Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium IVIember Libraries
http://www.archive.org/details/womenaspolicymakOOchat
DEWEY
.5
3l'
Massachusetts Institute of Technology
Department of Econonnics
Working Paper Series
WOMEN AS POLICY MAKERS:
EVIDENCE FROM AN
INDIA-WIDE RANDOMIZED POLICY EXPERIMENT
Raghabendra Chattopadhyay, Indian
Institute
of Management,
Calcutta
Esther Duflo,
MIT and NBER
Working Paper 01-35
October 2001
Room
E52-251
50 Memorial Drive
Cambridge, MA 02142
This paper can be downloaded without charge from the
Social Science Research Network Paper Collection at
http://papeis.ssrn.com/paper.taf7abstiact
id=XXXXXX
vv<
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
WOMEN AS POLICY MAKERS:
EVIDENCE FROM AN
INDIA-WIDE RANDOMIZED POLICY EXPERIMENT
Raghabendra Chattopadhyay, Indian
Institute
of Management,
Calcutta
Esther Duflo,
MIT and NBER
Working Paper 01-35
October 2001
Room
E52-251
50 Memorial Drive
Cambridge, MA 02142
paper can be downloaded witinout charge from the
Social Science Research Network Paper Collection at
Tills
http://papers.ssrn. com/paper. tafi'abstract
id=XXXXXX
rMAssACHusEns^rrutr
OF TECHNOLOGY
NOV
5 2001
LIBRARIES
Women
as Policy Makers:
Evidence from a India- Wide Randomized Policy Experiment
Raghabendra Chattopadhyay*and Esther Duflo^
October 2001*
Abstract
This paper uses political reservations for
women
in India to
leadership on policy decisions. In 1998, one third of
cils in
only
West Bengal were randomly
women
many
Villages Councils.
more
is
a
woman;
women
in rural areas.
in these councils
Using a data set we collected on 165
of public goods provided in reserved
the policy-making process
likely to participate in
woman. {JEL H4, H7,
and unreserved
directly relevant
and roads), while men invest more
(water, fuel,
118, J16,
Coun-
Village Councils are responsible for
We show that women invest more in infrastructure that is
to the needs of rural
village council
goods
we compare the type
Village Councils,
are
local public
leadership positions of Village
selected to be reserved for a
could be elected to the position of head.
the provision of
Women
all
study the impact of women's
if
in education.
the leader of their
Ol; Keywords Gender, Decentralization,
Affirmative action, Political Economy)
"Indian Institute of Management, Calcutta
tDepartment of Economics, MIT,
'We thank Daron Acemoglu,
Steve Levitt, Rohini Pande and
for
NBER
and
Abhijit Banerjee,
Emmanuel Saez
CEPR
Anne
Case, Mihir Ghosh Dastidar,
for discussions,
organizing and supervising the data collection, Lucia Breierova,
excellent research assistance,
Angus Deaton, Marie Lajus,
Prasid Chakrobarty and Mihir
Ghosh Dastidar
Shawn Cole and Jonathan Robinson
for
and the National Insitute of Health (through grant RO1HD39922-01) and the John
D. and Catherine MacArthur foundation for financial support. Chattopadhyay thanks the Institute for Economic
Development
at
Boston University
for its hospitality.
Introduction
1
women
Worldwide,
sented 13.8% of
nities,
are under-represented in
parliament members, up from
all
positions. In
all political
9%
men and women
Compared
in 1987.
education, and legal rights, political representation
is
June 2000, women repreto economic opportu-
the area in which the gap between
has narrowed the least between 1995 and 2000 (Norris and Inglehart 2000).
Increasing the political representation of
sition, increasing equity of efficiency:
women
is
equity because
often thought to be a "win- win" propo-
women and men have
different political
agendas, and increased representation would ensure a better representation of women's needs;
because
efficiency,
altruistic),
women
are supposed to be better politicians (less likely to be corrupt,
and because the
child health
political
agendas of
women
more
are thought to lead to investment in
and education, which have positive long term consequences on growth. Alleviating
deprivation and inequality in the political realm can thus have far reaching consequences, not
only for women, but for everyone (a point
as
Freedom (Sen 1999)). This
The World's
Women
made most
position, articulated for
Because women's representation
policies
to rapidly enhance
in assemblies or
(World-Bank
on
example
in the
in
Development
United Nations report.
and debates around the world.
advances slowly, and does not seem to
in the pohtical sphere
be greatly affected by economic development,
way
by Amartya Sen
(United-Nations (2000)), and in the World Bank's report. Engendering
Development (World-Bank (2001)), shapes
as a
forcefully
women's
political reservations for
ability to participate in policy
women
are often proposed
making. Quotas
for
women
parties' candidate lists are in force in the legislation of over 30 countries
(2001)),
and
European Union (Norris
in the internal rules of at least
one party in 12 countries of the
(2001)).
Reservation policies clearly have a strong impact on women's representation.-' This does not
necessarily imply, however, that they have a impact on policy decisions. In standard "Downsian"
political
economy models, where candidates can commit
reflect the preferences of
increases
the electorate.
to a specific policy, political decisions
In a Coasian world, even
women's bargaining power, pohcy
if
the reservation policy
decisions other than transfers to
women
should not
'See Jones (1998) for a study of the Argentinian case, and Norris (2001) for the impact of reservation in the
Labour Party
in
the
UK. Women's
representation
were eliminated during the transition from
fell
from 25% to
7%
Communism (World-Bank
in
Eastern Europe
(2001)).
when gender quotas
be affected. The
outcome should be achieved, and women would then be compensated
efficient
with direct transfers.
There
is
some evidence that the assumption behind these models
are violated. In the U.S.,
Levitt (1996) shows that politicians' decisions are strongly influenced by personal ideology.
Female
more energy
legislators devote
successful in passing legislation on
to women-specific issues than
women's
when they propose
issues
and Welch (1991)). In India, Pande (1999) argues that teservation
men
it
do, and are also
(Thomas
(1991),
more
Thomas
for minorities affects targeted
transfers.
However, despite the importance of this issue
about the causal
effect of
for the design of institutions, very little
women's representation on policy
based on cross-sectional comparison,
then
may
elects
The
them.
The
known
available evidence,
to interpret, because the fact that
is difficult
women
are
may
reflect the political preferences of the
correlation between policy
outcomes and women's participation
better represented in a particular county or locality
group that
decisions.
is
not reflect the causal effect of women's participation.'^
Furthermore, even
if
we knew more about the causal
knowledge would not necessarily extend to the
force greater participation of
through quotas
may
women
effect of
women's representation,
mechanisms
effects of quotas, or other
in the political process.
distort political competition
this
to en-
Ensuring women's representation
and have direct
effects.
For example,
it
may
lower the competency of the pool of eligible candidates, alter voter's preferences over pohtical
and increase the number of politicians that are new
parties,
in office.
This paper studies the policy consequences of mandated representation of
advantage of a unique experiment implemented recently
positions of chief (Pradhan) of the
GP) have been
in a reserved
reserved for
West Bengal
in India.
village councils
women by
Since 1998, one third of
for the position of
GP
GP
were selected randomly:
Pradhan
were ranked by
^For example. Dollar, Fisman and Gatti (1999) find a negative correlation between representation of
in
parliaments and corruption. Does this
are also
more
likely to elect
women
mean women
child support enforcement policies are
more
and Case (1997) show that worker compensation and
be introduced
parliament, after controlling for state and year fixed effects.
women
in
parliament
may be
a proxy for
women
are less corrupt, or that countries that are less corrupt
to parliament? Besley
likely to
all
(Gram Panchayat, henceforth
women: only women may be candidates
GP. Furthermore, the reserved
taking
women's involvement
in states
But they
where there are more women
in
explicitly recognize that the fraction of
in politics,
more
generally.
their serial
number, and every third
dramatic increase in the number of
and implement
was reserved
women
Pradhan makes
to provide them.
We
conducted a detailed survey of
(Birbhum,
and unreserved GP. Since
resulted in a
was enacted
five
development projects, and maintain local infrastructure, using state
local
GP
GP of one district
woman. This has
substantial power to the village councils to
funds. Thus, the
the
for a
elected as Pradhan."^ This policy
amendment that gave
years after a constitutional
define
GP
GP
in
decisions about which
all
pubhc goods to provide, and where
investments in local public goods in
West Bengal), and we compare investments made
were randomly selected to be reserved
for
women,
all
in reserved
differences in
investment decisions can be confidently attributed to the policy of mandated representation of
women.
The
results suggest that reservation does indeed affect policy.
political process in
GP that are reserved for women.
In those
Women
participate
GP, there are
more
in the
also significantly
more
investments in drinking water infrastructure, recycled fuel equipment, and road construction.
By
contrast, in unreserved
GP, there are
significantly
more investments
in education.
workers are monitored more closely in reserved GP, and teachers are monitored
These
results are interesting in their
did indeed modify' policy making.
legislation to investigate
whether
We
own
right,
can in addition exploit specific features of the reservation
its effects
be explained by the fact that
as less likely to be re-elected,
less closely.
because they show that the reservation policy
can be attributed to the gender of the Pradhan,
rather than to other consequences of reserving seats.
results can
Health
women
We
specifically investigate
whether the
may
be perceived
are inexperienced, that they
and that they tend to come from a more disadvantaged background
than men.
These
results thus indicate that a politician's gender does influence
generally, they provide
new evidence on
poHcy
decisions.
More
the political process. In particular, they provide strong
evidence, based on a randomized experiment, that the identity of a decision
policy decisions. This provides empirical support to political
maker does influence
economy models that seek
to enrich
the Downsian model (Alesina (1988), Osborne and Shvinski (1996), Besley and Coate (1997)).
The remainder
policy,
In
of this
and the data
set.
West Bengal, only
6%
paper proceeds as follows. Section 2 describes the
political context, the
how
political reservations
Section 3 reviews the literature and discusses
of
Pradhan were women
after the 1993 election.
for
women
could affect policy outcomes. Section 4 discusses these intervening mechanisms:
it
compares the characteristics of male and female Pradhan, and shows that the reservation policy
did affect
women's pohtical
participation.
Section 5 presents the central results of the paper,
the difference in public goods provisions in reserved and unreserved GP. Section 6 examines
whether this difference can be mostly attributed to the
effect of the policy
on the gender of the
Pradhan. Section 7 concludes.
