Remember me
▼ Content


By Kasper Sulkjær Andersen
This article is a part of a master's thesis from 2013.

< TheoryTable of ContentsThe Dependent Variable >

This section will elucidate the research foundation for the subsequent analysis of the shift in China’s climate policy (Section 5.0) i.e. explicating the move from theory to analysis. In order to do so, this section will be subdivided into two sub-sections. The first sub-section (Section 3.1) will elaborate on the philosophy of science underpinning this study; the second sub-section (Section 3.2) will elaborate on the ‘technical’ aspects underpinning the analysis; most importantly, the chosen research design and strategy as well as the definition and operationalisation of central concepts.

In the present dissertation, China’s climate policy is perceived as an objective phenomenon in social reality and as such is perceived as being constituted not by the way actors ‘think’ about this phenomenon but rather as independent from the ‘minds’ of the actors involved. This objectivist starting point implies that China’s climate policy can best be studied through the adherence to positivist research principles i.e. adherence to the principles of explaining rather than understanding the phenomenon under scrutiny. Although, the positivist approach to studying objective phenomena most commonly involve quantitative methodology this dissertation will contend that there are not necessarily any epistemological commitments attached to the chosen research design and vice versa, thus, refuting the notion that quantitative and qualitative research methodology belong to separate and incompatible paradigms (Bryman, 2004:452-454). Rather, the case-study design chosen here (see Section 3.2.1) can both be conducted in purely quantitative terms or in purely qualitative terms or, lastly, as a combination of both, which is the preferred method in the present dissertation.

The research design utilised in the present study can be classified as a case study. Given the bounded nature and the complexity of the empirical phenomenon under scrutiny this design provides the researcher with the most suitable logic of inquiry seeing that the objective of this study is to provide “(...) an intense study of a single unit for the purposes of understanding a larger class of (similar) units” (Gerring, 2004:342). Analysing the driving forces behind China’s climate policy involves the inclusion of and interaction between the multitude of actors embedded within a varied and complex political landscape, thus making other research designs less apt for analysis. In case study terminology, China’s climate policy will be analysed as a single unit64 at different points in time thus approximating what Gerring (2004) refers to as a Type I case study design. The case study employed here can be further distinguished from other case study designs by its analytical ambition and the underlying reasoning found in this study i.e. firstly whether the study seeks to describe or explain and secondly what role theory plays in the study. Following this line of thinking, the research design chosen here can be characterised as an explanatory clinical case study which can be distinguished from other explanatory case studies by the fact that the present study will neither seek to test theory (deduction) nor will it seek to generate new theory (induction). Instead, the present study adheres to the tenets of the clinical tradition of inquiry in which theories are seen as heuristic tools guiding research. The ambition here is to let the nature of the empirical problem under scrutiny define the role played by theory and as such the ambition of the clinical study is to find the fullest and/or ‘best’ explanation for the observed phenomenon (de Vaus, 2001:223ff). In other words, the present study occupies a middle position in the dichotomous area between pure induction and pure deduction in that an abductive logic of research is pursued. This position65 holds that due to the shortcomings of either the inductive reasoning or the deductive reasoning in generating knowledge a third and more dynamic perspective is needed; one which infers the ‘best’ explanation by integrating induction and deduction. Peirce argued that pure deduction cannot generate new knowledge as the deductive method is confirmatory in nature and thus the conclusion is embedded in the premise (Yu, 1994). Conversely, pure induction suffers from the methodological shortcoming of only being able to arrive at superficial conclusions and never to the bottom of things since induction arrives at empirical laws rather than theoretical laws (Ibid). Instead, social research would benefit from employing an abductive method which involves the integration and interplay of induction and deduction thus, combining the strengths from both approaches. From the perspective of abduction the research process is seen as circular rather than linear as the researcher goes back and forth between theory and data in order to arrive at the ‘best’ explanation of a particular empirical phenomenon. Thus, the present study will employ abductive reasoning by alternating between disciplinary studies on the one hand emphasising theory and area studies, on the other hand, emphasising in-depth empirical insights. STRENGTHS AND WEAKNESSES OF THE CASE STUDY DESIGN
The primary strength of case studies and the main reason for why this design is so prolific in much social research is its ability to attain in-depth knowledge of bounded phenomena. If the case study is solidly designed then this type of research can be very effective in providing causal insight (Gerring, 2004). Although severely limited in terms of providing details on the causal effects of a given association, case studies have a clear advantage over other research design in unearthing and developing the causal mechanisms between variables within a limited empirical area (Ibid). The primary strength of the case study, however, is also emphasised by critics as its biggest weakness. The bounded nature of the case study makes it susceptible to criticisms regarding the quality of research; particularly in terms of external validity (or generalisability) and reliability (Bryman, 2004; de Vaus, 2001). The most effective way to mitigate methodological concerns over validity and reliability is, partly, to emulate previous operationalisations of central concepts and if none exist then develop these concepts with a close eye to the operationalisations found in related studies. And partly, to select a case according to its distinct features; namely, choosing a case with similarities with other cases beyond the scope and boundaries of the case under scrutiny will increase external validity. Both considerations have been incorporated into the present research design as the operationalisation of central concepts has been carried out with close attention to the operationalisations employed in other studies. Particularly, the concept of climate vulnerability is rife with definitional disputes and depends to a large extent on which scientific community one belongs to. In order to increase both reliability and measurement validity the present study has utilised the ‘standard’ definition of climate vulnerability offered by the IPCC (2007) which is the most widely cited definition and chosen quantitative indices adhering to this standard definition66. Likewise, in terms of calculating China’s abatement costs the present study relies on the findings of the three most widely cited abatement cost functions.

