Human rights in Saudi Arabia: Finding a causal relationship between
dependency on oil and human rights violations
Bachelor Thesis Politicologie/IBO
Michiel de Vreede
s1851918
Supervisor: Violet Benneker MSc
17-06-2019
words: 8347
Introduction
In this thesis, I will investigate the causal relationship between dependency on oil and violations of human rights by states. Davenport (2007) observers that state authorities, when their power is being challenged, generally exert some form of repression over their citizens. He labels this observation the ‘law of coercive responsiveness’ (Davenport, 2007, p.7). A comprehensive literature has been written over the years to explain repression and human rights violations by states. Some look into the effectiveness of international human rights treaties (Hafner-Burton et al., 2008), others consider the structures of the state and domestic issues (Cole, 2015). One factor that has risen in importance in human rights literature is the relation between the economy, especially natural resources, and human rights. States that are more dependent on natural resources tend to be more inclined to violate human rights
(DeMerrit & Young, 2013).
Most research on this ‘oil dependency’ theory has been conducted using quantitative methods. These large-N methods have proven to be useful in determining a relation between oil and repression. The causal mechanism that they assumed is that states weigh the costs and benefits of violating human rights and decide accordingly. These assumptions are not further examined. Therefore, the main goal of this thesis is to fill this gap in the literature and establish whether the proposed causal mechanism between dependency on oil and violations of human rights works. In order to do this, I will perform a case-study on Saudi Arabia. Saudi Arabia has, since the first oil fields were discovered, build its economy on oil (Wilson, 2004). In 2016 Saudi Arabia announced their vision for 2030, in which they stated to become less dependent on oil in the coming years (Council of Economic and Development affairs, 2017). Two time periods before and after the announcement of Vision 2030, will be studied: 2007-2008 with a higher oil dependency and 2015-2016 with a lower oil dependency. Both the oil dependency theory itself and the causal mechanism are investigated.
The results of this undertaking show for 2007-2008 that both the oil dependency theory as well as the causal mechanism can explain rights violations. In 2015-2016 the theory itself did not did not align with the expectations of the model. This makes it difficult to assess whether the cost/benefit analysis works as a causal mechanism. Moreover, no hard evidence is found that a cost/benefit analysis has actually been made. The academic contribution of this thesis is that it fills the gap in the literature by studying the causal mechanisms of the oil dependency theory. The contribution to society is that it helps to understand why states violate the rights of their citizens. It can help to find ways to stop states from doing this.
Theoretical framework
The overarching question of this study is why states violate human rights. In this section I will give an overview of the literature on the relation between natural resources and human rights, as well as further elaborating on the causal assumptions made in this literature, and further specify the main goal of this research.
The relation between natural resources and human rights: an overview
The starting point of the literature on natural resources is the theoretical development of the so-called ‘natural resource curse’. Sachs and Warner (1995) found that states without natural resources experienced more economic growth than resource-rich states. They argue that the price of raw materials is lower than the price of manufactured goods. Countries that export many raw materials thus generate less income than states that export mostly manufactured goods.
The natural resource curse has later been expanded to explain other phenomenon, such as the presence of civil war or a certain regime type. Ross (2004) finds that oil can be linked to the offset of civil war. He bases this claim on a meta-study of earlier research on resources and civil war. Ross (2001) links the presence of oil with regime type. He finds that states depending on oil export are less democratic. He argues that governments keep taxes low and build up their internal security forces to ward off democratic pressures. Wright et al. (2013) expand on this and find evidence that the presence of natural resources helps to maintain dictatorships. They argue that regimes depending on natural resources do not need revenues from taxes, and are therefore less accountable to society, making it easier to stay in power. Other scholars expanded the resource curse to the field of rights and liberties. Wigley (2018) argues that the ‘Natural resource curse’ also holds for private rights of the individual. He finds that more resource wealth leads to a decrease in private rights. Poe et al. (1999) observe that economic growth leads to a decrease of repression. Egorov et al. (2009) link the abundance of natural resources with media freedom. They argue that free media can be beneficial for a dictator in a resource-poor country, since the media can monitor
bureaucrats and give incentives to improve their quality. In resource-rich countries,
bureaucratic incentives are less important for the dictator. Hence, media freedom is less likely to emerge. Smith (2008) finds that natural resources increase the threat of societal revolution. He argues that when state-income from resources increases, citizens want to have a part of
this. They are therefore more likely to join groups fighting for a more inclusive political system. Peterson (2015) finds that states with a more diverse export tend to have more respect for human rights. Diverse export leads to more welfare among citizens, who will be more likely to press for better human rights. Governments in turn will have more resources to accommodate citizens participation in the economy and improve human rights. Ross (2008) argues that oil has been a factor in gender inequality in the Middle East. According to Ross, oil has led to less labour-participation of women which has overall led to a worse position of women in society. DeMerrit and Young (2013) find that states with abundant natural
resources are more likely to repress their citizens, because they do not have to rely on income from citizens, making repression less costly.
The causal mechanism
The literature discussed so far discovered a relation between the presence of natural resources and violations of several human rights. Most of these authors assumed a causal relationship between the variables. In this section, I will provide a more in-depth discussion of the causal mechanism of the oil dependency theory.
The oil-dependency theory starts with the basic assumption that governments want to retain their power and that their decisions are based on staying in power (Bueno de Mesquita et al., 2003, p.9). In order to maintain their control and hold off possible threats to their power, regimes need income. Tilly (1985) describes how European States were formed. He argues that these states needed income to wage war to keep out enemy forces. They thus needed to impose taxes on their civilians. This increased the need for bureaucracy and helped build the state. There are different ways of earning this income. Bueno de Mesquita and Smith (2009) identify two of them. Firstly, there is the taxation of citizens, as we have seen earlier. The second is income derived from so-called ‘free revenues’ (Bueno de Mesquita & Smith, 2009, p.5). This is revenue for which the government does not depend on their citizens, such as natural resources and foreign aid.