2
2.1
The Policy and Design
The Panchayat System
The Panchayat
is
in
of the Study
West Bengal
a system of village level
district level (Zilla
(Gram Panchayat), block
level
(Panchayat Samiti), and
Parishad) councils, elected by the people, responsible for the administration
of local public goods.
The Panchayat system has
existed formally in most of the
major states of
India since the early 1950's. However, in most states, the system was not an effective
body
of
governance. Elections were not held and the Panchayats did not assume any active role (Ghatak
and Ghatak 1999). This changed
in
West Bengal
when
in 1977,
the Left Front Government
The most important
gained power on a platform of agrarian and political reform.
part of the
land reform was tenancy reform, which enabled tenants to register and restricted the eviction
of registered tenants (Banerjee, Gertler and
was to give
life
Ghatak Forthcoming). The major
The
to a three-tiered Panchayat electoral system.
1978 and elections have taken place at
five
first
year intervals ever since.
political
reform
election took place in
The
first
election took
place in the middle of a major political and social upheaval resulting from the land reform,
which arguably allowed the Panchayat to assume an important
traditional
landowning
class.
This
may
political role,
independent of the
account for the importance of the institution in West
Bengal, compared to most other states of India (Ghatak and Ghatak 1999).
Each Gram Panchayat (GP) encompasses 10
people.
The
GP
to 12 villages
do not have jurisdiction over urban
areas,
municipalities. Voters elect a council, which then elects
and a population of about 10,000
which are administered by separate
among
its
members a Pradhan
(chief)
and an Uppa-Pradhan (subchief). The council takes decision by majority voting (the Pradhan
does not have veto power). The Pradhan, however,
is
the only
member
of the council with a full-
time appointment. The major responsibilities of the
irrigation
excavate ponds (used for bathing, fishing and irrigation), and maintain and construct
The main
roads.
are to administer local infrastructure:
and construct drinking water and
repair public buildings (such as schools), maintain
facilities,
GP
source of financing for these projects are state grants (the 73d
stipulates that the ultimate objective
is
that
40%
of the each State's budget
is
amendment
to be spent by
the Panchayats). Until 1992,- district funds were distributed through "schemes" earmarked for
a specific purpose. In 1992, the 73rd amendment to the Constitution of India established the
framework of a three-tiered Panchayat System with regular elections throughout India.
GP the primary responsibility for implementing development programs,
the
the needs of the villages under
but the
money which was
broad schemes.
Money
is
its jurisdiction.
The main source
previously earmarked for specific uses
allocated under the Jawhar Rozgar
public infrastructure. Second, there
is
now
gave
as well as identifying
of financing
is
It
is still
the state,
allocated through four
Yojana (JRY) scheme
for
work
in
GP
a separate scheme for drinking water projects. Third,
have the responsibility to identify beneficiaries of social programs and disburse funds (widow's
pension, rehef, etc.), and receive funds for this purpose.
its
own operation
The
GP was
also given the right to collect
establishment of collective
GP
the
JRY
honorarium
(salary of secretary,
accounts for
fisheries.
30%
programs, 15%, the grant for
GP
its
receives
money
Pradhan, building maintenance,
for
etc.)
own, through taxation of assets, and the
income, the drinking water scheme, 5%, the welfare
functioning, 33%, and the
Following the 73rd amendment, the
First,
income on
GP
According to balance sheets which we could collect in 40
of total
GP
for the
Finally, the
GP
GP own
was given additional
revenue for 8%.
responsibilities in
West Bengal.
they were entrusted to establish and administer informal education centers (called SSK),
an alternative form of education
for children
who do
not attend school (a non-qualified instructor
teaches children three hours a day in a temporary building or outdoors).
adult literacy programs.
Most importantly,
since
May
They must
also oversee
1998, they are required to organize two
meetings per year, called "Gram Samsad". These are meetings of villagers and village heads
in
which
GP
all
voters
may
participate (one
Gram Samsad
council submits the proposed budget to the
in the
previous six months.
Gram Samsads
September 1998. While attendance
is
far
is
organized for each 700 voters).
Gram Samsad, and
The
reports on their activities
have been regularly held in most locations since
from universal (16% of
eligible participants),
a study
Gram Samsad
of 20
meetings by Ghatak and Ghatak (1999) shows that the meetings give
to active debates over the kinds of projects that should be undertaken
rise
and the effectiveness
of
implementation of past projects, as well as charges of corruption and mismanagement.
GP
The
is,
has no direct control over the appointment of formal teachers or health workers.
however, supposed to monitor their performance.
campaigns and immunizations.
and committees
frastructure,
management
for the
of
it is
helps organize health information
responsible for organizing
community-based
In 1992, the 73rd
fisheries, forests,
drinking water
in-
Women
amendment provided that one
third of the seats in
one third of the position of Pradhan must be reserved
well as
women's organizations
and schools.
Reservation for
2.2
Finally,
It also
It
all
Panchayat councils
as
women. Seats and Pradhan's
for
position were also reserved for the two disadvantaged minorities in India, "scheduled castes" (SC)
and "scheduled tribes" (ST), which mandated representation proportional to each minority's
population share in each
States were asked to modify their electoral rules to conform
amendment.
to this
In
district.
West Bengal, the Panchayat Constitution
third of the position of councilors in each
only
women
women
GP
to
rule
was modified
women:
in 1993,
in a third of the villages in
could be candidate for the position of councillor for the area.
elected to Panchayat councils increased to
36%
and reserved one
after the
each GP,
The proportion
of
1993 election. The experience
was considered a disappointment, however, because very few women (196 out of 3,324 GP)
advanced to the position of Pradhan, which
is
the only one that yields effective power (Kanango
(1998)).
To conform
to the 73d
amendment, the Panchayat Constitution Rules
of
West Bengal were
again modified in April 1998 (Government of West Bengal 1998). In addition of the reservation
at the village level within each
Pradhan:
in those
had to be reserved
office of
of
GP, the council members had
Likewise, a proportion of
where the
GP, one third
for
GP
GP
were selected to be reserved
to elect one of their female
equal to the share of the share of
SC and ST. A
Pradhan were
SC and ST
specific set of rules ensured a
to be reserved for a
woman.
members
All
GP
as
a female
Pradhan.
in the population
random
in
for
selection of
GP
a district were ranked
number (an administrative number
in consecutive order, according to their serial legislative
dating this reform). They were then ranked in two separate
the seats had been reserved for a
SC/ST
following a similar method). Using these
was reserved
Prom
for
a
woman
for the
according to whether or not
lists,
or not (these reservations were also chosen randomly,
lists,
every third
GP
starting with the
officials
GP
of
we
study,
we could
into those reserved for
GP
who implemented
verify that the policy
SC/ST and
woman by
reserved for a
1
the allocation of
was
indicates the
reserved for a
number
woman
of female
are female.
strictly
who
devised
More importantly,
in the
implemented. After sorting the
those not so reserved, we could reconstruct the entire Hst
sorting
GP
list
appears that these
in individual districts, it
all
GP
by their
before the law was passed), and selecting every third
verifies that
on the
1998 election.^
instructions were successfully implemented throughout the state.
district
first
discussions with the government official at the Panchayat Directorate
the system and district
pre-
to the reserved
Pradhan
GP
list
serial
number
(allocated several years
starting from the
first in
was indeed random,
each
This
list.
as intended.
Table
reserved and unreserved GP. All Pradhan in
is
Only 6.5%
of
them
are female in unreserved
GP
GP. The policy
thus had a large effect on the proportion of female Pradhan in the district.
2.3
Data Collection and Empirical Strategy
In the
summer
Birbhum
we conducted a survey
At the time of the 1991 census,
the main economic activity, and rice
rates were
The
50% and 37%,
district
is
known
There are 166
in
of all
GP in the district of Birbhum, West Bengal.
located in the western part of West Bengal, about 125 miles from the state capital,
is
Calcutta.
of 2000,
in
a-
had a population
of 2.56 million.
the main crop cultivated.
respectively, lower
to have
GP
is
it
The male and female
is
literacy
than the West Bengal average of 67% and 47%.
relatively well-functioning
Birbhum, of which
Agriculture
five
Panchayat system.
were reserved
for pre-testing, leaving 161
GP
our study. Table 2 shows means of the most relevant village variables collected by the 1991
census of India in reserved and unreserved GP, and their differences. Panel
variables,
and panel
B
shows village
level variables.^
''For the next election (in 2003), every third
etc...
The Panchayat
GP
A
shows
As expected given the random
starting with the second on the hst
Constitution rule has actual tables indicating the ranks of
was reserved
for
GP
level
selection
a woman,
GP to be reserved in each election.
GP level, but the statistical
''The standard errors, like in the rest of the paper, are adjusted for clustering at the
of
GP, there are no
significant differences
between reserved and unreserved GP. Reserved
At the
GP
of public health facilities (hospital
and
have a somewhat smaller total population, but this difference
level,
the total
number
number
of health facilities, the
is
not significant.
primary health centers and subcenters), and the number of schools of
reserved and unreserved GP. Unreserved
the village level as well,
very few villages (3%
drinking water being
91%
all
are marginally
more
all
types
is
very similar in
have a hospital. At
likely to
variables are very similar in reserved and unreserved GP. Note that
among
GP) have tap
the unreserved
handpumps and
Most
tubewells.
water, the most
common
villages are accessible only
sources of
by a
dirt road.
of villages have a primary school but very few have any other type of school. Irrigation
important:
43%
of the cultivated land
is
few villages (8%) have any public health
We collected the data for this
GP
GP
GP
We
Pradhan.
(some land
irrigated
is
irrigated in
all villages).
is
Very
facility.
study in two stages.
First,
we conducted an
interview with the
asked each a set of questions about his or her family background, education,
previous political experience, and pohtical ambitions, as well as a set of questions about the
activities of the
villages in the
GP
Pradhan
GP
May
since his or her election in
GP: two
resides.
randomly
villages
We
1998.
selected in each
then completed a survey of three
GP, plus the
During the village interview, we drew a resource
The map
a group of 10 to 20 villagers.
featured
all
map
1998.
We
mapping
For
and
is
all
of the village with
built or repaired since
Previous experience of one of the authors, as well as experimentation during the
method
pre-testing period, suggest that this
village.
which the
the available infrastructure in the village,
and we asked whether each of the available equipment items had been
May
village in
yields extremely accurate information
about the
then conducted an additional interview with the most active participants of the
exercise, in
outcomes
for
which we asked
which
at the village level.
it
The
was
in
more
possible,
detail
we
two years was
easy.
investment in the entire GP, and
is
insignificance of these differences does not
^The questionnaires are
collected the
village level information
not provided by the Pradhan, and because
village in the last
about investments in various pubhc goods.
available
is
same information
likely to
be more
for villagers, recalling
at the
reliable,
investments
However, the information given by the
GP
GP
'^
level
because
it
made
in their
head
refers to
thus free from sampling error. Therefore, when an outcome
depend on
this correction.
upon request or on
line at
http://web.mit. edu/eduflo/www/.
is
we perform
available at both levels,
Due to the randomization
the analysis separately for both, and compare the results.
built into the policy, the basic empirical strategy is straightforward.
Under the assumption that the reservation does not
reduced form
effect of
the pohcy can be obtained by comparing the
and unreserved GP. Denoting
interest in reserved
investments in unreserved GP, the
affect
Yi the value of the
investment in drinking water between 1998 and 2000) and Ri a
reserved for a
woman,
this
is
mean
of the
outcome of
dummy
equal to
outcomes of
interest (say,
1 if
GP
the
is
simply:
E[Y,\R,
=
-
I]
E[Y^\Ri
=
0]
In the village level regressions, the standard errors are adjusted for possible correlation
GP
within
using the Moulton correction (Moulton (1986)).^ Unless otherwise indicated,
village level regressions using only the
data
Pradhan's villages are not random and
for
we
the two villages
may be
we run
selected randomly, since the
selected differently in reserved
and unreserved
GP.
Since
all
the reserved
GP
reduced form coefficient
this
have a female Pradhan, and only 6.5% of the unreserved
is
GP
do,
very close to the coefficient that one would obtain by using the
reservation policy as an instrument for the Pradhan's gender.^
We
reduced form estimates, which are directly interpretable as the
effect of the reservation policy.
When
interpreting these results,
from men
likely to
in
many important
it
should be kept in mind, that
women
elected as
be new leaders, and they are probably
all
less likely to
on the
Pradhan
dimensions, other than gender. In particular, they are
The reduced form estimates capture
A
will therefore focus
differ
much more
be re-elected in the next election.^
these potential effects.
very interesting feature of the experiment in West Bengal
is
that
it is
possible to try to
disentangle the effect of gender per se from these other effects of reserving electoral seats to
specific groups.
The outcomes we consider
because the regressor
produce
coefficients
is
the
are jointly determined, since they are
same
in all
outcome equations, a
and standard errors numerically
hnked by a budget constraint. However,
joint estimation of the
identical to
OLS
system of equations would
estimation equation by equation.
^The instrumental variable estimate would simply be the reduced form estimate scaled up by a
(The
ratio of the reduced form effect
unreserved GP, which
is
equal to
and the
difference in the probability that a
woman
is
factor of 1.075.
elected in reserved vs.
93%
^Recall that the reservation rotates: seats that were reserved in 1998 will not be reserved again in 2003.
10
we can
First,
to
control for whether the
compare investments
non-random sample
unreserved
GP
GP
in all unreserved
the fact that the Pradhan
is
new may
Pradhan
where the Pradhan
is
always
to be Pradhan.
As part
reserved for a
woman
new
We
to office.
if
in the 1998 election,
can therefore compare
Pradhan
control for the fact that the
There
results.
new
is
Once
election.
Pradhan
as
Pradhan
of
member
(if
in those
GP
GP
run
for
a council
elected, the councilors choose
one of
one third of council seats (identified by
women Pradhan
to this subgroup of
list will
be
new Pradhan,
to
politicians.
be re-elected in 2003.
likely to
be reserved
will
does not fully control for the
in office. Clearly, this
is
GP
for that
Every
a female Pradhan for the 2003
for
should realize that they will not be able to stand for re-election
their particular seat
of the
may
we can be sure that the Pradhan
new
starting with the second in the
GP:
the previous councilor was a man, and his seat was
Second, we can control for whether the Pradhan
GP
in reserved
however, a random subset of
is,
Pradhan's experience: even new Pradhan could be experienced
third
would not be legitimate
It
Individuals
in ofRce.
of the reservation scheme,
were reserved for women: thus,
village)
or not.
unobserved characteristics of the GP, and this
seat only in the village in which she or he resides.
them
new
where Pradhan are new to those
reflect
would bias the
selection
is
council).
is
We
not reserved, they
may
still
be able to run
therefore restrict the sample to
GP
for
a position
reserved in 1998 and
those which will be reserved in 2003, to examine whether and to what extent the differences
observe are due to the fact that
Again,
men
could
still
women may
have a longer horizon
we
not think they have a chance to be reelected. ^°
in office
than women,
if
they plan to be elected on
another position or be elected again in 10 years.
Finall}',
we take advantage
of the reservation of about
reservation were also selected randomly, and within each
for
to
women.
Irrespective of their gender,
be new leaders and to be elected
'"There are other reasons
be reserved
table of
for a
SC
or
why
a Pradhan
it is
much more
indeed be reserved. His seat could also
we could not
reserved
in
may
list,
of the seats to
one third of positions were reserved
due to the quota system. They also tend to
not be able to be a candidate in the next election: His
not exploit this because the system of
We
reserved.
do not have the
exploit this either. This does not invalidate our strategy:
2003, the term limit does apply for
all
SC
reservation
is
GP
could
based on a
a Pradhan (and us) to figure out whether the seat will
difficult for
become
SC and ST. These
the leaders elected under this reservation policy tend
in large part
ST Pradhan. We do
random numbers, so
all
44%
Pradhan.
11
it
is
list
of serial
numbers within GP, so
clear that in the subset of
GP
to be
be poor, and to come from small
reserved for
villages.
SC and ST. Remaining
and those reserved
for
female
in
sample to
GP
GP
that are
reserved for
SC/ST
should be largely attributable to gender.
Why Would Mandated
3
restrict the
between investments
differences
SC/ST
we
Therefore,
Representation of
Women
Affect Pol-
icy Decisions?
Women
and men have
different policy priorities. In
known
to support liberal policies, a difference
voter pool
is
of investment.
The primary
are to fetch water
and
fuel
than
Thomas
men about
(1994),
likely
of the
women's and
needs and favor different types
in rural
West Bengal, besides working
and to take care
of children.
to men's
income
Child health has been
(see for
example Thomas
and Duflo (2000)), which indicates that women are more concerned
child health. It has also
in collective decision
more
different
women
responsibilities of
shown to be more responsive to women's income than
(1990),
are
The composition
as the "gender gap".
and they may therefore have
different,
fields,
women
therefore likely to affect policy outcomes. '^^ In developing countries,
men's roles are
on the
developed countries,
making
been suggested that the degree of women's involvement
affects policy decisions.
village forest conservation
committees
where women participated
in their elaboration
However, the fact that
in India
For example, a study of rules
shows that rules are very
set
up by
different in villages
(Agarwal (1997)).
men and women have
different preferences
is
not enough to imply
that policies would be influenced by direct manipulation of the gender of representatives through
quotas. In a Coasian world, public good allocation should be unaffected (the efficient allocation
should be achieved both before and after the policy).
policy confers to
women
are not available,
poficies,
and
if
if
The
additional bargaining power the
should translate into higher transfers to women. Even
if
these transfers
the candidates could commit during the electoral campaign to a set of
candidates knew the preferences of the voters, then electoral incentives would
cause candidates to commit to the policy bundle favored by a majority of electors. In this case.
"Lott and Kenny (1999) show that giving women the
They argue that women's preference
for bigger
government stems from
and Pande (Forthcoming) suggest that the gender gap
the gender gap
is
bigger in states where divorce
right to vote increased the size of state governments.
is
is
easier.
12
their greater need for insurance.
Edlund
indeed related to women's expected need for insurance;
the gender of the elected
official
would not
affect policy.
There are therefore two main
argument why mandated representation of women could
of
The
affect policy decisions.
based on the idea that elected representatives cannot commit ex ante to a set of
this context, political reservations for
selection.
for
Pande (1999)
discusses
women
affect
presence of
women
In
both the gender of the politicians and
its
in the
context of the reservations
and presents
for the legislative assemblies in India,
The second
evidence that the reservations did affect outcomes.
first is
policies.
most of these arguments
Scheduled Castes and Scheduled Tribes
lines
is
based on th6 idea that the
mechanism by which preferences are aggregated.
representatives affects the
Empirical evidence suggests that the preferences of legislators strongly affect the decisions
they make.
For example, Levitt (1996) shows that senators' individual ideology
most important factor explaining the way they
to a set of policies, the identity of the elected
(1988),
be a
Osborne and
woman may
preferences
is
Slivinski (1996), Besley
different policy priorities
The
idea that
women
than
men
it
the politicians do not commit ex ante
official will
if
determine policy decisions (Alesina
the overlap between women's and men's policy
has been shown repeatedly that female legislators have
Thomas
(see, e.g.
tend to be more public spirited,
advanced as an argument
in favor of active
and spend more time advocating, women's
welfare issues (such as child support), while
(1991), Carroll (1994), Thomas (1991)).
al.
and more
less corrupt,
measures to
pohtical decisions (World-Bank (2001), Dollar et
to,
If
facilitate
(1999)).
"fair" is often
women's participation
Women
issues (abortion, equal opportunity)
men
will
The
concentrate on business and economic
reservation pohcy does not affect only the Pradhan's gender, however.
is
chosen. First, since there were very few
the reservation policy was enacted,
women
policj'.
women than
the system better. Alternatively, newly elected leaders
'^Saltzstin (1986) also
women
It affects
the
active in politics before
elected under the policy are less likely to be incum-
bents. This could have positive or negative effects; incumbents
in
and child
be adopted (Berkman and O'Connor (1993), Besley and Case (1997)).
process by which she
women
in
attach more importance
Thus, the number of female legislators predicts whether policies preferred more by
men
the single
and Coate (1997)). Constraining the candidate to
then affect policy decisions,
not exact. In the US,
vote.
is
clout
and know
willing to start
more new
may have more
may be more
shows that municipahties with women as mayors tend to have a higher proportion of
municipal employment, which suggests that
women
13
placed in power try to act in women's interest.
projects.
