Introduction
As the world has become interdependent, so has the international economy. Linkages across the world's economies have grown more robust, as shown by indicators of economic globalization, which include the level of foreign direct investment (FDI), portfolio investments, and open trade policies.Footnote 1 Some researchers have argued that these factors have negatively influenced human rights and induced increased state repression.Footnote 2 Most of these authors consider the impact of foreign investment on the general human rights condition of the host state, as opposed to the state's reaction to contentious politics in particular. The public nature of protest and the government's response to protest are more public displays of the civil and political rights conditions of the state. The government's reaction to political contention fits better with the structural dependence theories that most of these arguments are based on, as protests provide a unique opportunity to measure government restraint in light of international scrutiny. While research has looked at government respect for human rights more generally, the causal mechanism places much emphasis on institutional change and the public spotlight. Research has not paid as much attention to public forms of repression or measured how globalization might influence this type of behavior by government officials. This article addresses one aspect of the globalizing economy, namely, FDI. I examine the effect of FDI stock on the violent repression of protest by host governments, using data drawn from 145 countries from 1990 to 2020.
FDI refers to the procurement of physical resources or stocks in a different country by international firms, giving them administrative control.Footnote 3 The number of companies with foreign affiliates increased from 63,000 in 2000 to 79,000 in 2007. Measured against the monetary value of goods and services produced—that is, gross domestic product (GDP)—the increase in FDI has been phenomenal. Among developing countries, FDI increased from 13 percent of GDP in 1980 to about 30 percent of GDP as of 2012.Footnote 4 By 2013, FDI stock made up 34 percent of global GDP.Footnote 5 The world's economy has grown continuously since the end of the European colonial era in the mid-twentieth century.
The economic interdependence that increased foreign investment represents has received its fair share of criticism, particularly in terms of human rights and state repression. There have been issues with multinational corporations (MNCs) seeking out and investing in countries where governments continue to violate human rights. The relationship between state repression and foreign investment has been explored using structural dependence theories, which accentuate the influence of investors on the government.Footnote 6 It is thus common for oil companies or other MNCs to face international shaming for inducing or encouraging state repression in their host communities.Footnote 7 An example of a prominent case linking MNCs to incentives for violent government repression is the oil companies located in the Niger Delta region of Nigeria.Footnote 8
Another example is the conflict between Peruvian indigenous groups and the foreign mining companies in their communities, particularly during the regimes of Alberto Fujimori and Alan García, from the late 1980s through the 1990s.Footnote 9 Most research questioning the actions of foreign investors has relied on qualitative case studies. Still, quantitative cross-country research has not generated a consensus on whether FDI leads to increased state repression of dissenting movements.
In exploring the relationship between FDI and the way that states use repression to respond to public protests, it is crucial to begin by posing the following question: do foreign investors prefer the governments of their host states to be violently repressive? Scholars have written extensively on the relationship between international economic relations and state repression and have come up with differing conclusions. Some researchers argue that the relationship is one in which international economic links increase repression,Footnote 10 while others have argued the opposite;Footnote 11 some have even claimed that there is no relationship at all.Footnote 12 Sorens and Ruger, for example, do not find any evidence that FDI affects state repression in any way.Footnote 13 Amid this contention, one question remains: does FDI inflow influence a state's decision to use repressive tactics in the face of overt dissent? State repression can be institutional or systemic; in these cases, repression does not easily change and might go unreported. While research has considered the repression of protests, not much attention has been given to how government violence against protesters may be affected by dependence on foreign capital. This article examines that relationship.
This work builds on the research on violent repression. However, it deviates from other studies as it looks at the effect of dependence on foreign capital on government response to public protests. A protest presents a public and sometimes volatile act of political dissent. However, it also puts a spotlight on the government and its activities. How the government reacts to protests could reflect its ability to reduce escalation and test its protection of personal integrity rights. In the event of a protest, violent repression has a bigger audience. Therefore, the repressing government is open to more criticism and external reaction. How much do economic concerns play into the government's reaction in such situations? The government might want to remain in power, but its dependence on international economic ties could threaten its economic well-being.
