams. And so I ended up chatting with my undergraduate advisor Laurie Santos. I’m sure you’re familiar with her.
SHANKAR: She has this great podcast called The Happiness Lab and she was my mentor starting from the time that I was a freshman. And I said, Laurie, “What do I do?” And she said, “There’s this amazing work that’s happening in the White House right now where they are using insights from behavioral economics to help enroll low-income students into the free lunch program.” The government offers this amazing program for kids to be able to eat at school, but millions of kids were still going hungry because the form was really burdensome and there was a stigma associated with signing your kid up for a public benefits program. The government used the power of defaults. And they basically used existing administrative data on these kids to automatically enroll them in the school lunch program. So now parents only had to take an active step if they wanted to unenroll their kids, not enroll them. And as a result of this policy change, over 12.5 million more kids were now eating lunch at school every day. And it was a light bulb moment for me. I thought, “Oh my gosh, I really want to do this kind of work. I want to be a practitioner of behavioral science.” But I had no connections in the political sphere. So I sent a cold email to Cass Sunstein. He is the co-author of Nudge. And he had served as the head of the Office of Information and Regulatory Affairs in the Obama White House. I just said, “I’m a postdoc, I’ve published nothing of significance. I have no public policy experience.” I even wrote in the email, Steve, that I didn’t think I was quote “cool enough” to work with the likes of Obama. “And so if there was a state or local government opportunity, I’d be totally game for that.” And thankfully, Cass ignored all of the insecurities and connected me with President Obama’s science advisor. And so a week later I was pitching my, soon would be boss, on the idea of building out a new role for a dedicated behavioral scientist in the White House.
LEVITT: So, eventually you have this job. It doesn’t sound like it comes with a very big budget or a lot of supporting staff.
SHANKAR: Actually it came with zero budget. There was no mandate, no resources of any kind. This felt like I was trying to build a startup in my parents’ basement, genuinely. I had to be as creative as I possibly could when it came to tactics. For example the only way that I was going to get government agencies to work with me on behavioral science pilots is if my colleagues could see the value of this work firsthand — it would be far more compelling if they saw that this was improving their programs and policies than if I just tried to convince them with beautiful prose and some sort of policy document. This involved basically knocking on every door that I could in government saying, “Hey, let’s try to align incentives here. What problems are you already trying to solve? And now let me brainstorm based on the tools in my toolbox as to how I can help you achieve those goals.” And one way that I tried to build a lot of excitement around these early behavioral pilots was to organize this meeting with a lot of luminaries from academia, so Danny Kahneman was there, and also luminaries within the federal government. I created an admissions ticket for the meeting. A one- or two-page proposal about a behavioral pilot that you would be able to implement within the next three months. And I would work with these folks on these proposals in advance of the meeting. It really generated a lot of enthusiasm. We probably got about 50 proposals in the door as a result — and of course everyone wanted to come. They’re like, “I want to meet Danny Kahneman. He’s a legend in the field.” And so that was a nice way of actually incentivizing people to take the risk of running a randomized controlled trial and in some cases, using behavioral science for the first time.
LEVITT: Let me go back though, because you’re a lone-wolf in the White House, you have no budget. And then the next thing I heard you had a conference with 50 different proposals from government agencies. There must’ve been something in between?
SHANKAR: I was just relentless. I left no stone unturned and I engaged with people at all levels of government. So, of course, I engaged with political appointees, but I really try to tap into the expertise of civil servants. People who had spent decades in the federal government and really understood the veteran’s experience, really understood the student loan borrower’s experience, really understood what it’s like to be a low-income student who’s struggling in school. And I try to gin up their enthusiasm for this work and also to learn, in return, exactly where the most promising opportunities were. And let me tell you, the denominator was massive. For every hundred conversations I had, it would lead to one potential collaboration.
LEVITT: That’s a pretty good ratio — better than I might’ve expected. And my own experience is that the lower you go in the organization, the better the information you get. I remember when I did management consulting right after college, my job was to try to get this drug approved at the F.D.A. And I was used to hanging out with the top people, the head of R&D and things like that. But what I learned is that if I got really into the guts of the organization, the people who were designing the studies and who analyzed the data, they knew all the answers. Somehow nobody ever asked them what they thought or what they knew. I found just giving them attention and respect was the secret to everything in making things happen.
