Eric Hanushek on the Education, Skills, and the Millennium Development Goals
Jul 27 2015

How important are basic skills for economic success and growth? Eric Hanushek of Stanford University's Hoover Institution talks with EconTalk host Russ Roberts about the importance of basic education in math and literacy and their relationship to economic growth. Hanushek argues that excellence in educating people in basic skills leads to economic growth, especially in poorer countries where years of education may be a poor proxy for learning. He argues that the U.N.'s Millennium Development Goals should emphasize outputs rather than inputs--performance on skill-based exams rather than years of education.

Eric Hanushek on Test-based Accountability, Federal Funding, and School Finance
Eric Hanushek of Stanford University's Hoover Institution talks with EconTalk host Russ Roberts about the current state of education and education policy. Hanushek summarizes the impact of No Child Left Behind and the current state of the charter school movement....
Jeffrey Sachs on the Millennium Villages Project
Jeffrey Sachs of Columbia University and the Millennium Villages Project talks with EconTalk host about poverty in Africa and the efforts of the Millennium Villages Project to fight hunger, disease, and illiteracy. The project tries to achieve the Millennium Development...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Phil Langton
Jul 27 2015 at 11:00am

An interesting conversation. I’ve listened to previous discussions between Rus and Rick and have often been left wondering if certain associations can be as directly related (dependent) as seems to be suggested.
Take the observation in this episode (starting at around 27 minutes) that immigrants from Korea, for example, tend to have outcomes in the US that reflect where they were educated – for example, if they were educated in the US then their performance in the workforce is correlated with other US students and NOT with Koreans who came to the US after their education (ie educated in Korea). Whilst I’m happy to accept that the difference could reflect the quality of the US education system relative to the system in Korea, there are other factors that may contribute. There are, for example, cultural norms to consider – the culture of respect for family, for elders and for authority (including teachers) and then there is the influence of other students ie peer-pressure. If there is in general a tendency to compete and comply then there will be positive feedback from the success of any given student, encouraging others to engage and succeed.
Consider that however important teachers are, it is the students that must demonstrate their learning, their skills. To do this students must engage – they must apply themselves and work at learning. If the students don’t engage then little learning will be evident. Perhaps the magic that teachers can demonstrate is there ability to manage each student in order to optimally motivate each one – something that every good sports coach knows how to do.

In summary, how is it possible to ignore all the cultural norms that may influence the pressure that students feel to apply themselves fully to the learning opportunities that are provided for them?


steve hardy
Jul 27 2015 at 2:00pm

A good example of a where educational excellence didn’t create economic growth is the Soviet Union.

Daniel Barkalow
Jul 28 2015 at 12:02pm

About that test question: while it does test a bit of mathematics, it’s also a test of whether you understand the way exchange rates are stated in test questions, which is somewhat harder to understand than the way that (for example) Google states exchange rates. I, for one, started off backwards the way the question was stated (but corrected myself before settling on an answer, in part because it was hard to calculate the wrong way), and wouldn’t have had that issue if it had said “1 US Dollar equals 1.1 Euro”. On the other hand, figuring out what the question means is a reasonable test of figuring out technical writing in your native language, which is another valuable skill, possibly even more so than the requested calculation.

Of course, correlating the test scores with growth means that the test must be measuring valuable skills even if they aren’t really math and science.

Michael Byrnes
Jul 28 2015 at 7:26pm

@ steve hardy

I think Hanushek’s response might be that as bad as growth was in the Soviet Union, it still could have been worse. (i.e., education had value even though growth potential was very low due to the system and institutions).

Chris Stone
Jul 29 2015 at 12:25pm

Listened to the podcast on my drive in this morning….Did I hear Mr. Hanushek twice refer to a place called “Silicone (sic) Valley” is this somewhere in California? Sounds like a very titillating place…

I really get tired of those in the educational establishment bashing US schools vs. those from (pick any random country: Finland, China, S. Korea, Japan, etc.). When the educational people bash schools, they are really asking for more money for public education. Has this solved the issue of poor performance, NYC?

Two points: 1. Didn’t the US just win the international math competition (last week)?
2. I just had a look at the CWUR international university rankings…take a look yourself, US has 15 of top 20 universities in the world, S Korea 2 in the top 100, #1 in China is 56th in world, #1 in Singapore is 65th in the world, #1 in Finland is #111 in world. Now exactly why are we using “Singapore Math?” Not to say the US system is perfect, but let’s stop bashing the US. Everybody is afraid to talk the truth about the US educational system, the poor performance (aggregated) is a “demographic issue…”

Jul 29 2015 at 4:01pm

Phil’s comment about culture made me think:

I went to high school in a town that was diverse (economically, ethnically, etc), but the population was ~70k. Not near a large metro area. There were two public and one catholic high school. (The Catholic school was by no means a prep school – just a private school, about as good/bad as the public schools.)

The diversity of the kids at this school was such that there were some with multi-millionaire parents and some from effectively dysfunctional homes. Some kids went home, had to do homework before play, talked about school at the dinner table, etc. On the other end of the spectrum, I remember one kid’s mom and aunt came and walked into a class to beat up some girl, throwing chairs around the room before being dragged out by cops. There were druggies, thugs, etc. Some kids went on to med school, got PhDs, etc. Lots dropped out. You get the idea…

This experience makes me skeptical of the relationship between the actual school and person’s educational success. Living in Chicago now, I see the same diversity with greater tails, but within a school district or neighborhood, there is much more homogeneity. You live in a great neighborhood, your school is “great” because the test scores are high and everyone goes off to college. You live in a bad neighborhood, you have a different outcome.

Charter/magnet schools have obvious selection biases, and I think make the data in large metro areas harder to use. My city had none, and parents obsessed with their children’s future had basically the same options as apathetic parents. From my personal experience, I don’t think the income of the parents is an ingredient for a pro-school culture. (But it’s highly correlated for obvious reasons.)

We might learn something about whether school quality is a good predictor of long-term success by looking specifically at medium-sized areas where there is a natural socio-economic/cultural diversity intra-school district.

Have others considered this? Is there literature about it?

Jul 29 2015 at 4:18pm

In my comment, I didn’t mention: the school I attended was one of the two public schools.

Aug 3 2015 at 10:09am

I would love to see a debate between Eric Hanushek and Bryan Caplan on the efficacy of education and the signaling theory of education.

Does Eric Hanushek believe that swapping good teachers from predominately white schools with teachers from predominately black schools and vice versa for coaches would greatly reduce the achievement gaps in academics and in football and basketball?

Eugene Swin
Aug 3 2015 at 9:13pm

Neither genetics nor IQ was mentioned in the podcast as a potential cause of educational or other success.