Second, the reservation constrains voters to choose
if
there
is
not a competent
to choose one in reserved
woman
GP if there
thus affect the leader's quality.
man,
and
less assertive,
Third,
less likely
women may have
little
than
hope of being
Khemani
may
may
GP
may
constituents in order to be re-elected
may
affect policy decisions
It
this
women's quotas could
men, or
the sex of the Pradhan
all
women
very limited (only
is
a
Pradhan'
demanding
vi^ho face
more
"vigilant"
(village meeting) institution
activities.
for
if
women
fielded
rights than
women
fielded
For example,
women's
may not be relevant
even
may
effect
how information
is
done in the
villagers.
is
if
It
could thus
in the context of reservation
elected
women had
maximize the same
is
Gram Samsad, and
during
However, the participation of
of participants
woman may
were women
women
the same preferences
social welfare function:
transmitted. Both legally and in practice,
and decide how to spend
to aggregate information
in the 20
office
women
hours at which the
at the
Gram Samsad
Gram Samsad
studied by
are present, they rarely raise questions.
then affect the participation of other
''Besley and Case (1995) present evidence that this effect
Governors
potentially
representatives act simply as proxy for their husbands. However,
and Ghatak (1999)), and even when
that a Pradhan
is
Pradhan, which suggests that the need to please
elected officials attempt to
This
9%
The Gram Samsad
of these factors
the most important role of the Pradhan
Pradhan meet with
This
ambition.
less pohtical
marginal male voters toward the right.
affect policy decisions
if all
relevant resources.
have
quota system, and since
even on dimensions that are not directly relevant to women.
because most
as elected
re-elected outside the
affect
less
may push some
has been suggested that
in India,
educated than the average
also less
alter political competition.
by parties on the right were known to be
left,
no competent man). This
is
(2000) suggests that Indian voters are
ensures a high level of accountability for
by parties on the
(while they are
provide an incentive for the elected representatives
in national elections.
Fourth, the reservation policy
is
GP
Even
to have worked outside her home.
important, as the desire to be re-elected
to behave while in office.'^
in unreserved
The average woman
the seats are reserved by rotation, they
in state elections
set of candidates.
woman, they have
always free to choose a competent
may
among a Umited
term limits make decisions that are
is
important for governors
less likely to please
14
women
is
Ghatak
The
fact
to this process
in the
United States.
voters (such as increasing taxes).
of aggregating information.
encourage other
office hours.
women
woman
fact that a
is
may
may have
be
less
Gram Samsad
very important effects on policy, even
very sensitive to women's issues (and even
if
she
who
is
more aware
Somanathan (2001) show that a
constituents
is
more
leader
likely to get
them
that materially alters her decision.
meetings or during her
Hkely to be taken desultorily.^'*
is
to talk and
Thus the
if
the leader
a proxy for her husband).
more
is
not particularly inclined towards women's causes
In the next section,
not herself
Banerjee and
likely to get
leader
information from them
knows the other women
Pradhan (including
their prior political experience
Pradhan on women's
and
their
of these intervening mechanisms:
background, their
their political ambitions)
participation. In section
5,
if
she
herself.
we provide evidence on some
characteristics of the female
to
of the underlying preferences of her
woman
fact that a
women
Enabling
better than the male leader she replaces, will help her elucidate better their needs, even
is
may
present in a position of authority
to raise issues, either during the
Their remarks
express their concerns
The
we turn
and the
to the
the
political affiliation,
effect of the
main question
gender of the
of interest: does
the gender of the Pradhan affect the the provision of public goods in the community?
Leaders and Voices: Characteristics of Female Pradhan and
4
Pohtical Participation of
4.1
Characteristics of Female
Pradhan
Table 3 presents diiferences by gender
political
(3) presents
taged background, which
may be
is
less likely to
some evidence indicating
woman
legislator
be
surprising:
this
(2)
presents the average in unreserved GP.
Standard errors are
the difference.
and
to which a
Pradhan's background (panel A), her previous
average in reserved GP, and column
significantly less educated
'""There
in the
involvement and political ambition (panel B), and other's perceptions of her. Column
(1) presents the
Column
Women
may be
literate.
in parentheses.
They
also
come from a more disadvan-
they are more likely to be
important: In the US,
Thomas
referred to earlier also indicates that
women
talk
more when there
15
officially classified as
(1991) show that the degree
supports women's issues depends on the number of other
the existence of a female caucus. In India, the study of joint forest
Female Pradhan are
women
in
the assembly, or
management committee by Agarwal
are
more women
in
the meetings.
(1997)
being below the poverty
and they own fewer durable household goods. ^^ They come from
line,
smaller villages than men. This does not contradict the fact that
from unreserved GP.
different
women come from
The
often
come from the
reserved for
Our
interviewers are
no
more
answering on their behalf).
results in panel
B
Pradhan have
of table 3 confirm that female
89%
of
women
(and
57%
less political
None
activity.
less likely to
to train the
are
them had served
of
as
Pradhan (12%
have received any training than
new Pradhan and counselors
more
likely to
be helped by
men
of
that they are helped by their spouse, compared with
of the
13%
Women
are also
more
women
any Panchayat
are only slightly
governments
(Kanango
efforts
(1998)).
in reserved positions report
of the men).
17%
women may be
former Panchayat counselor or Pradhan. The worry that
thus not without foundation.
reflects
elected under the reservation rule
(43%
in
Women
had).
men Pradhan, which
their spouse
experience
men) holding reserved positions had not
of
been elected to any Panchayat position and most had not even participated
is
are
female Pradhan are shy in answering the questions (and that someone else
prior to the 1998 election.
They
women
biggest villages within their GP, while
villages that are similar to the average village.
likely to report that
may be
Men
GP
are the spouse of a
their husband's
likely to claim that
they
proxy
not run
will
again.
The
in
reservation policy does not
each political group
seem
(left front, right,
to affect party affiliation: the percentage of
and Congress)
is
identical in reserved
Pradhan
and unreserved
GP.
4.2
Effects on the Political Participation of
Table 4 displays the
The percentage
of
the Pradhan
a
is
effect of
having a
women among
woman
woman Pradhan on
participants in the
women, and a
Gram Samsad
Gram Samsad,
is
women.
significantly higher
when
is
men. This
is
consistent with
influenced by the fact that citizens and leaders are of
'^We asked whether the Pradhan 's household owned a
car.
this corresponds to a net increase in
decline in the participation of
the idea that political communication
a bicycle, a motorcycle, and a
the political participation of
(increasing from 6.9% to 9.9%). Since reservation does not affect the
percentage of ehgible voters attending the
the participation of
Women
The durable goods
television,
variable
16
is
had
electricity in the
simply the
sum
of
all
home, had a telephone,
of these variables.
the same sex.
Women
more
are also
increases from 0.29 to 0.38, although the difference
women
in villages
complaint to the
that the
Pradhan
affairs of
the GP.
is
not significant at the
with reserved Pradhan are twice as
GP
is
Pradhan
woman
a
When women had
in
Gram Samsad
a question at the
likely to ask
likely to
95%
(the proportion
level). Finally,
have addressed a request or a
the last 6 months, and this difference
is
significant.^^
therefore significantly increases the involvement of
raised an issue,
questions were open ended, and
we
we asked the
classified
villagers
them ex-post
The
women
which issue they had
into broad categories.
shows the range of questions, the fraction of
by women
far
(in
column
(1)),
and by men
the issues most frequently raised by
The next most important
issue
(in
Our
then
surveyed a random sub-sample of 48 villages, and asked about issues raised by men.
5
in the
raised.
We
fact
re-
Table
total questions concerning each issue asked
column
(2)).
Drinking water and roads were by
women (26% and 24%
of the questions, respectively).
was welfare programs (these programs mainly target women:
they are maternity grants, widows' pension, and old age pensions), followed by housing and
electricity.
Road improvement
is
most frequently raised by men (22%). Three
also the issue
other issues are raised relatively frequently by men:
drinking water (14%). This
is
consistent with the evidence in
questions raised (by men) at 21
Gram Samsads
in
Ghatak and Ghatak (1999) on the
West Bengal: The questions that appeared
most frequently have to do with mis-management or corruption
about half the
villages), irrigation needs,
or of project locations.
Road
and
irrigation (17%), education (14%),
in
program implementation
(in
road building, and the choice of program beneficiaries
building thus seems to be a concern
common
to
men and women:
road construction projects are employment generating activities (especially for women), and
poor roads also make communication very
difficult
concerned about drinking water than men, and
In the next section,
less
during the rainy season.
if
the effect
are
more
concerned about irrigation and education.
we examine whether the gender
the type of public goods provided and
Women
is
of the
Pradhan
affects the quantity
and
consistent with the preferences described
above.
'^In the
up any
subsample of
villages
issue in the last 6
which we conducted follow up surveys we also asked whether men had brought
months. In
all
cases but one (a reserved GP), they had.
17
5
5.1
on Public Goods Provision
Effects of the policy
Public Action
In table
6,
we examine whether the
We
vation policy.
the Pradhan
is
have seen above that
a woman.
Is it
To answer
of organizations that the
GP
of water
and
forest
reser-
women were more
active politically in villages
where
because female Pradhan are generally more effective at inducing
collective participation?
is
by the
level of organization in the village is affected
we examine
this question,
variables that reflect the extent
supposed to sponsor: number of women's organizations, number
management committees, number
of village education committees, areas
brought under social fishery and social forestry, and proper organization of the
For none of these dimensions of organization
is
Gram
Samsads.
there a significant difference between male and
female GP.
For a subset of
GP
(100),
depend mostly on the state
collection),
we
collected information on their
for resources (less
than
8%
but they are authorized to collect revenues
population, the
women. Tax
amount
of funds collected
by the
collections as well as revenues
are smaller in
GP
GP
GP
headed by women,
possible explanation for these results
is
is
in
Even
collection.
come from
GP
GP
(fisheries, forestry,
tax collection
is
is
far
GP
reserved for
ponds, etc.),
small compared to other
only half as large as in non-reserved GP.
that revenue collection
GP
revenue
after controlling for the
significantly smaller in
from other sources
headed by women. The difference
source revenue, which, in
of their budget
locally.
is
own revenue
One
from being straightforward:
collecting revenues requires effective persuasion power, with respect to the individuals
the taxable properties, and to the representative of the villages where the
GP
who owns
ponds and
forest
are located.
GP, however, depend mostly on the state
GP
population. In the 40
funds), the total budget
5.2
Public
is
GP
for
for resources, allocated largely in
which we have complete budget data (including government
not significantly different in male and female GP.
