1

The Color of Money: Defense Resource Allocation and Social Spending Tradeoffs in NATO Countries, 1980-2003

By Brent M. Eastwood, PhD (2004)

Abstract
This study revitalizes the Guns and Butter debate by introducing the defense acquisition theory Color of Money. Three dimensions of defense expenditures are introduced: Procurement, Research and Development, and Personnel. Parochialism and regulation over these three pots of money help explain why tradeoffs, which include cutting defense expenditures and then increasing spending on other social programs, are so rare. The study also examines the reticence of some NATO countries to invest in defense. The free-rider concept explains why countries refrain in investing individually in a public good that is beneficial to other countries. Expenditure patterns are examined in a logistical regression analysis with all NATO countries from 1980-present. Comparative analysis controls for political, economic, and foreign policy factors. Cases showing evidence of trade-offs are selected for further qualitative study.

Introduction
The dawn of the 21st century has brought enormous challenges to international security. The global war on terror has incredible costs which would seem to force governments in NATO countries to closely scrutinize their spending on defense. At the same time, these countries have pressing needs for social welfare spending, especially as their populations grow in age. There are greater demands for increased spending on old age pensions, health care, and unemployment compensation. The existence of tradeoffs between defense and social spending would then seem intuitively appealing. However, the main scholarly research concerning comparative governmental spending on defense and domestic spending is mainly in agreement: there is little trade-

2 off between guns and butter. The current budget in the United States, with its large expenditures on defense and homeland security, while at the same time funding a long list of domestic programs, brings this topic to light. European NATO members feel the crunch as well. The United Kingdom faces costs from the Iraq invasion. Germany and France must confront a military that needs transformation to become more mobile and self-sustaining for missions outside of Europe. Members such as Poland and Turkey find different viability in the alliance, but it is not clear what effect this has on the trade-offs of domestic spending. With this background in mind, a new study is in order which focuses on the guns and butter trade-off in NATO countries. This comparative analysis would be timely and of vital interest to policy-makers and analysts. The following schools of research on this debate have emerged: ReaganThatcher, US domestic, incrementalism, comparative budgetary theory, and comparative defense. Research on domestic and defense spending reached heightened interest in the 80s after the Reagonomics budgets. Domke, Eichenberg, and Kelleher (1983) hypothesized conditions ripe for trade-offs but found no significant patterns of trade-offs in domestic and military spending. Other factors appear to have been driven by separate sets of determinants (Eichenberg et al. 1983, p. 33). Mintz and Huang (1991) looked for more indirect trade-offs between higher defense spending which they believed would crowd out economic growth and welfare spending. The authors hypothesized that military spending would have a negative impact on economic growth and would thus negatively affect education spending. Mintz and Huang also found no short-term effects on trade-offs, but they did find some significant linkage to the possibility of long-term trade-offs. Russett (1982) was also unable to find regular trade-offs in the data from the

3 US Office of the Management of the Budget records for the last four decades of American history (Russett, 1982, p. 775). Russett used OLS regression analysis to estimate the change in level of Education (DV) and change in military spending controlling for health, housing, productivity, capacity, GNP, taxes, population under 18, battle deaths, and Republicans. Even though it could be argued education spending comes primarily at local levels, it was a robust list of independent variables. But the analysis yielded no clear-cut patterns. Most of this literature originates from the US, is narrow in focus, and is relegated to the cold war. Murray and Viottis (1982) comprehensive work is also forged under the bitter peace of the cold war, but it offers some interesting conclusions. The authors claim structuralism to be the biggest constraint on policy makers concerning defense spending. The bipolar cold war structure drove spending decisions and individual actors made decisions based on these threats and perceptions. Murray and Viotti admit that the rules and norms of a regime within a security alliance, such as NATO, have more certainty and therefore less a need for arms expenditures. Wildavsky (1975) attempts a comparative look at budgetary theory and finds institutional differences can be attributed to differences in budgetary processes. The US has separation of powers and a strong presidency. The UK has a strong cabinet, but its members of the House of Commons have little control over spending. Agencies in Japan must appeal to the Ministry of Finance and to the Liberal Democrats. They also negotiate in a curious process with an extra-parliamentary party. France has a hybrid President and Prime Minister system in which, according to Wildavsky, is able to better close off the