Hanushek’s book, The Knowledge Capital of Nations: Education and the Economics of Growth (CESifo Book Series) does concede at P. 78 that “An extensive set of estimates of intelligence quotient (IQ) differences across nations exists that could potentially supplement, or even replace, the international achievement data.” However, Hanushek misses the latest two books by the authors of these data sets– Lynn and Vanhanen, Intelligence: A Unifying Construct for the Social Sciences, 2012, and Vanhanen, Global Inequality as a Consequence of Human Diversity, published in 2014, that analyzes education, economic growth and global inequality among 178 countries. “ Vanhanen concludes, P. xii, “Because human diversity measured by national IQ explains 74% of the variation in the combined Index of Global Inequality, it provides the most powerful theoretical explanation for the emergence and persistence of great global disparities in human conditions.”

Here is my comparison of Vanhanen’s estimates of the national IQs (rounded to the nearest point) P. 66-7, with PISA’s 15 year olds’ math scores in national ranking at Page 5.

Note the very great similarity of the rankings. I would appreciate comment by Hanushek or others on whether his book’s rankings are any closer to the PISA scores than the national IQ scores. Do his regression analyses add much explanatory power to a simple model relying on IQ?

Countries with national IQs higher than 100:
Shanghai China 105 (IQ)/ 613 PISA Math, Singapore 105/ 573; Hong Kong China 105/ 561; Chinese Taipei 105/ 560; Korea 105/ 554; Macao-China 105/ 538; Japan 105/ 536; Lichtenstein —/535; Switzerland 101/ 531.

Countries with national IQs generally 100-90:
Netherlands100 (IQ)/ 523 613 (PISA Math); Estonia 99/ 521; Finland 100/ 519; Canada 100/ 518; Poland 98/ 518: Belgium 99/ 515; Germany 98/ 514; Viet-Nam 94/ 511; Austria 100/ 506; Australia 99/ 504; Ireland 93/ 501; Slovenia 97/ 501; Denmark 98/ 500; New Zealand 99/ 500; Czech Republic 99/498; France 98/ 495; United Kingdom 100/ 494; Iceland 100/ 494; Latvia 97/ 491; Luxembourg 98/ 460; Norway 99/ 489; Portugal 95/ 487; Italy 99 /485; Spain 97/ 484; Russian Federation 97/ 482; Slovak Republic 97/ 482; United States 98/ 481; Lithuania 93/ 479; Sweden 99/ 478; Hungary 98/ 477; Croatia 94/ 471; Israel 95/ 466; Greece 98/ 453; Serbia 90/ 449; Turkey 90/ 448; Romania 93/ 445; Cyprus 91/ 440; Bulgaria 93/ 439; United Arab Emirates 86/ 434; Kazakhstan 90/ 432; Thailand 90/ 427; Chile 90/423; Malaysia 92/ 421.

Countries with national IQs generally below 90:
Mexico 89 (IQ)/ 411 (PISA Math); Montenegro 86/ 410; Uruguay 93/ 409; Costa Rica 88/ 407; Albania 86/ 394; Brazil 86/ 391; Argentina 93/ 388; Tunisia 84/ 388; Jordan 85/ 386; Columbia 83/ 376; Qatar 79/ 376; Indonesia 86/ 375; Peru 85/ 368.

Stephen Williams
Aug 6 2015 at 1:18am

Eugene (above) has bought up the thing that nobody wants to see much less talk about. For some reason economists see people ( and sometimes culture) as fungible. They aren’t, the raw material that one works with needs to be taken into account. Look at this map. I see there are no comments responding to Eugene, that’s to be expected. Those who have dared to consider this have generally had their careers destroyed. Such is the result of speaking facts.

Aug 11 2015 at 5:07am

@ Eugene Swin.
Great post. The numbers you provided suggest that there is a strong correlation between PISA test scores and IQ test scores. Such a correlation does not surprise me at all but I don’t think there is any evidence to suggest that the correlation is causal.

The brain is plastic, at least according to some recent books I’ve read such as “The Brain that Changes Itself.” Plastic means it changes in response to external stimuli, which means the brain is like a muscle in that use makes it stronger. It also means the lack of use leads to atrophy and decline. Thus, if you keep a child in a closet every day, it grows up a mentally retarded adult. If you raise that same child under a smart, attentive tutor it will grow in to a smart adult [assuming no major genetic defects]. Rearing a healthy child in a closet vs. raising him under an attentive tutor are two extremes of a continuous spectrum of possible learning environments.

Given these verifiable causal associations, I would like to suggest that the associations you have noted between IQ and PISA test scores are both positive reflections of the environment around the students in each country rather than a reflection of IQ causing PISA test scores.

@ Eric Hanushek
Eric Hanushek’s conviction that excellent schooling is positively and causally associated with economic growth may be true or false depending on circumstances.

Human beings behave like Pavlov’s dogs. Behaviors that get rewarded tend to occur more often over time and behaviors that get punished tend to occur less often over time. Based on that simple, verifiable observation, I will argue that economic growth is positively and causally associated with rewards and punishments.

I have heard it argued many times that “profits” are a reward for innovation. If true, then that would mean that societies which allow businessmen to generate and benefit from profits would see more economic growth over time than societies that discourage profit-making and profit-seeking. Importantly, that causal chain assumes that the profits are obtained in a manner consistent with the precepts of Justice. Because it is also possible to generate profits through Injustice. Treasure obtained through theft, for example, is almost pure profit. In societies where Injustice is allowed, profits made through innovation would simply attract thieves. Given that theft and robbery are punishments, Injustice predicts less innovation over time and thus negative economic growth.

Given this behavioral model of economic growth, how does education lead to economic growth? It depends. You can’t teach what is not yet known so it is irrational to imagine an education system can cause economic growth through innovation. You might reasonably ask an education system to teach economic stability. That is to say, it might be reasonable and desirable to ask an education system to teach its students to maintain the present level of economic production but teaching innovation is, by its very nature, impossible.

However, that argument only holds when innovation is the only means to economic growth. In third-world countries, prosperity is not limited by the absolute knowledge frontier. Catch-up economic growth CAN be taught through a school system since the information is already known—somewhere in the world—and is therefore teachable. Thus, an excellent school system could–and probably would–cause positive economic catch-up growth, even in societies where Injustice is rampant. “Excellent schools”, therefore, are defined by the access they provide to the world’s knowledge.

I appreciate Eric Hanushek’s efforts to review, compile, and relate present day studies on education on Econtalk.

Aug 14 2015 at 8:03pm

[Comment removed pending confirmation of email address and for irrelevance. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.–Econlib Ed.]

Aug 21 2015 at 9:17am

Bjørn Lomborg has come interesting views on the Millennium Development Goals, and development in general. I’d love to hear you interview him sometime.

Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

About this week's guest:

About ideas and people mentioned in this podcast episode:Books:


Web Pages and Resources:

Podcast Episodes, Videos, and Blog Entries:



Podcast Episode Highlights
0:33Intro. [Recording date: July 15, 2015.] Russ: Our topic for today is the Millennium Development Goals, especially related to education around the world, and how we might think about the best policies to improve wellbeing and the role education plays in that. We're going to draw on work that Rick has done with Ludger Woessmann, and we'll put a link up to the study you've done on the topic. And to get us started, tell us what the Millennium Development Goals are and why they matter--if at all. Guest: In 2000, the United Nations and UNESCO (United Nations Educational, Scientific and Cultural Organization) set out a range of goals that they thought would help to guide countries and development agencies in their policies. The goals were supposed to be met by 2015. Russ: This year. Guest: And they ranged from issues of health to poverty to inclusion of females, and the one that I pay most attention to is that all kids in the world should have basically an 8th or 9th grade education by 2015. Now, they're important because they actually seem to influence both what development agencies do and the way countries respond. Russ: Just fascinating. There's some desire to "meet the goals." They don't want to look like a failure--I guess that's part of it. Guest: Absolutely. And in fact there has been a large effort to have an annual report on both how individual countries are doing on separate goals and to provide narrative on better development policies. So, they've been producing reports for the last 15 years on what's been going on around the world in terms of development. Russ: So, here it is, 2015. How are we doing? Guest: Well, in 2014, they suddenly realized we weren't going to meet the goals. Russ: It's getting close. Hope reigns eternal, for a while. And then it starts to die. Guest: And so in 2014, they started having new meetings about what would be post-2015 development goals, and these were large international gatherings around the world to try to hammer down a new set of goals that would be met by 2030. They sort of have a thing about every 15 years, having a set of goals. Russ: Well, 15 years is long, and I have to think we've got plenty of time and also to think, 'Eh, it's so far in the future, we can just--.' Yeah. Guest: Well, it could be. But they work very hard at this, and in May of this year, May 2015, there was a large international gathering in Inchon, Korea to try to hammer out the final version of their post-2015 goals. This version is to be enacted at the U.N. (United Nations) in September of this year. Russ: So, how did we do--or how are we likely to do in 2015 in reaching the goals that were established in 2000, that every child in the world would have an 8th or 9th grade education? Guest: Well, I think that you have to conclude that there was remarkable progress in terms of the quantity of schooling that people were getting: how many kids were in the classrooms and how much school they were completing between 2000 and 2015. Particularly in South Asia and Africa, and to some extent Latin America, we saw some fairly dramatic increases in school attendance and in school completion. Russ: But there will be an actual number. Like 73% or 46%, right? Guest: Absolutely. Russ: And where is that number likely to fall? Guest: Well, I know it depends basically on the wealth of the country. So, for the developed countries of the OECD (Organization for Economic Co-operation and Development), that number is above 95%. Russ: The OECD being the United States, Europe, Canada, Japan? Roughly. Guest: And we've added, some added countries now: We've had Chile and Mexico and Turkey, are in there also. But it's basically the club of developed countries, is the OECD. If you get down to what the World Bank calls 'middle income countries,' you see that maybe 80% of their kids will have completed 8th or 9th grade education. Up from probably about 60% in 2000. So there have been these-- [?] Russ: Big increase. Huge increase-- Guest: but there is still, by this standard, a ways to go. Russ: And knowing, having read some of the report and having talked to you about this issue before, there was a 'but' that you didn't get to in your summary. You said 'there have been some incredible gains in the level of education, the number of people sitting in classrooms.' So, what's the 'but'? Guest: The 'But' is a huge but. Lots of kids had butts in chairs. They were sitting in classrooms. But they didn't learn anything. Because we have separate measures of math and science and reading that have been given to a large number of countries now internationally, including developing countries. And we see that amazingly little learning by some of the kids that had 8 or 9 years of education. Russ: Yeah. And we've talked about this on EconTalk before, with previous guests. And the challenge that we as economists typically are measuring something called years of schooling--which we assume perhaps naively till recently was related to growth of knowledge, what we call human capital. And it's dawning on us that perhaps there's not a lot of human capital accumulation going on, despite the hours that are spent there. Guest: That's absolutely the case. Once you have measures of what kids are learning, you see that years in the classroom is a very imperfect measure of the skills of people in different countries, in particular, because the difference between, say, kids in Peru and kids in Singapore, in terms of measured test performance translates into maybe 5 years' difference in [?] schooling. So that two kids-- Russ: That looks the same. Observationally. Guest: That looks the same. So, two kids in the 9th grade, if you call Singapore, the 9th grade--in Peru, they're in the 4th grade by Singapore standards. So that simply recording how much time people are spending in the classroom does not in any way give you information about whether they have the skills to compete in today's international world.
8:24Russ: So that raises the question that, when we think about the Millennium Development Goals going forward, if we said I'm going to put you in charge of those goals, obviously one way to set the new goals is to say we're going to keep the old goal. Which is within reach to getting every kid to have an 8th or 9th grade education. And we'll just get close to 100%. I assume that's not what we're doing. Guest: That's not what I would like to do. At the meeting in Inchon, Korea in May of 2015, the proposed goal was entirely, everybody getting an 8th or 9th grade education, with no mention of quality. Now, fortunately, a number of people are beginning to realize that you can't ignore quality. And so there was pushback at meetings. Russ: Were you at that meeting? Guest: I was not at that meeting. Russ: What kind of people are are the meeting? Are these education ministers from the countries we are talking about? Guest: There are education ministers. There are finance ministers. There are all kinds of people. Because the Development Goals span lots of different areas. But the Vice President of the World Bank for Education was there. The Head of the Education Section at the OECD was there. The Head of Education for UNESCO was there. So these are high-level people with all of their people to carry their materials for them at the meetings. So it was a very large meeting. But these are serious people who are trying to get some consensus. Russ: But when you say there's just some pushback against this goal of just years of education as a crude and unsatisfactory proxy for human capital accumulation, who is pushing back? Who is representing and lobbying for a subtler or richer appreciation of what's going on? Right now? Guest: I think it's now a fairly broad group. There are representatives of the World Bank that I believe now fully believe in qualitative goals for education. There are representatives of the Asian Development Bank that have the same perspective, that you have to mention quality even if you stick with attainment goals. The people from the OECD and the developing agencies like the Department for International Development of the United Kingdom are there, and they also are now quite aware of the difference between years of schooling and learning. Russ: So, I just want to mention the episode I was thinking of in the back of my mind is Lant Pritchett's, and his work that we've talked about here before on that disconnect between people in school versus actually learning something.
