Creation and Implementation of a Certification System for Insurability and Fire Risk Classification for Forest Plantations

Abstract

Currently, the Chilean insurance market sells forest fire insurance policies and agricultural weather
risk policies. However, access to forest fire insurance is difficult for small and medium enterprises (SMEs),
with a significant proportion (close to 50%) of forest plantations being without coverage. Indeed, the insurance
market that sells forest fire insurance policies does not have an evaluation system based on known
standards accepted by the market; rather, it is discretionary, with each company evaluating risk with their
own specific parameters. This lack of standardization reduces access and increases costs due to uncertainty,
assigning a high risk rating by default. At present, over 19,000 SMEs and more than 810,000 forested hectares
are affected by difficulties in getting fire insurance policies.
This situation is being addressed through a project funded by the Chilean Foundation for Agriculture
Innovation (FIA), along with government institutions, forest companies, insurance companies, and insurance
adjusters and brokers, and executed by the Chilean Forest Institute (INFOR). The proposed solution is a
certification system of fire risk classification for forest plantations (named SAFOR –Sello de Asegurabilidad
Forestal- as per the Spanish acronym), that will be created and implemented, allowing the SMEs to have access
to the insurance market.
The system has two elements, the standard, built as a formal national norm that describes objectively how
to evaluate the risk, and the certification system that will transfer the standard into the market based on an
appropriate accreditation of auditors through the fulfillment of certain procedures.
The implementation of SAFOR certification is expected to have several different benefits and positive impacts,
improving several indicators. Forest companies with plantations close to SMEs plantations may also benefit
from a decrease in fire risk. The adoption of fire management best practices that landowners may apply to
reduce the risk and therefore obtain lower insurance premiums could reduce the fire occurrence, contributing
to the environment and also to the State investments, since owners of plantations established with State
contribution could continue in the forest business after a fire, preserving the rural economy

Full text

USDA Forest Service Proceedings RMRS-P-73. 2015. 141
Keane and others
Introduction
Chile has a population of 16.6 million inhabitants, a per
capita Gross Domestic Product (GDP) of US$ 18,945, a total
continental area of 756,000 km2, a forest area covering 17.3
million hectares, with 2.3 million in plantations, and total
annual forest exports amounting to US$ 5,714 million (INE
2014).
Forest plantations are essential in the forest activity since
they supply 98% of the timber used by the industry. Forest
ownership is concentrated in two big companies that hold
over 50% of the total forest plantations (Op cit.).
Fire statistics indicate a burned area of 1.2 million hect-
ares between 1990 and 2013, with more than 137,000 res
and losses of forest assets worth more than $50 billion a year
market value, plus other losses that have been estimated in
8-10 times that amount. Forest res are caused mainly by
human activities. The annual average burned area reaches
an average of 49,000 ha, and the re forest insurance mar-
ket is US$ 9.1 million (CONAF 2013).The Chilean insurance
market offers two products for forest plantations: initial in-
surance, which operates during the rst three years after
plantation and deals with weather damage (drought, exces-
sive rainfall, ood, frost, hail, snow and wind), and re for
new forest plantations; and the re insurance policies and
agricultural weather risk policies. Nevertheless, it is very
difcult for forest SMEs (small and medium-sized enter-
prises) to obtain re insurance policies for forest plantations
because the insurance market has few specialized agents
in forest plantations. As little quality information is avail-
able regarding re risk in forest plantations, agents typically
assign a high risk qualication by default. This increases
Creation and Implementation of a Certication System for
Insurability and Fire Risk Classication for Forest Plantations
Verónica Loewe M., Chilean Forest Instute (INFOR). Metropolitan Oce. Sucre 2397, Ñuñoa, Sanago, Chile. vloewe@
infor.cl.; Victor Vargas, Chilean Forest Instute (INFOR). Bio Bio Oce. Camino a Coronel Km. 7.5, Concepción, Chile.
vvargas@infor.cl.; Juan Miguel Ruiz, FORESTSUR. Las Urbinas 53, Of. 81, Providencia, Sanago, Chile; Andrea Alvarez C.
Chilean Forest Instute (INFOR). Metropolitan Oce. Sucre 2397, Ñuñoa, Sanago, Chile. aalvarez@infor.cl.; and Felipe
Lobo Q, Chilean Forest Instute (INFOR). Bio Bio Oce. Camino a Coronel Km. 7.5, Concepción, Chile. obo@infor.cl.
Summary—Currently, the Chilean insurance market sells forest re insurance policies and agricultural weath-
er risk policies. However, access to forest re insurance is dicult for small and medium enterprises (SMEs),
with a signicant proporon (close to 50%) of forest plantaons being without coverage. Indeed, the insur-
ance market that sells forest re insurance policies does not have an evaluaon system based on known
standards accepted by the market; rather, it is discreonary, with each company evaluang risk with their
own specic parameters. This lack of standardizaon reduces access and increases costs due to uncertainty,
assigning a high risk rang by default. At present, over 19,00 0 SMEs and more than 810,00 0 forested hect-
ares are aected by dicules in geng re insurance policies.
This situaon is being addressed through a project funded by the Chilean Foundaon for Agriculture
Innovaon (FIA), along with government instuons, forest companies, insurance companies, and insur-
ance adjusters and brokers, and executed by the Chilean Forest Instute (INFOR). The proposed soluon is a
cercaon system of re risk classicaon for forest plantaons (named SAFOR –Sello de Asegurabilidad
Forestal- as per the Spanish acronym), that will be created and implemented, allowing the SMEs to have ac-
cess to the insurance market.
The system has two elements, the standard, built as a formal naonal norm that describes objecvely how
to evaluate the risk, and the cercaon system that will transfer the standard into the market based on an
appropriate accreditaon of auditors through the fulllment of certain procedures.