4 budgetary process to outside pressures. But despite these institutional differences, Wildavsky still concludes that the budgetary process is based on incrementalism. Although many schools of thought on the guns and butter debate are evident, I argue that two underlying theories have been neglected in this literature: the color of money argument, which is a favorite of experts in the field of defense acquisitions (Tack 2000) and the free rider concept, which is a tenet of microeconomics. The color of money notion refers to ownership of the total cost of defense spending. Defense departments or ministries have different pots of money which are controlled by different sectors, for example; procurement, research and development, personnel, or maintenance and operations (Tack 2000). Since each sector has different accounting rules and parochial interests which guard the pot, it becomes extremely difficult to chart savings or spending across the whole department. Further, not all these pots are adequately filled to perform the mission of preparing for war in order to keep the peace. These and other oversight constraints make it difficult to shift funds from one pot to another, which in turn make it more difficult to transfer saved funds to increases in social spending. In other words, institutional constraints (under this construct) make the guns and butter tradeoff more rare. Mintz (1988) recognized the problems of the color of money concept in his work on resource allocation and the US Department of Defense. He also divided total defense spending into procurement, R&D, personnel, and maintenance. Mintz (1988) used each dimension as a separate dependent variable and compared them to economic, political, and foreign policy shocks. He found that certain domestic pressures such as elections, public opinion, and party control of the White House had strong influence over

5 spending changes in military personnel and procurement. The presence of war had the strongest influence. Although Mintzs work was groundbreaking in the color of money aspect, it still did not address social spending tradeoffs. Free-riding has to do with getting a good or service without paying for it. In the case of NATO, member countries have no incentive for increasing defense spending individually, especially in areas like procurement or research and development, if they feel it will benefit other countries (Vlachos-Denglar 2002). Free-riding can help explain the large difference in defense spending between the United States and NATO countries (Russett 1970). It also helps explain some NATO countries reluctance to conduct peacekeeping operations with teeth (Lepgold 1998), although some scholars recognize the willingness for certain countries such as England and France to take a more active role in European security without the United States, i.e. Bosnia (Kramer 2002). In this respect, the free-rider concept could negate the guns and butter argument. If countries are already spending less on defense, policy-makers have to make fewer tradeoff decisions. The color of money and free-rider premises help better explicate the guns and butter debate by modernizing the framework of the argument. Since most of the original guns and butter literature occurred during or immediately after the cold war, it is time to update theory and confirm new hypotheses.

Purpose
I am replicating parts of the Domke (1983) et al study and including the color of money and free rider notions within their original models. I want to expand the number of countries analyzed (to full NATO membership) and expand the time-frame to account for the last 20+ years. I choose this timeframe because the Reagan-Thatcher era

6 was a watershed in defense policy, it includes the end of the cold war and the beginning of the peace dividend era, and it includes the beginning of the global war on terror. I will not only analyze individual countries, but also the NATO-Europe bloc as a whole (16-countries), the four-largest military spenders in NATO-Europe (UK, France, Germany, Italy), and the four-smallest military spenders in NATO-Europe (Luxembourg, Portugal, Greece, Belgium). I hypothesize that a fragmented trade-off pattern (null hypothesis) will present itself as most NATO countries will show very little change in social spending as defense spending increases. Cases in which trade-offs actually occur, that is when defense spending increases and social spending decreases (alternative hypothesis), will be selected as cases for further qualitative study. The purpose of this regression analysis is to test the theoretical presence of defense and welfare spending trade-offs in NATO nations from 1980 until present. Other theories tested include the defense acquisition concept of color of money and the microeconomic free-rider concept. The first comparative analysis compares changes in total defense expenditures (DV) to the changes in total welfare expenditure (Explanatory IV) controlling for total overall expenditures, chief executive political orientation, legislature political orientation, presence of an election year, war deaths, international tension index, inflation, and unemployment. Welfare expenditures are defined by Total Social Welfare Outlays from OECD countries as used by Swank (2002). Controlling for domestic political pressures, economic pressures, and foreign policy pressures will show the importance of competing interests for the budget dollars of various policy makers within the sample. However, I keep in mind the constraints of