Good Investments
Table 7 presents the effects of the Pradhan's gender on
the
GP
proportion to the
since the last election, in
May
1998.
all
public good investments
As we aggregated investments
18
made by
in categories, these
regressions reflect
GP
and
all
Both
GP
of the
Pradhan
results are consistent across the
affects the
water equipment and roads in
GP
GP
present both village
and
village level data.
type of pubhc goods provided.
GP level and at the village level,
at the
We
good investments.
collected on public
The main
level regressions.
The gender
we
the data
there
is
where the Pradhan
is
significantly
more investment
a woman. In addition, there
in drinking
is
also
more
investment in other labor intensive projects (culvert and minor irrigation canals), a variable we
have only
'at
the
much investment
GP
level.
The magnitude
in drinking
of these effects
GP
water equipment in
is
there
large:
where the Pradhan
roads are almost twice as likely to be in good shape. In
all
GP
is
is
almost twice as
a
woman, and the
where the Pradhan
is
a woman,
tubewells were built, and labor intensive construction work was undertaken. Major roads are
20% more
likely likely to
have been repaired in women's
GP
(dirt
roads were repaired in
all
GP).
The
village level regression
(though not the
GP
regression) also
in the introduction of biogas projects (biogas projects
GP, and 26%
used
for
of the reserved
where the
cooking
forest cover
Both the
GP
fuel, as well as
and the
it
6%
of the unreserved
and
primarily a woman's task, and in Birbhum,
particularly difficult.
woman Pradhan on
should be noted that
large in the village level
investment in sanitation.
women never
raised sanitation
issue.
Finally, there
Since 1998, the
is
GP
less
investment in informal education
school in a temporary building (or outdoors).
from the higher
and the
level of
village level, the
a Pradhan.
The
for children in
GP
headed by women.
have been given the responsibility of organizing informal education centers,
by hiring and paying someone without formal
is
is
show a negative (and
village level regression
This might seem somewhat surprising, but
an
in
are a substitute for cooking fuel
wood can be
regression) but insignificant impact of having a
as
They
water,
very limited, obtaining
is
were introduced
GP). Biogas projects are installations that produce methane gas,
cooking or lighting, from animal waste.
electricity. Collecting
shows a very large difference
qualifications to teach children not attending
They
receive the authorization to set a center
the panchayat or from the block level administration. Both at the
number
of centers
is
negatively associated with the fact that the
coefficient is significant at the
point estimate in the village level regression
is
95%
only in the
GP
level regression,
large in magnitude: villages located in
19
up
GP
woman
but the
GP
where
Pradhan
the
in
GP
a
woman
are only half as likely to have an informal education center than villages
where the Pradhan
number
the
is
a man. To confirm the validity of these results,
is
is
significantly negative for all these variables.
the repair of school buildings. However, in
GP
buildings used for adult education (of which
GP
level,
There
is
the coefficient of the
no difference
headed by women, there
women
more often
are
is
To shed some
on
GP
more investment
this result,
we
restrict the
sample to
GP
random): the
effect is smaller
women on
among
children's education
Note that
GP
headed by
^'^
for adult education.
where the Pradhan
is
is
literate
literate is certainly not
hterate Pradhan, suggesting that
may be
in
beneficiaries).
earmarked
receive an allocation specifically
has to be taken with care, this the fact that the Pradhan
(this result
effect of
light
all
woman
in investments in
there are almost no investments in the repair of the adult literacy centers in
males, even though
also regressed
and children per capita, at the informal
of teachers, teachers per capita, children,
center (the regressions are omitted to save space). At the
dummy
we
some
of the negative
related to the fact that they are themselves less
educated.
These
important
results suggest that the reservation policy has
at the local level.
These
effects are consistent
women, and a
priority of
are also important issues for
men.
If
It
priorities are closely
should be noted that women's main preoccupations
we combine women's and men's requests (and
give an equal
weight to both genders), drinking water and roads are by far the most important
decisions in reserved
GP may
The
than
issues.
The
thus correspond better to the entire's village preferences.-'^
'^Female GP, on the other hand, are no more hkely than male
^^It is
women
women, and education was a low
men) and with the view that women's
linked with their role in the household.
on policy decisions
with the policy priorities expressed by
(drinking water, road and power were the highest priorities of
priority of
effects
GP
to have had a hteracy
campaign
since 1998.
unfortunately not practical to assign a monetary value to these investments, and thus to compare them.
fact that
men
in
most investments are equal or greater
in
women's
spending the available resources.
20
GP
suggest that
women may be more
efficient
Influence
5.3
The
GP
has no control over the estabhshment of health centers, or administrative appointments
such as health workers, doctors, or school teachers.-'^ However,
control over health workers and teachers, although in practice
limited. Likewise, they have
centers
no
it is
its
supposed to exercise some
influence over
them
often
is
direct control over the establishment of pre-primary child care
(Angan Wadi), which are controlled by a centrally administered program (The Integrated
may
Child Development Service), but they
seek to influence their placement.
examine whether the gender of the Pradhan
The
indirect influence.
first
also affects
two rows indicate that the gender of the Pradhan has no influence on
since 1998, but they were equally distributed in female-headed
surprising, as child care
are
much more
GP
since 1993)
are
more
is
generally thought of as a
is
The next two rows confirm the
Pradhan
results
likely to report
Since this
is
the Pradhan's
that they have influence.
GP
woman's
education.
in the
report low teacher turnout as a problem. Since this
report, this could indicate that
The next row
all
men
suggests that the villagers do
when asked
primary schools, men were more
is
Male
education committee (instituted in
own
is
priority.
not find male Pradhan more effective in controfling the primary schools. However,
which problems they thought were present
on average
and male-headed GP. This
we obtained regarding informal
likely to report that the village
effective.
we
8,
outcomes over which she yields only
the opening of pre-schools. There were slightly fewer than two centers open per
somewhat
In table
likely to
one of the principal problems of primary
education, and one that effective monitoring by village institutions could most likely affect,
this suggests that
male Pradhan are better informed, and may be more concerned, than female
Pradhan about the quality
that schools perform less well in
GP
to report that teacher attendance
The next rows suggest that
is
headed by men: when asked, the
a woman. This difference
villages that
'^We
do receive
collected data
visits
is
fact
villagers are equally Hkely
a problem.
the opposite
by health workers to the village in the
is
GP. This does not seem to be due to the
of education in the
last six
is
true for health workers.
months
is
The number
significantly higher
entirely attributable to a difference in the
by health workers, not a difference
on these variables and
verified that the
by the gender of the Pradhan.
21
number
in the
when
the Pradhan
number
number
of visits
of visits to
of villages in a
of health workers or teachers
is
not affected
GP
may be due
that receive visits. This suggests the difference
to active monitoring:
have no influence on the route of village health workers, but female Pradhan seem to be
in getting health
workers to actually
Do Women
5.4
Our
visit villages
Invest Less in Their
far.
The
in the rest of the paper, since the
selection
may be
Pradhan's village
(i.e.,
is
and non-reserved
who
those
village,
which
not as straightforward as
is
not chosen randomly, meaning sample
As we have seen above, male
villages.
female Pradhan, especially those whose husbands
villages, while
are not politically influential
Villages?
interpretation of these results
different in reserved
Pradhan come from bigger
Own
eflFective
visit.
data on investments in the Pradhan's
last set of results in this section uses
we have not exploited thus
they are supposed to
Pradhan
are not "helped in their jobs")
come from more
ordinary villages.
Nevertheless, the results in this data are striking. In table
the same investment variables as in table
=
y'lj
where
y^, is
GP
is
and non Pradhan's
The
7*^.
village.
village,
first
all
We
* P,j
+
^Population
(3^ is
and
in
Pij
is
a
dummy
i,
men's village
is
+
Ri'is
a.
e'lj,
dummy that
indicates
that indicates whether village j
the expected difference between investment in Pradhan's
GP, and
7*^
indicates whether the difference
control for the size of the villages, which
striking fact
in particular.
that almost
all
Columns
(1)
is
and
is
the same
systematically different in
(2) of table 9 report
/3'^
and
investments are significantly larger in the Pradhan's
These differences are usually larger than the constant, indicating that the Pradhan's
village receives
size.
woman, and
village in unreserved
reserved GP.
Pradhan's
is
reserved for a
use as a dependent variable
run the following regression:
+ o/'Ri + /j'^Py + -i^R,
the Pradhan's village. Thus,
among
We
the investment in good k in village j of Panchayat
whether the
is
C*
7.
we
9,
more than twice
as
much
£is
other villages receive, even after controlling for their
Second, the coefficient of the interaction between the Pradhan's village and female Pradhan
negative for
all
outcomes but one, and
significant for three individual
investments taken together (shown in the last
of the
same magnitude
line).
as the coefficient of the direct
The
outcomes as well as
coeflncients of the interaction are
Pradhan
effect:
there
is
investment between the Pradhan's village and other villages when the Pradhan
22
for
no difference
is
a
woman.
in
As we discussed
especially
Pradhan tend
in section 4, female
when they
are not helped by their husband.
between male and female-headed Pradhan
To examine whether
in these regressions.
to
villages,
this
is
come from smaller and poorer
This
may
villages,
account for the discrepancy
even though we control
for
population size
the case, we ran the same regressions using only
male Pradhan and Pradhan who are helped by their husbands. Female Pradhan who are helped
by
their
husbands come from
villages
which are similar to those of male Pradhan. ^° The
are at least as strong in this subsample.
This suggests that the coefficients of the interaction
Pradhan do indeed
are not pure artifacts of a composition bias in the sample, and that female
invest less in their
This difference
own
villages.
may be
explained by different levels of political ambition
men. To be re-elected, the Pradhan must
own
Because women are
village.
first
less likely to
among women and
be elected as a council member in his or her
own
seek re-election, they need to please their
Levitt (1996) shows that senators are more likely to cast votes benefiting their
electors less.
own
results
constituencies
when an
position vis-a-vis the other
election
is
near. Alternatively,
GP member
women may have
a weak bargaining
on the locations of these investments.
Separating Gender and other effect of reservations
6
The
results
women.
we
discussed so far could be attributed to other characteristics of the newly elected
they are inexperienced politicians. Second, they are more
First,
(by themselves or others) as "lame ducks"
.
Finally,
from smaller villages than men. All these factors
we have
may
likely to
be perceived
seen that they are poorer, and
come
affect their decisions. In this section,
we
use specific reservation rules to control for these characteristics.