The implementaon of SAFOR cercaon is expected to have several dierent benets and posive impacts,
improving several indicators. Forest companies with plantaons close to SMEs plantaons may also benet
from a decrease in re risk. The adopon of re management best pracces that landowners may apply to
reduce the risk and therefore obtain lower insurance premiums could reduce the re occurrence, contribut-
ing to the environment and also to the State investments, since owners of plantaons established with State
contribuon could connue in the forest business aer a re, preserving the rural economy.
In: Keane, Robert E.; Jolly, Matt; Parsons, Russell; Riley, Karin. 2015. Proceedings
of the large wildland res conference; May 19-23, 2014; Missoula, MT. Proc.
RMRS-P-73. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky
Mountain Research Station. 345 p.

142 USDA Forest Service Proceedings RMRS-P-73. 2015.
policy costs because of the uncertainty. Therefore, the lack
of standardized information, which affects over 19,000
owners and more than 810,000 hectares of plantations,
poses serious barriers to forest SMEs, with important socio-
economic implications.
An ongoing project intended to solve the problem of mar-
ket asymmetry affecting forest SMEs, funded by the Chilean
Foundation for Agriculture Innovation (FIA) along with
government institutions, forest companies, insurance com-
panies, insurance adjusters and insurance brokers, is being
implemented by the Chilean Forest Institute (INFOR), with
a total budget of US$ 410,000, allocated in three years (2013
to 2016). The project is implemented by an advisory council
formed by representatives of each associated member, FIA
and INFOR, which contributes with and validates results
and decisions.
Forest plantations, a very valuable asset for the country,
are subject to multiple risks, including re. However, a sig-
nicant proportion are not covered by re insurance, with
several negative implications, especially for SMEs, whose
proportion of uninsured plantations is higher than that of
large companies, increasing the gap between socioeconomic
stakeholders. This gap is caused by a lack of a standardized
methodology, hindering assessment of re risks by insur-
ance companies.
A certication system of re risk classication for for-
est plantations (named SAFOR as per the Spanish acronym),
will be created and implemented, allowing the access of
SMEs to the insurance market.
The solution has two elements. The rst element is a set of
standards, built as a formal national norm generated by the
Chilean Standardization Institute (INN), which describes
how to evaluate the risk in a methodological and objective
way. The other element is the certication system, which
will enable plantations to demonstrate that they meet the
designated standards based on an appropriate accreditation
of auditors by following certain procedures, and facilitate
the commercialization process of re insurance policies.
The insurance market will adopt the certication system as
a source of information to qualify re risk associated with
insurance policies. In turn, the clients will pay for the certi-
cation to break the entrance barriers by evaluating the risk.
Thus, policy price should decrease and the number of SMEs
insured should increase, allowing forest SMEs to hire re
insurance policies at sustainable prices, according to the for-
est activity. The forest re insurance market is expected to
grow as a result of the adoption of the SAFOR certication
system, because it will provide reliable information for the
evaluation of re plantation risks. In addition, the following
potential benets have been identied:
-Increase of the insurance forest market size and greater com-
petition, because new companies are expected to join the
activity;
-Improvement of self-protective management, including an
increased fuels management by plantations, in order to get
a lower risk qualication;
-More controllable res;
-Improved access for SMEs to the nancial system and to
government promotion tools/funds due to the risk reduc-
tion;
-Better information on SMEs technology and productivity;
-The forest subsidy allocated to the establishment of new
forest plantations will be protected, increasing its socio-
economic protability;
-Environmental benets, such as reduction of impacts to soil,
water and biodiversity, as well as reduced air pollution and
greenhouse gases.
It should be noted that the need to evaluate risk to a level
of sophistication that will meet the needs of the insurance
industry reaches far beyond Chile.
Development Process
The proposed development process is organized in four
phases:
1. Feasibility analysis, considering forest sector conditions,
insurance market, legal requirements, technical issues to
be used in the standard construction, and the potential
client’s willingness to pay through a contingent valuation
methodology according to Hannemann (1984), Arrow and
others (1993), Vásquez and others (2006) and Tietenberg
(1996). Sample size was determined according to Piñeda
(2004).
2. Development of a National Standard to quantify the re
risk level in forest plantations. The standard development
process follows the national authority scheme, dened in
the national standard 1 (NCh 1 cR2010), and considering
the risk management standard (NCh-ISO 31,000).
3. Design, creation and validation of the certication system.
4. Implementation of the new certication as a pilot project.
Due to the project characteristics, special importance has
been given to joint public-private collaboration because of
its potential to expand and its long-term sustainability.
Figure 1—Demand target market segmentaon.
Loewe and others Creaon and Implementaon of a Cercaon System...

USDA Forest Service Proceedings RMRS-P-73. 2015. 143
Among the main stakeholders involved are the insurance
market, SMEs and small owners of different types (Figure 1)
from the private sector, the Forest Institute (INFOR), Forest
Service (CONAF) and the Ministry of Environment from the
public sector.
Results
Since the project has only recently begun to develop the
standard, here we present the results for the rst phase, which
consists of a feasibility analysis. This feasibility study con-
ducted and completed ve different analyses: 1) forest sector
and insurance market overviews, 2) legal aspects, 3) an exam-
ination of the insurance market in Chile, 4) socio-economic
issues focusing on willingness to pay, and 5) forest manage-
ment procedures. The main results (Loewe and Vargas 2014)
are summarized below:
Forest Sector and SMEs Baseline
The study conducted by Loewe and others (2014) showed
that forest plantation ownership is strongly concentrated,
with two companies owning more than 50% of all planta-
tions, distributed in four regions (72% from Maule to Los
Ríos). On the other hand, SMEs plantations are concentrated
from O’Higgins to Los Rios (88%) (Table 1) (INFOR 2013).
However, little information on SMEs was available concerning
how they are divided between SMEs and Small and Medium
Owners (SMOs), being the threshold, for statistical purposes,
among small and medium size owners 200 hectares.
Plantations of big forest companies are mostly devoted
to radiata pine (77.3%) and Eucalyptus globulus (12.1%),
whereas SMEs and SMOs have a more diversied resource
(50% of radiata pine and 35.9% of Eucalyptus globulus).
