Max Vargo
Introduction
In his farewell address, President Dwight Eisenhower warned of the increasingly pervading power of the military industrial complex and its “unwarranted influence, whether sought or unsought” on American liberties and governance. Eisenhower’s 1961 warning came during a time in which the United States was spending roughly $48 billion on its discretionary defense budget. In 2020, that amount made up just 6% of the $740 billion that U.S. plans on spending for this fiscal year (Department of Defense, 2020). Such a warning can be considered ahead of its time as perhaps one of the most intense topics of debate in modern domestic politics is the role of the military, even more specifically the allocation of the national budget towards these military expenditures. Such arguments draw not just the ire of those interested in the economic impact of defense spending, but for globally involved countries like the United States, the question is also often carried with connotations regarding the role of the country abroad. As multifaceted and interesting as this topic is, this paper endeavors to examine principally the economic effects such spending incurs. While this issue is of special importance in the contemporary American zeitgeist, one would be remiss to not mention that the influence and effects of the military budget are of chief concern for almost every nation on earth, not just the sole geopolitical superpower. With this in mind, the goal of the following research is to understand how military spending affects the economic well-being of a country.
Previous research on this topic is all too-often political in nature. Grounded in theoretical, not empirical basis, these research papers seek to “prove” whether or not a certain ideologue’s assertion holds water on a particular facet of the military-economic dynamic. While such research makes for an interesting read, these papers make the mistake of focusing too broadly on several countries over a very short period of time, all the while only testing a few arbitrary economic indicators. Whether this broad effort is purposeful or not, the end result usually amounts to the authors struggling to make substantive sense of a complex and varied dataset whose statistical trends are far from uniform. These papers also rarely reach beyond the 20th century in their scope, meaning that the effects of watershed moments like the dissolution of the Soviet Union or the 2008 financial crisis are almost never at play in these studies. This is not to say that the existing literature is devoid of lessons, nor is it a prescription to throw the baby out with the bath water. Rather, it is exactly these kinds of studies that this paper hopes to build upon. By providing a clear “before and after” dimension to this paper in the form of the 2008 Great Recession, the study aims to expand on the existing dialogue on military spending and economic well-being while also grounding itself in its own unique timeframe and list of countries. It is from this framework that the statistical relationships can be identified while still keeping cognizant of the clear and defined differences that separate this paper from those before and after it. Additionally, through the use of statistical analysis and breaking down the time frames of analysis, the effects of the 2008 financial crisis can be examined as well. Given the Great Recession’s especially pronounced effects in the “developed” world (the focus of this study), such a breakdown for this study will allow us to extrapolate the acute effects (if any) of the financial crisis on the hypothesized relationship between military spending and economic well-being. It is with these clearly articulated goals that this paper seeks to identify whether, in fact, a relationship does exist between military spending and economic well-being. With the added analysis of the Great Recession and its related effects, the findings this paper will produce are indicative of not just the much-discussed relationship in question but also how this perceived relationship is transformed when put through the conduit of the worst economic recession since the Great Depression. This layered approach will yield results that can be added to the discussion on the military-economic relationship as well as the broader study of how economic recession impacts nations at large. Perhaps most importantly, the focus on the OECD countries will put all these analyses in the frame of reference from which most researchers are most familiar with, that being the “developed” world. Once the related literature is explained and put into context, the specifics of the research design will be presented. After this has been completed, statistical and substantive analysis of these tests will be given, whereupon conclusions and final judgements can be made.
The Historical Analysis of Military Spending and Economic Prosperity
Despite its perceived importance in the realm of political debate, the subject of military spending and economic well-being has only been researched and analyzed relatively recently, notably in the last 40 or so years. Though initially disparate in nature, such studies reflect not only the growing importance of the political debate surrounding the issue but also the increased accessibility and reputability of the data necessary to conduct such studies.
On the topic of military spending and economic well-being there stands three broad schools of thought in the existing academic studies and research. The first camp, largely influenced by the writings of neo-Marxist economists Paul Sweezy and Paul A. Baran, believes that economic development or well-being is negatively impacted by increases or an increased military budget. The writings of Sweezy and Baran were founded in theoretical, not quantitative grounds. In their book Monopoly Capital (1966), the two economists put forth that the economic growth of western nations in the wake of the Second World War was sustained chiefly through large military expenditures to fund imperialistic ambitions abroad and in their respective deteriorating colonial empires. Whilst objective empirical figures were lacking, Baran and Sweezy’s work can be seen as the theoretical groundwork upon which the future military spending studies were built upon. As one would expect, those who subscribe to this view also believe that decreases in military spending will generally correlate to increases in economic development/well-being. For expediency purposes, this camp will be henceforth referred to as the Negative Relationship group.
Proponents of the negative relationship group, like John D. Abell in his study Military Spending and Income Inequality (1994), point towards the tendency of increased military budgeting to increases in income inequality. Comparing the aggregate difference between the highest and lowest quintiles of family income to the overall U.S. military budget from 1972 to 1991, Abell found a clear correlation between increases in military spending and a widening income differential between the top and bottom family income quintiles. Though in this case the relationship is positive (Military spending increases, income inequality increases), these findings clearly fall within the categorically defined negative relationship group since income inequality is widely considered a benchmark indicator for economic well-being.
In the same vein as Abell’s study, Malcom Knight, Norman Loayza and Delano Vilanueva put forth an argument for the negative relationship group. In their International Monetary Fund-backed study The Peace Dividend: Military Spending Cuts and Economic Growth (1996), they analyze the economic effects a universal moratorium on military spending increases would have over a ten-year timeframe. Utilizing a mix of both real-world cases (chiefly those in Western Europe through the 1980s) as well as simulations, the three economists found a marked increase in economic growth when military budgets were curtailed for extended periods of time. However, these findings were not entirely universal with certain regions seeing far more pronounced increases than others. Countries in Eastern Europe and the broader Middle East region saw increases far above those in sub-Saharan Africa, signalling that the so-called “peace dividend” purported by those conducting the study was as dependent on the region as it was on the military spending itself. Furthermore, the authors noted a considerable delay or “lag” in terms of tangible economic benefits seen from the cease in military spending increases, meaning that such economic growth was contingent upon the expenditures stopping for a specific amount of time. While such findings do not entirely contradict the negative relationship group, they do lend more credibility to the assertions made by the second category of researchers.
This second group of research literature pertaining to military spending and economic well-being can be aptly titled the Positive Relationship as they believe, to the contrary of the negative relationship group, that increased military spending positively correlates with increased levels of economic well-being as defined by their various chosen fiscal variables. In short, whereas the negative relationship group sees military spending as detrimental to economic well-being, the positive relationship group sees it to be a net benefit. Literature for this grouping is less prevalent and is subjected to increased individual scrutiny as a consequence.
Just as the negative relationship group had its grounding in a theoretical framework, laid out in Monopoly Capital (1966), so too does the positive relationship group. The first major proponent of increased military spending to boost economic well-being was none other than esteemed British economist John Maynard Keynes. Keynes advocated in 1933 for President Roosevelt to increase military spending in the United States as a means of combating the mass unemployment from the Great Depression. While it is important to consider that Keyne phrased his advocacy in terms strictly of that of public works, his urging nonetheless has been seen as the watershed moment of what is now (often disparagingly) called Military Keynesianism.
It is this Military Keynesianism that was sought to be tested in the study Military Spending and Economic Well-Being in the American States: The Post-Vietnam War Era (2010) by Casey Borch and Michael Wallace. In this comprehensive research paper, the two authors analyzed the impact military spending from 1977 to 2004 had on the economies of 49 U.S. states. Utilizing a litany of economic factors such as median family income, unemployment rate and poverty rate, the authors specifically chose to test the effects of military spending at the state level during a time of relative peace. The findings of these analyses indicated that consistent military spending increases correlated with improvements in several of their key economic variables at the state level. While the journal articulates this not to be a direct endorsement of Military Keynesianism, the data does suggest that the “permanent war economy”, as they put it, does in fact lead to greater economic prosperity in both the short and long term.
