My UCLA colleague Sandy Black wrote a great Harvard thesis. Google Scholar says that the paper has already earned 292 cites; Do Better Schools Matter? Parental Valuation of Elementary Education, SE Black - Quarterly Journal of Economics, 1999. In a nutshell, she looked at home price differentials for homes physically located very close to each other where the two homes were located in different school districts. She interpreted the equilibrium price differences of homes (quality adjusted) as representing how much the marginal household values good schools.
With this in the back of my mind, I returned to my childhood. I am a 1984 graduate of Scarsdale High School. While I am bum, I have been told that Scarsdale is a great school district. It certainly has expensive homes -- http://en.wikipedia.org/wiki/Scarsdale,_New_York. Were the schools good? I actually don't believe this. The School System sought to create a risk averse set of professionals (i.e doctors, lawyers, dentists, bankers) and paid little attention to encouraging "creative" types. Very few professors were created by the Scarsdale schools.
What is my point? Go to this zillow website and take a look at the real estate price differential between homes located on the Scarsdale side of the border (averaging $1.2 million) relative to homes located 50 feet away on the White Plains border (averaging $550,000).
Type Farley Road, Scarsdale 10583 Here
Now that's an enormous border discontinuity! What is the Scarsdale address buying you relative to the White Plains address? The homes are 50 feet apart!
Discontinuities at political boundaries is always a fun fact to note.