“efficient market” loses out to actual seller and actual buyer of 76 Madison Avenue loft

market extrapolation can be hard, but this loft sale comp is a non-sale of same loft

Stop me if you’ve heard this before: an “index” is a proxy for The Market, just as a market report is a summary of The Market, and “The Market” is a proxy for the value of any one property. You use such tools to “get a feel” for the overall market, then you use some brain power to apply those tools to the question of what this Manhattan loft is worth at this time. There are some proxies that are more … er … proximate than others. Conventional Wisdom is that the best predictor of what the 3rd floor in a loft building will sell for today is what the 4th floor loft sold for yesterday, then adjust if The Market changed from yesterday to today. And a loft sold in the same building a few months ago is a better predictor than a loft sold in a different building yesterday. Et cetera, et cetera, et cetera, as Yul Brenner might have said.

But there is an even better proxy for the value of Loft A in, say September 2013, and that would be if Loft A did not sell for some extended period while professionally exposed to the market; then you can say that that unsuccessful price was a ceiling (never reached) at that time. Adjust for different market conditions at the different times, and voila! In an efficient market you have a principled basis for determining the upper limits of value, more or less. You see where this is going, right? (Especially if you remember the headline.)

The “804 sq ft” Manhattan mini-loft #2C at 76 Madison Avenue did not sell when offered from September 21, 2012 until February 20 at prices of $1,049,500, $999,999, $990,000, and $960,000. The price drops tell you the seller was motivated, but here’s another way to tell: the seller paid $999,996 to buy the loft in April 2007 (avoiding the “mansion tax”, just). There were a great many buyers active in the overall Manhattan residential real estate market over those 5 months, which establishes that the small loft was not worth any of those prices at any of those times, with no extrapolation required to draw that conclusion.

The extrapolation part is figuring out what it should not be worth in a later market. Thus, if the StreetEasy Condo Index is up from 2,000 in February to 2,188 in September (it was, and that’s 9.4%), you’d think that the established ceiling is that #2C is worth something less than $1.05mm (remember, the ceiling is what it should be worth less than; it doesn’t tell you what it is worth) as of last month. You see where this is going, right?

Loft #2C came back to market 4 months after not selling at $960,000. It took only 4 weeks for it to find a contract off an asking price $100,000 higher than any prior effort, closing on September 11 at $1.1mm. On the one hand, that’s only 5% off the implied ceiling. On the other hand, that implied ceiling was still a maximum value; the true value in those non-sale months is unknown. On a third hand, you can’t argue with this fact: regardless of what the loft is worth (or not worth) on any day before the contract was signed, on that day the loft was worth $1.1mm. (And we use the closing date as the hard date for determining value because otherwise we would go insane trying to normalize values across often unreported contract dates.)

Let’s review. It is not theory, but fact that the loft was not worth those earlier prices at those earlier times. In theory, that factual statement gives us a basis for finding the maximum value at which the loft will sell at future times; indeed, in practical terms that factual statement can be used to predict that the loft should not sell if offered at $1.05mm in or before September 2013. But a couple of real-life buyers (presumably aware of the history, but who knows?) decided that the loft was worth $1.1mm to them, at least when faced with a seller who may have insisted that it was worth $1.1mm to him.

Rubber meets road, when real buyers have to negotiate with real sellers. All past, all comps, are theory. Facts is facts. Even if two facts cannot be reconciled. That, as they say, is what makes a horse race.

there is another recent and relevant comp for this loft

Here’s why the seller came back to market after having been unsuccessful for 5 months, and why he came back higher than the prices at which he had been unsuccessful. The capricious loft gods of Manhattan brought the substantially identical loft #7C to market on May 16 at $1.25mm and threw it into contract within two weeks at $1.275mm; it closed on July 8. Look at that again: #7C came to market at $1.25mm three months after #2C was pulled off the market while asking $960,000. And got $1.275mm, a one-third premium over the poor neighbor downstairs.