Bueno de Mesquita & Smith (2009) also discuss the relation between how states respond to domestic threats and the type of revenue. When states depend on taxes as revenue, they are more likely to improve the distribution of public goods such as democracy and human rights to please the public. When states depend on so called ‘free revenues’ such as oil and foreign aid, the state is less likely to provide public goods since do not have to please the public. DeMerrit and Young (2013) and Conrad and DeMerrit (2012) work from this theory and link it directly to human rights violations. They argue that states make a cost-benefit
analysis when they decide to violate their citizens’ rights. The major benefit of repression, according to DeMerrit and Young (2013) is that the state is able to remain in power. It can remove threats to state power from society. They can oppress dissidents and break up protests. Usually this is less costly than other measures, such as the increasing the distribution of public goods. However, in the long run it can be costly. Studies have shown that oppression of protests can backfire in the long run, leading to more, dissent, violence and even revolution (DeMerrit & Young, 2013, p.101).
There are also other costs related to repression. If a state relies on public taxes, they face higher costs when repressing. These states need to ensure compliance from their citizens. Citizens expect to get something in return from their taxes. When a state begins to repress citizens, it loses credibility. Citizens do not longer believe that the state will do something good for them with their taxes. These citizens may therefore be less willing to pay taxes in the short term than they would when the state would respect rights. A second cost is that states lose revenue when they repress citizens. When states torture, kill or otherwise remove citizens from society, these citizens are less able to pay taxes. The abuse reduces their income and therefore the amount taxes they pay. When states do not depend on citizens but on natural resources, they do not face these costs when repressing their citizens.
Determining the causal mechanism
In short, the oil dependency theory entails that states that depend on oil for their income are more likely to repress their citizens to stay in power. They do not face the costs of repressing citizens. On the other hand, states that rely on income from citizens are less likely to repress their citizens to stay in power, since they face certain costs when repressing citizens. DeMerrit and Young (2013) and others all assume this causal mechanism. However, they only test the relationship between the variables. Causality is assumed, not measured. Quantitative studies generally do a good job at finding relationships between variables. Qualitative studies are more suited to develop and establish a theory, whereas quantitative studies are better at testing and generalizing a theory (Toshkov, 2018). The oil-dependency theory has been tested with large-N quantitative research. However, a thorough qualitative study on the causal mechanism on the theory is largely absent in the literature. This thesis will fill this gap by applying the theory to a case by using qualitative methods. Thus, the goal of this thesis is to answer the question whether the proposed causal mechanism between dependency on oil and violations of human rights works
Research Design
In this section I will lay out and further specify my research design. First, I will elaborate on my case selection. After that I will formulate several hypotheses and operationalise core concepts. Then I will describe how I will search for a causal relationship and discuss several issues with data and validity.
Case selection
The case that will be used in this research is Saudi Arabia. I have chosen Saudi Arabia because this country has a long tradition with oil and recently announced reforms of the economy, making it suitable for a cross-time case study.
The Kingdom of Saudi Arabia has a long tradition with oil and other natural resources. In 1938, after several failed search operations, a major oil field was found. In the first years, foreign oil companies extracted the oil and shared revenues with the state (Wilson, 2004, p.40-41). From the 1970’s and onwards the Saudi Arabian state joined the OPEC cartel and increased their oil production to 10 million barrels per day in 1980 (Wilson, 2004, p.44). We can thus see that oil is very important for the Saudi Arabian economy. Therefore, it is a fitting case to determine the causal mechanism of the oil dependency theory.
In 2015, King Abdullah of Saudi Arabia died. He was succeeded by his son, King Salman. He appointed his nephew, Mohammed Bin Salman, as the chairman of the Council of Economic and Development Affairs (CEDA). In this position, he was responsible for ‘Vision 2030’. This document, released in 2016, lays out the future plans and targets for the Saudi-economy. One of the most striking features is that it was announced that Saudi Arabia would decrease their dependence on oil and diversify the economy. The document states that oil has been very important for the Saudi-economy. Despite this, the government plans to privatize the national oil company and turn it into an investment fund. This fund will be used to invest in the public sector to diversify the economy (CEDA, 2016).
This change in policy makes it possible to study two time periods in Saudi Arabia, one before and one after the announcement of Vision 2030. In order to determine these periods World Bank data on oil rents will be used1. In a later section I will further elaborate on this data. A problem arising here is that there is no data after 2016. Therefore, the post-vision time
1World Bank (2019). Oil rents (% of GDP). Retrieved from:
period will be 2016. This makes the design slightly more complicated. Preferably, 2015 would be studied as the pre-vision time period. However, oil rents as percentage of GDP was 22,7% in 2015 and 26,4% in 2016. Not only are they almost the same, oil rents are higher in the year we expect them to be low. Nonetheless, 2015 can still serve as an additional ‘base year’ to contrast possible findings from 2016 and providing more context to results. In order to determine whether levels of oil-dependency actually influence levels of rights violations we need more units with actual difference between levels of oil-dependency. The years 2015 and 2016 are amongst the lower levels of oil rents. Thus, a period with higher rents will be
chosen. The year of 2008 has one of the highest scores of the past decade with 53%.
Therefore, this year will be chosen as an additional pre-vision period. It is desired to obtain as much data as possible to achieve a full overview of Saudi Arabia. Therefore, 2007 will be added as an additional year to the pre-vision period. In 2007, oil rents as percentage of GDP was 45,9%. In short, 2008 and 2016 will be chosen as the main units of study within the case, with 2007 and 2015 acting as additional years to provide contrast and more possibilities for comparison.
Hypotheses and measuring causality
In order to really test the oil-dependency theory one needs to know why exactly Saudi Arabia violates personal integrity rights. In other words, a causality between oil-dependency and violation of human rights need to be found. Based on the theory of oil dependency one can say that in a period of high dependency on oil, the state is more reliant on free revenue and less dependent on citizens. We would therefore expect to see relatively more cases of human rights violations in the years 2007-2008 in Saudi Arabia (Hypothesis 1A). In a period when there is less dependency on oil, a state relies less on free revenue and more on citizen taxes. One would therefore expect to see relatively less cases of human rights violations in the years 2015-2016 in Saudi Arabia (Hypothesis 2A).
