The Census Bureau released its Characteristics of New Housing report recently. My biggest takeaway from the report was how much the gap between for sale and rental housing construction has narrowed in recent years – most of this has come from the massive decline in for-sale housing construction, but multi-family completions are up substantially as well. However, there is also a big story within the multi-family data: the units that are getting built are increasingly in larger buildings. Bloomberg’s Justin Fox highlighted this trend in a couple of fascinating graphs that illustrate this striking trend.

Fox goes on to give several reasons for the trend towards larger units including NIMBYs forcing construction out of the suburbs and into denser downtown areas, increasingly large investors desiring scale and changing consumer tastes. However, there is one factor that largely goes unaddressed in Fox’s Bloomberg View article – cost, especially when economies of scale in operations are taken into account.

In theory, building smaller new buildings sounds great – it’s an under-served niche and one would think that yields are superior since the unit counts are well below the massive projects where institutional investors tend to bid land values up. However, this rarely plays out in reality. The problem is that there is simply not the economy of scale in operating a 20 unit building that there is in operating a 150 unit + building. This means that expenses are often substantially higher as a percentage of revenue which leads to lower profitability on a cash flow basis.

Couple that with the fact that these projects are not desirable to large investors and often trade at higher cap rates than large buildings as a result and you have a recipe for under-performance which is why so few are getting built in the US. This isn’t nearly as much of an issue when construction costs are low and yields are relatively high, which helps to explain why it has gotten worse over time.

Unfortunately, the opportunity today is not to construct more small apartment buildings but to rehab older buildings and position them below the new larger projects. We have seen this strategy deliver strong returns over the past few years, in part because the barriers to entry from new small product are so high. US cities and suburbs could absolutely use more small apartment buildings which tend to be cheaper to rent and more desirable for families. However, I don’t see it happening without some major advances in operating efficiency.