Chapter Six

Twelve Month Moving Premonitions

Before we could make our mini indispensable, we
discovered that there were preliminary hurdles we had to
cross, because the sight of simple things like routine hardware
and software maintenance sent the natives scattering like
nervous sparrows. It was as if feeding a cartridge into the
machine might expose them to something contagious, obscene or
at least uncouth. To our relief, they were not overtly hostile;
they had no trouble at all in referring to it as rubber ducky, but
in a tone that revealed more jitters than pure disparagement. It
was more a case of being afraid to be curious. Their avoidance
did not extend to making the computer's presence unspeakable.
In fact, they almost vied for the opportunity to make a clever
remark about our little beast. This would be the hook we
needed to draw them in. We needed to be quick, to get the game
afoot, to keep the ball in play. Before the novelty wore off, we
had laid a threefold plan of marketing, service and
entertainment.


Our marketing objective was straightforward; create a
positive aura that would induce the natives to consider our
overtures. At the very least, it was important to keep the
tribes engaged so they could not ignore our feats, diminishing
them into insignificance, virtual oblivion. With the power to
solve so many of their dilemmas it seemed so frustrating that
we should need a campaign to overcome their reluctance. We
responded warmly to any comer but our main tactic was to
repeat the successful stir created by our mini's accidental
naming. Carefully, I christened each of our ventures to invite
their curiosity, to amuse and provoke delight, to calm their
tribal jitters, to stay always in their mind's eye.


The actuarial staff, by contrast, withdrew into their
officially defined spaces, aloof on good days and arrogant on
bad ones. It was not uncommon to hear a phone slammed down,
to meet a glowering countenance. There was no measuring the
bad grace of the trainee actuaries or their chief, none of whom
was ever seen to consult or mingle. This clearly helped us now
and again but it made it difficult to properly gauge their
strength.


My unofficial staff's natural sensitivity to the tribe's
feelings and our eagerness to ease the chiefs' discomfort, to gain
their trust, was a matter of considerable pride among us. These
skills were so undervalued in the market and their value so
apparent to us that we may have over-valued them, but only in
the same proportion as we felt would compensate for the
market's injustice. This part of our strategy clearly aided our
marketing effort but it made it impossible not to despise our
adversaries for their deficiency, an unhealthy attitude that
has frequently been a serious source of strategic error in
campaigns, and will certainly be again.


But for the moment, their bad grace might be made use
of, and so we undertook an extensive plan of entertainment...

... Since premium-year data was reported only annually, I
needed to find an approximation that was available more
frequently. The closest choice for my basis was the
accident-year data that I had encountered on my forays into claims
and accounting. Claims were reported monthly but were categorized
by accident-year, which, with some calculating sleight of
hand, could be used as a surrogate for the claims component of
premium-year data. The trick was to replace the written
premiums that accounting recorded with a construction called
earned premium which would then serve as the corresponding
surrogate for our needed ´accident-year premium´. Earning a
month´s premium was a straightforward process of allocating
the total premium written over the previous twelve months
into monthly segments and then tallying the segments that
belonged to the current month. This amount gave an estimate of
the risks to be covered in that month. It measured our exposure
and should have been comparable to the accidents occurring
over that time period if the premiums in force were adequate
for their intended risks.





Where the planning project had been fundamental
number-crunching, this bridge was a high-tech statistical
construction that gambled on the limited histories we had
gleaned from accounting as well as assumptions about
appropriate curves to fit them. The histories of accident year
data had one more drawback; they were recorded in year-to-date
form. Since this form was extremely unstable for the first
few months of the year, it was not only potentially deceptive
during those early months but it also made it difficult for the
chiefs to use this form as a basis for any corrective action until
much of their maneuvering time before year-end was gone.
Though it suited the chiefs in accounting, it was less than
satisfactory for the decision needs of senior chiefs. For those
decision purposes, which it was our intention to support, we chose to
calculate a more appropriate form called twelve-month-moving data,...

© JHRaichyk