This report finds that land values are higher in areas near apartments and reports that apartments lead to higher land values (or more charitably that apartments don't decrease land values). Unfortunately it is not super rigorous or fine grained but even as presented doesn't seem to support the claim.

The first clue that something funky is going on is the map on page 3. The apartments (and their 1/2 mile neighborhoods) tracked are highly concentrated along the main north south corridor in Salt Lake City and close to town. There is one anomalous (to me because I don't know Salt Lake City at all) local concentration in the Southwest region that appears to around the intersection of highway 154 and a few other highways. The strong connection between accessibility and increasing land values in a growing city would lead us to expect this land to see faster appreciation with or without apartments. They compound this by essentially comparing all land within 1/2 mile of major transportation corridors to all land further than 1/2 mile away from a major transportation corridor (I can't even tell if they are restricting their analysis to pre-existing single family or are including the new housing on the suburban fringe in their median pricing each year). Nothing fine grained like we have seen in recent studies finding the opposite. They trot out the amenity effect explicitly examined by Dr. Xiaodi Li as occurring but not outweighing the supply effect (They don't cite Dr. Li and oddly all of their academicish citations I can't find online).

The main thing I find interesting are the charts 7, 9, 11, 13 that report the average annualized 1/2 mile home price growth based on apartment vintage vs the price growth not within 1/2 mile of any apartments within that region. If you look hard enough and try hard enough one can convince oneself that there exists a pretty clear pattern. For all sub-regions besides the south east, homes within the 1/2 mile of earlier vintages of apartments have seen relatively lower relative price appreciation through 2019. While the 1/2 mile relative advantage for price appreciation becomes stronger for more recent apartment vintages (the southeast region just always shows lower price appreciation for apartments 1/2 mile neighborhoods). They throw away the obvious implications of this with "This resulted from the negative economic impacts brought on by the housing crash of the prior decade", but I see no obvious reason that the trend of relative pricing due to competition from apartments should be expected to be different just after the recession through to today.

Now, I have pretty much exactly the same role as these guys. I get the need to just get a report written. But, if my data guy had come to me with these charts and tables I would have written a completely different article. In the short term apartments tend to built in areas with higher land prices so as we see rapid appreciation we expect that apartments are more like to start being built in those areas. Over the mid to long term as apartments are built in a neighborhood property price appreciation will be depressed (from what it would have been otherwise). If five years later my data guy brought me the updated data what I would expect to see, is that in the areas where they were building apartments in 2017, 2018 and 2019 and still seeing relatively rapid price appreciation they continued to build apartments (if allowed) in 2020-2025. And looking back then to the average annual price appreciation by vintage from 2017-2019 through 2025 the relative annual price appreciation advantage of within 1/2 mile will have disappeared with increasing competition.

Looking that their (confused) conclusion I think I understand why this report was written the way it was. The people who complain about apartments claim to be concerned about housing values. So they wrote this report as a sop to them, "hey see density doesn't cause price to fall" but then wrap it up with "but we need to allow for more density so prices don't rise".