Both lofts boast of top notch finishes; perhaps #7C is a little more polished, with media systems that #2C lacks:

stunning one-bedroom prewar loft, complete with 55 [inch] flat screen, built-in sound system with IPod docking station, is turn-key ready to deliver a luxury lifestyle at hello. Designer James Rixner fuses modern with prewar mixing textured walls, window treatments and custom lighting to create a relaxed minimalistic elegance. High ceilings, 4  [inch] solid oak wood flooring, natural stone, opaque glass doors and gray cabinetry blend beautifully together. Chef’s kitchen features bar dining, Sub-Zero refrigerator, wine cooler, Miele cooktop and oven. Spa bath has soaking tub and separate shower. Other features include: Washer/dryer, HVAC air, custom closets and 23+/-SF storage unit.

But #2C is not slumming it:

gorgeous one bedroom has soaring 12 foot ceilings, giant windows with incredible light and all custom finishes. Kitchen comes loaded with custom European cabinets, Sub-Zero, Miele range and cooktop, Bosch d/w and a wine refirgerator. The bathroom is truly specatacular – custom vanity and marble finishes, a giant soaking tub and a rainwater shower. Hardwood floors throughout, Miele washer/dryer in the unit, your own storage unit

You’ll notice that StreetEasy has #7C at “850 sq ft” and #2C at “804 sq ft”. I have to assume that the numbers are right (Property Shark has the same data), even though the floor plan for #2C is exactly the same as that of #7C. Of course, the floor plans are for identification purposes only, blah blah blah, but the units surely seem the same.  Without knowing exactly where the extra 46 sq ft are, it is hard to say whether it would be significant to a buyer at all (meaning, #7C beat #2C by $175,000, or 16%), or whether the units should be compared as $1,368/ft and $1,500/ft, or (only) 10%.

Regardless, The Market that had rejected #2C as low as $960,000 as recently as February 20 closed on #7C at $1.275mm on July 8. The #2C seller must have heard about the #7C deal sometime between contract on May 30 and #2C coming back to market on June 29. The #2C seller evidently believed that the #7C deal was a cudgel with which to beat potential buyers (or, to use less violent imagery, as an eraser to eliminate #2C’s own listing history as a relevant comp); in this, he proved to be correct: $1.1mm.

That must have been an interesting negotiation.

a change in direction, from renting to selling, to renting, to selling

The rental listing history for #2C on StreetEasy shows what the owner was doing with this loft over recent years, and provides insight into his 2013 decision-making. The effort to sell that began in September 2012 would be well-timed to coincide with the end of a year lease that started around November 29, 2011 (note that the early marketing did not refer to a tenant, suggesting the tenant would have vacated or was on a month-to-month basis by the time a deal could close). When the unit still had not sold into 2013, the seller listed it as a rental again January 13, overlapping with the sales effort that featured a cosmetic price drop to $990,000 on January 8 and a further cut to $960,000 on February 4.

The seller gave up on the sales effort and took the loft off the market on February 20, finally renting it by April 7 for $3,795/mo. Very likely, the owner was resigned to holding through the end of the new lease and hoping for a sales market re-set by then.

But then … #7C happened, asking the market to pay $1.25mm in the month after the #2C owner had resigned himself to another year of renting. I’d be curious to know if the #2C owner thought that #7C had any chance of getting anything close to the ask, but the smart money would be on “no chance”.

But then … #7C signed a contract by May 30 at $1.275mm. That was enough incentive to try to sell #2C again, despite the complications of trying to sell with a tenant not only present but present long enough a new buyer could not take occupancy for quite a while. (The recent broker babble concluded with “Tenant in-place until April 14, 2014 [paying $3,795/month]”.) That effort started 10 days before #7C closed, close enough that the (new) #2C listing agent could credibly use that pending deal as a cudgel an eraser in discussions with interested buyers who would have argued that something under $960,000 was a good place to start discussions.

The result: $1.1mm. On the one hand, a larger discount from #7C than seems to be warranted by the floor height or condition (or the 46 extra feet); on the other hand, to a seller who had been unsuccessful at $960,000 as recently as February … ka-ching!

Fans of the efficient market theory should now rend their garments.