Arnold Kling  

Test Scores and Economic Performance

Proposals for Liberalization... Demographics of the Oligarchs...

Gary Becker writes,

One of the challenging paradoxes during the past several decades is that American teenagers have consistently performed below average on international tests in math and sciences, and not especially well on reading tests, yet the American economy is more productive than any other.

I can think of only two classes of explanations for this phenomenon.

(1) International tests fail to measure the superior aspects of the U.S. education system.

(2) Education is not such an important factor in comparative economic performance.

I lean toward (2). It's better to have strong entrepreneurialism and mediocre education than the other way around.

If anything, I think that the caliber of our education system is over-rated. Yesterday, one of the students in my class at George Mason came for help with some sample problems. I showed how the problem boils down to solving for x in the equation

0 = 1 + x - 5

The student had no idea how to proceed. And Alex Tabarrok's wife, who teaches biochemistry (or somesuch) at Mason, is willing to bet that my students are better than hers.

Comments and Sharing

COMMENTS (35 to date)
Anonymous2 writes:

I can't see how college students could be so dumb that they can't realize that a number which is added to 1 and then decreased by five to give nothing must be four. I'm guessing your student wouldn't be able to solve x^2 - 5x + 6 = 0 either, huh?

Jack writes:

In many countries these international standardized tests are cause for national celebration and pride. Further, I think they often teach to the test. But in the US, who cares? The SAT is all that matters. It's your basic incentive problem.

Gabriel M. writes:

That's unpleasant... How do people like your student manage to get through high-school? Are you supposed to teach him/her reccursive methods and optimum control theory in 2-3 years? How about neo-Walrasian general equilibrium? I suck at math but I get better before any exam.

I understand that many US universities don't test their prospective students but rather admit them based on applications documents. I think that's a mistake.

The issue of productivity is, I think, that relatively few jobs require science or math skills so the US had enough competent people (some of the most competent anywhere) but the few brilliant people you need and have won't pull up the average enough.

Independent George writes:

You should try playing Precalculus Bingo sometime.

Independent George writes:

Er, 'Precalculus' and 'Bingo' are hyperlinks; they don't show up as such for some reason.

Tim writes:

I think Jack may really be on to something! An excellent SAT score will get you into the best of colleges. In some cases it may allow a student to get a substantial aid packages(e.g. pay the student for an excellent test score); So the incentives to do well on the test are huge. It has also long been known that it is far more effective to teach the test and the stratgies to take the test then it is to teach in detail the material the test covers( If it were not the case Sylvan, Kaplan, et al. would be out of business.) So yes Veronica, the SAT is an imperfect. Also, the educational system has no real incentive to change given the current system, so we get lack luster high school grads both in and out of college.

El Presidente writes:

Maybe having some dumb kids mixed in with the smart ones aids division of labor and socio-economic stratification. Maybe income differentiation and commensurate variation in economic leverage/power is a more effective tool for growth than we might think and tests just don't measure this because they want everybody to be a genius. You only need a handful of geniuses to make the world go 'round, but you need an awful lot of muscle to get things done. Think Egyptian Pyramids. There's a limit to the utility of education in society just as there's a limit to the utility of physical ability. Maximizing the utility of each is important. Not to diminish the importance of having highly educated members of society (Pyramid Architects).

I like to think of it in terms of Keynesian theory (APS, APC). If your individual propensity to save equals your individual propensity to consume then you are a maximally efficient member of society (you remove no wealth from circulation through saving/hoarding, whatever you call it) and although long-run PPC growth is a goal, as Keynes put it, "In the long-run we are all dead." In the short-run it helps to play the zero-sum game a little bit to make the long-run palatable. I think standardized tests don't account for this source of prosperity. And besides, who's gonna tell a kid that it would really be best for society for him/her not to go to college and not to earn more money? Please don't misinterpret this comment to be racist, justifying, or pejorative but maybe ther's a reason we seem to have such a high appetite for slavery and other cheap labor. Maybe adding rungs to the bottom of the income ladder makes the top seem higher by comparison. Not very nice but perhaps it's true.

