Skip to main content

Advertisement

Log in

What drives the adoption of climate change mitigation policy? A dynamic network approach to policy diffusion

  • Research Article
  • Published:
Policy Sciences Aims and scope Submit manuscript

Abstract

The requirement of bottom-up action from all the countries to deal with climate change makes it necessary to analyze the factors influencing policy adoption. This article contributes to the policy literature by shedding light on the conditions, which incentivize countries to adopt more climate mitigation policies. The theoretical argument builds on the integrated approaches to study policy diffusion, which include both internal and external determinants as explanations for the adoption of policies. While previous applications typically operationalize the latter by regional proximity, this study highlights the added value of network dependencies capturing political and cooperative interactions across countries. The article finds that the adoption of climate policies is a matter of social influence. Countries are more likely to adopt policies if they cooperate with countries that have adopted more climate policies and are in a similar structural position to countries that are active in climate protection. This article not only is an important theoretical contribution to the policy literature but also enriches our methodological and empirical understanding of climate policy diffusion.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. Mitigation policies usually relate to energy demand, supply, transportation, energy efficiency measures, reducing emissions from deforestation and forest degradation in developing countries (REDD +), etc.

  2. See IPCC (2007), Chapter 11 for a more detailed discussion on this issue (https://www.ipcc.ch/publications_and_data/ar4/wg3/en/ch11.html).

  3. Adaptation is a process by which individuals, communities and countries attempt to deal with the consequences of climate change, usually on a local level (Hasson et al. 2010).

  4. Compare Leenders (2002) for an encompassing presentation of network autocorrelation models.

  5. Compare Eq. 1: ADOPTit = f (INTit = (Mit + Rit + ADOPTit−1) + EXTit = (PROXi + NETit)).

  6. An example of such a flagship or framework law will be Bulgaria’s Climate Change Mitigation Act. The act puts forward the principles of state policy for climate change, the rules relating to emissions trading and the modalities for financing green projects Nachmany et al. (2014).

  7. Compare: Notre Dame's Environmental Change Initiative (ND-ECI) (2015). Notre Dame Global Adaptation Index (ND-GAIN) Available online: URL: http://index.gain.org/ranking/vulnerability.

    The vulnerability component measures the extent to which a country is susceptible to the adverse effects of climate change. Specifically, it captures a country’s exposure to climate hazards, its sensitivity to climate impacts, and its adaptive capacity.

  8. Compare ‘Appendix 4’ where we control for time-varying vulnerability indicators. The results do not change.

  9. Data on CO2 emissions, fossil fuel consumption, and party governments are taken from the Quality of Government Standard Dataset 2018 (Teorell et al. 2018).

  10. In addition to these models, in 'Appendix 4', we present a number of robustness checks. We estimated a number of models testing other covariates, variable combinations, interaction effects, and models using different time slices. Overall, our results are robust across these models.

  11. Marrakech Accords. Available at: http://unfccc.int/cop7/.

References

Download references

Acknowledgements

We are grateful to Laurence Brandenberger, Paul Wagner, and our anonymous reviewers for supporting this analysis by providing valuable discussion and critique on how to improve this paper. In addition, we thank Markus Schwarzbauer for editing this text.

Funding

This work was supported by the Swiss National Science Foundation (Grant No. 137808)

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Marlene Kammerer.

