The Washington Post’s Valerie Strauss asserted yesterday that “public education is a civic institution” and laments that it is seldom talked about as such (kindly citing our upcoming Cloning “Superman” event in the process).

Certainly the way children are educated can have a powerful impact on the kind of society they go on to build. And there are many social goals on which Americans strongly agree: that schools should prepare children for the responsibilities of citizenship as much as for success in private life; that they should encourage harmonious relations among people of different backgrounds (or at least not foment conflict); and that they should ensure that every child, regardless of background, has access to a quality education.

But does anyone seriously believe that our existing school system is doing a satisfactory job in any of these areas? I doubt that Ms. Strauss herself believes that, and suspect that she was merely expressing the view that our education system should do these things rather than claiming that it already was. Consider the hundreds of community conflicts around the country documented by my colleague Neal McCluskey as having been caused by public schools in a single year. Consider, too, the literature review performed by the University of Arkansas’ Patrick Wolf showing that the civic outcomes of freely chosen (usually private) schools are consistently superior to those of public schools, after controlling for differences in student and family background. And one needn’t have seen the documentary Waiting for Superman to realize that public schools have been failing far too many children, especially poor and minority children, for far too long.

If we are to remedy these profound shortcoming in American education, our best hope is to set aside our preconceptions about what kind of school systems should produce the social goods we seek, and instead ask which systems actually do produce them.

Having reviewed the worldwide econometric literature of the past 25 years, I’ve found that it is the most marketlike education systems that have consistently done the best job of serving disadvantaged children (indeed, all children) both here and abroad. Wolf’s literature review also favors private schools in their civic outcomes. And when people can get the sort of education they value for their own children without being compelled to impose their preferences on their neighbors, the conflicts caused by public schooling are avoided. Even with regard to meaningful integration among children of different racial and ethnic backgrounds, the private education sector performs as well or better than the state sector.

If Ms. Strauss or anyone else has compelling evidence to the contrary, I’ll be interested to hear of it. And if she or anyone else would like to know what the social impact of decades of private school choice has been in a communitarian nation like Sweden, they’re welcome to come to Cato and ask Peje Emilsson on the 28th of this month.