Posted by: isrj | May 28, 2010

Hegemonic Stability Theory: An Empirical Analysis

By: Jesse Hubbard


This paper examines hegemonic stability theory, the concept in international relations that an assertive hegemon can act as a stabilizing force in the international system. Through a quantitative analysis looking at both American and British hegemonic governance, it was discovered that in both cases there is a strong negative association between the hegemon’s relative power and levels of military conflict in the international system. However, higher levels of engagement in conflict by the hegemon are also associated wit higher levels of conflict in the international system, and military power is a weaker predictor of conflict levels than economic power. This has a number of potential implications, the most notable of which is that hegemons may be more effective in promoting peace through economic power than through the exercise of military force.

I. Introduction

In his magnum opus The Decline and Fall of the Roman Empire, Edward Gibbon wrote admiringly of the benefits of the Pax Romana. It was the “period in the history of the world” wrote Gibbon, “during which the condition of the human race was most happy and prosperous” (Gibbon 1776). Ever since Gibbon, the idea of an imperial peace has permeated the intellectual discourse. In the 19th century, apologists for Britain’s colonial empire spoke of a Pax Britannica spreading peace and progress throughout the world. In our own time, it is the term Pax Americana that is used to describe the relative peace that has permeated the international system since the end of World War II. In international relations, this idea of an imperial peace has been honed into a more precise concept known as hegemonic stability theory.

The field of international relations is replete with contentious debates over both theory and policy, but hegemonic stability theory has elicited controversy from theoreticians and policymakers alike. Hegemonic stability theory posits that an assertive hegemon can act as a stabilizing force in the international system. While the theory itself may be straightforward, it has complex ramifications for the nature of order in the international system.  The theory also holds important consequences regarding the morality of hegemony itself. If hegemony does indeed produce such beneficial effects for the world, then hegemonic stability theory lends moral weight to the hegemon.

Research into the theoretical underpinnings of this topic revealed that there are two main subfields within the literature on hegemonic stability. One line of study, an avenue pursued by prominent theorists such as Kindleberger, Keohane, and Ikenberry focuses primarily on questions of related to the economic system. The other avenue, pursued by theorists such as Gilpin, looks at the role of hegemonic governance in reducing violent conflict. In my research, I focus on this aspect of hegemonic stability – its implications for military conflict in the international system. To research this question, I undertook a broad quantitative study that examined data from both the American and British hegemonic epochs, focusing on the years of 1815-1939 in the case of British hegemony, and 1945 to 1999 in the case of American hegemony. I hypothesized that hegemonic strength was inversely correlated with levels of armed conflict in the international system.

Using the data from the Correlates of War Project, I was able to perform a number of statistical analyses on my hypothesis. To measure hegemonic strength, I used the Composite Index of National Capability, a metric that averages together six different dimensions of relative power as a share of total power in the international system. I then matched this data with data cataloging all conflicts in the international system since 1815. I organized this data into five-year increments, in order to make statistical analysis more feasible. Regression analysis of the data revealed that there was a statistically significant negative correlation between relative hegemonic power and conflict levels in the international system. However, further statistical tests added complications to the picture of hegemonic governance that was emerging. Regression analysis of military actions engaged in by the hegemon versus total conflict in the system revealed a highly positive correlation for both American and British hegemony. Further analysis revealed that in both cases, military power was a less accurate predictor of military conflict than economic power. There are several possible explanations for these findings. It is likely that economic stability has an effect on international security. In addition, weaker hegemons are more likely to be challenged militarily than stronger hegemons. Thus, the hegemon will engage in more conflicts during times of international insecurity, because such times are also when the hegemon is weakest. Perhaps the most important implication of this research is that hegemons may well be more effective in promoting peace through economic power than through the exercise of military force.

II. Research Question

In examining hegemonic stability theory, there are several important questions to consider. First of all, an acceptable definition of what constitutes a hegemon must be established. Secondly, a good measure of what constitutes stability in the international system must be determined. Certainly, the frequency and severity of interstate conflict is an important measure of stability in the international system. However, other measures of stability should also be taken into account. Conflict in the international system takes on a wide range of forms. While military conflict is perhaps the most violent and severe dimension, it is only one of many forms that conflict can take. Conflict need not be confined to wars between traditional states. Terrorism, piracy, and guerilla warfare are also types of conflict that are endemic to the international system. Economic conflict, exemplified by trade wars, hostile actions such as sanctions, or outright trade embargos, is also an important form of conflict in the international system. States can also engage in a range of less severe actions that might be deemed political conflict, by recalling an ambassador or withdrawing from international bodies, for example. Clearly, “stability” as it pertains to the international system is a vast and amorphous concept. Because of these complexities, a comprehensive assessment of the theory is beyond the purview of this research. However, completing a more focused analysis is a realistic endeavor.

