Gender Diversity and Environmental Performance: A Quantitative Assessment

By Miriam B. Ott

Introduction

Scientists have been ringing alarm bells since the mid-20th century, when scientific evidence

started pointing towards a multitude of environmental problems caused by humans such as

groundwater pollution from hazardous wastes, the hole in the ozone layer, climate change and

the accelerating decrease in biodiversity.1 Ever since, environmental degradation has been on

the political agenda of many countries all around the globe as well as the subject of international

cooperation. As the 2010 Environmental Performance Index (EPI) demonstrates, countries vary

to a great extent in their environmental performance.2 Many scholars have attempted to explain

this variation and have found evidence for a variety of explanatory variables. According to Esty

and Porter, for example, environmental performance varies with income levels, the sophistication

of a country’s regulatory regime and its broader economic and social context.3 Other authors

link environmental performance to ethnic diversity.4 Specifically, Das and DiRienzo found an

inverted U-shaped relationship between the two variables.5 Additionally, the authors of the EPI

themselves find GDP per capita, corruption and government effectiveness to be drivers of envi-

1
Helmut Breitmeier, "International Organizations and the Creation of Environmental Regimes," in Global Govern-
ance. Drawing Insights from the Environmental Experience, ed. Oran R. Young (Cambridge: The MIT Press, 1997),
96; Daniel Somers Smith, "Place-Based Environmentalism and Global Warming: Conceptual Contradictions of
American Environmentalism," Ethics & International Affairs 15, no. 2 (2001): 118.
2
Jay Emerson et al., "2010 Environmental Performance Index," Accessed November 8, 2010,
http://epi.yale.edu/file_columns/0000/0157/epi2010_report.pdf.
3
Daniel C. Esty and Michael E. Porter, "National environmental performance: an empirical analysis of policy re-
sults and determinants," Environment and Development Economics 10(2005): 395.
4
Julio Videras and Christopher J. Bordoni, "Ethnic Heterogeneity and the Enforcement of Environmental Regula-
tion," Review of Social Economy 64, no. 4 (2006).
5
Jayoti Das and Cassandra E. DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analy-
sis," The Journal of Environment & Development 19, no. 1 (2010): 105.
 Journal of Political Inquiry 4 (2011)

ronmental performance.6 This paper deals with another more marginalized approach, which

links environmental performance to gender diversity.

Since the 1970s, gender theorists have argued that women’s participation in environmen-

tal policymaking is essential. However, the reasoning behind this claim varies. Eco-feminists

argue that women are inherently closer to nature than men.7 Others argue that, due to different

socially constructed societal roles and functions, men and women alike are important but distinct

agents of change in the processes of mitigation of and adaptation to global and local environmen-

tal change.8 The latter logic would imply that gender diversity, meaning a balance of women and

men as decision-makers in environmental politics, would enhance environmental performance.

Unfortunately, environmental politics, just like other fields of politics, is a male-dominated field.

For example, among the country delegates to the United Nations Framework Convention on

Climate Change (UNFCCC) in Copenhagen in 2009, only 27.7 percent were women.9

While such arguments have been around for several decades, investigation in this area has

been limited to case studies.10 There has been no attempt to estimate the degree to which in-

creased participation of women in environmental policymaking affects a country’s environmen-

6
Emerson et al., "2010 Environmental Performance Index," 34.
7
Irene Dankelman and Willy Jansen, "Gender, Environment and Climate Change: Understanding the Linkages," in
Gender and Climate Change: An Introduction, ed. Irene Dankelman (London: earthscan, 2010), 22-23.
8
Irene Dankelman, "Introduction: Exploring Gender, Environment, and Climate Change," in Gender and Climate
Change: An Introduction, ed. Irene Dankelman (London: earthscan, 2010), 2.
9
UNFCCC, "FCCC/CP/2009/INF.1 (Part 1). Conference of the Parties. Fifteenth Session. Copenhagen, 8-18 De-
cember 2009. List of Participants," Accessed November 20, 2010,
http://unfccc.int/resource/docs/2009/cop15/eng/inf01p01.pdf.
10
For examples, see: Thais Corral, "Gender Perspectives in Adaptation Strategies: The Case of Pintadas Solar in the
Semi-arid Region of Brazil," in Gender and Climate Change: An Introduction, ed. Irene Dankelman (London:
earthscan, 2010); Alice Fothergill, "The Neglect of Gender in Disaster Work: An Overview of the Literature," in
The Gendered Terrain of Disaster: Through Women’s Eyes, ed. Elaine Enarson and Betty Hearn Morrow (Westport:
Praeger, 1998); William E. Lovekamp, "Gender and Disaster: A synthesis of flood research in Bangladesh," in
Women and Disasters. From Theory to Practice, ed. Brenda D. Phillips and Betty Hearn Morrow (Philadelphia:
Xlibris, 2008); Koos Neefies and Valerie Nelson, "Responding to Climate Change in Vietnam: Opportunities for
Improving Gender Equality," in Gender and Climate Change: An Introduction, ed. Irene Dankelman (London:
earthscan, 2010); Sibyl Nelson and Yianna Lambrou, "Gender Dimensions, Climate Change and Food Security of
Farmers in Andhra Pradesh, India," in Gender and Climate Change: An Introduction, ed. Irene Dankelman (London:
earthscan, 2010); Omoyemen Odigie-Emmanuel, "The Gender Impact of Climate Change in Nigeria," in Gender
and Climate Change: An Introduction, ed. Irene Dankelman (London: earthscan, 2010).

