The proof is in: smart growth reduces risk of mortgage default

[thelakelander]

Interesting....

An exhaustive analysis of 37,000 mortgages on multifamily rental properties has found that, when other factors are appropriately controlled, those properties in smart locations are substantially less likely to incur default on payments than those in average locations......

The numbers are quite impressive:

Close-in locations matter.  Every additional minute of commute time raises the risk of default 3.7 percent.  (Put another way, every five minutes of additional commute time raises the risk 18.5 percent.)

Transit matters.  If 30 percent or more workers in a location commute by rail transit, the risk of default is reduced by 58.4 percent.

Walkability matters.  Every one percent increase in the share of workers who walk to work decreases the risk of default by 3.1 percent.

Mixed uses matter.  If there are 16 or more retail establishments nearby, the risk of default is reduced by 34.4 percent.

Affordability matters.  If the multifamily building qualifies as affordable housing under Fannie Mae criteria, the risk of default is reduced by 61.9 percent.

Parks matter.  If the property is located within one mile of protected green space, the risk of default is reduced by 32.5 percent.

Distance from freeways matters.  If the property is located within 1000 feet of a freeway corridor, the risk of default goes up 59 percent.

full article: http://switchboard.nrdc.org/blogs/kbenfield/the_proof_is_in_smart_growth_r.html

[JFman00]

Wow that's pretty awesome data.

[spuwho]

Transit matters, walkability much less so.

Affordability trumps them all, no shock there.

[JFman00]

Transit matters, walkability much less so.

Affordability trumps them all, no shock there.

Where are you getting that?

Looking at the actual study: http://www.fanniemae.com/resources/file/aboutus/pdf/hoytpivo_mfhousing_sustainability.pdf

If I'm reading it right, the variable for transit, SUBWAY30, is a binary variable of a tract having >= 30% of commuters by subway or elevated rail or not. PCTWALK is simply the percentage of people in a census tract who walk to work. There's not enough data in this survey say that subway/elevated rail or walkability matters more (to the risk of default) just from this study. If we assume the relationship in ridership percentage and risk of default is linear up to 30% ridership share (extrapolating liberally from Table 3) then walkability actually matters more, as a 30% walk commuter share would mean a 93% lower risk of default. Putting the transit score in the same terms as PCTWALK would give you a 1.95% decrease in default risk for every one-unit increase in subway/elevated rail commuter share from 0-30%.

[spuwho]

I translated "walkability" as "walk to work" not as "walk to transit".

[JFman00]

If that's the measure you're using, the survey suggests walkability matters more than transit. If I remember right the ACS question is something along the lines of "What is the primary mode of transportation you use to get to work?".

The study explains walk to work and subway/elevated to work jointly while the actual "walkability" variable is having 16 or more retail establishments in the census tract. If that's the only measure your looking at then yes transit (specifically subway or elevated rail) might look to matter more.

You really can't conclude one way or the other given the constraints and scope of the study.

EDIT: The ACS question is "How did this person usually get to work LAST WEEK? If this person usually used more than one method of transportation during the trip, mark (X) the box of the one used for most of the distance."

[thelakelander]

From what I can tell in the study, walkability does matter more than transit.  After all, you can't have decent transit without having strong origins and destinations of walkability already present on some level.