Focusing on international armed conflicts in two select periods will serve to increase the feasibility the research. I will focus on the period of British hegemony lasting from the end of the Napoleonic wars to 1939 and the period of American hegemony beginning after the Second World War and continuing until 1999, the last year for which reliable data is available. The proposed hypothesis is that in these periods, the hegemon acted as a stabilizing force by reducing the frequency and severity of international armed conflict. The dependent variable in this case is the frequency and severity of conflict. The primary independent variable is the power level of the hegemon. This hypothesis is probabilistic since it posits that the hegemon tended to reduce conflict, not that it did so in every single possible instance. One way to test this hypothesis would be through a case-study method that examined the role of Britain and the United States in several different conflicts. This method would have the advantage of approaching the problem from a very feasible, limited perspective. While it would not reveal much about hegemony on a broader theoretical level, it would help provide practical grounding for what is a highly theoretical area of stuffy in international relations. Another method would be to do a broader quantitative comparison of international conflict by finding and comparing data on conflict and hegemonic strength for the entire time covered by British and American hegemony. The hypothesis is falsifiable, because it could be shown that the hegemon did not act as a stabilizing force during the years of study. It also avoids some of the pitfalls associated with the case study method, such as selection bias and the inherently subjective nature of qualitative analysis.

III. Literature Review

Hegemonic stability theory is one of the most influential yet controversial ideas to emerge from the field of international relations. Because the subject is of obvious relevance to the role of the United States in the world, the debate over hegemonic stability theory is consequential for academics and policymakers alike. Since the theory’s genesis in the 1970s, a substantial body of literature has emerged critiquing and expanding upon the original idea. A review of the literature reveals that hegemonic stability theory has been analyzed quite thoroughly from an economic perspective. It also reveals a potential for more research to be done examining the effect of hegemony on armed conflict in the international system.

The origins of hegemonic stability theory can be traced back to the 1950s, when Organski postulated that there was a link between British hegemony and the rise of free trade in the 19th century (Organski 1958).  In 1973, Charles Kindleberger proposed the first true iteration of hegemonic stability theory when he argued that economic disorder in the years between the First and Second World Wars could be attributed to the lack of a hegemon (Kindleberger 1973). Kindleberger’s analysis raised some interesting questions for further study. Namely, to what extent was his analysis applicable to other historical epochs? Is the stabilizing effect created by hegemony quantifiable? And, perhaps most intriguingly, does this stabilizing effect extend beyond the realm of economics?

The scholarship of Robert Gilpin was instrumental in fleshing out hegemonic governance as a broader theoretical concept. In his 1981 work War and Change in World Politics, Gilpin argued that instability in the international system is inversely related to the extent of a hegemon’s relative economic and military capabilities (Gilpin 1983). However, Gilpin’s analysis has not gone unchallenged. David Lake argues that Gilpin overestimates the centrality of hegemony to international order (Lake 1993). While he believes that hegemonic stability theory still has important insights, it should not necessarily be regarded as the central stabilizing factor in the international system. The work of Kindleberger and Gilpin divided hegemonic stability theory into two distinct approaches – Kindleberger, as a liberal economist, focused primarily on the role of the hegemon in providing collective goods contributing to economic stability. Gilpin, on the other hand, focuses much more on the security implications of hegemonic governance.

A number of scholars of international relations have followed Kindleberger’s approach. Among the most influential work in this area was done by Keohane, Deudney, and Ikenberry. In his seminal text After Hegemony, Keohane argues that despite the relative decline of the United States, economic stability has endured because of the strength of the institutions the US helped create (Keohane 1984). Deudney and Ikenberry expanded upon this analysis, arguing that hegemony alone cannot account for the current stability in the international system (Deudney and Ikenberry 1999). Rather, it is just one among a number of factors contributing to stability. Co-binding security institutions, economic openness, and civic identity are among the other factors of importance in fostering stability. The arguments of Deudney and Ikenberry are persuasive, but would be served by greater empirical rigor.

Fortunately, other authors have stepped forward with their own more empirical research. In The Limits of Hegemonic Stability Theory, Duncan Snidal uses the tools of game theory to argue that cooperation does not necessarily decline with a hegemon’s decline – it can even be enhanced under some conditions (Snidal 1985). Snidal’s analysis is useful for demonstrating the logical feasibility of Keohane’s, Deudney’s and Ikenberry’s arguments, even if his numerical models are ultimately somewhat arbitrary. And while the works of Deudney, Ikenberry, and Keohane are very important, they basically follow Kindleberger’s path in their focus on international economics. It is also worth devoting a good deal of attention to the more security-focused studies of hegemonic stability theory that have been done.

Compared to the amount of scholarship focusing on the economic aspects of hegemonic stability theory, there is a relative dearth of research focusing on security implications, especially from an empirical perspective. This is an area where there is clearly room for new research. A number of scholars have challenged Gilpin’s thesis that instability in the international system is inversely related to the extent of a hegemon’s relative economic and military capabilities, but few have empirically tested it. Spiezio performed one of the few test cases in his analysis of British hegemony from 1815 to 1939 (Spiezio 1990). He chose Britain because it provides the only example of a complete hegemonic cycle for which reliable data are available.

Webb and Krasner wrote in Hegemonic Stability Theory: An Empirical Assessment that one of the major challenges confronting researchers is in defining what constitutes hegemony (Webb and Krasner,1989). By examining Britain’s relative power year-by-year rather than simply measuring the amount of conflict during British hegemony, Spiezio circumvents this problem. Relying on the Correlates of War data set, he found that incidence of major war was inversely related to magnitude of hegemonic power, defined as a combination of military and economic capacity as a proportion of total capacity in the international system. The results also indicated that British hegemony was not the most important factor creating variation in the levels of international conflict. Another surprising anomaly was that hegemony appeared to account for more variation in frequency of all wars than in great power wars. Furthermore, Spiezio noticed a spike in conflict in the years when Britain was near its peak in terms of relative power. Thus, while Spiezio’s work provided support Gilpin’s basic hypothesis, it also called into question the centrality of hegemony in fostering stability. Spiezio’s research, while very useful, left a lot of room for further analysis. For instance, what would a similar empirical test reveal about United States dominance?

A similar, though not identical, empirical test was done by Volgy and Imwalle examining the correlation between the power of the United States and international conflict (Volgy and Imwalle 2000). Like Spiezio, Volgy and Imwalle used the data set provided by the Correlates of War project. Also like Spiezio, Volgy and Imwalle operationalize power by finding the share of US economic and military output as a proportion of the military and economic output of all major nations in the international system. The study, which examined US power from 1950 to 1992, found a statistically significant negative correlation between the relative strength of the United States and incidence of international conflict and terrorism. An interesting follow-up to Volgy and Imwalle’s research would be to expand the data set to the years after 1992. Since almost all the data in the original study was from the Cold War era, adding more data from the unipolar era would be a valuable exercise. The work of Spiezio and Volgy and Imwalle has been invaluable for providing more rigorous empirical analysis of Gilpin’s thesis.        However, despite the advances made by their research, there is still far more analysis that can be done to test the robustness of hegemonic stability theory. While existing research has been useful in proving correlations between hegemony and stability, demonstrating causation has proven to be more elusive. Given the complexity of the international system, this is understandable. However, the work of Spiezio, Volgy, and Imwalle presents some opportunities to look at causal mechanisms in some specific instances. For example, Spiezio identified an increase in conflict during Britain’s peak power years. Further research examining Britain’s role in the major international conflicts of this period could prove to be a worthwhile addition to the literature.

Hegemonic stability theory has produced its fair share of scholarship in the years since Kindleberger’s research of the interwar period. While most of the literature has focused on hegemony’s effects on international economic relations, the work of Gilpin has led to some serious scholarship on hegemony’s effects on international security. Studies like those of Spiezio have been particularly valuable in advancing scholars’ understanding of the theory’s relevance to international security. Further research building on these quantitative studies would represent a meaningful contribution to the literature on hegemonic stability theory. A more thorough understanding of the effects of hegemony on international security would have implications far beyond the ivory tower. Studying the link between hegemony and violence has the potential to affect the practice as well as the theory of international relations.

IV. Variables and Data Sources

Broadly speaking, my hypothesis addresses two distinct questions. The first question is whether there is in fact, as Gilpin posited, an inverse relationship between the strength of the hegemon and the level of conflict in the international system. The second question I address is whether there is an inverse relationship between the quantity of military force used by the hegemon and the level of conflict in the international system. I study both of these questions using the test cases of British and American hegemony. Like Gilpin, I define British hegemony as encompassing the period from the end of the Napoleonic wars to the outbreak of World War II. Excluded from this analysis are the years 1915-1919, due to some highly suspect anomalies in the data regarding composite index of national capability. The American hegemonic period encompasses the years since the end of World War II until 1999, the last year for which reliable data is available. Determining the exact start and end dates of hegemony is something of a subjective enterprise. These dates were chosen because they seem to be the most commonly used in the existing literature, but finding a perfect start and end date is inherently controversial.

Operationalization of the dependent and independent variables is a challenge of critical importance to the project. To determine the strength of the hegemon, I use the Composite Index of National Capability published by the Correlates of War Project. The Composite Index of National Capability is a statistical measure of national power that averages six different dimensions of relative strength to produce a single number between 0 and 1. Each dimension is weighted equally. A state with all the power in the international system would receive a CINC score of 1, whereas a state with absolutely no power in any dimension would receive a 0. For many years, international relations scholars were forced to rely on crude measures like GDP to gauge national strength. The Composite Index of National Capability is an improvement over these one-dimensional measures because it measures three different categories of power: economic, demographic, and military. While GDP may be acceptable as a measure of potential power, it does not reveal much about many important aspects of hegemony, such as a state’s capacity to undertake military action.

Economic power is measured through the iron and steel production ratio and the primary energy consumption ratio. The iron and steel production ratio measures the nation’s iron and steel production in kilotons in a given year against the total amount produced in the world in that year. Since iron and steel are the primary products of blast furnaces, they are one of the best proxies for overall industrial strength that exists in a quantifiable form. The primary energy consumption ratio measures a nation’s energy consumption in a given year compared to total energy consumption in the world in that year. Energy consumption can take many forms, from electricity to gasoline to coal burning. The Correlates of War project converts all types of energy production into kilotons of coal equivalent. For example, if a nation consumed 1,000 cubic meters of natural gas, the energy represented by this gas would be represented in kilotons of coal. This makes it compile a single ratio that takes into account all energy types. Relative energy consumption is a useful gauge of economic capacity since it measures a quantitative commodity that can be compared from state-to-state and year-to-year.          Demographic strength in the Composite Index of National Capability is measured using the total population ratio and the urban population ratio. The total population ratio measures the population of the state in question against the population of the world in a given year. There are several advantages a state with a larger population has in the international system. From a military standpoint, such a state can suffer greater losses during a time of war and can experience less acute labor shortages on the home front during such times. But simply measuring the number of people in a state fails to capture some of the more subtle benefits accrued by population. The Correlates of War project thus looks at urban populations for a more complete view, since they are “associated with higher education standards and life expectancies, with industrialization and industrial capacity, and with the concentrated availability of citizens who may be mobilized during times of conflict.” To count as an urban population, a city must have a population of at least 100,000 people. Of course, it is impossible to come up with completely objective definition of urban. Especially in the first part of the 19th century, many smaller cities could be unfairly discounted by this definition. However, the purpose of the urban population index is not to compare cities from different decades or centuries – its role is as a relative measure within a given year.

Any analysis of hegemonic power would not be complete without a measure of military strength. Military strength in the Composite Index of National Capability is measured through the military expenditures ratio and the military personnel ratio. For purposes of the index, military personnel are defined as soldiers under the control of the federal government, meant for use against foreign enemies. Military police and reserves, for instance, are not counted in the total. Military expenditures are defined as the military budget for a given state in a given year. Rather than look at official military budgets, which can be unreliable, the Correlates of War Project focuses on what constitutes true military spending. For example, pension to veterans and war widows are excluded from the calculations, even if they were included in the state’s official military budget.

My dependent variable, level of conflict, is also measured using data from the Correlates of War Project. Previous quantitative analyses of hegemony, such as that done by Spiezio in the 1980’s, examined only incidence of major power war. Unfortunately, this means that there are long “dry spells” in the data in which there are zero major power wars. At other points, the data contains only one or two major power conflicts during any given period. For statistical purposes, these data are less than ideal. By looking at frequency of military conflict, I avoided this statistical conundrum. The Correlates of War Project defines conflict as threat, display, or use of military force short of war. As an example of what constitutes such an institute, here is the Project’s summary of one dispute in 1997:

USA patrol boats stopped and detained two Russian tankers suspected of carrying

sanctioned Iraq oil from the Persian Gulf. Russian Foreign Minister demanded the

immediate release of the tanker (“Dispute Narratives” 2004).

Their database catalogues over 2,000 interstate militarized disputes in the years since 1815. Thus, it is a far less crude way of quantifying conflict in the international system. It also increases the reliability of quantitative analysis of hegemonic stability.

To operationalize my second independent variable, use of military force by the hegemon, I simply extracted every incident where the hegemon used military force from the Correlates of War conflict data set. In other words, I took each instance of conflict in which the hegemon was a party and plotted it against the total amount of conflict in a given five-year period.

To analyze these data, regression analysis is an appropriate tool. Using ordinary least-squared regression, I determined whether a linear relationship existed between the strength of the hegemon and the level of conflict in the international system. I also used regression analysis to determine the relationship between quantity of military force used by the hegemon and the level of conflict in the international system.

There are some shortcomings to the approach presented above. None of the measures of national power that make up the Composite Index of National Capability are completely flawless. Particularly in the early years, data may not be as precise as would be ideal. Additionally, the index does not account for some of the more subtle psychological and political aspects of power. However, this is a problem inherent in doing quantitative analysis. It is impossible to measure intangible factors using hard numbers. The Composite Index of National Capability is a good way to measure a concept that is very difficult to quantify. The results of my statistical analysis should not be looked at as laws of international relations. The analysis should be valued for the questions it raises rather than the tentative answers it provides.

V.  Data Analysis

The statistical tests performed resulted in some potentially interesting findings. The first set of data analyzed was the data on British hegemonic strength in comparison to armed conflict. As can be seen in the table below, British hegemonic strength is relatively high at the outset of the period covered, then begins a gradual decline in the later part of the century. By the 1930’s, Britain’s relative power is far lower than it was even at the outset of the hegemonic period. Correspondingly, international conflict is also higher in the final decades of focus than at any other time during the period of British hegemony. A graph of the relationship between the two variables puts this relationship into a visual context:

Years Covered Conflicts     CINC

1815-1819 5 0.327765
1820-1824 9 0.318672
1825-1829 9 0.328776
1830-1834 9 0.302717
1835-1839 13 0.303765
1840-1844 8 0.304156
1845-1849 15 0.305510
1850-1854 20 0.306360
1855-1859 24 0.292667
1860-1864 26 0.255963
1865-1869 15 0.251457
1870-1874 17 0.237007
1875-1879 24 0.231121
1880-1884 20 0.211896
1885-1889 24 0.197775
1890-1894 15 0.175833
1895-1899 29 0.167495
1900-1904 23 0.158162
1905-1909 30 0.120888
1910-1914 65 0.120462
1920-1924 46 0.105306
1925-1929 29 0.085634
1930-1934 32 0.079843
1935-1939 78 0.081023

Subjectively, the data in the table and the graph look as if they support the hypothesis, and more rigorous statistical analysis confirms this. Regression analysis reveals that the Pearson’s r-value for this data is -.754, a very strong negative correlation. Additionally, the data is highly significant, with a p-value of less than .0001. One objection that was raised to these results during their presentation was that including the years after 1914 in the British hegemonic period is controversial. Indeed, a strong argument could be made to end the period classified as hegemony in 1914. When the years after 1914 are excluded from the data, the results are still very robust. The correlation coefficient using data from this time period is -.732, a value very close to that obtained from the entire data set. The p-value is slightly higher at .0002, but is still statistically significant by even the most rigorous measures. Clearly, there is a strong negative correlation between British hegemonic strength and violent conflict.

The next step was to determine if this relationship held true for the other period of hegemony captured by reliable data- that of United States hegemony. As can be seen in the table below, there are fewer data points to work with simply because there are fewer years of hegemony to work with. However, visually, the graph of this relationship looks quite similar to the one found for British hegemony:

Years Covered     CINC      Conflicts

1945-1949 0.3251239 53
1950-1954 0.3014824 88
1955-1959 0.2491925 135
1960-1964 0.2095642 153
1965-1969 0.2041942 145
1970-1974 0.1659338 132
1975-1979 0.1412986 144
1980-1984 0.1350661 163
1985-1989 0.1396454 202
1990-1994 0.1468614 126
1995-1999 0.1464107 163

Statistical analysis confirms that the relationship found for British hegemony also holds for American hegemony. In fact, Pearson’s r is -.819 for this data set – indicating an even stronger correlation than that found between British hegemonic strength and violent conflict. Probably due to the smaller data set, the p-value is .002, still well beyond the threshold of statistical significance, but higher than the values found for British hegemony. In both of these cases, a clear negative correlation can be established between hegemonic power and violent conflict; the next question I sought to answer was why this relationship exists.

The next logical step in terms of data analysis was to look at the relationship between the hegemon’s use of force and total force used in the international system. I had originally hypothesized that the relationship between the two variables would be negative, matching the results obtained regarding hegemonic power and conflict. However, this turned out not to be the case. As can be clearly seen in the table and graph below, there is actually quite a strong positive relationship between use of force by Britain and total force used in the international system:

Year                  Brit. Conf.  Tot. Conf.

1815-1819 0 5
1820-1824 0 9
1825-1829 3 9
1830-1834 5 9
1835-1839 7 13
1840-1844 4 8
1845-1849 7 15
1850-1854 4 20
1855-1859 7 24
1860-1864 8 26
1865-1869 1 15
1870-1874 1 17
1875-1879 4 24
1880-1884 6 20
1885-1889 10 24
1890-1894 4 15
1895-1899 11 29
1900-1904 7 23
1905-1909 2 30
1910-1914 9 65
1915-1919 17 102
1920-1924 10 46
1925-1929 4 29
1930-1934 3 32
1935-1939 12 78

Statistical analysis provides quantitative backing for this observation. The correlation coefficient is .772, a strong positive relationship. The data is also highly statistically significant, with a p-value of less than .0001. While the data is sound, it is also puzzling – despite the strong negative association between British strength and violent conflict, actual use of force by the hegemon was associated with higher levels of conflict. This result added new complexity to the research puzzle, but before drawing any conclusions, I decided it was advisable to apply the same test to American hegemony. The data presented in the table and graph below tells a similar story to that shown above, albeit with significantly fewer data points to work with:

Year  US Con. Tot. Con.

1945-1949 9 53
1950-1954 5 88
1955-1959 26 135
1960-1964 25 153
1965-1969 27 145
1970-1974 17 132
1975-1979 20 144
1980-1984 29 163
1985-1989 18 202
1990-1994 17 126
1995-1999 16 163

However, despite the smaller amount of data points, the relationship is still strong statistically, evincing a Pearson’s r-value of .617 – weaker than that found for British hegemony, yet noteworthy nonetheless. The p-value, at .043, is significant at the 5% but not the 1% level. Thus, although the data is not quite as strong, the results do confirm the puzzling results found for British hegemony.

These results, though interesting, did not conform to my prior expectations. The assumption that hegemons act as a calming force in the international system by using its military strength had been severely questioned. Searching to further elucidate the complicated picture of hegemony that was emerging, I broke down the Composite Index of National Capability into its component parts, separating out military strength from overall strength. In so doing, my objective was to try and determine what exactly it is about hegemonic strength that associates it so well with conflict levels. The results of this analysis were very interesting indeed. For both British and American hegemony, military spending was actually a much weaker correlate with military conflict in the international system than the overall measure of national capability. With even a cursory glance at the British data, it is apparent that this relationship is somewhat less robust:

Year         Mil Spending  Conflict

1815-1819 0.283492672 5
1820-1824 0.223980542 9
1825-1829 0.244338151 9
1830-1834 0.161932957 9
1835-1839 0.159536552 13
1840-1844 0.157939797 8
1845-1849 0.165964312 15
1850-1854 0.236642445 20
1855-1859 0.199187872 24
1860-1864 0.105697529 26
1865-1869 0.158806708 15
1870-1874 0.120699628 17
1875-1879 0.132513581 24
1880-1884 0.136076375 20
1885-1889 0.154986914 24
1890-1894 0.143970819 15
1895-1899 0.156745702 29
1900-1904 0.25129528 23
1905-1909 0.127524639 30
1910-1914 0.207172546 65
1920-1924 0.158597164 46
1925-1929 0.105356094 29
1930-1934 0.067395886 32
1935-1939 0.13870381 78

Statistical analysis confirms this impression: the correlation coefficient is -.245, still significant, but much less so than a measure that the overall measure of power. The weaker correlation does not necessarily mean that military strength has no value in predicting conflict, it just indicates that it is far from the complete picture. A similar story plays out when the data on American hegemony is analyzed:

Year            Military Spending  Conflict

1945-1949 0.396074487 53
1950-1954 0.471776875 88
1955-1959 0.425786838 135
1960-1964 0.365564346 153
1965-1969 0.378224458 145
1970-1974 0.285955618 132
1975-1979 0.222681832 144
1980-1984 0.259814345 163
1985-1989 0.320110141 202
1990-1994 0.360196036 126
1995-1999 0.341913666 163

Statistical analysis of this data reveals that Pearson’s r-value is lower than in the data that looks at total national capability, measuring in at -.511, in comparison to the r-value of -.819 found when comparing total American capability to violent conflict levels. In fact, the relationship is even more tenuous than this still fairly large r-value suggests; the p-value is .108, and thus these results are not even statistically significant. If it is not military strength that is producing the strong association between violent conflict and hegemonic power, than what is? Data analysis of economic strength provides a potential answer to this question.

Analysis of the economic data revealed a further surprise: the hegemon’s economic strength, measured in energy output, is a more accurate predictor of conflict level than military strength. In the case of Britain, the raw data looks similar to that for military strength:

Year          Econ Strength   Conflict

1815-1819 0.853538089 5
1820-1824 0.840639128 9
1825-1829 0.812610895 9
1830-1834 0.788362747 9
1835-1839 0.749862405 13
1840-1844 0.720635622 8
1845-1849 0.687336635 15
1850-1854 0.644237745 20
1855-1859 0.579908461 24
1860-1864 0.545911899 26
1865-1869 0.500740641 15
1870-1874 0.465214556 17
1875-1879 0.442218083 24
1880-1884 0.390582117 20
1885-1889 0.34851341 24
1890-1894 0.309727031 15
1895-1899 0.282145134 29
1900-1904 0.238163843 23
1905-1909 0.199316659 30
1910-1914 0.176313819 65
1920-1924 0.138778427 46
1925-1929 0.114489304 29
1930-1934 0.127935668 32
1935-1939 0.117977629 78

Statistical analysis shows that the relationship is actually stronger than the relationship found between conflict and military spending. Pearson’s r-value is -.7398, compared to the -.245 r-value found in the analysis of British military spending. This result added more nuance to the picture of hegemonic governance that was emerging. When this analysis is applied to American economic power, a similar pattern emerges:

1945-1949 0.563718464 53
1950-1954 0.489165442 88
1955-1959 0.373312828 135
1960-1964 0.30090454 153
1965-1969 0.282157745 145
1970-1974 0.196508257 132
1975-1979 0.21354551 144
1980-1984 0.182688805 163
1985-1989 0.167266347 202
1990-1994 0.172330375 126
1995-1999 0.18656161 163

Regression analysis of this data shows that Pearson’s r-value is -.836. In the case of American hegemony, economic strength is a better predictor of violent conflict than even overall national power, which had an r-value of -.819. The data is also well within the realm of statistical significance, with a p-value of .0014. While the data for British hegemony was not as striking, the same overall pattern holds true in both cases. During both periods of hegemony, hegemonic strength was negatively related with violent conflict, and yet use of force by the hegemon was positively correlated with violent conflict in both cases. Finally, in both cases, economic power was more closely associated with conflict levels than military power. Statistical analysis created a more complicated picture of the hegemon’s role in fostering stability than initially anticipated.

VI. Conclusions and Implications for Theory and Policy

To elucidate some answers regarding the complexities my analysis unearthed, I turned first to the existing theoretical literature on hegemonic stability theory. The existing literature provides some potential frameworks for understanding these results. Since economic strength proved to be of such crucial importance, reexamining the literature that focuses on hegemonic stability theory’s economic implications was the logical first step. As explained above, the literature on hegemonic stability theory can be broadly divided into two camps – that which focuses on the international economic system, and that which focuses on armed conflict and instability. This research falls squarely into the second camp, but insights from the first camp are still of relevance. Even Kindleberger’s early work on this question is of relevance. Kindleberger posited that the economic instability between the First and Second World Wars could be attributed to the lack of an economic hegemon (Kindleberger 1973). But economic instability obviously has spillover effects into the international political arena. Keynes, writing after WWI, warned in his seminal tract The Economic Consequences of the Peace that Germany’s economic humiliation could have a radicalizing effect on the nation’s political culture (Keynes 1919). Given later events, his warning seems prescient.

In the years since the Second World War, however, the European continent has not relapsed into armed conflict. What was different after the second global conflagration? Crucially, the United States was in a far more powerful position than Britain was after WWI. As the tables above show, Britain’s economic strength after the First World War was about 13% of the total in strength in the international system. In contrast, the United States possessed about 53% of relative economic power in the international system in the years immediately following WWII. The U.S. helped rebuild Europe’s economic strength with billions of dollars in investment through the Marshall Plan, assistance that was never available to the defeated powers after the First World War (Kindleberger 1973). The interwar years were also marked by a series of debilitating trade wars that likely worsened the Great Depression (Ibid.). In contrast, when Britain was more powerful, it was able to facilitate greater free trade, and after World War II, the United States played a leading role in creating institutions like the GATT that had an essential role in facilitating global trade (Organski 1958). The possibility that economic stability is an important factor in the overall security environment should not be discounted, especially given the results of my statistical analysis.

Another theory that could provide insight into the patterns observed in this research is that of preponderance of power. Gilpin theorized that when a state has the preponderance of power in the international system, rivals are more likely to resolve their disagreements without resorting to armed conflict (Gilpin 1983). The logic behind this claim is simple – it makes more sense to challenge a weaker hegemon than a stronger one. This simple yet powerful theory can help explain the puzzlingly strong positive correlation between military conflicts engaged in by the hegemon and conflict overall. It is not necessarily that military involvement by the hegemon instigates further conflict in the international system. Rather, this military involvement could be a function of the hegemon’s weaker position, which is the true cause of the higher levels of conflict in the international system. Additionally, it is important to note that military power is, in the long run, dependent on economic strength. Thus, it is possible that as hegemons lose relative economic power, other nations are tempted to challenge them even if their short-term military capabilities are still strong. This would help explain some of the variation found between the economic and military data.

The results of this analysis are of clear importance beyond the realm of theory. As the debate rages over the role of the United States in the world, hegemonic stability theory has some useful insights to bring to the table. What this research makes clear is that a strong hegemon can exert a positive influence on stability in the international system. However, this should not give policymakers a justification to engage in conflict or escalate military budgets purely for the sake of international stability. If anything, this research points to the central importance of economic influence in fostering international stability. To misconstrue these findings to justify anything else would be a grave error indeed. Hegemons may play a stabilizing role in the international system, but this role is complicated. It is economic strength, not military dominance that is the true test of hegemony. A weak state with a strong military is a paper tiger – it may appear fearsome, but it is vulnerable to even a short blast of wind.

Reference List

CARR, EDWARD HALLET. (1939) The Twenty Years’ Crisis, 1919-1939: An Introduction to the Study of International Relations. New York: Palgrave.

Correlates of War Datasets. Correlates of War Project, Available at (Accesed March 19, 2010)

DEUDNEY, DANIEL., AND IKENBERRY, JOHN. (1999) Nature and Sources of Liberal International Order. Review of International Studies 25: 179-196. Available at (Accessed February 9, 2010.)

GIBBON, EDWARD. (1776) The Decline and Fall of the Roman Empire. New York: Alfred A Knopf.

GILPIN, ROBERT. (1988) The Theory of Hegemonic War. Journal of Interdisciplinary History. 18: 591-613. Available at (Accessed February 9, 2010.)

GILPIN, ROBERT. (1983) War and Change in World Politics. Camdridge University Press. Available at–Zvp473gE5Dk8IdVLrLdRQhI#v=onepage&q=Hegemonic%20Governance%20G(Accessed February 9, 2010.)

KEOHANE, ROBERT. (1984) After Hegemony: Cooperation and Discord in The World Political Economy. Princeton, NJ: Princeton University Press.

KEOHANE, ROBERT. (1980) The Theory of Hegemonic Stability and Changes in International Economic Regimes, 1967-1977. In Change in the Internationa System, edited by O. Holsti et al., pp. 131-62. Boulder, CO: Westview.

KEYNES, JOHN MAYNARD. (1919) The Economic Consequences of the Peace. Project Gutenberg. Available at

(Accessed April 20, 2010)

KINDELBERGER, CHARLES. (1973). The World in Depression 1929-1939. Berkeley: University of California Press

LAKE, DAVID. (1984) Beneath the Commerce of Nations. Available at (Accessed February 10, 2010.)

LAKE, DAVID. (1993) Leadership, Hegemony, and the International Economy: International Studies Quarterly, 37:  459-489. Accessed February 10, 2010, Available at

ORGANSKI, ABRAM FIMO KENNETH. (1958) World Politics. New York: Alfred A. Knopf. Available at (Accessed February 13, 2010,)

SNIDAL, DUNCAN. (1985) The Limits of Hegemonic Stability Theory: International Organization, 39: 579-614. 2010, Available at (Accessed February 9.)

SPIEZIO, EDWARD. (1990) British Hegemony and Major Power War: International Studies Quarterly, 34: 165-181, Available at (Accessed February 9, 2010)

VOLGY, THOMAS JAMES., AND IMWALLE, LAWRENCE E. (2000) Two faces of hegemonic strength: Structural versus relational capabilities – International Interactions: Empirical and Theoretical Research in International Relations. Available at (Accessed February 15, 2010.).

WEBB, MICHAEL C., AND KRASNER, STEPHEN D. (1989) Hegemonic Stability Theory: An Empirical Assessment : Review of International Studies, 15: 183-198. Available at

(Accessed February 9, 2010.)



  1. In your Abstract, you have a typo. Around line 9, ” However, higher levels of engagement in conflict by the hegemon are also associated wit higher levels ” wit = with? and in Paragraph 8 under Literature Review section there is a large space before ” However, despite the advances made by their research…” is this a copy paste format error, or did you intend to have a new paragraph?

  2. […] un distacco permanente rispetto alle altre potenze regionali. Secondo la teoria neorealista della stabilità egemonica (TSE) la presenza di un egemone aumenta la stabilità del sistema internazionale, anarchico per […]

  3. […] the decision had more to do with maintaining influence abroad then we must consider whether American hegemony is of benefit to the world economy because of the stability it brings or of detriment to it because of the negative consequences (and […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: