Is fire insurable? 117
to estimate rebuilding costs, but it is often impractical to refer to builders,
architects or quantity surveyors. Many insurers now provide consumers
with access to web-based calculators.’ (ASIC 2007, p. 10). Or more simply
put, most of the time the estimate of the rebuilding cost of a house, if there
is one, comes from the insurer. If the estimate is wrong then the insurance
is wrong.
But that does not, in and of itself, explain underinsurance. Because there
is no precise, mathematical, objective cost for the rebuilding of a home.
Until a home is actually being built, it always is an estimate that will to one
degree or another, in one direction or another, be wrong.
One would expect, however, that estimates of coverage to break evenly
high and low, meaning the rate of underinsurance should group around the
50% line, and distribute evenly above and below. And it does not. Figure 9.1
illustrates that there are dramatically more instances of underinsurance
than of over-insurance.
One way to understand underinsurance rates disproportionately clustered above the 50% line is to think of them as akin to the idea in mathematics called a mathematical fallacy. In broadest terms, a mathematical
fallacy is when the conclusion of a proof suggests that there is a flaw in the
proof, even if the flaw cannot be identified. Think, for example, of a coin
flipped 1000 times. 700 times it comes up heads. 300 times it comes up tails.
The experiment is repeated. Now the results are 650 heads, 350 tails. It is
repeated again. 800 heads. 200 tails. Something is wrong. Maybe the coin is
weighted unevenly. Maybe something else is going on. But it bears investigation. Because something may be amiss.
What may be amiss with dwelling insurance? Why are more homes underinsured than over-insured?
In 2007, ASIC reported, ‘Even if a consumer correctly estimates what it
would cost to rebuild their home in a one-off total loss, it is almost impossible to know what it will cost to rebuild a home that is destroyed in a mass
disaster. The surge in building prices that occurs after a mass disaster can
be very unpredictable.’(ASIC 2007, p. 13). This idea – demand surge – also
is proffered in the United States (Klein 2019, pp. 69–71). The premise of the
proffered explanation –’the surge in building prices that occurs after a mass
disaster can be very unpredictable’ – bears further study. Catastrophe modelers (creating data streams for vendors who sell costs estimators) contend
they can predict natural disaster with granularity down to a specific home
address (Raizman & Pratt 2021, 1:12;10–1:31:55).
Whether this granularity of modelling is real or not, however, the data
suggests that demand surge alone is an inadequate explanation. In the wake
of the 2017 Tubbs Fire in California, CoreLogic studied demand surge and
found it averaged 15%–30% (Kopperud 2019). In a Market Conduct Study
in 2010, the CDI found, however, that approximately 57% of homes that had
purchased an extension of their full replacement coverage still were underinsured (CDI 2010, pp. 1027–1030).
to estimate rebuilding costs, but it is often impractical to refer to builders,
architects or quantity surveyors. Many insurers now provide consumers
with access to web-based calculators.’ (ASIC 2007, p. 10). Or more simply
put, most of the time the estimate of the rebuilding cost of a house, if there
is one, comes from the insurer. If the estimate is wrong then the insurance
is wrong.
But that does not, in and of itself, explain underinsurance. Because there
is no precise, mathematical, objective cost for the rebuilding of a home.
Until a home is actually being built, it always is an estimate that will to one
degree or another, in one direction or another, be wrong.
One would expect, however, that estimates of coverage to break evenly
high and low, meaning the rate of underinsurance should group around the
50% line, and distribute evenly above and below. And it does not. Figure 9.1
illustrates that there are dramatically more instances of underinsurance
than of over-insurance.
One way to understand underinsurance rates disproportionately clustered above the 50% line is to think of them as akin to the idea in mathematics called a mathematical fallacy. In broadest terms, a mathematical
fallacy is when the conclusion of a proof suggests that there is a flaw in the
proof, even if the flaw cannot be identified. Think, for example, of a coin
flipped 1000 times. 700 times it comes up heads. 300 times it comes up tails.
The experiment is repeated. Now the results are 650 heads, 350 tails. It is
repeated again. 800 heads. 200 tails. Something is wrong. Maybe the coin is
weighted unevenly. Maybe something else is going on. But it bears investigation. Because something may be amiss.
What may be amiss with dwelling insurance? Why are more homes underinsured than over-insured?
In 2007, ASIC reported, ‘Even if a consumer correctly estimates what it
would cost to rebuild their home in a one-off total loss, it is almost impossible to know what it will cost to rebuild a home that is destroyed in a mass
disaster. The surge in building prices that occurs after a mass disaster can
be very unpredictable.’(ASIC 2007, p. 13). This idea – demand surge – also
is proffered in the United States (Klein 2019, pp. 69–71). The premise of the
proffered explanation –’the surge in building prices that occurs after a mass
disaster can be very unpredictable’ – bears further study. Catastrophe modelers (creating data streams for vendors who sell costs estimators) contend
they can predict natural disaster with granularity down to a specific home
address (Raizman & Pratt 2021, 1:12;10–1:31:55).
Whether this granularity of modelling is real or not, however, the data
suggests that demand surge alone is an inadequate explanation. In the wake
of the 2017 Tubbs Fire in California, CoreLogic studied demand surge and
found it averaged 15%–30% (Kopperud 2019). In a Market Conduct Study
in 2010, the CDI found, however, that approximately 57% of homes that had
purchased an extension of their full replacement coverage still were underinsured (CDI 2010, pp. 1027–1030).