7 incrementalism (Wildavsky 2004) and how certain facets of these relationships depend on how much each country spends on defense in the first place. Due to the free-rider concept, I hypothesize certain tradeoffs occurring among the Big Four European NATO countries (UK, France, Germany, Italy), but few if any tradeoffs occurring with the low-spenders such as Belgium, Luxembourg, Greece, and Portugal. I set up a second group of equations to factor in consideration of the color of money theory. Defense expenditure is broken down into three dimensions: procurement, research and development, and personnel. This separate model centers on three separate equations which compare total welfare spending to the changes in spending on procurement, R&D, and personnel. These equations also control for total overall expenditure plus the same domestic, economic, and foreign policy variables as in the first equations.

Literature
The guns and butter literature can be broken down into five main categories: The Reagan-Thatcher defense build-up, US domestic, Incrementalism, Comparative, and the Post-Cold War/peace dividend school of thought. Kamlet, Mowery, and Su (1988) devised simulations based on federal budgetary outcomes during the Reagan administration to judge fiscal impact from 1982-86. The authors hypothesized that presidents following Reagans term still would have spent runaway amounts on defense even without the tax cuts. They found that the tax cuts were the major contributor to the deficit. This type of deficit spending is an important economic argument against tradeoffs. The Keynesian dynamic of deficit spending during times of recession allows

8 for governments to keep spending on social programs, even in times of increased defense spending. Other researchers search for tradeoffs in US spending on social programs and defense expenditures. Beck (1983) examined various tools of analysis for estimating time series models in US budgetary policy and identified spending shifts before or after politically important times such as elections or administration changes. Kamlet and Mowery (1987) looked at more specific macroeconomic and political factors which may affect budget priorities between the US executive branch and Congress. They found defense budgets to have more expenditure interdependence with economic factors such as inflation and unemployment. These factors of interdependence were more pronounced in the executive branch than in Congress. Mintz and Huang (1991) looked at the indirect spending effects of military spending on education spending in the US from 1953-1987. They found no short-term effects, but significant negative long-term effects. Peroff and Podolak Warren (1979) tested four different tradeoffs concerning defense and health spending in the US from 1929-1974. These tests were based on time, size of defense share of spending, the stage of the budgetary process, and type of financing. They found evidence of health and defense spending tradeoffs during the Vietnam War, but found no overall tradeoffs occur during aggregate times of war and peace. Russet (1970) found the aftereffects of war to be significant in post-war spending habits. He called this phenomena the ratchet effect referring to a countrys inability to reduce defense spending to pre-war levels following conflict. Russett (1982) also examined health, education, and defense tradeoffs from 1941-1979 controlling for economic, political, and demographic changes. He found

9 no major evidence of tradeoffs and no evidence of Republican Party influence on tradeoffs. Perhaps the most compelling portion of the literature is based on Wildavskys (2004) theory on incrementalism in the budget process. Auten, Bozeman, and Cline (1984) challenge this notion with their sequential budget model which focused on topdown budget parameters, influence of interagency competition, and the connection between revenues and appropriations. Berry (1986) identified one of the main limitations of empirical studies in the budgetary process: methodological and theoretical problems in the independent and dependent variables (ratio variable problems). He offered an alternative approach using a null hypothesis combined with the concept of incrementalism and other built-in structural variables. Berry found that using simulated data in these alternative approaches helps alleviate the ratio variable problem. Other researchers find no real difference between competing models and Wildavskys. Fischer and Kamlet (1984) unveiled a new model called the Competing Aspiration Levels Model (CALM), which explored trade-offs in defense, non-defense, and fiscal policy. They test it on US presidential budgetary data from 1955-1980. The authors found that each ensuing year provides a secure budgetary base for the next, which supports the traditional incremental argument. Gist (1982) compared the incrementalist theory of Davis, Dempster, and Wildavsky with the competition theory of Natchez and Bupp. Gist found that the levels of analysis, theoretical construct and variables used in each model, were interchangeable. He concluded that the two models are not really in competition with each other. Baumgartner and Jones make the best argument against incrementalism.