Regarding age structure, most plantations are 0-20 years
old. Age structure is more spread across age segments in
the big companies than in SMOs/SMEs, reecting a lack of
planning due to fragmentation by the latter.
Tenure of plantaons in SMOs/SMEs at the municipality
level
In O’Higgins region, four municipalities (Pichilemu,
Marchigüe, Litueche and Paredones) accounted for 70% of
the forest plantations that belong to SMOs, located in the
regional coastal zone. In Maule region, the municipalities
Cauquenes, Constitution, Curepto, San Javier and Curepto
accounted for 46% of SMOs/SMEs forest plantations, also
located in the coastal region (INFOR 2013).
In Bio Bio region, about 25% of forest plantations with-
in this socioeconomic segment were concentrated in ve
municipalities (Quirihue, Mulchén, Los Angeles, Santa
Barbara and Cobquecura), whereas in the Araucanía re-
gion, six municipalities located in the central valley (Los
Sauces, Traiguen, Lumaco, Galvarino, Nueva Imperial and
Angol) accounted for nearly 50% of plantations.
Finally, in Ríos region, six municipalities (La Union,
Los Lagos, Corral, Valdivia, Mariquina and Paillaco) ac-
counted for 70% of forest ownership, and in Los Lagos
region, four municipalities (Fresia, Purranque, San Juan
and Rio Negro) accounted for 65% of their forest patr imony
at the regional level.
Fire Stascs
Fire statistics are produced annually by the National
Forest Service (CONAF) from data on re occurrence and
affected area by season (from July 1st to June 30th of the
following year).
From 1990 to 2013, the burned area totaled 1.2 million
ha, with 137,000 individual res, and losses of assets with
a commercial value averaging over US$ 50 million per year
(Table 2) (CONAF 2013). However, non-commercial losses
were much more critical because of their impact on the en-
vironment and quality of life for inhabitants, both in rural
and peri-urban areas of towns and cities. These impacts,
Loewe and others Creaon and Implementaon of a Cercaon System...
Table 1—Area of forest plantaons by owner and region.
Big Companies SMEs
Region (Hectares, %) (Hectares, %)
O´Higgins (VI) 7,725 7 105,039 93
Maule (VII) 247,831 55 204,971 45
Biobío (VIII) 654,496 72 252,860 28
Araucanía (XIX) 274,198 57 206,240 43
Ríos (XIV) 128,654 68 61,557 32
Lagos (X) 23,101 33 46,928 67
Total 1,336,005 60 877,595 40
Source: INFOR (2013)
Table 2—Fire occurrence and historical naonal damage (1990–2013).
Forest Service Companies Total
Total Fire N° 93,070 44,092 137,162
Average (N° of res/season) 3,878 1,837 5,715
Incidence (%) 67.9 32.1 100
Total aected area (ha) 983,420 192,537 1,175,957
Average (ha/season) 40,976 8,022 48,998
Incidence (%) 83.6 16.4 100
Protected area (ha) 34,000,000 3,000,000 37,000,000
Incidence (%) 91.9 8.1 100
Source: CONAF (2013)

144 USDA Forest Service Proceedings RMRS-P-73. 2015.
whose valuation is not straightforward, were estimated to
be 8-10 times higher than direct losses, being US$ 400-500
million/year, not including the incalculable value of loss of
human lives (Julio 2012).
On average, 49,000 ha are burnt each season, in about
5,700 events, with an average of 9 ha affected per event.
Nevertheless, the range of burnt area historically varied
widely, from a minimum of 10,000 ha in the 2000-2001
season to over 100,000 in the 1998-1999 season. There is
no relationship between occurrence and area affected; for
example, in the 2011/12 season, in Quillón about 25,000 ha
were burnt only in one re event.
The four seasons that caused the highest damage co-
incided with big res, and although they are known to have
been originated intentionally (simultaneous outbreaks), in-
vestigations have failed to identify culprits.
Interestingly, big forest companies had 32% of re oc-
currence but only 16% of the affected area, because they
control a smaller area and have detection and control sys-
tems (Op cit.).
In Chile, wildres principally originate from anthropo-
genic activities and are strongly linked to attitudes of the
value of the forests. If people assign a low value to the for-
est, they behave carelessly, negligently or recklessly; on the
contrary, if they assign a high value to the system they live
in, they will behave wisely. At the national level, for the
period 2003-20131, accidents accounted for 57% of the
area burned (including agricultural and forest residue
prescribed res, trash burning, neglected cigarettes
and res, transit and recreational activities, among
others); intentional res accounted for 26.8%, un-
known causes, 16% and natural causes accounted only
for 0.2%. Even though this is a complex reality, it has
the advantage that through education human behavior
related to re and re management can be improved.
The re statistics of the Forest Service comprises
326 municipalities, not including geographically ex-
treme regions. Of these, 29 concentrate over 50% of
wildres and have received prevention, control, sup-
pression and attack budgetary allocation since the
2010/2011 season.
Forest companies focus their efforts on the basis of
knowledge of physical parameters (vegetation, topog-
raphy, weather and re behavior) in order to protect
their forest resources. Forest companies intend to fos-
ter their protection programs and continue increasing
coordination and collaboration agreements with other
companies to strengthen self-protection. Accordingly,
the most relevant aspects for companies are coordina-
tion and collaboration agreements with other forest
companies, early detection systems, minimization of
re spread and attack resources.
Legal Analysis for the Development of SAFOR
Cercaon System in Chile
The implementation of a forest plantation insurability
certication was analyzed from a legal point of view (Lobo
2014), and was found to be quite feasible as no impediments
were identied in our legislation.
Given the legal nature of the insurance contract, the par-
ties may agree to include in the insurance policies terms and
conditions to quantify re risks in forest plantations through
a certication process, which could be accepted by insurance
companies when deciding on issuing a given policy and when
setting the premium cost.