Just a thought.

RogerM writes:

McKinsey and Co did a study recently on education and productivity and found the link to be the reverse of conventional wisdom. In the US, most job training takes place on the job, not in school. Also, economic growth causes a demand for better education. The USSR had a highly educated society.

robin writes:

I have another class of explanation: the distribution of test scores is broader in the US than in other countries. Consider a country where everyone has a slightly above average ability and imagine the US is 45% geniuses and 55% illiterate. They do better in the test, we get more innovation and smart business leaders. That's obviously not quite realistic, but I do think we have a lot of smart people, and, on the upper end of the scale, I'd bet we aren't behind at all. Unfortunately, there are also a bunch who have been 'left behind'.

The mean by itself is a poor indicator of overall performance, especially since national performance is nonlinear in the individual performances. I would like to see comparisons to other countries that involve, say, the 90th or 95th percentiles on these tests.

Tim Lundeen writes:

What does it mean that the average US test score is less than the average test score from other countries?

Part of this difference is because the US does not consist of a homogenious group of people, so our test score average is a blend of results from different groups, each with a different average/variance.

Part of this comes from the education system and cultural norms (how important it is to do well in school).

The important thing is that the top kids in terms of ability are getting a good-enough education and are incented by our economic system to continue our technological and entreprenuerial success when they join the work force. Averages don't measure this.

Although your story about not knowing basic algebra is pretty scary...

eric writes:

Didn't Easterly find that education was unrelated to growth in Africa, and argued that it was really the vibrancy of the private sector that affected GDP, as opposed to some 'grassroots' competency in school.

It makes sense, because most people (ie, not teachers) learn most of their valuable skills on the job, not from a textbook.

Fazal Majid writes:

A much simpler explanation is that the US imports much of its top talent from abroad. That's why the much more stringent immigration provisions, specially for students, are a growing concern for business leaders.

In any case, growth in China and India makes it more attractive for smart people there to stay there (in fact, given the cost-of-living differences, there is an active reverse brain drain going on currently), so the long-term viability of depending on other countries to solve our education problems for us is questionable.

Alcibiades writes:

There is very little I have learned during my tenure at school that I have ever applied to a job. Arithmetic, I suppose...

Bill writes:

When you say the problem boiled down to that simple equation, I am willing to give your student the benefit of the doubt. I remember my intermediate micro class and having to solve for the equilibrium amount of cocunuts Robinson Crusoe would trade bannas with Friday for given the time it took him to pick banannas, his utility for coconuts and hours spent sleeping in his hammock. A lot of times they boiled down to simple equations (although there was a lot of taking derivatives of negative exponents, which was always hard for me, I don't know why), but you had no idea how to boil them down, or what the overall purpose or real world application of the exercise was.

Fran writes:

The american teenage population will only form a sizeable proportion of the job force in say a decade or so from now .. maybe even more . We are considering the performance of the economy right now so how can we use math & science skills of indivisuals who are in their teens when you could presume that their contribution to the economy right now is not that significant . I think the study should have concentrated on the past test performance of people who currently form a major portion of the workforce .

dearieme writes:

Your foreign fans have always assumed that "the superior aspects of the U.S. education system" consisted of only (1) the best of your graduate schools, and (2) your cunning arrangements for recruiting clever foreigners. It's no wonder that tests for school children wouldn't pick these up.

Marcus Burrell writes:

I agree with Fazal. American productivity is no longer simply based on America's "home-grown" talent. America is home of some of the wealthiest countries in the world, and they have the money to hire intellectuals from around the world. They are skipping over the uneducated American students to pursue the lucrative prospects from other countries. Therefore, I would have to disagree with both classes of explanations for the growth of American productivity. 1) Our educational system is not superior to many other developed countries around the world. The international test scores are indicative of our education system. 2) Education is an important factor in comparative economic performance. We just are not remaining homogenized in our labor force as the 2 hypothesis suggest.

Dezakin writes:

"I lean toward (2). It's better to have strong entrepreneurialism and mediocre education than the other way around."