11:41Russ: So, what should we be doing? What do you recommend? Talk about what your work has been in this area and what it suggests should be a better goal? Or a strategy, more importantly, obviously. Because the goal itself is not the ultimate. Guest: In a larger sense, Ludger Woessmann and I have been working on issues of economic growth and how that relates to the skills of people. And also on earnings of individuals and how that relates to skills now for a dozen years. And recently put together a book on this subject called The Knowledge Capital of Nations, which tries to look at the relationship with growth. But because of the published 2015 Development Goals and because of this meeting in Inchon, the OECD contacted Ludger and me to see whether we could provide some better guidance on quality. And over the last spring we put together a small book called Universal Basic Skills, where we essentially propose keeping the 8th and 9th years of schooling standard, but that somebody at the end of that schooling should be able to meet a level of basic skills which can be defined now by the tests that are available. As we've discussed in the past, there have been international tests of math and science now for 50 years. But recently, since 2000, the OECD has, every three years, tested kids in a wide variety of countries. These are called the PISA tests. Which stands for the Program for International Student Assessment. Every 3 years they take a math problem, a math problem that's aimed at 15-year-olds, which is about the 8th or 9th year of schooling--so it's the right level. They take this math problem; they translate it into local languages and march it around the world. Russ: One problem. You say a math problem. You are thinking philosophically. They have set of math problems. Guest: They have a set of math problems. Russ: Okay. For sure, getting that one right when your whole future, you whole country's self esteem is riding on it, it seems too much. Guest: Yes, indeed. So, but let me give you one problem, to give you an idea of what the standard we propose is. We propose what's caused what's called Level 1, which for math means if you are given all of the variables and all of everything you need to solve a problem, you can solve it. So, an example would be that the people flew from Washington, to Inchon, Korea, and that airfare cost them $4600 dollars. If the exchange rate between the dollar and the euro is 1-to-1.1, how much did this ticket cost in euros? And so the idea would be, that's a Level 1 problem, or a problem similar to that, that people at 15 could reliably solve that problem. Well, it turns out that lots of kids age 15 can't; and including 24% of U.S. students age 15 can't reliably solve problems like that at Level 1. But it gets up to much larger numbers as you talk about developing countries. And what we propose in this work is that the post-2015 development goal gets and 8th or 9th grade education and can solve a Level 1 problem. Which to us seems to us like what you might call functional literacy in today's world. In order to compete in an increasingly international labor market and product market, people have to be able to solve some very simple problems like that. Russ: So, I'm not sure that's true. Let's talk a little bit about that. I'm also curious about Level 2 and 3. But that particular problem is something you'd want to solve if you are going to be an effective tourist, and spend your money carefully and wisely. It's not obvious to me--in many, many jobs, virtually no mathematical sophistication is necessary. Obviously in either the United States or in a poor country. On the other hand, there are many jobs where it's essential; and Level 1 would be trivially unhelpful. Before we go on: How many levels are there? Guest: There are 6 levels. Russ: Six levels. So, what would be a higher level type of problem? Guest: I don't have any readily available examples. Now, there was ones that require you to make some inference, or make-- Russ: make a connection. Guest: a connection, between the information you are given and the problem that you are trying to solve. So they would require direct inference. But not too much more. That they have to be able to extract the relevant information and use that. Russ: [?] You might not know what the exchange rate is, or you have to find it. Guest: Yeaah. Applying basic algorithms to solve problems. So, coming back to, just as an aside to your problem of being a tourist--I said that 24% of U.S. students couldn't reliably solve that Level 1 exchange rate problem. I actually mis-stated that, because 44% of U.S. students can solve similar problems to that. But 54% of U.S. students cannot solve an exchange rate problem. Russ: Right. Because they've never gone overseas, they've never-- Guest: No, that's not right. It's that all international transactions are in dollars. And we never have to worry about exchange rates in the United States. But other countries do. Russ: For sure. Obviously that's a culturally biased question, against U.S. students. Which--obviously I'm joking about it a little bit, but it's a serious problem, I assume, when you make these problems--translating them into the domestic language is not sufficient in many of these cases, I presume. Guest: Absolutely. And that's why, I think they actually test reading. But I fail to see how you can actually get comparable reading problems that you can get a good estimate of the difference in people. But math problems seem much more straightforward and [?] science problems.
18:54Russ: So, going back now to the, the goal and the policies. So the goal would be you have to get a certain number of people to get a passing grade or a certain level of achievement that certifies you effectively as literate--mathematically or numerately or reading-wise. And the question is, I guess--I guess there's two thoughts. One is you might be, with that standard you will of course teach to the test. Which maybe is a very good thing; we are talking about very fundamental levels of proficiency. And then the question is: How valuable is this going to be? And I know you've looked at that. So let's talk about that. Guest: This is going to be extraordinarily valuable. And we'll come back to teaching to the test. In some sense, you want people to teach to the test in terms of-- Russ: the right test-- Guest: having these basic skills, if they are measured well. The work that Ludger and I did here follows on [?] analysis of economic growth of countries and what the implications are. What we've done in the past is to show that essentially growth rates around the world are almost entirely dependent upon the skills of the population, and where skills are measured by performance on these international math and science tests. So that we can explain in 75% of the variation of growth rates across countries by simply knowing measures of the math and science ability of the population. So, these tests have a big impact on growth rates; and it's growth rates that determine wealth of a country in the future. It's economic growth that makes you better off today than your parents were. And if you don't have growth, you are very much different than your parents. So to give you an example-- Russ: In most human history [?] growth [?] parents. Guest: So, yeah, no, exactly. If you look since 1960 and you contrast what has gone on in East Asia versus what has gone on in Latin America, the average person in East Asia today is 9 times more wealthy than an average East Asian in 1960. Whereas in Latin America, they are 2 and a half times better off than 2 generations ago in 1960. Russ: Big difference. Guest: And this is all a simple manifestation of what growth rates or compound interest does for you. Russ: So the fundamental question is: Is the connection between--there's sort of a number of links in this chain. We have test scores as a proxy for skill/knowledge--ability, human capital. Obviously the test scoring in and of itself has nothing to do with growth. It's just merely an attempt to measure effectiveness of past education for the students. So we have that. That has to then link to economic outcomes. The presumption then is that people have more knowledge or are going to be more effective. And I guess my question would be reverse causation. Is it also not the case that richer countries are going to have more education? They are going to have more skills that will be measured by the test. But maybe they are really not so important. What I have in mind is, an issue we've talked about before--you know, it's funny--I always wonder whether listeners are frustrated when we go over topics that we've talked about in the past. You've been on many times, Rick. We've talked about many of these issues before. But I've forgotten almost everything we've talked about. So I assume many of the listeners have as well. And not only that; I'm much smarter than I was the last time we talked, because I've listened to--I've talked to Lant Pritchett and I've talked to James Tooley. And so I like to come back to these topics and chat about them again. Because I find that's actually the way education takes place. Guest: This is the bane of teachers, because I have given you all of these pearls in the past and they've gone away. Russ: Yeah. Guest: Disappeared. But I'm always willing to come back and talk about these, because those are the right questions. Russ: And these are the fundamental questions that don't have easy answers, and so for me it's very useful to think about them again, in a new way. So, go ahead. Guest: So, let me rephrase your question slightly. One way to put in your question is: If a country found a way to improve its schools and the achievement of its population, could it really expect to