The latest issue of The Economist praises president elect Obama’s pick of Arne Duncan to lead the department of education. In particular, it predicts that “Mr Duncan may restore the spirit of co-operation that helped pass NCLB in 2001.”

But bi-partisanship is not intrinsically desirable. It is only good when it results in good policies. NCLB has not proven itself a good policy. There are four nationally representative sources of information on academic outcomes over the last 6 to 10 years: America’s own National Assessment of Educational Progress (NAEP), and the international testing programs: PISA, PIRLS, and TIMSS. NAEP shows some gains since NCLB was passed, but they are, for the most part, smaller than those that occurred in the years immediately preceding the law. TIMSS shows gains in math over the past decade, but stagnation in science, and much of the math gain seems to have preceded NCLB — and we’re still just in the middle of the pack among the top 30 rich countries despite being the second highest spender among them. Both PISA and PIRLS show stagnation or decline across subjects and grades.

These lackluster results have come at an additional cost of perhaps $100 billion over our already world-leading per-pupil spending. I don’t understand why an “economist” would think more of this kind of bipartisanship desirable.