I doubt this; I think its more accurate to say its better to have strong market structures. Entrepreneurialism isn't significantly stronger in the US than in other countries, but the structures exist that allows markets to work better in the US on average. And still we need more and more competant engineers and service workers every year as the duller sections of the economy become more and more automated, and it certainly helps that the other countries educate fine workers for the US labor market.

Andy Wink writes:

'The Millonaire Mind' is a really interesting book that analyzes this phenomenom. It's from the same author of 'The Millionaire Next Door,' which was probably more popular.

Anyway, the guy analyzed the traits of many millionaires and found that education wasn't really that indicative of wealth.

Most of the millionaires in his sample, and according to him, in the country, are involved in sales (IIRC).

Anyway, very interesting book.

Rob writes:

For some examples of humorous math errors:

Max writes:

I have always been disgruntled about our own educational system here in Germany. But I have always felt that the US system was not soo much driven by excellence as the German is. I have compared tests and A-Levels with a friend of mine living in the US and we agreed that it was much easier in the US.

It can't be that anyone pursuing secondary education is not able to solve a linear equation, especially not when he is doing so in economics.
I can understand when students are not able to work with higher order differential equations, or if they have to learn statistical methods or probability schemes.
I'd think that you picked an example, which shows that not everyone picks the right subject in his first years :)

Shawn Mallison writes:

I don’t see why there would be a correlation between average test scores and productivity. Obviously average test scores are not a good gauge of a groups economic potential. In every organization I have ever worked for there were a few intelligent and highly educated leaders at the top and a much larger number of not so intelligent and less educated individuals being led. If every employee in an organization had the social and intellectual skills to run it yet were forced to work below there abilities, job satisfaction and productivity would plummet.

patrick Garrett writes:

ok i do not understand how you can say that mediocre education is ok. I agree that the international tests are not geared toward the US education system. However, the reason for that i they are more educated then us. So i guess we should continue to complain about the tests being to hard instead of making us americans smarter!

Jeremy writes:

comparing the test scores of teens here in america and analysing america's economic success doesnt really show much relation to me. when i was in high school i wasnt really conserned with making the best grades or my future success in the business world. it wasnt until college that i stepped it up and really focused on my work. college is where i became concerned about my future economic performance, and my future economic success will ultimately translate to the U.S.'s economy. so comparing some teens test scores to those of others in the worlds and not being able to see how the U.S. is the worlds most productive economy doesnt really add up to me.

Abigail writes:

How did this student make out of high school? The number of students who are slipping through the cracks in our public school system is outrageous. What I find to be even more unreal is the number of these students who are admitted to universities. How are they making it? These concepts that are part of the education system are going to appear in almost any field of work that these students fall into. I agree that a mediocre education might be sufficient for some, but if you want to be eligible for promotions you must be able to move beyond that. So letting these students move forward with their education, without learning all of the proper "tools" along the way will only harm them in the long run.

Antoine Bideau writes:

Well you wonder how students in the US are doing so bad but the economy is doing well.
Your answer is that the united state bring foreign people in the country to do the job that Americans can't do. I am speaking from experience because i am from another country and our standards of education are 10 times higher than yours.
And that is why American students fail the international tests.

Mr. Econotarian:

You wrote a comment that I accidentally deleted this morning while removing a massive spam flood from EconLog. It began like this:

As a professional broadcast engineer, I rarely use any advanced math. Despite...

Please feel free to repost it! I apologize for the error.


Grzesiek writes:


How many American students are going into R&D in the sciences?

To find that a student in a higher learning environment cannot solve the equation '0=1+x-5' is disturbing to say the least.

"It's better to have strong entrepreneurialism and mediocre education than the other way around."

Are there statistics out there that buttress your claim the little "Johnny and Joannie Fraternity" college students are starting small businesses?

I saw a study out "there" that stated that people are going to college when they're older either to:
* learn what should have been learned as an
* improve their lot in life
* continue their love of learning

"If anything, I think that the caliber of our education system is over-rated."

Might I add criminally over-priced too?