2
 Journal of Political Inquiry 4 (2011)

tal performance. The lack of quantitative studies in this area is partially due to the fact that many

gender theorists do not believe quantitative analysis can adequately capture the most important

elements of gender dynamics. Instead, quantitative analysis is viewed as objectifying and reduc-

ing humans to social facts and leads to a univariate, atomized analysis.11

This paper aims to demonstrate that quantitative analysis can strengthen gender theorists’

claims without obscuring basic feminist beliefs. As Majorie L. DeVault argues, the equation of

feminist methodology with qualitative methods is a misconception. Instead, feminist methodol-

ogy seeks to shift attention away from the primary focus on men’s concerns and to ‘bring women

in.’12 While this has mainly been done through qualitative methods such as interviews and narra-

tive analysis, this focus on qualitative methods is not stringently derived from feminist beliefs.

Instead, it needs to be recognized that quantitative techniques can contribute to ‘bringing women

in,’ and can be more compelling than personal testimony or case studies, which are vulnerable

towards criticism regarding selection bias. In other words, what is important is not whether a

study is qualitative or quantitative, but whether a study is committed to women and their con-

cerns.13

The goal of this paper is to quantitatively assess the empirical validity of the claim that

gender diversity is an important determinant of environmental performance. My research ques-

tion is: Does a causal link exist between increased participation of women in environmental poli-

tics and a country’s environmental performance? If the answer is yes, how strong is the effect?

To answer this question, I will estimate five Ordinary Least Squares (OLS) regression models

11
Elaine Enarson and Brenda Phillips, "Invitation to a New Feminist Disaster Sociology: Integrating Feminist Theo-
ry and Methods," in Women and Disasters. From Theory to Practice, ed. Brenda D. Phillips and Betty Hearn Mor-
row (Philadelphia: Xlibris, 2008), 60.
12
Majorie L. DeVault, Liberating Method. Feminism and Social Research (Philadelphia: Temple University Press,
1999), 30.
13
Ibid., 30.

3
 Journal of Political Inquiry 4 (2011)

testing the validity, robustness, and causality of the hypothesized positive relationship between

women’s participation in environmental politics and a country’s environmental performance.

The paper proceeds as follows: First, I will offer a brief literature review and derive a

testable hypothesis. The following section will describe the operational capacity of the variables

used. Then, I will offer some descriptive statistics and present and interpret the empirical evi-

dence. This will be followed by a conclusion.

1. Theoretical Background

Radical feminist theory is based on the notion that men and women are essentially different and

that gender inequality is rooted in a universal system through which men seek power and control

over women - and over the environment.14 Eco-feminists, whose beliefs are grounded in radical

feminist theory,15 argue that women are inherently closer to nature than men. Male domination

of nature is viewed as equivalent to male domination of women, which has led to a marginaliza-

tion of women’s knowledge in general and women’s knowledge of nature in particular. There-

fore, eco-feminists argue that male dominated “mal-development,” which is often assigned to

“white” men, has caused major social and environmental problems.16 This essentialist stance has

been criticized for its failure to address power and economic differences, to differentiate women

themselves by class, ethnicity, and caste and finally, to recognize that the concept of gender var-

ies across cultures and regions.17

Taking these criticisms into account, liberal feminist theory posits that gender differences

are not based on biological difference. Instead, they are thought to be socially constructed and

14
Enarson and Phillips, "Invitation to a New Feminist Disaster Sociology: Integrating Feminist Theory and Meth-
ods," 50.
15
Ibid., 50.
16
Dankelman and Jansen, "Gender, Environment and Climate Change: Understanding the Linkages," 22-23.
17
Ibid., 23.

4
 Journal of Political Inquiry 4 (2011)

reproduced by a gendered structure and practice of social institutions, which ensure a lifelong

socialization into gendered cultures. The resulting socially constructed differences disadvantage

women in a socially constructed world that privileges ‘male’ characteristics.18 This goes hand in

hand with an evolution of the definition of gender. Instead of thinking of gender differences as

stemming from inherent characteristics of women and men, many gender theorists now think of

‘gender’ as the manifestation of the context-specific relationships between women and men, and

define it as the socially acquired notions of masculinity and femininity by which women and

men are identified.19

Following liberal feminism’s definition of gender, this paper links gender diversity in en-

vironmental politics to environmental performance in the following ways—first, environmental

degradation has gender-differentiated impacts, especially in the developing world, which gives

women a distinct perspective. This is due to the fact that the asymmetry of power between

women and men is a pervasive trait all over the world, although to varying extents. In fact, gen-

der is a social stratifier, similar to other stratifiers such as class, ethnicity, religion, place and

age.20 For instance, women make up 70 percent of the 1.3 billion people living in extreme pov-

erty.21 Due to gender inequality, women tend to have less access to resources that are essential

for adapting to environmental degradation.22 It has been shown that women are more vulnerable

to natural disaster than men because they are over-represented in agricultural and informal sec-

tors, lack access of energy sources, clean water, safe sanitation and health care, and are thus dis-

18
Enarson and Phillips, "Invitation to a New Feminist Disaster Sociology: Integrating Feminist Theory and Meth-
ods," 47.
19
Janet Henshall Momsen, Gender and Development (London: Routledge, 2004), 2.
20
Dankelman, "Introduction: Exploring Gender, Environment, and Climate Change," 11.
21
UNDP, Human Development Report 1995. Gender and Human Development (New York: Oxford University
Press, 1995), 4.
22
Dankelman, "Climate Change, Human Security and Gender," 59.

