Why Do 12" Tonearms Cost So Much More Than 9"?

As regards your intimation that a pivoted tonearm per se generates a skating force, that is not really correct, if you consider that an underhung tonearm, at the one point on the surface of the LP where the stylus is tangent to the groove, generates no skating force, because no headshell offset angle.  It is also not correct for the class of overhung pivoted tonearms that maintain tangency to the groove by virtue of altering the position of the pivot, a la the Schroeder LT.  All other overhung pivoted tonearms do generate a skating force by virtue of their overhanging the spindle, combined with their having a headshell offset angle that is not zero.

What "measured error" are you talking about?

Cleeds.
I gave you an example in my post.
I thought it was obvious my statements were just related to maths and I was not discussing any actual tonearms or tracking errors as related to turntables per se.

Lewm.

I say that it does not seem to be borne out in practice just from my own experience with 9", 10" and 12" tonearms.
I cannot prove it one way or the other and similar to yourself I truly do not care overmuch.....

Cleeds.
Just to clarify my thoughts maybe a little better and to follow where I believe Lewm is heading.

Maths would seem to indicate that any INITIAL error in setup on a longer arm would result in a greater playback error.
So a basic change in vta by eye just looking at, say making the tonearm more tail down would result in more change at the stylus the longer the arm is.

But as I have said from practice this just does not seem to occur.

And this maybe because I am missing something ?

The Reed 5T and the Schroder LT are very similar arms. They work on the same principle which you can review here  https://www.youtube.com/watch?v=oIoHvCE9F0I
The Reed is animated while the Schroder is entirely passive. Both arms are pivoted. They do not suffer from horrendous horizontal effective mass. Both arms are dead straight and if set up correctly do not generate any skating force. They have to be right on the tangent, the turntable has to be dead level and the record has to be dead flat. All of these parameters are unlikely to be met so there will always be a little skating force one way or the other but magnitudes lower than any offset arm. 
The Schroder arm is brilliant. It uses groove friction to operate it's horizontal pivot platform and a magnet to maintain the correct position. It is very hard to describe but you can see the patent here   https://patents.google.com/patent/US8576687B1/en.
Lewm, both tonearms maintain tangency within 0.05 minutes which is pretty incredible. Both will easily out perform any offset arm not to mention any straight line tracker of any type because cartridges do not have to suffer under the high horizontal effective masses these arms have and are subject to very little skating force.
I am leaning towards the Schroder LT. It is 1/2 the price and just an elegant design. The Reed has one major advantage. It has removable arm wands like Graham arms and you can get them in a range of effective masses so you can tailor it to any cartridge. Switching from one cartridge to another is a breeze. Obviously the Schroder is going to be more durable in the long run. 
Uberwaltz, you are not missing anything. 12" arms are less critical of set up. We all agree they have less tracking error I hope. This means that if a cartridge is misaligned by 1 degree this will have less of an effect on a 12 inch arm relative to a 9 inch arm. Frankly, you either get the cartridge exactly where you want it or let someone else do the job who can. There is no magic here. Anyone with two functional hands and two functional eyes can do it. Just make sure the stylus guard is in place and save the beer for after the job is done unless you have an essential tremor than drink the beer first!  

mijostyn

The Reed 5T and the Schroder LT are very similar arms ... and if set up correctly do not generate any skating force. They have to be right on the tangent, the turntable has to be dead level and the record has to be dead flat. All of these parameters are unlikely to be met so there will always be a little skating force one way or the other but magnitudes lower than any offset arm.

I do not have firsthand experience with either of those arms and have not even seen them up close. However, any pivoted arm with a fixed horizontal pivot will generate some skating force, even though an underhung arm will show much less than any arm with an offset.

By way of real-world example, consider the lowly Garrard Zero100. Its zero-tracking-error "pantograph"- type pickup arm included an antiskate mechanism. Even then, it was understood that skating force was a byproduct of both offset and the pivoted arm itself.

I continue to be surprised at the level of confusion that exists here regarding pickup arm geometry, because all the math has already been calculated. Here’s a fair treatment on skating force that includes some excellent references.