10 Jones, True, and Baumgartner (1997) tested their own concept of punctuated equilibrium on US budget authority since World War II. They found spending can be categorized under three epochs: postwar adjustment until 1956, robust growth from 1956-1974, and restrained growth beginning in 1974. The authors test each epoch against rival explanations (changes in growth of postwar economy, partisan divisions, and public opinion) and find the punctuated epochs hypotheses survive each test. The comparative literature of western democracies focuses mainly on the Cold War period. Murray and Viottis (1982) analysis was admittedly affected by Waltzian structuralism and foreign policy pressures on defense spending. Caputo (1975) suggested using thicker case studies instead of relying on statistical analysis. He recommends these case studies be focused on economic and political factors. Chans (1995) review of postCold War literature found that political and economic constraints discourage change to the pre-Cold War paradigm of no tradeoffs in military and social spending. Domke, Eichenberg, and Kelleher (1983) make the most ambitious and complex comparative model, but found no real short-term evidence of tradeoffs, but they found evidence of long-term tradeoffs in the aggregate model. Eichenbergs (1984) later study on West Germany focused on a disaggregate model of West German social spending. He found little evidence of tradeoffs and any tradeoff could be attributed to bureaucratic politics.

Nature and Goal of the Research

Through deductive theory development using the defense and social welfare tradeoff (guns and butter) literature, plus the literature on free-riders and alliances and color of money, I attempt to answer the following questions on defense and social

11 welfare spending. I explore these questions using logistic regression analysis (Stata Reference Manual 1995) and replicate work done by Domke, et al (1983). These regressions test whether a significant relationship exists between the changes in total defense spending and total welfare spending controlling for total expenditures, chief executive political orientation, legislature political orientation, and presence of an election year, war deaths, international tension, inflation, and unemployment. The second group of equations test whether a significant relationship exists between total welfare spending and spending on procurement, research and development, and personnel controlling for total overall expenditures plus the same political, economic, and foreign policy variables.

Theory
Model One: Aggregate Defense Spending
Dependent Variable: Total Defense Spending-TOTALDEF
I use Table 1 Defense Expenditures of NATO Countries 1980-2003 from the NATO web page. Each countrys total defense spending is listed in their home currency or in the Euro depending upon the year. Total are given for aggregate units of analysis such as NATO-Europe, North America, and NATO-Total. I change each raw aggregate into total defense spending as a percent of GDP for each country. Null Hypothesis: Ho=There is no significant difference in comparison between total defense spending and total welfare spending in NATO countries. Alternative Hypothesis: Ha=As total welfare spending goes up, total defense spending goes down (negative relationship).

12

Explanatory Variable: Total Welfare Spending-TOTALWEL

I use Swanks (2002) measurement of Total social welfare outlays, comprised of OECD publications: Social Expenditure Statistics of OECD Member Countries; Labour Market and Social Policy Occasional Papers, No. 17, Paris: OECD; New Directions in Social Policy in OECD Countries, Paris: OECD, 1994. This measurement is in percent GDP. Null Hypothesis: Ho=There is no significant difference in comparison between total defense spending and total welfare spending in NATO countries. Alternative Hypothesis: Ha=As total welfare spending goes up, total defense spending goes down (negative relationship).

Control Variable: Total Government Expenditures-TOTALEXP

The following control variables replicate work done by Domke et al (1983). The total expenditure data is taken from current editions of national statistical yearbooks such as the Statistical Abstract of the United States. It reflects total government expenditures for a particular year as percentage of GDP. I am careful to consider uniform expenditure categories across the various countries and keep in mind the institutional differences in record keeping, tabulation, and other operational concerns.

Control Variable: Chief Executive Political Orientation-CHIEFPOL

This dummy variable controls for the political ideology of political executives as described by Domke et al (1983). The reasoning behind the variable is that the ideological positioning of chief executives is of paramount importance in driving

13 budgetary decisions. Leftist leaning executives (who are more likely to spend on social programs), i.e. Democrat, Labour, Social Democrat receive a 1; Republicans, Conservatives, Christian Democrats receive a 0.