SHANKAR: I completely agree. I felt like the civil servants were an untapped resource. Working very closely with people who had been in the government for decades and were planning to be in the government for future decades was a key to the team’s success because it would mean that every time a new project idea emerged these colleagues of mine were already bought in. They already felt ownership of the process. And our initial pilots would naturally evolve into sustained multi-year collaborations with government agencies.
LEVITT: In my mind, I’m seeing an army of civil servants lining up behind you, excited to change things, excited to do the hard work it would take to make the programs really great. But my own experience, both in government and in private sector, is that mostly people just want to do today what they did yesterday. And people like you are so annoying. When people like you come around with enthusiasm and really want to change things it just makes life hard. Did you run into a lot of that resistance?
SHANKAR: Oh my gosh. All the time. I remember sending an email to the Department of Education in 2014 proposing an idea and they wrote back saying, “Yeah, absolutely. We can try this in 2017 when our contract renews.” Okay. Oftentimes there’s very little upside for a government employee to take risks in their job. For example, maybe it’s revealed through our randomized control trial that their program isn’t as effective as they think. And like you said, it can be annoying to be the one saying, “Hey, try this new thing, try this new thing.” I would basically go to the annual budgets, figure out what agencies were already being tasked with or were already deciding they wanted to accomplish. And then working with secretaries, people at all levels of government to try to figure out how insights from our field could help them achieve those existing goals. That was the only workable model. If I had come in and introduced a new goal, a new outcome metric, no way in hell we would have gotten anything done. Had I come in and I was a top ranking official and I just issued a mandate, sure, we might’ve been effective in those four years. But I’m not sure it would have inspired the kind of lasting cultural change that we continue to see today in government agencies.
LEVITT: I think that’s a really wise insight because — I did a consulting firm with Danny Kahneman and others, and we generally came in at this top level. The C.E.O. would say, “Hey, I would love to hire Kahneman-Levitt to go do this.” But the middle-level managers would almost always fight us tooth and nail until they finally, just through a war of attrition, outlasted us. And so we rarely got anything done. The problem is, honestly, it is the most painful unpleasant task imaginable to go knock on hundreds of doors and try to build groundswell support for things. It’s a really scarce resource to find someone who’s both visionary and willing to do that kind of work. Maybe you have to be 26 to be willing to do those things.
SHANKAR: Absolutely. It was very hard work, absolutely demoralizing at times. When you get 80 no’s, you’re not sure actually that the 81st time you’re going to get a yes, you don’t have that hindsight telling you “Keep at it, Maya.” There were so many times where I thought “There’s no way this is going to be successful. I’m one person trying to get the entire federal government and Obama leadership bought into this mission.” But I will say, some of my most emotionally touching experiences certainly came from working with civil servants. And a good example of this is — I met a woman named Rosemary Williams who worked in the Department of Defense and she had worked in the department for years, and she was just on the cusp of retirement. And then a month into partnering with us she said, “I no longer plan to retire. You guys have ignited a fire in me that government can in fact be innovative.” And she started taking executive education classes. I mean, we saw that happen so many times where people were so excited about behavioral science and it led them to new career paths. And in another example of this, I remember working with a team at the Department of Veterans Affairs. We were trying to get veterans signing up for an employment and educational counseling benefit that they were eligible for after their years of service, but very few veterans were signing up. And so we were working with the V.A. They said, “Look, we basically have no budget, fine. We’re okay working with you, but you can only change this one email.” Instead of telling vets that they were eligible for the program, we simply reminded them that they had earned it through their years of service. So this is a play on the endowment effect. We might value things more if we own them or feel we have earned them. And that led to a 9-percent increase in access to the benefit. Now, that was a fun result, but actually the true victory was getting the V.A. to run the first A/B test it had ever run. They threw a pizza party when they got their early results and it was so exciting and energizing for this community to now have built the technological apparatus that would let them use randomized control trials in the future.
LEVITT: So, you probably don’t know this about me, but I am a skeptic when it comes to behavioral economics.
