This has been a strange week. I spent monday and tuesday in deep thought about the UCLA Statistics' Department's Past, Present and Future. There is nothing more fun than to serve on a Departmental Review. While we were locked in a room for 12 hours a day talking about Statistics (which is distinct from doing statistics), I made five new friends. UCLA's Vijay Gupta, Dorthy Wiley and Carol Bakhos are all very sharp, very fair colleagues and I hope to stay in touch with them. Our two external statisticians, Adrian Raftery, and Terry Speed, carried us to greatness. As you might guess, they are very smart but they were also very funny. I spent the last two days at USC in deep thought.
But now, I have actually read this newspaper article from the LA Times. Google is getting ready to "wire" certain liberal cities with fast cable that will allow users to download data 100 times faster than they can now. I have several questions for the Google Guys who regularly read this blog;
1. Have they bought land in the cities where they will lay down the cable? This unique amenity will be capitalized into land prices in the cities where they build. You could use a spatial regression discontinuity design to test this claim. There will be zip codes just inside the "Google Zone" and they will gain fast access while there will be a nearby control group of zip codes without access. So, the cliche of real estate is "location, location, location" but usually what has differentiated parcels of land are locally tied "god-given" attributes such as climate or proximity to the coast. Google is introducing a "man made" difference across parcels (namely access to extremely fast download tech).
2. What businesses will self select to locate in the "Google Zone"? In english, how will access to this unique amenity affect the migration patterns of households and firms? What types of households will seek out the "Fast Google Zone" areas even though home prices will be higher?
3. Should Detroit's mayor pay Google to designate Detroit as a "Google Zone"? Should Obama use Stimulus $ to pay Google to lay fast cable there?
4. General equilibrium effects -- as Google lays down more cable will diminishing returns kick such that if we repeat step #1 above we will see declining capitalization into real estate effects? Or is this an increasing returns technology such that the demand for "Google Zones" actually increases as Google lays down more cable?
What's mildly interesting here is that up till now, the Internet has been viewed as a substitute for Cities --- but returning to the old theme; in this case the Fast Google Zone Internet will increase the demand for the specific subset of cities that have access to the fast cable.
Look for San Fran to be one of the guinea pigs rather than detroit?