share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
SPEAKER 1: Good afternoon. I'm a real loudmouth, so I'm just going to [INAUDIBLE]. Hi, my name's Penny [INAUDIBLE]. It's a pleasure to meet you all here today for our program on, let's see, Insights from Behavioral and Development Economics. So if you're here for something else, please step out now.
It's my pleasure to introduce our moderator, George Boyin. George Boyin's a professor of economics and industrial and labor relations, has been at Cornell since 1982. He is an economic historian. His research examines social welfare policy, labor markets, and globalization in the 19th and early 20th centuries. He currently is working on a book manuscript entitled, The Winding Road to the Welfare State-- Economic Insecurity, Social Welfare Policy, and Living Standards in Great Britain 1840 to 1940. Please join me in welcoming Mr. Boyin, who will introduce our other speakers.
[APPLAUSE]
GEORGE BOYIN: Thank you. I'm going to begin just by saying a couple of words about the Economics department, and then I'll introduce our speakers. The Economics department is, in a sense, a new department that was formed two years ago by merging two former economics departments on campus, the Department of Economics in the College of Arts and Sciences that Ted and Jim were from, and the Department of Labor Economics within the School of Industrial and Labor Relations that I come from. It's now one department, making decisions as a single group that reports to two deans.
Our faculty is now about 50 members. We have about 750 undergraduates, over 150 PhD students. And we service more than 5,500 students each year.
The rationale behind the merger was to create an institutional structure that would permit a dramatic increase in the quality and visibility of Cornell Economics. Our goal is to make the new department a top 10 Economics department and to provide the best undergraduate education in the Ivy League. Toward that end, the department currently is engaged in an ambitious fundraising campaign, with great support from the University. Thank you, those of you who are here.
Our Chair, Kevin Hallock, has been on endless road trips this year. He's always interested in talking with alums about the future of the department and, unfortunately, about the Red Sox.
[LAUGHTER]
He's traveled so much, that he has actually earned-- on frequent flyer miles, he's earned a trip to the moon.
[LAUGHTER]
Unfortunately, he still has to keep traveling, so we can get him back.
[LAUGHTER]
Frequent flyer things are weird. [INAUDIBLE]. We also have begun a massive faculty recruiting effort that we expect to last for several years. And we've already had some great successes. And we have the most significant effort to reviewing our undergraduate program and identifying ways to improve.
In March of 2013, recently, we announced the new-and-improved undergraduate Economics major that will replace our existing program. The new curriculum will first apply to incoming freshmen this fall. A number of additional improvements are on our agenda for the upcoming year, including building a new undergraduate lounge and study space, hiring a new student experience coordinator, who will help with advising, research experience, and the like.
And Ted O'Donoghue, who has been leading this effort, is one of our speakers today. If you'd like to hear more about it, I'm sure he'd be willing to talk to you after the session. But enough about the department. Today we're here to learn about some of the research being done by two of our faculty members.
I get to introduce both Ted and Jim. And then they'll speak in that order. You can tell, by the way, how long we've been here by the color of our hair.
[LAUGHTER]
I'm not saying that you, as undergraduates, did this to me, but--
[LAUGHTER]
--it's possible. Ted O'Donoghue is a professor in the Department of Economics who has been a Cornell since 1997. His specialty is behavioral economics, and he is one of the leaders of the generation of behavioral economists who helped bring the research into mainstream economics in the late 1990s and early 2000s. He is a member of the Russell Sage Foundation's Behavioral Economics Roundtable.
He was the team leader for the Cornell Institute for Social Sciences theme project entitled Judgment Decision Making and Social Behavior. He co-founded and co-organizes the top annual meeting of behavioral economists in the country-- and maybe in the world. Starting on July 1st, he'll be the Associate Chair of the Department of Economics.
James Berry is an assistant professor in the Department of Economics. His research addresses questions of development economics, often touching on themes from behavioral and labor economics. He studies poverty in Latin America, sub-Saharan Africa, and South Asia.
He is currently conducting several field experiments to study household subjective value of education and household decisions to purchase and utilize health products and educational services. So Ted will speak first and then Jim.
[APPLAUSE]
TED O'DONOGHUE: Thanks, George. And thanks to all of you for coming. Very excited to see so many people interested in this topic. I actually teach behavioral economics in this room frequently, never seen this many people in my class.
[LAUGHTER]
So the plan is that I'm going to talk for about 15 to 20 minutes about behavioral economics. And then Jim will talk for about 15 to 20 minutes about development economics. And then we'll have lots of time for questions at the end, hopefully.
So behavioral economics and development economics are two prominent fields in modern economics, also two fields where Cornell's played a major role in the history of these fields. I think one of the things you'll get a sense, as you hear me and then Jim talk, is these are two fields that really ought to be talking to each other a lot. And it's only recently have they been starting to talk to each other. One of the things that Jim and I are hoping to do in the years to come is to really use our joint expertise at Cornell to bring these fields together.
But our talks are going to focus a bit more on what each of us have been doing. So let me talk about behavioral economics and let me just start with just giving you a sense of what behavioral economics is, what it's all about. So this is from the description of the program. Behavioral economics uses insights from psychology to better understand and improve economic outcomes.
So I want to parse this for you. And actually the way I want to get there is actually, first, just being clear on what is the science of economics? How do we think about economics? So two main pieces that I want to put forward in terms of the science of economics, the first is just to be clear that economists have two separate goals.