SHANKAR: Oh, I knew this about you, Steve. Don’t worry. And I still agreed to do this podcast. No, I’m just kidding.
LEVITT: For the most part, I think that the behavioral stories, they sound great and they often work well in contrived settings in lab experiments, but with a notable exception of choosing defaults, it seems to me that mostly the impacts of behavioral interventions turn out to be small.
SHANKAR: First of all, I absolutely agree with you. I think your skepticism is well-founded. It’s a humbling space to work in, the application of behavioral science, because fundamentally behavioral economics isn’t that powerful in a lot of contexts. So, our focus was working on lifting barriers for people who wanted to sign up for a program or take a specific action, but weren’t doing so because they didn’t have access to clear information or they were deterred by small frictions or features of the program.
LEVITT: When you went to Washington, were you as aware of the limitations in behavior science? Or did you learn along the way, how hard it was to make change?
SHANKAR: I think I was certainly aware of some limitations, certainly that a lot of the effect sizes might be small. Especially because in that first year, I was in quote, proof of concept mode. We had all these grand ambitions of serious program change. You’re actually changing incentives. You’re changing the structure of choices. You’re changing the default design of the policy. But that was all only going to come after I built a strong business case for why the team ought to exist in the first place. And all of that work felt much more in the domain of small tweaks. But of course, many of those effects are modest. That said, if the government were to adopt at scale many of these behavioral insights, and you aggregate all of those smaller results, that can be millions of people who now have access to a benefit who didn’t previously. And so the scale of the government is what buoyed my enthusiasm. Because even when you’d see a 3-percent increase in something you would also know that was 80,000 veterans, for example.
LEVITT: Yeah, the scale is a huge, powerful factor. I run a little center here at the University of Chicago called R.I.S.C. And what we’ve ended up doing over and over is partnering with usually really big firms, because even if we can make only a little impact, if we’re working with a firm that touches 50 million people we have a lot bigger impact than we would have if we tootled along on our own. So could you just give us examples of the big wins you had and what was the value of them?
SHANKAR: So, when I was first conceiving the team, the vision of the behavioral sciences team that I’d had was to help actually make the programs themselves easier to access or change the way the programs were administered or structured. So that instead of helping people navigate complicated choices, you actually just make the choices simpler or you offer better choices. One example of this was with the Department of Defense. The Department of Defense was really eager to drive military enrollment in the government’s retirement savings plan. But unlike their civilian counterparts, they were not automatically enrolled into these plans. You’d see roughly 87-percent enrollment among civilian government employees, but you’d only see about 42 percent among military service members. The ideal behavioral economics solution would be to just change the default setting. Now at the outset of our time, that was very difficult to achieve. You can imagine that pitching people on a substantial change in retirement policy for 1.3 million employees is a hard sell on day one. So we started with a pilot in which we prompted an active choice. When service members were changing bases, they would need to take a series of enrollment trainings or classes, alcohol and drug abuse counseling, learning about what their benefits were, et cetera. And we inserted into that enrollment process a yes or no choice about whether or not they’d like to sign up for retirement savings. And that was very effective.
LEVITT: How big was that impact right away?
SHANKAR: It was a 4-percentage point increase at the time.
LEVITT: That’s typical behavioral economics stuff.
LEVITT: Small but noticeable.
SHANKAR: Long after my tenure the government decided to automatically enroll all new recruits into the retirement savings plan, which was the ultimate big win. I think the small pilots can help inspire big cultural change or help build support for an argument that’s already underway.
LEVITT: So, what’s interesting about that to me is the story you just told transcended the Obama administration. And I would have suspected that every trace of anything you did got wiped clean from the government the day the Trump administration took over. It doesn’t sound like that’s true.