These findings are not in complete agreement with earlier attempts at finding the relationship of military spending and economic well-being, even within the positive relationship group. Errol Anthony Henderson’s Military Spending and Poverty (1998) had a very similar approach as that of Borch and Wallace but its findings were substantially different. Whether it was because of the top-down methodology Henderson used (he looked at the aggregate economic figures of the U.S. rather than at the individual state level) or the specific focus on poverty as opposed to overall “well-being”, Henderson’s findings suggested that it was only during times of war that military spending had a positive effect on reducing poverty rates. While it is true that Henderson’s analysis covered a much broader span of time than that of Wallace and Borch (1952 to 1992 as opposed to 1977 to 2004), these findings do suggest that the military situation is a significant factor in whether or not the military expenditures are being applied for the good of the overall economy or not. As was the case with Borch and Wallace, Henderson emphasizes that such data interpretations are not meant as justification for increased military spending or wars abroad but instead are indicators that times of war do tend to see increased levels of merit-driven social mobility to the poor and marginalized.
The final, and perhaps most broad, category of literature concerning the relationship of military spending on economic well-being is what is termed the Non-Influential Group. For one reason or another, the studies and research in this group have found that no such relationship, be it negative or positive, can be extrapolated from the data when comparisons between military expenditures and economic development are made.
Chief among those in the non-influential group is the work by economist Albert Szymanski. The influence of military spending on economic stagnation was examined in his study Military Spending and Economic Stagnation (1973). A professor from the University of Oregon, Szymanski sought to measure whether the assertions made in Monopoly Capital (1966) by the aforementioned Paul Baran and Paul Sweezy were true. Generally, these assertions credited the economic growth of industrialized capitalist nations to their large and extensive military industries. Even more importantly, Baran and Sweezy claimed that only military spending could prevent economic stagnation in capitalist societies. Baran and Sweezy cited the United States as a prime example of such economics in practice, with its huge military discretionary budget and its economic growth since WWII (Baran, 1966). It is this framework of assertions that Szymanski sought to test. He did this by compiling a list of the top 18 countries by Gross National Product (GNP) and comparing their rates of growth to their military and arms spending. It is important to note that Szymanski emphasized that the military spending would be measured as a percentage of their relative GNP. Economic heavyweights like the USA could have potentially skewed the results if these per capita adjustments were not made. Though Szymanski’s data and analysis found that increased levels of military spending did lower unemployment levels, the supposed positive relationship between spending and economic growth that was identified by Baran and Sweezy could not be found. Even more crucially, the hypothesized positive relationship could not be systematically identified either. Szymanski’s analysis suggested that nonmilitary spending played a much more significant role in the growth of a nation’s economy and that the correlation between the United States’ substantial military-industrial complex and its economic success had much more to do with the diverse and versatile nature of the country’s economy than its military spending (Szymanski, 1973). In this regard, Szymanski’s study lent credibility to those in the Non-influential camp, whose theories found military spending to be inconsequential to the overall economy of the nation.
But as some may point out, this study is only measuring the effects of military spending on the growth of the economy as defined by GNP and not necessarily the overall well-being of the economy (as measured by negative indicators such as income inequality or poverty rate). Does the non-influential group hold water in this regard as well? Such a measurement was tested in the 2009 study Military Spending and Inequality: Panel Granger Causality Test by Eric Lin and Hamid Ali. Examining 58 countries over a period of 13 years (1987 to 1999), the authors sought to determine whether or not statistical significance could be extrapolated when military spending was analyzed against median family income inequality. Their findings saw no such statistical significance, even when the countries were grouped along OECD and non-OECD lines. Such a distinct lack of a relationship in both categories of countries indicates that even among so-called “developed” and “developing” economies there exists no econometric relationship between defense spending and economic inequality.
With the primary fields of thought on the subject summarized and contextualized, it becomes obvious that a lack of consensus is perhaps the most common trait found throughout the above literature. The variability with methodology, variables used for economic well-being and even the countries being examined leave no clear indication as to whether or not the hypothesized negative relationship will in fact exist or not. What’s more is that the time period my paper endeavors to examine is far more contemporary in nature relative to those studies shown in the literature review, with most research papers limiting themselves to the 20th century. This 21st century time frame has implications beyond the more obvious social and technological consequences that are often associated with more recent studies, that being the inclusion of post-Soviet states such as the Baltic countries.
But as is the case with any research project that seeks to build off those before it, methodological consistency is rarely ever practiced in full. Such orthodox research design would be at odds with the stated goal of this paper and as such, any criticisms based on these concerns can be disregarded. Rather, the purpose of this research is to analyze the relationship between military spending and economic well-being and to further contrast this relationship within the context of the 2008 Great Recession. The model by which these relationships will be tested is as follows:
Hypothesis 1: Countries that spend more on military spending (as a % of GDP) will see lower levels of economic well-being than those that spend less.
