Rand report here.. thought I’d excerpt some of this on topics (economic analysis of fires) we have discussed on this blog…
from pages 27-30.
There are several important costs and resource value changes that we
did not include in our analysis because the data did not exist or because
the data or methods required to estimate the values were not yet sufficiently
well developed to justify including them. These included nonmarket
values, federal disaster assistance, timber losses, and public
There is a growing literature on how fires affect nonmarket resources,
such as wildlife habitats, watersheds, public health, cultural heritage,
and recreational services. But the data and methods required to quantify
these resource changes are not yet adequate to generate compelling
estimates of these nonmarket values (Abt, Prestemon, and Gebert,
2008; Hesseln and Rideout, 1999; Venn and Calkin, 2007).
Results from the small number of studies that have attempted
such valuations reveal complex effects of wildfires that may correspond
to large positive or negative value changes and to changes that
evolve dynamically over time. For instance, Englin, Holmes, and Lutz
(2008) reported complex time-varying effects from fires on recreational
demand, with demand increasing in the years immediately following a
fire but significantly decreasing in later decades.
Similarly, some fire effects result in both positive and negative
changes, such as the effects of fire on watersheds. Typically, fires
increase water yield, but they also increase sedimentation in the water.
Potts, Peterson, and Zurring (1985) attempted to value these positive
and negative outcomes, finding that the benefits of additional water
availability exceeded the costs of additional sedimentation by a factor
of more than 1,000 in some regions.
Other studies have suggested that the social value of preventing
large fires may be extremely high. For instance, using a contingent
valuation approach to estimating willingness to pay, Loomis and
Gonzales-Caban (1998) found that the societal value of protecting
the first 1,000 acres of northern spotted owl habitat in California and
Oregon amounted to $25 per household, a figure that Venn and Calkin
(2007) note is “greater than the annual national fire suppression expenditure
by the Forest Service in recent high cost firefighting years.”
The uncertainties in nonmarket values have led recent Forest Service
guidance on cost-risk analyses to suggest estimating the minimum
value of nonmarket effects that would be implied by available interventions,
rather than trying to estimate the nonmarket effects themselves
(Calkin, Hyde, et al., 2007).
Federal Disaster Assistance
In principle, federal disaster assistance funding to states, localities, and
individuals affected by wildfires should be known to the federal government.
However, a recent Congressional Research Service report on
wildfires reported that public data on Federal Emergency Management
Agency fire disaster assistance were not available (Gorte, 2006).
A portion of this funding goes to reimbursing states and local
governments for the costs they incur in suppressing wildfires. Because
we have already estimated state and local suppression costs, including
the costs of federal reimbursements in our calculations would result in
double-counting of some suppression costs. Therefore, this portion of
the federal disaster assistance budget is omitted from the analysis.
While past studies have attributed considerable financial value to lost
timber (e.g., 20 percent of total losses for four large fires reviewed by
Abt, Prestemon, and Gebert, 2008), such estimates reveal widely varying
assumptions about the commercial viability of the lost timber, the
proportion of existing timber lost in burned areas, and the value of salvageable
burned timber. Some estimates of timber losses consider only
Forest Service timber lease values or lost Forest Service timber sales,
while others have considered the likely wealth transfers resulting from
the market effects of shocks to the timber supply.
Butry et al. (2001), for instance, examined the welfare effects of
the timber price reductions resulting from a short-term glut of salvaged
timber and later price increases resulting from local shortages of timber,
which they hypothesized would follow the 1998 wildfires in northeast
Florida. Their analysis suggested that short-term gains to consumers
are roughly matched by long-term gains to producers, with a net loss
resulting from the effects on owners of damaged timber.
We adopted a social cost perspective for our analyses, as opposed
to considering just the cost to the Forest Service. As such, to include
timber losses in our analysis, we would have required an estimate of the
welfare changes resulting from timber losses that are net of the types
of wealth transfers described by Butry et al. (2001). While that study
produced such an estimate for one fire in 1998, we do not believe that
these results can be generalized to timber losses nationally, and other,
more general estimates of timber resource value changes of this kind
were not available for us to incorporate into our analysis.
Public Health Effects
Fires degrade air quality in ways that are harmful and can exacerbate
asthma and bronchitis, reducing quality of life, increasing hospital
admissions, and contributing to deaths. But attributing the prevention of morbidity and mortality to fire suppression is complicated by uncertainties and great variation in the numbers of people affected by
individual fires, the severity of harms they might be expected to suffer,
and the valuation of those harms.
Sorenson et al. (1999) noted that, during the 1998 Florida wildfires,
admissions at some regional hospitals increased 91 percent for
asthma and 132 percent for bronchitis over the same period in the
prior year, though the atmospheric conditions that contributed to fire
risk in 1998 could have affected respiratory conditions as well. Butry
et al. (2001) used treatment costs as a proxy for the societal costs of
smoke from the same Florida fires, concluding that they represented
a small cost relative to other costs of the fires. Similarly, Rittmaster et
al. (2006) evaluated the health effects of a large fire in Canada. Unfortunately,
the results of each of these studies are likely unique to the fires
they investigated. We know of no analyses offering national or peracre-
burned estimates of the public health costs of smoke, so we could
not include these values in our analysis.
Next, we discuss the estimated costs of prospective aircraft. Ultimately,
it is these aircraft costs against which we compared the estimated
costs of large fires.