Control Variable: Legislative Political Orientation-LEGPOL

This variable controls for the political ideology of each countrys legislature as described by Domke et al (1983). Different political parties have different ideologies concerning the nature of budgetary spending concerning defense and social welfare. Each country is assigned a percentage of leftist seats from the total number of seats in the main legislative body.

Control Variable: Election Year-ELECTION

It has been argued in budgetary and electoral literature that election year spending on social programs and defense increases during election years as policy makers attempt to win votes through transfer payments (Domke et al 1983). A dummy variable is then included to control for the presence of a major executive or legislative election in each country.

Control Variable: War Deaths-WARDEATHS

Domke (1983) et al asserts that the participation in war positively correlates to increased defense spending. I control for this factor by using annual combat casualties as a control variable. The data are taken from various statistical yearbooks such as Statistical Abstract of the United States or Facts on File.

Control Variable: International Tension Index-TENSION

14 NATO has undergone the ebbs and flows of tension and crises since the end of the cold war, Bosnia and Kosovo, and 9/11. Domke et all (1983) proposes using a tension index developed by Goldmann and Lagerkranz (1977). The measure consists of a coefficient of imbalance taken from a content analysis of statements from the leaders of various countries in the NATO alliance. This index forms a baseline for tension in the alliance and a benchmark to contrast the strength of relationship for each polity.

Control Variable: Unemployment-UNEMPLOY

Variable measures percent change of unemployment from previous year.

Control Variable: Inflation-INFLAT

Variable measures percent change of cost of living (inflation) from previous year. These economic variables reflect the Keynesian countercyclical spending theories practiced by NATO countries, namely, during recession execute deficit spending; during recovery execute austerity measures. These variables combine to form the equation for aggregate defense spending as follows: %TOTALDEF= - %TOTALWEL+ %TOTALEXP+CHIEFPOL+LEGPOL +ELECTION+WARDEATHS+TENSION+ %UNEMPLY + %INFLAT

Model Two: Resource Allocation (Color of Money)

To account for problems of multicollinearity between the variables total defense spending and total expenditures and to account for more complex theories of resource

15 allocation and the color of money, I introduce a separate four-equation model which disaggregates defense spending and addresses these issues.

Dependent Variable: Total Welfare Spending-TOTALWEL

I use Swanks (2002) measurement of Total social welfare outlays, comprised of OECD publications: Social Expenditure Statistics of OECD Member Countries; Labour Market and Social Policy Occasional Papers, No. 17, Paris: OECD; New Directions in Social Policy in OECD Countries, Paris: OECD, 1994. This measurement is in percent GDP. Null Hypothesis: Ho=There is no significant difference in comparison between resource allocated defense spending and total welfare spending in NATO countries. Alternative Hypothesis: Ha=As resource allocated defense spending goes up, total welfare spending goes down (negative relationship).

Independent Variable: Procurement-PROCURE

Procurement spending denotes appropriations used to purchase weapons systems and other investment items which exceed $100,000 in terms of systems unit cost (Roberts 2002). According to Vlachos-Dengler (2002) European-NATO countries showed very little spending growth in procurement throughout the 90s. Vlachos-Dengler attributes this to little incentive for investment in new weapons system since individual county spending will benefit other countries in the alliance. Along with free rider issues is the color of money problem in which the incremental nature of spending in separate pots is constrained by standard operating procedures and bureaucratic politics.