Hypothesis 2: Any negative economic relationships experienced due to this military spending will be even more pronounced after the 2008 recession.
Research Design
As outlined in the introduction, the stated goal for this research paper is to compare the military spending of a country as percentage of its overall GDP to that of its economic well-being to determine whether the two are correlated. While military spending as a percentage of a country’s overall GDP is exactly as it sounds (sourced from the World Bank), economic well-being is far more abstract and can be measured by a litany of different financial or empirical values. Beyond measurements, the phrase can also be interpreted in a variety of ways, ranging from an all-encompassing look at the economy of a nation to the minute details of certain fiscal or economic sectors and how these areas are performing on a year-to-year basis. Indeed, such broad phrasing would certainly invite confusion and it is for that reason that three specific measurements were used to bring method to the proverbial madness. Rather than list off the reasons that certain measurements weren’t chosen this section will instead look at the reasons these three measurements were utilized.
The first of these measurements is the Human Development Index (HDI). Developed, compiled, and sourced through the United Nations Development Programme (UNDP), the HDI is an index that measures the overall development of domestic human capital in a country. The index itself is derived from three primary sub-indexes. The first of these is the country’s health, specifically its citizen’s average life expectancy at birth. Education of a nation’s populace is also factored in with the expected versus actual average number of years of schooling being implemented as part of the education sub index. Lastly, the country’s Gross National Income (GNI) per capita is used as the third and final sub-index. The confluence of three such important domestic indicators is what makes the HDI a valued measurement to use as a baseline for a country’s economic well-being and it is for that reason that it was chosen. Some may argue that the sub-indexes of education and life expectancy are not specifically indicative of “economic” well-being in the traditional sense of the word. Such a point may be valid if one construes only hard-financial figures to be of note to a country’s economy but for the sake of this paper, a workforce’s health and education is absolutely a factor in the economic well-being of that nation. Such sub-indexes certainly have tertiary effects on an individual’s employment or economic consumption among other economic indicators.
The next measurement used was the Gross Domestic Product (GDP) per capita. GDP is perhaps the most commonly used indicator of a country’s economic health and has remained a steady measurement in terms of a country’s ability to produce goods in competition with other nation’s industries. The measurement signifies the total value of all products and services within a nation’s borders. Some might wonder with the use of GDP if some redundancy is taking place with the inclusion of GNI as being a sub-index of the HDI measurement but there exists a stark divide between GDP and GNI, namely the difference between income and total production of a country. Whereas GDP measures all goods and services in a nation regardless as to whether or not they are produced by foreign nationals or native citizens, GNI solely measures the average income of citizens of their respected nations, even outside their home country. Such a distinction may not seem significant, but a country like China can see a dramatic difference from GNI to GDP chiefly due to its large portion of citizenry living and working abroad. In this context, GNI and GDP are most certainly separate and distinct measurements.
The last economic measurement utilized is the OECD-sourced Harmonized Unemployment Rate (HUR). Whilst the first two economic variables measured the economy in terms of growth or development, the HUR is a specific indicator for the negative aspects of an economy. The HUR is different from a typical unemployment rate since that measurement is usually arbitrarily defined by the country conducting the measurement. Whereas, the HUR uses a uniform and seasonally adjusted equation to find the unemployment rate of a country, thus the “harmonized” portion of the measurement. The inclusion of this measurement comes with the hope that its addition provides a well-rounded look at a nation’s economy.
With the three measurements defined and justified an explanation of the target countries as well as the timeframe from which they will be measured is in order. The Organization for Economic Cooperation and Development (OECD) countries were chosen as the targets for this study due to their consistent industrialized nature as well as their shared political characteristics of being democratic open societies. The organization’s stated goal is to strive to develop and coordinate domestic policies that would be to the benefit of its member nations. This consistency of free-market economics and democratic values makes analysis of such a grouping of nations far more uniform and without unintentional variables than a study on all nations of the world would. The OECD’s foundational history is linked directly to the military fallout of the Second World War and the implementation of the Marshall Plan in Europe. This means that the confluence of military spending and economic well-being is also of special historical importance to this specific grouping of countries. Of the 36 OECD countries, 35 were utilized, with Iceland being omitted due to the lack of reliable military spending data.
