Well, who do you think should run Rochester's schools?

The Rochester Business Alliance wants the mayor to do it, and that has set off the predictable protests from School Board members, who say it's an assault on democracy.

It isn't, of course; we elect the School Board. But we also elect the mayor. And a lot more of us vote for mayor than for School Board.

I don't think the board members' objections are purely self-serving; they're certainly not in that job for the money, the power, or the prestige.

Still, it's time to take a good hard look at this.

"Mayoral control" could take any number of shapes: the mayor could appoint the superintendent and we could continue to elect a School Board (similar to the way the mayor appoints the police chief and we elect City Council). The mayor could appoint some or all of the School Board members.

The school district could become a department of the city, and the School Board could cease to exist.

In almost any iteration, the mayor would be in charge of the schools.

That will appeal to a lot of people. The question, of course, is what's best for the children.

The Business Alliance commissioned the Center for Governmental Research to look at mayoral control - specifically, where it's been used and what the result has been.

In its response, CGR didn't offer a recommendation for Rochester. It simply compiled information from the handful of cities where mayors are in charge of their school districts. And it listed the arguments that have been given for and against mayoral control.

Frankly, you don't learn much that's helpful.

The Business Alliance is clinging to a statement in CGR's memorandum saying that mayoral control "is linked to increases in student achievement in the elementary grades and improved performance in the lowest-performing schools."

But none of the studies drawing those links were scientific: comparing cities with exactly the same circumstances.

And Rochester - like many cities - has been seeing improvement in the elementary grades - without mayoral control.

In addition, as the CGR report notes, mayoral control is "relatively new." Researchers know that you can get temporary increases in children's achievement simply by doing something new.

What we all want is long-term improvement - and improvement in high schools. And CGR found little indication that things were better in the high schools when the mayor was in charge.

In short, there seems little evidence that giving the mayor control of the school district will help students learn more.

The folks who want mayoral control cite other reasons, too: better "accountability." "Improved financial management." What that means, I assume, is that the supporters think City Hall is run better than the school district. This is an argument the district can't win, since - barring financial scandal - the proof is in things like test scores and the graduation rate.

Nobody's charging that Rochester has a high rate of violent crime because City Hall isn't well run, of course. Everyone seems to understand that the mayor can't control people's behavior and that he can't do much, on his own, about the poverty that breeds that violence.

School districts, on the other hand, are supposed to overcome the influence of the poverty and the crime.

I've been opposed to mayoral control in the past, but I'm beginning to bend a bit. For one thing, it would eliminate the argument that City Hall could do a better job. At some point, we've got to stop blaming everybody and everything except the real culprit: the city's concentrated poverty.

I'd love to see a mayor say to the business community: "I've tried my best. And unless we do something about the poverty, we can't do anything about the children's education."

Then, at last, we might stop looking for a scapegoat and start looking for a solution.