Test Scores and Economic Performance

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. Of course, an economy's productivity depends not only or even mainly on schooling, but also on its physical capital stock, institutions and laws, and various other variables as well. Even regarding human capital, however, the United States does better than suggested by its rankings in international tests.

Frequently cited are the results of tests in 2003 of 15-year-olds in math, sciences and reading. The U.S. ranked about 25th in math, 20th in science, and was above average in reading out of the about 40 countries that participated. I believe these results correctly reflect that most American elementary and high schools are much less challenging in math and science than are schools in Finland, Japanese, Hong Kong, and other high scoring countries. However, the creation of valuable human capital for an economy depends on much more than is measured by tests of fifteen year olds in different subjects.

The philosophy behind U.S. education is to build up, so that later levels of schooling are more challenging than earlier levels. This means that more is expected of students at college than was expected of them in the high schools they attended, even though both high schools and colleges vary greatly in their degree of difficulty. On the other hand, secondary schools in most countries that rank high in these test scores give a lot of home work and expect a lot, while their colleges are often easier. In Japan, for example, getting into the top universities is very hard, but these universities are easy compared to the secondary schools that their students attended. As it were, Japanese students rest in college after the exhausting demands of their high schools. Japan is an extreme version of a build down system, but less extreme versions are found in many of the other participating countries. This difference in education approach implies that a more relevant international comparison of the production of human capital would be to test not teenagers but young adults, say at age 22. The U. S. would probably still perform below average in math and science, and might not excel in reading, but the relative performance of older Americans would be, I expect, considerably higher than that of fifteen or eighteen year olds.

Another important consideration in international comparisons of human capital is that a larger fraction of Americans than is common in other countries continue their learning after high school at junior colleges, trade schools, non-profit four-year colleges and universities, for-profit online education and universities, such as the University of Phoenix, adult education classes, and in other ways. This vast array of learning opportunities allows young (and not so young!) persons to pick out programs that suit them, and to change where and what they are studying if they are dissatisfied. In most other countries, later as well as earlier schooling is not flexible. This again suggests that the human capital of Americans would look better in international comparisons if comparisons were not of teenagers but of adults in their late twenties and thirties.

U.S. education in junior colleges, many four-year colleges, at trade schools, and for-profit universities is more oriented toward improving job-relevant skills than is common in post secondary education in other countries. I refer not only to schools that teach how to drive a truck, use a computer, or cut and shape hair, but also to junior and four year colleges that provide instruction in landscape gardening, bookkeeping, and other practical subjects. This type of education may not help students know much about the world at large, but it does raise their productivity at work.

Another factor is intangible, but nevertheless is relevant in helping American men and women become innovative at work and in other parts of life. American schools are less oriented toward rote teaching than are schools in many other countries, and they are more oriented toward giving students practice in thinking through issues and expressing themselves in discussions. Japan and the United States are outliers at opposite ends among rich countries in the degree of emphasis that schools place on thinking for oneself rather than memorizing information. The United States may go too far in its emphasis on "self expression" at the expense of teaching valuable knowledge and skills, but still international tests of subject matter knowledge, such as the 2003 tests, do not even try to capture originality and related important aspects of human capital accumulation.

Note in this regard that despite the mediocre record on international tests, American trained scientists do extremely well in garnering Nobel prizes and other international awards. American CEO's and investment bankers are ranked very high in the international business world for their energy and creativity, which is why many foreign companies have chosen Americans to head their operations. American workers also rank high when international businesses rank the quality of the workforce in different countries.

To be sure, a significant number of prominent "American" scientists and some business leaders were born abroad and immigrated to this country. This attraction to skilled immigrants must be taken into account in assessing the quality of the human capital that enters the American economy, although it may reflect unfavorably on the quality of education provided to American students in math and the sciences. Still, if America allows India and other countries to pay the cost of training many of the engineers and other skilled workers who end up in the American economy, that is a pretty effective human capital production "system" when considered in its totality.

I will try to address a few of the various interesting comments.

The comments take very different positions on the importance of mathematics in a modern economy. A couple of things are pretty clear: most employees do not need a lot of math, but math is becoming more valuable for a larger fraction of the tasks in more advanced economies. For example, it is difficult to use regression analysis in ways that illuminates without understanding some of the basic mathematical concepts that lie behind regressions. The statistical theory of estimation need not be mastered, but someone limited to only cookbook knowledge of regressions can make serious mistakes.

The fear of math among many Americans is common, and this fear prevents them from reading simple figures, analyzing say the relation between housing prices and interest rates, understanding probabilities in medical and other areas, etc. Still, many jobs are designed such that only very basic math, even only arithmetic, is required.

The advantage of the American "build up" learning system is that people can more readily change their minds about what interests them as they try out different programs. It also implies that there are second and third chances as students can decide their interests are different than what they had believed, and workers can shift to entirely new lines of work. That flexibility is less common in the learning structures of many other countries.

Obviously, memorization is important to learning, and it is said, for example, that first year medical school is mainly about memorizing various pieces of information about the human body. Perhaps too I exaggerated the importance of rote learning in the Japanese and other educational systems. But Japanese businessmen and scientists frequently complain about an excessive emphasis on rote learning. My experience with Japanese students in the U.S. is that especially at the beginning they have difficulty in thinking on their own, and rely too much on what their teachers told them.

I believe schools systems should establish minimum standards for graduation that are meaningful, but not very difficult. Then individual high schools can top that up to an extent determined by their student bodies.

As I argued in previous posts, inequality in American earnings has grown greatly during past 25 years, in large measure because of increasing demands in a modern economy for workers who have considerable knowledge. Graduates of elite universities have benefited the most from the greater emphasis on knowledge, but all groups that invest in knowledge and information do better in modern knowledge-based economies. I believe America has adopted better than most other rich countries to these changes, in part because of the flexibility in its learning system, and in part because of its emphasis on lifelong learning.