It’s like the bad penny that keeps appearing, only it costs hundreds of millions of dollars. The city teachers union has begun pushing a new property-tax proposal tied to a union employment program. Everyone would be better off if they just stuck to teaching kids. The union’s proposal, announced with great fanfare by United Federation of Teachers boss Michael Mulgrew, is to increase property-tax receipts by placing an added burden on housing owners who do not use their city apartment as their primary residence. The revenues, which they estimate would add up to some $900 million annually, would be used to fund a new low cap on class sizes in grades K-3. The idea is getting Albany legislators to lower current class size restrictions, which are built into the union-negotiated contract — that is, using the Legislature to force new teacher hiring. Of course, the unions do not have strong opinions on the specific tax used. Any tax will do, as long as it can funnel money into labor’s coffers. So instead of focusing on the levy itself, it is more useful to focus on the true objectives: lowering class sizes in order to hire many new (dues-paying) educators. This is an idea whose time came, and went, many years ago. In the decade and a half that has passed since the beginning of the class-size-reduction craze, virtually everybody concerned with student achievement has come to realize that further reductions are not a wise way to spend school money. Support can be tracked back to a single experiment in the mid-1980s, Project STAR in Tennessee. While this was a poorly implemented experiment, it showed some achievement gains in small kindergartens. Nobody has wanted to repeat that experiment for fear of a different answer. Nationwide, pupil-teacher ratios for the country fell from 17.8 in 1995 to 15.8 in 2008. With the recession, they rose back to 16.4 in 2010 — but this simply reflected the fact that this was the least harmful way to deal with the budget shortfalls. For New York City, the equivalent 2010 number is 14.5. (Class-size numbers are typically higher than pupil-to-teacher ratios, but that’s a subject for another day.) Nobody has shown that the substantial class-size reductions of the past 15 years have paid off in terms of student achievement. Instead, the two main effects of past class-size reduction have been more teachers and more expensive schools. Education research is essentially unanimous: The effectiveness of the teacher in the classroom is far, far more important than how many students are in the classroom. But this is not the message that the union wants to hear, because it would involve evaluating teachers and making personnel decisions based on the quality of the work they do. Interestingly, before calling for this new tax to reduce class size, the union was in Albany to stop any attempts to improve the evaluation of teachers. There is substantial evidence of the value of top teachers: They improve students’ skills and boost their students’ labor-market outcomes. Estimates by Raj Chetty, John Friedman and Jonah Rockoff based on New York City teachers and students show that a top teacher can add a quarter of a million dollars to the future incomes across her class. And she does this every year. Similar estimates that I have done show the other end of the picture: A low-quality teacher subtracts a similar amount or more from her students. Instead of acknowledging this, the union is fighting any changes that might produce more thoughtful teacher evaluation and retention decisions. This isn’t an education policy push by the union; it’s a political one. By going back to the tried-and-true rhetoric of class-size reduction, the union is distracting attention from policies that would trouble their members — and really help kids. Hanushek is a senior fellow at the Hoover Institution of Stanford University and the co-author of “Endangering Prosperity: A Global View of the American School.