5
 Journal of Political Inquiry 4 (2011)

advantaged in disaster preparedness, mitigation, and rehabilitation.23 While this link is stronger

in developing countries, the same pattern can be found in developed countries. Studies have

shown that Hurricane Katrina, which hit the United States Gulf Coast in 2005, affected women

more than men due to women’s distinct roles within the family, society and the labor market. As

a result, women were often left poorer, less able to find jobs and therefore unable to move back

to their previous homes.24 Due to the gender-differentiated impacts of environmental degrada-

tion, integrating women’s perspectives into environmental policymaking is an essential prerequi-

site to finding adequate responses that protect both men and women.

Second, women’s perspectives on environmental degradation can complement men’s per-

spectives in a variety of ways. Due to their different upbringing, different socially acquired

roles, norms, values and ways of thinking, an increase in the participation of women in environ-

mental politics may be accompanied by the introduction of new and innovative ideas concerning

how to deal with environmental degradation. Jayoti Das and Cassandra E. DiRienzo have made

a similar argument when explaining the inverted U-shaped relationship between ethnic diversity

and environmental performance. They suggest a moderate level of ethnic diversity offers an ad-

vantage to countries in finding solutions to environmental problems, since ethnic diversity leads

23
For a more elaborate discussion of the causes of women's increased vulnerability to natural disasters, see: Robert
Bolin, Martina Jackson, and Allison Crist, "Gender Inequality, Vulnerability, and Disasters: Issues in Theory and
Research," in The Gendered Terrain of Disaster: Through Women’s Eyes, ed. Elaine Enarson and Betty Hearn Mor-
row (Westport: Praeger, 1998); Dankelman, "Climate Change, Human Security and Gender."; Enarson, Fothergill,
and Peek, "Gender and Disaster: Foundations and Directions."; Fordham, "Gendering vulnerability analysis: To-
wards a More Nuanced Approach."; Alice Fothergill, "The Neglect of Gender in Disaster Work: An Overview of the
Literature," in The Gendered Terrain of Disaster: Through Women’s Eyes, ed. Elaine Enarson and Betty Hearn
Morrow (Westport: Praeger, 1998); Eric Neumayer and Thomas Plümper, "The Gendered Nature of Natural Disas-
ters: The Impact of Catastrophic Events on the Gender Gap in Life Expectancy, 1981-2002," Annals of the Associa-
tion of American Geographers 97, no. 3 (2007); Jennifer Wilson, Brenda D. Phillips, and David M. Neal, "Domestic
Violence after Disaster," in The Gendered Terrain of Disaster: Through Women’s Eyes, ed. Elaine Enarson and Bet-
ty Hearn Morrow (Westport: Praeger, 1998).
24
Rachel Harris, "Gender Aspects of Climate Change in the US Gulf Coast Region," in Gender and Climate
Change: An Introduction, ed. Irene Dankelman (London: earthscan, 2010).

6
 Journal of Political Inquiry 4 (2011)

to more creative and innovative solutions.25 If the same is true for gender diversity, the increased

participation of women in environmental decision-making should improve environmental per-

formance.

Third, increased participation of women in environmental politics would enhance envi-

ronmental performance by introducing previously neglected capacities.26 In many countries,

women are now more likely to graduate high school and college and have better grades than

men. In Europe, for example, women account for 55 percent of university graduates.27 The no-

tion that women’s excellent performance in education translates into professional performance

has been put forward by a study of McKinsey & Company, which shows that companies with

three or more women in senior management positions tend to score more highly on indicators of

organizational excellence and financial performance than companies with no women at the top.28

Since mitigation of and adaptation to environmental degradation require large amounts of re-

sources and capacities, neglecting women’s capacity would be a big mistake. Especially in the

fields of promoting sustainable lifestyles, reforestation, education and raising awareness, wom-

en’s rich contributions and potentials have been particularly valuable for environmental conser-

vation and management.29

On this basis, I theorize that the positive effect of gender diversity and the participation of

women, as found by McKinsey & Company, will persist at the aggregate country level and trans-

late into improved environmental performance. This leads me to the hypothesis that there is a

positive relationship between women’s participation in environmental politics and a country’s

25
Das and DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analysis," 93.
26
Dankelman, "Introduction: Exploring Gender, Environment, and Climate Change," 11.
27
McKinsey&Company, "Women Matter. Gender diversity, a corporate performance driver," Accessed October 30,
2010, http://www.mckinsey.com/locations/swiss/news_publications/pdf/women_matter_english.pdf.
28
Ibid., 12-13.
29
Dankelman and Jansen, "Gender, Environment and Climate Change: Understanding the Linkages," 47-48.

7
 Journal of Political Inquiry 4 (2011)

environmental performance. In the following section, the operational capacity of the variables

will be laid out. This will be followed by descriptive statistics, regression analyses, and interpre-

tation of findings.