have higher growth in the future and better economic outcomes? And that's the fundamental question. Russ: So, make me more like Finland--make my education system more like Finland--will I be more like Finland--a wealthy, happy country? Guest: Not you--you probably won't be cold and all the other things that go along with Finland. So, we spend a lot of time in the knowledge capital of nations that summarizes all the underlying research trying to pin down this question of causation and whether achievement or the skills that are embodied in our achievement test really do cause differences in growth rates. It's very hard in international work to be conclusive about this and to have an airtight case, but what we do is take the series of most likely arguments such as the one you make: Does higher income cause more achievement, as opposed to achievement causing more income, higher income? And we try to test a series of these. So there are, first, the finding that changes in spending on schools by countries is basically unrelated to any changes in their test scores. So that if you thought that higher income countries could then put more money back into their schools as they grow faster, put more money into their schools, you don't see it in terms of the simple answers. We look at whether in fact institutions about the schooling system like the presence of external examinations for students that give incentives to students to work harder affect growth by affecting higher achievement. So you wouldn't expect external exams necessarily to affect growth directly. But in fact, they lead to higher achievement; and that by itself leads to higher growth, we can find. So there's a series of institutional structures of choice in the schools and examination-accountability systems that in fact affect achievement; and the achievement that is related to those differences affects growth. We look at whether in the past 30, 40 years when countries have in fact managed to increase their test scores over time, is it related to changes in their growth rates at the same time? And in fact it is, with the limited number of countries that we can look at those for. Then, there's a question about whether it's all just the economic institutions of a country that determine growth in-- Russ: That's my bias. Guest: countries with good economic institutions also have good schooling institutions-- Russ: Yup. That's my worry. Guest: and so it's something left out.
27:35Guest: So one way to get around this in a cultural argument is to look at test scores that are available for different countries and follow immigrants into the United States from different countries with different average level of skills. And compare them to immigrants from the same countries that got their schooling in the United States. Does it make a difference when we look at the test scores of those educated in their home countries as compared to the United States? And what you see is that it directly affects the income levels of the immigrants. So, an immigrant from Korea will earn as if he had the test scores of Korea if in fact he got educated there, but not if he got educated in the United States. So, comparing two[?]-- Russ: So, I've got two--this is my thought experiment. So I have two Korean immigrants. One of them comes here, comes to the United States at 18. They have been educated in Korea. The other comes here at 6 and gets educated here in the United States. So they both arrive into the same set of economic opportunities, in theory. And yet the one who is educated in Korea doesn't do as well as the one who is educated here--if that's what you are saying. Guest: Well, not quite. He does better. He does better because the Korean education is better, and that his income is directly related to the skills on average that he would have gotten from Korea. So we are looking at the same economic institutions, the United States, and looking at how skills are treated there; and we are looking at the same culture, because we are comparing Korean to Korean. Russ: Right. Guest: And the difference is not culture or economic institutions but skills that appear to matter in terms of U.S. earnings. So again, it's not-- Russ: So that test, just stick with this for a sec. That test, attempt to control for that statistically, what kind of breadth are we talking about in terms of number of countries, observations? And how would you--where are you getting those data from? Because that is an [?] interesting example but it's a fascinating example. Guest: We take the data from the same set of international tests that we use in these growth models, that we have for 63 countries. Russ: But how do you have data on Korean immigrants, incomes, versus Mexicans, for people who arrived at age 6 versus 18? That's a lot of detail there. Where do you get that from? Guest: The U.S. Census gives us several hundred thousand observations of immigrants from different countries, and we know when they came to the United States and where they would have been educated from the Census information. And we know their earnings [?]-- Russ: from the Census, sure-- Guest: the labor market, so we can look directly at these issues. Russ: Pretty noisy[?] though, I'd guess. Guest: Well, you get estimates that look very similar to the estimates you get from individual panel data for native-born U.S. workers on the impact of measured test scores on U.S. workers. It's not just the function of the large Mexican immigrant population because it holds if we ignore Mexican immigrants. It is not something that is only a language issue, because-- Russ: That's another challenge. Guest: if we look at just the immigrants from countries where English is the first language we find exactly the same thing. So again, there are reasons to distrust some of this. But all of the evidence we have is consistent with this being a causal story. We get the same answer from these different ways of looking at the problem. Russ: So, tell me what you think, then, the implication is. Let's take that as a given. We'll come back and challenge it some more in a minute. But let's take that as a given. What are the implications, then, for the goals? You have some--it's more than just--and again, we don't literally care about the goals; we can also care about policy improvements. And this is not just about the developing world; it's about the United States as well. So, what do you see as the key things to remember, thinking about bang for the buck in education policy? Guest: Well, let's to back to the estimates in universal basic skills, which was that document that Ludger and I put together to relate to the first[?] 2015 development goals. During the following experiments, given our relationship between economic growth and achievement, first, think of getting everybody to have 8 or 9 years of education at the quality level of each country. Russ: So, no improvement in quality. Just get more people in the seats. Guest: Exactly the basic Millennium goals-- Russ: Of the past 15 years. Guest: Of the past 15 years. And the quantitative aspects of post-2015 goals. So, Experiment 1 is, get everybody with an 8th or 9th grade education of current quality. Experiment 2 is, take only those that are currently in school in each country and bring the bottom up to Level 1. Don't touch anybody else; just bring-- Russ: Keep the proportions attending the same but improve the quality of the experience so that they actually learn some of these basic skills. Guest: Exactly. Russ: They are apparently not learning from the evidence we have on the exams. Guest: In many cases, like the 24% of U.S. students that aren't there. Russ: And what would be--give me another country, if you have it off the top of your head. How bad is it in some places? Guest: 50% in Mexico-- Russ: Cannot do Level 1. Guest: A 15-year-old who are in school cannot do Level 1 problems. Russ: So, for Mexico, we can imagine going to 100%? Or take the 50% that are already there and get them up to Level 1. Guest: Exactly. So that's the comparison. Guest: Or doing all three. The third experiment is doing [?] Russ: Yeah. Guest: So you find that we can look at 76 countries in the world: 76 is chosen by the fact that we have test score information on 76 different countries in roughly 2013. So that we can project out to the future. And so we project-- Russ: Seventy six is a big number. You've got a big range of development and levels there. Guest: We have all 31 OECD countries, plus a bunch of other wealthy countries like the Arab oil countries are in there and not in the OECD. And then we have a lot of developing countries like 7 or 8 countries in Latin America who are developing. What you find for the middle income countries, which are the ones below the developed OECD level, about 80% of their students currently get to 8th or 9th grade of schooling. But larger, much larger proportion don't get to Level 1. Getting everybody in schooling has an impact on their future income, which we calculate into the future and then take present values--it would roughly be worth double their current GDP if they got that other 20% on average-- Russ: into the schools. Guest: Into the schools. Take the 80% that's currently in school and lift the bottom end up to Level 1 for these countries, you'd get 6 times the value of GDP, or 3 times more than just putting people in the classroom at the current level of learning. Do both of it together and you get 13 times GDP. So there are two summary statistics that come out of this. One is: Improving the skills of the population has at least by historical observations has a huge impact on the future economic wellbeing of countries. That's the fact that East Asia has been [?] times as rich as 2 generations ago as opposed to Latin America with much lower growth is 2 and a half times as rich. So, it has a huge impact. Secondly, if you just focus on getting people into seats, you find that that is insufficient. That that doesn't get you what you want. What you really have to worry about is what people know. And their skills in school.