SHANKAR: No, that’s not true. My boss, Tom Khaleel. He had worked in the Clinton administration for eight years. He left for Bush and then he came back for Obama and he said to me, “When Bush came into office, it was as though I had spent the last eight years building this elaborate sand castle at the beach, and then one big wave came and crashed it — like basically destroyed the whole apparatus.” And so his advice to me was, “You need to make sure that whatever it is you do outlives your personal tenure at the White House. Bake this work into more non-partisan parts of government that will be committed to using behavioral science, even if there’s not a Democrat in the White House.” So Obama did sign an executive order in 2015, which institutionalized the team that I’d built and issued a mandate to government agencies to apply behavioral insights to policy. Now that executive order persisted during the Trump administration and continues to exist today. But I made sure not to build this team in the White House. And that’s because I knew it would be far more transient. So instead, I found a nonpartisan part of government called the General Services Administration and we actually built the team there. So that team actually persisted during Trump and continued to do great work with government agencies. They worked on like G.I. benefits and wildfire assessments and the opioid epidemic. But it is true that I disbanded the White House component on my way out the door.
LEVITT: So, how awful did it feel to you when Trump won? Here you were having the time of your life building something amazing and then it was just taken away from you. What was that like?
SHANKAR: Yeah, I was completely crushed and devastated. I had built this team thinking, if you move away from small tweaks and actually do system level changes, it can have a huge impact. But we had to build so much credibility before we could get to that point of influencing policy that I knew it would only come under a Hillary administration. I was drafting all these policy proposals of things that we would love to see happen and the campaign was so excited about this. And so to see those dreams crushed was really sad. I was actually more just reacting on a personal level, which is, I just felt absolutely heartbroken while also dealing with the fact that I was now going to quit my job and I needed to figure out what my next path would be.
LEVITT: I’m surprised the Biden administration hasn’t already reinstated a behavioral sciences team. Would you go back to D.C. if they offered you the job?
SHANKAR: I don’t think I would go back. One because I’m very happy with the work I’m doing right now. And I also think the team would benefit from fresh ideas and fresh blood and a new vision for what behavioral science could look like in the context of the Biden administration. Certainly all the documents are in place, but one thing I learned when I was in the government is executive orders don’t always self execute. Two years into my tenure, we’re briefing President Obama in the Oval and he ends up signing this executive order. And I remember very naively thinking, oh my gosh, this is it. Life is going to be so much easier after this change. But actually no. Certainly it was easier to now point to a document that had Barack Obama’s name on it versus Maya Shanker’s name on it. But did we still have to work just as hard to try and convince our agency partners to take on these pilots amidst so many competing demands and constraints and risk assessments? Absolutely. So, I think reinstating this team will require finding and making an active effort to recruit people whose sole job is to translate insights from behavioral science to policy. Otherwise, I just don’t know if enough activity will happen.
LEVITT: It’s interesting to me how differently the U.K. and the U.S. governments approach behavioral economics. In the U.K., it was a central part of what Prime Minister David Cameron was trying to do, politically. It was high profile, lots of staff and support. And in the U.S. you had to send an email, a cold email, to the White House to get hired. Are you surprised by that difference?
SHANKAR: I kept pointing to the successes of the U.K. Nudge Unit, but of course, every country is like, “Oh, but we’re different.” The behavioral insights team invited me to 10 Downing and to visit their group early on in my tenure at the White House. And I remember thinking, oh man, this is so lovely. Everyone seems like they’re on board with this. And then I came back to my home environment and I just felt like every barrier was sitting in front of me. There were so many times, Steve, where we would have worked on a project for eight months, 12 months, two years. And something would happen either politically or due to the budget or someone left. And the whole project would be canned. The failure rate was so high all the time. In order to build trust and transparency with the American people we made sure that in every policy report we shared all the pilots that worked, that were successful, and had positive results. But we also shared all the pilots we ran that had null results or negative results. Cause we wanted to make sure that people were aware of all the activity. I wish we’d also shared how many failed attempts we had to change things.
LEVITT: Forget about the ones that don’t work, but the 90 percent of the time and effort that goes into things that never even launch — that’s invisible to everyone. Maybe it would be counterproductive — my thought is it will be useful to show people how many failures there are. But I actually think maybe the discouragement effect would be so great that all the people like you, who are doing this inspirational work — if they knew how hard it would be in the beginning, I bet a lot of them wouldn’t even begin launching off in that direction. So maybe keep those failures secret.
SHANKAR: I think you’re right. I think — this can just be your and my little secret. Okay.