Appendices

Appendix 1: country list and coding

Country name

UN country code

Status in UNFCCC

Gain vulnerability rank

Policy adoption

t1

t2

t3

t4

United Arab Emirates

ARE

Non-Annex I

34

0

0

1

1

Argentina

ARG

Non-Annex I

44

1

5

0

3

Australia

AUS

Annex I

1

1

1

1

8

Austria

AUT

Annex I

17

1

2

0

5

Belgium

BEL

Annex I

59

4

2

0

3

Bangladesh

BGD

Non-Annex I

143

1

0

2

4

Bulgaria

BGR

Non-Annex I

39

3

1

2

3

Belarus

BLR

Annex I

53

1

4

2

3

Bolivia

BOL

Non-Annex I

79

1

0

1

4

Brazil

BRA

Non-Annex I

23

2

4

3

3

Canada

CAN

Annex I

2

1

0

1

4

Switzerland

CHE

Annex I

12

2

0

1

3

Chile

CHL

Non-Annex I

23

3

0

3

6

China

CHN

Non-Annex I

28

1

1

0

5

Democratic Republic of Congo

COD

Non-Annex I

161

0

0

0

4

Colombia

COL

Non-Annex I

42

2

1

0

5

Costa Rica

CRI

Non-Annex I

65

1

0

3

4

Cuba

CUB

Non-Annex I

82

1

1

1

0

Czech Rep

CZE

Annex I

18

2

0

0

2

Germany

DEU

Annex I

13

3

3

2

5

Denmark

DNK

Annex I

14

0

1

4

2

Dominican Republic

DOM

Non-Annex I

109

2

1

1

2

Algeria

DZA

Non-Annex I

69

3

3

0

6

Ecuador

ECU

Non-Annex I

74

0

0

1

4

Egypt

EGY

Non-Annex I

75

2

3

0

15

Spain

ESP

Annex I

15

1

0

0

3

Ethiopia

ETH

Non-Annex I

146

0

0

1

8

European Union

EU

Annex I

0

6

0

10

8

Finland

FIN

Annex I

10

2

1

0

4

France

FRA

Annex I

9

1

2

1

5

Gabon

GAB

Non-Annex I

86

0

0

0

4

United Kingdom

GBR

Annex I

3

6

4

2

10

Ghana

GHA

Non-Annex I

126

1

1

0

5

Greece

GRC

Annex I

32

2

2

0

7

Grenada

GRD

Non-Annex I

91

0

1

0

4

Guatemala

GTM

Non-Annex I

119

1

0

1

2

Guyana

GUY

Non-Annex I

135

0

0

1

0

Hungary

HUN

Annex I

42

0

0

3

4

Indonesia

IDN

Non-Annex I

87

0

3

8

9

India

IND

Non-Annex I

115

3

4

1

2

Ireland

IRL

Annex I

19

3

2

1

6

Iran (Islamic Republic of)

IRN

Non-Annex I

53

1

1

2

4

Israel

ISR

Non-Annex I

56

2

0

0

6

Italy

ITA

Annex I

23

8

5

3

7

Jamaica

JAM

Non-Annex I

94

1

0

1

2

Jordan

JOR

Non-Annex I

101

1

0

0

2

Japan

JPN

Annex I

27

3

1

0

2

Kazakhstan

KAZ

Non-Annex I

30

1

2

1

1

Kenya

KEN

Non-Annex I

148

0

1

1

2

Republic of Korea

KOR

Non-Annex I

32

0

1

1

3

Kuwait

KWT

Non-Annex I

103

1

0

0

0

Morocco

MAR

Non-Annex I

91

0

0

2

6

Madagascar

MDG

Non-Annex I

160

0

0

0

2

Maldives

MDV

Non-Annex I

0

0

0

0

0

Mexico

MEX

Non-Annex I

47

1

2

0

4

Myanmar

MMR

Non-Annex I

134

0

0

0

1

Mongolia

MNG

Non-Annex I

76

1

2

0

5

Mozambique

MOZ

Non-Annex I

145

0

0

4

4

Malaysia

MYS

Non-Annex I

31

0

1

0

4

Nigeria

NGA

Non-Annex I

130

0

1

0

3

Netherlands

NLD

Annex I

49

3

2

0

2

Norway

NOR

Annex I

4

1

0

1

3

Nepal

NPL

Non-Annex I

131

1

0

0

1

New Zealand

NZL

Annex I

5

5

0

0

0

Pakistan

PAK

Non-Annex I

111

0

0

0

8

Peru

PER

Non-Annex I

71

3

1

2

7

Philippines

PHL

Non-Annex I

95

1

2

3

6

Poland

POL

Annex I

22

1

1

2

3

Portugal

POR

Annex I

45

1

3

0

4

Romania

ROU

Annex I

84

2

1

0

7

RUSSIA

RUS

Annex I

6

2

0

4

2

Rwanda

RWA

Non-Annex I

168

1

0

0

4

Saudi Arabia

SAU

Non-Annex I

63

0

0

1

1

Singapore

SGP

Non-Annex I

40

4

0

0

3

El Salvador

SLV

Non-Annex I

118

0

2

0

2

Slovakia

SVK

Annex I

37

1

3

2

8

Sweden

SWE

Annex I

11

1

0

2

5

Thailand

THA

Non-Annex I

70

0

2

1

7

Tajikistan

TJK

Non-Annex I

102

0

1

1

1

Trinidad and Tobago

TTO

Non-Annex I

66

0

2

0

3

Turkey

TUR

Annex I

41

0

4

0

3

Tuvalu

TUV

Non-Annex I

0

0

1

0

1

Tanzania

TZA

Non-Annex I

140

0

0

0

3

Uganda

UGA

Non-Annex I

166

1

1

0

1

Ukraine

UKR

Annex I

50

3

1

0

1

United States of America

USA

Annex I

7

0

3

4

5

Uzbekistan

UZB

Non-Annex I

62

0

0

1

1

Venezuela

VEN

Non-Annex I

38

0

1

1

1

Vietnam

VNM

Non-Annex I

117

0

3

0

7

Vanuatu

VUT

Non-Annex I

156

1

0

0

0

South Africa

ZAF

Annex I

51

0

0

2

2

Appendix 2: coding of event types

Event types nominal

Event types

Description

Event dummy

Positive statement

1

Optimistic or emphatic comments, symbolic acts, express accord, consider policy option, acknowledge responsibility

Cooperation

Concrete action

2

Consult, make or host a visit, meet, mediate, negotiate, discuss

Cooperation

Appeal positive action

3

Appeal for material cooperation (economic, judicial, information), diplomatic cooperation, aid, political reform, to yield (e.g., easing sanctions, dissent), to negotiate, to settle dispute, to mediate

Cooperation

Intend positive action

4

Express intends to engage in material cooperation (economic, judicial, information), diplomatic cooperation, aid, political reform, to yield (e.g., easing sanctions, dissent), to negotiate, to settle dispute, to mediate

Cooperation

Yield cooperation

5

Ease sanctions, political dissent, agree on political reform

Cooperation

Substantive cooperation

6

Provide aid or engage in material cooperation (economic, judicial, information, intelligence), engage in diplomatic cooperation (praise or endorse, rally support on behalf of, grant diplomatic recognition, apologize, forgive, sing agreement)

Cooperation

Negative statement

− 1

Decline or make pessimistic comment, deny responsibility

Conflict

Demand cooperation

− 2

Demand for material cooperation (economic, judicial, information), diplomatic cooperation, aid, political reform, to yield (e.g., easing sanctions, dissent), to negotiate, to settle dispute, to mediate

Conflict

Criticize, accuse, disapprove

− 3

Disapprove, criticize, accuse, rally opposition against, complain officially, lawsuit, find guilty or liable

Conflict

Reject cooperation, veto

− 4

Reject material cooperation (economic, judicial, information), diplomatic cooperation, aid, political reform, plant, proposal, to yield (e.g., easing sanctions, dissent), to negotiate, to settle a dispute, to mediate. Defy norms, laws, and to veto

Conflict

Threaten

− 5

Threaten to reduce or stop aid, with sanctions, boycott, embargo, with political dissent or repression, to halt negotiations or mediation. Give ultimatum

Conflict

Substantial conflict

− 6

Protest, strike, or boycott, engage in political dissent. Reduce relations, stop material aid, halt negotiations or mediations, impose an embargo, boycott, or strike and coerce and assault

Conflict

Appendix 3: Time periods

T1: Negotiating the Kyoto protocol (1995–2004)

The first (t1) period is characterized by events leading to the ratification of the Kyoto protocol. The release of the first report of the Intergovernmental Panel on Climate Change’s (IPCC) in 1990 was integral to the drafting of the UNFCCC in 1992. The UNFCCC consequently came into force 2 years later during the year 1994. The predominant key principles enshrined in the Climate Convention are the North–South divide and the principle of ‘Common but Differentiated Responsibilities and Respective Capabilities (CBDR/RC).’ It distinguishes between Annex I countries (with greater historical responsibility and capability to combat climate change) and non-Annex I countries with relatively less (or no) such responsibility or capability to combat climate change (Blaxekjær and Nielsen 2014). However, the UNFCCC did not contain binding or specific targets and brought to the fore the need for a more stringent agreement. Consequently, the first Conference of Parties (COP-1) was expected to take strong action leading to the adoption of the Berlin Mandate whose focus was to promote legally binding reduction commitments intended to be adopted at COP-3 in 1997 at Kyoto (Gupta 2010). The adoption of the Berlin Mandate ultimately led to countries becoming a party to the UNFCCC and adopting what came to be known as the Kyoto protocol. The Kyoto protocol was the first internationally binding treaty that contained provisions for developed country parties to take up legally binding emissions targets for period 2008–2012. Unfortunately, enforcing the ratification was much more difficult than expected with the USA, which was a key player in the negotiations, pulling out in 2001. However, parties tried to salvage the effects of the USA pulling out of the Kyoto protocol during the negotiations in Bonn in 2001 leading to the adoption of the Marrakech accords.Footnote 11 The Marrakech Accords set the rules for implementing the Kyoto protocol and detailed the flexibility mechanisms such as the Joint Implementation (JI), Emissions Trading and Clean Development Mechanisms (Betzold et al. 2012). This hallmark phase in the climate policy architecture ends with the entering into force of the Kyoto protocol on February 16, 2005.

T2: Implementing the Kyoto protocol (2005–2007)

During the second period (t2), the most important issues were to implement the Kyoto protocol and to negotiate its successor. With respect to the institutional framework conditions, two important milestones must be mentioned: (1) the adoption of the Bali Road Map in 2007, which paved the way for a post-2012 agreement, and (2) the release of the IPCC’s Fourth Assessment Report in 2007, which brought the climate change issue on top of the international agenda. It instilled great enthusiasm among the parties ahead of COP 15 in Copenhagen with respect to agreeing on a new international legally binding agreement and a second commitment period of the Kyoto protocol. This period is also significant due to the fact that the CDM witnessed a growth and increasing focus on adaptation-related issues (Gupta 2010).

T3: Post-bali enthusiasm (2008–2009)

The high expectations in the run-up to the Copenhagen (COP-15) summit are an important characteristic of period three (t3). However, they remained unfulfilled due to the parties failing to agree on a legally binding follow-up agreement to the Kyoto protocol. With the Copenhagen Accord, the parties only submitted non-binding emission reduction to be implemented at a later point in time. In general, this phase started with high political and public attention toward the climate change issue because of the release of the fourth IPCC report and former US Vice-President and environmentalist Al Gore winning the Nobel Peace Prize in 2007. It, unfortunately, ended with disappointment over the ‘Copenhagen disaster’ (Blühdorn 2012). Similar to the previous phase, the static North–South divide between countries positions remained, with developing countries and emerging economies seeing themselves as having little (or no) responsibility as well as the capability to combat climate change.

T4: Toward a new agreement (2010–2015)

Period 4 (t4) spans from 2010 to 2015. It started with the adoption of the Cancun Agreements in 2010, which advanced important mechanisms such as the Green Climate Fund, the Technology Mechanism, and the Cancun Adaptation Framework. Despite the failure of the Copenhagen conference in 2009, countries continued negotiating with the goal to achieve a legally binding international treaty that is applicable to all parties and comes into effect from 2020. Negotiations on the design of the agreement mainly took place under the Ad hoc Working Group on the Durban Platform for Enhanced Action (ADP). Its main goals were to achieve progress toward implementing clear mitigation contributions by all parties and assisting parties to adapt to a changing climate (Blaxekjær and Nielsen 2014). The time after COP-15 brought a proliferation of institutions and arrangements under the umbrella of the UNFCCC. Moreover, it also called for a reinterpretation and questioning of the UNFCCC key principles, as well as a rearrangement of country groups (Blaxekjær and Nielsen 2014; Brenton 2013). The divide now remained between three main antagonistic camps. The emerging powers stuck to the key principle of CBDR/RC and the North–South divide. They demanded that industrialized countries must carry the heavier burden, as they are historically responsible and are comparatively more capable in combatting climate change. Alongside a broad range of vulnerable (least) developing countries, the EU pressed for sharp emission reductions and called for joint action of all involved countries. In this perspective, all parties must take action and adopt national contributions. The USA and other developed nations such as Russia and Canada were more reluctant to accept legally binding emission reductions. However, due to the new bottom-up approach, where each country now adopts its own legally binding contributions in contrast to the top-down formulated targets of the Kyoto protocol, negotiations finally led to the Paris Agreement being adopted in 2015.

Appendix 4: Robustness checks

In the following, we present a number of robustness checks. Table 5 presents the parameter estimates (presented as coefficients) with standard errors in parentheses for a number of pooled models, i.e., negative binomial regression models that do not control for period fixed effects. Model 1 includes all internal determinants. Model 2 also comprises the same region variable that operationalizes regional proximity. The results are robust to those presented in the paper. Model 3 includes instead of the ND-Gain vulnerability Index two different vulnerability indicators, namely the percentage of agricultural land and the forest area from the total land area of a country. Both variables are taken from the Quality of Governance dataset 2018. In contrast to the ND-Gain index, which reflects the same vulnerability level for all periods, these variables vary of over time. Both indicator variables are not significant just as the vulnerability variable taken from the ND-Gain index, and the model results do not change. We are therefore confident to include the ND-Gain Index into our model as an effective operationalization of country vulnerability. In Model 4, we include the network autocorrelation terms. We find that both terms are not significant and positive, whereas the Same Region variable is still significant. In Model 5 and 6, the Same Region variable is removed. Model 7 controls for the time-lagged versions of the network autocorrelation terms (i.e., NETit−1), but the estimates are not significant. Model 8 includes interaction terms for all relevant covariates interacting with the Cooperative Interaction network autocorrelation variable. These interaction terms indicate if the effects of the respective covariates (i.e., CO2 pc emissions, GDP pc, Vulnerability, and the level of democracy) interact with the intensity of cooperation between two countries. The parameter estimates related to these four interaction terms are all not significant. Finally, Model 9 controls for structural effects of the political interaction networks. In-Degree centrality (i.e., the number of political interactions reflecting the popularity of a country) and Betweenness centrality (i.e., the number of times a country is the link between a pair of countries that are themselves not interacting) are measures for the centrality a country in the political network. The larger the number the more central and powerful an actor is in the network. The terms control whether countries that are more central in the political network adopt more climate policies. The Cliques term captures whether countries that are closely connected in subgroups adopt more climate policies. Thus, all these network statistics test whether the structural position of a country in a political network affects its policy adoption behavior. However, the variables are all not significant.

Table 5 Pooled regression model (***p < 0.001, **p < 0.01, *p < 0.05)

Table 6 presents the results for the fixed effects models with the exact same model setup. The results are more or less unchanged.

Table 6 Fixed effects regression model (***p < 0.001, **p < 0.01, *p < 0.05)

Finally, Table 7 presents the results for different periods with the exact same model setup. For this purpose, we created a version of the dataset with only two periods. In the alternative dataset, the first phase spans from 1995 to 2009 and reflects the developments prior to the Copenhagen summit. The second phase (2010–2015) covers the negotiations ultimately leading to the adoption of the Paris Agreement. We decided to split the dataset into these two phases, as the failure of the Copenhagen summit in 2009 remarked an important hallmark of international climate change politics and we expect different causalities for the later period. In fact, the presented results underpin this assumption. The outcomes show no significant effects for most of the independent variables. In particular, the typical internal determinants explaining climate policy adoption, such as the level of democracy, GDP, or population size do not matter anymore. Instead, countries that have adopted more policies prior to Copenhagen (Policy Adoption t1) more often also adopted new policies later on. We also find that countries with high per capita emissions have adopted fewer policies. With respect to the autocorrelation terms, we see that in the two period setup countries adopt more policies if they have interacted (Cooperative Interactions) in the past (t−1) with others that have adopted more policies. This means that cooperation prior to Copenhagen affected the policy adoption behavior after Copenhagen.

Table 7 Regression models based on two periods (1995–2009) and (2010–2005) (***p < 0.001, **p < 0.01, *p < 0.05)

Appendix 5: marginal effects

figure a

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kammerer, M., Namhata, C. What drives the adoption of climate change mitigation policy? A dynamic network approach to policy diffusion. Policy Sci 51, 477–513 (2018). https://doi.org/10.1007/s11077-018-9332-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11077-018-9332-6

Keywords

Navigation