And we always try to be very clear which goal we're thinking about when we talk. So one goal is to try to understand the economic outcomes that we observe in the world. A second goal is once we understand, then we can try to design policies to improve the economic outcomes that we observe in the world.
So much of what we do is the first. But in some sense, our big picture goal is the second. In terms of the first goal, what's the methodology that we use in economics? Well, it's a big back and forth between observing data and then developing theories and observing data and developing theories.
So we start by observing some economic outcomes. Then we try to develop a theory from what might explain those economic outcomes. Once we've done that, we can't go back to the original data because showing the theory explains the original data just says we're good modelers, doesn't say anything more than that.
So [? what ?] we need to do is to develop additional or ancillary predictions of our model and then go to the data to test those ancillary predictions. Inevitably, we go back to the data. We identify flaws in the initial theory. We then develop a new modified theory, or perhaps we jettison the first theory and come up with a completely new theory that performs better. Then we go back to the data, so on and so forth. And that's sort of the science of how economics proceeds.
So why am I bringing this up here? Well, behavioral economics is really just doing economics. So it's going to use the exact same goals, exact same methodology, with one new wrinkle. Traditionally in economics, there's been a set of assumptions that had to be part of any theory you write down.
And in some sense, what behavioral economics is doing is saying, look, in fact, we can relax those. We can play with those assumptions as well as we develop our theories. So to give you a sense-- so now moving to the science of behavioral economics-- here are some standard assumptions that traditionally had to be part of any theory you write down.
So perhaps the most famous that you all know is the assumption of pure self-interest, that people care only about themselves. But there's a bunch of others. So another standard assumption is that people treat gains and losses symmetrically. Winning $100, losing $100 feels roughly the same in terms of magnitude.
People behave exactly as they planned. So economic models say you sit down. You form a plan for be, might be a conditional plan, depending. You could have a plan to behave differently if it's sunny versus rainy. But whatever your plan is, you carry it out.
People understand probabilities perfectly.
[LAUGHTER]
Lots of problems there. And finally, assumption that people behave in their own best interests. Whatever people do has to be in their own best interests.
The starting point for behavioral economics is to say, in fact, if you go to evidence from psychology and then some evidence from economists that came in later that built on that evidence, it casts doubt on all of those standard assumptions, as well as additional standard assumptions. Now, I want to highlight here this starting point. This is really the easy part.
It's sort of easy to show that those standard assumptions aren't quite right. Amazingly, economists were incredibly resistant to this, so it actually took people like Dick Thaler, Bob Frank, both here at Cornell in the 1980s, taking 10 or 15 years of going around and generating anecdotes or little bits of evidence to convince economists that those assumptions perhaps we should relax. Finally, by the 1990s, economists were convinced.
And then the hard work began. The hard work, getting to the more the science of behavioral economics, where sort of my career started, is now once we have this evidence from psychology, now we want to go back to this methodology of economics. We want to try to build better models based on that evidence from psychology, that is, models that help us better understand economic outcomes and models that help us improve economic outcomes.
So to illustrate this, let me give an example of some work from my own research history. And this is a phenomena that's known as present bias, the behavioral economics literature. So what do we mean by present bias? So evidence from psychology suggests that people have a sort of bias for now. They always, relative to thinking about their lives, they put a little extra weight on now relative to the rest of their lives. And of course now is changing because tomorrow will be now.
So let me give you this sort of phenomena. Present bias, even if it's relatively small, can have big implications. So let me illustrate this with a simple example. So suppose you receive an email. And this email says, click this link to take a 15-minute survey. And if you complete this survey, we'll send you a check for $100. So if you complete the survey today, we'll send you a check for $100.
Seems like a good thing to be doing. If that were the full stop, probably a lot of people would do this. But suppose the email goes on to say that if you complete the survey tomorrow, we'll send you $95. And if you complete the survey in two days, we'll send you $90, and so on and so forth.
So just to sort of set this up as an example, suppose you feel it's well worth 15 minutes to earn $100, spending 15 minutes on a survey to earn $100. Indeed, you might feel it's well worth spending 15 minutes to earn even $40. So this is something you plan to do.
But it could also easily be that today you decide the best plan is you don't want to do that 15 minutes right now. You'd rather do it in the future. So the best plan is I'm not going to do it today. I'm going to do it tomorrow.
And yeah, I get $5 less, but that's worth it. So I don't have to do this sort of pain of 15 minutes today. So today I wait, planning to do the survey tomorrow.
Now the problem arrives because tomorrow, now is now. So tomorrow arrives. You change your mind. Now you decide you don't want to spend the 15 minutes now. You'd rather spend it tomorrow, a $5 cost of doing that. Again, you decide it's OK. So you wait again, planning to do it the second day.
And hopefully you can see what's going to happen here. Then the next day arrives. Now that day's now. So you change your mind again. And this can continue until you end up never completing the survey.
So clearly a stylized example. You can imagine, lots of other stuff that might come in. But it's just meant to illustrate this basic idea of present bias and the types of problems that can emerge.
So just to be clear about this, the standard model, again, would assume-- the standard economic model would assume that people have to behave as planned. So if I go back to this sort of suppose you feel it's well worth-- sorry, this line here, where today you feel the best plan is to complete the survey tomorrow, there's nothing wrong with that, according to the standard model. But the standard model would say, when tomorrow arrives, you have to carry out that plan.