LEVITT: Behavioral economics has this element of manipulation to it. It’s trying to take advantage of people’s mistakes and inattention to get them to do what you, the behavioral scientist, wants. Do you have any uneasiness about doing that?
SHANKAR: I think there’s a few safeguards in place that prevent this work from having a deleterious impact. Number one is the one we already talked about, which is there are just limits on how effective any given nudge can be. The second is just the reminder that there is no default-less state in the world. So every program and policy has a default design that will influence people one way or the other. So, in the same way that, if you go to a restaurant, menu options are listed in a particular order. And we know from research that people are more likely to pick the first option they see from a set of options. Now, in the context of government, if you’re a veteran and you’re asked to fill out a burdensome application form that requires referencing 16 different tax documents, that’s a default too, and chances are that those requirements are nudging them away from accessing the benefits. So, it’s really important to remember what the status quo is and whether that’s also serving as an implicit nudge, but is actually counterproductive. It’s not achieving the program’s goals. And then I think the third thing is transparency. All the best nudges work just as effectively when they’re transparent. One of my favorite projects in the wild was around helping to curb the opioid epidemic. When a doctor’s prescribing opioids for the first time, they go into this online system where they can subscribe a fixed number of pills for that first prescription. And researchers experimented with changing the default number in the system. So rather than it being, for example, 30 days, they changed it down to 14 days. Now doctors can see that there’s this difference but it makes them a little bit more thoughtful about what that original prescription is and that led to a significant decrease in prescriptions across that particular healthcare company. When you are transparent with folks, it doesn’t backfire. In fact, I think it’s a very healthy part of the social contract we’re all in.
You’re listening to People I (Mostly) Admire with Steve Levitt, and his conversation with behavioral economist Maya Shankar. After this short break, they’ll return to talk about learning Chinese and Maya’s new podcast A Slight Change of Plans.
* * *
LEVITT: Today’s listener question comes from Elkin Cruz and Elkin writes, “Hi, Steve. Aren’t you afraid of survivorship bias? You keep asking successful people for advice when maybe they’re not even capable of comprehending what made them get where they are. What about all of those who did the same as them, but still, never succeeded?” So Elkin, that is a great question. Let me start by making sure everyone knows what we mean by survivorship bias. Survivorship bias occurs when you base conclusions off of a subset of the population who meet a particular set of criteria, assuming that the results will generalize more broadly. For example, let’s say you study successful startups and you find that 80 percent of them were founded by men under the age of 30. You might be tempted to conclude that young men are the most successful entrepreneurs. But what if you expand your sample to include not just successful startups, but also all the failed ones, and you find that 90 percent of failed startups were founded by men under 30? Then you’ll draw a very different conclusion that young men are less successful on average with startups, they just take a lot more shots at it. So back to Elkin’s point. I often ask my guests for advice. Is their advice any good? My guests are extraordinary people and while their approach worked for them, maybe if more regular people like you and me followed their exact same path, we would have failed miserably. That’d be a classic example of survivorship bias. But their advice might be awful for other reasons also. Maybe successful people don’t even understand why they’ve been successful or maybe their strategies were useful for the particular challenges they faced but aren’t broadly generalizable. Nonetheless, I really enjoy hearing their advice. I always try to imagine what advice each guest will give. And I’m almost always surprised by what they say. Like when Yul Kwon, the winner of Survivor, suggested that he and others should go live on an island for a while to reconnect with their gratitude for life. Or when Ken Jennings said to make sure not to neglect the things that make you weird. So more broadly, Elkin’s question got me thinking — maybe advice from people who haven’t achieved their full potential in life might be particularly valuable and interesting. We get advice from successful people all the time, but no one ever asks not so successful people for advice. So I’d like to try something. If you were listening to this and you feel like you were someone for whom things didn’t turn out nearly as well as they should have, your talent was squandered or unappreciated, then write me with your answer to the question: What advice would you give to a young person trying to find their place in the world? I will do my best to pull together the answers I receive and report back what I hear. So, send me your advice. The address is email@example.com. And now back to my conversation with Maya Shanker.
