8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
1/32
Contradictions and Constraints in Chile's Health Care and Education DecentralizationAuthor(s): Mary Rose Kubal
Source: Latin American Politics and Society, Vol. 48, No. 4 (Winter, 2006), pp. 105-135Published by: Distributed by Wileyon behalf of the Center for Latin American Studies at theUniversity of MiamiStable URL: http://www.jstor.org/stable/4490494.
Accessed: 07/10/2014 12:46
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp
.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
.
Wileyand Center for Latin American Studies at the University of Miamiare collaborating with JSTOR to
digitize, preserve and extend access toLatin American Politics and Society.
http://www.jstor.org
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=miamihttp://www.jstor.org/action/showPublisher?publisherCode=miamihttp://www.jstor.org/stable/4490494?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/4490494?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=miamihttp://www.jstor.org/action/showPublisher?publisherCode=miamihttp://www.jstor.org/action/showPublisher?publisherCode=black8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
2/32
Contradictions
nd Constraints
n
Chile'sHealth Care and
Education
Decentralization
Mary
Rose Kubal
ABSTRACT
Proponents
of
social
service
decentralization
ften claim
that trans-
ferring
ervice
administrationo lower
levels of
government
acili-
tates increases
citizen
participation
nd
governmental
ccountabil-
ity
while
improving
allocative
efficiency
and
equity
in service
distribution.Usingthe cases of healthand educationdecentraliza-
tion
in
Chile,
this
article
evaluateswhether
and under
what condi-
tions
social
service
decentralization
rograms
are
likely
to deliver
on
these
promises.
It discusses
the tensions
between
equity
and
efficiency
goals
and how
these
may
play
out
given
different
accountability
mechanisms n
local
public
choice,
principal-agent,
and
real-world
hybrid
ecentralization
models. The
case studies
illustrate
he
difficulty
of
balancing
he
need for central
standards
and
funding
with
local
autonomy,
but
suggest
that
accountability
mechanisms hat
emerged ollowing
Chile's
ransition o
democracy
in
1990
ed to
improvementsn bothequityandefficiency
n
decen-
tralized ervice
administration.
As
Latin
American
countries
struggle
to democratize
their
political
institutions while
pursuing
market-oriented
economic
strategies,
international financial
institutions have
encouraged
decentralization
as
an
administrative and
political strategy
for
rationalizing
the
two
processes.
If
properly
designed,
proponents
posit,
decentralization
measures
may
facilitate
increased citizen
participation
and
governmen-
tal accountability while improving allocative efficiency and equity in
service
distribution
(Agrawal
and
Ribot
1999, 473;
IDB
1994;
Litvack
et
al.
1998;
Peterson
1997).
In
addition,
international
lending
institutions
and
government
technocrats have
promoted
administrative
and fiscal
decentralization as a
means
of
providing
public
services
in
order
to
relieve
the
growing
financial burdens
on
central
treasuries
and
to
reduce
high
levels of
public
employment
(Rondinelli
1989,
77;
see
also
IDB
1994).
In
theory,
improved efficiency
and
central
budget
cuts make
free
market reforms economically sustainable; participation, accountability,
and
equity
make
market reform
politically
sustainable
by creating
stake-
holders,
as well as
deepening
democracy
or,
at
the
very
least,
moving
a
country
toward
good governance
(Graham 1998).
Using
the cases
of
105
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
3/32
106
LATIN
AMERICAN
OLITICS
AND
SOCIETY
48: 4
health
and
education
decentralization
n
Chile,
this
article
evaluates
whether
and under
what
conditionssocial
service decentralization
ro-
grams
are
likely
to
deliver
on
these
promises, particularly
n
terms
of
improvements
n
equity,
efficiency,
and
government
accountability.
A
number of
scholars
have
attempted
to outline
the
optimal
arrangements
or social
service
decentralization-especially
in the area
of
fiscal
federalismand local
public
choice
(Bennett
1990;
IDB
1994;
Lit-
vack et
al.
1998;
Peterson
1997;
Rondinelli
1989).
These
analyses
focus
on
administrative
nd fiscal
decentralization
nd
normallyemphasize
theoretical
gains
in
efficiency.
They
often
acknowledge,
but fail
to deal
adequately
with,
the
potential
trade-off
between
efficiency
and
equity
goals. Otherauthors have identifiedpoliticaland institutional actors
that
encourage
or
discourage
decentralization
Murillo
1999;
O'Neill
2003;
Williset al.
1999)
and influencethe
design
and
implementation
f
decentralization
policy
(Eaton
2001, 2004;
Garman
et
al.
2001).
These
studies
focus
on
the
political goals
of decentralizers
and
help
predict
when and to what
degree
decentralization
an be
expected.
They
paint
a more
complex picture
of the interactionbetween
decentralization
nd
the
political process
in
developing
countries,
but
they
often
focus on
decentralization
utcomes in
terms
of
the
balance
of
power
and
respon-
sibilitiesbetween levels of government,ratherthan policy outcomes,
such as
equity
and
efficiency.
This
leaves a numberof
questions.
What
happens
to
accountability
when
central
politicians
and
administrators
maintain control
over
dimensions of
decentralized
ervices,
such
as
funding,
oversight,
and
evaluation,
but local
officials
are
subject
to electoral
and exit
pressures?
Is it
possible
to
balance the need for
central tandards
nd
funding
with
local
autonomy
and
accountability?
What
are
the
implications
or
equity
and
efficiency
n
service
delivery?
Recently,a numberof quantitative,ross-nationaltudieshaveiden-
tified
factors
exogenous
to
decentralization
olicy,
such as
regime
type,
that
help
to
explain
equity
outcomes
in
the social
sectors. The
findings
suggest
that
political
democratization,
ven
in the context
of
market
reform,
has
a
positive
impact
on
aggregate
levels
of social
spending
(Brown
and Hunter
1999),
or
at least results
in increased
spending
(Huber
et al.
2004;
Kaufman
nd
Segura
2001)
and levels
of
service
pro-
vision
(Brown
1999;
Lakeand Baum
2001)
in the
health
and education
sectors. While
these studies do
not consider
decentralization
rrange-
ments as explanatory rcontextualvariables, heyraisequestionsabout
the
significance
of
such
arrangements.
Primarily,
ow
do factors
other
than
funding
levels-such
as
the
way
financing
is
allocated and
... how
delivery systems
are
organized
(Kaufman
and
Segura
2001,
581)-shape policy
outcomes?