With regards to case selection, and thus the study’s external validity, China’s climate policy (or China in more general terms) is perceived as a critical case. Not critical in the conventional terminology of case selection as testing a specific hypothesis but critical, nonetheless, in terms of the study’s potential to extend generalisability beyond the confines of China as a single case. Out of all the countries participating in climate negotiations China is the only developing country67 with enough negotiating leverage to match such countries as the US and the EU which are significantly more powerful in terms of material capacities. Gaining an insight into this critical case can contribute to the knowledge on how to better co-opt not only China but also the other BASIC-countries68 in future climate negotiations. Cultural, historical and linguistic differences aside it seems reasonable to assume that the drivers behind the BASIC-countries’ climate policies share more similarities than dissimilarities, thus making China the key to a better understanding of these rising powers.

3.2.2 DATA
The present study relies on a mix of quantitative and qualitative data, however, given the qualitative nature of the case study design employed here qualitative data will take primacy as the main source of information regarding the analysis of China’s climate policy. All of the qualitative data is primarily gathered from secondary sources69. Primary sources of information only play a marginal role in the present dissertation due to a lack of language abilities, namely, given the fact that I do not speak or read Chinese. However, this will not have a bearing on the quality of this study as the most important government white papers outlining China’s climate policy have, for the most part, been translated into English70. Also, in order to at least partially mitigate the lack of primary data I have conducted a semi-structured interview71 with Professor Jørgen Delman who is considered one of Denmark’s foremost capacities on China.

The quantitative data employed in this study (primarily utilised in Model A) are collected from a mix of quantitative indices available on the internet and supplemented by qualitative studies relying on the quantification of data.

Both models introduced in the previous theoretical section72 rely on concepts which cannot easily be defined or operationalised. Thus, for the sake of the quality of the subsequent analysis, namely the validity and reliability it becomes paramount to explicate what is meant by different concepts and how these conceptualisations are measured. Admittedly, due to the complexities of conceptualising and measuring ‘ecological vulnerability’ and ‘abatement costs’, for which Model A relies upon, most of the present sub-section will be dedicated to this work. However, Model B will not be omitted but touched upon in the end of this section. MODEL A: DEFINING ECOLOGICAL VULNERABILITY (CLIMATE VULNERABILITY)
Before proceeding it should be noted that the concept of ‘ecological vulnerability’ used by Sprinz & Vaahtoranta (1994) in their theoretical framework (Model A in this dissertation) is adequately broad to encompass the many guises that environmental impacts come in which gives the framework sufficient flexibility to be applied across environmental contexts. However, for the present purposes where the focus is on analysing the shift in China’s climate policy it seems logical to narrow down the meaning of ‘ecological vulnerability’ to ‘climate vulnerability’ as it helps focus the analysis. Thus, ecological vulnerability will from here on out be equated with climate vulnerability.

The utilisation of Model A hinges importantly on the ability of the researcher to quantify the level of China’s climate vulnerability (and its abatement costs) in order to be able to meaningfully analyse China’s objective national interest in regards to climate change. However, the term ‘climate vulnerability’ is by no means uncomplicated to define or operationalise. In fact, looking at the literature73 it seems that there are as many notions of climate vulnerability as there are studies. In order to effectively circumvent this methodological contingency it is paramount to utilise a generally applicable standard definition. In the academic literature the conceptualisation of climate vulnerability that most clearly approximates a standard definition is what some scholars characterise as an ‘end point’ definition (O’Brien et. al., 2004; Kelly & Adger, 2000). A prominent example of an ‘end point’ definition is the widely used conceptualisation of climate vulnerability employed by the IPCC74. Here climate vulnerability is defined as:

(...) the degree to which a system is susceptible to, and unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate change and variation to which a system is exposed, its sensitivity, and its adaptive capacity (IPCC, 2007:883).