2. Empirical Analysis

2.1 Operational Capacity of Environmental Performance

As a dependent variable, I will use the 2010 EPI, which is available for 163 countries.30 The EPI

uses a target-to-proximity methodology to track a country’s success in meeting ten environmen-

tal policy goals, which are either based on international treaties and agreements or derived from

environmental and public health standards developed by international organizations and national

governments. The EPI is a weighted average of these environmental policy goals and ranges

from 0 to 100, with higher values indicating greater success at meeting environmental targets.31

2.2 Operational Capacity of W omen’s Participation in Environmental Politics

The main independent variable is women’s participation in environmental politics, which will be

illustrated as the percentage of women in country delegations to the 15th Conference of the Par-

ties, (COP-15) to the UNFCCC. This operational capacity assumes that the percentage of wom-

en in a country’s delegation to United Nations environmental conferences reflects the gender dis-

tribution in other bodies of environmental policymaking. The data was obtained from the List of

Participants published by the UNFCCC Secretariat. This list contains the names of all members

30
Emerson et al., "2010 Environmental Performance Index," 13.
31
Weights: Environmental Burden of Disease (25%), Water (effects on humans) (12.5%), Air Pollution (effects on
humans) (12.5%), Air Pollution (effects on ecosystems) (4.2%), Water (effects on ecosystems) (4.2%), Biodiversity
& Habitat (4.2%), Forestry (4.2%), Fisheries (4.2%), Agriculture (4.2%), Climate Change (25%) ibid., 15..

8
 Journal of Political Inquiry 4 (2011)

of each country delegation.32 It is easy to recognize which delegates are women and which are

men since the appropriate gender-indicative title precedes every name that is listed. For every

country, I counted the absolute number of delegates and the number of women delegates and

created a variable, which indicates the percentage of women on each country delegation. This

data is available for 194 countries and ranges from 0, (Eritrea, Honduras, Jordan, North Korea,

Libya, Mauritius, Myanmar, Qatar, Saudi Arabia, Somalia, and Yemen) to 83.3, (Antigua and

Barbuda). A full list of the percentages of women in each country delegation can be found in the

Annex.

Additionally, I created a variable that measures the percentage points by which the per-

centage of women on a country’s delegation deviates from the “ideal” level of 50 percent in ei-

ther direction. The inclusion of this variable reflects the idea that gender parity rather than a

larger amount of women is the critical variable.

To test the robustness of my results, I performed additional regression analyses using two

proxies of women’s participation in environmental politics: first, I use a more indirect measure-

ment, namely data on gender equality, which is taken from the International Development Asso-

ciation’s 2005 Resource Allocation Index. This variable assesses the extent to which the country

has enacted and put in place institutions and programs to enforce laws and policies that promote

equal access to human capital development, promote equal access to productive and economic

resources and give men and women equal status and protection under the law.33 The variable

ranges from 2 to 4.5 with higher values indicating higher levels of equality. Second, I use the

32
UNFCCC, "FCCC/CP/2009/INF.1 (Part 1). Conference of the Parties. Fifteenth Session. Copenhagen, 8-18 De-
cember 2009. List of Participants."
33
Jan Teorell et al., "The Quality of Government Dataset, version 27May10. University of Gothenburg: The Quality
of Government Institute," Accessed November 11th, 2010, http://www.qog.pol.gu.se; Emerson et al., "2010 Envi-
ronmental Performance Index" 49; International Development Association, "IDA Resource Allocation Index
(IRAI)," Accessed November 24, 2010, http://go.worldbank.org/FHNU4A23U0.

9
 Journal of Political Inquiry 4 (2011)

Gender Gap Index, which uses a scale from 0 to 1, with 1 representing a maximum level of gen-

der equality. The index measures the degree to which women have achieved equality in the are-

as of economic participation, economic opportunity, political empowerment, educational attain-

ment and health and well being.34

To test whether the link between women’s participation in environmental politics and im-

proved environmental performance is a causal link or whether the two variables are merely cor-

related, since both are associated with higher levels of development, I use the year in which

women gained the right to vote as an instrumental variable. This variable is obtained from the

Organisation for Economic Cooperation and Development (OECD) Gender, Institutions, and

Development Database.35 Instrumental variables are a commonly used solution to the problem

of an endogenous independent variable such as women’s participation in environmental politics.

An instrumental variable can establish a causal relationship if it is correlated with the independ-

ent variable and it can reasonably be expected to only impact the dependent variable through the

independent variable, meaning that it is uncorrelated with the error term.36 The first criterion is

easy to establish; the year in which women received the right to vote is negatively correlated

with the percentage of women in a countries’ delegation to the COP-15 (r=-0.27). Thus, the ear-

lier women received the right to vote, the more women are involved in environmental politics

today. The second criterion is trickier. It seems reasonable to assume that in countries where

women have been able to vote and have been elected into office for a longer period of time,

women’s participation in environmental politics is better established and norms of gender equali-

34
Jan Teorell et al., "The Quality of Government Dataset, version 27May10. University of Gothenburg: The Quality
of Government Institute," Accessed November 11, 2010, http://www.qog.pol.gu.se; World Economic Forum, "The
Global Gender Gap Report 2005," Accessed November 24, 2010, http://www.weforum.org/gendergap.
35
OECD, "The Gender, Institutions and Development Data Base," Accessed November 20, 2010,
http://www.oecd.org/document/0/0,3343,en_2649_33731_39323280_1_1_1_1,00.html; Teorell et al., "The Quality
of Government Dataset, version 27May10. University of Gothenburg: The Quality of Government Institute."
36
Jeffrey M. Wooldridge, Econometric Analysis of Cross Section and Panel Data (Cambridge: The MIT Press,
2002), 83-84.