The
case studies
of Chilean health
and
education
decentralization offer some tentative
answers.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
4/32
KUBAL:
CHILE'SDECENTRALIZATION 107
Given
the limits
of
a
one-country
study,
this articledoes not
set out
to
present
and test a
comprehensive
xplanation
or
decentralized
ocial
sector
outcomes.
Instead,
t
explores
the
implications
or
equity
and
effi-
ciency
outcomes that often are not addressedin the literatureon the
politics
of
decentralization.
n
doing
so,
it
contributes
o
the
literature
on
social
service
decentralizationand fiscal
federalism
by examining
two
cases of
hybrid
ecentralization,
omparing
ideal
ype
models
with
real-world
policies
and
exploring mplications
or
efficiency-equity
trade-offs
and
accountability
mechanisms.
Furthermore,
t
helps
to
answer
questions
raised
by
cross-national,
uantitative
tudies
by
iden-
tifying
accountability
mechanismsthat
affected
the distribution
of
the
additional unds allocated to the health and education sectorsin Chile
following
the
democratic
ransition.
Decentralizationn
Chile
was
initiated
n
1980
under
military
ule,
and it
continuedafter he
transition o
democracy
n
1990.
The variation
in
regime
type
allows
for an
exploration
of
conditions associated
with
democratization,
mainly
increased electoral
accountability,
that
are
hypothesized
to
affect social
policy
outcomes.
In
other
Latin
American
countries,
ocial
servicedecentralization
ften
coincided
with democratic
transitions,
or,
as in
Colombia and
Venezuela,
there
was
no
regime
change. Health and educationdecentralizationn Chilebegan almosta
decade
before
similar reforms
n
the
region,
and
the Chilean
reforms
have
been
held
up
as
a
model
(Delannoy
2000;
DB
1994;
Wiesner
2003).
The
first
consideration
s
the
tensions between
equity
and
efficiency
goals
and
how
they may play
out,
given
different
accountability
mecha-
nisms in
local
public
choice,
principal-agent,
nd
hybrid
ecentraliza-
tion
models. This
discussion
takes into account
contextual
variables,
such as
regime
type
and
government
deology,
which are
likely
to
shape
the
functioning
of
decentralized
ervicesand
thus,
policy
outcomes.
This
frameworks appliedto health and educationdecentralizationn Chile
with
a focus
on how
the initial
arrangements
hatwere established
under
the
Pinochet
dictatorship
were
(and
were
not)
altered
ollowing
the tran-
sition to
democracy
and
how
this
shaped
policy
outcomes.
Changing
lines of
accountability
n the
decentralized
ectors,
along
with
increased
funding
due
to
factors
exogenous
to the decentralization
rocess,
helped
improve
service
equity
and
quality,
and in some
instances,
efficiency.
THREEDECENTRALIZATIONODELS
Before
turning
to
the Chilean
cases,
it is useful
to
establish
how differ-
ent models
of
social service
decentralization
might
be
expected
to
improve
the
efficiency
and
equity
of
service
delivery
along
with increas-
ing
government
accountability.
These
particular
social
policy
outcomes
are
highlighted
for three related
reasons.
Proponents
of social
service
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
5/32
108 LATIN
AMERICAN OLITICSAND
SOCIETY
48:
4
decentralization ften
focus their
claims
on
these
areas
(as
do theircrit-
ics).
These
outcomes affect
people's
everyday
lives
in terms of
being
able to
access crucial
services such
as
health
care and education
and
having
avoice in matters hat
directly
affecttheir
quality
of life. At
times,
the
mechanisms
chosen to
achieve these
goals may
create
contradic-
tions within
the
decentralization
rocess
and constrain
he
achievement
of the
desired outcomes.
Decentralization
efers to a
variety
of
administrative, iscal,
and
political
changes
that involve
the transfer
of some combination
of
resources,
responsibilities,
r
decisionmaking
uthority
rom
the
central
government
o
lower-level
units,
be
they
subnational
overnments,
units
of the centralgovernment,orstate-ownedenterprises.Thismayinvolve
deconcentration
f
functions o
subnational
units of the
centralbureau-
cracy,
delegation
of
responsibility
to lower-level
actors
upwardly
accountable
to the central
government,
or devolution
of decisionmak-
ing,
financial,
and
managementauthority
o
lower levels
of
government
(which
may
or
may
not
be
downwardly
ccountable
o local
residents)
(Agrawal
and
Ribot
1999, 475;
Rondinelli
1989,
85).1
Decentralization an
furtherbe
categorized
n terms
of the substan-
tive area
being
reformed.
For
example,
Willis
et al.
(1999,
8)
define
political decentralizationas the establishment r reestablishmentof
elected autonomous
subnational
governmentscapable
of
making
bind-
ing
decisions at least in
some
policy
areas.
They
differentiate
his from
what
they
term
functional
ecentralization
r the transfer
f
policy
responsibilities
nd
expenditure
and
revenue-raising
owers
Willis
et
al.
1999,
8).
Social
service
decentralization
necessarily
involves
func-
tional
decentralization
ut,
as
in
the
Chilean
cases,
may
be achieved
in
the
absence of
political
decentralization
hrough
the
delegation
of
responsibilities
o
upwardly
accountable ocal
administrative
nits.
The two most common decentralizationmodels advocatedby the
internationalinancial
nstitutions,
ocal
public
choice
(LPC)
and
princi-
pal-agent
(P-A),
roughly
correspond
to devolution
and
delegation,
respectively.
In the LPC
model,
services are
locally
funded
and
locally
provided. Accountability
s
ensured
by
consumer
choice:
citizen-con-
sumers
vote with their feet
by
moving
to localities
that
offer their
pre-
ferred tax
service
package
(Tiebout 1956).2
The threats
of
exit
resi-
dents
moving
out
of
the
locality)
and voter
backlash
orce
local officials
to
formulate
policies
that
respond
to
constituent-consumer
emands.
Individualsrecognize that they getwhat they pay for (and pay for
what
they
get),
thus
preventingoverconsumption
nd
encouraging
effi-
cient resource allocation.
According
to this
logic,
continued central con-
trol over
resources undermines
efficiency
because fiscal
correspon-
dence,
the
connection between service
provision
and the
origin
of
the
funds
to
pay
for those
services,
is broken.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
6/32
KUBAL:CHILE'SDECENTRALIZATION
109
The
public
choice
model,
with
its focus
on
allocative
efficiency,
assumes
that service
demand
across
local
jurisdictions
aries
mainlyby
tastesor
preferences
rather
than income
(Prud'homme
1995,
208).