Using data from 2000 to 2017, this research project intends to examine nine years before and after the 2008 financial crisis as its primary window of analysis. Broken into respective blocks of 2000-2008 and 2009-2017, these two periods can be analyzed and contrasted with one another to add another layer to the research question regarding military spending and economic well-being; expressly that of the role of the 2008 financial crisis on this military-economic relationship. Both timeframes will have their economic measurements and military spending analyzed for relationships and this data can then be juxtaposed against one another to see just how much of an impact the Great Recession had on these hypothesized relationships. Given the Great Recession’s distinct impact on the developed world more so than the developing or undeveloped nations, such an analysis makes all the more sense given the specific choice of analyzing the OECD countries. Once the two aforementioned datasets are compared and contrasted, analysis and findings can be given based on the hypothesized relationships and trends.
The statistical method employed will be a correlative graph as well as a linear regression analysis consisting of two separate but distinct parts. The first will utilize the 2000 to 2017 data of the OECD country averages. The collective data for each variable will be averaged on a yearly basis for the first portion. Once these OECD averages are compiled, the averages for each year will be compared. Given that there exist three economic variables by which the military spending index will be compared to, this testing will result in six regression analyses from which we can extrapolate findings and conclusions. The key indicator for ANOVA regression analyses is the p-value, which indicates the overall significance of the relationship between the compared variables. The correlative graphing will supplement these regression analyses by providing a frame of reference for which direction the relationship is trending toward. If the first hypothesis is expected to be correct, then there would be a negative correlative graph in the case of HDI and GDP and a positive correlative graph in the case of HUR.
Once a generalized view of the OECD countries can be created, a more in-depth understanding of where outliers may be can be found. This will be conducted by using the ANOVA regression analysis for each individual country for both timeframes. Once a country has had its three p-values produced, these will be averaged, and a mean p-value will be assigned to that specific country.
Macro-Level Statistical Analysis
2000 to 2008 OECD Averages
The first tests conducted were that of the averaged OECD indexes for the 2000 to 2008 timeframe. Once the yearly averages were compiled from each country, the various datasets of HDI, GDP per capita, HUR and Military Spending (as a % of GDP) were compiled. Each of the dependent variable datasets to be analyzed against the Predictor (Constant) Military Spending independent variable. It should be noted that for this first portion, some data entries from certain countries were omitted due to the lack of reliable data for their first few years of this timeframe. Countries such as Turkey and Israel lacked sufficient HUR data for a few years of this timeframe and were thus not included in the OECD averages for those years.
The statistical tests for this first portion were created as a means of giving a point of reference for the analysis that would later be done on the post-recession block of data. With a sample size of 35 (n=35) for this nine-year timeframe, the ANOVA regression analysis yielded significant statistical insights into the relationship between the three economic indicators and their relationship with military spending. Perhaps the most significant of these was that all three of the regression tests had p-values less than the benchmark 0.05, which gives clear credence to there being a statistical relationship between all three of these economic well-being indexes and the principal military spending index. With each ANOVA producing varying p-values, the Harmonized Unemployment Rate regression analysis by far produced the starkest affirmation of this hypothesized relationship with a p-value of 0.000 (See Table 1.). It should be noted that unlike HDI and GDP per capita, HUR was the only “negative” index by which economic well-being was judged. Despite this difference, the HUR null hypothesis can be rejected with confidence for the first timeframe. This rejection is compounded by the correlative graph for the same relationship, which shows a clear positive correlation between the compared variables (See Graph 1.). From this graph we can extrapolate that in the 2000 to 2008 timeframe, countries with lower military spending as a proportion of their GDP saw less unemployment overall.
While less extreme than HUR, both HDI and GDP per capita also rendered similar ANOVA regression table results, with both indexes having 0.012 and 0.004 p-values respectively (See Table 2. & Table 3.). As was the case with HUR, the null hypothesis for these datasets can be rejected as well. The correlative graphing also produced matching negative correlation between the economic well-being variables when tested against the military spending variable (See Graph 2. & Graph 3.). Both graphs indicate that as military spending increases, the economic indexes of both GDP per capita and HDI decrease during the 2000 to 2008 time period.
