There seems to be some disconnect or bit of talking past one another on the issue of a housing recovery.

There are people in the Recovery Winter camp, like myself, who argue that the upturn in housing will contribute to sustained growth in the United States, baring major unforeseen shocks. That is, unless something identifiably contractionary happens we should expect growth to become more solid, with fewer disappointments and a steady path back towards our long run trajectory. In short, a recovery.

However, I am receiving a steady stream of comments noting the continued downward movement in home prices along with evidence that the uptick in new home sales and starts at the beginning of this year was a fluke.

So, to be clear, I don’t think there is any reason why we should expect strongly rising single family home prices in the near future. Nor, is my baseline assumption that we will see a strong increase in single family construction. The upturn at the beginning of this year was surprising to me as well.

In contrast, my thesis is first that single family construction will not fall any further. Currently new single family homes built-for-sale are running slightly behind new single family home sales. Inventory is at record lows and declining. Thus even with continued weakness in the existing home market we are not likely to see significant declines in new single family home construction.

This is in large part because new single family home construction is close to or outright dominated now by custom home construction. That is homes built by owner or built by contractor for a specific owner. There is no substitute for a custom built single family home and there is no reason to expect that people in this market are going to be doing worse economically. So, we think those builds are safe.

In contrast the baseline scenario involves a pick-up in multifamily construction as rents rise. At core the reasoning is that a breakdown in the mortgage market and/or folks desire to invest in single family homes does not change the need for housing. Currently America’s population growth is well outstripping its housing stock so that we have an abnormally large number of people per household and an abnormally small number of households.

Barring a large cultural change – which is possible – this will mean a steady increase in the demand for rental housing. We have seen this as rents are currently rising quite rapidly and reports from apartment owners show what are approaching record low vacancies.

This makes constructing apartments a great value proposition.

Now, lots of people point out that multifamily is a small part of the overall housing market. This is true but two things:

First, the small size of multifamily is a result of a 20+ year expansion of single family homes. Go back to the 70s and 80s and multifamily construction was on par with single family.

Second, for growth what matters is not the level but the absolute change. If multifamily went from its current pace of roughly 200K units a year to a pace of 800K units a year that would constitute massive growth, even though it would be well shy of the record pace that single family construction hit during the boom.

A residential industry that was pumping out 800Kmultifamily units along with roughly 400K single family units a year would also be in a robust recovery, with rapidly declining unemployment among construction workers.

This scenario, in which new home construction comes to be dominated by multifamily built-to-rent units does not require the single family housing market to normalize anytime soon. Indeed, it predicated on the assumption that the single family home market will not normalize soon.

If the single family home market does normalize it will put a dent in the multifamily market. I suspect that part of the reason multifamily isn’t accelerating any faster is concern that the single family market will repair itself ahead of schedule and knee-cap rising rents.