RogerM writes:

Most of the posts here seem to assume that productivity increases come from smarter workers. But that's not the case. Productivity increases come from better equipment and organization. The equipment and organization that increases productivity is very specific to each industry and business. There is no way that the general education that takes place in public schools can teach the requirements of every industry. So most productivity enhancing education must take place on the job and it does.

The only area in which better education will help is in innovation, but few people choose R&D as a career and if we need more geniuses, we can always import them.

The one area in which our educational system endangers our prosperity is its constant pushing of socialism as the answer to all questions. Highly educated people in a socialist government will always be poor.

TGGP writes:

No mentions of La Griffe du Lion's "smart fraction" theory yet? Guess I'm not as late to the party as I thought.

Jadagul writes:

Jack, I think you're right about the incentive problem. A year ago I blogged about an article in Slate that said pretty much that: American students don't even find out how they did, it doesn't affect their lives, so they don't care. Shows more intelligence, not less, to me.

tp writes:
Rajan R writes:

Perhaps, just perhaps, such test scores aren't actually representative of intellectual development - the American education system is vastly different from most other countries, and that's largely by one critical fact that most American highschoolers (or elementary schoolers, for that matter) aren't taking 2-4 years preparing for a national examination which would account for *all* (or in some cases, *most*) of their grade.

So I would assume the manner of teaching, for better or for worse, would change in adaption.

Rajan R writes:

BTW, to those suggesting importing as a feasible option, note:
- Other Anglophone countries, principally UK and Australia, regeared their rather protectionist immigration policy to favour professionals (an accountant with a Commonwealth-standard chartered accountancy diploma? Australia loves you).
- But America? If you have been issued a student visa, you *cannot* work in the United States for 2 years after.

SO the easiest way to get to US (to settle down) is to study in some other Anglophonic country like UK, Australia, New Zealand or Canada (heck, or like me, Singapore!) and then try to get into America. Even that is a bit tough - a brilliant computer programmer? ALl you get is a *temporary visa* - while places like Singapore begin pressuring you to take citizenship.

Personally, as a libertarian, I don't quite like any discrimination in immigration policy and practice amongst worker class - but if you have them, it is ridiculously stupid it might soon be easier for uneducated Mexicans to work in the United States (legally) than highly educated Indians.

brian simmons writes:

Math is not really my strong point, but I think I could probably find my way to an answer for a simple algebra problem. To my knowledge George Mason isn’t at the bottom of the education bin by any means, although is does amaze me how someone like that could slip through the cracks. Maybe it had just been a while since Algebra II; I’ll give the student the benefit of the doubt.

I also find it quite interesting that American teenagers have lower international tests scores in math and science, yet the U.S. economy is arguably the most productive. I definitely agree will the authors second explanation of how this phenomenon could occur. It’s no secret that education is taken more seriously in other countries. Other countries, such as India, have identified our weaknesses in math and science and began to take forge ahead in these areas. The amount of work outsourced to other countries is easily noticed by anyone that has ever tried to contact customer service at Dell or any other IT-based company. You can’t blame companies for taking advantage of a growing pool of tech graduates, a swelling workforce, and booming exports of IT related services. The companies are able to get the same amount of work or more for considerably less money. A graduate from an India tech school earns a great deal less than a graduate from a U.S. which makes an even stronger case for large corporations outsourcing.

Education may not be the most important factor in economic performance, but that doesn’t change the fact that the U.S. is falling behind in that regard. Our education system is over-rated in my opinion and the example at the bottom of this article reinforces that theory. The first question is why is education not taken as seriously in this country? I guess part of the answer to that question is the absolute necessity of an education in over country in order to escape a poverty stricken past. This may be an extreme example, but true is some cases. The need to learn in this country isn’t really as pressing on today’s youth. The strength of our economy should not being an excuse for laziness in terms of education. I think that the strength of international forces in the economy is becoming more and more of an issue to the U.S. economy and education system. If the U.S. wants to keep pace with other countries they will rework some parts of our education system and begin to take education as a whole more seriously. The strength of our economy in the future is directly related to how we chose to educate our future leaders of tomorrow and this responsibility should not be taken lightly.

Comments for this entry have been closed
Return to top