* * *
LEVITT: The answer that surprised me the most so far in talking with Maya is when she said she wouldn’t go back to Washington if invited by the Biden administration. It sounded like at the time she loved her government work more than anything. She only stopped because Trump won, she lost her position. But when forced into a change, now she doesn’t want to go back. Interestingly, the only other big change I know of that Maya has experienced in her life was when she was forced to abandon a promising career as a concert violinist, due to a hand injury. Despite her doctors advising her to quit, she kept playing for almost another year with just one good hand, before finally giving up at the age of 17. I wonder if she’s someone who’s the opposite of me, someone who never quits by choice. I’ll start the second half probing her about that. And also she’s got a new podcast called A Slight Change of Plans. It’s generating a huge buzz. I want to find out what that’s all about.
LEVITT: So, you were a violin prodigy as a child, a student at Julliard, practicing five hours at a time, and private lessons with the legendary Itzhak Perlman. But for an injury that you suffered, that likely would have been your life’s focus. Do you remember what motivated you as a 10-year-old or a 12-year-old to work so hard on something like that? Was it being on stage, or joy in perfection or, enormous amounts of positive feedback from adults?
SHANKAR: Oh, I think I got a lot of negative feedback from adults.
SHANKAR: If you’re studying at Julliard, your teachers are going to be very critical. I’d have about 10 hours of classes over the course of the day. And every teacher has the ability to tell you, “Ooh, this thing, and this thing, and this thing, and ah, you should’ve practiced this more,” whatever. But I think that was actually really useful because it means that I’m very open to being wrong in a lot of situations and updating my beliefs. Because you’re not going to get through that kind of intense environment if you can’t take a lot of criticism. Fundamentally, one of the things that I love engaging in are pursuits where your inputs feel like they really matter because they’re expressed in outputs. The more you practice, the better you become as a violinist. And that’s not true in every discipline. You can try your absolute hardest on this latest startup, but then all these market factors and exogenous factors play a role and you just don’t have control over the system. But it felt like I could see the translation of my hard work and see it manifested in better playing. And when you choose a pursuit like that, it can be endlessly satisfying because you’re not always concerned with the absolute quality of playing, you’re concerned with the delta — how much progress you’re seeing over time. And that’s certainly relevant to my life today. My husband is Chinese, his whole family in China doesn’t speak English. And so one of the activities I took up a few years ago was to try to learn Mandarin.
LEVITT: Good luck with that. I tried that too.
SHANKAR: It’s very hard. I am elementary at best. But it is so gratifying to see improvement over time. I understood a sentence my Chinese teacher told me today and I know that I wouldn’t have been able to do that two months ago. So my drive came from just the joy of getting to see myself be better at something.
LEVITT: So, I adopted two daughters from China, and so I tried to learn Mandarin. I tried really hard. I would say I spent a half an hour a day for a year or more trying to learn Chinese. One of my graduate students was tutoring me and my ear is so bad I actually went to one of those online sites where they would say two tones in Chinese for you. And then you’d have to say which of the tones you had just heard. And I was so bad at it that I decided to start collecting data and it turned out that I literally could have guessed. I was no better than chance.
SHANKAR: I love that.
LEVITT: And I found that very discouraging. That really hurt my progress. Have you tried with your Chinese relatives? Because after all that practice, I went to China and I neither understood a single word that anyone said to me or had a single word I said understood.
SHANKAR: Yeah. So I think in general, I suck at learning foreign languages. I just don’t pick things up effortlessly. But Chinese does actually play to my comparative advantage, which is I have a finely tuned ear from my violin days. I would say that my ability to speak is far outpacing my ability to comprehend. So I did go with my husband to China. It was our post-wedding trip. And all these relatives were meeting me for the first time and I was able to produce language and they were able to understand it. But one thing I was struck by were all the regional differences in accents when speaking the language. So, I felt completely demoralized after the trip.