The causal mechanism of the theory assumes that states make a cost-benefit analysis to decide whether the costs of violating rights are higher or lower than the benefits. One would therefore expect, for both time periods, that Saudi Arabia has weighed the costs of violating human rights versus the benefits. In 2007-2008 one would expect that this has led to more violations of personal integrity rights (Hypothesis 1B) In 2015-2016 one would expect that this has led to less violations of personal integrity rights (Hypothesis 2B). Also, since Saudi-Arabi announced to decrease oil dependency in 2016, it is expected that this will lead to
less violations of human rights in 2016 compared to 2015 (Hypothesis 3). Taking this all together, there are three hypotheses:
H1: It is expected that there will be relatively more cases of human rights violations in 2007-2008 (A). It is also expected that Saudi Arabia made calculation of the costs and benefits of violating rights and decided that the benefits outweighed the costs (B)
H2: It is expected that there will be relatively less cases of human rights violations in 2015-2016 (A). It is also expected that that Saudi Arabia made a calculation of the costs and benefits of violating rights and decided that the costs outweighed the benefits.
H3: It is expected that there will be relatively less cases of human rights violations in 2015 compared to 2016
The question that now needs to be addressed is how this process of making a cost-benefit analysis can be measured. We therefore need to break down the elements of the cost/benefit analysis. The cost-element refers to the costs of repression. In the theory, this is seen as lost revenue from citizens. These costs can be measured by determining the amount of taxes the Saudi citizens have to pay in the years 2007-2008 and 2015-2016. This works as follows: When taxes are low, they are not very important for the total revenue of the state, making it less costly to repress. When a state has higher taxes, it is costlier to repress. The benefits of repression can be numerous, but the most important assumption, as seen before, is that states repress their citizens to remain in power. States can repress dissent and eliminate sources of challenge to their power. This can be measured by looking at whether there are challenges to state powers in 2007-2008 and 2015-2016. In other words, I will look whether there are large-scale protests or social unrest in these periods. When they are present, the benefits of violating rights are higher than when there are no challenges to state power.
In order to assess whether a calculation of these costs and benefits has been made, one needs to retrace the process of making this decision. Process tracing can be described as ‘the use of evidence from within a case to make inferences about casual explanations of that case’ (Bennet and Checkel, 2015, p.4). I will use this method to distinguish the ‘chain of events’ that happen when Saudi Arabia makes a cost/benefit analysis. This means that I will look at what exactly the costs and benefits are for Saudi Arabia and how the policy making process works have worked in this case. From all this information the decision-making process will be reconstructed. When this chain has a direct link from higher oil-dependency leading to a decision to violate more rights or the other way around, the causal mechanism is present. By
comparing the two time periods one can get more certainty about causality. Oil-dependency is the key variable here. When the chain of events leads to the expected level of human rights violations, it is plausible that this is due to the change in oil rents, since this is the variable that changed. A general remark on causality is that all other variables should remain constant (Ceteris-paribus) (Gerring, 2005) Since this research is not confined to a laboratory
environment, it is impossible to keep all other variables constant. It is possible that a change of leadership or developments in the international realm have influenced the decision-making process. Therefore, even if it I found that this causal mechanism works in Saudi Arabia, there still remains uncertainty about it.
Conceptualisation and operationalisation of core concepts
In order to properly examine the hypotheses stated above it is necessary to conceptualise and operationalise several core concepts that are identified in the oil dependency theory. The two most important concepts are oil dependency and human rights violations.
In the previous sections, the concept of oil-dependency has already been used in the case selection. This concept will now be further defined. DeMerrit & Young (2013) submit that states depending on oil largely rely on oil revenues for their income instead of revenue acquired from citizens. There are different measures for oil dependency. Some have measured it by looking at the export value of oil relative to the GDP of a nation (Sachs and
Warner,1995; Ross, 2001; Conrad & DeMerrit, 2012). When the export of oil makes up a larger part of the total GDP, a nation is more dependent on oil. Ross (2008) has some critical remarks on this method. He argues that richer states use more of their oil domestically and therefore export less. Because of this it seems that states are less dependent on oil (Ross, 2008, p.121). Ross therefore looks at oil rents. This is the value of the production of oil, minus extraction costs, as percentage of the GDP (Ross, 2008, p.121). The higher this number is, the more a nation depends on oil. I will use this indicator, since it looks at the importance of having oil for the entire economy. When one only looks at export of oil, it does not take into account the total domestic effects of oil on the economy.
The second core concept is violation of human rights. So far, I have not made a distinction between repression and human rights violations. Repression is a broader concept that can be defined as “the use of physical sanctions against an individual or organization, within the territorial jurisdiction of the state, for the purpose of imposing a cost on the target” (Goldstein, 1978, pp p. xxvii. In Davenport, 2007). Human rights are rights that we have for the sole reason that one is a human being. They are used to protect our human dignity and
ensure we can live a good and meaningful live (Griffin, 2002, p. 310-311).
I will focus on human rights, and especially human rights violations. The rights that I will focus on are personal integrity rights. I have chosen this because Saudi Arabia has signed and ratified the UN Convention Against Torture (CAT). This treaty forbids the use of torture. Therefore, it is a violation of human rights when Saudi Arabia tortures someone. Moreover, torture is also forbidden under the Sharia law of Saudi Arabia, as well as using confessions in trials that were obtained under torture (US State Department, 2008). The CAT provides a workable definition of torture. Article 2 of CAT says that any form of torture is forbidden at any time under all circumstances, except for the addition that torture ‘does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions’ (United Nations, 1984).