This evidence of predict present bias, though, suggests no, maybe people don't do that. Maybe they change their mind again and again and again. So this is a simple example, actually related to some stuff I did sort of 15 years ago. This idea has been applied, by myself and a lot of other behavioral economists, to a lot of economic applications, this sort of incremental procrastination type intuition.
Just to give you some examples, so one example, where some very nice work has been done, is procrastination and going to the gym. So that is people sort of planning to go to the gym. You're sitting on the couch. Or you sort of plan to go to the gym tomorrow. But when tomorrow arrives, you're sitting on the couch. You say, eh, I'd rather go the next day.
In fact, this research also has nice evidence of procrastination in canceling memberships. So they have a panel data set of people, and what they see is a big chunk of people who eventually stop using the gym and eventually cancel. And in that group of people that eventually stop using the gym and canceling, there's actually 2.29 months of continuing to pay their gym membership without ever using the gym before canceling.
Another nice example, not exactly procrastination but the same sort of incremental logic, is addiction. Standard economic model is behaving as planned. What this says is people that end up with a cigarette addiction, for instance, according to standard model, that would have to mean they decided I'm going to be a cigarette addict, or at least probabilistically decided I'm going to start smoking, and there's a 50% chance I'm going to be a cigarette addict. You have to plan on that.
This sort of incremental present bias intuition, though, suggests an alternative story, which is when people are young or with friends, maybe they decide they're going to smoke cigarettes for the next week or the next month. But then as time passes, you have these sort of incremental decisions to smoke a little bit longer or to smoke a little bit more per day or per week. And then they just end up with an unplanned addiction. It seems, in some sense, like a more natural explanation for many addictions.
Another example I'll say less about, but there's evidence out there of procrastination and job search, that people sort of taking time to find a job not because they can't find a job, but because they sort of don't search enough from day to day. And then perhaps the sort of most important example where this has been applied very successfully is procrastination and saving behavior. So that is people want to save more. They plan to save more. But they keep putting off starting that saving. And indeed, some very nice research by Dick Thaler and Shlomo Benartzi where they actually generate a plan that many firms have implemented to help overcome this problem that they trademarked, called the Save More Tomorrow Plan.
And essentially what they did is they offered employees the option to start increasing their savings via automatic deductions starting in a month or two. And indeed, lots of employees took that up, and it was very successful. So again, examples of successfully applying these ideas.
So let me move on to talk about a little bit more research, a slightly different flavor. And how I want to get there is I want to just highlight that within behavioral economics, there are two types of research, two types of approaches to research. One type of approach, which the research I've discussed to this point sort of falls into, is you take an idea from psychology, like this idea of present bias, you develop its predictions and various economic applications, and then you go look for evidence. So take present bias, develop the implications for addiction, go look for evidence. Take present bias, develop the implications for savings, go look for evidence.
An alternative approach that sometimes use this to go sort in the opposite direction. That is to go out in the world, look for economic outcomes that seem odd from the perspective of those standard economic assumptions that I said before, behavior that's hard to explain with the sort of approach we usually use in economics, and then try to investigate whether insights from psychology can help us understand those behaviors. So let me give you another type of research that sort of illustrates this.
This is more preliminary research, something I'm working on right now with a former Cornell grad student that's actually in the context of development. So this will make a nice transition to Jim. So actually what this research is about is bulk discounts in Tanzania.
So in fact, looking at the data, our sort of first step was seeing in Tanzania, households face significant bulk discounts. But I want to be clear here. This is not the type of bulk discounts that you think about in the US. So this isn't sort of going to Costco and loading up with toilet paper, things like this and whatnot. This is very small bulk discounts.
So to give a few examples here, this is something like choosing to buy one onion for 50 Tanzanian shillings versus 4 onions for 100 Tanzanian shillings, sort of very small quantities. Buying a little ladle, which is about 3 tablespoons of cooking oil, often put in a little plastic baggie to take it home, versus buy a 1/2 liter for 1,000 Tanzanian shillings. If you're not good at this translation, I did it online.
1/2 a liter is about 33 tablespoons. So about 11 of these will get you up to 1/2 liter. A similar thing for kerosene, buying the little ladle of kerosene versus a 1/2 liter of kerosene, or for charcoal, buying-- I'm not quite sure how it ends up being 0.36, but buying this small quantity for 200 shillings versus four times that for 300 shillings. These are actually examples that we're seeing for some households facing.
So roughly speaking, these are sort of anecdotal. There's variation in the data. But roughly speaking, these are decisions of purchase every day for today's usage versus purchase every week for the week's usage. So the next step we did after identifying this stuff is we actually went out-- and this is where we are right now-- and try to quantify to what extent are households taking advantage of these bulk discounts? And if they're not, how much are they losing out in terms of money?
And in fact, they're losing out. Households are not taking advantage of it, for the large part. They're losing out quite a lot. So in fact, on average, households are losing something like 20% to 40% of money, or spending 20% to 40% more than they need to for these goods. And this is big.
So we've done a very preliminary calculation, sort of back of the envelope. Before the households in our data set, 8% those households, if they just rearranged their consumption purchases, would move from below the poverty line to above the poverty line. So this is like big stuff, and yet they're not doing it.
And so the question comes up, why? And this is where we are now, at this sort of investigation stage. So we're sort of going through. We're laying out a list of possible things that might explain this.
There's sort of two standard stories that you could think about. Maybe this has something to do with transportation costs-- it's heavy to carry stuff home-- or storage costs at home. Maybe it something to do with cash or credit constraints, that you don't have enough money to buy at the weekly level. You need to buy every day.
