The Miller on absorption of Manhattan coops and condos

hint: it is worse as you go up
One of the nice things about the blogoshere is that it provides a forum for dialogue, even (especlally) with oneself. Here’s an article last week from the Wall Street Journal analyzing the "absorption rates" for different Manhattan real estate market segments, based on numbers reported by The Miller, Condos Rule Over Coops at High End, June 22. Here is The Miller’s further consideration, on his own Matrix blog, The Differences Between High End and Low End Absorption, June 22. First, let’s define the term.

defining absorption
The Miller defines absorption as "the number of months it takes to sell available inventory", but does not add what is implied: that the rate of sales is the rate for the month under consideration. I.e., the chart in the WSJ and on Matrix showing that the absorption rate for coops between $3mm and $5mm is 16.2 months, assumes a rate of sale based on May 2010 recorded sales of coops in that range. Thus, absorption is a function of two metrics: size (how much inventory in a price range at any given time) and rate (how many sales in that price range in the month under discussion). Thus, if the rate of sales increases with no change in inventory, absorption declines; if inventory declines with no change in rate of sales, absorption declines. And vice versa.

The Miller argues with The Miller’s chart
The take-away from the WSJ article is evident from the title: Manhattan high end condos were absorbed more quickly in May than high end coops. In a sense, this is ‘evident’ from the graph, which clearly spikes at different levels at the top of the price scale, breaking the (rough) symmetry between coops and condos at all other price points. Which is an opportunity for The Miller on his blog to muse that

The article sort of suggests that condos are doing better than co-ops as a generalization, which isn’t quite correct or I am over analyzing the results.


As he explains on his blog,

[Absorption rates for] Co-ops over $10M are significantly higher than condos but this segment is about 1% of all sales so the results are easily skewed.


In other words, after slicing and dicing one month’s worth of sales data, the slice for coops over $10mm is awfully thin, and not worth investing too much reliance on as a trend point.

2 points to rely  on
The Miller makes two important contextual points in his blog post that are missing from the WSJ , presumably because the WSJ was more interested in making a point about the difficulties of Wall Streeters ("It is more difficult for the Wall Streeter who wants the big prewar apartments to buy one than before," said Mr. Miller.):

[1] Co-op and condo absorption is generally on par with the 10 year 9.9 month average overall rate of absorption

[2] Lower priced property generally absorbs faster


I would add that under $3mm the absorption rates for coops and condos are in close agreement, and that at each price point below $3mm absorption rates are roughly even.

a final word of caution
Lovely  bar graphs with striking data points from a single month of sales divided into 16 unequal segments are easy to over-analyze. (The Miller does not provide an N, but my guess is that he will show around 900 overall sales for May, heavily weighted toward lower price points.)

© Sandy Mattingly 2010


Tagged with: , , , , ,

Leave a Reply