compare second home information to housing crisis performance

[garrettgman]

choose a measure(s) of an area's performance during the housing crisis and compare it to the number of second homes in the area with linear regression. Is their a relationship between:

• the number of second homes and the time of the peak?
• the number of second homes and the rate of the decline in prices?
• the number of second homes and the max increase in prices since (2000)?
• was there a change in the number of second homes between 200 and the peak?
• between the peak and now?

[garrettgman]

At the state level: Hawaii, California, New York, and Washington D.C. had the highest rates of second homes (~ 40%) In other states, including Florida, second homes make up about 25% of all homes. The percentage of second homes fluctuated between 2000 and 2007 for all states. The largest fluctuation was 8% (Wisconsin and Maine)

At the state level: an examination of Florida shows that different puma's have different rates of second home ownership. The highest three rates occur in and near downtown Miami (4409, 4010) and downtown Orlando(2204). These puma's are also among the smallest in the state, suggesting areas of high population density.

Conclusion: the ACS owner occupancy variable appears to be a reliable measure of second (or third, etc.) homes. These homes appear to more often be rental properties, such as apartments, than vacation homes.

TO DO: there is not a 1 to 1 correspondence between PUMAs and MSAs. How should we combine this PUMA data with the MSA HPI and construction data?