Here's the first:
The per-capita funding to the state comprehensive attended by a good friend of mine was four times that of the state comprehensive school I attended. The schools were of a similar size, though mine was in a (relatively) affluent area, hers next to a sink estate. Guess which one provided the best exam results. Clearly funding is not the key issue.
This is anecdotal evidence, somewhat like arguing: my granny smoked forty a day for forty years, and she never got lung cancer, but Auntie Betty, who never smoked, did, so when it comes to lung cancer, smoking "is not the key issue".
Funding may not be the only issue, of course, when it comes to education. But then when did I say it was? I am quite sure that peer group, home background etc. all play a very significant part too. And the system I suggested is intended to affect these things too, by ensuring that we don't get all the posh kids at one school, all the working class kids at another.
The suggestion that levels of funding have little to do with the quality of eduction provided by a school is often made by defenders of private schools. But it is ludicrous claim. If it was true... well, let's save ourselves a mint by slashing the funding of state schools by half!
Jonathan, you also said:
I think we are agreed that those currently paying for private education or buying houses in the catchment areas for good schools are those most interested in a good education for their children.
Good fu**ing grief. Is this really what you meant to say?
Jonathan also says:
Others have talked about possible gaming of the system - one more anecdote. Another friend, at university this time, was distinctive as a first year who owned both a new car and a mobile phone (at a time when such phones were rare amongst the working population), yet still drew a full means tested grant (her parents were divorced and she took the expedient of declaring the income of only her non-working mother). And she had received, of course, the benefit of private schooling.
Yep, more anecdotes (a very "Daily Mail" style of argument, this). All systems are "played" to some extent. What we want is to minimize the playing.
Now notice that the system I suggest actually has the advantage of minimizing any playing of the system. In my system, the voucher's value does not depend on the parent's income. It depends on the income of all the other parents.
Let's use the tax system to determine income. To fiddle my system, parents will have to fiddle their taxes. Substantially. And collectively. For, there is no incentive for any parent to do so in isolation. If I fiddle my taxes to make myself look poorer, that won't effect the value of my voucher. Or the vouchers of the kids attending my daughter's school. So what's my incentive to fiddle, then?
There is none!
Rather than my system "not being thought through", it seems that this particular criticism is not well thought through.
As for schools not being able to grow, actually, they do so all the time. And companies setting up schools in this system can of course design for growth (like building a house in such a way that it can easily be extended). Not much of a problem I'd say.