This is an interesting discussion at a convenient moment, as I am
about to sell my house using an REA. That in itself is not very
interesting and certainly not very unique (so why mention this here?).
I am however trying to figure out what model to use for the contract
with the REA to get him "aligned" (assumption: with the right model I
will be able to find an REA who is willing to take the deal).

The remark of Justin on REA's selling their own homes at a higher
price (though taking a little longer) triggered me as it is a symptom
I would expect when the goal of an REA is not aligned with the goal of
the customer (until the customer is the REA himself).

Although this is slightly off-topic as I am approaching this thread
from the angle of the customer, I think it touches just about enough
on the sales process to post it here.
I wonder if you could comment on my thought experiment – whether it
makes any sense, whether it contains fallacies, wrong assumptions or
negative effects, or whether it has already been tried and tested in
some form or another.

In line with the previous post there are two variables that are of
prime interest to me: (1) selling price (as high as possible) and (2)
certainty about getting the place sold (asap). I think there is a
third one that is related to the second but not entirely the same,
namely the exact date at which the house will change ownership. I
myself am not very fond of moving, so I would prefer to keep my old
home say one month while I already own my new home.


=|JRM: So here's a problem right away. A (goal-seeking) system can only have
one goal. That's the nature of reality. In the TOC world, we are careful to
differentiate between *the* goal and necessary conditions (there can be more
than one of the latter).

This realisation should lead you to re-word your objective above. It may be
that you want to maximise your sales price (goal), while ensuring that the sale
occurs prior to x date (NC1) and settlement occurs at x + 1 month (NC2).

Of course, you may chose to switch the goal I've nominated with the first
necessary condition. In which case you would want to sell your home as soon as
possible (goal) for not less than $y (NC1), with settlement occuring 30 days
after the sale date.

If you approach a REA with conflicting goals, you are practically asking to be
'gamed'!

This clarification should make it easier to (in theory) align the REA's
interests with yours.

In practice, as you recognise below, it's necessary to take into account that
the REA has more properties to sell than yours and that -- quite possibly -- he
has finite capacity.

If you really want to maximise your equation, it may be that it's necessary to
sell the property yourself!|=


Background info: Over here an REA generally receives somewhere between
1 - 2% of the selling price (for private homes). I assume his goal
will be something like "maximizing profit" with behaviour that is
focused on short term transactions and not so much on long term
relationships.
What worries me is that in this classic model a plus or minus of say
10k (Euro in this case) in the selling price is not a very impressive
amount if you look at how much it contributes to the REA's goal. On
the other hand the effect it has on my monthly OE is significant
enough from where I stand.
I suspect that the REA will have a hunch that his time is dear if he
is to reach his goal, so spending less time on a deal is more
important than getting the best possible price. The assumption here is
that there is enough business for the REA "out there". I suspect that
this is the case for the reputable REA's. I know this does not
necessarily mean that he will accept a ridiculous price –he will still
have to get my OK – but I also think that the chance of getting the
best price is not optimal either.

So I am going to take alignment step 1: rewarding based on selling price.
I think that getting a better alignment on selling price is not too
difficult. It is possible to get a figure for the market value of the
house that both the REA and I can agree on. An REA should be able to
sell the house for that price in little time (within a few months)
without performing magic. If he is able to sell the house for a higher
price he is adding value in my perception and I am willing to reward
that. So, what I'm thinking of is to give the REA a rather significant
share of the extra value generated (say 25% for all extra generated
value, or a variable, increasing percentage). I think that should make
it far more interesting to perform some magic. But that is only one
dimension covered. Looking at his goal the REA might now be tempted to
hold off potential buyers for the next 15 years hoping to get an even
better deal.

Alignment step 2: security
I am happier when the REA closes a deal sooner. Now how am I going to
add this to the equation? I am not totally sure yet. I am thinking of
setting a time limit for him to close a deal after which the contract
is canceled without the REA receiving any payment. Maybe there should
be some room for extending this deadline in a situation in which
serious candidates are still in negotiation. I suspect that when the
REA is interested in my payment model in the first place, he is
someone who expects to get a high bonus. In that case losing this
contract is a pain. I will have to make sure that the deadline is set
in such a way that the REA has a fair chance to get a good price while
I have sufficient time to start all over again with a different REA if
need be. I might also have to add an option to cancel the contract
sooner if there is no progress at all.

What I still haven't secured in this model is my ideal transaction
date. I don't want to move twice and I don't want to keep two houses
too long, as that is a costly hobby. This need feels less important
and is negotiable when price and security are covered well enough.
Still, looking at rewarding magic, I could introduce some sort of
penalty based on costs incurred (two houses too long, or extra costs
incurred by moving to a temporary home).

Reactions, ideas?
Do you think this will accomplish anything (positive)?

TIA.

---Lars.

=|JRM: See my embedded comments above.|=