Our initial best guess, but we still need to do more work, is that these are not going to play a major role here because, again, these are very small quantities and very frequent purchases. And when we typically talk about these things, we talk about larger quantities and less frequent purchases. So we're also trying to explore possible behavioral explanations, things like inattention or not really understanding just how much they could save if they bought things in different patterns, or maybe that is just requires too much effort to manage your inventories if you try to buy at the weekly level.
And then perhaps the most interesting, and again, we're trying to see to what extent they might play, one is present bias. Maybe you're afraid that if you buy at the weekly level, you'll bring the stuff home, and you'll consume it in two or three days, and you end up consuming a lot more than you intended to. Or then a second thing, which is consuming more quickly than intended for but for a different reason, maybe you're afraid if you bring home the weekly amount, and the neighbors see it sitting on the shelf, the neighbors are going to be knocking on the door quite frequently asking if they can borrow something.
So as I said, this is preliminary, no clear answers yet. We're working on it. Check back in a year from now. So on that note, let me turn it over to Jim.
[APPLAUSE]
JAMES BERRY: All right. Great. I'm real excited to be here. I'm going to be talking about development economics, and specifically the type of work that I do, to kind of just frame a little bit what development economics is. And to kind of basically take what Ted's definition generally of what economics is, it's to understand the economic conditions and behavior of poor households and firms in developing countries.
So we just focus on the set of developing countries, which is actually most of the countries in the world, and specifically we want to use economic analysis to inform policy to alleviate poverty. I'm going to take a slightly modified version of the way Ted defined economics. And there's actually some feedbacks, and especially in development economics. And we do a lot of policy evaluation, and we do policy evaluation to understand behavior and evaluate policy. And I'll show you a little bit more about what I mean by that in a minute with an example of some work that I'm doing.
So just to-- oops. Uh-oh. Well, is it frozen? So just to kind of drill down, and the kind of work that I do is to really take some specific issues--
SPEAKER 5: I'll get rid of that.
JAMES BERRY: --take specific issues in developing countries and try to understand them from an economic perspective and to kind of inform policy and evaluate different policies to help with these certain issues. So one of the things I work on quite a bit is education. So we know, just looking at the data, that households in developing countries have worse education outcomes.
So you have a great deal of a drop out after the primary level of education. It's a big issue of trying to keep kids in school after basically primary school. And in terms of learning, think about kind of teaching basic reading and math skills to kids, is a big challenge in developing countries. We know this.
But how do households really make education decisions? And so how do households, given the resources that they have, actually make decisions to educate their kids or not is something that we actually don't know a ton. And so one of the things I'm working on is trying to understand whether the households undervalue education, whether the actual benefits education are actually perceived by households. Maybe they lack access to quality education.
And if that's true, how can they be encouraged to motivate their children to learn? Or if there's some kind of economic failure there, how do we fix that? And then we have to educate the household side, but there's also the provider side of understanding education systems and how education systems might contribute to outcomes.
So along with poor education outcomes, we have, obviously, very lousy systems in a lot of countries. And so we want to understand how the inputs are insufficient. So what are the inputs that actually produce a quality education in developing countries, and how can they be improved? OK, now it works again.
So you can basically go through the same logic for health. I also work on some stuff in health. And so again, we know in developing countries, households, in general, have worse health outcomes, so in terms of things like life expectancy, child mortality, general morbidity, some maternal mortality. So those are all lower obviously in developing countries.
But understanding the economics and understanding how households make health decisions is something that, again, we don't know enough about. So is it just that-- first we want to just know how much they demand goods and services. Is it just lack of access, and they would actually buy things if it was provided to them or if they had access to them?
Do they lack information about things? They know that their health is poor, but they have no idea how to improve it. Do they lack the means to pay? Is it just a virtue of being poor? And so all these things kind of suggest different policies.
And then, again, from the provider side, we also have poor health systems in developing countries. So there might be certain types of access that just aren't there. So what are the inputs and the type of access that will actually provide a health system that's sort of functional and useful to people? And how can these things be improved from where they are?
So this is very broad, right? So this is just kind of how do we think about just generally these kinds of issues. And obviously there are many, many more that development economists think about, so just drilling down. I'm not going to show you the picture yet.
So one of the issues in education is that you have poor education outcomes among kids. But the household, if you look at that sort of education inputs that are being provided by the household, those are also very poor. So households don't have books. I'm thinking, in particular, about India because I did a study in India, which I'm going to tell you about in a second.
So the households don't have many books. They don't have other kinds of education assets. The parents are very poorly educated. So a lot of areas where the median level of education among parents is zero, so parents have never been to school.
So one of the policies that people have suggested here is to educate the parents. And so one of the policies that I've just finished evaluating is an adult literacy evaluation for mothers. So one of the nice things about being a development economist is I get to go to developing countries and take pictures and then show them to people. So this is a picture of a women's literacy class in India.
And so we evaluated this program, one, to see if it's a useful program in terms of educating women. But we want to use it to kind of understand how we can influence the household education production, so that how can we influence the education of the kids. So the questions that we were asking is, do these women's literacy classes teach women? That's the kind of the main policy question.
Does it empower women, which is also a kind of a big issue in development? Does it encourage child learning? And does it improve the home learning environment, for example, by encouraging mothers to be more involved with child learning? Or maybe you get more education materials in the home, et cetera.
