A blog post on a blog I read, Urbanophile, analyzed the housing stock of some Midwestern cities. The author pointed out the uniqueness of Detroit's housing stock, particularly its relatively high amount of older units and detached single family homes. The author argues that older homes are more difficult to renovate and that having a large proportion of single family detached homes in a city limits the ability of developers to construct newer housing that meets the preferences of today's would be city dwellers e.g. areas zoned for single family detached housing make it difficult to construct lofts, apartments, etc.
With this in mind I decided to analyze the housing stocks of Dayton and Greenville. Below is a chart showing occupied housing units grouped by the age they were built for both cities. This data is at the city level, not the MSA level.
As the graph shows Dayton has a much older housing stock than Greenville. This is not surprising considering the Greenville MSA's population, which includes the city itself, has been growing since 1970 while Dayton's has declined. A growing population means new housing has to be built, while a declining or stagnant population means that old houses built in the past when the population was larger are still around.
Roughly 90% of Dayton's housing was built prior to 1979, compared to approximately 67% for Greenville. The real difference is in the amount of housing built prior to 1950; roughly 50% for Dayton compared to only 21% for Greenville. If older housing is truly more difficult and thus costly to renovate or rebuild, Dayton has a larger financial hurdle to clear than Greenville in the coming years.
Below is a graph showing they types of housing units in each city.
Greenville actually has relatively more 1 unit detached housing units occupied than Dayton. Dayton has relatively more occupied large housing complexes consisting of 5 to 20 or more units. I was a little surprised to see that Dayton has a larger percentage of large housing complexes. Having walked around Greenville quite a bit over the last 4 years, especially the downtown, they appear to have a lot of large complexes compared to Dayton's downtown, though I admit my familiarity with Dayton's downtown is not particularly high.
Based on the criteria laid out in the blog post it appears that the outlook for Dayton is mixed. On the one hand they have a relatively large amount of older housing; on the other hand they also have a relatively good mix of unit types. It is worth mentioning that the numbers used for these charts are occupied housing units, which means Dayton could have a lot of newer unoccupied housing (if developers over built) or a lot of empty 1 unit detached homes (people have left and no one has bought the property). Both of those scenarios seem more likely to be the case in Dayton relative to Greenville, and both would be bad signs for Dayton.
When I have more time I might try to get data on total housing units by age and type. This data was easy to grab so I started here. Here is a link to the tool I used to get the data. It is fairly easy to use and has a lot of data on various geographies, including cities, counties, census tracts, and blocks.