there’s no such thing as maintaining the status quo.
However, this analytically driven competition has fed a cycle
of consolidation among personal auto writers. Figure 1 illustrates the number of groups writing personal auto insurance
over time. For many years, the number of groups was relatively constant. But in 1995—at the dawning of the analytics
arms race—the trend shifted sharply. Since 1995, the number of
groups has decreased by more than 100—more than a third have
disappeared. And the decreasing trend has yet to bottom out.
The consolidation activity is further evident when we look
at the portion of personal auto premium written by the top 10
and top 50 insurers over time (see Figure 2). Not only are fewer
groups writing personal auto insurance, but the biggest companies are becoming bigger. In part, this is due to the competitive
advantage that larger carriers have gained by investing in and
implementing new predictive analytic capabilities in their rate-making and underwriting processes.
How predictive analytics fuels competition
A simple example helps to illustrate how the development of
more refined loss estimates can create such competition. It’s a
case study in how adverse selection works in a competitive marketplace. In this example, there are only two insurers—we’ll call
them Laggard Insurance and Luminary Mutual. At the beginning,
both have identical books of business consisting of three policies
each. Each policy has a different expected cost, and recent predictive information can accurately identify those expected costs.
Laggard hasn’t implemented a refined rate plan using the information, so it charges the same premium for all policies. Luminary
has implemented a refined rate plan, so the rate it charges equals
the actual expected cost for the policy (see Figure 3, Page 48).
In the initial scenario, both insurers have the same revenue
and profit expectations. However, as a result of the difference in
policy premiums, marketplace dynamics will change this picture.
Because Luminary has the refined rate plan, it’s able to offer a
$660 premium to the policy with the $600 expected cost, which
is a much lower premium than the $880 Laggard charges. This
policy will be lost to Luminary. On the flip side, Luminary will lose
its highest-cost policy to Laggard because Laggard underprices
this risk. See how the books of business now look in Figure 4.
FIGURE 1
FIG#
FIGURE 4
FIG#
FIGURE 7
FIG#
number of personal auto insurance groups
FIGURE 1
FIG#
FIGURE 2
FIG#
FIGURE 3
FIG#
FIGURE 4
FIG#
FIGURE 5
FIG#
FIGURE 6
FIG#
FIGURE 1
600
FIGURE 7
FIG# 500
400
300
200
100
0
1980 1985 1990 1995 2000
Source: ISO analysis of data acquired from A. M. Best Co. Inc.
2005
consolidation of auto insurance markets FIGURE2
FIG#
FIGURE 3
FIG#
FIGURE 5
FIG#
FIGURE 6
FIG#
FIGURE 2
100
Percent market share
90
80
70
60
50
40
1995 2000 2005