So what we did, and a lot of the kind of methods that I use in my research is conduct a field experiment. So we assigned villages to receive mother literacy classes or not. So half the villages in our study received these classes, and the other half didn't. It was randomly assigned. So we compare the two groups, and that gives us the causal effect of these classes.
So what's the answer? Well, all these things were actually affected by the classes. They were affected not a huge amount, but a statistically detectable amount. All these things show improvements.
One of the reasons why the effects are small is because it's extremely challenging to get women to attend these classes. And so what we're kind of showing is that if you are able to get these classes to work, like to get women to come, then you can sort of open up a whole bunch of possible inputs that might go into child education within the household. Exactly how this works is something that we're working on. And we're kind of trying to use our experiment to tease out all these mechanisms and understand a little bit more about how households make education decisions.
So here's another study. This is more way more in progress, something I just started. So think about private schooling. So public schooling in developing countries, it's usually free. It's usually kids have a lot of access to it. There's been a lot of money spent to give kids access, especially at the primary level, to schooling.
But it's usually very poor quality. So as a reaction to that, a lot of households in developing countries have turned to the private market. And there are basically two channels for this. One is going to private school, very cheap private schools, so not like a real fancy private school, but such as like a $3- or $4-a-month level of private school. And the other is to go to classes after school, and pay for [? that. ?]
So in this research, and the thing that we're trying to pick apart, is that nobody really understands this market. This is something that's come up in the last maybe five years or so. It's something we don't really understand how households, given that they're relatively poor in means, and all the policy energy has gone into getting these households access to public schools, but households are still going and purchasing education for their kids anyway, even though they're very poor.
So we want to know first, very simply, what types of households demand these classes. Are they just more wealthy than the other households around? Or is it something else? And another is, it's kind of more interesting, I think, question that kind of starts to touch on behavioral economics, that we see in the organization that we're working with, as they sort of started raising their fees, they've noticed that kids start attending the classes more regularly.
So why is that? That could just be that they are attracting different types of kids, so households that are kind of more committed to educating their kids. It might be that the price of these classes is a signal of quality, right? So higher price means that you think that it's a more valuable service that you're getting.
Or something I find most interesting, might be kind of a commitment device. You might actually pay for the class. Once you've paid for the class, that induces you to go afterwards. So you don't want to-- if it was free, then you kind of don't care. But if you've paid for it, it makes you think that you want to continue going, even though it might be something that you wouldn't normally do.
And in economics, that's kind of called-- or in the behavior literature, that's called a sunk cost effect. So after paying for the class, the cost is sunk. So it shouldn't affect your decision to send your kid, but it might.
So these are all things that we're looking at in terms of attendance. And we want to also know whether parents actually know whether their kids are going to benefit, and that causes them to pay more. So the classes are supposed to help their kids in school. Hopefully they do.
And if they do, is it that the parents can kind of perceive that, and are they willing to pay more for the classes? So these are all things that we have basically zero answers to right now but I'm kind of starting to work on. And that's it.
[APPLAUSE]
GEORGE BOYIN: OK, we have about 20 minutes for questions. Yes, ma'am?
AUDIENCE: Thank you. First of all, I have one comment. And what [INAUDIBLE] Dr. Berry said about paying for classes and being more committed is sort of the opposite of what Dr. O'Donnell said with gym membership. So that just came to mind. The other thing, basically my question, in the United States, we have similar situations with low-income families as far as economics and health care.
I'm a retired teacher in alternative ed. And unfortunately, it's a very severe problem in the US. Is anyone in the department addressing these same issues in the US?
JAMES BERRY: So I think for the first question, I was actually thinking about that issue a little bit as I was kind of making these slides. And I think in terms of the thinking about whether people are actually-- the question is really whether people are over paying for these classes, whether they're paying for them, and they're not using them, right? That's the kind of gym membership example.
And that might be going on. But these are basically two different kinds of effects. And I think we can probably even test for them. What usually happens in these classes, if you look at the data, people stop paying before they stop attending. So once they start getting chased about paying, then they kind of drop out. So it's not as enforceable as gym memberships, so that might not be as big an issue.
In terms of thinking about whether people are doing [INAUDIBLE] in the US, that's something that there's a lot of people in the department, in the economics field-- the graduate field of economics is very large. I think there's 100 or so economists in the graduate field that kind of advise grad students across all the different units. And there's a large number of economists who actually work on-- the biggest group is probably in education. There's also quite a few who work on health, like health behaviors and health care and those kind of issues. Those aren't things that I work on myself. But it's actually quite a strong area for Cornell economists in general.
AUDIENCE: Yeah. When you talk about the importance about literacy classes for the moms and for the children, we, in this room and [INAUDIBLE] intuitively know that more education equals better chances at a better life, and we all value education, right? But my question is, do you have any evidence, or have you looked at how they value-- if they value education, and if not, how they value it? In other words, what's their incentive of their attending classes or sending their kids to school?
JAMES BERRY: Yeah. That's actually an area that I'm actively working on. I think there are a couple of issues. One is even in the US, we don't know exactly what the benefits of education are. They're very hard to quantify, extremely hard to quantify. We all know they exist. But knowing exactly what they are is very hard.
In developing countries it's the same. Conditional on that, there's a few things you can do to just think about whether people just perceive benefits at all or perceived benefits that are kind of in the ballpark of what we think are the benefits of education, which in developing countries it's harder because, say, women are less likely to work in developing countries, and trying to understand with the benefit of education for a girl is going to be a little different than a boy, stuff like that. But what we're doing is we measured people's perceived benefits of education across a bunch of different contexts.