10
 Journal of Political Inquiry 4 (2011)

ty are more ingrained in the system. While it is impossible to prove that women’s participation

in environmental politics is the only path through which the earlier extension of franchise influ-

ences environmental performance, it also seems to be a reasonable assumption. Data is available

for 153 countries; among these countries, women’s right to vote was first introduced in 1893

(New Zealand) and last in 2005 (Kuwait).

2.3 Control V ariables

When considering environmental performance as a dependent variable, other variables, which

have been identified to be drivers of environmental performance in prior studies, need to be con-

trolled for. As noted above, these are factors such as income levels, the sophistication of a coun-

try’s regulatory regime, the broader economic and social context37 and ethnic diversity.38 Addi-

tionally, Das and DiRienzo, suggest economic freedom, corruption, democracy, government ef-

ficiency and economic development as additional control variables.39 However, including too

many control variables exacerbates the small-N problem. Therefore, I will limit the amount of

control variables used in this study to the three main drivers of environmental performance found

by the authors of the EPI, namely GDP per capita (measured by its logarithm), corruption and

government effectiveness.40 Additionally, I will control for the level of democracy.

Economic Wealth

According to prior studies, it can be expected that more economically developed countries, as

measured by their per capita GDP, tend to be more successful in meeting environmental tar-

37
Esty and Porter, "National environmental performance: an empirical analysis of policy results and determinants,"
395.
38
Das and DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analysis," 105.
39
Ibid., 99-102.
40
Emerson et al., "2010 Environmental Performance Index," 34.

11
 Journal of Political Inquiry 4 (2011)

gets.41 Data will be taken from the United Nations Statistics Division, which published GDP per

capita at constant 1990 prices in US dollars in 2002 for 192 countries.42

Corruption

Based on previous studies, higher corruption levels can be expected to negatively affect a coun-

try’s environmental performance. Therefore, this study uses the Corruption Perception Index

published by Transparency International for 180 countries.43 This index measures the degree to

which politicians and officials are believed to accept illicit payments or bribes in public pro-

curement, embezzle public funds, or commit offenses. It is measured on a scale ranging from 1

to 10, with 10 indicating the lowest level of corruption.44

Government Effectiveness

The measure of government effectiveness is taken from the Worldwide Governance Indicators

project dataset and combines responses on the quality of public service provision, the quality of

the bureaucracy, the competence of civil servants, the independence of the civil service from po-

litical pressures, and the credibility of the government’s commitment to policies in 194 coun-

tries.45 Higher values indicate higher levels of government effectiveness.

41
Das and DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analysis," 102.
42
Teorell et al., "The Quality of Government Dataset, version 27May10. University of Gothenburg: The Quality of
Government Institute."; United Nations Statistics Division, "National Accounts Main Aggregates Database," Ac-
cessed November 13, 2010, http://unstats.un.org/unsd/snaama/.
43
Das and DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analysis."; Teorell et al.,
"The Quality of Government Dataset, version 27May10. University of Gothenburg: The Quality of Government
Institute."
44
Das and DiRienzo, "Is Ethnic Diversity Good for the Environment? A Cross-Country Analysis," 100.
45
Emerson et al., "2010 Environmental Performance Index," 69; Daniel Kaufmann, Aart Kraay, and Massimo
Mastruzzi, "The Worldwide Governance Indicators (WGI) project," Accessed November 19, 2010,
http://info.worldbank.org/governance/wgi/index.asp.

12
 Journal of Political Inquiry 4 (2011)

Democracy

The level of democracy is measured by the un-weighted average of political rights and civil lib-

erties as published by Freedom House.46 I use the data for the years 2002-2006, which is availa-

ble for 194 countries and which I recoded in such a way that 1 is the lowest possible degree of

democracy and 7 is the highest.

3. Descriptive Statistics

Table 1 below provides a summary of the data used in this analysis as well as the descriptive sta-

tistics for each variable.

Table 1: Descriptive Statistics

Variable Name N Mean Std. Dev. Min Max
Environmental Performance Index 163 58.37 12.41 32.1 93.5
Percentage of Women Delegates 194 27.21 15.11 0 83.3
Divergence from Gender Parity 194 23.88 13.31 0 50
Gender Equality 76 3.47 0.65 2 4.5
Gender Gap Index 128 0.67 0.06 0.45 0.81
Year of Extension of Franchise 153 1949.10 20.18 1893 2005
GDP per capita 192 6309.03 10711.46 97.77 72924.18
Corruption Perception Index 180 3.99 2.08 1.2 9.7
Government Effectiveness 194 -0.60 0.99 -1.93 2.25
Level of Democracy 194 4.63 1.95 1 7

3.1 Regressions A nalyses

Table 2 presents the results of five regression analyses testing my hypothesis. Column (1) sug-

gests that the higher the percentage of women in a countries’ delegation to the COP-15, the high-

er its environmental performance. This supports my hypothesis, which predicts a positive rela-

tionship between women’s participation in environmental politics and a country’s environmental

performance. Specifically, my results suggest that for every percentage point, by which the par-

ticipation of women in environmental politics increases, environmental performance improves

by 0.163 on a scale from 0 to 100. As indicated in Table 1, the average percentage of women on

46
Teorell et al., "The Quality of Government Dataset, version 27May10. University of Gothenburg: The Quality of
Government Institute."