Taken together, these statistical findings clearly support the hypothesis that countries that allocate more to their military spending see less economic well being altogether. With all three regression analysis tests possessing p-values of <0.05, the argument for this hypothesized relationship becomes much stronger and supports the rejection of the null hypothesis. But this only proves half of what this paper intends to examine. A full confirmation of the hypotheses presented can only be made once the second timeframe is analyzed and the acute effects of the Great Recession upon this relationship are seen.
2009 to 2017 OECD Averages
In the same vein as the first timeframe, so too was this second time frame subjected to the regression and correlative statistical analysis tests. Taking the averages of all OECD countries (except Iceland) in both the economic and military indexes on a yearly basis, these values were then compared with one another using the ANOVA regression table to garner an understanding of the projected p-values. These statistical findings were of particular importance since they would be compared to the initial testing done on the 2000 to 2008 timeframe. The comparisons between the two would allow for the development of a judgement as to whether the second hypothesis regarding the impact of the Great Recession could be accepted or not.
As was the case with the previous three regression tests, the analysis completed on the HDI for the 2008 to 2017 period also had a p-value under the 0.05 threshold. With a p-value of 0.01, the null-hypothesis for this specific relationship can be rejected (See Graph 4.). The correlative graph keeps with this trend and depicts a decisively negative correlation (See Graph 4.). It is interesting to note that the HDI regression analysis possessed the highest p-value in the 2000 to 2008 timeframe but possessed the lowest p-value in this latest period. This, however, would not prove to be the last of stark contrasts between the two analyzed timeframes.
The relationship between HDI and military spending is just about the only commonality between the two time frames that are being tested. While every economic index thus far has been in-line with the original hypotheses of this paper, the remaining two economic indicators of the 2009 to 2017 appear to buck this trend. When conducting the ANOVA regression analysis, the p-values for the HUR and GDP per capita tests were 0.414 and 0.307 respectively (See Table 5. & Table 6.). This puts both far above the <0.05 benchmark and the null hypothesis cannot be rejected as a result. Essentially, this means that there is no statistical significance of any hypothesized relationship between the two economic variables and military spending. For this reason, no correlative analysis was needed since the p-value yield showed no indication of there being a true relationship between the values in question.
Micro-Level Statistical Analysis
2000 to 2008 Country p-values
The second set of statistics focused on the p-values of the various countries economic indexes when put through the same regression tests as the OECD averages. This was done in order to create a better understanding of the military-economic relationship at a more detailed level. Once the three p-values were acquired, these ANOVA inputs were registered in a table and compared amongst each other (See Table 7.). Relationships that fell below 0.05 p-value were highlighted according to their correlative trend when graphed and applied with a fit line. Finally, a country’s individual “economic well-being” score was obtained by averaging out the three p-values. This score could then be compared to the OECD average “economic well-being” score so that a clear understanding of where that country placed in relation to the OECD average could be gained.
Once compiled, this table provides a reliable reference toward each individual country and its associated relationship between military spending and economic well-being. Of the 35 countries tested, 10 possessed average economic well-being scores below the benchmark. GDP per capita saw the most statistical significance with 23 countries falling below 0.05 for that associated p-value relationship. HDI followed closely with 21 and HUR with 17. Only seven countries had no statistical significance whatsoever (Austria, Finland, Greece, Korea, Poland, Portugal and the United Kingdom).
While the majority of the countries tested yielded results, whose correlative trends were in-line with the hypothesized negative relationship between military spending and economic well-being (18 showed this relationship), there were a few that showed a positive relationship. Such a positive relationship manifested itself in a negative trend for HUR p-values and positive trends for HDI and GDP p-values. Countries who showed two or more signs of such a relationship amounted to about five, total (Canada, Estonia, Latvia, Slovenia and the United States). While this is clearly a minority of the countries tested, this does indicate that such positive relationships do in fact exist among this particular dataset.
2009 to 2017 Country p-values
The same statistical tests that the previous dataset was put through were also applied to the post-Great Recession 2009-2017 timeframe. These values were then compiled and listed under the same listing arrangements as the last timeframe. Listed together, this table also provides a comparison to the OECD average p-value score that was obtained in the Macro-level analysis (See Table 8.).
Even a brief glimpse of this compilation when compared to the previous 2000-2008 table shows just how pronounced the effects of the Great Recession were on the military-economic well-being relationship at hand. For one, the number of countries that saw total average economic well-being scores below the 0.05 benchmark was down to six from the previous timeframe’s mark of 10. This stark contrast is compounded by the fact that GDP per capita went from having the clearest relationship with military spending in the last timeframe to having the least in common with military spending this time. Only 10 countries saw p-values below 0.05 when regression tests were running comparing GDP per capita to military spending. HUR was not far behind with 15 and HDI with 22. In essence, this second time frame saw a complete reversal in terms of the relationship that was established between GDP per capita and military spending that was found in the last timeframe. The total number who saw zero statistical relationships was also down from the previous dataset, sitting at just five total (France, Hungary, Korea, Mexico and Poland).
As for correlative trends, the effects of the Great Recession were felt but they did not necessarily alter the directional makeup of most countries. There were only three countries that yielded two or more p-values that supported the positive relationship hypotheses between military spending and economic well-being; Greece, Italy and Lithuania. While most countries who possessed p-values of statistical significance were supportive of the negative relationship hypothesis, the overall trends indicate that the relationship between economic well-being and military spending is no longer as strong as it was during the 2000 to 2008 timeframe.
Substantive Discussion
Taken together, the Macro and Micro level analyses can be compared so as to establish whether or not the two posited hypotheses are acceptable or not. While the regression analyses clearly indicate (in both cases) that the 2000 to 2008 timeframe has clear evidence of a relationship between military spending and economic well-being, this relationship is not present in the 2009 to 2017 datasets. The p-values of all three regression tests for the OECD averages were well within the 0.05 statistical significance threshold, but HDI was the only regression test to fall below this benchmark for the second time frame. What’s more, when analyzed at the country-level, the effects of the Great Recession also manifested in less overall p-values within the target 0.05 range in all three economic indicators. Additionally, the Mean p-value score for the OECD averages was also up to 0.24 in 2009-2017 from 0.005 in 2000-2008.
These statistical trends would suggest that while there may have existed some residual relationship between military spending and economic well-being left in the 2009-2017 timeframe (in the form of HDI), the overall tangible relationship between the two related variables was substantially altered. Even the number of countries whose statistical data supported the positive relationship hypothesis saw decreases in their numbers. Though not the focus of the research, it is interesting to note that of the countries that did see a positive relationship between military spending and economic data, they were oftentimes post-Soviet or post-Yugoslavian states. States such as Lithuania, Latvia, Estonia and Slovenia all saw correlative trends that suggested that as military spending increased, so too did their economic well-being indicators. Of the remaining four countries who saw similar positive relationship results, they, too, had geographic commonalities. Canada, the United States, Italy and Greece all possessed correlative trends that supported the positive relationship counter-hypothesis. Whether the North American or Mediterranean commonalities had anything to do with these trends is unsubstantiated, but the coincidence is worth noting, nonetheless.
A few notable outliers for both datasets were found in the two countries that showed no statistical relationships in any of the three economic variables, those being Poland and Korea. Countries such as Hungary, Mexico, France and Portugal also possessed zero p-values below 0.05 but Poland and Korea were consistent in this regard in both of the referenced timeframes. However, almost every country displayed some degree of a statistically significant relationship between military spending and economic well-being, especially in the 2000 to 2008 timeframe.
Conclusion
The primary goals of this paper were to identify and analyze whether or not there existed a relationship between military spending and the economic well-being of a country, specifically within the context of before and after the 2008 Great Recession. Given that its very nature spans the intersection of political, economic and military interests, this topic is one of key significance in contemporary international relations and domestic policymaking. At its most fundamental level, the primary obligation of the state is toward the protection of its territory and people, meaning that the concerns of military spending are of serious importance even in these relatively peaceful times. But as this paper displays, so too are the concerns of the economic sphere important, and it is this struggle between the two that was being analyzed during this research.