Although this is criticised by some for being too broad (Füssel, 2007) it captures the core idea that climate vulnerability is not a process with a starting point but rather a static ‘end point’ – i.e. countries are either vulnerable to climate change or not, depending on the net impact of changing climatic conditions. Mathematically, this conceptualisation of climate vulnerability is captured by the following equation75:

The equation denotes that vulnerability is a function of both the direct impact (I) of climate change and the adaptive capacity (AC) of a system (country, region, locality etc). However, no scientific consensus exists on which of these two variables should be accorded the most quantitative weight. For the present purposes, the primary focus will be on gauging the impact component of climate vulnerability as this conforms better to the original conceptualisation of ecological vulnerability in Sprinz & Vaahtoranta’s framework (1994). Also, given the fact that the impact component temporally precedes adaptive capacity further speaks to the fact that this should be accorded with primary analytical emphasis. In logical terms, it makes more sense to my mind to assert that a country confronted by the effects of climate change will base its political decisions on the actual impact confronting it and not its available stock of adaptive capacity. In other words, I would argue that it is the impact component of climate vulnerability that countries first and foremost base their decisions on and not its adaptive capacity. Adaptive capacity matters, of course, but can be considered a post hoc rationalisation. MEASURING CLIMATE VULNERABILITY
In order to quantify China’s level of climate vulnerability in accordance with the conceptualisation of vulnerability presented above, the present dissertation will utilise the economic costs associated with changing climatic conditions as the primary vulnerability indicator. This indicator is widely used to gauge climate vulnerability (O’Brien, 2004) and can be considered the most suitable for the present purposes as it is the ratio between the costs of climate vulnerability and abatement costs that matters, according to Model A, when countries decide on which set of policies to adopt. Given that the economic costs of climate change are likely to impact China through a complex set of channels these costs will be sub-divided into a direct and an indirect cost-component. Where the former component refers to the costs of the first-order impact of climate change in terms of loss of infrastructure, loss of arable land etc., the latter component refers to the associated impact of climate change in terms of second-order and long-term disruptions to the economic system76.

Estimating the direct economic costs associated with climate change impacts in China is a very complex and at best a tentative undertaking (Hanemann in Schneider et. al., 2010). However, the Emergency Events Database (EM-DAT) constructed by the Centre for Research on the Epidemiology of Disasters (CRED) 77 seeks to aggregate the total damage costs associated with natural disasters incurred by countries across the globe. This database contains detailed information regarding every natural disaster that has occurred over the last 30 years and is continuously updated on a daily basis. Given the fact that the CRED database is comprised of raw data the analytical section will undertake efforts designed to only include the economic costs caused by those natural disasters that can be linked directly to climate change and will, based on these costs, calculate the total direct costs of climate-induced disasters in China78.

The indirect costs of climate change are significantly more complex to gauge than the direct cost-component given the protracted and discreet nature of these costs. For this reason, the analysis will rely on the findings of climate change experts79 when totalling indirect costs. DEFINING ABATEMENT COSTS
Although considered fixed in the short term due to structural characteristics (see Section 2.1.1) the level of abatement costs matters to China in as much as Model A suggests that the higher the level of abatement costs the lower the inclination towards an ambitious climate policy. For the present purposes abatement costs are simply defined as the: “(...) economic costs of reducing carbon dioxide emissions (...)” (Cline, 2011:2). In general, abatement costs are related to different reduction scenarios given a specific baseline year. Cline (2011) calculates abatement costs based on the emissions reduction scenario pledged to by countries encompassed in the Copenhagen Accord. Cline (2011) extends the time period until 2050 assuming that countries will incrementally curb their emissions until 2020 and will afterwards remain on a low carbon emission trajectory. MEASURING ABATEMENT COSTS
In order to utilise Sprinz & Vaahtoranta’s framework (1994), China’s level of abatement costs needs to be determined. Put simply, abatement costs measure the economic costs associated with reducing the emission of GHG gases (most often carbon dioxide) and are often expressed in terms of cost functions. Based on three leading abatement cost functions (RICE, EMF-22 and McKinsey) Cline (2011) calculates the costs associated with different abatement scenarios. Cline’s results will serve as the analytical backdrop for the estimation of China’s abatement costs in this dissertation.