37:30Russ: It seems indisputable. The question I guess is what the magnitudes are, and then the question would be how to get there from here. So when I think about these issues, one of the things that comes to mind is a recent episode on EconTalk with Morten Jerven. And he gave an example of how Congo did not use the plow. And everybody knows the plow is a very effective way to use agriculture, a way to do agriculture, and the fact that they didn't use the plow is obviously a handicap, and their productivity could have been so much higher. But when you look more closely, despite these claims, it turns out the topsoil in Congo is not so amenable to a plow. And it was a bad investment--it would have been a bad investment. So, when I think about these desperately poor countries, I certainly accept the correlation that we started with, which is that in the richer countries they have higher test scores, lower countries they have lower test scores. It's not obvious to me that when you improve those test scores they are going to be more like Korea or the United States or Denmark or Finland. And I guess the issue would be, one way of thinking about it is, Well, they are on the ground. Why is there is less--why isn't there more clamor for improving these skills, than you might think? So, in the case of the plow, you'd say, Well, why aren't they using the plow? What's wrong? You'd say, 'Give 'em the plows and they'll have agricultural productivity.' Or give them a wheat combine. Give them some real equipment. So the question is: Are you fooling yourself that by adding these skills they are going to translate into productivity in their home country the way they translate, say, in a developed country? Guest: I don't think we're fooling ourselves on that score. When we've done the analysis of growth, we find that just looking at non-OECD countries or poorer countries, you get an even larger estimated impact of skills on growth rates. So within the set of poorer countries, does skill matter? And what we see is that it has a bigger impact there. So that's the first bit of evidence. The second bit is sort of anecdotal evidence. If you start looking at Chinese factories--we've spent a lot of time looking at Chinese growth rates. If we look 20 years ago we would have seen a lot of labor-intensive activities. If we look today, we see much more mechanized activities. Sort of like thinking of U.S. car manufacturers--some time ago there was a big production line and you handed everybody a wrench or a hammer and they made a car. And today you hand them a computer that allows them to control the production line; and then there's a few interventions by people but it's really checking up on whether the computer is doing its job. And what you see is that around the world, there is skill-intensive change in production, so that over time, countries are demanding more and more skills. You see it in the United States, when you look at U.S. earnings of highly skilled people versus medium skilled people you see a huge difference in terms of their earnings in the United States. And that's because the most effective technologies are ones that at least right now use a lot of skill. And so I think that's what you would see, start to see, in developing countries, at the same time. Now, there's another--perhaps people think of as an anomaly: The value of highly skilled, rocket scientists, is even greater in developing countries than it is in developed countries, because developing countries are trying-- Russ: Scarcer. Guest: It's scarcer, and they are trying to imitate what goes on, and imitation takes skill.
42:11Russ: Those are all relevant and plausible. I guess the question is whether the range of stuff that comes along with a good education and a good economic system--a functional labor market, a functional dynamic business sector, whether that's really much more complicated than these basic skills. So, I love the point that you made. I remember a friend of mine used to be involved in the manufacturing of clothing overseas for American retailers. He said, 'You know, a sweater factory in China is a bunch of women with knitting needles. That's the capital: each woman gets two needles and they spend all day knitting.' Well, that was 30 years ago, or so; and it's not true any more. So, knowing how to knit, which is a very easy skill to acquire--you need some dexterity--that doesn't get you anywhere any more. So, as you point out, there's been an escalation in what skills are required. And I want to mention the episode we did with Adam Davidson where manufacturing in the United States, exactly like you talked about, is not a guy with a wrench. It's somebody who actually knows a little bit of calculus and is doing some very subtle things in stamping and metal work that would have been unimaginable 20, 25 years ago. So I think that's all true. I think the question is--I want to go back to this earlier question of teaching to the test. If I teach, if I improve the Level 1 skills of millions of children in a dysfunctional economy, are they really going to find the application of their better mental acuity at that point? I worry about that. In fact, my bigger worry is that by focusing resources--and we're going to turn in a minute to how you get there from here--on this particular aspect of the problem, are we giving everybody a plow? Are we saying--or the equivalent--in an extreme case, I think about a Connecticut Yankee in King Arthur's court--you know, you've got back from the future and you've got the only calculator. You've got the only Texas Instruments calculator. That's a big edge, if [?], if you really know what to do with it, could be [?]; or it could be absolutely useless, because you don't have the full range of other infrastructure you need to make it powerful. What are your thoughts on that? Guest: I think that you raise legitimate questions but I think what history shows as we look at growth rates and so forth is that when the skills are there, the economies develop to use those skills. And it's a slower process in some places than others, obviously: if there's no capital equipment in a country it takes a while to develop that and to figure out how to use it and so forth. But all of history that we can see, which is kind of limited, but looking from 1960 until today we see that we can explain three quarters of the variation in growth rates across countries both within the developing world and within the developed world by the skills of the population. And so what that suggests is that economies around the world have found ways to use the skills that are available. So if you are a smart investor and you know that all of a sudden there is a spurt of skills and knowledge in Peru, that isn't being used by the local manufacturers it might pay for you to have direct foreign investment in Peru to absorb these skills and use them competitively and to make better profits. Russ: So let's take that as true. So, again, that leap is--let's assume that there are some pretty general universal skills that make you more productive than you otherwise would be and that your economic system is able, where you are, to use those skills. How do you get there from here? Guest: Can I stop one second? Russ: Yeah. Guest: What I should also point out is that the character of the economic system does impact the returns on these skills, so that if I have a completely closed economy to the outside world, there is a return to skills, but it's about half the return that we see in completely open economies. And in countries that don't have secure property rights and all the other things that we think are important economic institutions, the returns to skills are lower, but they are still positive. And it takes a while to change these institutions along with them, although we've seen that countries that both change skills and the institutions that follow get the gains. Russ: Okay, well said.
