[anamax]

> But school choice is somewhat illusory when there are limited and largely fixed numbers of places and it honestly has real, significant problems.

Since we can't have an infinite number of schools, any scheme will have a limited number of schools.

However, I think that you mean something different, that there can't be "enough" school choices. Since a school can be a room with a few people in it, and we have lots of rooms, it's unclear why school choice is necessarily limited in any relevant way.

[eftpotrm]

Education tends to be funded per head, but with strong network effects - a larger school can have a wider range of facilities because they can defray the cost per student over a larger base to cover the lower-interest offerings.

Also, a school running heavily under capacity is _very_ expensive and likely to spiral down, rather quickly, for simple financial reasons.

We're seeing an introduction of a new 'free schools' policy in the UK at present. I'm not in the least saying every last school should have tight government control, but the side-effect of this policy as currently implemented is the impoverishment of existing facilities, to the detriment of their pupils.

Infinite choice is clearly not possible. Neither is the capacity for all students to get their first choice, for physical infrastructure reasons if nothing else. Hence total choice isn't deliverable, and any degree of choice is almost guaranteed to leave some schools over-occupied and needing quick (expensive) hiring and building to cover the gap, while others have too many facilities and resources for their per-head income and now have a financial black hole which, combined with the social stigma of not being the 'preferred school', tends to drag them further down.

Note I'm talking here from the perspective of how school choice works in England, as that's what I know. I'm not entirely anti the princple at all, but as I hope I've shown it does have some significant undesirable side-effects that can both increase cost and at best drive up the gap between best and worst by pushing at both ends of the spectrum - it doesn't just improve the top.

[anamax]

> a larger school can have a wider range of facilities because they can defray the cost per student over a larger base to cover the lower-interest offerings.

Yes, a given smaller school can't have as many different things as a large school, but that doesn't imply that the range of things at a set of small schools is necessarily smaller than the range at a large school.

When 10 kids at a school with 400 kids want something, it may not happen. When 10 kids at a school with 100 want something, it's more likely.

That's the advantage of choice - those 10 kids can "gang up" on a small school if they get to choose where to go.

> significant undesirable side-effects that ... at best drive up the gap between best and worst by pushing at both ends of the spectrum - it doesn't just improve the top.

Just improving the top would increase the gap, which you seem to think is bad.

I'm not convinced that choice hurts the bottom. I think that it exposes the real bottom, the folks who drag down the average. When they're split out, they're more obvious.

The big advantage of separating them is that then they don't drag down other folks.

There are lots of poor parents who do all that they can to keep their kids away from trouble. Why are we forcing them to send their kids to school with trouble?

[eftpotrm]

Your example of 10/400 v 10/100 implies early specialisation if that's to be a realistic scenario, which I confess I'm not a great fan of. If you'd found me at 11, or even 14, I was near enough top of the chart on everything bar sports. Plenty of others were in a similar position to me, or would have been equally flat at a different level. Early specialisation forces pupils to close off options before they may realistically be ready to.

I have no problem with the gap per se - I went to a state funded selective school and I'm perfectly happy that that sort of school has a place in the system. I believe I've illustrated though how school choice as implemented in Britain necessarily impoverishes the schools perceived as poorer - through the inefficiencies and excess capacity it requires to operate while giving anything like true choice, school choice gives less popular schools higher per-pupil expenses for worse opportunities and outcomes. Someone's kids have to go there, they're paying the same taxes as everyone else to fund them, but they're getting a rotten deal.

Like I said in my first contribution - I don't think there is a perfect solution and school choice may well be the least worst option. It is not a panacea though, and we should be honest in appraising its failings.

[anamax]

> Your example of 10/400 v 10/100 implies early specialisation if that's to be a realistic scenario, which I confess I'm not a great fan of.

Young kids have interests. They change over time, but they have interests.

> Early specialisation forces pupils to close off options before they may realistically be ready to.

You assume too much.

> I believe I've illustrated though how school choice as implemented in Britain necessarily impoverishes the schools perceived as poorer

Actually, you've proposed a mechanism. Even if we assume that things always work that way (and they don't), there should be nothing keeping kids at those poorer schools, so what's the problem?

> Someone's kids have to go there

Why?

If no one wants to go there, why keep it open?

> It is not a panacea though,

Strawman.

> we should be honest in appraising its failings.

We should apply that to all schemes.