The RICE cost function (Nordhaus, 2008) and the EMF-2280 cost function are examples of ‘top-down’ approaches to emission reduction meaning that modelling of abatement costs takes place at the aggregate economy wide level as a nonlinear function of the extent of the target carbon cutback (Cline, 2011:26). The cost function in these models can be expressed as kt = αtμtβ, where k is the abatement costs at time (t), α is a parameter that declines over time to reflect the widening menu of technological alternatives, μ is the proportionate reduction in emissions from the baseline year to the projected policy target, and β is the exponent showing the degree of nonlinearity in costs for deeper cuts (Ibid).

The McKinsey cost function81 is an example of a ‘bottom up’ calculation of abatement costs which models costs based on specific operations (e.g. shifting the fluorescent lighting in buildings, increasing the efficiency of light bulbs, introduce hybrid cars etc.) and typically asserts that some initial portion of emissions can be reduced at zero or even negative cost. The McKinsey cost function can be expressed as

where z’ represents marginal cost per unit of carbon dioxide reduction, R is the amount of the reduction, B is the quantity of the reduction at which the marginal cost curve turns vertical (infinite costs for an additional unit of abatement), and A is the cost parameter (Ibid:30).

Cline (2011) notes that where the EMF-22 cost function has a tendency to overestimate the true level of abatement costs the McKinsey cost function, conversely, has a tendency to underestimate the level of abatement costs. Therefore, the analysis will be primarily based on the RICE cost function as the estimation of costs tends to lie in between the two other cost functions thus placing this cost curve closer to the true costs of climate mitigation (Ibid:33ff). MODEL B: OPERATIONALISING A ‘CONDUCIVE POLICY ENVIRONMENT’
The prerequisites necessary for Model B to be employed is markedly different from Model A in the sense that the former model does not rely on quantitative data to the same extent as the latter model and as such, does not include concepts which presuppose a particular methodology. Instead, Model B relies primarily on qualitative data from either primary and/or secondary sources in order to obtain information regarding whether or not a conducive policy environment82 exists in the context of China’s climate policy-making. The specific theoretical proposition derived from Model B (see Section 2.2.1) is that an ambitious climate policy will only be adopted when the net aggregation of interests of China’s most powerful policy actors is tilted in favour of such a proactive policy. No preordained method exists on how to ascertain the existence of a conducive policy environment, however, for the present purposes a conducive policy environment is assumed to be causally associated, according to Model B, with two central variables. The interests of policy actors involved in China’s climate policy-making process, on the one hand, and the configuration of power/influence between different policy actors, on the other, are particularly salient for ascertaining the existence of a conducive policy environment.

The contents of the former variable83 can be extremely difficult to operationalise, let alone, obtain reliable information about due to either an unwillingness to disclose such sensitive information for strategic purposes or due simply to a lack of data. Therefore, interests have to be assumed – at least to some extent. Following in the vein of public choice theory84 the present study assumes that the actors involved in China’s climate policy-making process are focused on maximising in terms of their self-interest. As a proxy for self-interest actors embedded within the official Chinese state apparatus are assumed to pursue acts of utility maximisation in accordance with their organisational interests. Most importantly in this regard are the contents of the organisational mission statements of different actors as the success in attaining these objectives will, at least partially, determine the size and extent of future budgets as well as the political influence of these organisations. With regards to actors located outside of the Chinese state apparatus such as business interests these are assumed to be interested in maximising their profits by minimising their costs of production.

The second main variable pertinent to ascertaining the existence of a conducive policy environment is power. The academic field of political science contains many conceptualisations of power, however, only a few of these are relevant for the present purposes. As it is not possible, due to a lack of transparency, to gain access to information regarding what actually goes on during consultations between different policy actors involved in China’s climate policy-making process, this dissertation will focus on the most tangible expression of political power. Thus, following Dahl’s (1957) classic conceptualisation of power: “A has power over B to the extent that he can get B to do something that B would not otherwise do” (Dahl, 1957:202ff). In regards to gaining information on the distribution of power among China’s climate policy actors the present dissertation will have to primarily rely on the insights provided by secondary studies on the characteristics of Chinese policy-making. As has already been mentioned above, the literature specifically dealing with climate policy-making is relatively sparse which I have sought to partially mitigate by conducting an expert interview.

< TheoryTable of ContentsThe Dependent Variable >

Posted by branner on 24. May 2014, 17:06 0 comment(s) · 2508 views

No comments yet.

Join the debate on Methodology:

Remember me

▲ Top of page
Public Poll
Who is leading the renewable energy race?








Don't know

Thanks for supporting
Copyright © 2009-2020 | About | Contact