Then the Superintendence of Securities and Insurance
(SVS), based on its legal powers, will validate these new terms
and conditions to be endorsed in case the amount of the annual
insurance premium is lower than a stated limit in support of
a transparent business for SMEs; if the parties do not comply
with this obligation, they cannot use a new policy or term. If
the annual insurance premium exceeds the previous amount,
and the insurance beneciary is a legal entity, the mentioned
obligation is no longer valid because it will be a private agree-
ment, with those parties being free from incorporating the
certication.
This conclusion is reinforced by the general rules of the
Commercial Code, whose Article 516 concerning the infor-
mation that policies must contain for all kinds of insurances,
states that the policy must describe “the situation of the in-
sured property,” without indicating how this statement should
be expressed.
Thus, there are no legal impediments for the parties to
agree that part of the condition of the insured resource is de-
termined by a third party who technically certies the re risk
of a forest plantation through the SAFOR standard.
The Insurance Market in Chile
The analysis performed by Forestsur (2014) includes two
parts. The rst part describes the structure of the insurance
market in general terms, including stakeholders (insured,
regulator, brokers, reinsurance, adjustors and support agen-
cies), the regulatory environment, business practices, market
dimension and insurance types, in order to explain the insur-
ance market functioning and role in the economy of Chile.
The second part describes the insurance market from the point
of view of supply and specic operation for forest plantations
in Chile.
General Insurance Market
An increasing trend in the industry is observed: life in-
surance companies have increased over time, exceeding the
general insurances.
Specically in insurances for SMOs and SMEs, there
would be a disadvantage in the products offered by the insur-
ance market (coverage), because they were not designed for
a specic customer segment or size; rather, they are generic,
so the insurance market can access a broad market (clients or
insured) but has the disadvantage that often small potential
customers face entry barriers due to business size or risk rate.
1 Stascs from CONAF, 2013.
Loewe and others Creaon and Implementaon of a Cercaon System...

USDA Forest Service Proceedings RMRS-P-73. 2015. 145
Moreover, the supply of insurances is growing, with a
variety of increasingly specic interests to satisfy insured
needs. In addition, there is a fairly detailed legislation on
what the various stakeholders in the industry may or may
not do which regulates the amount of investment, diver-
sication, policy contract components, etc., regulated and
supervised by the SVS.
The incidence analysis of the total annual premium dis-
tribution for the different insurance branches shows that
re and earthquake reach 35%, and excluding the compul-
sory automobile insurance, it would reach 50%; given the
high variety included in “others” (theft, liability and sever-
al others), re and earthquake branch is the most important
of all general insurances.
This market will probably continue to grow during the
next years, both in variety and size, i.e., increasing the num-
ber of actors, both insurance and reinsurance companies
and brokers, and it will improve the offered insurances, di-
versifying the investment portfolio.
The insurance industry in Chile is transparent and has a
stable institutional framework, governed by clear laws and
regulations, which has led to a sustained increase in invest-
ment value, reaching US$ 5 billion in 2012.
This has particular positive implications for SAFOR
project, because when demand is added to the insurance
market, it will nd a market ready for taking risks quanti-
ed by the certication. This opportunity means that forest
SMOs and SMEs will gain access to this market.
Insurance market for forest plantaons in Chile
In the case of forest plantations, insurance companies
are classied as General Insurance Companies. In the
case of forest res coverage, even if it is a standard pro-
cedure from the point of view of the product (policy type),
its value chain relies heavily on reinsurance capacity and
on the analysis of the inherent risks to forest plantations.
Reinsurance capacity is made up of the involved reinsur-
ance companies that cover these risks and their nancial
capacity. Regarding risks, each reinsurer has dened ca-
pability coverage (coverage limits) for forest plantations in
Chile, and depending on the risk analysis, a structure of
coverage is set, i.e. rates, compensation limits, deductibles,
and special additional coverage. Risks become insured
capital, then placements and nally insurance policies.
To analyze the risks, the insurance market uses dis-
cretional procedures without standardized information;
among the considered elements are plantation location,
forest management and operational performance, histori-
cal re occurrence; as well as forest neighbors’ protection
programs, cooperation agreements and business volume.
The main stakeholders are insurance companies, in-
surance brokers and sales agents, reinsurers and adjusters
(SVS 2013a).
Of the total plantation area insured, amounting
1,234,190 ha, eucalyptus species represent 17%, radiata
pine 81%, and only 2% correspond to other species.
The three most impor t ant regions for insured plant ations
are Bio Bio (52%), Araucania (22%), and Maule (15%),
making a total of 89% (Table 3).
The SAFOR certication system is oriented to SMOs
and SMEs, including several actors, as shown in Figure 1.
An important nding was that big forest companies have
a correct administration of their re insurance policies
through catastrophic coverage, concentrating in coverage
for large losses, which is logical given the substantial ap-
preciation of their resource. Therefore, if premiums were
higher, the cost would be much higher given the high self-
protection costs.
It was conclu ded tha t the tot al insura ble plant atio ns, only
in the regions of O’Higgins, Maule, Bio Bio and Araucanía,
exceed 1,950,000 ha, equivalent to 82% of the insurable
plantations at a national level. Of these, over 1,170,000 ha
are already covered, with only 18.5% belonging to SMOs
and SMEs: In these four regions, 750,000 ha have not been
insured, and most of that area would belong to this stake-
holder segment.
The area without insurance, on which SAFOR is focus-
ing, is located in four regions (O’Higgins, Maule, Bio Bio
and Araucanía), because the insurance market has 95% of
forest re insurance policies in these regions and it is where
re hazards are higher. It has been estimated that in those
regions there are 15,700 forest owners, 400 being SMEs
and the remaining ones, SMOs (SVS 2011).
Regarding the estimated business volume, a conser-
vative scenario that considers incorporating 30% of this
potential area in 5 years would represent a real increase for
the insurance forest market of 20%, which is relevant for a
general insurance market that grows annually at a rate of
6 to 7%.