6.1
Women
In our sample,
of
are
all
new Pradhan
the
women who
were elected
in reserved position are
new to the
Pradhan, whereas 12% of male Pradhan have previously served as Pradhan.
have more experience as member of the
^°The difference between the population
their
husband
is
only 250, and
is
not
GP
position
Males
council. This could have a direct influence
of villages of
significant.
23
also
on the
male Pradhan and that of female Pradhan helped by
investments
made by men and women:
On
their objectives.
They may
To
the other hands, they
also favor
new
may
have a hard time achieving
be more energetic and want to start new project.
than continuing existing projects.
projects, rather
investigate whether the effects of reservation
newly elected, we compare them to another
reservation policy at the village level:
may
inexperience Pradhan
due to the
in part
is
Pradhan who were newly
set of
we compare investments
in reserved
where the previous Pradhan was prevented from re-election because
a
women
fact that
GP
his village
are
elected due the
to those in
GP
was reserved
for
woman.
These
results are presented in
column
(2) of table 10.
For comparison, column (1) reproduces
Pradhan are elected
the results for the entire sample. In this sub-sample,
all
Men
they are
are, nevertheless,
have served in the
more experienced
GP
in the entire sample).
politicians:
council before, but the difference
The
point estimate for
all
now
is
more
still
for
likely
the
than
first
time.
women
to
smaller (0.24 instead as 0.32
the other results are unaffected, except for
the probability that a metal road was repaired, where the differences between reserved and
unreserved
GP
falls
from 0.18 to
0.09.
The
coefficient
on the quality of the roads
in the village,
however, remains unchanged, suggesting that day-to-day repairs remain more frequent in villages
in reserved
GP. This may suggest that repairing a main road
Pradhan does. This may be a very pubhc gesture, since
it is
among
is
the
first
new
thing a
at the top of the priority of
both
men and women.
However, none of the other results change. This suggests that the
has on policy cannot entirely be attributed to the fact that these
Women
6.2
we
present the difference between
GP
will
be reserved in 2003. In these GP, the Pradhan faces a term
will
be reserved
the
in
will
be reserved
for a
likely
than
new Pradhan.
reserved in 1998 and those
limit.
We
women
Pradhan themselves
woman. Moreover, men could have a
career than just the next election. In fact,
more
are
determine which
who
GP
2003 by following the allocation rule (which was strictly followed in 1998).
course, this does not necessarily imply that the
GP
women
women
are lame ducks
In column (3) of table 10,
Of
effect reservation for
to say that they
Pradhan
in
want to run
24
GP
effectively realize that
longer time horizon for the
that will be reserved in 2003 are
for the
GP
council again (they can
still
still
be a candidate for the council as long as their seat
The
results for
above, this
still
Other
6.3
It is also
women,
all
some extent
women
domly
selected to be reserved for
in reserved
women. Column
GP. Since
SC and ST
reserved for
this policy
tistically significant.
Men
is
thus
are
still
was
22% have
much
(4) in table
more
women. They
likely to
effect
on the number
are unreserved for
The
less
what we found
last
row
hkely than
reserved for
SC
are
more
to invest in their
or ST,
men
are
Irrespective of their gender,
more
all
member. The
may run
same
GP
visits
own
is
by doctors
villages.
GP
The
in
most
results
was a consequence
size,
result
own
is
Pradhan
SC/ST GP
difference in the
Importantly,
and men are not
25
or ST.
The
results
the weakening of the
fact that
even
GP
that
villages).
SC and
ST,
not changed:
village,
while
women
are
even in
GP
women
aren't.
are elected because of the quota system,
remain unchanged suggest that the difference
of the Pradhan's gender.
is
SC and ST
reserved for
likely to invest in their
SC/ST Pradhan
due to the
most of them have no experience, and women and men share similar
fact that
in
again.
SC
reserved for
GP. The only notable exception
of table 9 examines whether, even in
men
male Pradhan
villages of the
of doctor visits in the village (which
women, there
GP
of the
however, better educated.
are,
in all
of the
say that they
In column (4) of table 10, we compare outcomes in
are very similar to
most
smaller (14%) than in the entire sample, and not sta-
among SC/ST Pradhan, women and men come from
significantly richer than
10 shows the differences female
also enacted in 1998,
served as
stratified:
must be ran-
"general" seats, one third
Only 2% (one)
are new.
has previously served as Pradhan, and
proportion of incumbents
among
and 11.5%, respec-
women be
prescribe that the randomization for
seats reserved for each group as well as
GP
tend to come from poorer families and smaller
their population in the district (34.5%
among
in
and other differences between men and
for these
minority reservations. Positions of Pradhan were reserved for Scheduled Castes
The 1998 Panchayat Rules
GP
the caveats mentioned
all
effects of reservation
in particular for the fact that
and male
With
suggests that the results are not driven by women's low immediate prospects.
and Scheduled Tribes, proportional to
tively).
not reserved).
other outcomes are essentially unaffected.
possible to control to
villages, using
is
social background.
in public
The
good investments
Conclusion
7
Mandated representation
Women
of
women has important
effects
elected as leaders under the reservation policy invest
women's concerns: drinking water,
closely linked to
fuel,
at the
government.
more
in public
goods most
visit
in
Men
informal schooling.
invest
also
the villages more often.
increase in investments in drinking water, fuel, and roads seems to
expense of investments
They
and road construction.
appear to exercise influence over health workers to induce them to
The
in local
on policy decisions
come
at least partly
more resources
in informal
education centers, and seem also more aware of the need to exercise influence over teachers to
elicit
is
more regular attendance. This
not as surprising as the
good
for children
relative lack of interest of
commonly
women
held view that
results
the Pradhan, rather than on
identical in the
other
its
GP
subsample of
effects:
to those where the
is
also
of the entire
These
priorities.
Leaders
new due
community than
in reserved
an important issue
GP
results contradict the simple intuition
which
GP
will
be
reserved to
to the reservation policy.
women
invest
more than men,
are also
GP.
behind the Downsian model, or the idea that
outcomes of a Coasian bargaining process:
the world, the fact that a
woman
the head of the
Indirectly, these results also confirm that the
GP
when we compare
political decisions are the
decisions at the local level.
men.
thus appear to better represent the preferences
leaders in unreserved
is
for
we
the policy on the gender of
SC/ST, when we compare
Drinking water and irrigation, the two issues in which
among men's
in the village
is
with only two exceptions, the results are virtually
reserved for
Pradhan
it is
effect of
reserved in 2003 to those that were reserved in 1998, and
women
women
might suggests: among the problems mentioned by
seems to be largely attributable to the
in charge for education
strongly support everything that
surveyed, education appears relatively infrequently, whereas
These
women
GP
in
both of these views of
should not influence pohcy decisions.
panchayat has effective control over the policy
These results suggest that direct manipulation of the identity of
the policy maker can have important effects on policy. This has important implications beyond
reservation policy, suggesting that
eligibility conditions, etc)
whether
this is
may
all
mechanism that
affect policy decisions.
affect politician's identities
These
due to the increased participation of women
26
results leave
(term limits,
open the question of
in the political process, or to the
fact that the decision
maker's preferences
now
women's
reflect
preferences, rather than men's.
References
Agarwal, B. (1997) 'Environmental action, gender equity and women's participation.' Develop-
ment and Change
28(1), 1-44
Alesina, Alberto (1988) 'Credibility and policy convergence in a two-party system with rational
voters.'
American Economic Review
Banerjee, Abhijit, and Rohini
78(4),
Somanathan
796-805
model of
(2001) 'A simple
voice.'
Quarterly Journal
of Economics 116(1), 189-227
Banerjee, Abhijit V., Paul L. Gertler, and Maitreesh
efficiency:
The economics
Berkman, Michael
legislators
B.,
of agrarian reform.' Forthcoming, Journal of Political
and Robert E. O'Connor (1993) 'Do
and state abortion
Besley, Timothy, and
icy choices?
Ghatak (Forthcoming) 'Empowerment and
Anne Case
policy.'
American
women
legislators
Economy
matter? Female
Politics Quarterly 21(1), 102-124
(1995) 'Does electoral accountability affect economic pol-
Evidence from gubernatorial term
limits.'
Quarterly Journal of Economics
110(3), 769-798
(1997)
'Women
in pohtics:
Finding instruments
for incidence analysis.'
Mimeo, LSE and
Princeton University
Besley, Timothy,
and Stephen Coate (1997) 'An economic model of representative democracy.'
Quarterly Journal of Economics 112(1), 85-114
Carroll,
Susan
J.
(1994)
Women
as Candidates in
American
Politics, 2 ed. (Indiana University
Press)
Dollar, David,
Raymond Fisman, and Roberta
Corruption and
women
in
Gatti (1999) 'Are
women
really the "fairer" sex?
government.' Mimeo, Columbia University
27
Duflo, Esther (2000) 'Grandmothers
allocation in South Africa.'
and granddaughters; Old age pension and intra-household
Working Paper 8061, National Bureau
of
Economic Research,
December
Edlund, Lena, and Rohini Pande (Forthcoming) 'Gender
riage.'
politics:
The
political salience of
mar-
Forthcoming, Quarterly Journal of Economics
Ghatak, Maitreya, and Maitreesh Ghatak (1999) 'Grassroots democracy:
A
study of the pan-
chayat system in West Bengal.' Mimeo, Development Research Group Calcutta and University of
Government
Chicago
of
West Bengal (1998) The West Bengal Panchayat (Constitution)
Department
Jones,
Mark
of Panchayats
P. (1998)
&
Rules, 1975
Rural Development
'Gender quotas, electoral laws, and the election of women: Lessons from
the Argentine provinces.' Comparative Political Studies 31(1), 3-21
Kanango, Sukla Deb (1998) 'Panchayati
In People's
Power and Panchayati
raj
and emerging
women
An
leadership:
overview.'
Raj: Theory and Practice, ed. Bansaku (Indian Social
Institute) chapter 5, pp. 77-95
Khemani, Stuti (2000) 'Decentralization and accountability: Are voters more
vs. national elections?'
Levitt, Steven D. (1996)
party
Lott,
John
affiliation,
R.,
vigilant in local
Mimeo, Development Research Group, World Bank
'How do senators vote? Disentangling the
and senator
ideology.'
role of voter preferences,
American Economic Review
86(3), 425-441
and Lawrence W. Kenny (1999) 'Did women's suffrage change the
of government?' Journal of Political
Economy
Moulton, Brent R. (1986) 'Random group
effects
size
and scope
107(6), 1163-1198
and the precision of regression estimates.'
Journal of Econometrics 32(3), 385-397
Norris, Pippa (2001) 'Breaking the barriers: Positive discrimination policies for
Liberalism Failed
Women:
Parity, Quotas
and
and Charles Maier (St Martins Press)
28
women.' In Has
Political Representation, ed. Jyette
Klausen
and Ronald Inglehart (2000) 'Cultural
Norris, Pippa,
barriers to
women's leadership:
A
world-
wide comparison.' IPSA 2000 paper
Osborne, Martin
J.,
and Al
model of
Slivinski (1996) 'A
candidates.' Quarterly Journal of
political
competition with citizen-
Economics 111(1), 65-96
Pande, Rohini (1999) 'Minority representation and policy choices: The significance of legislator
identity.'
Saltzstin,
Discussion Paper
16,
LSE,-STICERD
women
Grace Hall (1986) 'Female mayors and
in municipal jobs.'
American Journal
of Political Science 30(1), 140-164
Sen,
Amartya (1999) Development
as
Freedom (Anchor Books)
Thomas, Duncan (1990) 'Intra-household resource
of
Human
allocation:
An
Resources 25(4), 635-664
Parental education and child
(1994) 'Like father, like son, like mother, like daughter;
health.' Journal of
Human
Resources 29(4), 950-988
Thomas, Sue (1991) 'The impact
53(4),
inferential approach.' Journal
of
women on
state legislative policies.' Journal of Politics
658-976
Thomas, Sue, and
legislators.'
S.
Welch (1991) 'The impact
Western
Political Quarterly 44,
United-Nations (2000) 'The world's
United Nations,
women
of gender
on
activities
and
priorities of state
445-456
2000:
Trends and
statistics.'
Technical Report,
New York
World-Bank (2001) Engendering Development: Through Gender Equality
and Voice (Oxford University Press and World Bank)
29
in Rights, Resources,
Table
1:
Fraction of
Women among
Pradhans
Reserved
Total
number
% Female
in
Non
Reserved and Unreserved
reserved
GP
GP
(1)
(2)
Difference
(3)
54
107
100
6.5
93.5
(3.38)
GP
Table
2:
Village Characteristics
GP
Reserved and Unreserved GP, 1991 Census
Mean, reserved
Mean, unreserved
GP
GP
(1)
(2)
12606
13579
-973
(570)
(329)
(614)
Dependent variables
A.
in
Difference
(3)
Level
Total Population
Number of public
health facilities
0.76
1.10
-0.34
(.17)
(.17)
(.26)
7.15
7.10
0.05
(.90)
(.67)
(1.14)
Number of Hospitals
0.00
0.09
-0.09
(.05)
(.06)
Number of High Schools
1.19
1.37
-0.19
(.17)
(.18)
(.29)
0.69
0.66
0.02
(.11)
(.08)
(.13)
Number of health
facilities (total)
Number of Middle Schools
Number of Primary Schools
12.26
12.07
0.18
(1.00)
(.37)
(.88)
974
1022
-49
(60)
(46)
(75)
B. Village level
Total Population
Female Literacy Rate
Male Literacy Rate
% Cultivated land
that
is
irrigated
Dirt road
Metal road
Bus
or tram stop
Number
of Public health
facilities
Number of tuwells and hand pumps
Number of drinking water
wells
Tap Water
Number of Primary Schools
Number of Middle Schools
Number of High Schools
0.35
0.34
0.01
(.01)
(.01)
(.01)
0.57
0.58
-0.01
(.01)
(.01)
(.01)
0.45
0.43
0.02
(.03)
(.02)
(.04)
0.92
0.91
0.01
(.02)
(.01)
(.02)
0.18
0.15
0.03
(.03)
(.02)
(.03)
0.05
0.31
0.26
(.04)
(.02)
(.04)
0.06
0.08
-0.02
(.01)
(.01)
(.02)
1.13
1.20
-0.07
(.11)
(.08)
(.14)
0.44
0.47
-0.02
(.07)
(.04)
(.08)
0.05
0.03
0.01
(.03)
(.02)
(.03)
0.95
0.91
0.04
(.07)
(.03)
(.08)
0,05
0.05
0.00
(.01)
(.01)
(.01)
0.09
0.10
-0.01
(.01)
(.01)
(.02)
Notes:
1)
There are 161 observations
in the
GP level
regressions and
2120 observations
in the village level
regressions
2) Standard errors given in parentheses. In the village regressions they are corrected for clustering
at
the
GP
level
Table
3:
Pradhan's Characteristics
Depedent variables
A.
Reserved and Unreserved
GP
Mean, reserved
Mean, unreserved
GP
GP
(1)
(2)
31.87
39.72
-7.85
(1.08)
(.87)
(1.45)
-2.79
Difference
(3)
PRADHAN'S BACKGROUND
Age
Education
Literacy
Married
Number of children
Below poverty
line
Number of household
assets
own
Population of pradhan's
Hesitates
when answering
village
the questions
(interviewer's impression)
B.
in
7.13
9.92
(.48)
(.29)
(.54)
0.80
0.98
-0.19
(.06)
(.01)
(.04)
0.89
0.87
0.02
(.04)
(.03)
(.06)
2.45
2.50
-0.05
(.2)
(.15)
(.26)
0.46
0.28
0.18
(.07)
(.04)
(.08)
1.72
2.36
-0.64
(.18)
(.14)
(.23)
1554
2108
-554
(204)
(179)
(291)
0.75
0.41
0.34
(.06)
(.05)
(.08)
-0.32
PRADHAN'S POLITICAL AMBITION AND EXPERIENCE
Has been
GP
0.11
0.43
before 1998
(.04)
(.05)
(.07)
Has been
0.00
0.12
-0.12
elected in the
council
elected has pradhan
before 1998
Took
part in Panchayat activities
before elected
Knew how GP
functioned
Did not receive any formal
(.03)
(.04)
0.28
0.78
-0.50
(.06)
(.04)
(.07)
0.00
training
0.06
0.35
-0.35
(.05)
(.07)
0.00
0.06
(.03)
Was
spouse ever elected to the
(.02)
0.17
0.02
0.15
Panchayat?
(.05)
(.01)
(.04)
Does spouse help?
0.43
0.13
0.30
(.07)
(.03)
(.07)
0.33
0.21
0.13
(.06)
(.04)
(.07)
Will not run again
C.
PRADHAN'S POLITICAL PARTY
Left Front
Right (Trinamul or BJP)
Observations
Notes:
(1) Standard errors are given in parentheses
0.69
0.69
-0.01
(.06)
(.04)
(.08)
0.19
0.18
0.01
(.05)
(.04)
(.06)
54
107
=
o o
,^
=
o o
c
o
"(5
oo
Q.
"o
'c
o
0-,
O
CO
^ o
Q.
o
ro
o
o
0.
(/>
"c
E
o
>
_aj
-a
«
>
Oh
c
o
c
o
O
CM
^
O
m
oo
oo
iS
o
r-1
>
<L)
^
c
"(5
C/5
o >
K
£
C/3
(/)
0)
CD
q:
OO
<D
D-,
a
<D
J3
"c
a>
E
o
60
>
td
OJ
00
C3
<u
c
00 *c
cl>
(S
o
0)
LU
00
o
O
00
•a
TO
c
c
1-
c
D
Q
cA)
o E
E
3
O
T3
f3
c
o
a.
E
CO
ON
a3
g
o
E i
o S
»
Q
O
JS
o
.S
jf
a
j-
T3
T3
^^ fN
r^
C
—
C
3
O
Table
5:
Issues Raised by
Women
and Men
in tiie
Last 6 Months
Fraction of issues raised
Issue
Women
by
Men
(1)
(2)
Drinking water
0.26
0.14
Road improvement
0.25
0,22
Welfare programs
0.10
0.04
Housing
0.09
0.05
Electricity
0.08
0.08
Child care
0.05
0.01
Health
0.05
0.02
credit/employment/women's group
0.04
0.08
0.03
0.17
Education
0.02
0.14
Other
0.04
0.06
Irrigation
and ponds
Notes:
(1) Each cell is the number of times an issues was mentionned, divided
by the total number of issues mentionned.
(2)
The data
for
men comes from
a subsample of 48 villages
Table
6: Effect of
Outcome
Women's Reservation on
Organizations
Mean, reserved
Mean, unreserved
GP
GP
in
GP
Vne
Difference
no control
controlling
for population
A.
(1)
(2)
0.05
0.03
0.01
(.02)
(.01)
(.02)
(3)
VILLAGE LEVEL
Women organisations,
self help
groups
created since 1998
Number of tubewells management
0.04
0.03
0.01
comittees created since 1998
(.02)
(.01)
(.02)
Village education and Village attendance
l.IO
1.04
0.06
committees created since 1998
(.07)
(.05)
(.09)
-0.11
Number of health
information campaigns
2.69
2.81
(.27)
(.19)
(.33)
0.64
0.70
-0.06
(.05)
(.04)
(.06)
0.26
0.34
-0.08
(.06)
(.05)
(.08)
0.72
0.79
-0.06
committee or a drinking water committee
(.06)
(.04)
(.07)
Additional acreage brought under
0.93
0.60
0.33
and immunization campaigns since 1998
All the previous 5 planned
were
B.
(4)
Gram Samsad
held.
GP LEVEL
Women
organisations, self help groups
and micro-credit groups created since
Does
the
GP has a
forest
1
998
management
social forestries or fisheries since
1998
(.53)
(.21)
(.48)
Village education and Village attendance
10.17
10.57
-0.40
committees created since 1998
(.83)
(.74)
(1.19)
Number of health
5.43
6.48
-1.05
information campaigns
and immunization campaigns since 1998
Total
Own
fund collected
in
1
998
1000)
(rs.
Taxes
Other revenues
(1.01)
(.80)
(1.34)
40.63
59.11
-18.48
-17.79
(6.28)
(6,06)
(9.58)
(9.79)
25.83
34.34
-8.51
-7.63
(2.84)
(3.26)
(5.01)
(5.12)
9.92
26.53
-16.61
-16.36
(2.28)
(4.54)
(6.59)
(6.76)
Notes:
( 1
Standard errors
(2)
There are 322 observations
There are only
1
in
parentheses
in the village level regressions,
00 observations
in the last three
(3) Standard errors are corrected for clustering at the
Moulton (1986) formula
and
1
61 in the
GP
level regressions.
rows
GP
level in the village level regressions, using
Table
7; Effect of
Women's Reservation on
Public
Goods Investments
Dependent
Mean, reserved
Mean, unreserved
variable
GP
GP
(1)
(1)
(2)
(3)
Number of drinking water facilities
23.83
14.74
9.09
newly
(5.00)
(1.44)
(4.02)
3.31
3.67
-0.36
(.81)
(.81)
(1.27)
0.41
0.23
0.18
(.05)
(.03)
(.06)
0.14
0.42
-0.28
(.08)
(.13)
(.19)
0.26
0.06
0.20
(.11)
(.02)
(.08)
0.06
0.12
-0.06
(.02)
(.03)
(.04)
0.10
0.01
0.09
(.04)
(.01)
(.03)
0.59
0.51
0.07
(.10)
(.05)
(.10)
1.00
0.93
0.07
(.02)
(.03)
A.
VILLAGE LEVEL
built or repaired
Number
newly
of irrigation
facilities
built or repaired
Condition of roads
(1 if in
good
condition)
Number of latrines and
newly
drainage pits
built or repaired
Number
of new biogas
Number
of informal education centers
facilities
Nuber of of adult education
newly
newly
of formal school buildings
built or repaired
GP LEVEL
new
tubewell
1
if a
1
if at least
1
1
if a
0.17
0.09
0.07
(.05)
(.03)
(.05)
0.67
0.48
0.19
(.06)
(.05)
(.08)
1.00
0.85
0.15
built or repaired
()
(.03)
(.05)
0.50
0.57
-0.07
were constructed
(.07)
(.05)
(.08)
new
0.50
0.45
0.05
(.07)
(.05)
(.08)
0.67
0.82
-0.16
(.06)
new
drainage
biogas equipment was introduced
if
1
if there is
an informal education
center in the
if there
is
if there is a
or if there
GP
(.04)
(.07)
0.74
0.83
-0.08
(.07)
(.04)
(.07)
0.91
0.87
0.04
(.04)
(.03)
(.05)
an informal education
center in the
1
pump was built
minor
was
if new toilets or
1
1
built
metal road was built or repaired
if a culvert or a
pits
was
one irrigation
irrigation canal
1
centers
built or repaired
Number
B.
Difference
GP
(literate
pradhans)
continuing education center
was
a literacy
campaign
Notes:
(1) Standard errors in parentheses
(2)
There are 322 observations
in the village level regressions,
There are only 100 observations
in the last
(3) Standard errors are corrected for clustering at the
Moulton (1986) formula
and 161
in the
GP
level regressions,
row
GP
level in the village level regressions, using
Table
8: Effect of
Women's Reservation on Formal Education and
Health
Outcomes
Mean, reserved
Mean, unreserved
GP
GP
(1)
(2)
(3)
(5)
160
Number of pre-school
Difference
obs.
0.18
0.54
-0.36
(.05)
(.37)
(.52)
0.18
0.54
-0.36
(.43)
(.3)
(.46)
Pradhan reports that the Village education Conjmittee
0.80
0.89
-0.09
has influence over the primary school (GP)
(.06)
(.03)
(.06)
centers
opened since
1
998 (GP)
Number of pre-school
centers
opened since
1
998 (Village
level)
Villagers report that the Panchayat has influence over
0.37
0.32
0.05
the running of the primary school (Village)
(.05)
(.04)
(.07)
0.44
0.65
-0.21
(.07)
(.05)
(.08)
0.47
0.66
-0.19
(.08)
(.05)
(.09)
0.31
0.34
-0.03
(.05)
(.04)
(.06)
Pradhan reports that low teacher attendance
is
a
is
a
problem. (GP)
Pradhan reports that low teacher attendance
problem. (GP,
litterate
Villagers report that
pradhans only)
low teacher attendance
is
a
problem
(village)
Number of health workers
in the last
Number of health workers
in the last 6
months
12.11
9.00
3.11
(1.44)
(.95)
(1.67)
15.03
11.01
4.02
(1.55)
(1.09)
(1.63)
or doctor visits in the villages
6 months
or doctor visits in the villages
(for village with at least
1
visit).
Notes:
(1)
Standard errors in parentheses
(2) Standard errors are corrected for clustering at the
GP
level in the village level regressions, using
Moulton (1986) formula
(3) All equations are estimated jointly in the
for interequation correlation.
GP
level regressions
and the standard
errors account
Number of
252
161
322
161
148
322
322
262
Table
9: Effect
of
Women's Reservation on
All
GP
GP
reserved for
Pradhan's village
the Location of Public
woman
Goods
where
Pradhan's village
Irrigation
Road
Sanitation
Biogas
Informal education
Adult education
Formal education
helped
woman
(4)
(3)
(2)
0)
Drinking water
is
reserved for
*pradhan's village
*pradhan's village
Dependent variable
woman
38.9
-33
37
-51
(12)
(16)
(11)
(22)
0.52
3.05
0.27
-2.71
(1.5)
(2.4)
(1.6)
(2.2)
0.19
-0.13
0.19
-0.1
(0.058)
(0.098)
(0.058)
(0.150)
1.69
-1.56
1.67
-1.23
(0.88)
(0.94)
(0.86)
(1.04)
0.42
-0.081
0.45
-0.11
(0.15)
(0.28)
(0.15)
(0.40)
0.029
0.02
0.012
-0.0076
(0.085)
(0.061)
(0.062)
(0.120)
0.31
-0.23
0.32
-0.26
(0.080)
(0.110)
(0.081)
(0.110)
0.47
-0.5
0.47
-0.66
(0.12)
(0.18)
(0.12)
(0.22)
All invesments
42.6
-32.2
40
-55.9
cumulated
(12)
(17)
(11)
(23)
All invesments
21.7
-21.8
22.2
-33
GP
(6.7)
(11.4)
(6.7)
(18.8)
cumulated,
reserved for
SC/ST
Number of observations
477
384
Notes:
(1)
each row gives the result from
a separate regression
of the dependent variable on
dummy for woman reservation, a dummy for pradhan's village and the interaction between
village and woman reservation dummies. Only the last two coefficients are reported.
gp population, a
pradhan's
(2) standard errors in parentheses
(3) Standard errors are corrected for clustering at the
GP
level
(4) Variables in this table are defined as in table 8
(5)
There are 218 observations for the
last line,
column
(1)
and
(2),
and 170
for the last line,
column
(3)
and
(4)
Table 10: Effect
of
Women's Reservation
in
Selected Sub-Samples
Difference between
women
All
GP
reserved for
GP will
Previous pradhan's
constituency reserved
(1)
GP
and unreserved
in
GP
be reserved
2003
GP
is
for
(4)
2)
{2}
reserved
SC/ST
PANEL A; PRADHAN'S BACKGROUND AND EXPERIENCE
-2.58
-2.65
-3.31
Pradhan's education
-2.79
(.54)
(.68)
(.61)
(.86)
Number of assets
-0.64
-0.70
-0.60
-0.37
(.23)
(.26)
(.26)
(.27)
-554
-482
-357
14
(291)
(312)
(349)
(381)
-0.31
-0.14
Population of Pradhan's village
GP
-0.32
-0.24
before 1998
(.07)
(.08)'
(.08)
(.09)
Elected as Pradhan before 1998
-0.12
0.00
-0.08
-0.02
(.04)
(.)
(.04)
(.03)
0.13
0.14
0.13
0.16
(.07)
(.09)
(.09)
(1)
Elected in
council
Will not run again
PANEL B: WOMEN'S PARTICIPATION
Have women addressed a complaint to
the GP in the last 6 months
PANEL C: PUBLIC GOODS
Revenue collection in the GP
(controlling for GP population)
0.09
0.10
0.11
0.10
(.05)
(.06)
(.06)
(.06)
-18
-14
-7
-11
(10)
(11)
(10)
(14)
Village level:
Number of drinking
newly
water
Condition of roads
9.09
8.44
10.14
10.59
(4.02)
(5.5)
(5.25)
(6.01)
facilities
built or repaired
(1 if in
good
condition)
Number of new
biogas facilities
Nuber of of adult education
newly
centers
built or repaired
0.18
0.21
0.21
0.25
(.06)
(.07)
(.06)
(.08)
0.20
0.17
0.21
0.22
(.08)
(.12)
(.12)
(.15)
0.09
0.09
0.08
0.09
(.03)
(.04)
(.04)
(.04)
GP level:
1
1
1
if a
if a
new tubewell was
built in the
metal road was built or repaired
in the
0.08
0.06
0.09
(.04)
(.03)
(.05)
0.19
0.11
0.17
0.22
(.08)
(.09)
(.1)
(.12)
-0.16
-0.14
-0.13
-0.14
(.07)
(.09)
(.09)
(.11)
-0.21
-0.19
-0.22
-0.24
(.11)
GP
an informal education
if there is
0.07
(.03)
GP
GP
MONITORING
center in the
PANEL C;
Pradhan reports
that
low teacher attendance
is
a
(.08)
(1)
(1)
3.11
2.34
3.40
0.87
(1.67)
(2.14)
(1.82)
(2.48)
problem. (GP)
Number of health workers
in the last 6
months
or doctor visits
(for village with at least
1
visit).
Note
mean of the dependent variable in GPs
was prevented from re-election due to a reservation of his boot
Pradhan
previous
reserved for women and GP where the
51
GP (102 villages) where the previous pradhan's boot is reserved
and
women,
There are 55 GP (1 10 villages) reserved for
(1)
Column
2 presents the difference between the
(2)
Column
3 presents the difference
reserved for
women
There are 55
GP
(3)
Colum 4
( 1
!
and
GP that will
between the mean of the dependent vanable
be reserved for
villages) reserved for
women
presents the difference between the
woman
in 1998,
mean
woman SC/ST and GP reserved for a SC/ST
There are 78 GP (146 villages) reserved for SC and
in
in
GP
2003
and 52
GP
(146 villages) that will be reserved
of the dependent variable
in
2003
m GP reserved
for a
women as well
GP level in the village
ST, including 28 reserved for
corrected for correlation at the
(4) Standard errors are in parentheses, and are
level regressions
Date Due
3/2^/ o3
Lib-26-67
MIT LIBflARIES
3 9080 02240 4831