13
 Journal of Political Inquiry 4 (2011)

country delegations to the COP-15 was 27.21 percent. Thus, if women’s participation was in-

creased to constitute an average of 50 percent, the model would predict an increase in global en-

vironmental performance by (50-27.31)*0.163 ≈ 3.7 points on a scale from 0 to 100. While this

increase may not be exceptionally large, it is significant. It strengthens claims for the necessity

of increased participation of women delegates and decision-makers in environmental politics.

Table 2: Regression Analysis - Percentage of Women in Country Delegations to the COP-15

(1) (2) (3) (4) (5)
EPI EPI EPI EPI EPI
Percentage of Women Delegates 0.163*
(0.02)

Divergence from Gender Parity -0.166*

(0.02)

Gender Equality 6.514***

(0.00)

Gender Gap Index 65.704**

(0.00)

Year of Extension of Franchise -0.155***

(0.00)

Government Effectiveness 3.567 3.596 2.293 -0.233 2.747

(0.08) (0.08) (0.43) (0.93) (0.22)

Level of Democracy 0.752 0.753 -0.460 1.132 0.531

(0.27) (0.27) (0.59) (0.13) (0.43)

lg(GDP per capita) 2.455* 2.441* 1.961 3.175** 3.513***

(0.02) (0.02) (0.21) (0.01) (0.00)

Corruption Perception Index -0.210 -0.224 1.653 -0.236 -0.441

(0.81) (0.80) (0.19) (0.82) (0.62)

Constant 33.084*** 41.499*** 14.159 -12.307 332.768***

(0.00) (0.00) (0.29) (0.43) (0.00)
N 152 152 61 123 141
R2 0.476 0.476 0.331 0.462 0.543

p-values in parentheses
*
p < 0.05, ** p < 0.01, *** p < 0.001

14
 Journal of Political Inquiry 4 (2011)

Critics may argue that these results do not reflect a positive effect of increased participa-

tion of women on environmental performance but rather that both women’s participation in envi-

ronmental decision-making and improved environmental performance are independent conse-

quences of high levels of human and economic development. To counter such arguments, I have

included the logarithm of a country’s per capita GDP as a control variable in all regression anal-

yses. However, it is interesting to note that the percentage of women in a country’s delegation to

the COP-15 is not as highly correlated with human and economic development as one might

think; the correlation coefficient between the percentage of women on a country’s delegation and

the 2002 Human Development Index is r=0.29.47 The correlation coefficient between the per-

centage of women on a country’s delegation and per capita GDP is slightly higher but still only

r=0.34. The ten countries with the highest percentage of women on their delegations are Trini-

dad and Tobago, Uzbekistan, Bulgaria, Saint Lucia, Latvia, Saint Kitts and Nevis, Moldova,

Cook Islands, Niue, and Antigua and Barbuda, none of which are highly developed and industri-

alized countries (see Appendix). This reinforces the notion that the increased participation of

women in environmental politics is an independent determinant of a country’s environmental

performance and not simply a covariate of human or economic development and is in line with

common criticisms of GDP as a measure of human progress.

Column two (2) suggests that the deviation from gender parity in the composition of a

country’s delegation in any direction is associated with a decrease in environmental perfor-

mance. However, it is questionable whether this coefficient gives us any information about the

potential effect of having too many women on board, since out of the 194 countries in the sam-

ple, only Bulgaria, Saint Lucia, Latvia, Saint Kitts and Nevis, Moldova, Cook Islands, Niue, and

Antigua and Barbuda had a delegation that consisted of more than 50 percent women. There-
47
Ibid.

15
 Journal of Political Inquiry 4 (2011)

fore, the significantly negative coefficient most likely reflects the negative effect of having a low

level of participation of women in environmental politics.

Columns three (3) and four (4) suggest that my argument is robust to the use of different

measurements such as the more general measures of gender-equality. Column five (5) presents

the results of using the year in which women were given the right to vote as an instrumental vari-

able to test whether there is a causal relationship between women’s participation in environmen-

tal politics and environmental performance. The significantly negative coefficient suggests that

the later women were given the right the vote, the poorer a country’s environmental performance

today. This highly significant instrumental variable supports the contention of a causal effect of

women’s participation in environmental policymaking on a country’s environmental perfor-

mance.

Conclusion

The goal of this paper was to test whether there is quantitative support for the argument made by

gender theorists, according to whom an increase in the percentage of women in environmental

politics would have a positive effect on environmental performance. While strong support for

this argument has emerged from case studies, I argue that quantitative evidence could strengthen

the argument put forward. Therefore, the research question that this paper sought to answer was

whether there is a causal link between increased participation of women in environmental poli-

tics and a country’s environmental performance, and if the answer is yes, how strong the effect

is. My findings suggest that the percentage of women in a country’s delegation to international

conferences indeed has a positive influence on environmental performance. Furthermore, the use

of an instrumental variable supports the idea that this is a causal relationship. Under the assump-

16
 Journal of Political Inquiry 4 (2011)

tion of linearity, my results suggest that if the percentage of women at international environmen-

tal negotiations would be increased from current levels to constitute an average of 50 percent,

global environmental performance would improve by 3.7 points on a scale from 0 to 100. There-

fore, the overall results reinforce the importance of increasing women’s participation in envi-

ronmental politics, which is not only an end in its own right but also a means to achieve higher

levels of environmental performance.