I investigate the relationship between economic globalization and government repression in the face of public protests. As a measure of the influence of foreign investors, I examine how much a country relies on inward FDI stock and how this affects the government's use of violent strategies. I argue that because of the fear of violent conflict and general international pressure on foreign investors, investors would prefer that the government not use violence in times of nonviolent civil resistance. Thus, I argue that with higher levels of dependence on FDI inflow, foreign investors have more influence on the government. This will make the government less likely to use violence against protesters.
Protest and government repression
Scholars have long treated protest and repression as two sides of the same coin. This argument considers dissent as a reaction to repression or repression as a reaction to dissent.Footnote 14 In this work, the concern centers on the government's decision to repress after protests have broken out. Most researchers have considered this from the perspective of the costs attached to using repression, as well as the benefits the government can gain from using repression. If the cost of using repression outweighs its benefits, one would expect the government to shy away from repressive responses.
Gartner and Regan argue that the decision to use repression depends upon the level of threat that the government feels from the dissenting groups.Footnote 15 However, the cost of using repression restrains the government from doing so. Costs here could be domestic or international. Democratic institutions, for instance, increase the price of repression, and pressure from the broader international community could constrain the government in applying repressive techniques in response to dissent. The causal relationship can be considered from actual agency constraints or backlash effects.Footnote 16 Democracies, for instance, are less likely to get away with repression domestically, so they are less likely to use repression in response to a protest.Footnote 17 In the same vein, we can consider international reactions to state repression and see potential condemnation as raising costs. The government thus makes decisions in ways that are most beneficial in the end, or at least it makes the decision that it has the means to implement.
As mentioned earlier, the inherent characteristics of the political regime in the country can affect its citizens’ decision to protest. Regime characteristics have also been observed to influence governments’ decisions to repress protests. These characteristics provide the information that is the basis for both the protesters and the government's decisions.Footnote 18 When the government is a stable autocracy, repression is available and at the government's disposal. In democracies, violent repression is less of an option. As a result, autocracies will be less likely to experience protests, and generally there will be fewer cases of violence against protesters. Because of institutional constraints, democracies are also less likely to experience violence against protesters.Footnote 19 Pseudo democracies, on the other hand, are most likely to end up using repression against protesters because there is less information about the behavioral norms of the government.Footnote 20 The mixture of institutional characteristics in such governments makes them less predictable, making violence more likely as there is a less precise definition of the constraints of government and the rule of law as it pertains to liberal rights. In other words, both sides misconstrue the other's preferences, and strategies based on imperfect information lead both to protests and to violent repression.
One factor that may influence the occurrence of acts of repression could be a product of prior interactions. Moore argues that previous behavior could be a source of information.Footnote 21 The government substitutes tactics between accommodating rights and repression based on which tactic was met with dissent. Since the government wants to maintain political order or control, what is essential is that there is no dissent. The government can deal with this by measuring the cost of the tactics it uses against the benefits. The cost-benefit analysis considers the government's use of repression to maintain political survival and deal with political threats while dissidents protest to effect political change.Footnote 22 For instance, Ritter posits that repression will likely increase as the danger to political survival grows. Stable forms of government are less likely to introduce repression. However, when a government is very secure and uses repression, it will likely use more severe forms of repression.Footnote 23 This moves slightly from the perceived threat and constraints arguments to consider repercussions. Ritter argues that when the government has a firm hold on its power, it can get away with more repression; therefore, if it uses repression, it will use more severe forms of repression.Footnote 24
There are also more nuanced characteristics of government that could regulate how or when repression is used. In their analysis of African states, Hendrix and Salehyan find that though identity-based contention is most threatening, when the military is factionalized, it becomes less likely that the government will use repression to deal with such issues.Footnote 25 This is because the agency charged with repressing the contentious event violently is absent. Divisions in the military along ethnic or religious identity lines make it difficult for the military to be an agent of repression in such situations. Therefore, aside from the threat level, other domestic realities of the regime that go beyond democratic institutions could affect the decision to repress. These characteristics could make it difficult for the government to use repression even if it has the political will to do so.
Researchers have successfully created a bedrock for analyzing the factors that influence a government's decision to use repression more generally. Their argument points to factors that can be summed up as the constraints on the government's ability to use repression and the threats to the government. These arguments create an avenue to consider how much external factors could increase the cost of repression for the government or how they might increase the benefits of state use of repression. Aside from domestic institutions, other factors may influence a government's decision-making even in times of domestic unrest. International organizations, allies, and superpowers have been shown to influence states greatly, and more recently, attention has been drawn to firms’ influence.
Foreign investment and state repression of human rights
Debates on brutal business interests posit the conventional wisdom that repressive governments are best for business, but most researchers argue to the contrary. Blanton and Blanton argue that normative principles, as well as business practicalities, make repression a disincentive for investors, both directly and indirectly.Footnote 26 There have been calls for corporate responsibility, and the international system is increasingly able to sanction and shame firms that are complicit in human rights abuses in a way that may affect stock value.Footnote 27 It is thus only practical that firms are more likely to want the countries they invest in to have better human rights records. Therefore, firms are more likely to invest higher amounts in states that protect political and civil rights to avoid any future backlash or instability.Footnote 28 Though they find differences across economic sectors, Blanton and Blanton still find that human rights represent the most consistent sociopolitical predictor of foreign investment across industries.Footnote 29
However, a few authors regard foreign investments as inversely related to rights. Adam and Filippaios argue that foreign investors are attracted to the repression of civil rights as it reduces challenges to the management of firms, and they would prefer such a business climate.Footnote 30 Though they identify a positive relationship between the protection of political liberties and FDI, there is a negative relationship between the protection of civil liberties and FDI. Taking a step back to consider a broader causal path, Timberlake and Williams look at the relationship as more indirect.Footnote 31 According to them, dependence on FDI increases political exclusion, and where political exclusion thrives, so does the use of negative sanctions domestically against citizens. Azarvan is also pessimistic about this relationship and counters what he considers the neoliberal perspective of FDI.Footnote 32 Using the case of Algeria, Azarvan argues that FDI indirectly induces state repression. The accumulation of economic wealth by the government increases its ability to fund its repressive infrastructure. As for external sanctions and condemnation, Azarvan theorizes that trade relations and Western dependence on oil producers make international pressures less likely, especially pressures from the United States. Azarvan's work considers the global dependency on specific resources as reducing their power to influence the owners of such resources.
Studies of firm preferences have shown another side of the relationship. Firms consider some risks before they decide to invest; these include purely financial and economic risks, but also political risks.Footnote 33 Violent repression of civil resistance raises the risk of conflict. For instance, Blanton and Blanton cite the cases of Shell in Nigeria and Chevron in Sudan.Footnote 34 In both cases, repression made MNCs the targets of political violence by the citizens. Another prevailing shift is the call by the current international regime for firms to consider human rights in their spheres of influence.Footnote 35 This creates a picture that foreign firms are increasingly likely to be uncomfortable with public shows of state repression by their host governments. This forms the basis of our framework for explaining the impact of foreign investment on the domestic behavior of the government in light of public unrest.
Dependence on foreign investment and state repression
The direct relationship between foreign investors and state repression draws mainly from structural dependence theory. Structural dependence theory is built on the argument that societies are dependent on the owners of capital, and how these individuals choose to invest has far-reaching consequences on the society.Footnote 36 Thus, state leaders must consider the impact of their policies and actions on these owners of capital because the state and all its citizens are directly or indirectly dependent on them. In other words, the government likely wants to please these investors.
The literature on FDI has established that the inflow of foreign investments is desirable for most states.Footnote 37 FDI has been linked to prospects for economic growth and development, and governments of prospective host nations go to great lengths to attract foreign investors.Footnote 38 Political risks create a disincentive for foreign firms by creating an added cost to doing business in such a country.Footnote 39 Where there are protests, the threat of escalation is more apparent, and this perceived threat increases if the government uses violent repression.Footnote 40 The escalation risk is even higher when the government is already repressive.Footnote 41 If we assume that this is the understanding of the host government, it would likely seek to avert any form of violent conflict as this may cause foreign firms to disinvest or scare potential investors away.
Since the 1980s, foreign investors have been put under another sort of spotlight that forces them to consider state repression unfavorably. The current practice of regime and corporate shaming puts firms in an uncomfortable position, as it is increasingly difficult for them to hide their heads.Footnote 42 Firms that are seen to be complicit in human rights violations or firms that are seen to be supporting an overtly repressive regime could be shunned internationally or at home. An example of this is Talisman Energy in Sudan, which was shamed and pressured by shareholders into making a decision to exit the country in 2002. The company's decision to disinvest came amid significant profits that they had secured during their four years of operation in Sudan.Footnote 43 Calls for companies to disinvest also played a role in the fall of the repressive apartheid regime in South Africa. In the case of South Africa, it is argued that firms experienced a rise in their stock value in the aftermath of announcing their decision to disinvest.Footnote 44 Even then, in the 1980s, when human rights norms were not as universally supported as they are today, firms could still suffer backlash for investing in repressive states. The case of US firms in South Africa also points to the economic repercussions of remaining with overtly repressive regimes. This case interestingly also shows the gains of walking away. Some countries, such as Norway, even have a mechanism to punish companies that are complicit in human rights abuses abroad. The Norwegian government's pension fund has an exclusionary mechanism in its ethical guidelines, which serves as a punishment for firms that fall short of human rights norms. Foreign firms thus increasingly do not only consider the stability of the business environment in their decision whether to invest or not, but they also have an incentive to consider international human rights norms.Footnote 45
When a government opens itself to host foreign investments, it ushers in new players in its domestic policies. As a country is able to secure more foreign investors, the government becomes more dependent on them. An increase in foreign investment also shows the government's willingness to open its economy to foreign investors. Since the state is dependent on FDI inflow, the host government will want to at least consider the preferences of investors and their perceptions of the country as a business environment. The host government will be more likely to maintain order without escalating the conflict, as escalation would create a disincentive for businesses to keep investments in the country. States will also shy away from overt repression, as this would shine a light on them and put pressure on foreign investors. The host government will not want to be seen as overtly oppressive so as not to exacerbate the fears of the firms.
Based on these arguments, when the government is dependent on foreign investments, it is expected that the relationship will be one that reduces overt forms of repression. So, where there are protests, the government will be less likely to use violent repression in response to protests. This is because the government will become more sympathetic to political and civil rights and because such acts shine a spotlight on the government. So, even for governments that might want to choose repression, the overt nature of protests and the attention given to how the government responds will create a disincentive for violent repression in such cases.
The arguments are based on two assumptions. First, states are aware of the disincentives, and so are investors. Second, states find FDI to be attractive, and they prefer higher levels of FDI inflow and fear disinvestment. The relationships and the public nature of protest behavior and international reactions make a choice to use violent repression in these circumstances costly. Based on these arguments, this article proposes that more foreign investment will lead to less use of violent repression against protesters.
Data and measurement
To investigate the arguments put forward, I rely on a few datasets. For the data on protests, I rely on the Mass Mobilization (MM) dataset. The MM dataset, funded by the US government, has records of protests covering 168 countries from 1990 to 2020 drawn from media coverage of the events. It includes all protest events with at least fifty participants. The MM dataset also codes for government responses, which is the source of the dependent variables.Footnote 46 For the analysis in this article, I collapsed the dataset to the country-year level. The unit of analysis is the country-year, and the sample is limited to years in which at least one protest took place.
Data from the United Nations Conference on Trade and Development (UNCTAD) was used for the independent variable. I use the UNCTAD measure of FDI stock as well as the World Bank's data on GDP to create the measure of FDI stock as a fraction of GDP.Footnote 47 Inward FDI stock as a percentage of GDP is my primary independent variable; it captures, to an extent, the dependence of a country's economy on inward FDI. UNCTAD measures FDI as an investment for a management interest with 10 percent or more voting stock. The World Bank dataset also provides the GDP, a measurement of a state's wealth in terms of the total value of goods and services produced within that country's borders. It represents the monetary value of all final goods and services produced by the country's residents and serves as an indicator of the country's economic performance. I use the log of this measure for the analysis.
The data on democracy come from the Varieties of Democracy (V-Dem) Project. V-Dem includes more than 30 million data points covering 202 countries from 1789 to 2021. With the involvement of more than 3,700 scholars and experts, V-Dem assesses hundreds of democracy-related characteristics.Footnote 48 All observations with missing data were dropped from the analysis.Footnote 49 Table 1 shows the descriptive statistics of the variables used for the main analyses in this article.
Descriptive statistics of variables of interest.
Dependent variables
The dependent variable is violent government repression of protest. I consider government repression in terms of violent, negative sanctions imposed by the government to counteract a perceived threat. In this article, government repression is operationalized to include two forms of violent, negative sanctions used by the government in response to protests. For this work, repressive violent government actions include two government responses: shooting at protesters and killing protesters. These two variables are measured as the share of protests in a given year in which the government used these sanctions. I use a count of the total number of times the government used these forms of repression in its response to protests in a given year and divide it by the total number of protests that year. Because the dependent variables are continuous, I use ordinary least squares (OLS) regressions to test my hypotheses.
Independent variable
For dependence on foreign investment, I use a measure of FDI as a fraction of GDP from the previous year, FDI(t–1). I use the FDI stock from the year before the protest as doing so helps mitigate threats to proper causal identification.Footnote 50 In other words, this refers to the ratio of inward FDI stock hosted in a country the year before the protests to its GDP over the same period, expressed as a percentage. This measure indicates the level of foreign investment in the country relative to the size of its economy. Though economies like the Cayman Islands have amassed FDI stock amounting to more than 100 times their GDP, the countries included in the analysis here range from 0 to 1967.75 percent. I would expect this to have a negative relationship with the dependent variables. This is to say that as the dependence on FDI stock increases, there will be a lower number of incidents in which the government used violent repression on protesters.
Controls
The level of democracy is based on the electoral democracy variable from the V-Dem dataset. It is a continuous variable ranging from 0 to 1 that considers the following factors: clean elections, elected executive, suffrage, freedom of association, and freedom of expression. I use this as a control variable, which is expected to have a negative relationship with the dependent variable, in line with arguments made in political science research. Seeing that we view democracies as being more respectful of human rights, most authors argue that they are less likely to use repressive techniques.Footnote 51 The division of power, competitive elections, and viable democratically elected political opposition all constrain the government's use of violent repression. In other words, democracies are less likely than nondemocracies to repress protests violently.
I also include a control for violence by protesters. I calculate the share of protests in which protesters used violence in a given year from the MM dataset to produce the protester violence variable. This ranges from 0 to 1, and this occurs in about a quarter of all protest cases in the data compiled for this work. The MM dataset records protester violence as any case in which the protesters used force against the authorities or destroyed public property. In line with extant arguments, a positive relationship can be expected, as more hostile activities are more likely to lead to violent repression, and the use of violence by protesters might reduce the backlash on the government if it decides to respond severely.Footnote 52
The analyses also include other control variables that researchers have argued predict violent government repression. I employ a control variable for the state's wealth as measured by the log of the GDP (log), population, and fuel exports calculated as a percentage of all merchandise exports. The World Bank's database provided the data for these three variables. Researchers have found that population and fuel resource dependence to be associated with an increased probability of violence. In contrast, a higher GDP is expected to be associated with a reduced likelihood of violence.Footnote 53
Researchers have found that protesters’ demands are fundamental in predicting whether the government will use violence.Footnote 54 The analysis thus includes a control for the share of the total number of protests that were protests against police brutality; protests against political policies or processes, which I label political protest; protests calling for the removal of an official; and protests against social restrictions, which I refer to as social rights protest. I expect these protest demands to experience more use of violence than other protest demands, such as tax or wage disputes, as they are more threatening. Protest demands data were drawn from the MM dataset. I also use the lagged version of the dependent variables in some models. Shooting at protesters(t–1) and killing protesters (t–1) are used to capture previous violently repressive responses to protest.
Results
The outcomes of the regression analysis are presented in Table 2. The two initial models examine the correlation between the independent variable and the frequency of government forces shooting at protesters. On the other hand, Models 3 and 4 illustrate the correlation with government involvement in the killing of protesters. All four models show a consistently negative relationship between dependence on FDI and the main independent variables (i.e., shooting at protesters and killing protesters). Models 2 and 4 include lagged versions of the dependent variables as control variables; they are lagged by one year, indicating the government's response to protests in the previous year. This is to capture the expected response from the government based on the share of protests during the last year that were met with violence from the government.
Results of violent repression, OLS regressions.
Notes: Standard errors in parentheses, clustered by country. * p < .1; ** p < .05; *** p < .01.
Even in these models, in line with my main argument, there are statistically significant negative relationships. The statistically significant negative relationship allows us to reject the null hypothesis, supporting the primary hypothesis that dependence on FDI inflow negatively affects a state's decision to use violent repression against protesters. Every unit increase in FDI as a share of GDP is associated with a decrease in the share of protests in which the government responds violently against protesters while holding the other variables in the models constant. The expected share of protests in which government forces shot at protesters decreased by 0.0000329 and 0.0000349 (i.e., 0.0033 percent and 0.0035 percent), respectively with each unit increase in the level of FDI dependence. The values are similar for killing protesters, as Models 3 and 4 show a decrease of 0.0000279 and 0.0000321 (i.e., 0.0028 percent and 0.0032 percent), respectively, with each unit increase in FDI(t–1). This would mean, all else being equal, that a country like Cyprus, whose dependence on FDI was close to 0 in the early 1990s, had almost no likelihood of such acts of government repression. By 2004, that FDI stock in Cyprus made up approximately five times its GDP. All things being held constant, this shift could lead to a significant change in the expected rate of violent government repression in times of contentious mass mobilizations.
For the control variables, the analysis shows some of the expected results as well. Regarding violence by protesters, the results show statistically significant and positive relationships between violent protest and government use of brutal repression. The results are positive and statistically significant in all four models. A higher number of cases of violent protest in a particular year corresponds with an increase in the share of annual protests in which the government shot or killed protesters. As the fraction of protests in which protesters used violence increased, this corresponds with a larger share of protests in which violent repression occurred. With each case of protester violence, the expected share of protests in which the government responded with violence increases by 0.156 and 0.167 for government shootings and approximately 0.135 and 0.109 for the killing of protesters.
The control for wealth is in the expected direction; richer countries, as operationalized by higher GDP, have a lower expected share of protests in which the government used violent repression. However, there is no statistically significant relationship in Model 3, indicating no statistically significant relationship with the killing of protesters, though we find a statistically significant relationship in Model 4, where we include a lagged version of the dependent variable, Killing protesters(t–1). The results are similar for the level of democracy, which shows statistically significant negative relationships, as expected, but no such relationship with the killing of protesters in Model 4. Also, The control for fuel exports only reaches statistical significance in Model 4. The evidence here also suggests that I cannot reject the hypothesis that a country's population has no bearing on whether a government will respond to protests with violence. As a robustness check, I include controls for the protester demands in Table 3.
Results of OLS regressions, including controls for protester demands.
Notes: Standard errors in parentheses, clustered by country. * p < .1; ** p < .05; *** p < .01.
Table 3 shows the results of the models; however, this time, controls are included for four types of demands in the protests. Including these variables does not change our results much. The level of democracy, however, loses statistical significance in Models 7 and 8, which deal with the killing of protesters. The results remain similar for the correlation between FDI stock and violent repression, and protester violence remains significant in all the models.
Regarding demands, when it comes to the number of protests calling for the removal of officials or protests against social restrictions, there is no statistically significant relationship with incidents of killing or shooting of protesters. However, as the number of protests against police brutality increases, so does the share of episodes with the government shooting at protesters, as seen in Models 5 and 6. Protests against political processes correlate with instances of shooting at protesters, but only in Model 6, where we control for the shooting of protesters in the previous year.
The results suggest that the analysis supports the central hypothesis. I also find supporting evidence for other variables included as controls based on existing research on violent government responses to protests. As foundational work in this field would suggest, I find evidence for the impact of democracy in two models. Still, I failed to reject the null hypothesis when controlling for previous violent responses from the government. Though the coefficients of the effect of FDI may seem relatively low, the results are consistent across all specifications of the models. The growing dependence on foreign capital to extents that exceed the size of GDP gives cause for optimism in a world where firms increasingly have globalized outlooks. The fact that it is a consistent predictor of state violence against protesters provides more support for arguments that point to the informal influence of foreign firms.
Conclusion
The growth in FDI and the globalization that can now be observed in the intra-firm activities of MNCs could spell changes in global and domestic political interactions. In this work, I consider the impact that increased dependence on FDI could have on state repression, and the results give cause for hope. Contributing to a larger body of research supports the argument that increasing levels of international economic integration are linked to less government oppression. My work looks at a specific form of repression amid public expression of dissent, presenting a narrower yet essential analysis of how governments may weigh the benefits of foreign capital against the threat of public dissent.
The results show that increased dependence on FDI inflows reduces the violent repression of protesters. However, this article looked at the reaction to protest directly, not other forms of systemic oppression like state control of the media. Future research must concentrate on the different forms of repression and the limits of the preferences of foreign investors. Violent, repressive tactics against protesters are reduced when the government is subject to the influence of foreign investors. These findings support the optimistic argument that the presence of foreign firms could discourage repression—at least severe, overt repression of dissent. Many factors shape the preferences of foreign investors. Still, these results generally suggest that their preferences align with international antiviolence and pro–human rights norms.
Past research tended to consider repression more generally as it relates to the influence of foreign investment or economic globalization. By looking particularly at protests, I was able to streamline the repression of protests to publicly observable instances. The sample of cases is quite representative, drawing from 145 countries instead of selecting a single region. Using protests from across the world creates a test of the most basic forms of repression that could trigger an external reaction making them much more difficult to ignore. Nevertheless, this does not include the indirect effect that foreign investment might have; if it increases economic growth, it is safe to say that it also indirectly decreases the use of violent repression by the government. Civil and political rights advocates and human rights organizations may thus find that foreign investors could be valuable allies as opposed to adversaries.
The findings here also contribute to research on the influence of economic globalization on democratization. In a way, the impact of foreign investors creates a sort of protection for the protester, even as it dissuades the government from using extreme violence in response to a protest. It could increase the ability of the opposition to assemble and gradually influence the government to make concessions. This could incentivize the government to be more open to accommodating dissent or making compromises. With this in mind, dissidents could gradually have more impact on the government and its policies. However, this could also incentivize the government to mitigate such behavior. Further research could explore this dynamic more, as one can argue that other restrictive government policies could increase with more foreign direct investment.
As it relates to repression, there are more dynamics to the relationship that can be explored in the future. It could be argued that rather than becoming less repressive, in light of increased amounts of foreign investments, states instead shift to systematic forms of repression. To avoid the publicity of violently repressing protests, states could use other repressive techniques, such as shutting down radio stations or attempting to place barriers on opposition movements. The public nature of protest makes repression open to external backlash, but subtle tactics could remain outside the international spotlight. FDI could have a negative impact by making the state shift from overt violent repression to severe systemic repression. Besides, because I consider violent repression, the government could still use arrests or less coercive negative sanctions to deal with protesters. This article does not consider other indirect coercive responses to protest, and therefore, I cannot make claims about the government's decision to use them.
A final area for more research would be the home countries of investors and their role in how much of an impact their investments might have on state repression. Not every country or investor might be like Canada and Talisman Energy; thus, investors from more autocratic states may be less likely to be fazed by repressive techniques employed by the host government. Smaller companies with fewer international subsidiaries could also be less interested in influencing state repression. Differences in the human rights records of investors’ home governments are outside this article's scope, but the findings here present an opportunity for further research. However, in this article, I found that even in the face of contentious challenges, foreign investors exert a level of influence on the government of their host states. This finding brings some optimism as this influence is more in line with international human rights norms. As such, it suggests that state repression is negatively influenced by foreign capital.
Appendix. Results of violent repression, OLS regressions.
Open access