Loewe and others Creaon and Implementaon of a Cercaon System...
Table 3—Insured plantaon area by region and insurance company.
Region
Insurance Company IV XIII VI VII VIII IX XIV X XI XII Area (ha)
Magallanes 75 98 3,515 33,072 35,613 11,149 1,566 2,342 514 87,944
Mapfre 521 1,626 1,741 7,138 654 251 11,931
Penta 50 1,548 8,11 21,134 7,02 25,294 835 63,991
RSA (SMEs) 679 12,088 30,208 19,116 1,713 454 7 64,265
RSA (Big companies) 63,796 135,31 550,538 236,688 10,167 9,561 1,006,060
Insured plantaon area (ha) 75 669 71,164 190,321 644,631 274,627 38,991 13,192 514 7 1,234,191
Region weight (%) 0 0 6 15 52 22 3 1 0 0 100
Source: Elaborated with informaon from the Superintendence of Securies and Insurance (SVS)

146 USDA Forest Service Proceedings RMRS-P-73. 2015.
Socioeconomic Analysis and Willingness
to Pay for SAFOR Cercaon
Forest res are currently a very important concern given
the occurrence of large disasters in the country in recent
years, which have affected forest plantations and surround-
ing communities, causing irreversible damage to the heritage
of many Chilean families, and also to companies that sup-
port many families. In this scenario, given the externalities
that directly affect people’s welfare, a study was conducted
to assess the willingness to pay for SAFOR certication by
potential users (SMOs and SMEs). The study was performed
by García (2014) in two stages: rst, surveys were conducted
among SMOs and SMEs to gather data on their socioeconom-
ic characteristics, perception of forest re risks, reasons for
not buying re insurance policies, as well as to explore their
interest in SAFOR certication. Second, the survey results
were analyzed using the contingent valuation methodology to
assess the willingness to pay for the new certication.
This methodology evaluates the possible willingness to
pay for SAFOR certication as a private concept that de-
pends on the income of potential users, considering that it
would remove entry bar riers through an objective risk assess-
ment, reducing existing asymmetries in the current market.
The model simulates through surveys and hypothetical
scenarios a market for an asset or group of assets for which
no market exists. This method is used to estimate changes
in welfare of people, especially when these changes involve
goods or services that do not have a price. One case applied to
prescribed res in the wild land urban interface in Colorado
(USA) was reported by Kaval and others (2007).
The methodology recommended by the National Oceanic
and Atmospher ic Administration (NOAA) (Arrow and others
1993) was used to design and implement the surveys and the
sample size (72 surveys) was calculated for 90% condence.
The analysis allowed us to characterize socially and eco-
nomically the objective SAFOR stakeholders, who were
found to be of low educational and low-income levels.
Surveys were concentrated in the municipalities of Nueva
Imperial (18%), Hualañe (16%), Santa Barbara and Cauquenes
(14% each) and Arauco (13%).
Regarding social characteristics, the educational level
shows that 50% of forest owners have basic education level
(years 1 to 8) and only 10% received university education;
5% have no formal education at all. Of all the respondents,
66% are married and 3% divorced. On average they have 2
children, and 23% have 3 children.
As per the income level, 66% earned less than US$ 330/
month, 11% between US$ 331-640; 9% between US$ 641-
1,090; 6% from US$ 1,091-3,630 and only 5% earned more
than that amount. It should be considered that when an-
swering income-related questions, people tend to report
smaller values, leading to bias in the responses.
Concerning forest characteristics, respondents own on
average 261 ha of plantations, and are concentrated in the
Maule and Bio Bio regions, with the former locality being the
one with the highest forest plantation area.
Of all of the owners, 65% indicated not having been af-
fected by forest res in the last 5 years, whereas 35% gave a
positive answer, reporting an affected area ranging between
1 and 480 ha. The re origin was diverse, including arson,
controlled burns by neighbors, neglect, illegal burning, elec-
tric lines, etc. Roughly 20% of the total area surveyed was
affected by a re.
Regarding damage caused by re, 88% of the respondents
declared to be aware of it and, only 9% were not, showing that
although they know about re damage, they do not use this
type of insurance.
The analysis of re insurance showed that only 7% of re-
sp ondents have an insu ranc e policy, indicating the existence of
buying barriers; among the mentioned reasons for not hiring
are high costs (35%) and ignorance (22%); only 3% consider
that the re risk probability in their property is low.
Of all the respondents, 85% saw value in buying a policy
and only 15% did not. Potential customers were willing to pay
for a re insurance policy US$ 5.5-12.7/ha/year (74%), 12.7-
25.5/ha/year (4%), 25.5-42/ha/year (4%); only 3% would pay
a higher amount.
Interestingly, 74% did not request a quotation from an in-
surance company, despite the importance given to buying a
policy. Therefore, it is inferred that they ignore the require-
ments for accessing an insurance policy; in fact, 78% of the
respondents declared not being aware about the requirements
for buying a re insurance policy.
Regarding the SAFOR certication, 59% of the respond-
ents would get it, whereas 38% are not interested and 3% did
not answer.
The willingness to pay for the certication declared in 50%
of cases ranged between US$ 1.3-3.3/ha/year; 15% would pay
US$ 3.3-4.2/ha/year; 23% would pay US$ 4.2-6.5/ ha/year; 8%
would pay US$ 6.5-9.8; and 4% over US$ 9.8.
The results of the contingent valuation analysis, using the
Probit model (Pucutay 2002) (Table 4), to determine the will-
ingness to pay for the certication, corrected for the relevant
variables and salary, indicated an average willingness to pay
of US$ 5.1/ha/year. Considering that the average price for re
insurance is currently US$ 36/ha/year, the certication would
correspond to 14%.
The signicant variables that affect the decision to buy the
certication correspond to region of residence (especially Bio
Bio and Maule), marital status (especially married), and the
interest in buying insurance policies. The willingness to pay
is strongly positively correlated to the total plantation area.