16 Data Sources: See Mintz (1988). Stockholm International Peace Research Institute, World Armament and Disarmament: SIPRI Yearbook, United Nations Document A/35/479, Reduction of Military Budgets. New York. 1981. Null Hypothesis: Ho=There is no significant difference in comparison between changes in spending on procurement and total welfare spending in NATO countries. Alternative Hypothesis: Ha= As percent change in spending on a countrys defense procurement goes up, total welfare spending in that country goes down. Independent Variable: Research, Development, Test, and Evaluation-R&D R&D includes appropriations used for research, development, test and evaluation efforts. R&D is historically a laggard in spending compared to the other resource allocations in defense budgets. This sector is complicated since European countries have varying degrees of interest and involvement in military R&D. Some countries, such as France, spend upwards of 50 percent of the defense budget on R&D. However, the incremental base for budgeting in this pot of money is low for many European countries (VlachosDengler 2002). This is problematic because it is then unlikely that any tradeoff with social spending will occur. There are low incentives for European governments to invest in R&D, therefore any tradeoff occurring in social spending is likely negligible or accidental. Null Hypothesis: Ho=There is no significant difference in comparison between changes in spending on defense R&D and total welfare spending in NATO countries. Alternative Hypothesis: Ha= As percent change in spending on a countrys defense R&D goes up, total welfare spending in that country goes down.

Independent Variable: Personnel Spending-PERSONNEL

17 This variable includes appropriations used to pay military personnel costs such as basic pay, allowances, special pay, bonuses, and moving costs. Of all the resource allocation variables, personnel is most likely to vary compared to social spending. According to Vlachos-Dengler (2002, p. 149), Europeans spend a larger proportion of their budgets on personnel versus what they are spending on arming themrather than reallocating funding toward equipment budgets to provide forces with transport, precision weapons, and command and control. What does this have to do with social spending? It would

appear that decision makers in European countries feel more domestic pressure to provide for social concerns, so the political and economic control variables may have a greater effect on the personnel explanatory variable. Null Hypothesis: Ho=There is no significant difference in comparison between changes in spending on defense personnel and total welfare spending in NATO countries. Alternative Hypothesis: Ha= As percent change in spending on a countrys defense personnel goes up, total welfare spending in that country goes down. The remaining control variables are identical to the variables in the first model. The above factors can be taken together in the following equations: %TOTALWEL= - %PROCURE + %TOTALEXP+CHIEFPOL+LEGPOL +ELECTION+WARDEATHS+TENSION+ %UNEMPLY + %INFLAT

%TOTALWEL= - %R&D + %TOTALEXP+CHIEFPOL+LEGPOL +ELECTION+WARDEATHS+TENSION+ %UNEMPLY + %INFLAT

%TOTALWEL= - %PERSONNEL + %TOTALEXP+CHIEFPOL+LEGPOL +ELECTION+WARDEATHS+TENSION+ %UNEMPLY

18 + %INFLAT

Findings
I predict that the null will hold in all hypotheses and that no real significant relationships will emerge in these variables. The color of money institutional constraints of incrementalism, standard operating procedures, and bureaucratic politics are too much to overcome. There are also the cultural and normative constraints of low defense spending due to the free rider problem. However, these findings are still important because knowledge is still advanced on this topic: budgetary decision making is greatly influenced by resource allocation constraints. It is an important update on work done by Mintz (1988) and Domke et al (1983).

19

Bibliography Auten, Gerald, Barry Bozeman and Robert Cline. 1984. A Sequential Model of Congressional Appropriations. American Journal of Political Science. 28:503523. Beck, Nathaniel. 1983. Time-Varying Parameter Regression. American Journal of Political Science. 27: 557-600. Berry, William D. 1986. Testing Budgetary Theories with Budgetary Data: Assessing the Risks. American Journal of Political Science. 30: 597-627. Berry, William D. and David Lowery. 1990. An Alternative Approach to Understanding Budgetary Tradeoffs. American Journal of Political Science. 34:671-705. Caputo, David A. 1975. New Perspectives on the Public Policy Implications of Defense and Welfare Expenditures in Four Modern Democracies: 1950-1970. Policy Sciences. 6:423-446. Chan, Steve. 1995. Grasping the Peace Dividend: Some Propositions on the Conversion of Words into Plowshares. Mershon International Studies Review. 39: 53-95. Domke, William K., Richard C. Eichenberg, and Catherine M. Kelleher. 1983. The Illusion of Choice: Defense and Welfare in Advanced Industrial Democracies, 1948-1978. American Political Science Review. 77: 19-35. Duval, Robert D. 2001. Trading Bases: Resolving the Guns versus Butter Tradeoff Puzzle via Full Specification draft paper. Eichenberg, Richard C. 1984. The Expenditure and Revenue Effects of Defense Spending in the Federal Republic of Germany. Policy Sciences. 16:391-411. Fischer, Gregory W. and Mark S. Kamlet. 1984. Explaining Presidential Priorities: The Competing Aspiration Levels Model of Macrobudgetary Decision Making. American Political Science Review. 78:356-371. Gist, John R. 1982. Stability and Competition in Budgetary Theory. American Political Science Review. 76: 859-872.

