So, I got curious and did some rough calculations:

Karen Coyle <[log in to unmask]> wrote:
> What I haven't seen discussed here is the frequency with which this
> data is needed.

The NCSU catalog has 2125036 bibliographic records.
43273 have brackets in 260$a but NOT in the start position.

(on the theory that if there's a bracket in the start position, the city 
itself was supplied.  The higher jurisdiction wasn't supplied alone. 
Give or take whatever error rate might naturally exist in the data ... 
In case you are also curious, 404391 records have brackets of any sort 
in the 260$a.)

so 2.04% of the NCSU records have higher jurisdiction supplied.

Erin Stalberg
Head, Metadata and Cataloging
North Carolina State University Libraries
[log in to unmask]