Achieving
allocative
efficiency,
however,
is not a
primary
concern,
at
least for
the
majority
without
the resources o
exit,
n
countrieswhere
the tax
base
in
many
localities is
not sufficient
to
cover even minimal
service
provision.
Without
centralredistributive
mechanisms,
nequality
between
jurisdictions
ncreases
due
to
the variation
n local tax and
serv-
ice levels
that
theoretically
acilitates
efficiency
(Wolman
1990,
36).
A
trade-off
emerges:
central
ntervention o
improve
equity upsets public
choice
incentives,
and
efficiency
s
sacrificed
L6pezMurphy
t al.
1995,
14). Some analysts posit thatorganizedparticipation f poor residents
at
the local
level and the
targeting
of limited
resources
toward
those
individuals
and
communities
most
in
need can
increase
equity
in decen-
tralized
service
delivery
without
sacrificing
efficiency
(Graham
1998;
IDB
1994,
194-95;
Litvack t
al.
1998,
8).
In
addition
o
equity
concerns,
centralofficials
may
limit ocal fiscal
autonomy
because
they
believe that macroeconomic
stabilization
requires
central
control over
tax and
spending
levels.
They
cite
exam-
ples
such
as
the
contribution
of the
provincial
debt to the 2002
Argen-
tine fiscal meltdown. Accordingto analysts from the Inter-American
Development
Bank and
the
World
Bank,
national
governments
can use
a
principal-agent
model
of
decentralizationo create
ncentive
structures
that
encourage
efficient
resourceadministration
t the
local
level while
maintaining
entralcontrol
over resourceallocation
and
technical
over-
sight
(IDB
1994;
Litvack t al.
1998).
Rather han
being
held accountable
by
local
residents,
local
officials are
upwardly
accountable
to central
authorities.
Thus,
citizen
participation
that
is,
demand for increased
service
levels)
threatensthe
efficient
operation
of a P-A model
when
localauthoritiesare accountable o the principal centralgovernment]
and not to
the local
taxpayer
or voter
IDB
1994,
181).
The risk is that
such
arrangements
will
decrease incentives for
efficient
local adminis-
tration
by
allowing
local
officialsto blame the center
for
poor
services
and
use
political
pressure
to extract additional
resources.
Upward
accountability
lso
decreases the effectiveness
of
poor
residents'
organ-
ized
participation
t the local
level-one
of
the
proposed
mechanisms
for
overcoming
he
equity-efficiency
rade-off.
In
Latin
America,
hybrid
ecentralization
rrangements ombining
elements of both LPCand P-Amodels are common. Nationalgovern-
ments
maintain
a
role
in
various dimensions
of
decentralized
ervice
provision,
including
funding,
policymaking,
oversight,
and
program
evaluation,
while
local residents are
encouraged
to
hold local officials
accountable
through
their
choice of
local
services
and their choices at
the ballot
box. The
literature on the
politics
of decentralization
policy,
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
7/32
110
LATIN
AMERICAN OLITICS
AND SOCIETY
48:
4
with its
focus
on
the
dynamics
of center-subnational
ower
relations,
suggests
several reasons
for
the
emergence
of such
hybrid
arrange-
ments.
Recent
research
has
emphasized
the incentives
created
by
lines
of
accountability
within
politicalparties
Garman
t al.
2001;
Williset al.
1999).
In
countries
where
national-level
politicians
control
nominations
and make
party policy,
such
as
Mexico,
it is
unlikely
that
power
and
resourceswill be
handed down to
the state
or
local
level
with no
strings
attached.
In
countries
where subnational
politicians
have a voice
in
party
decisions,
such as
Brazil,
a
higher degree
of
devolution
can
be
expected.
Additionally,parties
that
expect
to lose control
over national
political
institutions
but maintain
support
at the subnational
evel
(see
O'Neill2003)and individualnational-levelpoliticianswho expect to run
for
subnational
office
in the
future
(Samuels
2000)
may
reasonably
be
expected
to
devolve
significant
autonomy
and resources.
As
long
as these
politicians
or
partiesexpect
to hold on to
national-
level
positions,
central
officials
will
generally
oppose,
or exhibit
an
ambivalent
ttitude
oward,
he decentralization
f
political
power
and
resources
o lower levels of
government
Eaton
2001;
Williset al.
1999).
National
executives
and
legislators
are
therefore
likely
to formulate
hybrid
decentralization
rrangements
which
correspond
neither o
pure
LPCnor to P-Aforms).Sucharrangements llow themto take credit or
improvements
n
efficiency
and
governance
while
maintaining
s much
controlover
resourcesas
possible
and
avoiding
responsibility
or
prob-
lems
relatedto
poor
service
delivery.
This
clearly
poses
a
challenge
to
efficient
service
administration,
as lines
of
accountability
become
blurredand
important
lements of service
delivery
may
be
neglected
or
duplicatedby
central
and
subnational
bureaucrats
as
Eaton2001 docu-
ments in the
case
of
the
Philippines).
The
tendency
to
establish
hybrid
forms
of social
service decentral-
ization s reinforcedby the demandsof publicsectorunions,whichgen-
erally oppose
decentralization.Union
leaders
can
be
expected
to
push
for
decentralization
rrangements
hat
guarantee
heir
right
to
negotiate
salary
evels and
working
conditions
with
central
officials,
thus
limiting
local
control
over
decentralized
ersonnel.
Centralized
unions
with
large
memberships,
eachers'unions in
Chile
and
Mexico,
for
example,
are
likely
to
negotiate
such
arrangements.
Weak
or
fragmented
unions
and those
that do not
have the
political support
of
central
politicians,
such
as
teachers'
unions
in
Nicaragua
and
Argentina,
are less
likely
to
influencedecentralizationutcomes(Gershberg 999;Murillo1999).Yet
even
where
teachers'
unions
fail
to
maintain centralized
bargaining
arrangements,
it is common
for national education
ministries to
main-
tain
or even
increase
involvement
in
education
policy
development
and
oversight
following
the
decentralization of school
administration
(Gersh-
berg
1999;
Gorostiaga
2001,
562).
This is
largely
because,
like central
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
8/32
KUBAL:CHILE'SDECENTRALIZATION
111
politicians
and
public
sector
unions,
centralbureaucrats
ave a stake
in
decentralization
rrangements.
In
addition
to
partisanmaneuvering,
the
struggle
to maintain
or
expand
bureaucratic
prerogatives
affects lines
of
accountability
n the
decentralized
sectors. Central
ministries
commonly
have
conflicting
objectives;
for
example,
health and
education
ministry
officials
con-
cerned
with
improving
ervice
quality
and
preserving
heir own
bureau-
cratic
prerogatives
will
conflict
with finance
ministry
officials
preoccu-
pied
with
macroeconomic
tability
and
limitingspending.
Beneath this
layer
of
conflict,
subnational
governments
are
likely
to
have
difficulty
asserting
control
over
decentralizedhealth clinics and
schools,
espe-
cially fhealthandeducationministry fficialsmaintain noversightrole
(which
is
usually
the
case).
Service
providers
face
one
more
layer
of
bureaucracy
and
may
feel that
municipal
or
regional
officials
are less
familiar
with their
problems
than
central
ministry
functionaries.
Still,
local
health
providers
and teachers
may align
themselves
with
local
offi-
cials
to
demand
more
resources romthe central
government.
This
type
of
politickingupsets
efficiency
outcomes
in
both the local
public
choice
and
principal-agent
ecentralizationmodels. In
the
former,
entral
rans-
fers
would
threatento
underminefiscal
correspondence;
n the
latter,
efficiencytargetsestablishedby centralplannersneed to be protected
from
demands of
their
subnational
agents.
Yet
if
subnational
actorsdemand
transparency
nd
equity
from
cen-
tral
politicians
and
bureaucrats
instead
of,
or in addition
o,
pork
barrel
transfers),
hey
may
introduce
an
additional ource of
vertical
ccount-
ability,
nverting
he
upward
accountability mphasized
n
the
P-Amodel
and
distinct rom
the
downward
accountability
ey
to the
functioning
of
the
LPC
model. This
vertical
accountability
s similar o
the
horizontal
accountability
described
by
O'Donnell
(1996,
44)
as
the
controls
that
stateagenciesaresupposedto exercise over other stateagencies, but it
involves
subnational
actors'
calling
centralactors
to
account,
and these
subnational
actors
may
or
may
not
have
legal authority
o
sanction
their
central
counterparts.
Local officials
may
work within
their
parties
or
through
nonpartisan
municipal
and
regionalgovernment
associations
o
hold
central
officials
accountable or theirside of
the
bargain.
Of
course,
central
officials
interested
in
claiming
credit for
improvements
n
the
functioning
of
decentralized
services also have incentives
to
follow
through
and hold
subnationalofficials accountable
or efficient service
administration.Both central and subnational officials have a greater
incentive
to
guarantee
mprovements
n
quality
and
equity
of
decentral-
ized
services
when
they
are
subject
to electoral
pressures.
Thus
the
larger political
context,
particularly
regime type
and
gov-
ernment
ideology,
will
affect both lines of
accountability
in
and levels
of
funding
to
decentralized service sectors. The
findings
of the
quantitative,
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
9/32
112
LATINAMERICAN OLITICS
AND
SOCIETY
48:
4
cross-national tudies cited
above
suggest
that
regardless
of administra-
tive
arrangements,
emocratic
egimes
are
likely
to
allocate
more
funds
to health
and
education
than their
authoritarian
ounterparts.
Lake
and
Baum
(2001)
argue
that
democracy
contributes
o increasedlevels of
health and
education
provision
because
political competition
and
the
low
cost
of
participation
imit
politicians'
ability
to seek rents
at the
expense
of
public
service
delivery.
Brown
1999)
posits
that
elected
offi-
cials
perceive increasing
primary
chool
enrollment
as
an
effective
way
to
expand
heirelectoral
base,
hus
explaining
his
finding
that
among
low-income
countries,
enrollment s
significantly
igher
n democracies.
Kaufman nd
Segura
2001,
583)
speculate
that
electoral
considerations
mayexplain why democraticpoliticians n LatinAmerica eem less will-
ing
than their
authoritarian
ounterparts
o cut
health
and education
spending. Puttinggreater
emphasis
on
government
deology,
Huber et
al.
(2004,
6)
find that
democracy
correlates
with
increased
health
and
education
spending
levels in
LatinAmerica
because
it
allows for
the
emergence
and consolidationof
political
parties
representing
he
inter-
ests of
the
underprivileged.
Their
analysis
shows
that
spending
increases
with
left and
center-left
majorities
n the
legislature.
Whether
or not
greater
national
government
nvestment
n health
and education will improveequity outcomes in decentralizedservice
sectors
depends
partly
on
accountability
mechanisms
n these sectors.
The
following
case
studies
suggest
that
where central
officials
maintain
control over
resource
allocation,
he
emergence
of vertical
accountabil-
ity
mechanisms
n
the context of democratization
an create
incentives
for
equitable
distribution.
Still,
even the most
skillful
democratic
politi-
cians
and administrators
ave to
weigh
efficiency
and
equity
outcomes
as
they design
and
modify
decentralization
rrangements.
n
Chile,
the
transition o
democracy
did
not resolve
the internal
contradictions
of
decentralization rograms,but insteadchangedthe politicalcontext in
which
these were
played
out
at
the national and
local levels
and
in
administrative
nd
political
arenas.
HEALTH
AND
EDUCATION
DECENTRALIZATIONN
CHILE
Social service
decentralization in Chile
was initiated
in
1980,
under
authoritarianule,with the transferof the firstpublicschools to munic-
ipal
administration;
he
first
primary
health
care clinics
were
transferred
in
1981.
Although
he
rhetoric
urrounding
ealth
and education
decen-
tralization
emphasized
choice and market
mechanisms,
the
military
model more
closely
followed
principal-agent
lines,
with administrative
responsibilities
deconcentrated to
centrally
controlled
municipalities,
continued central
oversight,
and
extremely
limited fiscal
autonomy.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
10/32
KUBAL:
HILE'S
ECENTRALIZATION
113
Additionally,
consistent with
the
findings
of the
quantitative
studies
noted
earlier,
unding
for
health and education
was cut
by
the
right-
wing
military egime,
which
pursued
a marketreform
agenda,
thus
hurt-
ing
equity.
In
the
1990s
and
2000s,
elected
center-left
governments
ncreased
health and
education
funding
and
emphasized equity
and citizen
par-
ticipation
n service
delivery.
Despite
improvements
under
democracy,
some
of
the
old
contradictions
proved
difficult
to
resolve.
Tensions
between
municipal
and
centraladministrators
nd
politicians
ncreased,
and to
some
extent,
lines
of
accountability
became
blurred,
but the
return o
municipal
autonomy
and elections
generated
new accounta-
bilitymechanisms hatimprovedboth equityand efficiency.