I've worked on it in India. I'm doing a project in the Dominican Republic that's actually measuring these. And it turns out that on average, people's beliefs are pretty accurate about at least what salaries are for different levels of education. What's less clear is that there's a huge dispersion. And there's a lot of people who undervalue education or seem to undervalue education. And there's even some people who look like they overvalue education.
Why that is very unclear. And that's something that I'm actively working on. I think it might have to do with just the kind of information that they get.
GEORGE BOYIN: Man in the blue?
AUDIENCE: Yeah, I wasn't trained in economics. But I was trained in psychology. And [INAUDIBLE] history [INAUDIBLE] behavior and behavioral economics to get accepted by economists. You mentioned the resistance. [INAUDIBLE] mentioning [INAUDIBLE] psychology [INAUDIBLE] [? for ?] [? centuries, ?] [? although it kept ?] [? developing. ?] And the other question is, I'm aware of Daniel Kahneman winning the Nobel Prize in economics. And he's a social psychologist. And I wondered what [INAUDIBLE] he had [INAUDIBLE] psychology?
TED O'DONOGHUE: So the former, I think the major reason it took so long is that I think economists get a bad rap in the sense of lack of understanding what we do. Economists are not in the business of trying to develop a good model of individuals. Economists primarily are in the business of having a tractable model of individuals that allows us to get decent models of markets, economy, things like that.
And so the individual, in some sense, we're just trying to get something that's reasonable. But then we're moving up to a higher level of analysis. So in some sense, a lot of what Dick and Bob were trying to do was to convince economists that-- I think a lot of economists, if you ask them, would say, well, those assumptions are wrong. But they're mostly just going to be noise. It's not going to be systematic effects.
And it's a little bit convincing them that they actually, no, it's going to be systematic effects. So it was their work to convince economists to listen. And then the next generation's work, then, to convince economists, to demonstrate it formally the systematic effects. And that's sort of been [? the story ?]
In terms of connections at Cornell, Cornell has perhaps the best connections anywhere of economists and psychologists. So I spend lots of time talking with the psychologists here. We run joint workshops, where we alternate between more econ-oriented talk, more psych-oriented talks and have a mixed audience.
The theme project that George mentioned that I lead was also bringing a group of economists and psychologists together to spend the year together. And so this is an area that's active, and we're continuing to try to build upon.
GEORGE BOYIN: [INAUDIBLE] the back.
AUDIENCE: One of the questions which comes to mind in many of these situations is, what sort of economic problems are related to lack of information which one would need to proceed to make an informed choice? Versus which sorts of behavior truly represent a paradoxical or irrational behavior based on psychology, such as whether you state something is being first an opportunity or just simply restate the problem linguistically in an inverse sense, even though it's isomorphic? And classically there have been many sorts of situations where lack of information has been understood and dealt with long before behavioral economics [INAUDIBLE].
TED O'DONOGHUE: Yeah. That's a very good point. And it's something that is being taken more-- the economists are looking at it more and more carefully. So I think at some level, the early generation of behavioral economics got away, I think, with being a little too loose on ignoring the informational story, I think because it was new.
But the literature's been getting, in some sense, more and more rigorous, more and more careful. There's more and more work trying to say-- in some sense, exploring. If I say, if we provide people with certain types of information, does that make the effects go away? And sometimes it does, and you're exactly right. So it's something that's getting thought about more and more.
AUDIENCE: I'm standing up and I'm talking as loud as I can, because one of the problems with economists like myself is we don't [? plan ?] very well and don't have microphones.
[LAUGHTER]
[INAUDIBLE] I didn't hear the question. That was why. And [INAUDIBLE] I want to go back. And I've been doing research on these questions for years, in the 1970s. And why didn't people adopt new technology? Why didn't they? And it always comes up to the fact that there are at least half a dozen [INAUDIBLE] why. So you left us with a graph that says, oh, these are [? reasons ?] why, and then you [? stop. ?]
And the point is-- the question is, how do you go beyond that to say, which of these six are important? And are you going to research on that? Or is that still the [INAUDIBLE] question of behavioral economics, which we were doing in the '70s before we knew it was called behavioral economics?
[LAUGHTER]
[SCATTERED APPLAUSE]
TED O'DONOGHUE: So I'm sorry if that [INAUDIBLE]. What I'm trying to say with that last slide of the six things is that's exactly what the co-authors and I are currently working on is trying to think about how we would tease those six things apart, to try to say what additional data we would need to see if it's social taxation or if it's present bias or if it's something else. In fact, our list is about another six items. And so that's exactly, in some sense, what that project is about is trying to nail into that.
I'm sympathetic to what you say, because economists do tend to sort of raise more questions than we answer, I think. But I think we're getting better.
AUDIENCE: In another field of economics, we see [INAUDIBLE] the huge disagreement between economists about whether [INAUDIBLE] et cetera, is working, if it ever worked, will it work, et cetera. And [INAUDIBLE] not only [INAUDIBLE] but among economists as well just have totally divergent views. Are there [? similar ?] issues, in addition to what this gentleman behind me was talking about, that [INAUDIBLE] in development economics?
TED O'DONOGHUE: So there's definitely controversy. I'm not sure it's as large as in some of the sort of bigger things. But where it's been harder to model that, we're really nailing down the individual. In some sense, it's possible to get better data.
But that said, there's, I think, still controversies. One thing that I like about being in the behavioral economics literature is those controversies typically don't devolve into fights. But rather they get into constructive conversations and people trying to say, I believe this. You believe that. How could we try to figure this out? But very much, we [INAUDIBLE] [? advance ?] in a lot of questions [INAUDIBLE] about development.
JAMES BERRY: Well, I have to say very quickly that the stuff on private-- private provision of education, and private provision of services generally, in developing countries is very controversial in the policy-- from the policy perspective. Because, by charging people for things, you're basically eliminating the extremely poor from certain types of service delivery. And whether-- the extent to which that actually matters is really important.
I mean, it's something that I've spent a lot of time on. But if you just think about an education, why wouldn't the answer to this-- you could just reasonably ask, why are you studying private education markets if the thing that you really need to do is improve the government system? I mean, understanding the way the market works here is extremely important.
AUDIENCE: [? Four ?] years ago, I was [INAUDIBLE]. You could get by with just high school algebra. Recently, a son of mine studied economics at Stanford. I saw his book, and I thought it was calculus with a little economics thrown in it. How quantitative are you going to make the new Cornell economics curriculum?
TED O'DONOGHUE: [INAUDIBLE]
GEORGE BOYLIN: Can we hold that off until afterwards-- can you ask because-- well, I mean [INAUDIBLE] the question, but I think there's lots of other questions about the [? work ?] today.
TED O'DONOGHUE: The short answer is that we're trying to make it a mix of-- it does need to be calculus-based, but we also want [? courses ?] that aren't purely calculus-based, because economics really is both.
GEORGE BOYLIN: We certainly have a large-- we have a large number of electives that don't require calculus, as well as electives that do require calculus.
AUDIENCE: [INAUDIBLE]
AUDIENCE: I'd just point out to the gentleman in the back that he doesn't have a microphone, and he's only person. So obviously [INAUDIBLE] My question really is about loss avoidance, loss aversion. A, where does that come from in our psychology, and B, do we see an uptick in that when there's a lot of turmoil in the markets?
TED O'DONOGHUE: That's a very good question. So the question is about loss aversion, which is an idea that, sort of following up on what I said about gains and losses traditionally being seen symmetrically, the evidence suggests that maybe people react a lot more to losses in the gains. So $100 loss feels a lot worse than $100 gain feels good. So to be honest, I'm not sure sort of what underlying mechanisms drive that.
There's interesting new stuff in neuroscience suggesting that the gains and losses, when people experience a gain versus experience a loss, different areas of the brain get activated. But I never quite know how to fully interpret that. So I don't quite know where to go there. It's actually sort of interesting. The behavioral economics literature has really picked up on loss aversion as something that was thought to be very, very important.
And it still feels is very, very important. In recent years, there's been a little bit of sense of maybe loss aversion isn't as important as we thought, and maybe we should be paying a bit more to what's going on with probabilities. And maybe gains or losses are actually not about how we feel about gains and losses, but how we interpret probabilities differently in those [? debates. ?] So it's something that's being explored recently, but that's sort of new. In terms of past experience, it certainly seems [INAUDIBLE] past experience changes things. So experiencing recent losses tends to make people more upset about more losses and more conservative.
GEORGE BOYLIN: [INAUDIBLE]
AUDIENCE: Any studies that you know of involving extended unemployment benefits and how they affect either the willingness or the timing of people to take employment?
TED O'DONOGHUE: So not really my area, although [INAUDIBLE].
JAMES BERRY: I don't know it--
TED O'DONOGHUE: So the question was any evidence of the impact of extended unemployment benefits and how that impacts decisions.
JAMES BERRY: Yeah, I don't-- I shouldn't-- I don't know enough to speak to that.
GEORGE BOYLIN: Ron Ehrenberg in ILR used to work on that. He hasn't worked on that stuff recently. There's a very interesting article on this that's just came out by a guy named Hank Barber at Princeton who got his master's degree at Cornell. Yes.
AUDIENCE: Going back to what you were saying about the loss aversion as [INAUDIBLE] how [? it's affected-- ?] is the economics department planning to do any individual studies using the MRI machine to see whether or not there are different areas of the brain that are used when looking at loss versus gain?
TED O'DONOGHUE: So the question is, as you might know, Cornell has recently acquired an MRI machine so we can start doing some of this neuro work. And certainly, the psychology and human development groups are getting very into that. And the question is, are people in the economics department going to get involved? So my best guesses is at least initially we're going to be involved in a little bit more of a consultancy way.
So part of the reason-- and a little bit this is a perspective amongst many economists, including myself, which is I think in terms of neuroscience work, there's a lot to be done in terms of understanding what's going on in the brain before we can really start taking that evidence to inform economics too much.
Because right now, people are trying to, but I think they're basically inventing stories. But at the same time, why I think the consultancy role is important is as they're trying to figure what's going on in the brain, hopefully they'll be asking the types questions that economists are interested in. So I actually am involved talking with like Valerie Green and other [INAUDIBLE]
AUDIENCE: To what extent are your studies pushing against culture of the country, the tribe, the family. And I don't think you ever use the word culture. It must overlay all these things. And so, to some extent, a mother isn't going to apply for education, because the tribe or the family says, no, no, no-- it's not going to happen, even though [INAUDIBLE]. So it seems to me our culture is probably not a positive impact on the studies [INAUDIBLE]
JAMES BERRY: I think that's an excellent, excellent point. I think that it's actually probably something that economists are-- I would say might be the worst at, in terms of understanding and sort of merging that with the way we analyze things. I mean, the one thing that people talk about with these things is empowerment, and thinking about how it shifts-- it shifts the dynamics of decision making in the household. And we do actually see that a little bit.