I will operationalise torture with two indicators based on the definitions by Conrad et al. (2018). They examine two types of torture. The first one is scarring torture, this leaves visible marks on the victim of torture. Secondly, they define clean torture. These are ‘painful physical techniques of interrogation or control that leave few marks’ (Conrad et al., 2018, p. 5). I will base my indicators on these, but rephrase them to make them more usable for this research. I will make a distinction between physical and non-physical torture. Physical torture can be scarring torture, but includes all types of torture that arises from physical contact. This can be, amongst other things: Beating (with or without leaving marks), lashing (but not as a lawful sanction), sexual abuse and electrocution (Conrad et al., 2018, p.9). The other indicator will be non-physical torture. This includes all forms of torture where there is no physical contact of objects applied to the body. These, amongst others, include: sleep deprivation, withholding food or medication and using noise or light as torture (Conrad et al., 2018, p.9).
Data sources, measurement and validity issues
I will be using a variety of data to test the hypotheses. Most of the data is related to one of the concepts that I have discussed. Data for oil dependency has already been discussed briefly. I will use data from the world bank. The World Bank has data on oil rents as a percentage of GDP. They define oil rents as ‘the difference between the value of crude oil production at world prices and total costs of production’. This is the same as the operationalisation that Ross (2008) uses. A possible problem with oil rents, is that it only looks at income from oil. Perhaps, the other part of the total state income is sufficient for a state to survive. A state may largely rely on oil, but does not depend on it. This uncertainty could influence the results of
this study. However, since this data is easily available and reliable, I will remain using World Bank data.
Human rights violations are more difficult to measure. States are, for obvious reasons, reluctant to share their records on rights violations. For practical reasons, it will be impossible to do field research. I therefore have to rely on external sources. I will be using human rights reports from several sources: Amnesty International, Human Rights Watch and the United States Department of State. These organisations publish annual reports on human rights practices worldwide that I will use. In order to determine high or low levels of rights violations, I will use the coding guide from the CIRI-dataset. They describe three levels of torture. Torture is either practiced frequently, occasionally ore there is no mention of torture in the reports. These codes are mostly based on language in the reports. Words such as widespread, extensive or routine can indicate that torture is practiced frequently. I will therefore research the reports and look for these signalling words. The problem here is that it is difficult to distinguish between frequent practice and occasional practice. CIRI advises to rely on interpretation and overall language of the report. This subjectivity could lead to biased or misinterpreted results. Therefore, it is necessary to carefully consider each report and be sure to uphold a certain continuity. This is achieved by coding all reports that contain obvious signal words as frequent and reports that mention torture without using signal words as occasional.
A downside of using external sources, such as reports, is that information might be not always be accurate, since they do not always have full access to the state. Another possible issue is that this data can be one-sided. External sources can have their own view on Saudi Arabia and on the Middle East in general, which can lead to biased data., Amnesty
International and Human Rights Watch are independent NGO’s, and thus can be expected to act neutral. The US State Department is a political institution and therefore not fully
independent from national and international politics. Therefore, I will look into all three sources to have a most complete overview possible. Also, these reports are written for a wider purpose and might not contain all the information that I am looking for, or focus on other human rights that are not of interest for this research. In the ideal situation, interviews or other field research would be conducted that focused solely on torture. Unfortunately, this is not possible due to the smaller scope of this thesis.
Data for the costs in the cost/benefit analysis will be found in databases on tax
revenues. The world Bank has data on taxes as part of total state revenue. Unfortunately, there is no data on Saudi Arabia. Another dataset, the Government Revenue Dataset contains data
on government revenue and taxes as a percentage of GDP, including Saudi Arabia. This data will be used. Data on the benefit part will mostly be found in news articles and scholarly work. From these articles I will derive whether there were any major challenges to state power, such as uprisings. In order to reconstruct the cost/benefit analysis, I will look into policy documents, speeches and news articles. Policy documents and speeches might be hard to find, but may contain information on policy decisions and arguments why a certain policy was chosen. News articles are a more accessible source. They are useful because journalists might have had more insight into the policy-making arena. I will use articles from respected newspapers, such as The New York Times and The Guardian, to ensure quality and
credibility.
Results
In this section I will discuss the results of the analysis. For each time period, I first discuss whether there is a relation between oil-dependency and violations of human rights
(Hypothesis 1A and 2A). Then I will provide a general overview on policy-making in Saudi Arabia and retrace the process of a cost/benefit analysis (Hypothesis 1B and 2B). Based on this, I decide whether there is a causal mechanism with also using taxation levels (Hypothesis 3A and 3B).
Political decision making in Saudi Arabia
Before we look into the actual results it is first necessary to gain more insight into Saudi Arabian policy making. In Saudi Arabia, there is hardly any separation between politics and religion. The official constitution of the Kingdom is the Sharia and the Quran. All laws and governance are based on the Sharia (Vogel, 2012). Besides the Sharia, there is also the basic law, which contains information about the state and the political system that are usually found in a constitution. The basic law states that Saudi Arabia is a kingdom, and the monarch is chosen from the sons of the first king. He can then appoint a crown prince and a council of ministers who assist him. The basic law also states that the king is both head of state and prime minister (Teitelbaum et al., 2019).This means that the king has almost unlimited power to make decisions. Officially, the king has to adhere to the Sharia and govern the country according to this. He is also surrounded by a council of ministers, the crown prince and his other family members. However, most decisions are not made inside these official channels but between the members of the royal family (Teitelbaum et al., 2019). This makes it difficult
to reconstruct decision-making, since most decisions are made behind closed doors by
multiple men. In the realm of human rights, there is a non-government body to oversee human rights practices. the National Society for Human Rights (NSHR). This agency is founded by the king, but, at least in theory, is independent from the monarch. The goal of the society is to protect and enforce human rights in Saudi Arabia. The formation of this society can be seen as a step in the direction of decreasing human rights violations. However, forming an independent human rights commission was demanded by the CAT. Besides, the presence of this body does not mean that actual rights practices will change.
