Trainability is not the same as education

The reason we do not have to wait for significantly different educational systems is that today's  workers around the world can be trained on the job to achieve much higher labor productivity, even to a level close to global best practice in many cases. Twelve years ago, the biggest economic question was the performance of the U.S. economy relative to Japan and Germany. The conventional wisdom at the time was that the U. S. economy was going down the drain. The poor skills of the U.S. workforce were often cited as one of the primary reasons. U.S. managers complained regularly about the inability to hire workers who could read, write, and do arithmetic. U.S. high school students scored poorly on standardized international tests in science and mathematics. Public officials in Europe and Japan labeled the U.S. workforce lazy and incompetent.

It came as a great surprise to everybody, both in the United States and abroad, when our early work showed in industry after industry that the U.S. workforce achieved higher labor productivity than anyplace else on earth. The only exceptions were a handful of manufacturing sectors where Japan had the global lead. How could this be? Well, of course other factors such as more capital, or more technology, or bigger scale, or better organization of operations could be offsetting a labor disadvantage in the United States. In general, the United States did not apply more capital, did not have access to different technology, and did not build operations on a larger scale. The primary U.S.  advantage was better organization of operations. These better organizations are usually more demanding of workers. The U.S. workforce was able to work in ways that were more productive.

The most  important evidence came from the performance of Japanese automotive transplant factories in the United States. Japan was, and remains, the global leader in automotive assembly productivity. The factories of Toyota, Honda, and Nissan in the United States achieve about 95 percent of the productivity they achieve in Japan. They achieve this productivity by training U.S. workers on the job. They do the same thing with Japanese workers in Japan. They estimate the additional training required by the U.S. workforce causes them about 5 percent productivity penalty.

Of course, these factories are virtually all in the U.S. Midwest and not in big industrial cities. Maybe the Japanese got the best of the workers in predominantly rural areas. However, these workers were the product of the U.S. educational system. Any lack in their education did not penalize their trainability to any significant extent. The Japanese probably would not have had as good an experience with workers educated in the school system in the urban core of some large cities. However, when the transplant evidence is combined with the finding that U.S. workers across the board almost always work at the highest labor productivity in the world, we have the basis for a strong conclusion. Whatever shortcomings the U.S. education system may have, it is not an important constraint on U.S. relative economic performance.

The trainability of workforces is not a constraint today on economic development in poor countries. Brazil, Russia, and India could all clearly double their GDPs per capita with today's  workforce. The disappointing returns on investment in education in poor countries are documented by statistical evidence in chapter 4 of William Easterly's recent book The Elusive Quest for Growth. The evidence on which I rely comes from studying why labor productivity in most sectors in poor countries is so low. A whole variety of factors explain this, without ever having to get to the trainability of the workforce. Of course, these other factors could just be masking a constraint from labor force trainability. If countries fixed all the other factors, then they still might not make significant economic progress because of labor force trainability.

Uneducated Workers can achieve best practice and educated ones often do not

Fortunately, there are tests of whether this is true. In every economic sector in developing countries, there is reasonably wide range of productivity across firms. Most important evidence is from firms that achieve productivity much higher than industry average. Sometimes this productivity is even close to global best practice. These examples indicate that the local labor forces are capable of achieving much higher labor productivity than they do now. However, the other factors have to be fixed for them to do so.

Perhaps the most important evidence of this conclusion came from the United States and Brazil. In the U.S. construction industry, there are workers equivalent to many Brazilian construction workers. In Huston, Texas, illiterate agricultural workers from Mexico not speaking any English are achieving best practice labor productivity in housing construction. In Brazil itself, the two leading private retail banks are owned locally and achieve near global best practice productivity. Similarly, a Honda transplant factory there is near Japanese performance  levels. Carrefour, the French hypermarket, achieves in Brazil about 90 percent of its productivity at home.

In Russia, the evidence is somewhat different. Education ministries and the World Bank would love for poor countries to have an educational system as good as Russia's. However, in  industry after industry the productivity of whole sectors and individual firms is uniformly low. The productivity is so low that Russia's overall GDP per capita is significantly below that of Brazil. Brazil clearly has an educational system inferior to that of Russia. Russia shows for sure  that economic policies are more important than education for economic performance.

Even in India, there are a couple of examples that help to prove the point. India's labor productivity overall is about 8 percent of the United State's. However, India got the policies roughly right for one sector, automotive assembly. In this sector, India allowed foreign direct investment by one of the smaller Japanese automotive companies, Suzuki. The  resulting joint venture company, Maruti, has 60 percent of the car market in India and achieves about 55 percent of U.S. automotive assembly labor productivity. India stopped the licensing restriction on automotive companies in 1993. Since then,  many of the world's best automotive companies have invested in India, including Honda, Daimler-Chrysler, Ford, General Motors, and Hyundai. Since 1993, labor productivity in automotive assembly has grown 20 percent per year. This performance is being achieved with the Indian workforce and the educational system as it now exists. Even the illiterate wheat and dairy farmers are doing the right thing economically. They are investing in tractors and combining both wheat and dairy farming in the same household. Given rural wage rates, it is not economic for them to invest in harvesting combines or to do wheat or dairy farming alone.

Finally, I should mention Korea. Korea is neither rich nor poor. Korea is one of the few countries in the middle. Since 1960, Korea has made a massive investment in education. In 1960, 80 percent of the Korean workforce had no education beyond primary school. In 1995,  80 percent of the Korean workforce had education beyond primary school. The corresponding figure  for the United States was 90 percent. Thus, it doesn't look like education would be a constraint for Korean labor productivity. However, the Korean workforce achieves only about 35 percent of U.S. labor productivity. Korea achieves 50 percent of U.S. GDP per capita primarily by working 40 percent more hours. That's not the effect education advocates have in mind.

what is the role of education?

This book is not about education. However, I'll add here a few speculative thoughts. After reading Diane Ravitch's book Left Back, A Century of Failed School Reforms, about the U.S. education system, I was left wondering if anybody knew what education was really about. I have begun to suspect that economic development causes education to develop even if governments don't force it as Korea has done. After all, that's how education got started. When we were all hunters and gatherers 10,000 years ago, we did not have time for education. We did not even have time to wonder much about anything besides finding enough food. Only when our productivity for food production increased did we have time for other things.

Jared Diamond gives us a good explanation of this transition. We wanted to do things that would make us richer and more powerful. We also wanted to know things about which we were simply curious. Who are we? What are we? Where are we?  why are we? Education arose naturally as successive generations wanted to learn what those going before them had learned and thought. We began to make the trade-off between simply working more now and learning more to work in more valuable ways later. We also wanted to learn more to satisfy our curiosity. The more productive we became, the more we had time for education. We demanded that education.

No doubt, as education increased, our productivity potential increased. If our policies pushed us to achieve that potential, then we could afford more education. As I will come back to in chapter 11, this increasingly high level of education probably is necessary, but not sufficient, for the complex political  systems necessary for advanced economic performance. It's possible that poor countries today will not get out of their poverty traps without political changes. Those political changes may only be possible with broader education. The point is, however, that education is not a constraint on the ability of today's workforces to achieve substantial productivity improvement around the world. It's the other way around. Constraints on productivity improvement are the reason education is not developing faster around the world.

Look for our next newsletter. We will include Education an Democracy and The Roles of Education from Chapter 11.

William W. Lewis was a partner at McKinsey & Company for twenty years and the Funding Director of the McKinsey Global Institute. He held several policy making positions in the U.S. Departments of Defense and Energy and also served in the World Bank for  four years earlier in his career. His work has appeared in the Wall Street Journal, The New York Times, and the Economist.