77586 Experiential
Briefing
Note
 
 Rainfall
Index�Based
Insurance
for
Maize
Farmers
in
Thailand:
 Review
of
Pilot
Program
2006�2008
 
 January
2009
 
 Prepared
by
Ornsaran
Pomme
Manuamorn,
Operations
Analyst
 Agriculture
and
Rural
Development
Department,
the
World
Bank Context
 Agriculture
 remains
 a
 major
 economic
 activity
 in
 Thailand
 despite
 the
 growth
 of
 industrial
 and
 service
 sectors
 in
 the
 past
 decades.
 
 Agriculture
 accounts
 for
 10
 percent
 of
 Thailand’s
 GDP1
 but
 supports
 directly
 and
 indirectly
 the
 majority
 (68
 percent)
 of
 the
 Thai
 population
 who
 live
 in
 rural
 areas.
 
 Exposure
 to
 natural
 disasters
 causes
 significant
 financial
 losses
 to
 Thailand’s
 agricultural
 sector
 each
 year,
with
droughts
and
floods
being
the
most
recurrent
shocks.
At
the
micro�level,
 these
 extreme
 weather
 events
 negatively
 impact
 farmers’
 livelihoods,
 resulting
 in
 significant
income
losses,
an
inability
to
repay
loans,
and,
ultimately,
poverty.

At
the
 national
level,
the
government
absorbs
the
costs
of
these
natural
disasters
through
 various
spending
channels
including
public
relief
operations
funded
by
the
Ministry
 of
 Finance
 (MOF),
 and
 a
 natural
 disaster
 compensation
 scheme
 funded
 by
 the
 Ministry
of
Agriculture
and
Agricultural
Cooperatives
(MOAC).2
 
 Among
 developing
 countries,
 Thailand
 has
 a
 relatively
 more
 sophisticated,
 commercially
 oriented
 agricultural
 sector
 and
 a
 strong
 agricultural
 bank
 with
 extensive
 outreach.
 
 While
 the
 insurance
 industry
 is
 rather
 developed
 in
 both
 life
 and
 non�life
 business
 lines,
 it
 has
 limited
 rural
 penetration,
 and
 agricultural
 insurance
 remains
 absent.
 
 The
 most
 recent
 efforts
 in
 developing
 agricultural
 insurance
 were
 initiated
 by
 the
 Government
 with
 collaboration
 from
 private
 insurance
 companies
 during
 the
 late
 1970s
 and
 the
 early
 1980s.
 
 However,
 these
 efforts
 only
 led
 to
 a
 few
 experimental
 schemes
 that
 were
 short�lived
 and
 discontinued
due
to
financial
losses.
 
 Though
 the
 country
 had
 a
 very
 brief
 past
 experience
 with
 another
 form
 of
 index
 insurance
(area�yield),
the
current
lack
of
an
agricultural
insurance
system
gave
rise
 to
 interest
 in
 Thailand
 to
 test
 index�based
 weather
 risk
 management
 instruments.
 This
 interest
 was
 catalyzed
 among
 key
 Thai
 institutions
 when
 the
 Department
 of
 Insurance
(DOI),
now
the
Office
of
Insurance
Commissioners
(OIC),
and
the
Bank
for
 Agriculture
 and
 Agricultural
 Cooperatives
 (BAAC)
 were
 approached
 by
 some
 1
World
Development
Indicator
Database,
April
2007.
 2
For
more
information,
see
Vichit
Lorchirachoonkul
and
Weena
Chaisilaparungruang,
“Crop
 Insurance
Covering
Production
Cost�
(in
Thai),
National
Institute
of
Development
Administration
 (NIDA),
2002,
p.
9�
10
.

The
MOAC
website
also
provides
information.
 1
 foreign
market
players
about
the
possibility
of
developing
index�based
weather
risk
 management
products
for
the
Thai
market.
The
opportunity
to
implement
a
weather
 index
 insurance
 project
 in
 Thailand
 materialized
 in
 2005
 when
 the
 World
 Bank
 ��
 which
had
been
working
on
a
pilot
scheme
in
India
and
exploring
the
applicability
of
 the
index
approach
to
Southeast
Asia
��
came
in
contact
with
key
Thai
stakeholders
 including
 the
 DOI,
 The
 General
 Insurance
Association
(GIA),
 and
BAAC.

These
key
 stakeholders
 requested
 technical
 assistance
 from
 the
 World
 Bank
 and
 the
 project
 ultimately
resulted
in
a
pilot.
 
 In
 this
 context,
 the
 Commodity
 Risk
 Management
 Group
 (CRMG)
 of
 the
 World
 Bank’s
 Agricultural
 and
 Rural
 Development
 (ARD)
 department
 began
 to
 provide
 technical
 assistance
 to
 Thai
 institutions
 to
 develop
 an
 index�based
 weather
 insurance
 pilot.
 
 The
 primary
 focus
 of
 the
 CRMG’s
 technical
 assistance
 was
 on
 contract
design
for
risks
related
to
rainfall,
while
the
local
partners
led
and
financed
 other
 aspects
 of
 the
 pilot
 program
 development.
 
 The
 World
 Bank
 also
 provided
 advice
 on
 administrative
 procedures,
 pilot
 program
 monitoring,
 and
 international
 experiences.
 
 Implementation
Approach
 There
 were
 four
 facets
 of
 work
 that
 led
 to
 pilot
 implementation
 –
 operational,
 technical,
 financial
 and
 legal.
 
 Roles
 and
 responsibilities
 of
 institutions
 involved
 in
 the
pilot
are
defined
in
relation
to
the
four
aspects
as
follows:
 
 • Operational
�
 BAAC
 suggested
the
pilot
locations,
 led
the
field
work
and
data
 collection,
 and,
 as
 an
 agent/intermediary
 of
 the
 insurance
 companies,
 conducted
 farmer
 education
 sessions,
 marketing
 and
 sales
 of
 the
 index
 insurance
 contracts.
 BAAC
 was
 also
 responsible
 for
 other
 operational
 work
 including
 collecting
 and
 transferring
 premiums
 to
 insurers,
 distributing
 insurance
 certificates,
 creating
 databases
 of
 participating
 farmers,
 and
 distributing
daily
rainfall
information
to
insured
farmers
during
the
period
of
 insurance
coverage.
 
 • Technical
 �
 CRMG
 funded
 and
 provided
 guidance
 for
 designing
 a
 weather
 index
for
maize
in
the
pilot
areas
that
led
to
a
prototype
insurance
contract.

 The
local
insurance
team
later
refined
the
index
contract
based
on
feedback
 received
 from
 farmers,
 BAAC,
 and
 a
 local
 maize
 expert.
 The
 local
 insurance
 team
 also
 conducted
 premium
 pricing
 based
 on
 standard
 international
 methodology.
 
 • Financial
�
The
 GIA
 represents
ten
insurance
companies
that
participated
in
 the
project.

The
ten
companies
jointly
underwrote
all
the
drought
contracts.

 
 • Legal
 �
 The
 DOI
 (now
 OIC)
 advised
 on
 legal
 and
 regulatory
 issues
 and
 approved
the
index
contract
as
an
insurance
product
eligible
for
sales.
 
 2
 Apart
 from
 the
 above,
 other
 organizations
 are
 also
 part
 of
 the
 steering
 committee
 that
supported
the
project:
 • MOAC
provided
data
and
policy
advice.
 • The
Thai
Meteorological
Department
(TMD)
provided
data
and
technical
and
 operational
 advice,
 and
 supported
 the
 installation
 of
 an
 additional
 weather
 station
during
the
first
year
of
the
pilot.
 
 Pilot
Program

 The
 index�based
 insurance
 pilot
 project
 began
 in
 �the
 Pak
 Chong
 District
 of
 the
 Nakorn
Ratchasrima
Province
of
Thailand.

Maize
is
the
primary
crop
in
Pak
Chong,
 thus
 was
 chosen
 as
 the
 focus
 crop
 of
 the
 pilot,
 and
 drought
 is
 the
 principal
 risk.



 Nakorn
Ratchasrima
is
a
major
maize
production
site
in
Thailand,
and
a
large
part
of
 maize
 growing
 areas
 in
 this
 province
 lack
 irrigation,
 thereby
 leaving
 the
 crop
 exposed
 to
 drought.
 
 Maize
 is
 a
 major
 economic
 crop
 in
 Thailand
 and
 is
 grown
 by
 more
 than
 320,000
 farm
 households
 in
 Thailand
 (2004
 numbers)3.
 Maize
 ranks
 sixth
highest
in
loan
volume
for
BAAC’s
lending
portfolio.

BAAC
has
a
large
group
of
 maize
 growing
 customers
 in
 Nakorn
 Ratchasrima
 and
 suggested
 its
 Pak
 Chong
 Branch
 as
 the
 initial
 pilot
 site.
 
 In
 this
 district,
 there
 is
 high�quality
 historical
 weather
data
and
reliable
real�time
communication
��
a
prerequisite
for
developing
 index�based
weather
insurance
contracts
–
from
the
Pak
Chong
Agro�meteorological
 Station.
 
 The
 station,
 which
 belongs
 to
 the
 TMD,
 had
 approximately
 40
 years
 of
 rainfall
data
for
use
in
contract
design
and
premium
pricing.
 
 The
2006
Dry­Run
 
 Preparatory
technical
work
for
the
pilot
primarily
involved:
1)
collecting
of
rainfall,
 yield
 and
 other
 key
 agro�meteorological
 data;
 2)
 interviewing
 farmers
 about
 the
 importance
 of
 rainfall
 and
 losses
 they
 have
 suffered
 due
 to
 rainfall
 deficit
 risk;
 3)

 running
 crop
 model
 to
 derive
 quantitative
 relationships
 between
 maize
 yield
 and
 rainfall;
 and
 4)
 designing
 a
 prototype
 rainfall
 index
 insurance
 contact.
 Upon
 completion
of
this
technical
work,

a
dry�run
of
the

pilot
was
carried
out
from
June
 to
 August
 2006.
 
 The
 clientele
 consisted
 of
 BAAC
 clients
 (borrowers)
 who
 grew
 maize,
lived
within
the
20
km
radius
of
the
Pak
Chong
Agro�meteorological
Station,
 and
were
interested
in
the
insurance
product.


 
 The
 dry�run
 provided
 an
 opportunity
 for
 the
 GIA,
 BAAC,
 and
 TMD
 to
 practice
 product
 marketing
 and
 customer
 enrollment,
 and
 to
 set
 up
 a
 rainfall
 monitoring
 system
 for
 insurance
 payout
 purposes
 so
 that
 the
 system
 will
 be
 fully
 functional
 when
the
pilot
was
operational.

The
dry�run
also
provided
the
pilot
team
with
input
 from
 farmers
 to
 improve
 the
 prototype
 rainfall
 index
 developed
 for
 Pak
 Chong.

 Both
 BAAC
 and
 farmers
 in
 the
 pilot
 area
 had
 a
 good
 understanding
 of
 their
 risk
 environment.
 
 As
 a
 result,
 during
 the
 dry�run,
 they
 were
 able
 to
 apply
 careful
 scrutiny
to
the
usefulness
of
the
insurance
product,
and
the
role
which
index�based
 insurance
 can
 play
 to
 complement
 existing
 risk
 management
 measures.
 
 Feedback
 3
MOAC,
Thailand
 3
 from
farmers
and
BAAC
managers
showed
that
the
simple
cumulative
rainfall
index
 did
 not
 reflect
 the
 disproportional
 impact
 on
 yield
 of
 a
 prolonged
 water
 stress
 during
the
vegetative
phase
and
that
the
contract
would
need
to
be
altered
for
the
 full
 pilot.
 
 This
 resulted
 in
 changes
 that
 would
 provide
 coverage
 during
 the
 identified
“dry
spell�
period
of
the
2008
contract.
 
 The
2007­2008
Pilot
Implementation
 
 With
 the
 refined
 product,
 the
 partners
 implemented
 the
 first
 year
 of
 the
 full�scale
 pilot
 during
 the
 2007
 maize
 growing
 season.
 A
 relatively
 small
 total
 amount
 of
 US$42,400
was
insured
in
two
locations
in
the
Pak
Chong
district.

To
facilitate
the
 expansion
 of
the
 pilot
to
 cover
 two
 additional
 sites,
 the
 GIA
 agreed
 to
sponsor
the
 installation
of
a
new
automatic
weather
station.
Following
the
2007
season,
the
pilot
 partners
 decided
 to
 expand
 the
 project
 to
 other
 maize
 producing
 locations
 with
 significant
deficit
rainfall
risk.

In
the
2008
season,
the
project
grew
to
cover
eleven
 weather
 stations
 in
 five
 provinces
 including
 Nakorn
 Ratchasrima,
 Petchaboon,
 Lopburi,
Nakorn
Sawan
and
Saraburi.
The
insurance
companies
also
offered
farmers
 with
more
contract
variations,
including
choices
in
contract
start
date,
sum
insured,
 and
 premium.
 US$300,000
 was
 insured
 in
 the
 second
 year.
 
 
 In
 2008,
 a
 group
 of
 farmers
 around
 one
 weather
 station
 received
 a
 large
 payout
 due
 to
 rainfall
 deficit
 measured
during
the
first
phase
of
the
contract.
 
 Lessons
Learned
 While
 it
 is
 too
 early
 to
 fully
 evaluate
 the
 pilot
 experience
 in
 Thailand,
 an
 early
 review
of
the
pilot
offers
the
following
observations.
 
 Additional
work
on
contract
design
needs
to
be
carried
out
to
design
contracts
 appropriate
for
Asia.

Most
previous
work
on
contract
design
had
been
carried
out
 in
 Africa.
 
 While
 the
 methodology
 used
 in
 Africa
 works
 well
 for
 climates
 with
 pronounced
 seasonal
 variations,
 it
 needs
 to
 be
 better
 adapted
 to
 suit
 tropical
 climates
such
as
Southeast
Asia,
and
perhaps
wetter
areas
in
Africa.

New
work
on
 contract
design
could
be
used
in
other
areas
with
similar
climatic
patterns.



 
 Feedback
from
the
Thai
farmers
suggested
institutional
trust
plays
a
key
role
 in
product
take­up.

During
qualitative
interviews
after
the
pilot
seasons,
farmers
 stated
 motivations
 for
 purchasing
 insurance
 included
 risk
 management,
 experimentation,
 and
 word�of�mouth
 from
 peers,
 but
 most
 importantly,
 trust
 in
 BAAC
–
an
institution
with
long�term
relationships
with
the
farmers
–
which
played
 a
key
role
in
introducing
the
insurance
product.

This
lesson
is
consistent
with
the
 experience
of
BASIX,
a
microfinance
institution
that
first
introduced
rainfall
index� based
 insurance
 for
 farmers
 in
 India,
 and
 suggests
 that
 identifying
 partners
 from
 organizations
 that
 farmers
 trust
 could
 be
 similarly
 critical
 to
 successfully
 reaching
 target
clientele.


 
 The
growth
of
weather
index�based
insurance
in
Thailand
has
been
gradual
and
the
 number
of
farmers
insured
modest,
but
even
with
the
limited
experience
of
just
two
 4
 seasons,
 the
 project
 also
 has
 had
 a
 significant
 demonstration
 effect
 by
 generating
 wider
 interest
 from
 other
 institutions
 in
 the
 Southeast
 Asian
 region,
 as
 well
 as
 international
 reinsurers.
 More
 importantly,
 this
 project
 has
 laid
 an
 early
 institutional
 foundation
 for
 future
 market
 development
 through
 the
 investment
 that
 the
 local
 partners
 have
 made
 in
 building
 operational
 infrastructure,
 raising
 public
 awareness,
 educating
 farmers,
 and
 engaging
 in
 dialogue
with
the
government
on
broader
policy
support.


 
 Despite
 the
 operational
 infrastructure
 and
 local
 capacity
 in
 actuarial
 premium
 pricing,
 it
is
observed
that
 the
primary
constraint
in
the
current
size
of
the
pilot
 program,
as
well
as
independent
market
development
in
Thailand,
is
the
need
 for
an
agro­meteorological
expertise
to
work
with
insurers
in
designing
index­ based
 insurance
 contracts
 for
 new
 crops
 and
 new
 areas
 of
 the
 country.
 
 This
 suggests
 that
 more
 engagement
 in
 the
 contract
 design
 process
 from
 national
 institutions
 with
 relevant
 agro�meteorological
 expertise
 is
 needed,
 along
 with
 continued
 investment
 in
 weather
 stations
 and
 ongoing
 cooperation
 among
 the
 insurance
sector,
BAAC,
and
the
government.
 
 Next
steps
 Local
program
partners
are
planning
to
expand
the
insurance
coverage
to
all
maize
 growing
districts
in
the
existing
5
provinces
in
2009
(weather
stations
permitting).

 They
 are
 also
 discussing
 the
 possibility
 of
 new
 weather
 station
 installation.
 
 The
 World
 Bank
 will
 continue
 to
 provide
 limited
 technical
 advice
 and
 support
 to
 the
 Thai
 partners
 to
 ensure
 the
 smooth
 transfer
 of
 knowledge
 for
 locally
 designed
 contracts.


 
 

 
 
 5