negotiation science / The Anchoring Effect

indeed, "we" are not so smart
Interesting post from an interesting blog that I have been mulling over this past week, as it had great personal relevance this week (story below) and has a real estate nexus. The post is
Anchoring Effect; the blog is You Are Not So Smart with the clever subtitle, A Celebration of Self Delusion. Here’s the lede, where the key factor (after you understand how it works) is where that "initial conception" comes from:

The Misconception: You rationally analyze all factors before making a choice or determining value. 

The Truth: Your first perception lingers in your mind, affecting later perceptions and decisions.

Read the whole thing. I am pretty sure that at least one of his examples will resonate with you (the leather coat, the population of Venezuela, African countries in the UN, car sales, auctions and social security numbers, Vuitton vs. Wal-Mart handbags, summer camp volunteers). The Anchor Effect definitely exists, and understanding how it works helps us better able to overcome it.

Let me give you one (true!) story about the car we bought this week (after I read this post), before getting to some highly relevant Manhattan real estate issues.

me and my Anchors buy a car
Long story, short: I did tons of research, over-preparing to buy a certified pre-owned Prius, had Net Number (after trade-in) in mind we were comfortable with, and identified two dealers with interesting low-mileage inventory. Both had a pair of a fully-loaded 2008 and a 2010 with fewer options. At the NJ dealer, the 2008 and 2010 were offered at essentially the same price and we preferred the 2008 (partly due to the exterior and interior colors). Valuing the trade-in, that dealer and I got within $550 of each other, net.

I have to believe that was the dealer’s best price, because we said NO, walked out, and did not hear from them again.

At the CT dealer (90 miles north of the first dealer!), the web price for the fully loaded 2008 (which we again preferred) was $3,000 less than the 2010 but (it turned out) this dealer valued the trade-in much lower than the NJ dealer, complicating the Net Number calculation.

I became aware of three Anchor Effect problems. First, my expectation based on the lower 2008 sticker price in CT was that I should do much better there than in NJ, so I was disappointed (distracted?) when I did not. Second, my expectation from the NJ experience was that the 2008 fully loaded should be equivalent in price to the 2010 with fewer options, while in CT there was a relative over-valuing of the 2010 (a car I did not prefer). Third, I had negotiated a much lower discount on the 2008 in NJ than it looked as though I was getting in CT (absolute prices aside), so it did not feel like a better deal.

As it played out, I negotiated a Net Number in CT for the fully loaded 2008 that was clearly going to be better than the price I would have been willing to pay in NJ (but that dealer would not meet it). The mistake I almost made was to walk away in CT to return to NJ to buy the 2010, even though we originally preferred the 2008 to that 2010, and even though I could get the 2008 in CT for less than we would have been satisfied paying in NJ. Weird behavior.

Fortunately, I had enough time to reflect through (and get over) my anchor biases, and to return to my original goal: to buy the Prius we preferred within our original dollar comfort zone, net. We preferred the 2008 fully loaded, so the 2010 valuations were only a distraction. I could get a better deal (net) in CT even though that dealer was willing to give me (significantly) less on trade-in. The fact that my negotiating margin in CT was smaller than in NJ was irrelevant, so long as the result was better.

a 5% solution
We are picking up our new-to-us certified pre-owned 2008 Prius (did I mention that it is fully loaded?) in CT on Tuesday. The Net Number in CT is 5% lower than the number I would have taken in NJ but that dealer let me walk away without offering.

Now I spend too much time second-guessing my CT negotiating strategy since I only pushed back once when the dealer said I had squeezed him to his "final" number. Yes, I got $250 less than his "final" number, but could I have gotten another $100 off his final final number??

Never mind. What does this have to do with real estate?

difference for real estate buyers and sellers: who sets the anchor
There are the obvious points of relevance for buyers to recognize the Anchor Effect of an asking price, and maybe someday I will do a post about how sellers of (truly) unique lofts set arbitrary anchors (perhaps using as a point of reference the June 2008 sale of #9S at 130 West 17 Street, which I hit when it it came to market, 130 W 17 9S is new + really going for it, and again when it went into contract, don’t sneeze at 130 West 17 Street contract, as I believe that seller created an arbitrary anchor that succeeded in generating an above-market deal). For now, I want to focus on the difference between anchors for sellers and buyers.

As Raney suggests in his blog post, a buyer is more likely to be aware that the seller has created a likely-to-be-arbitrary asking price than the seller is to be aware that she has been (irrationally?) influenced by an anchor in setting (and sticking to, stubbornly) a possibly-too-high asking price. As he puts it:

External anchors, like prices before a sale or ridiculous requests, are obvious enough you can sidestep the actual price, the real appeal. Internal, self-generated anchors, are not so easy to bypass.


I take that to mean that a seller who paid $1.5mm for a loft at any earlier time will probably use that $1.5mm as the anchor from which to make adjustments for current market conditions but that seller is prone to making poor adjustments because of the anchor. As Raney quotes some other smart folks:

In many situations, people make estimates by starting from an initial value that is adjusted to yield the final answer. The initial value, or starting point, may be suggested by the formulation of the problem, or it may be the result of partial computation. In either case, adjustments are typically insufficient…that is, different starting points yield different estimates, which are biased toward the initial values.

 – “Judgment Under Uncertainty” by Kahneman, Slovic and Tversky

Or, a seller may (unintentionally) be anchored by the number she has to have in estimating what a buyer will rationally pay. Yes, that seller may say she is willing to make adjustments offered by a professional real estate agent based on the current market, but may be over-committed by her anchor to an unrealistic price for a "reason" she does not fully appreciate.

Fascinating stuff.

h/t Andrew Sullivan

© Sandy Mattingly 2010

Posted in psychology of the market Tagged with: , , , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *