Thrifty Authoritarians: U.S. Regime Change 1945-Present

Introduction

Given the rampant instability in various parts of the world today, and the almost constant friction emanating from the Middle East amidst the Civil War in Syria, many leading voices in the American political sphere have floated the idea of overthrowing the regime of Syrian head-of-state, Bashar al-Assad, in favor of a government run by moderate rebel forces. These postulations have caused uproar from a war-weary nation that has begun to strongly subscribe to the belief that it is no longer in the best interests of the United States to intervene in the affairs of foreign nations, especially those in the Middle East.

The United States has a long, well-documented history of overthrowing regime to install others that align with US foreign policy interests. These efforts can be seen as far back as the presidencies of William McKinley and Theodore Roosevelt. The US used these efforts to establish themselves as an expansive power in the world and to actively defend their interests abroad. Some of these ideologies persist in today's geopolitical world.

However, an equally long history of discontent with these pursuits exists as well. Many in the political world have expressed outrage over human rights abuses perpetrated by regimes in Chile and Argentina, amongst others, and targeted the American government for helping take down democratically elected governments in order to defend certain perceived interests in these states. These sentiments shift the dynamic of discussions on potential regime change efforts in the present day. The US has all but completely abandoned the Central Intelligence Agency (CIA) as an engine for covert regime overthrows, and any suggestion of military intervention is met with hostility and scrutiny. The harsh criticisms drawn by the intervention in Libya and potential intervention in Syria illustrate the shifting discussion.

This raises the important question of how US regime change affects nations in which it decides to alter the current power structure. This study seeks to provide statistical insights to that question by analyzing the current schools of thought surrounding the issue in detail, delving deep into the statistics of the cases where the US has intervened both democratically and economically, and to draw conclusions based on the findings. This study is not meant to be a definitive concrete answer to this or any question. However, it seeks only to provide informative data to further the conversation in this area.

Primacy, Power, and Isolation: Theories of US-Sponsored Regime Changes

Most experts and academics in international relations advocate for one of three theories when it comes to United States intervention and regime changes. Skeptics, seemingly in the majority, say that United States interventions have been largely unsuccessful in the past, leaving no reason to think they can or will successfully benefit US foreign policy in future attempts (Fukuyama 2006; Kinacioglu 2012; Kinzer 2006; Meernik 1996). Those who rebut the skeptics directly claim that the United States, if it takes certain steps, can reshape almost any nation and force beleaguered regimes that may be unfriendly to the US to swiftly crumble (Litwak 2008). Those who fall into this camp, argue that it is not the concept of regime change that produces failure, but rather tactical errors that are wholly avoidable. The third, more nebulous camp consists of those who look at the context of the situation before advocating action. Writers, such as these, suggest certain efforts are doomed to failure due to the context and tactics used and others represent the possibility of success if it is approached in the right manner (Schulte 2013). There is a large-scale focus on the international surroundings and geopolitical climate. In this study, various aspects of each school of thought are adopted to form the over-arching conclusion.

The seemingly dominant theory holds that the United States is inept when it comes to an interventionist foreign policy and, therefore, creates more enemies and threats to US interests than there were previously (Kinacioglu 2012). There are, indeed some merits to this line of thinking, especially if one were to look at the more recent endeavors the United States has made. For example, non-interventionist theorists would easily point to the propping up of leaders like Saddam Hussein at the hands of the United States as an example of the fragile nature of these efforts. The US propped up Saddam to fight a radical Iranian regime, but then overthrew him when interests conflicted. The resulting war created an unstable Middle East, according to these theorists, and allowed the various threats emerging from the destabilized region to become an even more dangerous enemy to both the US and its interests. These theorists aim to show how the interests of the United States are always changing and a regime change effort can only be temporary. Thus, any regime ouster will require the allocation of far more resources and manpower than the US can spare and these efforts are spurious at best and cannot be reliably successful.

Theorists who subscribe to this camp will use even wider ranging historical examples going back as far back as the days of Hawai'i, Cuba, and the Philippines to show that the United States cannot be successful in any context and that, despite the accomplishment of certain, specific, minor goals, these attempts will always be incredibly costly as far as resources and manpower are concerned (Fukuyama 2006). The skeptic camps, while often dismissing or ignoring the universally lauded regime change successes, such as those in Germany and Japan, highlight the important aspect of this discussion that many of these efforts are perceived as failure. This study will take that perspective into account when defining the scope of success and failure.