DECENTRALIZATION
NDER
MILITARY uLE:
MUNICIPALIZATION
ND
ITS CONTRADICTIONS
In its
rhetoric,
the
military dictatorship
headed
by
General
Augusto
Pinochet
(1973-89)
was
committed
to
implementing
a
decentralization
program
based
on
public
choice
assumptions.
Pinochet
cast his
lot with
a
group
of
neoliberal
technocrats
known as
the
Chicago Boys.
While the
Brazilian
military
regime
was
investing
in
huge,
state-led
development
projects,
in
Chile there
was
massive
privatization
of
state-owned
enter-
prises
as reformers
attempted
to decrease the
level of state
intervention
in
the
economy.3
The
decentralization of
social service
administration
facilitated
this
state-shrinking
genda
by reducing
central
funding
for health
care and
education and
setting
the
stage
for
eventual
privatization.
Fiscal
devo-
lution was never
considered,
however,
and the
Chicago
Boys
main-
tained
central
authority
over
tax
policy
and
severely
constrained
local
autonomy
in
spending
decisions. With
municipal
officials
accountable
to
the
military
government
rather than
to local
populations,
and demo-
cratic
political participation
proscribed,
there
was
little
hope
for the
proper functioning
of a
local
public
choice
model.
In
reality,
the decen-
tralization mechanisms
implemented
more
closely
resembled
a
P-A
model,
with
continued
government
involvement
in
program
funding
and
oversight,
although
with
an element
of consumer
choice,
particu-
larly
in
education.
Decentralization rrangementslearlyreflected he military overn-
ment's
priority
of
decreasing
(or
eliminating
through
privatization)
spending
on social
services. The
1979
New
Municipal
Revenue
Law
ini-
tiated health care and
education decentralization
in
Chile.
It created
a
Common
Municipal
Fund
(Fondo
Comtin
Municipal,
FCM)
to assist
the
municipalities
in
creating, maintaining,
and
providing
services
to local
communities.4
The authors of this
legislation
intended
the increase
in
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
11/32
114
LATIN
AMERICAN OLITICSAND SOCIETY 48:
4
municipal
revenue to be
spent
on
newly
transferred
services,
such as
health care
and education. The
enabling
legislation
for the
municipal
revenue law
specified
that the
amount of
central funds
accompanying
the
transferred services
could
not
exceed
the amount
spent during
the
year
prior
to the transfer.
Any growth
in
spending
would
have to
come
from the
municipalities.
The
transfer
of
administrative
responsibilities
to the
municipalities
was seen as an intermediate
step
toward
privatization.
As
the
architect
of health care
decentralization
put
it,
To the extent that
it
was
possible
to
decentralize health care
administration,
it was easier to make it com-
petitive
with the
private
sector and this would facilitate
a
gradual
trans-
fer of state actions to private actions in the future (Hernfin Bichi,
quoted by
Duarte
1993,
298).
Education decentralization
actually
involved an
element of
privatization.
Parents
were
given
the
option
of
sending
their children to either
municipal
or state-subsidized
private
schools.
(Each
school received the
same
central
subsidy
per
student
enrolled).
More than a thousand new
subsidized
private
schools
were
established in
the first five
years
after the
reform,
with
2,643
operating
in
1985,
compared
to
1,627
in
1980
(Montt
and Serra
1994,
13).5
Policy-
makers
hoped
that
those
schools and clinics that could not be
privatized
would operate more efficiently as a result of new incentives for com-
petition
(subsidies
were allocated on the
basis
of students enrolled
and
number of
medical services
provided).
Funding
for the
transferred schools and clinics came
from two
main
sources: a
subsidy
from
the central
government
to
cover
operating
costs
(violating
the LPC
assumption
of fiscal
correspondence)
and
municipal
investment
in
infrastructure. In
reality,
the
municipal
contribution
went
to cover the
operational
deficit left
by
insufficient
central
governmental
funding
(Carciofi
et
al.
1996, 98, 107;
Friedmann
1995,
78).6
Local
sources of revenue, mainly driver permit and commercial license fees,
were
minimal,
and
significant
only
in
wealthy
urban areas with
sub-
stantial
commerce;
redistribution
through
the
new
FCM,
although
rep-
resenting
a
step
toward more
equitable
funding,
did not
provide
enough
resources
to
ensure local fiscal
autonomy.
Meanwhile,
the value
of cen-
tral transfers to
municipalized
clinics decreased
by
more than
30
percent
between
1982
and
1986
(Carciofi
et al.
1996,
127).
The trend for educa-
tion was
similar.
During
the
1982-83
recession,
the
government
dein-
dexed the
subsidy
levels
from the inflation
rate,
causing
a decline in the
real value of transfers for primaryeducation of 26 percent and a 30 per-
cent decrease for
secondary
education
from
1981 to
1985 (Carciofi
et al.
1996,
98).
The actual
transfer
of
schools
and clinics occurred
by way
of a
con-
tract
(convenio)
signed
by
the
mayor
(at
that
time
appointed
by
General
Pinochet)
and
the relevant
central
government ministry
or
public
entity.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
12/32
KUBAL:
CHILE'SDECENTRALIZATION
115
The
ministriescontinued
to
transmit
policy
goals
and
programs
o the
local
level and to
supervise
program
implementation.
Mayors
were
responsible
or
overseeing
day-to-dayoperations,
ncludingbudget
allo-
cationsand
personnel
decisions.
Essentially,
municipal
officialsreceived
central
mandates
concerning
the
programs
o be
implemented
in
the
schools and
clinics,
and were
responsible
or
hiring
and
overseeing
edu-
cation and health
personnel
to
carry
out
those
mandates,
all
on
a
budget
determined
by
the
value of
central
government
ransfers.
Mayors
ould,
in
theory,
raise doctors'
and
teachers'
salaries
to
compete
with
other
municipalities
and
private
providers.
In
reality,
to do
this,
they
would
have to
contributescarce
municipallygenerated
fees.
As a
municipal
functionaryobserved in the mid-1980s, thegovernmentalministries
(health,
education)
transfer
esources to
the local
level,
oversee
these
resourcesand enforce
the
completion
of technical
norms.There is
a lot
of
rigidity
....
This
discourages
ocal
initiative
quoted
by Raczynski
and Serrano
1987,
141).
Thus
potential
efficiency
gains
from
ocal
choice
were undercut
by
continued
central
mandates.
Because local
financing
was
necessary
to
make
up
the deficit
in
central
transfers,
municipalization
ed to
increased
disparities
n health
and education
spending. Despite
the
redistribution
f income
through
the FCM,poor municipalitieswere unable to contributethe same
amount of resources
in
absolute terms
(not
as a
percentage
of their
budget)
to
fund schools and
clinics.
Additionally,
lack
of
transparency
in
settingsubsidy
levels resulted n
funding
some
municipalities, ener-
ally
wealthier ocalitieswhere
supporters
of the
military
egime
ived,
at
higher
ratesthan
other,
often
poorer
areas
(Duarte
1995,
307-12).7
Official
figures
on
municipal
spending
were
not
kept
during
the
militaryregime.
A
survey
conducted
in
Santiago
in
1982,
however,
revealed
great disparities
n
education
spending
between
wealthy
and
poormunicipalities.Thedifferencewas apparent nteachersalary evels
and
expendituresper
student.
One
wealthy
municipality,
Providencia,
spent
1,657
pesos monthlyper
student,
while
Renca,
a
poor
municipal-
ity,
spent
813
(Parry
1997,
115).
In
1990,
the
first
year
for which
official
figures
are
available,
the 10
percent
of
municipalities
with the
lowest
revenue
per capita spent
an
average
of
2,660
pesos
per
student,
while
the 10
percent
with the
highest
revenue
per
capita
spent
42,080
(Parry
1997,
116).
In
the same
year,
the 20
percent
of
municipalities
with the
least
revenue
per capita
spent
an
average
of
87
pesos
per
month
on
health clinicusers,and those municipalitiesn the top quintilespent489
pesos
(for
both
groups
this
represented
4.2
percent
of
municipal pend-
ing)
(Duarte
1995,
312).
Despite adding
an
element of local
choice,
funding
mechanisms
created
perverse
incentives
that hurt
efficiency.
The
funding
mechanism
for
health
clinics,
which
were reimbursed
for
each service
provided,
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
13/32
116 LATIN
AMERICAN
OLITICS
AND SOCIETY
48:
4
resulted
n
serious
productive
and
potentially
allocative)
nefficiency.
t
encouraged
the
clinics to
provide
unnecessary,expensive
services and
discouragedlow-priced,
often
preventive
measures,
such as
inocula-
tions. Thisdecreasedthe
quality
of care for
patients,
who were rushed
through
appointments
and often submitted
o
unnecessaryprocedures.
The
quasi-voucher
unding
system
for education
presented
efficiency
problems
because
parents
lacked sufficient information
about
their
choices
(such
as
school
test
scores),
and the
private
subsidized
schools
could
deny
students
entry,
leaving
difficult
students
to the
municipal
schools.
Competition
did
not
necessarily
ncrease
the
quality
of educa-
tion,
because
private
schools focused
on
status
lements
like
fancy
names(preferablyn English)anduniforms.Bothprivateandmunicipal
schools
resorted
o
practices
uch as
grade
inflation o
maintain tudent
enrollment
Carciofi
t al.
1996;
Gauri
1998).
The
quality
of
services also
declined
as
poor
working
conditions
and
low
salaries resulted in
a
shortage
of
physicians,
and teachers
worked
double
shifts
with
virtually
no
preparation
ime
(Collins
and
Lear
1995).
Once
a
municipality
assumed
control of
a
school or
clinic,
staff
members
signed
new
employment
contracts
with
the
municipality
under
the
terms of the
private
sector labor
code
(Cddigo
de
Trabajo).
These workers ost theirstatus as centralgovernmentemployees, elim-
inating
their
job
security,
opportunities
for career
advancement,
and
other
benefits,
such
as
administrative nd
maternity
eave
(Carciofi
t
al.
1996,
96;
Frank
2002,
40;
Heyermann
1994,
52).
Primary
health
care
workers,
in
addition,
were
cut off from the
rest of
the
public
health
sector
(hospitals
and
specialized
clinics),
further
hurting
heir chances
for
career
advancement.
Primary
are
physicians
could
no
longer
hope
for
more
prestigious
appointments
o
hospitals,making
jobs
in munici-
pal
clinics
much less
attractive.
By requiring
health care
and education
workers o negotiatewithmorethanthreehundredmunicipalitiesather
than
collectively
with
the
relevant
central
ministry,
he
military
egime
achieved its
not-so-publicized
oal
of
decreasing
he influence
of
public
sector
unions,
in addition
to
reducing
social
spending.
The failuresof
health
care and education
decentralization
uring
ts
first
decade-administrative
rigidity,
nequity,
ack
of
transparency,
nd
poor
design
of
funding
mechanisms-have
been attributed
o the
con-
straints reated
by
authoritarian
olitical
nstitutions,
articularly
he
lack
of
citizen
participation
Carciofi
et
al.
1996;
Heyermann
1994;
Marcel
1994). It is interesting hat while electoralpressuresdid increase the
accountability
f both
local
and
national
politicians,
community
partici-
pation
in the
decentralized services did not
emerge
in an
organized
and
sustained fashion at the
local level after the
transition.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
14/32
KUBAL:
CHILE'SDECENTRALIZATION 117
DECENTRALIZATION
NDER DEMOCRACY:
RESOLVING
HE
CONTRADICTIONS?
Following
the
presidential
elections
of
1989,
the broad outlines
of the
military's
municipalization
program
remained
in
place.
President
Patri-
cio
Aylwin
(1990-94)
made it
clear that central
officials
had
no
interest
in
reassuming
responsibility
for
administering
health clinics and schools:
We
believe that
primary
health
care and
primary
and
secondary
edu-
cation
must
remain
at the
municipal
level. . .
.
The
serious
financial
problems
of
the
municipalized
health and education
systems
will
not
be
resolved
by
transferring
hese
services to the
[central]
tate
interview
in ElMercurio,November30, 1993,quoted by Heyermann1994, 47).
Equity
outcomes
improved
nevertheless
as the
posttransition overn-
ments'
deological
commitment o
growth
with
equity,
he
emergence
of
municipal
officialsas
watchdogs
ver central
policy
concerning
he
municipalized
ervices,
and
electoral
pressures
on both
centraland local
officials
facilitatedan
increase
in the level
of central ransfers
and their
more
equitable
distribution.
till,
democratization
id
not
resolve
all
the
contradictions
n
the
military's
ecentralization
model,
as
problems
with
administrative
igidities,
difficulties
creating
community
stakeholders,
and tensions within and across local and nationaladministrative nd
political
arenas
persisted,
constraining
progress
in terms
of
efficiency
and
equity.
Democratization
and
Changing
Lines of
Accountability
The first
three
posttransition
governments-all
from the center-left
Con-
certaci6n
coalition-wanted
to address the
social
debt left
by
the
Pinochet
regime,
but
without
returning
to the level
of
popular
mobi-
lization
before the
1973
military coup
or
endangering
the
military's
neoliberal
economic
project
(Garret6n
1994;
Weyland
1997).