What the results have shown is that while there existed a clear and present negative relationship during the 2000-2008 timeframe, this relationship all but evaporated in the subsequent 2009-2017 timeframe. This macro-level analysis was supplemented by a Micro-level assessment of each individual country and its varied relationship between military spending and economic well-being. By utilizing multiple economic indicators, an all-encompassing view of each individual country’s economy could be formed and judged against that specific country’s military spending. Even at the country level, the effects of the 2008 Great Recession materialized in the form of a distinct lack of statistical significance between the two variables at hand. While the reasons for such an occurrence were not hypothesized, nor were they expected, the general effect the recession had was that it up-ended the existing hypothesized negative relationship. To reference the literature, the 2000-2008 timeframe results clearly supported the findings of those in the Negative Relationship camp while the second half of the results were more in line with the research of scholars such as Szymanski in the Non-Influential group. While the findings do demarcate those identified relationships, it does not do so with a sense of ascriptive blame nor does it seek to advocate for any changes in existing military or economic policy.
Those seeking to expand upon this research (and research before it) would do well to keep in mind the serious limitations that come baked-in with such a controversial topic. The confluence of methodological differences in countries, economic indicators and even timeframe will inevitably lead to critique if one feels that a researcher went too narrow or too wide on a certain aspect of the study. This paper mitigated such a response by explicitly focusing on certain countries and economic indicators all on the backdrop of the 2008 Great Recession and justified these positions as such. But if the tumultuous literature review reveals anything, it is that the range of opinions and ideas surrounding this subject are as mired in ideology and theory as the actual policymaking surrounding military spending and economics. Lest one try to play the role of the policymaker when they are really a scholarly researcher, approaching this topic with humility and justified restraint is advised.
References
Abell, John D. “Military Spending and Income Inequality.” Journal of Peace Research, vol. 31,
no. 1, 1994, pp. 35–43. JSTOR, www.jstor.org/stable/425581. Accessed 13 Mar. 2020.
Baran, Paul Alexander, and Paul Marlor Sweezy. Monopoly Capital. ) Penguin Books, 1970.
Benoit, Emile. “Cutting Back Military Spending: The Vietnam Withdrawal and the Recession.”
The Annals of the American Academy of Political and Social Science, vol. 406, 1973, pp.
73–79. JSTOR, www.jstor.org/stable/1039663. Accessed 13 Mar. 2020.
Borch, Casey, and Michael Wallace. “Military Spending and Economic Well-Being in the
American States: The Post-Vietnam War Era.” Social Forces, vol. 88, no. 4, 2010, pp.
1727–1752. JSTOR, www.jstor.org/stable/40645956. Accessed 13 Mar. 2020.
“DOD Releases Fiscal Year 2021 Budget Proposal.” U.S. DEPARTMENT OF DEFENSE, www.defense.gov/Newsroom/Releases/Release/Article/2079489/dod-releases-fiscal-year-2021-budget-proposal/.
Duchin, Faye. “Economic Consequences of Military Spending.” Journal of Economic Issues,
vol. 17, no. 2, 1983, pp. 543–553. JSTOR, www.jstor.org/stable/4225328. Accessed 13
Mar. 2020.
Hamid Davoodi, et al. “Military Spending, the Peace Dividend, and Fiscal Adjustment.” IMF
Staff Papers, vol. 48, no. 2, 2001, pp. 290–316. JSTOR, www.jstor.org/stable/4621671.
Accessed 13 Mar. 2020.
Henderson, Errol Anthony. “Military Spending and Poverty.” The Journal of Politics, vol. 60, no.
2, 1998, pp. 503–520. JSTOR, www.jstor.org/stable/2647920. Accessed 13 Mar. 2020.
Knight, Malcolm, et al. “The Peace Dividend: Military Spending Cuts and Economic Growth.”
Staff Papers (International Monetary Fund), vol. 43, no. 1, 1996, pp. 1–37. JSTOR,
www.jstor.org/stable/3867351. Accessed 13 Mar. 2020.
Korotkin, Arnold. “Impact of Military Spending on the Nation’s Quality of Life.” Social Work,
vol. 30, no. 4, 1985, pp. 369–372. JSTOR, www.jstor.org/stable/23714509. Accessed 13
Mar. 2020.
LIN, ERIC S., and HAMID E. ALI. “Military Spending and Inequality: Panel Granger Causality
Test.” Journal of Peace Research, vol. 46, no. 5, 2009, pp. 671–685. JSTOR,
www.jstor.org/stable/25654463. Accessed 13 Mar. 2020.
Szymanski, Albert. “Military Spending and Economic Stagnation.” American Journal of
Sociology, vol. 79, no. 1, 1973, pp. 1–14. JSTOR, www.jstor.org/stable/2776707.
Accessed 13 Mar. 2020.