Women, when they a little bit of education, it doesn't take much. I mean, they learn write their name and they like very basic reading, and they start to actually make more decisions for their-- education decision for their kids. And so we can kind of see at a very narrow scale. But I definitely take your point that probably the culture-- trying to build that into an economic framework is hard. And it's something we need to do a better job at.
GEORGE BOYLIN: Yes, sir--
AUDIENCE: Doesn't that question raise the question of whether your research research is worth [INAUDIBLE] at all since [INAUDIBLE] human behavior?
TED O'DONOGHUE: Well, but apparently, there are purely cultural explanations. In some sense, I think the key is economists are looking at how certain types of forces push behavior around. I wouldn't want to claim we're explaining everything. And I think Jim's points about-- the reaction to that question in developing countries-- the same thing applies in the US. Often, economists say it's--
AUDIENCE: [INAUDIBLE]
TED O'DONOGHUE: Yeah, and [INAUDIBLE] oh, maybe it's driven by some norm, which I think, much like culture, we have no model for that. So [? we're bad ?] [? at that. ?]
GEORGE BOYLIN: We only have time for a couple more questions. Yes, sir.
AUDIENCE: Sorry, just a quick question. On the behavioral side, how could we-- [INAUDIBLE] bringing behavioral insights in makes sense. [INAUDIBLE] you run into trouble in having a workable model [INAUDIBLE]. And the just on the development side, if you look at things like in [? Brazil, ?] there's conditional cash transfers, so families get cash benefits only if children are in school. [INAUDIBLE] that kind of incentive structure.
TED O'DONOGHUE: So the first question was about in behavior, how close are we to having a workable model? And I think we're sort of going and realizing that we can't have a model that has absolutely everything in it. But this isn't new to economics. Economics has always been about I'm looking at this particular application, what are the things I think are important for that application. That needs to be my model.
Other stuff that I think influences people in another application, that doesn't need to be in my model, and behaviorists are doing that. But you're right, right now, we are very micro focused. A little bit, we're still trying to really figure out the implication at the micro level or the market level and policy at the market level. But there's some initial forays going on in the literature, trying to push a bit more towards macro. But it's hard.
JAMES BERRY: So the question on conditional cash transfer, conditional cash transfers is basically this program-- a program that gives some cash amount to a household if their kid enrolls in and attends school for certain amount of days in a month. And it's hugely popular. It originated in Mexico. It's kind of expanded all over the world. You have these programs. And it's not something that I've studied a ton directly. I've studied some programs with incentives on education behavior, but in terms of the economics and really the business ideas on how people react with incentives.
And that's something that people are looking at a lot. I don't think people have really dug into the mechanisms behind these sorts of things enough. But it's something that people are starting to look into. It's a very active program. They do get a lot of kids in school. They tend not to-- kids tend not to learn a lot when they get to school, partially because the infrastructure of schools tend not to be improve when that happens. But that's something that people have been working on from the policy side.
GEORGE BOYLIN: Yes, sir.
AUDIENCE: Did you do any research around in economics on how family size affects the decisions for education and spending? Whereas if you have like six kids, who do you decide to school, who works on the farm? Or do you have more of a need to do the buying in bulk things you talked about with a bigger family? Any interesting findings there?
JAMES BERRY: I haven't looked at that directly. There's a large literature on-- I guess it's called sibling rivalry literature. And things about looking at quality versus quantity of kids. That's the way economists talk about it. There is quite a bit of research on that. I haven't actually applied it directly to what I'm doing, but yeah, there does seem to be-- especially when households have a large number of kids, how to decide who to send to work. And those are actually-- it depends on a lot of things. The short answer is it depends on a lot of things, but it matters to the household.
AUDIENCE: And ma'am, and that's going to be the last question.
AUDIENCE: So besides the areas of your own research, where do you hope the field goes next, and what do you think the most pressing questions to look into are?
TED O'DONOGHUE: So I think mainly where behavioral economics has been moving in the last year and I think very exciting, and [INAUDIBLE] we'll see where it goes, is the impact of salience. Certain things are salient, certain things aren't. Now, you say that, and it's like, yeah, yeah, yeah. But the hard part is how do you make any progress on that without being able to say, I see something, that must be salient, which is very bad way of [INAUDIBLE]. People are trying to generate careful ways to think about salience, with some success. But I think that's a very exciting area for the next five years, for people to think about.
JAMES BERRY: For development, I'm very junior, so I don't have a great answer. But I think understanding a little bit and being able to really kind of reconcile the intersection of policy and behavior, and using the behavior-- trying to model behavior and do it in a rigorous enough way, you can inform policy. And when you go to evaluate policy, it actually works in the way that you model. And having that exact sort of chain of things is not something that we've sorted out as well as I think we should.
TED O'DONOGHUE: OK, It's 2 o'clock, so we have to stop. But let's thank Ted and Jim.
[APPLAUSE]
Behavioral economics uses insights from psychology to better understand economic outcomes. Development economics generates solutions to world poverty through economic analysis. Hear how these two fields are working together to better understand household behavior and improve economic outcomes in developing countries.
Presented by Ted O'Donoghue, professor of economics, and James Berry, assistant professor of economics; moderated by George Boyin, professor of economics.