The
post-
transition
preoccupation
with
achieving
growth
with
equity
led
to a
focus on
increasing
social
spending.
During
its first
year
in
office,
the
Aylwin
administration
egotiated
a tax increase
equivalent
to
2
percent
of
GNP,
which
was
spent
on
targeted
social
programs;
he health
care
and
education
budgets
rose
by
7
and
23
percent,respectively,
as
a
result
(Hershberg
1997,
352;
Weyland
1997).
Accordingto FinanceMinistry igures, centralgovernmentsocial
spending
increased,
on
average,
7
percent
annually
during
the decade
of
the
1990s,
for a
total
increase of
112
percent
(MIDEPLAN
d,
chap.
2,
sec.
1).
Central
redistribution
was made
possible by
a
period
of sus-
tained
growth
following
the
democratic transition.
During
the
1990s,
rising per
capita
GDP
was the
most
important determining
factor for
per
capita
social
spending
levels
in
Chile,
more
important
than the
level of
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
15/32
118
LATIN
AMERICAN OLITICSAND
SOCIETY
48:
4
public spending
relative o GDPor
the shareof
social
spending
in
terms
of
total
public
spending
(Mostajo
2000,
14-15).
Democratization
reated
incentives
for
the
posttransitiongovern-ments to
modify
the
decentralization
rrangements
nherited
from the
military
egime
in
ways
that
increased
the
central
government's
role in
terms of
funding,
oversight,
and
personnel
administration,
et
newly
autonomous
municipal
officials
were
able to
take
advantage
of
the
opportunities
fforded
by
the
democratic ontext
to
lobby
for increased
resources.
Increasedcentral
supervision,
guidelines,
and
funding
after
the
democratic ransition
elped
improve
he
quality
and
equity
of
serv-
ice
delivery,
but
local control
remained imited.This
point
is
important
not so much in explaining the relative weakness of subnationalgov-
ernments in
postauthoritarian
Chile,
but for
the
implications
of this
weakness as
it
affected he
functioning
of Chile's
hybrid
decentralization
arrangements
nder
democracy.8
The
visible
intervention
of
President
Ricardo
Lagos
(2000-2006)
in
health
and
education
policy
before the October
2000
municipal
elec-
tions
(the
resultsof
which would be
interpreted
s
a
referendum n
the
administration's
erformance)
highlights
he
political
ncentives or
cen-
tral
officials
o intervene
n
supposedly
decentralized
ervices.
In March
2000,the UnitedNationsreleasedthe resultsof a 1999survey revealing
that
almost
70
percent
of
Chileans
were dissatisfied
with the
country's
situation.
Respondents
anked
educationand
health care
as the
top
two
areas
where
changes
were
needed
(Santiago
Times
2000a).
Soon
after-
ward,
President
Lagos pledged
to
end
waiting
lines in all
of
the
coun-
try's
municipally
administered ealth
clinics
within
three
months
(San-
tiago
Times
2000c).
This
included
a
much-publicized
but less-than-
successful
attempt
by
the
Health
Ministry
o draw more
physicians
nto
municipal
clinics.
In
education,
the
president
promised
to double
the
per-studentsubsidy, reduce dropout rates, and improvethe perform-
ance of
disadvantaged
students
(Santiago
Times
2000b).
The central
government
responded
to
public
opinion
(not
organized popular
demand),
but not
by
increasing
ocal control
over clinics
and
schools;
just
the
opposite.
While the
desire
to
improve
the
quality
and
equity
of
public
serv-
ices
was
certainly
a factor n
these
decisions,
political
calculations-the
inclination of
central
politicians
to take credit
for
improvements
in
locally
administered
ervices-also
probably
played
a
part.
This
trend
correspondswith the earlierobservation hatcentralofficials nterested
in
credit
claiming
have incentives
to
hold
subnational
officialsaccount-
able for
effective service administration. Instead of
holding municipal
governments
accountable for
improved
services, however,
in Chile the
central
government
has
acted for
them,
partly
to
help
the electoral
chances of
the
governing
coalition's candidates
for
municipal
office.
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
16/32
KUBAL:
CHILE'SDECENTRALIZATION
119
Thus,
in
the context
of the
administrative ecentralization
f
important
services,
municipal
elections
became
a
mechanism
or
local
communi-
ties
(albeit
it somewhat
indirectly)
to hold
the
central
government
accountable for service
quality
when their own local
governments
lacked the
autonomy
and
resources
o act.
A
government
headed
by
the
right,
while
likely
to
respond
with different
policies,
would have
the
same
incentives
during
the
local election season.9
Newly
elected
municipal
officials
entered
their
first
term in
1992
at
a
disadvantage
elative
o
central
politicians
and administrators.
ecause
mayors
were
appointed
during
he
two decades of
military
ule
(and
the
firsttwo
years
afterthe
transition),
he
prestige
of
the
office was some-
whattainted.10 ome of the reforms hat ncreasedcentral nfluenceover
health
and education
administration
ere
passed
before
the
municipal-
ities
gained
political
autonomy.
Control
over
fiscal
resources,moreover,
did not
accompany
(and
thus
constrained)
the renewal
of
municipal
political
autonomy
(IDB
1994, 209;
Yafhez
and
Letelier
1995,
138).
Somewhat
paradoxically,
t
may
be
that the
very
weakness
of Chile's
municipalities
orced
municipal
officials
o
become
more effective
advo-
cates for
the
local
consumers
of decentralized
ervices,
introducing
an
element of
vertical
accountability,
which
certainly
did
not
exist
under
military ule.
Municipal
officials
and
their
nonpartisan
lobbying group,
the
Chilean
Municipal
Association
Asociaci6n
Chilena
de
Municipalidades,
AChM),
ormed
in
1993,
had
some
success in
changing
the transfer
mechanismsand
increasing
he
funding
or
the decentralized
linics
and
schools.
They
framed
this
issue
in terms
of
accountability.
One AChM
official
observed
that from
the
municipal
standpoint,
it s
irresponsible
on our
part
to
maintaina
system
where we cannot
guarantee
he
pop-
ulation a
situation
of
quality
in
terms
of the services
we are
offering
(Luna1997). This sentimentwas echoed by the association'ssecond
president
in
his
inaugural
peech:
Wedo
not
want
to continue
lying
and
sacrificing
our
neighbors.
They
have
the
right
to
count on
good
education
and health
care.
. .
This
is
why
we are
demanding
resources
[from
the
central
government].
It is because
we
feel an
obligation
towardthe
people
whom
we swore
to serve
El
Mercurio
1997a).
Because the
original decentralizing
legislation
obligated
central
authorities o
fund the
operating
budgets
of health
and
educationestab-
lishments,
municipal
efforts
to
secure
this
funding
represent
an
attempt
to hold the nationalexecutive accountable.Becausethey did not have
legal
authority
o
sanction
central
ministry
officials,moreover,
his was
accomplished mainly by
using
the
media to
get
municipal
issues
onto
the
public
agenda
(Belmar 2003).
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
17/32
120
LATIN
AMERICAN OLITICSAND SOCIETY
48:
4
Education
Reform
During
the
1990s
and
2000s,
central
education
reform
focused
on
improving
the
quality
and
equity
of
municipalized
education
through
expanded
ministry support,
both financial and technical.
Between
1990
and
1993,
spending
on
decentralized educational establishments
increased
by
nearly
25
percent.
Under
the
Aylwin
administration,
sev-
eral
support programs
were
implemented,
including
the
program
for the
Improvement
of
Education in Poor
Zones
(P-900),
initially
funded
partly
by
the
Swedish and
Danish
governments,
and
the
Program
to
Improve
Educational
Quality
and
Equity
(MECE).
The
MECE
program
received
a
$170
million
loan
from the World Bank and
$73
million
in
matching
funds
from
the Chilean
government
(Allard
1994,
37).
During
the
second half of
the
1990s,
the
government
of Eduardo
Frei
1994-2000)
dedicatedeven more
resources o
education
under the
rubric
of its
comprehensive
Educational
Reform
nitiative
La
Reforma
Educativa).
The
MECE nd
P-900
programs
were
expanded,
along
with
other
programs
aimed
at
pedagogical
improvement
at the school
level,
initiatives
to
improve
teacher
training
and
motivation,
and
a
gradual
transition o
a
full
schoolday
(as
opposed
to
dual
morning
and after-
noon shifts).11The Lagosadministration ommitted tself to continuing
the
reforms,
particularly
xpanding
the
P-900
program,
which
targets
underprivileged
hildren;
completing
the transition
o the full
school-
day;
and
increasingmandatory
ducation from 8 to
12
years (Santiago
Times
2001a,
2003,
2004).
The
administrative
tructure
of
the
ministryprograms
conveys
the
tensions
between
centraland
local bureaucrats.
The
implementation
f
initiatives
under the MECE
and the
P-900
programs
did not involve
municipal
education
administrators;
he
relationship
was between
the
municipalized chools and the centralministry through ts provincial
departments).
One Education
Ministry
fficial
remarked
hat
legally,
the
ministry's
responsibility
was to
support
the
educational
establish-
ments-it
was not
obliged
to
go through
the
municipalities,
although
there was
movement
in
this direction
(Herrera
1997).
In
reality,
the
arrangements
f
municipalization
fficially
eft all
technical-pedagogi-
cal decisions
to the
ministry.
Nevertheless,
many (particularly
arger,
urban)
municipalities
moved to create
their
own
technical
eams,
which
worked
alongside
the
provincial
ministry
upervisors.
Whenmunicipaleducationdepartmentswere involvedin technical
and
pedagogical
matters,
heir
decisions often were
subordinated
o cen-
tral
planning priorities.
For
example,
the
centrally
mandated
annual
municipal
education
plan
(PADEM)
was executed
by
the
municipal
edu-
cation
departments
at
the
direction of
the
ministry
and had to be
adapted
to the
technical-pedagogical
norms,
policies,
and
programs
of
This content downloaded from 158.109.185.182 on Tue, 7 Oct 2014 12:46:22 PMAll use subject toJSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/10/2019 Contradictions and Constraints in Chile's Health Care and Education Decentralization.pdf
18/32
KUBAL:
CHILE'SDECENTRALIZATION
121
the
Education
Ministry
Ministerio
de
Educaci6n
nd,
13).
A
researcher
who worked on an
Interior
Ministry
valuation
of
the
PADEM
bserved
that tensions between
ministry upervisors
and
municipal
officials
were
particularly
isible in
municipalities
where
mayors
or
municipal
educa-
tion
departments
ook a
strong
nterest
n
technical
ssues
(Zambra
997).
At
the school
level,
significant
progress
was made
toward
increas-
ing
the
autonomy
and
participation
f administrators
nd
staff,
which,
along
with
resources
accompanying
central
initiatives,
contributed
o
increasing
he
quality
of
education
n
targeted
chools. Some
central
ni-
tiatives,
such as the
educational
improvement
projects
(PME)
unded
through
the MECE
rogram,
gave
a
great
deal
of discretion
to
school
officials.12 he PMErepresented he firststep in what has been termed
pedagogical
decentralization
y
creating
a
space
for teachers
collec-
tively
to
evaluate their
schools'
needs
and to
design
projects
to
meet
those needs.
Examples
of
projects mplemented
n
the schools
include
folklore
workshops,
innovative
programs
o
improve
math
skills,
and
facilities or
art nstruction.
The teachers
received
technical
support
rom
the
ministry's
provincial
departments
n
formulating
heir
proposals,
which were then
submitted
for
the
ministry's
onsideration
n
annual
competitions.
Althoughthe PME mprovedteachers'participationn the schools
and contributed
much-needed materials
and
inancial
support,
there is
little
empirical
evidence to
support
the Education
Ministry's
laim
that
the
projects
have
significantly
ltered he
schools'
relationship
with the
larger
community
Courard
1998;
Kubal
2001;
Ministerio
de
Educaci6n
1996,
57).
The
PADEM
lso
succeeded
in
increasing
eacher
nvolvement
in
school
management
but
generally
failed
to include the
larger
com-
munity
in
meaningful
decisionmaking,
as
originally
intended
(Kubal
2001;
Ramirez
nd
Zufiiga
1997).
Recognizingthe frustrations f municipaleducation officials and
hoping
to
spur
local
innovation,
the
Frei administration
moved to
increasethe
municipal
role
in
education
policy
decisions;
for
example,
by giving
municipal
education
departments
he
option
to
modify
the
curriculum
under
the
supervision
of
provincial
education
administra-
tors).
This increased ocal
autonomy,
however,
increased
choices
only
in
those
jurisdictions
with
the resources
to
innovate.
The one
case of
large-scale
nnovation in
municipal
education
administration
n
Chile
illustrates his
point.
Under the administration
f
Mayor
Joaquin
Lavin
(1993-99), control over public schools in Chile'swealthiestmunicipal-
ity,
Las