Quantitative Analysis of Factors Hindering or Enabling the Protection of Human Rights

Thumbnail Image
Date
2016-03
Authors
Marx, Axel
Soares, Jadir
Journal Title
Journal ISSN
Volume Title
Publisher
FRAME
Abstract
Different factors influence the protection of human rights. They are extensively discussed in work package 2. This report adds to previous deliverables by focusing on a quantitative analysis of factors hindering or enabling the protection of human rights at the country level. It aims to explain variation between countries in terms of the protection of human rights. Although this kind of approach misses deep comprehension of the specifics of a given country, it offers a better understanding of how different factors influence the protection of human rights around the world, favouring a comparison between countries. The first part of this report starts with a discussion of the different measures of human rights (i.e. the dependent variable) available in the quantitative literature on the protection of human rights. Some of these measures or dependent variables try to capture the overall protection of the many different human rights in a specific country, while other measures are far more exact, capturing the respect of a specific human right, e.g., rights of freedom of association and collective bargaining or the right to a fair trial. The authors favour more specific measures for which data is collected on the basis of extensive data-collection protocols (see Section 1.1). Next, Section 1.2 discusses different political, socio-cultural, economic and international factors that have been used in quantitative studies to explain variation in the protection of human rights. A literature review on the determinants of civil and political rights and workers’ rights (the right to freedom of association and collective bargaining) is presented in Sections 1.2.A and 1.2.B, respectively. Factors such as the ratification of international agreements, democracy and development level have a positive net effect on the protection of civil and political rights. On the other hand, civil war, state religion, and economic growth have a negative effect on the protection of civil and political rights. Concerning workers’ rights, the main factors positively associated with protection of freedom of association and collective bargaining are democracy, development level, investment flows, and neighbour’s pressure, while civil war and large population have a negative effect. These studies illustrate a standard approach for quantitatively investigating the protection of human rights, which focus on the net effect of one explanatory factor (independent variable) on a measure of human rights (dependent variable) (see Section 1.2.C). This approach often uses statistical techniques to measure, on average, the net-effect of each independent variable on the protection of human rights. Although the average effect is relevant information, this approach neglects the fact that these variables can have different effects depending on their interaction with other variables. Hence, focusing on isolated factors and searching for one or two explanatory variables with most explanatory power neglects the fact that protection of human rights can be the result of several, mutually complementary models (or causal paths). As the discussion below will show, some studies find a positive effect for a variable while others do not find any effect or even a negative effect for the same variable. For example, in the case of FACB rights, different authors find different effects for the ratification of ILO conventions and trade openness. This is explained by the fact that, depending on the interaction with other factors, a certain variable can have a different effect. In order to understand these dynamics, we need to focus on the interaction of explanatory variables. In order to address this issue, this report uses a different set of analytic techniques, Qualitative Comparative Analysis (QCA), to analyse variation in the protection of human rights. Part 2 starts with an introduction to Qualitative Comparative Analysis (QCA). Instead of measuring net effects of explanatory factors on the protection of rights, QCA introduces ideas such as conjunctural causation (i.e., combinations of factors produce an outcome), multifinality (i.e., same factor can have different outcomes), equifinality (i.e., different factors can produce the same outcome) and asymmetric causality (i.e., the presence and absence of the outcome have different explanations). Different from the net effect of each factor, QCA provides a set of analytic techniques allowing researchers to identify necessary explanatory factors (necessary conditions) for the protection of human rights as well as to identify sufficient (combinations of) conditions for the protection of human rights. Section 2 introduces QCA as a research approach and technique and illustrates its modus operandi. The third part of the report then presents an application of QCA to the study of human rights with a specific focus on the rights of freedom of association and collective bargaining (FACB rights), which includes both a longitudinal as well as a cross-sectional analysis. First, the report introduces the outcome/dependent variable (FACB-rights) and presents the results of newly collected data for a set of 73 countries over 30 years. The analysis presented in Section 3.2 shows a downward trend for the protection of freedom of association and collective bargaining rights. This trend is persistent for all regions investigated over the period 1985-2012. Although the results might be cause for concern, the authors also highlight that this trend can be influenced by many different dynamics, including more accurate reporting of violations. Next, QCA is applied to the case of rights of freedom of association and collective bargaining (Section 3.3). The QCA approach is explained step by step, from the calibration of