Once cross-national data on women’s participation in other bodies of environmental poli-

cymaking, particularly the ministerial level, becomes available, it would be interesting to per-

form similar analyses using this new data to run further quantitative tests of my hypothesis. Ad-

ditionally, it would be interesting to take into account the role that women play within such bod-

ies, e.g. whether they are in leading or supporting positions. In addition, a distinction between

developing and developed countries might give us insights into whether women’s participation

has a different impact at different stages of development. Similarly, it would be interesting to

look at the sub-indexes of the EPI to determine for which environmental policy goals women’s

participation matters the most. Since hardly any quantitative empirical evidence on the impact of

women’s participation in environmental decision-making has been published, much work still

needs to be done. This paper made an important first step in this direction.

Environmental problems such as climate change and the rapid decrease in biodiversity ur-

gently need to be addressed to prevent catastrophe. My paper shows that women need to be a

crucial part of the solution. My analyses offer evidence for this claim at the level of intergov-

ernmental policymaking. However, an increase women’s participation at the level of intergov-

ernmental policymaking needs to be initiated from the ground up—in policymaking processes at

the city, state and national level, as well as in the educational system and the private sector. De-

17
 Journal of Political Inquiry 4 (2011)

spite this clear evidence, women in many parts of the world still face unequal access to decision-

making, formal financial systems, land ownership, education and information, which undermines

their way into positions of decision-making power. This needs to change for the well being of

women and for the well being of the planet.

18
 Journal of Political Inquiry 4 (2011)

APPENDIX

Women in Country Delegations to COP 15 in Percentage

19
Eritrea 0.00 Japan 14.89 Venezuela 26.19
Honduras 0.00 Nauru 15.38 Fiji 26.32
Jordan 0.00 Central African Papua New
Korea, North 0.00 Republic 16.13 Guinea 26.39
Libya 0.00 Lebanon 16.39 Azerbaijan 26.67
Mauritius 0.00 Gambia 16.67 Micronesia 26.67
Myanmar 0.00 Ghana 16.67 China 26.73
Qatar 0.00 Guinea 16.67 Zambia 26.79
Saudi Arabia 0.00 Sierra Leone 16.67 Namibia 27.08
Somalia 0.00 Cote d'Ivoire 18.18 Benin 27.27
Yemen 0.00 Djibouti 18.18 Estonia 27.27
Bangladesh 2.44 Congo 19.05 Macedonia 27.27
Chad 2.56 Tunisia 19.05 Grenada 27.78
Bhutan 6.25 Korea, South 19.16 Hungary 28.13
Ethiopia 6.38 Nigeria 19.73 Austria 28.30
Mauritania 6.98 El Salvador 20.00 Luxembourg 28.57
Cuba 7.41 Guatemala 20.00 Marshall Islands 28.57
Algeria 7.69 Liechtenstein 20.00 Brazil 29.55
United Arab Monaco 20.00 Bahamas, The 30.00
Emirates 7.69 Niger 20.00 Ukraine 30.23
Brunei 8.33 San Marino 20.00 Ireland 30.43
Panama 8.33 Seychelles 20.00 Serbia 30.43
Timor-Leste 8.33 Singapore 20.34 Belgium 31.18
Bahrain 8.70 Malaysia 20.37 Guyana 31.25
Egypt 9.09 Gabon 20.45 Tuvalu 31.58
India 9.09 Maldives 21.88 Kiribati 31.82
Mali 9.64 Sudan 21.95 Bolivia 32.00
Senegal 9.90 Indonesia 21.98 Dominican Re-
Oman 10.00 Malta 22.22 public 32.00
Vanuatu 10.00 Kenya 22.30 Czech Republic 32.08
Kuwait 10.20 Russia 22.69 Turkey 32.79
Cameroon 10.34 Iraq 22.73 United States 32.91
Zimbabwe 11.63 Angola 22.92 Barbados 33.33
Equatorial Guin- Liberia 23.08 Bosnia and Her-
ea 11.76 Palau 23.08 zegovina 33.33
Togo 11.76 Morocco 23.33 Comoros 33.33
Pakistan 11.90 Swaziland 23.33 Georgia 33.33
Afghanistan 12.50 Chile 23.40 Guinea-Bissau 33.33
Nepal 12.77 Switzerland 24.32 Paraguay 33.33
Cambodia 13.04 France 24.45 Samoa 33.33
Vietnam 13.75 Congo, Demo- Turkmenistan 33.33
Burkina Faso 13.79 cratic Republic 24.47 Finland 33.78
Iran 14.14 Malawi 24.53 Uganda 34.29
Cape Verde 14.29 Haiti 25.00 New Zealand 34.38
Mongolia 14.29 Israel 25.00 Slovakia 34.48
Tajikistan 14.29 Tanzania 25.33 Canada 35.27
Tonga 14.29 Portugal 25.93 Kazakhstan 35.29

20
European Union 35.39 Sao Tome and Australia 46.94
Netherlands 35.53 Principe 40.00 Ecuador 47.22
Belarus 35.71 Sri Lanka 40.00 Suriname 47.62
Nicaragua 35.71 Syria 40.00 Sweden 48.31
Costa Rica 35.90 Croatia 40.91 Mozambique 50.00
Botswana 36.00 Argentina 41.18 South Africa 50.00
United Kingdom 36.04 Peru 41.18 Trinidad and To-
Denmark 36.05 Germany 41.41 bago 50.00
Armenia 36.36 Philippines 41.44 Uzbekistan 50.00
Uruguay 36.36 Jamaica 41.67 Bulgaria 53.33
Laos 36.84 Montenegro 41.67 Saint Lucia 54.55
Albania 37.50 Romania 42.19 Latvia 55.17
Lesotho 37.50 Solomon Islands 42.31 Saint Kitts and
Cyprus 37.93 Lithuania 42.42 Nevis 55.56
Spain 38.27 Norway 42.69 Moldova 57.14
Madagascar 38.71 Kyrgyzstan 42.86 Cook Islands 66.67
Burundi 38.89 Slovenia 42.86 Niue 80.00
Thailand 38.89 Greece 43.75 Antigua and Bar-
Iceland 39.29 Saint Vincent and buda 83.33
Poland 39.51 the Grenadines 44.44
Colombia 40.00 Italy 45.90
Dominica 40.00 Belize 46.15
Mexico 40.00 Rwanda 46.15

21
REFERENCES

Breitmeier, Helmut. "International Organizations and the Creation of Environmental Regimes."

In Global Governance. Drawing Insights from the Environmental Experience, edited by
Oran R. Young. 87-114. Cambridge: The MIT Press, 1997.
Dankelman, Irene. "Climate Change, Human Security and Gender." In Gender and Climate
Change: An Introduction, edited by Irene Dankelman. 55-71. London: earthscan, 2010.
Dankelman, Irene. "Introduction: Exploring Gender, Environment, and Climate Change." In
Gender and Climate Change: An Introduction, edited by Irene Dankelman. 1-18. London:
earthscan, 2010.
Dankelman, Irene, and Willy Jansen. "Gender, Environment and Climate Change: Understanding
the Linkages." In Gender and Climate Change: An Introduction, edited by Irene Dankel-
man. 21-54. London: earthscan, 2010.
Das, Jayoti, and Cassandra E. DiRienzo. "Is Ethnic Diversity Good for the Environment? A
Cross-Country Analysis." The Journal of Environment & Development 19, no. 1 (2010):
91-113.
DeVault, Majorie L. Liberating Method. Feminism and Social Research. Philadelphia: Temple
University Press, 1999.
Emerson, Jay, Daniel C. Esty, Malanding Jaiteh, Christine Kim, Marc A. Levy, Valentina Mara,
Alex de Sherbinin, and Tanja Srebotnjak, "2010 Environmental Performance Index," Ac-
cessed November 8, 2010,
http://epi.yale.edu/file_columns/0000/0157/epi2010_report.pdf.
Enarson, Elaine, and Brenda Phillips. "Invitation to a New Feminist Disaster Sociology: Integrat-
ing Feminist Theory and Methods." In Women and Disasters. From Theory to Practice,
edited by Brenda D. Phillips and Betty Hearn Morrow. 41-74. Philadelphia: Xlibris, 2008.
Esty, Daniel C., and Michael E. Porter. "National Environmental Performance: An Empirical
Analysis of Policy Results and Determinants." Environment and Development Economics
10 (2005): 391-434.
Harris, Rachel. "Gender Aspects of Climate Change in the Us Gulf Coast Region." In Gender
and Climate Change: An Introduction, edited by Irene Dankelman. 152-58. London:
earthscan, 2010.
Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi, "The Worldwide Governance Indicators
(Wgi) Project," Accessed November 19, 2010,
http://info.worldbank.org/governance/wgi/index.asp.
McKinsey&Company, "Women Matter. Gender Diversity, a Corporate Performance Driver," Ac-
cessed October 30, 2010,
http://www.mckinsey.com/locations/swiss/news_publications/pdf/women_matter_english
.pdf.
Momsen, Janet Henshall. Gender and Development. London: Routledge, 2004.
OECD, "The Gender, Institutions and Development Data Base," Accessed November 20, 2010,
http://www.oecd.org/document/0/0,3343,en_2649_33731_39323280_1_1_1_1,00.html.
Smith, Daniel Somers. "Place-Based Environmentalism and Global Warming: Conceptual Con-
tradictions of American Environmentalism." Ethics & International Affairs 15, no. 2
(2001): 117-34.
Teorell, Jan, Nicholas Charron, Marcus Samanni, Sören Holmberg, and Bo Rothstein, "The Qual-
ity of Government Dataset, Version 27may10. University of Gothenburg: The Quality of
Government Institute," Accessed November 11, 2010, http://www.qog.pol.gu.se.

22
UNDP. Human Development Report 1995. Gender and Human Development. New York: Ox-
ford University Press, 1995.
UNFCCC, "Fccc/Cp/2009/Inf.1 (Part 1). Conference of the Parties. Fifteenth Session. Copenha-
gen, 8-18 December 2009. List of Participants," Accessed November 20, 2010,
http://unfccc.int/resource/docs/2009/cop15/eng/inf01p01.pdf.
United Nations Statistics Division, "National Accounts Main Aggregates Database," Accessed
November 13, 2010, http://unstats.un.org/unsd/snaama/.
Videras, Julio, and Christopher J. Bordoni. "Ethnic Heterogeneity and the Enforcement of Envi-
ronmental Regulation." Review of Social Economy 64, no. 4 (2006): 539-62.
Wooldridge, Jeffrey M. Econometric Analysis of Cross Section and Panel Data. Cambridge: The
MIT Press, 2002.
World Economic Forum, "The Global Gender Gap Report 2005," Accessed November 24, 2010,
http://www.weforum.org/gendergap.

23