Vivek Wadhwa, writing in Bloomberg BusinessWeek, echoes a point Tino Sanandaji made a few weeks ago:
Hal Salzman, a professor at Rutgers’ John J. Heidrich Center for Workforce Development, debunked myths about these in a May 2008 article in Nature magazine. Salzman noted that international tests use different sampling criteria from country to country, so we’re not always comparing apples to apples. As well, the tests compare select populations of small countries such as Singapore and Finland, which each have about 5 million people, with the U.S., which has 310 million. These countries achieve the top rankings on the PISA list. Compare these countries to similar-sized U.S. states, however, and you find that some of those states, including Massachusetts (population 6.5 million), produce the top students. Additionally, we’re comparing America’s diverse population—which includes disadvantaged minorities and unskilled immigrants with little education—with the homogeneous populations of countries like Finland, Japan, and New Zealand. [Emphasis added.]
That is the Tino point.
Much is made of the PISA test scores and rankings, but the international differences are actually quite small. Most of the U.S. ranking lags are not even statistically significant. The U.S. falls in the second rank on some measures and into the first on others. It produces more highest-performing students in science and reading than any other country does; in mathematics, it is second only to Japan. Moreover, one has to ask what the test results actually mean in the real world. Do high PISA rankings make students more likely to invent the next iPad? Google (GOOG)? I don’t think so.
There is the germ of an important point here. In a provocative paper on labor market balances, Richard Freeman floated the following idea:
How can low income countries with few scientists and engineers relative to their work forces compete in high tech?
These countries have moved to the technological frontier because success in high tech depends on the absolute number of scientists and engineers rather than on the relative number of S&E workers to the workforce. It isn’t how many engineers perperson that produces a technological breakthrough as much as the total number ofengineers working on the problem. Put differently, there is an economy of scale in R&D and innovation that enables large populous countries to reach the scientific and technological frontier. China and India can have a large footprint in high tech because they will have many highly educated scientists and engineers, not because they approach the advanced countries in S&E workers per capita. I have called the process of moving up the technological ladder by educating large numbers as “human resourceleapfrogging” since it uses human resources to leapfrog comparative advantage from low tech to high tech sectors, contrary to the assumption of the North-South model.
Freeman’s point is that this will prove advantageous to China and India, but it also applies to the United States relative to small homogeneous countries with high average levels of educational performance: even if the Shanghainese or Singaporeans or Finns perform well on average, the absolute number of educational U.S. high-fliers is always going to be very high, for a variety of reasons. And these super-skilled workers concentrate in a handful of metropolitan areas, which become hotbeds of S&E-driven entrepreneurship.
Now let’s revisit Tino, who offers an important insight: when we (crudely) correct for demographic factors, weak U.S. performance becomes very strong U.S. performance.
I pushed back against Tino’s point, arguing that the U.S. was still performing worse than it should given our affluence. Tino very kindly replied here:
It turns out that the U.S educational advantage is approximately what the U.S economic advantage would predict.
I can’t fault Mr. Salam for wanting more output for the input American tax payers are spending. However my analysis suggests that unlike conventional wisdom on the political right, public education spending is not just wasted. It appears successful in buying America better test scores, for each given demographic group, albeit of course with diminishing marginal returns.
I still think we’re talking past each other. When I express disappointment or dismay over our relative performance, it’s not because I think rigid, unionized public education systems in western Europe are outperforming U.S. schools. Rather, it is because the U.S. has experience in overcoming Baumol’s cost disease in a number of domains — the retail sector comes to mind — and yet we’ve failed to replicate this feat in education.
It is widely understood that the U.S. retail sector is more productive than the retail sector in most of western Europe, though the gap has been closing in recent years. Part of our retail advantage lies in the fact that U.S. labor markets are less rigid than labor markets in most of western Europe, and because U.S. firms tend to have higher levels of capital productivity. (These phenomena are related, obviously.)
The U.S. public sector, in contrast, is far less productive than the public sector in most of northern Europe, as we discussed in November:
According to the World Economic Forum’s GCR, the U.S. is ranked 68th in the world in terms of wastefulness of public spending. That is, 67 of the 139 countries in the index are doing a better job of channeling public dollars into productive activities, including Singapore, Rwanda, and Qatar at the top of the list and Australia, New Zealand, and Canada among our large English-speaking peers. Not surprisingly, to those of us who follow work on the quality and cost-effectiveness of public spending across countries, the northern European market democracies perform well, with Sweden at 12, Finland at 14, Denmark at 16, and Norway at 20. Switzerland is at 9, Holland is at 17, Germany is at 33. There are no real shockers on the list.
If the U.S. public sector were as productive as the Canadian public sector at 32 — not an unreasonable goal, I should think — we’d presumably have more resources to direct towards public education for non-native children and children from minority backgrounds. And if our education sector were organized around more responsive, market-like lines, it is at least possible that productivity gains would see the pattern we’ve seen in retail and other service industries.
So again, I think we’re underperforming in light of our core economic strengths. But I do think Tino is telling us something important: the U.S. has nothing to learn from Finland.