The power of the database is its depth. Assuming you could get 10% of
pilots
or so to use it, it would give us a lot more insight into this market.
Assuming that insurers are indeed milking us, or certain segments of the
market, this would also provide a data set that would help convince a new
company to enter the market.



Well, what I am shooting for is for a more responsive insurance market that
provides information on the safety of the planes through pricing. In the
long run, the more dangerous planes would be reduced, while the safer ones
would thrive, and the overall result would be added safety with lower costs
for everyone. The insurers would supposedly benefit from lowered claims
unless you believe in the theory that they LIKE larger claims.



I doubt it has that much effect. We're talking about toys that cost
anywhere
from a quarter to half million dollars, and it is a pretty rarefied group
of
people that can afford that. Yes, there are probably a few marginal
customers who just can't justify another $500 per month to own an SR-22
versus a new 182, but I doubt it's significant.


You know, I hate to admit this, but you seem to be absolutely correct.
People ARE buying a new Cirrus despite the price of insurance. However, I
think buyers may be different from the used ones.

This makes me wonder what happens to the resale values. How much is the
free training worth as part of a new Cirrus. If you want a used one, do you
haev to pay for the school to get insured?


A much more interesting insurance question right now is the Sport
Pilot/LSA
segment of the market. This is going to be much more price-sensitive and
potentially a lot larger. It will be interesting to see how this evolves.


I can't figure out how it will get very large unless Sport pilots are
allowed under the class B umbrella. I know they are not allowed in the
Bravo, but can they go under? Can the LSA's go into B with a PP as PIC? Is
that all decided?


Okay, I hear you. But what if Avemco only discounted the Mooneys, and
charged the same or more for the others (based on Richard Collins data
being
proved out in claims)? Would they not then be more profitable than the
competition by attracting more than their fair share of the better
retract
business?

Well, I suspect this *is* going on, particularly with light twins. If you
read insurance threads here you'll often see cases where one insurer
offers
a significantly (15%) lower rate than the others. The issue is that
information moves much more slowly. Airlines, for instance, change fares
constantly, but they are all published onto the SABRE network in near-real
time so competitors see very quickly what's going on and can respond in
kind
if desired. Insurers probably need a minimum of a few months to see these
sorts of trends. Again, this is a case where a master database could help
accelerate things.

However, it does have a possible downside in that it diminishes the value
of
price-cutting. If an insurer starts offering significantly-reduced rates
on,
say, Mooneys, it will take some time before the other insurers notice. In
this time they will scoop up a lot of customers. Then the others will cut
their rates too, at which point the advantage will disappear. So the more
time it takes for your competitors to realize you cut your prices, the
higher the ROI on your price cutting. Now, you also need to consider that
cutting prices will initially cost you money since you're also going to be
offering lower rates to customers you already have. So the key is to catch
enough new customers to make up for lost revenue from existing ones. If
your
competitors can respond to price cuts more or less instantly, then it
eliminates the incentive to do so. This, coincidentally, explains why the
"we will not be undersold" guarantees you see in ads are actually ways of
discouraging price competition.

So in the end the key is to have a lot of companies in the market. This
way
you always have someone upsetting the cozy equilibrium that favors the
insurers and forcing everyone else to come along. The
four-is-few-six-is-many rule is derived by observation, and there remains
a
Nobel to be won by the economist who gives a good crisp mathematical
justification for it.


That's all good stuff, but it seems to me that if you lower the price on the
"good eggs" then you might have to raise it on the "bad eggs" to make up for
that. If you can successfully drive the more claims ridden planes to your
competitor you can really stick it to him, and he may never catch up. Even
if he does, you will have a stack of cash for your next move that he will
not have.


Also, what you say brings up an idea. Perhaps one year is not enough
data
for claims because each insurer does not have a wide enough pool.
Perhaps
they all need to provide their claims data to a third party, and then buy
back the overall fleet results in order to change rates to reflect the
total
fleet results.

I don't see how this would benefit the insurers. Assuming they are
overcharging, they have no reason to want to stop.


They are not overcharging, they are not being discriminating enough. This
has too affects. One, it raises claims because it does not discourage the
use of poorly designed planes. Two, it reduces overall safety by the same
mechanism.


with predictable results. Time will tell. Either way, rates will not
come
down without a pretty substantial reduction in accident rates and no
one
is
predicting that for anybody.

Hasn't Diamond had a reduced incident as well as fatality rate?

Perhaps, I don't know. Again, the key is to figure out why. Perhaps the
reputation of the Diamond as a "safe" airplane attracts safety-oriented
pilots who are going to be safer no matter what airplane they fly. I am
very
open to believing that a plane can be made more crash-worthy, and the
Diamond clearly is. I am less persuaded that there is that much more to be
done to make planes safer to fly in day VFR conditions. Is there that much
that can be done to improve stall-spin characteristics? Can we make planes
that much easier to land in crosswinds? Of course, there is always
something, but most accidents begin with bad judgment.


I have to say that Volvo does not make the safest cars by crash statistics,
yet I believe they do benefit from exactly this phenomenon. There is sense
in the safer population theory, but I think its so marginal as to be almost
a non factor. After all, did Cirrus NOT attract safe minded individuals
with the parachute and all their marketing?


Ahhh, but they did treat Diamonds like Cessna's and appear to be making
out
like fatcats. Besides, if they really do take that approach, isn't it
just
proving my point that they reduce innovation, and are therefore reducing
safety?


Well, this is the way of all flesh. Prices tend to go up quickly, and come
down slowly. Yes, no question insurers do occasionally milk certain market
segments.


PS I really appreciate your perspective on this, you are helping me
reshape
my opinions and sharpen some of my arguments.


IMHO the real problem is not insurers, it's the FAA certification process.
To answer one of my own questions, it would seem that the data stream
available in the G1000 ought to be sufficient to construct a warning
device
that could predict many of the potential stall-spin scenarios. For
instance,
if you're buzzing around pattern altitude near a field and have the
traffic
frequency tuned, you're probably flying the pattern. Now, let's watch the
airspeed trend, and sound an alarm if it starts slowing down rapidly on
the
base-final turn. "Speed up, speed up!" would probably prevent a
non-trivial
number of such accidents, though surely not all. You could construct
similar
routines for plenty of other scenarios. However, getting this approaved by
the FAA and your company's legal department would be a nightmare. Neither
of
these have *anything* to do with the insurance companies.

-cwk.


I here you there.