variables into sets (i.e., the transformation of the original raw data into sets of countries for which FACB rights are well protected and a whole range of explanatory factors) to the analysis of necessary and sufficient conditions. The results show the persistence of three necessary conditions (democracy, ratification of ILO conventions and absence of civil conflicts), independent of the period of analysis (2002 or 2012) and calibration (one set of countries with high level of protection or another set of countries where the protection of rights is above the mean). These necessary conditions are aligned with theoretical expectations and previous research, since democracy, ratification of ILO conventions, and the absence of civil conflict were also identified in the literature as being significantly correlated with high protection of FACB rights. The QCA analysis adds to this discussion that these conditions are not only merely highly correlated with the protection of these rights, but they are necessary (although not sufficient) factors for protecting FACB rights. The analysis of necessary conditions was further separated into two analyses; one focusing on the factors influencing the protection of these rights in law and one focusing on the factors influencing the protection of these rights in practice. When comparing necessary conditions for FACB rights overall, in law and in practice, it is evident that the absence of civil conflicts is the most important necessary condition, showing highly consistent results. It is also interesting to note that ratification is a necessary condition for the protection of FACB rights in general (overall) and in law, but not for the protection of these rights in practice. This result confirms the presence of so-called false positives, i.e., countries that ratify conventions without the intention of enforcing them. The comparison also shows that the presence of democracy in a country is a strong necessary condition for the protection of rights overall as well as rights in law. However, contrary to expectations, democracy is not necessary for rights in practice, which might be related to the difficulty of enforcing compliance with standards (i.e., the compliance gap). Next, an analysis of sufficient (combination of) conditions (i.e., different causal paths to an outcome) for FACB rights in 2002 and 2012 are presented in Section 3.3.C. Considering all explanatory conditions investigated in this study, many different possible explanatory models can be developed to explain the outcome. Four models were selected for further analysis based on measures of consistency, coverage, and complexity, which are explained in the report. Results show that a combination of five conditions is sufficient to reach high protection of FACB rights: democracy, ratification of ILO conventions, absence of civil conflicts, the level of economic development, and the pressure from neighbouring countries. These results are aligned with evidence presented by previous research and are the core conditions associated with the protection of FACB rights. Other conditions, such as left-wing executive, trade openness and flows of foreign investments are also relevant, but their effect depends on the context (interaction with other explanatory factors). For example, high trade openness is present in some combinations of conditions leading to high protection of FACB rights, but in other combinations of conditions it is the absence of trade openness that contributes to the outcome. Results like this are characteristic of a QCA analysis. QCA presents different paths, or different combinations of conditions, that together can lead to the outcome (i.e. multiple conjunctural causation). This particular aspect of QCA can help us to understand why some correlation based statistical studies find a positive relationship between trade and the protection of labour rights, while other studies find the opposite relationship. In Section 3.3.D, we analyse in-depth this relationship between economic indicators and protection of FACB rights. These results indicate that one cannot draw any general conclusions, but that they are highly dependent on the context. Finally, Part 4 presents our overall conclusions. First, it discusses the importance of rights of freedom of association and collective bargaining, highlighting that the protection of these rights have gained increased attention in international fora. Secondly, Section 4 discusses the main findings of this study and their policy implications, giving special attention to the three necessary conditions for the outcome “high protection of FACB rights” (democracy, ratification of ILO conventions, and absence of civil conflicts), and discusses the five conditions that are sufficient for this outcome: democracy, ratification of ILO conventions, absence of civil conflicts, the level of economic development, and the pressure from neighbouring countries. Lastly, Part 4 discusses the advantages and disadvantages of using QCA for a quantitative analysis of factors. QCA allows the identification of non-trivial necessary conditions and the analysis of multiple conjunctural causation (see the case of trade openness above). This study also reveals the presence of equifinality, namely the fact that different paths, each consisting of different combinations of conditions, lead to the same outcome. On the other hand, the complexity of this result is discussed as a drawback of QCA. Although it uses minimization procedures to reduce complexity, this can be still a problem, particularly when many explanatory factors are analyzed. In sum, this report identifies factors that influence the protection of human rights, such as democracy and ratification of treaties, and absence of war. The results are well aligned with previous studies and add new insights for how different combinations of conditions/factors can affect the protection of human rights.
Description
Keywords
Citation