2007-2008
In this period, oil rents are relatively high with 45,9% in 2007 and 53% in 2008. It is thus expected that torture is practiced frequently in this period. All reports that were studied disclosed practices of torture, and overall it seemed that torture was practiced frequently. For both years, all three reports had an elaborate section on torture2. Physical torture seems to be more widespread and practiced more frequently than non-physical torture. Physical torture included beatings and lashings, which in 2007 even led to the death of one prisoner. Non-physical torture included humiliation, sleep deprivation and insults. Physical torture is more prominent in the US department of State reports, whilst non-physical torture is more
prominent at Human Rights Watch and Amnesty International. Another notable point is that most reports state that confessions obtained by torture were frequently used in courts. This practice is strictly forbidden by law in Saudi Arabia. In the end, both the amount of text and cases devoted to torture, as well as strong signal words, provide evidence that torture was practiced frequently in both 2007 and 2008. This result provides support for hypothesis 1A. In 2007 and 2008, Saudi Arabia was ruled by king Abdullah bin Abd Al‑ Aziz Al Saud (US State Department (2007). The King came to power in 2005. In his first years of king, he initially spoke out in favour of reforms. He announced to evaluate the status of women and religious minorities and allowed more media freedom (Human Rights Watch, 2015). Also, under his rule the National society for human rights was formed. In order to assess whether he has made a cost-benefit analysis on violating human rights, we need to know what the costs and benefits actually were in these years. The costs of benefit can be derived from the amount of taxes that Saudi-citizens have to pay. When these are higher, the costs for repression are also high. For the years of 2007 and 2008 there is no data for Saudi
Arabia. The amount of taxes in 2004 was 1,37% of GDP, and 2010 this was 1,27%3. Only
public taxes are relevant for this research. These taxes are only several small taxes for companies and expats; Saudi residents are exempt from paying any taxes at all (Santander, 2019.). Thus, the costs of violating human rights are absent. The benefits of violating rights are to stay in power and oppress dissent. In 2008, there were worldwide protests over high fuel prizes and rising inflation. Including Saudi Arabia, where inflation rose over 6% in 2008 after being almost zero in the past decade. This has sparked public protests and boycotts in the country (Worth, 2008). With a potential crisis lying ahead, Saudi Arabia decided to increase oil output. The king organized a conference of oil producers to discuss the high oil prices. It was decided that they would increase oil production, thereby hoping to decrease oil prices and avert public unrest (Mouawad, 2008). This proves that there was unrest in Saudi Arabia in 2007-2008 and that the government was worried about it. Thus, there would be benefits when repressing and violating rights.
To what extent these costs and benefits are directly linked is hard to say. Saudi Arabia has increased oil-production to avert protests, but it is unclear whether this has had any influence on rights violations. Nevertheless, the costs of violating rights seem to have been lower than the benefits, making it at least plausible that there is a causal connection between the two. Unfortunately, hard evidence for this has not been found. One can therefore conclude that the oil-dependency theory applies to Saudi Arabia in 2007-2008, thereby supporting hypothesis 1A. The presence of a causal mechanism is plausible, but there is no evidence that a cost/benefit analysis actually took place. Hypothesis 2A is thus neither confirmed nor invalidated.
2015-2016
In 2015 and 2016 oil production was lower than in 2007-2008, with 22,7% in 2015 and 26,4% in 2016. This period is interesting since ‘Vision 2030’ has been released in 2016. To see if the release of this document has made any difference, I will discuss 2015 and 2016 separately and compare them later.
For 2015 it was expected that torture will be practiced occasionally and therefore practiced less compared to 2007-2008. Reports on human rights differ in their account on
3United Nations University (2019). Saudi Arabia total taxes. Retrieved from:
human rights, but in general torture is practiced at least occasionally4. Two reports state that
confessions obtained by torture were frequently used in court and Amnesty claims that torture is widespread and common. Based on this, torture is practiced frequently in 2015. However, the US State report states that physical torture is uncommon in prisons and non-physical torture sometimes used. Human Rights watch does not have a section on torture and only mentions that some prisoners are beaten by guards. Physical torture mostly included beatings, whilst non-physical torture included physiological abuse and withholding food and medicine. This differing account makes it difficult to assess the practice of torture. Overall it seems that torture is at least practiced occasionally, but it would be more fitting to label it occasionally to frequent practice. This means that there is some support for hypothesis 2A, albeit not very strong.
In 2015, the leadership has changed. King Salman has taken over from King Abdullah who died early in 2015 (Human Rights Watch, 2016). Although Salman has initiated several government reforms, he mostly continued on the lines that King Abdullah set out (BBC, 2015; Human Rights Watch, 2016). To see whether the new king made a cost/benefit analysis on human rights violations, we again need to identify the costs and benefits. Tax levels in 2015 were slightly higher than in 2007-20085. Again, this is only corporate tax, so public tax levels are still zero in 2015, making the costs of repression low. In 2015, the economy steadily declined in Saudi Arabia. As we know, income from oil declined. Saudi Arabia had growing budget deficits and there were fears that they could no longer uphold social spending (Barnato, 2016). This was a potential danger for the Saudi-government, since it could lead to unrest and protests. Especially in the east of Saudi Arabia, where mostly minorities live. There were fears that once protests started here, they would spark protests al over the country (BBC, 2014). Based on this, the benefits of violating rights were relatively high. This means that the benefits were higher than the costs. It is thus plausible that Saudi Arabia decided to violate rights based on weighing these costs and benefits. This also fits the outcome that levels of rights violations have not really declined. Still, as in 2007-2008, no hard evidence has been found that directly linked oil dependency with human rights violations. This means that torture is practiced slightly less frequent in 2015, providing some support for hypothesis 2A. There is evidence that a cost/benefit analysis could have led to this outcome, but hard
4 US State Department, 2016, Amnesty International (2016), Human Rights Watch (2016)
5United Nations University (2019). Saudi Arabia total taxes. Retrieved from:
evidence is missing to fully support hypothesis 2B.