Although some of the respondents did not have the neces-
sary education to use insurance policies, they showed interest
in improving their conditions and having access to both cer-
tication and re insurance policies.
Technical aspects related to forest management
Forest management planning provides a framework to es-
tablish priorities, set objectives and devise strategies to deal
with risks (Day and Pérez 2013). Several aspects related to
forest management and res were addressed by Pinilla (2014).
Every year the country is affected by forest res that con-
sume large areas of native forests and plantations; therefore,
once a re starts, much of the efforts are focused on losing
the smallest area possible. In the case of forest plantations,
the signicant area planted with radiata pine and several
Loewe and others Creaon and Implementaon of a Cercaon System...

USDA Forest Service Proceedings RMRS-P-73. 2015. 147
eucalyptus species create a scenario where the probability of
re occurrence is affected by a number of factors. Whenever
a forest re occurs, a serious problem arises that affects for-
est resources, infrastructure, services, ecosystems, as well as,
and most importantly, loss of lives. In addition, forest res
cause the release of greenhouse gases, potentially increasing
the negative effects of climate change.
Due to the anthropogenic origi n of res, either unintent ion-
al, through carelessness or negligence, or directly intentional,
all prevention measures and early control are based on silvi-
cultural and community actions, and recently, on increasing
the penalties for the ones guilty of causing forest res.
Therefore, efforts to mitigate damages and losses from
forest res in plantations include community awareness cam-
paigns, education, preventive actions, availability of human
and technical resources, and joint collaboration between in-
stitutions, forest companies and owners (Urzúa and Cáceres
2011).
This scenario has led to consider that, in general, higher
forest re risk areas are associated with roads and nearby
population centers, recreational zones and areas where agri-
cultural and forest activities are developed. This is associated
with the chance of re, as determined by the presence and
activity of agents of ignition, known as re hazard (CIFFC
2013).
The amount of fuels associated with forest management
is very important. The occurrence likelihood and factors in-
volved in the generation of a wildre largely determine the
re insurance price, the investments made in forest planta-
tions protection and the expected returns. Insurers then need
elements to decide premiums for selling re insurances.
This demonstrates the need for information and analysis of
the main factors that are necessary to describe objectively the
risk of re in a forest plantation to be protected.
There are multiple forest and environmental factors that
can be used to explain re occurrence and spread; however,
many of these are difcult to quantify, estimate or model,
especially those related to environment, climate, topography
and vegetation, as well as the relation ships among them, which
limit their usefulness as elements when insurance companies
model or estimate these risks.
In this scenario, there are some forest or environmental
factors that are more objective when determining whether to
consider them as wildre risk descriptors.
Among the most frequently mentioned silviculture factors
at a national and international level are forest composition
(species, age, distribution, land use planning), presence of
agricultural activities, existence of forest operation planning
(establishment, management, harvesting), early prevention
and attack plans, fuel management, existence of perimeter
rebreaks as well as the implementation of Best Management
Practices (BMP) (INFOR 2006).
Other factors to be considered as part of the setting are
nearby towns, topography and climate, as well as land man-
agement elements and protection used by associations, forest
companies and local government.
It is concluded that the proper fuel biomass management
through silviculture reduces wildre probability when focus-
ing on the denition and assessment of risk areas, minimizing
damage at the local, personal, environmental and economic
levels. Through forestry practices it is possible to reduce the
number of res and their magnitude, preventing erosion and
enhancing soil protection, water movement, wildlife, etc., and
if the re has already started, to limit the associated damage
(Peña and others 2004).
In this context, Preventive Forestry was dened with the
aim of reducing the threat of and vulnerability to forest res,
seeking to reduce the number of res, and more importantly,
their size. It integrates all protective measures to achieve this
objective by implementing a plantation management oriented
to minimize this type of damage.
Accordingly, in 2006, the USDA noted that the decrease
in stand density, understory and surface fuel loads are factors
that explain historical re rates. Hence, the efforts to prevent
damage and losses in plantations aim at including measures
related to rebreak use, decreasing horizontal and vertical
fuel continuity, prevention measures in harvest or thinning
areas, fuel management, proper zoning and management of
plantations, among others, that contribute to the reduction of
wildre likelihood, minimizing losses and supporting actions
for a rapid extinction if it occurs.
WWF (2013) recommends identifying critical points and
most vulnerable forests, and designing specic re prevention
plans. These plans should aim, among other objectives, to:
• Rene urban uses in the forest, reducing the urban-wild-
land interface.
• Require that local actors located in forest areas develop
and implement prevention and defense plans against res,
including selective vegetation control and access control
protocols.
Regarding investigations to predict re behavior, location
and intensity in real time and the possible effects and con-
sequences on forest resources, infrastructure and security,
signicant efforts have been made, concluding that more in-
formation is required to reduce model uncertainty, and that
Loewe and others Creaon and Implementaon of a Cercaon System...
Table 4. Conngent valuaon results through Probit model applicaon
ξ Coe. Std. Err. z P>z 95% Conf. Interval
β
o -10.2336 6.3628 -1.6100 0.1080 -22.7045 2.2373
β
1 (Bio Bio region) 4.6831 2.7814 1.6800 0.0920 -0.7683 10.1345
β
2 (Maule region) 7.0493 2.7047 2.6100 0.0090 1.7482 12.3504
α
2 (marital status) -2.3498 1.2508 -1.8800 0.0600 -4.8013 0.1017
θ
3 (interest in buying re policy) 7.1935 3.4821 2.0700 0.0390 0.3688 14.0183
Source: Esmaons made from survey informaon

148 USDA Forest Service Proceedings RMRS-P-73. 2015.
modeling is complex, involving signicant uncertainty
(Taylor and others 2013).
Therefore, forest management factors should be included
in the development of the proposed insurability certication
system, which associated with a new Chilean standard will
lead to an objective forest re risk classication. This clas-
sication will be recognized by insurance companies, which
would take into account the risk level when determining re
insurance premiums.