The direct counter posits that the United States is better off being the world's authority in other parts of the world as a there is moral degradation happening in certain areas that threaten to invade the Western world and harm American interests both at home and overseas. The argument states that some foreign aggressors who are dangerous to the world, and US foreign interests, can pose as victims in certain lights and create a moral relativism that prevents many on the global stage to be wary of outright calling out these ideologies as blatantly wrong (Kristol 1995). The only way to neutralize these threats, most commonly taking the form of Communism or radical Middle Eastern regimes, though this is not always the case, is to take them down by force. Many of these theorists, who can generally be classified as "neoconservatives" can point to the successes in nation building in the aftermath of World War II. In rebuilding the German and Japanese governments after the leadership of radical authoritarians, the US played an integral role to prop up the new regimes. Germany and Japan quickly became re-integrated into the global community and became stable democracies and improving economies soon thereafter. These theorists do recognize, as Litwak does, that there is no such possibility of a universal success and that the type of unilateral intervention where the United States come riding in on white horses overwhelming these other countries with force and instantaneously forcing their worldview upon them as people like Meernik would like to suggest (Litwak 2008; Meernik 1996). These are not irrational jingoists, but pragmatists who recognize that the best interests of the United States might lie in the exerting of effort to alter the world stage.

However, in recognizing the fact that the possibility of failure is present in the regime change efforts these theorists praise, they fail to elaborate on why failure occurs. Litwak, for example, argues that regime change, in order to be successful, needs to be accompanied by targeted, progressive sanctions to undermine the legitimacy of the regime (Litwak 2008). However, this falls short in being a generalizable statement as it does not account for why regime change efforts have been successful without these policies and why efforts utilizing these policies, like the Cuban embargo leading to the attempted ouster of Fidel Castro, have failed in unseating a sitting regime. Gregory Schulte, who does not fall into this camp, alludes to this as well in his theory in his assessment of the overthrow of Slobodan Milosevic in the former Yugoslavia, arguing that there are certain factors that need to be present for success (Schulte 2013). Optimist theories contribute to the hypothesis of this study, in that it asserts the potential for success of regime change efforts. The study will seek to ascribe a definition for what constitutes a failure, as optimist theorists generally fail to do so.

The third camp mixes these two approaches. It goes to say that there needs to be a favorable international context in play for a regime change effort to work. For example, Schulte details the United States' overthrow of Yugoslav dictator, Milosevic. He discusses how the US ensured it had allies in the region and used its regional allies as well as a precision, limited airstrike campaign with other non-military measures such as sanctions, to tear down the Milosevic regime (Schulte 2013). Theorists such as Schulte argue that these international factors need to be in place for a regime change and that the US should look carefully at what is in place and use specific tactics to make it work.

This school of thought also fails, as do the rest of the theories, to define what exactly is a success. There is no mention of the democratic, humanitarian, or economic factors necessary to be able to concretely define, for the purposes of the theory, one case as a success and another as a failure. Thus, most theorists in this area fail at making their theories generalizable. It then makes it difficult to apply them to real-world cases of this phenomenon and assess the validity of the theories.

In a case study of the Islamic Revolution in Iran, Ervand Abrahamian opined that it was the failure of the US-sponsored regime to modernize institutions in the long-run that led to the collapse of the regime (Abrahamian 1980). Therefore, this study will use Abrahamian's specific theory, in conjunction with aspects of the aforementioned three to analyze whether or not success is achieved in the short term to ultimately posit that regime change does not fail due to the effort in and of itself, but rather the lack of effort for an extended period of time.

Research Design

To analyze modern instances of United States sponsored regime change, a logical starting point would be the United States nation building after the end of the Second World War. Any efforts of the United States to alter the regime of a nation had very much to do with its expansive interests. The goals of regime change efforts after such point were very much similar in rhetoric, regardless of specific differences on a case-by-case basis. Therefore, the baseline year for the study was determined to be 1945, as the year in which the regime was altered is the basis for each case study, not the year in which the United States initiated its efforts. It must be noted that there are differences that exist between US regime ousters, where the US actively assists in the overthrow of a sitting regime and "nation building", where the US takes more of a passive role in changing the existing system, but took a more active role in shaping the regime going forward. Both sets of efforts are lumped into this study. However, it is an important distinction. The cases identified were done so through an article written by William Blum for the Centre for Research on Globalization. There are 38 cases that fit said criteria listed:

• Germany, 1945
• Japan, 1945
• Syria, 1949
• South Korea, 1953
• Iran, 1953
• Guatemala, 1954
• Congo, 1960
• Laos, 1960
• Iraq, 1963
• Brazil, 1964
• British Guiana, 1964
• Bolivia, 1964
• Dominican Republic, 1965
• Indonesia, 1965
• Ghana, 1966
• Greece, 1967
• Cambodia, 1970
• Bolivia, 1971
• Chile, 1973
• Australia, 1975
• Portugal, 1976
• Argentina, 1976
• Jamaica, 1980
• Turkey, 1980
• Chad, 1982
• Fiji, 1987
• Nicaragua, 1987
• Afghanistan, 1989
• Panama, 1989
• Bulgaria, 1990
• Albania, 1991
• Yugoslavia, 2000
• Ecuador, 2000
• Afghanistan, 2001
• Venezuela, 2002
• Iraq, 2003
• Haiti, 2004
• Libya, 2011

When the United States initiates a campaign to oust a sitting regime in a foreign land, it is usually for one of two reasons, if not both. The first of which would be an intervention to foster democracy, like those seen after World War II in Germany and Japan after the devastation of the recently finished war and the havoc that the Hitler and Tojo regimes imposed upon their people. The United States maintained a military presence and had an active role in shaping the new governments of those two nations. The other reason would be if vital American interests were at stake in the nation in question and the sitting regime threatened those interests. This would lead the United States to take action to ensure their interests, usually economic ones, were best suited in the nation. Therefore, when analyzing the impact of a regime ouster by the United States, it must, first and foremost, be viewed through the lenses of democracy and economic growth.

For data with regards to democratic growth or decline, the primary data set used will be data from the 2013 Polity IV Country Reports. This report rates a nation based on how democratic its regime was regarding free and fair elections, political participation, and the presence and degree of institutionalization of democratic institutions. The ratings are on a scale from negative ten to ten, with negative ten being a full autocracy, and ten being a full democracy (Polity IV 2013). To analyze the effects of US intervention, the score from the year in which the US unseated or fostered a new regime will be recorded and then subtracted from the same rating five years afterwards. The five-year benchmark allows enough time for the effects of the regime change to set in, but does not allow too much time for outside factors to obscure the direct impact the US may or may not have had. Negative scores will indicate a more repressive regime after the intervention and positive scores will indicate a more democratic regime.

The differences that were calculated will then be submitted into a onesample t-test to determine the significance of the data. The test will test to see what the odds of achieving similar, or more extreme results, if the true mean of the data set were zero, indicating that the US had no direct impact on the democratic future of a nation. This test will produce a p-value that will state the aforementioned odds. The threshold for significance is a p-value of less than, or equal to 0.05. Any p-value registering above that indicates insignificant data to reject the hypothesis that the US had no direct impact in the area being tested. This threshold will be universally applied to each and every one of the tests.

A different, but similar test will be performed on adjusted data. Any country that improves in democracy will be assigned a score of one, and any country that declines will be assigned a score of negative one. Countries remaining the same will be scored a zero. The same t-test will be performed on this adjusted data set to see if there is significant correlation having isolated factors such as drastic improvements or declines that may skew the test mean and alter the p-value. The same standards for significance apply.

For economic data, data was retrieved from the Maddison Project database which records the GDP per capita data for each nation by year in Geary-Khamis dollars, a universal currency (Maddison 2015). The study will then analyze three specific numbers: The GDP per capita ten years before US intervention, the GDP per capita in the year in question, and then the GDP per capita ten years later. The rate of growth will then be calculated for the ten years prior to and the ten subsequent years after intervention. Both data sets will then be submitted into a two-sample t-test to test for the significance of the data. The null hypothesis is that the means for rates during both periods would be the same and the alternative suggested hypothesis is that the mean is significantly higher in the ten years after US intervention.

The ten-year benchmark for economy was used as opposed to the aforementioned five-year one used for democracy, to allow control for shortterm exogenous shocks to the economy and hopefully get a better view of the trajectory of the economy. The data focuses on GDP per capita, as it generally provides an accurate depiction of both standard of living within that nation, and its economic output. It is an encompassing statistic unlike various other economic data indicators, which are more specific, but do not paint the overall picture to answer the question this inquiry poses.

To compare both data sets, to see if there exists a correlation between nations that do well democratically, and those that do well economically, a chisquared goodness-of-fit test will be performed using actual and expected counts of nations who improved or declined democratically and economically. The two tables of values will produce a p-value that will be subject to the same threshold of significance as the previous t-tests mentioned for this study.

Due to the fact that much unrest in the Middle East revolves around religious struggle, it would not be unreasonable to theorize that Islam is inherently hostile towards democracy and capitalism and, therefore, US regime change efforts are futile in the Muslim world. In order to account for this in the study, two simple chi-squared goodness-of-fit test will be used to determine whether or not there is statistically significant data to show that results are notably different in Muslim countries, both as far as democracy and economics are concerned, than they are in nations that revolve around other cultures.

This study will operate on the following model:
US Intervention -> Changed Economic Growth Rate (GDP per capita)
US Intervention -> Altered Democratic State (as defined by Polity IV)

The running hypothesis for this study is the following: A country, after the United States intervenes, will experience more rapid economic growth in the short-term, but will suffer democratically in the short-term.

Research and Analysis

Figure 1 shows the results of the democratic subtraction. Most nations experienced a decrease in the effectiveness of their democracy five years after United States intervention. Many were small and fell close to zero. However, some nations, like Argentina 1976 and Chile 1973, experienced decreases in their scores of 15 and 13 points respectively. Conversely, however, it can be seen that Germany 1945 experienced an almost complete transformation, making the jump from negative nine to ten. The table also shows that six nations were neither affected positively, nor negatively when it came to democracy (Polity IV 2013).

Hypothesis test results:
μ : Mean of variable
H_O: μ = 0
H_A: μ < 0

The results in Figure 2 do not produce a statistically significant result to reject the hypothesis that United States intervention does not affect the course of a nation's democracy. The p-value achieved shows that there would be about a 55.8 percent chance of obtaining those results if the hypothesis that US intervention has no effect on democracy was true. In fact, the sample mean for these cases was a positive number, indicating that, on average, a United States regime change will lead a nation more towards being a full democracy than it will the other direction. This seems to run counter to the common narrative amongst scholars today in the skeptics' camp and the alternative hypothesis that the United States very often fails at creating more democratic societies abroad.

However, the skew that nations like Germany, Argentina, and Chile had on this test was taken into account, and the appropriate, aforementioned adjusted t-test was completed as described in the previous section to adequately isolate that skew to find out whether the significance was affected.

Hypothesis test results:
μ: Mean of variable
H₀: μ = 0
H_A: μ < 0

With the null and alternative hypotheses remaining the same, the statistics still showed there to be no significant data to reject the null hypothesis that the United States intervention does not have a direct effect on the trajectory of that nation's democracy. The p-value did, however, decrease showing, now, only a 7.32 percent chance that similar results would be achieved given that the null hypothesis was true. However, even given the data has been adjusted to account for possible skew, the evidence to show that the United States has an adverse effect on democracy is not there. This is likely because, as seen in Figure 4, half of the interventions undertaken by the United States resulted in either a better democracy, or a democracy equally as strong as it was. However, the sample mean in the first test indicating an average gain in democracy is misleading since far more nations got worse democratically than in either of the other two categories, regardless of the significance of the data. Given that the United States is only detrimental, to various extents, to the democracy of its target nation in half of cases, it is a bridge too far to say that this is a general rule in these pursuits.

The economic tests performed showed a large number of nations improving their economic fortunes in the ten years after US intervention, in relation to the ten years prior. Sticking out amongst this data is Japan, whose economy grew at a rate in the ten years after the US fostered a new regime after World War II, almost 150 percent faster than it did in the ten years before the end of the War. The data also shows that eleven nations experienced a growth rate increase of over 39 percent. To put this number into perspective, the most drastic decline in economic fortunes was Portugal, which experience a rate decline of 36.83 percent (Maddison 2015).

Hypothesis test results:
μ1 : Mean of Growth Pre
μ2 : Mean of Growth Post
μ1 μ2 : Difference between two means
H₀ : μ1 μ2 = 0
H_A : μ1 μ2 < 0
(with pooled variances)

The data in Figure 6 shows that there is, in fact, statistically significant data to reject the hypothesis that US intervention does not affect the finances of the affected nation. The data shows that, on average, the nation that was the subject of the intervention fared almost 26 percent better, as far as GDP per Capita is concerned, in the ten years after the intervention, than it did in the ten years prior. These results are even further illustrated in Table 6 (below), where it can be seen that in over 70 percent of cases, a state that was the subject of intervention, experienced heightened economic growth. The p-value for this test came in at less than .0001. This indicates that there would be less than a .01 percent chance of obtaining these results if the means of the rates during the different periods were, in fact, the same. Given the data we have achieved enough data to reject that hypothesis.Continued on Next Page »