In 2016, oil rents were higher than in 2015 but still higher than in 2015 but still much lower than in 2007-2008. It was therefore expected that there would be relatively less cases of torture in 2016 than in 2007-2008. Overall, human rights reports did not differ much from 2015. Torture is still common, but there are differences between the reports. All three reports note that prisoners are frequently convicted based on confessions obtained by torture6. Amnesty further states that torture in general is widespread. The US state department report again emphasizes that physical torture was uncommon, whilst non-physical torture is sometimes practiced. Physical torture included beatings, non-physical torture included psychological abuse. This provides the same problem as in 2015: on the one hand torture seems to be widespread, but other parts of the reports hint to a less frequent use of torture. Therefore 2016 will receive the same label as 2015: torture is practiced occasionally to frequently. This means that there is again some support for hypothesis 2A, although it is not very strong.
In 2016, not much has changed politically since 2015. Salman is still king, although his son Mohammed Bin-Salman takes up a more prominent position. He responded to the declining income from oil and rising dissent in 2015 by releasing ‘Vision 2030’. In this document Saudi Arabia proposes several reforms of the economy. Income from oil is
declining. to make sure economic prosperity is secured in the future, Saudi Arabia announced to reform to become less dependent on oil (CEDA, 2017). In order to see whether this might have influenced a possible cost/benefit analysis, we need to know the costs and benefits. Tax levels in 2016 are the same as in 2015, leaving the costs of repression unchanged. Social unrest and protests continued to be an issue over 2016. In the beginning of the year, Saudi Arabia executed a sharia cleric, leading to protests in Eastern Saudi Arabia (McVeigh, 2016). Also, many civilians signed a petition asking for more rights for women, after years of protest against the strong laws for women (Sidahmed, 2016). Thus, the benefits for repression were again relatively high. This means that the benefits of repression again outweigh the costs. This aligns with the observation that repression is still relatively high. This means that it is possible that this cost/benefit analysis has been made. However, no hard evidence has been found for this. In sum: the occurrence of torture is only marginally lower than in 2007-2008, providing little support for hypothesis 2A. It is plausible that a cost/benefit analysis has been made, providing some support for hypothesis 2B. If one compares 2015 to 2016, it is seen that
the introduction of Vision 2030 has not made a difference. There is almost no difference in the practice of torture between 2015 and 2016. Therefore, there is no support for Hypothesis 3.
Conclusion and Discussion
In the previous sections, results were drawn from the data to support the hypotheses. In this section I will bring the hypotheses together and answer the main question of this thesis. After that I will draw several conclusions and discuss issues and implications arising from this research.
Results and conclusion
Beginning with 2007-2008, it is seen earlier that oil revenues as a percentage of GDP in Saudi Arabia are relatively high. In both 2007 and 2008 torture was practiced frequently in Saudi Arabia, both physical and non-physical torture. This is in accordance with hypothesis 1A. This confirms the theory that higher oil dependency and violations of human rights are related. The assumed causal mechanism was also examined. The benefits of violating human rights were higher than the costs, resulting in the expected outcome of high levels of torture. This means that it could have taken place, although there is no evidence it actually did. This partly confirms hypothesis 1B. In 2015 and 2016 oil revenues were much lower. This did not result in the expected results. Torture seems to be practiced less frequently, but is still
relatively frequent. The causal mechanism, on the other hand, does align with this result. The benefits of repression were higher than the costs in both 2015 and 2016. This means that the causal mechanism is at least plausible, partly confirming hypothesis 2B. There was no
difference in use of torture between 2015 and 2016, thus there is no support for hypothesis 3. There is one issue that has not been resolved yet. The proposed causal mechanism seemed to work and yielded logical outcomes. Nonetheless, these outcomes did not
completely align with the oil dependency theory. In 2015 and 2016, oil rents were lower than in 2007 and 2008. According to the theory, this should have led to less use of torture in 2015 and 2016. However, torture is still widespread and common in 2015 and 2016. There seems to be a small decline, but overall use of torture is still occasional to frequent. How then
Can we explain that income from oil changed, but the use of torture remains relatively stable? In the results section, it became apparent that Saudi Arabia had to deal with protests and public dissents in both 2007-2008 and 2015-2016. In both time periods, they reacted by
changing the production of oil. In 2008, oil prices were high. Saudi Arabia increased oil production, hoping to decrease oil prices and please the public. In 2015, income from oil declined, leading to budget deficits and worries of public unrest. The government reacted in 2016 by publishing vision 2030, announcing reforms to become less dependent on oil. Based on these narratives, it seems that there is a different relation between oil and rights violations. Instead of repressing public unrest and violating human rights, Saudi Arabia chose to reform the economy and increase or decrease the dependency on oil. In 2008 this led to a higher output of oil to dampen oil prizes, in 2016 this led to less dependency on oil when oil revenues declined. When necessary, the state can still repress and violate human rights. Repression is then independent from economic factors. This explains why income from oil changed but the use of torture remained constant. An additional problem is how to explain why the causal mechanism still lined up with the results, when oil dependency and repression are not linked? The answer lies in the definition and operationalisation of the costs of
violating human rights. These were seen as lost revenue from citizens. This was
operationalised by the amount of public taxes. In Saudi Arabia there are no public taxes, meaning that the costs are always nihil. Thus, the costs are always lower than the benefits. Since rights violations are constant, the cost/benefit analysis always leads to high repression in the case of Saudi Arabia.
The main question of this thesis was whether there is a causal relationship between dependency on oil and violations of human rights. The answer difficult to give. As is seen, the results show that the causal mechanism can work. However, it seems like the oil dependency theory does not really suit the case of Saudi Arabia. In 2007-2008 the expectations of the theory lined up, but not for 2015-2016. Therefore, the causal mechanism cannot really be proven; if the theory does not work, the causal mechanism also does not work.