In order to develop the standard, different factors have
been identied. Below are some of those factors grouped by
category, according to their characteristic of risk, hazard, or
re control which are analyzed as potential contributors to re
risk:
1. Forest Risk, including history of past res, intentionality
level, proximity to main roads, proximity to population and/
or recreational centers, electric lines and high risk adjacent
focusses.
2. Hazards, including those that will likely determine the re
development and behavior once already started, such as
vegetation fuel model (considering silviculture practices),
slope and weather parameters.
3. Fireghting organization and prevention measures, includ-
ing rebreaks, fuel treatments, owned available protection
resources, distance to specialized reghter teams, distance
to big enterprises plantations and agriculture activity in or
close to the property.
Based on these factors a re risk index will be built con-
sidering the national situation and other information such as
that reported by Schneider and others (2008) based on rela-
tive greenness; by Alexander and De Groot (1988) based on
weather information; by Koetz and others (2008) and Hawkes
and others (1995) based on fuel types and properties; by
Rodríguez and Silva and others (2012) based on economic
vulnerability; and by Holecy and Hanewinkel (2006) based
on a forest insurance model.
Conclusions
Based on several studies, the feasibility analysis concluded
that:
1. Forest plantations are a highly valuable resource for the
country, and are subjected to multiple risks, with res
among the most important. Over 50% of forest plantations
are not covered by re insurances, with potentially negative
implications, especially for SMOs and SMEs, in which the
proportion of non-insured plantations is higher, increasing
the existing gaps between the sectorial stakeholders.
2. The legal analysis did not identify administrative or legal re-
strictions that could affect the development of the proposed
certication system in Chile.
3. The industry of insurance market in Chile is transparent and
framed within a stable institution system, ruled by clear
laws and procedures. This is particularly positive for the
project because the increasing demand will nd a prepared
insurance market ready to deal with the certied risks. This
means allowing forest small owners and SMEs to gain
access to the market, which is also an attractive business
opportunity.
4. The insurance companies lack a standardized methodology
for evaluating the occurrence of forest res risk, which is a
market disadvantage.
5. The potential insurance business is signicant even in a con-
servative scenario, with an estimated increase of 20% for
the market, which is relevant in the general insurance mar-
ket that grows by 6-7% annually.
6. There is agreement both in national and international stan-
dards about the parameters that can be used as indicators
of forest re prevention and that therefore can be used for
generating the risk evaluation standard.
7. From the risk point of view, there are non-controllable fac-
tors, such as climate and vegetation, but also controllable
factors that can be used as technical parameters in risk
evaluation.
8. Several studies have identied important technical and com-
mercial gaps that affect the project beneciary sector; that
information turns out to be relevant for the future imple-
mentation and adoption of the SAFOR brand certication.
9. The socioeconomic analysis performed through a Contin-
gent Valuation analysis indicated, with a 90% condence
level, that although 88% of SMEs are aware about the re
risk in their plantations, an important proportion does not
buy insurance policies. However, 74% of them declare to
be interested in contracting insurances, and 59% is will-
ing to pay for the certication on average US$ 5.1/ha/year
when corrected by salary level. Variables that resulted in
signicant differences were: (1) legal status (married); (2)
region (higher in regions with greater forest plantation ar-
eas, especially Maule and Bio Bio); (3) interest for buying
re insurance policies.
10. Forest SMEs will need to build partnership relations to buy
re insurances in order to protect their heritage, to articulate
networks of state support and to develop agreements of de-
tection, reghting and suppression systems.
The performed feasibility analysis validated the initial as-
sumptions and at the same time recommended the national
standard development, currently under construction. Once
ofcially registered, the certication system will be created,
validated and implemented to transfer the standard into the
insurance market.
The analysis also indicated the existence of elements
that provide the opportunity of studying the establishment
of a government subsidy, both for re insurance and for the
SAFOR certication system.
References
Alexander, M.E.; De Groot, W.J. 1988. Fire behavior in jack pine
stands as related to the Canadian Forest Fire Weather Index
System. Canadian Forest Service, Northern Forestry Centre,
Edmonton, AB.
Loewe and others Creaon and Implementaon of a Cercaon System...

USDA Forest Service Proceedings RMRS-P-73. 2015. 149
Arrow, Keneth; Solow, Robert; Portney, Paul; Leamer, Edward;
Radner, Roy; Schuman, Howard. 1993. Report of the NOAA
Panel on Contingent Valuation, National Oceanic and Atmospheric
Administration (NOAA) [Online] http://www.darrp.noaa.gov/li-
brary/pdf/cvblue.pdf [December 12, 2013].
CONAF. 2012. Plantaciones forestales establecidas durante el año
2011. Informe Técnico 2012, Corporación Nacional Forestal,
Santiago, Chile. 65 p.
CONAF. 2013. Estadísticas de incendios. [Online] www.conaf.cl/es-
tadisticas. [October 27, 2013].
CIFFC. 2013. The Canadian Interagency Forest Fire Centre. [Online]
http://www.ciffc.ca/ [December, 2013].
Day, J. Kenneth; Pérez, David Manuel. 2013. Reducing uncertainty
and risk through forest management planning in British Columbia.
Forest Ecology and Management 300: 117-124.
FORESTSUR. 2014. Análisis del mercado asegurador en Chile y los
seguros para plantaciones forestales. In: Loewe, Verónica; Vargas,
Víctor (Ed.). 2014. Análisis de prefactibilidad de creación, desar-
rollo e implementación del sello SAFOR. 128 p.
García, Gladys. 2014. Evaluación de la disponibilidad a pagar de me-
dianos y pequeños propietarios de bosques y empresas PYME por
el sello SAFOR. In: Loewe, Verónica; Vargas, Víctor (Ed.). 2014.
Análisis de prefactibilidad de creación, desarrollo e implement-
ación del sello SAFOR. 41 p.