LEVITT: After I adopted my first daughter from China, I literally did not practice Chinese for one minute for the next three or four years until I adopted my second daughter. And on the plane to China, I just memorized and practiced really hard four sentences. I am American. This is my daughter. I adopted her from the orphanage. And then I had my father with me and then — this is her grandfather. And I really worked to pronounce them well. So, every conversation I had in China, I started with my first sentence and no matter what anybody said to me, I would say my second sentence. And then I would say my third and then my fourth. And it was phenomenal. It worked unbelievably well. Every conversation was a huge success. People were like, calling their friends over. And the funny thing was the other Americans who were with me, watched this happen and watched every conversation be so vibrant. They’re like, I cannot believe how much Chinese you know. And I never told them that I literally said the same four things in every conversation.
SHANKAR: I do find, actually, I mislead people often because my pronunciation is relatively good because of the musical training. I’ll say a few things and then they assume that I must have some degree of proficiency or fluency and I’m like, “No, there’s a cap on how much I know here.” But I live in the Bay Area, so there’s a lot of Chinese-speaking folks here. There’s been a few times where I’ve been walking on the street and a person who might not speak English is lost. They’re looking for directions and they see me and obviously you’re thinking this person can be of no help to me. So they’re trying to sign to me — how do I get from here to there? And I suddenly start speaking in Chinese and this beautiful human connection forms almost instantly. And you feel like you’re bonding with people, literally having a conversation that you could not have had in the counterfactual world had you not tried to pick up the language. And those interactions have definitely touched me on a deep level.
LEVITT: So, in the White House and at Google, your job was to change other people’s behavior. And in your new podcast, A Slight Change of Plans, it’s all about how people respond to life changes. Danny Kahneman always told me that people study things that they themselves personally are really bad at. So is Danny right? Are you bad at change?
SHANKAR: I think I am bad at change. That was probably my big motivation for launching A Slight Change of Plans. In 2020, I was feeling quite overwhelmed by the rapid pace of change around me. I think many people were feeling that and — that you just don’t have control. I thought, “Okay, maybe the specifics of what 2020 is throwing our way are unprecedented, but the human ability to navigate change is not. And it’s possible that we recruit a very similar type of psychology when we confront various types of change. And so if we can mine people’s stories, we can potentially learn some very valuable things that we can use in our own life.”
LEVITT: So, what kinds of things do you think we can learn from change?
SHANKAR: I think in my own life, obviously I had a very privileged form of change, which was simply that I was this budding concert violinist and I couldn’t play the violin anymore. But it did lead me as a 15-year-old to realize — cause you don’t really think about these concepts when you’re a kid necessarily, at least in my case, I just let my life unfold. And I think it taught me this control thing is a bit of an illusion. I am the personality type that loves having control, but it takes a while to fully appreciate that. You have to have a few more experiences with change and unexpected change and unwanted change, I think for it to really drill in. There’s this concept in psychology called identity foreclosure. It basically refers to the idea that we can settle into a self identity early on and close ourselves off to other alternatives. To your point earlier, I think this is one of the reasons why we as people have a really hard time quitting things and knowing when it’s time to move on to the next chapter. I think the fact that I lost the violin at an early age forced me, as a child, to think of my identity as more malleable. That served me well as I’ve gotten older, because it means that I attach my identity less to the thing that I’m doing at that particular moment of time and instead attach it to a broader cluster of personality traits and the friends that I have, and the activities that I enjoy in life, and what my mindset is, or my beliefs about things. I don’t know if you’ve had this experience too, where you figure out — what does it really mean for me to have a certain identity as a person? And maybe I am just the aggregate of all the things that I think and do.
LEVITT: I think it’s such a risky strategy, to put all your identity eggs in one basket. A good example — so my son was a world-class fencer as a high school student. And probably I had as much identity wrapped around that as he did. It was so much fun. We shared so much time with it and then he hurt his knee. And even though it was fencing and he was 14-years-old at the time and who cared, it was a devastating loss to me. I just imagine what it’s like for the parents of a professional basketball player or a college football player to suffer such a devastating injury. And it was interesting cause I hadn’t really even understood that about myself. How much enjoyment and identity I really got from being, not even the world-class fencer, but the father of a world-class fencer, which is like, incredibly pathetic. But it was eye opening.
SHANKAR: Thanks for sharing that example, because it does remind me of a conversation I had for my new podcast with Tommy Caldwell. He’s considered one of the greatest big-wall climbers in the world. He had been taken captive in Kyrgyzstan for six days and nearly died of hypothermia and starvation. He ended up pushing his captor off of a cliff and he had to reckon with the fact that he had done this and he surprised himself. He said he’s a meek, softer type and he doesn’t believe in killing people. I was just so confident that was going to be the crux of his change story and that that would have informed his climbing identity in the future. Actually, the big change moment for him was that he said his body entered survival mode when he was in the depths of this hypothermia starvation and that he experienced the greatest mental clarity and focus he had ever had in his life. And so since that day, the true motivator for him has been chasing this mental state. And he once even tried to starve himself on a climb to recreate those conditions. But it’s propelled him to become the world’s greatest big-wall climber. And it has been the thing that he’s been searching for and seeking over the course of his life, even outside of the climbing domain. After I had this interview, I thought to myself, Tommy Caldwell is the type of guy who seeks genuine flow states. Climbing is the secondary thing in his life. That’s just as a means to an end. But that’s actually the thing that’s very core to his identity and that he can achieve in a bunch of different disciplines or domains. Like you said, it can be risky to attach your identity to a sole pursuit. And if you identify features of the activity that light you up. I mentioned earlier in the interview that I love the idea of seeing progress — seeing that delta between how I sounded on the violin yesterday and how I sound today. And if you can identify those features, then you can find them in many spaces in your life. And that I think can lead to a sturdier identity.
LEVITT: Have you tried playing violin again since your injury? Or did you just permanently retire?
SHANKAR: I haven’t played the violin for what feels like ages at this point. And then my producer said, “Hey, Maya. What if you played the violin for the soundtrack of the podcast?” I was thinking, “Oh my gosh, I don’t know if I can do this.” But I called my parents, they were going to be flying out here to see us after they got their vaccines, and I said, “Hey, Pops, do you mind just bringing my violin with you?” And I opened it up and it was amazing to me. It is a little bit like riding a bike. I lost so much of my technique, but the fundamentals were still there and I was able to record original music for the soundtrack. So, if you hear the first episode of my show, it’s a conversation with Michael Lewis in which I give a preview of what’s to come in the season of A Slight Change of Plans. You’ll actually hear my violin playing at the end of the episode. And it was incredibly fun to do.
LEVITT: So, I have a theory about leading a happy life. And some of Maya’s answers I think were consistent with my theory, which is that the single best predictor of who will be most unhappy in life versus those who will be happiest is that the unhappy people need to feel control over things. Whereas the happy people accept the fact that much of life is beyond their control. Now, admittedly I don’t have much evidence in support of this theory. It was initially based on casual observation of the people in my life, some of whom crave control and others who don’t seem to. A second piece of evidence, again, not particularly convincing, is that over the last 20 years, I’ve intentionally and pretty successfully transformed myself from being someone who craves control to someone who accepts the impossibility of having control. And I’m much happier now than I was back then. Of course, a thousand other things have changed in my life over the last 20 years so who really knows. Have others followed my path and let go of the need for control? If so, are you happier now? Has anyone gone the other direction? Not needing control when you were young, but having the desire to control everything grow with age. Is there someone out there who absolutely needs to control everything, but does so with a great sense of joy? I’d be really curious to hear your personal experiences on the subject. If you feel like sharing, you know how to reach me. It’s firstname.lastname@example.org. I read every email you send. Thanks for listening and take care.
* * *
People I (Mostly) Admire is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and The Freakonomics Radio Book Club. This show is produced by Freakonomics Radio and Stitcher. Morgan Levey is our producer. Dan Dzula and Greg Rippin were the engineers. Our staff also includes Alison Craiglow, Joel Meyer, Tricia Bobeda, Zack Lapinski, Mary Diduch, Brent Katz, Rebecca Lee Douglas, Emma Tyrrell, Lyric Bowditch, Jasmin Klinger, and Jacob Clemente. All of the music you heard on this show was composed by Luis Guerra. To listen ad-free, subscribe to Stitcher Premium. Thanks for listening.
SHANKAR: Hi. Niihau. I’m Maya and I speak a little Chinese and today I’m excited. I’m very excited because I, with Steve, get to chat.