Critical remarks and implications
It is not reasonable, however, to completely disregard the oil dependency theory. Several critical remarks can be made that might have influenced the outcomes of this research. First of all, there might be issues with the operationalization of oil dependency and human rights violations. One can ask the question whether the difference of oil dependency between the cases is enough to result in a difference in the use of torture. The highest oil revenues were 53% in 2008, the lowest 22,6% in 2015, a difference of 30,4%. One can do research with cases that have a bigger difference in oil revenues. It is possible that this will yield a difference between the use of torture in the time periods. Another problem is the use of
torture in this research. Torture is mostly used behind closed doors, such as in prisons and during interrogations. This makes it difficult to see whether a decision to torture is based on a cost/benefit analysis, since the decision to torture is made behind closed doors. Moreover, most people are tortured by prison guards and other officials. They usually decide to use torture, not the king or the state. The king can influence this, by prosecuting state officials who are accused of torture, or deciding not to do anything about alleged cases of torture. This type of influence has not been studied in this research, possibly altering the outcomes.
Another issue with torture was the data. The Human Rights reports differed in their focus. The 2015-2016 reports contained less info on torture and mistreatment than the 2007-2008 reports. This can mean that torture is less used in these years. However, it could also mean that the reports focused on other issues that had more priority at the time. Thus, the way one interpreted can influence the results.
Secondly, there might have been problems with assessing causality. I have already discussed the problem of using public taxes as an indicator for the costs of repression, especially in the case of Saudi Arabia. Another indicator for this that goes beyond monetary costs might be more suitable. This could, for instance, consider lost productivity from a worker who is being tortured in prison. For the benefits of repression, I looked at the extent at which the regime’s survival is being threatened. In the theory, the benefits of repression are seen as the ability to remove threats to the regime. I thus looked at whether there were protests and uprisings threatening the regime. The pitfall of this indicator is that it provides the benefits of oppressing uprisings, not necessarily to torture these people in prisons. Thus, choosing another indicator for this could lead to different results. A final word on causality is that , as said before, it is very difficult to determine it. I have not been able to find evidence for an actual cost/benefit analysis. And even if I did, the Ceteris Paribus rule would not have been met. There are undoubtedly factors that influenced the relation between oil dependency and violations of human rights that I have missed.
Nonetheless, this research has proven that, although difficult, it is possible to study the causal mechanism of a theory, thereby filling a gap in the theory. The results from this
undertaking shed a new light on the oil-dependency theory, since it does not seem to apply to Saudi Arabia. This might be due to factors in this research, as I have said earlier. Future research could take up these points to see whether this leads to different result. They could for example focus on other years with even lower oil revenues or use other operationalisations for the cost/benefit analysis. Also, future studies could perform field studies to obtain better data, for example interviews with cabinet members. Assessing causality is then more likely be
possible. Also, it is difficult to generalise these results beyond this single case. Preferably, the causal mechanism will be applied to more cases to get a better overview on whether the mechanism works.
Besides these academic contributions, this research also has implications for society in general. This research contributes to the question of why nations violate the human rights of their citizens. By getting more insight in how decisions are made regarding human rights, we get more knowledge on how to deal with rights violations. It will become possible to identify key moments in a decision-making process where can be intervened to stop violations of human rights. This research has contributed in identifying this process. In general, it is
another peace in the puzzle of why states violate human rights and brought us a little closer to understanding why the law of coercive responsiveness is as everlasting as it is.
Bibliography
Amnesty International. (2008). Amnesty International Report 2008. The state of the world’s
human rights. Retrieved from:
https://www.amnesty.org/en/documents/pol10/001/2008/en/
Amnesty International. (2009). Amnesty International Report 2009. The state of the world’s
human rights. Retrieved from:
https://www.amnesty.org/en/documents/pol10/001/2009/en/
Amnesty International. (2016). Amnesty International Report 2015/16. The state of the
world’s human rights. Retrieved from:
https://www.amnesty.org/en/documents/pol10/2552/2016/en/
Amnesty International. (2017). Amnesty International Report 2016/17. The state of the
world’s human rights. Retrieved from:
https://www.amnesty.org/en/documents/pol10/4800/2017/en/
Barnato, K. (2014, 13 January). Could Saudi Arabia’s austerity spark social turmoil? Retrieved from:
https://www.cnbc.com/2016/01/13/could-saudi-arabias-austerity-spark-social-turmoil.html
British Broadcasting Corporation (2015). Saudi Arabia’s King Salman unveils cabinet
reshuffle. Retrieved from: https://www.bbc.com/news/world-middle-east-31052071
British Broadcasting corporation (2014). Reporting Saudi Arabia’s hidden uprising. Retrieved from: https://www.bbc.com/news/world-middle-east-27619309
Bennet, A. & Checkel, J. (2015). Process tracing: from metaphor to analytic tool. New York: Cambridge University Press.
Bueno de Mesquita, B., Morrow, J.D., Siverson, R.M. & Smith, A. (2003) The Logic of
Political Survival. Cambridge, MA: MIT Press.
Bueno de Mesquita, B., & Smith, A. (2009) Political Survival and Endogenous Institutional Change. Comparative Political studies, 42(2), pp. 167-197.
Cingranelli, D.L. & Richards, D.L. (2013). The Cingranelli-Richards (CIRI) human rights
Data project coding manual. Retrieved from:
http://www.humanrightsdata.com/p/data-documentation.html
Cole, W.M. (2015). Mind the Gap: State Capacity and the Implementation of Human Rights Treaties. International Organization 69(2), pp. 405-441.
Conrad, C.R., Hill Jr, D.W. and Moore, W.H. (2018). Torture and the limits of democratic institutions. Journal of Peace Research, 55(1), pp. 3-17.
Conrad, C.R. & DeMerrit, J.H.R. (2012). Constrained by the bank and the ballot: Unearned revenue, democracy, and state incentives to repress. Journal of Peace Research, 50(1), pp. 105–119
Council of economic and development affairs (2017). Vision 2030. Retrieved from:
https://vision2030.gov.sa/en
Davenport, C. (2007). State repression and political order. Annual review of political science, 10(1), pp. 1-23.