47:32Russ: So, going back to this question of how we get there from here: I think it's easy to see the appeal of the 2000 Millennium Development Goals, that there's going to be 100% getting at least an 8th or 9th grade education, because most countries keep track of that. It's easily measured. I think we've spent the first half of this conversation; we understand it's not the best measure but we understand the appeal of it. Now--you're kind of push a more advanced, richer, thoughtful measure. I may be skeptical about the magnitude of the impact but I think it would be a better thing for the world if more children learned how to think and read and absorb information. So, [?], I don't want anyone to misunderstand me. So the question then is: How do you get there from here? Many people would agree with that on paper, which is where this might be ultimately, the question is: You want to make your schools better. You've got a bunch of dysfunctional schools where this learning is not taking place. How are you going to hope that this might make an actual difference in what they learn? How would you have that happen? What would it take to get there? Guest: Well, what it really takes, in simplest terms, is better teachers. We've talked about this before on EconTalk. As far as I can tell in both developed and developing countries the thing that makes the largest difference in schools is the quality of the teachers. Now, parents are important and so forth, but in a lot of these developing countries parents don't enter in much in the direct education, and it's the quality of the teachers. And here, how you get from here to there, is more uncertain because it depends upon lots of local institutional structures. The education sector everywhere in the world is heavily unionized and heavily protective of their current teachers, and tries to resist many changes. That holds for the United States; it holds for Peru, which I keep coming back to, and Brazil and other places. And part of this is educating both policymakers and the population in different countries that really makes a difference. A huge difference. We have that problem in the United States, where everybody nods their head and says, 'Yes, education is important.' But if you say, 'Well, you have to do dramatic things that change the quality of your teachers or how you pay and reward teachers,' they say, 'Oh, well, but that's pretty tough.' And so there's resistance to the idea of making major changes. Russ: For a whole bunch of reasons. Some of them just fear. Power of the status quo. Guest: Yeah, I'm oversimplifying this. But part of the purpose behind the work that Ludger Woessmann and I have done looking at individual countries and the gains they can make is to lay out, in the open: Here's the benefits you could get if you find a way to improve your schools. We don't work [?] at laying out specific policies, but we do work at convincing people that the gains are so large-- Russ: that you've got to try-- Guest: that you really need to be willing to consider a broader set of policies than having one student fewer in each class. Russ: Yeah.
51:21Russ: So let me raise a philosophical issue here about education generally that I think about a lot. If you are thinking about--parents care deeply about their children. Obviously true. And if you said to a parent, 'Something's wrong with your child physically. Does it matter the quality of the doctor you go to?' And every parent understands that it matters. Every parent wants the best doctor for their kid. They want a minimum standard. They are willing to pay a lot of money if they have to; they prefer not to, of course, but they are willing to pay money. And they are going to try to find out whether it's a good doctor or not. If people, neighbors and others, have bad experiences with that doctor, they are not going to want to go to that doctor [?] for long. What's interesting to me is that--even though parents know nothing about medicine--zero; most of us, most parents know very little technically about medicine--we are very careful about how we consume the people who provide medicine and health for our children. It's interesting to me that we struggle to do that with education. Just as we trust the doctor to know what's best, we also trust the school board, the curriculum, the administrators of the school. And we sort of assume--many people do; I don't, obviously, as listeners know--[?] they are all experts and they know what's best. Well, if you look at any reasonable history of education, what's considered best at any one point in time is rarely what's considered best 5 years later. There's fads. And any thoughtful parent realizes that this is a very tricky thing. So, the reason I raise this, I'm thinking about James Tooley's work, who was a guest on the program, where parents are aware that the public schools they are in are atrocious even though they are without charge, there's no fee. They are willing to pay to go to private schools even though they are desperately poor people. And I just wonder about this as an end around for the current problems that we are talking about. Is that really a plausible--given how hard it is to measure and for parents to measure and figure out quality--how important is that? It seems pretty important that these private schools are coming along. I'm obviously a big fan of competition. So what are your thoughts on that? Guest: Well, again, I don't think there's a simple answer, but in Pakistan where Tooley is working the government schools are so dysfunctional that you can put together these low-cost private schools that just dominate the government schools. They do better with teachers--they are paid a quarter of [?] Russ: A fraction of. Yeah. Guest: As both of us agree, competition among schools would be helpful. And there's lots of resistance to that. But charter schools are starting to make inroads and starting to have an impact. Russ: In the United States. Guest: In the United States, I'm talking about. What you see when you look at the research is that there are no silver bullets. There are no of the [?] that you talk about that are going to take over everything. But having good accountability systems, which are kind of threatened today in the United States but which are catching on in other countries, where you have better measurement of the performance of schools that you can present to parents, that counts, having more choice, rewarding teachers that are seen to do well and not rewarding those that do poorly--are all things that have been shown to have systematic positive impacts. How you introduce those in different countries or even within different States or communities in the United States is uncertain because you are starting at different points in trying to improve them. But we know that there are these overall themes of things that are going to improve. But there's resistance to every one of them. Russ: Yep. Part of what we mentioned before--the vested interests have a strong incentive to-- I was going to finish that sentence. I'm going to stop, because I think there is a cliché there about, that I think I've come to be a little bit uncomfortable with: that the teachers' unions try to stop all change. I think the unions do that. I think there are many teachers in those unions who desperately want to do better, would like to do better. I'm think now about episodes we've done with Doug Lemov and others about the pedagogy of how to become a better teacher. I think most teachers get into education because they care about kids. They'd rather work less hard than hard--as all human beings do. But if I said to them, 'You're going to work a little bit harder but it's going to really change these kids' lives and have an enormous impact on them,' I think they'd love to see it. Guest: You're absolutely correct. I think in general they are--it's a very good teaching force in the United States. I can't say that for all countries in the world of course. But there are still improvements that could be made. And I think you're absolutely correct to say it's not all teachers' unions. If you have States in the United States where unions are not important, where there is not collective bargaining, you don't see dramatically different outcomes. The leadership in schools, principals in the United States or headmasters in other schools, clearly is very important, we're saying [?] good evidence on that. There's a whole series of things. Ultimately, the policy that I would like to see is having a clear description of the quality of each school with their value added--that is, how much they are adding to the achievement of kids. Make that very public, and start making policies not on the basis of what is class size or what should we do about this or that or the other thing, but: Are you producing the outcomes? Because we know good teachers, clever, innovative, motivated teachers can do this. But they'll do it in very different ways. Russ: Correct. Guest: There's no--what we've seen in research is it's very difficult to describe what distinguishes a particularly effective teacher from a mediocre teacher or even an ineffective teacher. Because it's hard to pick out individual aspects. But we know that there are some that are much more gifted than others, and those are the people we want in our classrooms. Russ: Yeah, and that really brings up one of the biggest challenges of this which is that the temptation to look for explicit measures of this that are quantified is very powerful. And, as you say, a great teacher is not somebody who has a Master's Degree; it's not somebody who has been teaching for at least 7 years. Those things may matter. But it's often intangible. But that doesn't mean you can't find it. And I think great Principals often understand who their good teachers are, great teachers, and who their not so great ones are and it comes back to our point that without competition, without accountability, in my mind of the marketplace, very hard to expect that those processes are going to work well on their own. That's what I come back to. An example--I think I probably asked you this before, but I just got another example of it the other night. We had some friends over for dinner from Israel, and I have family in Israel; I have friends in Israel. They uniformly decry the Israeli public school system. They say it's horrible. And you think, well, everyone likes to complain, and how horrible can it be? And a guest last two nights ago, 'Well, here's how horrible it is. My 9-year-old finds her schoolwork boring. They ask her to spit back stuff that they ask her to memorize. She doesn't learn anything. And I told her, don't worry about it; if it's not interesting, you don't have to work so hard because grades aren't that important. But she gets As and Bs anyway, because the teacher doesn't want it to look bad for the teacher.' The student isn't learning. And you think, like, that level of dysfunctionality is so depressing. And maybe it's not symptomatic--I'm sure it's not symptomatic of all Israeli public schools. But the fact that that exists at all is so depressing to me. Then I step back and say, well, Israel is one of the most innovative countries in the world. They have an unbelievable workforce of creative people. It could be twice as creative, if their schooling system is better? Or is it really that so many other parts of life are what are doing the educating? So, why don't you talk about that? Guest: Well, first, I want to reinforce your friends. Israel doesn't do very well on these international math and science tests. And partly reflecting what's going on. What you see in Israel is the same thing you see in Silicon Valley, where we're sitting today. There is a culture and a set of incentives that encourage people to innovation, to do different things-- Russ: take risks-- Guest: to take risks, to have a dynamic economy. And so it's not all skills. But in fact, more skills, I think lead to fine-tuning the economic institutions that provide these rewards. And that encourage people to do better. And so, I think that Israel could do better if in fact it improved its schools. I was quite surprised to start seeing results on Israeli schools, because I know a bunch of smart Israelis. But they are all in the United States. Large numbers of them are.
1:01:59Russ: I think that's--you mentioned Silicon Valley. This is, one of the things that strikes you when you are here is how international it is. And we in this area tracked the most skilled people in the world. If they can get here, they often will come. Because--for two reasons. Obviously there's synergies and complementarities that we've talked about on other episodes, on economic development. But the other part is what we've talked about before, that their home country may not have the scope--they might do better in their home country than the less skilled folks who are in their home country, but to come here is even better. You get a multiplicative effect. Guest: So, that's been a U.S. advantage. We do better than you would expect given this performance of our kids and their measured skills, we do better than you would expect. Part of it is good institutional structure for encouraging economic growth and development. Part of it is historically we have much better schools than every place else in the world, which is not quite the same thing true today. A part of it is that we attract smart people from abroad: our immigrant population and our immigration population has been important in terms of providing much of the innovation and we do in fact see that. Silicon Valley is a mecca for foreign food places because all of the restaurants catering to their immigration population makes it a wonderful place. Russ: Markets work. Let's close and talk about 2030, just for fun. So, again, I think you and I, we've sparred a little bit in this conversation but I think we both agree. You obviously agree there's more things than skills that matter. I obviously agree that skills matter. And I think what we're debating to the extent we disagree is over the magnitudes and how accurately we can measure it. The next question will be: What are the probabilities that the world will move more in the direction that you are encouraging? So, put on your, take out your crystal ball and tell me--God willing, we'll both be alive and at the Hoover Institution in 2030, and it will be--I'll be 75; I don't know how old you'll be. I hope I'm still doing EconTalk then; we'll look back on this conversation as we review the 2045 Millennium Goals. And what do you think we'll have observed? Because as you point out, between 2000 and 2015 we made a lot of progress on the measured goal. Not so much progress on the human capital, but the world's a better place. I don't think it has much to do with the Millennium Goals. I [?] lots of other things. But just for fun, talk about what might be coming down the road. Guest: Well, what I see is that a number of countries will in fact improve their economic institutions and improve their schooling and be extraordinarily competitive for the U.S. schools, I mean U.S. economy in the future, unless we get our own schools up to the challenge. A number of other countries see what has made the United States rich. We have good economic institutions. We have lots of human capital. And they are working very hard to do both of those things in their own countries. And they are getting better in terms of quality. The United States now is 6th from the bottom in the OECD in terms of high school graduation rates. And we're about the same ranking in terms of math and science scores internationally. So other countries are pushing very hard and they are going to become much richer by it. Then there's a set of countries that we look at in all of the stories of the sort of political, societal disasters in sub-Saharan Africa, are ones that maybe they'll never make it. So, what I think we're likely to see in 2030, unless there's a big change, is more bifurcation in the world--that we'll see the rich countries pulling away from the poor countries in many ways. And that's going to cause all kinds of problems, for the world order to--how do you manage the fact that the rich are getting richer and the poor aren't? Russ: Well, on a more cheerful note I would point out, somewhat in line with the recent panel conversation on the Magna Carta that if we'd been back in 1975, or I'll go back to 1985, 15 years before--I don't know what year the Millennium Goals started; I assume 2000; I assume that's the 'millennium' part. If you had said in 1985 what does the future hold, you would have missed almost certainly the extraordinary transformation of China and India. Which has really made an enormous difference in the aggregate numbers; obviously there still is a difference. I'll say it differently. There's a bifurcation in the developing countries between a handful that have changed policies dramatically--and thank God there's billions of people who live in those countries who are benefiting from that. Guest: Well, I think those are two examples. If you asked me to predict what happens in 2030, I would predict that China is way, way ahead of India. And that's largely on the basis of the investments that the two countries are making in the human capital and skills of their populations. Both China and India, I see as having these double-digit growth rates from taking an extraordinarily bad economy and fixing it somewhat. Russ: A little bit. Guest: A little bit. And you can get tremendous growth out of taking an institutional structure that strangles business and lightening up a little bit; and you can get extraordinary growth. But in the future, both economies, as they improve the structure of their economy, is going to have to fall back on their skills of their population, too. And in that regard, China is making much, much larger investments than India. In India, there have been a lot of work, looking at their schools. There's one NGO (Non-Government Organization) that has tested large parts of the population and they find that of the 8th graders in India, the people that are in their seats in 8th grade, 25% of them cannot pass a 2nd grade reading test. And that's the kind of thing that we're talking about in universal basic skills. If you have that population left behind, it's going to come back to haunt India in the future. Russ: And we have reason to think China is doing better. Guest: We do. China is only participating in a few of these international tests, in very specific ways. But the city of Shanghai, a mere 25 million people, is at the very top of the lead tables[?] on international math and science tests. And the students that are tested within Shanghai, which is nowhere near representative of China, but show that they have invested largely in these skills and that their kids in Shanghai high schools are doing extraordinarily well.