Hanemann, W.M.; Morey, E. 1991. Separability, Partial Demand
Systems, and Consumer’s Surplus Measures. Universidad of
California, Berkeley, USA. Journal of Environmental Economics
and Management 22: 241-258.
Hawkes, Brad; Goodenough, D.; Lawson, B.; Thompson, A.; Sahle,
W.; Niemann, O.; Fuglem, P.; Beck, J.; Bell. B.; Symington, P.
1995. Forest re fuel type mapping using GIS and remote sensing
in British Columbia. Proceedings of the Ninth Annual Symposium
on Geographic Information Systems, GIS95. Vancouver, BC. Pp.
647-656.
Holecy, Jan; Hanewinkel, Marc. 2006. A forest management risk in-
surance model and its application to coniferous stands in southern
Germany. Forest Policy and Economics 8: 161-174.
INE. 2014. Variables de estado del medio ambiente [Online] http://
www.ine.cl/canales/chile_estadistico [April 29, 2014]
INFOR. 2006. Especicaciones Técnicas de buenas prácticas para
plantaciones forestales. Ministerio de Agricultura. Comisión
Nacional de Buenas Prácticas Agrícolas. 47 p.
INFOR. 2013. Base datos programa de actualización de plantaciones
forestales de Instituto Forestal. INFOR.
Julio, Guillermo. 2012. Limitantes y desafíos en el desarrollo del
manejo del fuego en Chile. XXVIII Jornadas de Manejo del
Fuego, October 17-18, 2012, Concepción, Chile.
Kaval, Pamela; Loomis, John; Seidl, Andy. 2007. Willigness-to-pay
for prescribed re in the Colorado (USA) wildland urban inter-
face. Forest Policy and Economics. 9 (2007) 928–937.
Koetz, B.; Morsdorf, F.; Van der Linden, S.; Curt, T.; Allgoewer, B.
2008. Multi-source land cover classication for forest re man-
agement based on imaging spectrometry and LiDAR data. Forest
Ecology and Management 256: 263-271.
Lobo, Felipe. 2014. Factibilidad legal sobre la formación e imple-
mentación del sello SAFOR. In: Loewe, Verónica; Vargas, Víctor
Loewe and others Creaon and Implementaon of a Cercaon System...
(Ed.). 2014. Análisis de prefactibilidad de creación, desarrollo e
implementación del sello SAFOR. 15 p.
Loewe, Verónica; Alvarez, Andrea; Vargas, Víctor; Ruiz, Juan. 2014.
Recurso forestal y clientes potenciales del sello SAFOR. In:
Loewe, Verónica; Vargas, Víctor (Ed.). 2014. Análisis de pref-
actibilidad de creación, desarrollo e implementación del sello
SAFOR.46 p.
Loewe, Verónica; Vargas, Víctor (Ed.). 2014. Análisis de prefactibi-
lidad de creación, desarrollo e implementación del sello SAFOR.
39 p + an.
Peña, Eduardo; Pedernera, Patricio. 2004. Silvicultura preventiva
para combatir incendios forestales. Chile Forestal 302:12-14.
Pinilla, Juan Carlos. 2014. El manejo forestal y su efecto sobre los
incendios forestales. In: Loewe, Verónica; Vargas, Víctor (Ed.).
2014. Análisis de prefactibilidad de creación, desarrollo e imple-
mentación del sello SAFOR.57 p.
Piñeda, Leticia Esther. 2004. Probabilidad y estadística. 9th Ed.
Barcelona, Spain. 623 p.
Pucutay, Franck. 2002. Los modelos Logit y Probit en la investigación
social. Centro de investigación y desarrollo del Instituto Nacional
de Estadísticas e Informática (INEI). Lima, Perú.
Rodríguez y Silva, Francisco; Molina, Juan Ramón; González-Cabán,
Armando; Herrera, M. Miguel Angel. 2012. Economic vulnerabil-
ity of timber resources to forest res. Journal of Environmental
management 100: 16-21.
Schneider, P.; Roberts, D.A.; Kyriakidis, P.C. 2008. A VARI-based
relative greenness from MODIS data for computing the re poten-
tial index. Remote Sensing of Environment 112 (3): 1151-1167.
SVS. 2011. Seguros. [Online] http://www.svs.gob.cl/sitio/estadisti-
cas/seguros.php [December 12, 2013].
SVS. 2013a. Normativa. [Online] http://www.svs.gob.cl/sitio/legis-
lacion_normativa/legislacion_seguros.php?mercado= [December
12, 2013].
SVS, 2013b. Estructura de mercados. [Online] www.svs.cl/sitio/mer-
cados/seguros_estructura_mercados [December 12, 2013].
Taylor, Steve; Woolford, Douglas; Dean, C.; Martell, Dave. 2013.
Wildre Prediction to Inform Fire Management: Statistical
Science Challenges. Statistical Science 28 (4) 586–615. Institute
of Mathematical Statistics.
Tietenberg, Tom. 1996. Environmental and natural resources eco-
nomic. Fourth Edition. HarperCollins College Publishers. New
York. 614 p.
Urzúa, Nadia; Cáceres, Francisca. 2011. Incendios forestales: prin-
cipales consecuencias económicas y ambientales en Chile. RIAT
Revista Interamericana de Ambiente y Turismo 7 (1): 18-24.
USDA. 2006. Wildland Fire and Fuels Research and Development
Strategic Plan: Meeting the Needs of the Present, Anticipating
the Needs of the Future. Washington, DC: U.S. Department of
Agriculture, Forest Service, Research and Development. 50 p.
Vázquez, F.; Cerda, A.; Orrego, S. 2006. Valoración económica del
ambiente. Fundamentos económicos, econométricos y aplicacio-
nes. Argentina, 1ª ed. Thompson.
WWF. 2013. Bosques vulnerables a grandes incendios. Análisis de
WWF sobre el riesgo de los bosques a sufrir grandes incendios
forestales. Madrid, Spain. 47 p.
The content of this paper reects the views of the authors, who are responsible for the
facts and accuracy of the information presented herein.