DeMerrit, J.H.R. & Young, J.K. (2013). A Political Economy of Human
Rights: Oil, Natural Gas, and State Incentives to Repress. Conflict Management and
Peace Science 30(2), pp. 99-120.
Egorov, G., Guriev, S. & Sonin, K. (2009). Why Resource-poor Dictators Allow Freer Media: A Theory and Evidence from Panel Data. American Journal of political science, 103(4), pp. 645-668.
Gerring, J. (2005). Causation. A unified framework for the social sciences. Journal of Theoretical Politics 17(2), pp. 163–198.
Griffin, J. (2002). First Steps in an Account of Human Rights. European Journal of
Philosophy, 9, pp. 306-327.
Hafner-Burton, E.M., Tsutsui, K. & Meyer, J.W. (2008). International Human Rights Law and the Politics of Legitimation: Repressive States and Human Rights Treaties. International Sociology 23(1), pp. 115-41.
Hopkins, N., Kirchgaessner, S. & Shaheen, K. (2019, 31 March). Leaked reports reveal severe abuse of Saudi political prisoners. Retrieved from www.theguardian.com.
Human Rights Watch (2008). World Report 2008. Retrieved from:
https://www.hrw.org/world-report/2008
Human Rights Watch (2009). World Report 2009. Retrieved from:
https://www.hrw.org/world-report/2009/country-chapters/saudi-arabia
Human Rights Watch (2016). World Report 2016. Retrieved from:
https://www.hrw.org/world-report/2016/country-chapters/saudi-arabia
Human Rights Watch (2017). World Report 2017. Retrieved from:
https://www.hrw.org/world-report/2017/country-chapters/saudi-arabia
McVeigh, T. (2016, 3 January). US warns Saudi Arabia’s execution of prominent cleric risks inflaming sectarian tensions. Retrieved from:
https://www.theguardian.com/world/2016/jan/02/suadi-arabia-cleric-execution-unrest-predicted-shia-areas
Mouawad, J. (2008, 14 June). Plan Would Lift Saudi Oil Output. The New York Times. Retrieved from:
https://www.nytimes.com/2008/06/14/business/14oil.html?rref=collection%2Ftimestopic%2F Saudi%20Arabia
Peterson, T.M. (2015). Export Diversity and Human Rights. Journal of Conflict Resolution,
61(8), pp. 1740-1767.
Poe, S.C., Tate, C.N. & Keith, L.C. (1999). Repression of the Human Right to Personal Integrity Revisited: A Global Cross-National Study Covering the Years 1976-1993. International Studies Quarterly, 43(2) pp. 291-313.
Ross, M.L. (2001). Does Oil Hinder Democracy? World Politics, 53(3), pp. 325-361.
Ross, M.L. (2004). What Do We Know About Natural Resources and Civil War? Journal of
Peace Research, 41(3), pp. 337–356.
Ross, M.L. (2008). Oil, Islam, and Women. American Political Science Review, 102(1), pp. 107-123.
Sachs, J.D., & Warner, A. 1995. Resource abundance and economic growth. Saudi Arabia: Tax System (2019) Retrieved from:
https://en.portal.santandertrade.com/establish-overseas/saudi-arabia/tax-system
Sidahmed, M. (2016, 26 September). Thousands of Saudis sign petition to end male
guardianship of women. Retrieved from:
https://www.theguardian.com/world/2016/sep/26/saudi-arabia-protest-petition-end-guardianship-law-women
Smith, A. (2008). The Perils of Unearned Income. The Journal of Politics, 70(3), pp. 780 793.
Teitelbaum, J., Ochsenwald, W.L. & St. John Bridger Philby, H. (2019). Saudi Arabia. In Encyclopaedia Britannica. Retrieved from:
https://www.britannica.com/place/Saudi-Arabia
Tilly, C. (1985). War Making and State Making as Organized Crime. In P.B. Evans, D. Rueschemeyer and T. Skocpol (eds.). Bringing the state back in. Cambridge: Cambridge University Press.
Toshkov, D. (2016). Research design in political science. New York: Palgrave Macmillan.
United Nations. (1984). Convention against Torture and Other Cruel, Inhuman or
Degrading Treatment or Punishment. Retrieved on 17-06-2019 from:
https://www.ohchr.org/EN/ProfessionalInterest/Pages/CAT.aspx
US State Department. (2008). 2007 Country Reports on Human Rights Practices: Saudi
Arabia. Retrieved from: https://2009-2017.state.gov/j/drl/rls/hrrpt/2007/100605.htm
US State Department (2009). 2008 Country Reports on Human Rights Practices: Saudi
Arabia. Retrieved from:
https://20092017.state.gov/j/drl/rls/hrrpt/2008/nea/119126.htm
US State Department (2016). Country Reports on Human Rights Practices for 2015: Saudi Arabia. Retrieved from:
https://20092017.state.gov/j/drl/rls/hrrpt/humanrightsreport/index.htm?year=2015&dlid=2529 45
US State Department. (2017). 2016 Country Reports on Human Rights Practices: Saudi
Arabia. Retrieved from:
https://www.state.gov/reports/2016-country-reports-on-human-rights-practices/saudi-arabia/
Vogel, F.E. (2012). Sharia in the politics of Saudi Arabia. The review of faith and
international affairs, 10(4), pp. 18-27.
Wigley, S. (2018). Is There a Resource Curse for Private Liberties? International studies
quarterly, 62, p.834-844.
Wilson, R. (2004). Economic Development in Saudi Arabia. London: Routledge.
Worth, R.F. (2008, 25 February). Rising inflation creates unrest in Middle East. The New
York Times. Retrieved from:
https://www.nytimes.com/2008/02/25/world/middleeast/25economy.html?rref=collection%2F timestopic%2FSaudi%20Arabia
Wright, J., Frantz, E. & Geddes, B. (2013). Oil and autocratic regime survival. British