20 Hartley, Thomas and Bruce Russett. 1992. Public Opinion and the Common Defense: Who Governs Military Spending in the United States. American Political Science Review. 86: 905-915. Jones, Bryan D., James L. True, and Frank R. Baumgartner. 1997. Does Incrementalism Stem from Political Consequences or From Institutional Gridlock? American Journal of Political Science. 41:1319-1339. Jones, Bryan D., James L. True, and Frank R. Baumgartner. 1998. Policy Punctuations: U.S. Budget Authority, 1947-1995. Journal of Politics. 60: 1-33. Kamlet, Mark S., and David C. Mowery. 1980. The Budgetary Base in Federal Resource Allocation. American Journal of Political Science. 24:804-821. Kamlet, Mark S., and David C. Mowery. 1987. Influences on Executive and Congressional Budgetary Priorities, 1955-1981. American Political Science Review. 24:804-821. Kamlet, Mark S., David C. Mowery and Tsai-Tsu Su. 1988. Upsetting National Priorities: The Reagan Administrations Budgetary Strategy. American Political Science Review. 82:1293-1307. Kramer, Stephen P. 2002. French Foreign Policy: The Wager on Europe. In Foreign Policy in Comparative Perspective. Washington, D.C.: Congressional Quarterly Press. Lepgold, Joseph. 1998. NATOs Post-Cold War Collective Action Problem. International Security. 23:78-106. Mintz, Alex. 1989. Guns vs. Butter: A Disaggregated Analysis. American Political Science Review. 83: 1285-1293. Mintz, Alex and Chi Huang. 1991. Guns versus Butter: The Indirect Link. American Journal of Political Science. 3:738-757. Mintz, Alex. 1988. The Politics of Resource Allocation in the US Department of Defense. London: Westview Press. Mok, Jin Whyu. 1987. The Elusive Choice: A Comparative Study of Budgetary TradeOffs Between Defense and Social Spending. PhD Diss., West Virginia University.

21 Murray, Douglas J. & Paul R. Viotti. 1982. The Defense Policies of Nations: A Comparative Study. Baltimore: Johns Hopkins University Press. Peroff, Kathleen and Margaret Podolak-Warren. 1979. Does Spending on Defense Cut Spending on Health? British Journal of Political Science. 9:21-39. Roberts, Julian. 2002. Financial Management Terms. Defense Acquisition University: Business, Cost Estimating, & Financial Management Department, Fort Belvoir, Va. Russett, Bruce. 1982. Defense Expenditures and National Well-Being. American Political Science Review. 76: 767-777. Russett, Bruce. 1970. What Price Vigilance? New Haven: Yale University Press. Swank, Duane. 2002. Global Capital, Political Institutions, and Policy Change in Developed Welfare States. Cambridge, UK: Cambridge University Press. Tack, Siobhan. 2000. Color of Money 101. A Primer for the Who, When, Where, and Why of the Restrictions on Congressional Appropriations. Fort Belvoir, Va.: Defense Acquisition University. Vlachos-Dengler, Katia. 2002. From National Champions to European Heavyweights: The Development of European Defense Industrial Capabilities Across Market Segments. Santa Monica, Ca.: RAND. Wildavsky, Aaron & Naomi Caiden. 2004. The New Politics of the Budgetary Process. New York: Pearson, Longman.

22

The Color of Money: Defense Resource Allocation and Social Spending Tradeoffs in NATO Countries 1980-2003

Brent Eastwood POLS 555 Dr. Studlar 19APR04