$5 million data points: unrealistic Manhattan loft sellers or market softening?

the numbers are intriguing

One of the story lines coming out of the Manhattan Media Division of the Real Estate Industrial Complex has been about supply in the luxury market (specifically, in new developments), with the (often implicit) question about whether demand will keep up. I can’t claim a systematic review, but I have noticed a string of data points at the upper end in my Master List of Manhattan Loft Sales that is interesting enough to share. (As long-time Manhattan Loft Guy readers know, my Master List caps at sales of $6,000,000.)

The heady asking prices aside, if only one of these had caught my eye I might have done a blog post like my July 14, 2015, so many incorrect prices finally sell 73 Worth Street penthouse loft above ask, or my February 17, 2015, challenging 474 Greenwich Street loft proves a challenging sale, not quite $2 million off, or my September 2, 2014, 81 White Street loft takes a year to sell as the very not-prime Tribeca loft it is, as I am often drawn to extreme price drops and long marketing periods. In an individual post, I’d likely look at history, comps, and the specific layout or finishes of the loft to try to come up with a theory about why this loft took so long to sell, or that one needed so many price chops.

I may yet do that for some of these, but there were four in December alone, with big spreads from a big asking price, and a long time between coming to market and finding a contract. So I started counting ….

As of earlier this month, there are 18 loft transactions on the Master List that closed since October 1, 2015 that started asking $4.5mm or more. A bunch took deep discounts to eventually close (and by “a bunch”, I mean in this case eight), as we’ll see below.

Repeat after me: anecdota are not data. But a series of anecdotes are data points. Here’s some from that bunch:

not every expensive loft suffered this experience, of course

In the interest of completeness, here are the ten that did not fit into that bunch:

A couple of notes: the DoM (“Days on Market”) in my Master List is measured from first offering to contract (you probably know that other people use different metrics); my cut-off for the first bunch was a sale price at least 10% lower than the first ask, though as it breaks out I could have used 15% and gotten the same split; I used $4.495mm as the low cut-off for original asking price for this exercise, as a fair rounding to $4.5mm.

Some observations:

(a) It takes a long time to sell a mis-priced Manhattan loft at these price points. All but one in the first bunch took more than six months to find a contract.

(b) It can take a long time to sell even a well-priced loft at these price points. The 52 Laight Street entry took almost a year to do a deal within 2% of the first ask; the Spice Warehouse loft at 481 Washington Street took almost 16 months to sell within 8% of the first ask.

(c) While ten of the 18 lofts on the two tables took at least 183 days to find a contract, only eight other lofts on the Master List that sold since October 1 took at least that long to sell (out of a data set of 90 other lofts).

(d) The data from my Master List are all resale units, as I keep the new development sales on a separate sheet. I don’t pay a great deal of attention to the new development portion of the market, for a variety of reasons (principally that I don’t trust the asking price and marketing dates reported, but that’s a long story), but for present purposes I will simply note that loft buyers for new developments are very different from loft buyers in (what can be termed) the current market.

a theory, not a data point, about the top of The Manhattan residential sales market

Getting away from data completely for a moment, I have heard a number of experienced and knowledgeable agents say lately that even resale buyers at this price point have been ‘spoiled’ by new developments, even if they are not interested in buying a new development property. Subject (always!) to empirical testing, that makes sense to me: people who pay close attention to the market see levels of finishes and amenities packages in new developments that few lofts in the resale market can match. (A related point: I always tell my buyer clients not to look at lofts at price points they don’t want to pay, as the collection of appealing attributes of such lofts will [almost by definition] exceed that of lofts that are truly within their reach. This is true whether the budget is one million, two million, three million, etc.)

New developments simply soak up so much media attention that they can insidiously reset norms and expectations. That’s my (working) theory, and I am sticking to it (until debunked).

are more sellers over-valuing their downtown Manhattan lofts?

This post focuses on the seller side, of course. The intriguing possibility is that these data points suggest that loft sellers at this point are more unrealistic about The Market than sellers at lower price points. All that hedging language is required for many reasons, not least that my data set has only 18 sales going back to October 1; a further caution is that The Miller did not find any great discrepancy between the market expectations of the ‘luxury’ portion of the market (he calls that the [rolling] top 10% by price) and the overall market, as measured by Listing Discount (he measures from the last listing price). See, for example, his Manhattan Market Report for the last quarter of 2015, which noted a Listing Discount of 3% for the overall market and 2.9% for the Luxury portion.

The Miller’s data for Days on Market are difficult to compare to even my limited data set, because he (again) measures from the last listing price. I understand his theory (that’s when the seller really found The Market for the property), but philosophically I prefer to count from the earliest hard date on which the seller announced an intention to sell. The Miller has to ignore agents who make small changes in listing prices. (Some Manhattan agents are notorious for doing this, apparently believing they get a fresh look each time.)

The Miller’s DoM numbers show an overall market that is much quicker measured year-over-year but a little slower measured quarter-over-quarter (82, 73, 105, for 4Q15, 3Q15, 4Q14 respectively) but a sorta opposite ‘trend’ at the Lux level, much quicker quarter-over-quarter but only a little quicker year-over-year (150, 116, 160, respectively). The upper end takes longer to sell in The Miller’s world, as in mine, but there’s not much comfort in apples to oranges.