sign in sign up
<<
Home , Conscription

Research Reports

ENDING Figure 2 presents military regimes world- wide. Conscription is still dominant in Asia and Africa. CONSCRIPTION* In America the use of conscription has been in a steady decline, with (de facto) abolitions in Uruguay (1989), (1990), Honduras (1994), PANU POUTVAARA* AND (1995), (1999), (2005) and (2008).

NDREAS AGENER A W ** While the decision to abolish or maintain the mili- tary draft has country-specific aspects in each single Current trends case, certain groups of motivations for the abolition of conscription can be identified: Throughout most of the 20th century, military con- • Conscription in the US was abolished in 1973, to- scription (the draft) was a prominent feature of na- wards the end of the Vietnam . Liberal defer- tional military doctrines. Both world were ment rules and, later, the use of lotteries had gen- fought mainly by conscripts. Among the 12 founding erated a strong sense of the unfairness and biased countries of NATO in 1949, only did not rely selectiveness of the draft system, which added to on conscription (Iceland did and still does not have the public discontent with the Vietnam War. In armed forces). While the (1960), the debate about the pros and cons of abolishing (1967) and the (1973) the draft, economists (most notably Milton Fried- adopted all-volunteer forces quite early during the ) had substantial impact, arguing that con- , military conscription was the dominant scription is “inequitable, wasteful, and inconsis- mode of peacetime recruitment in the alliance until tent with a free society” (Friedman 1974, 253). mid-1990s. Likewise, throughout the entire existence (1955–91) of the Warsaw Pact, all its members used • In continental , with its dominant histori- conscription. cal view that every citizen has an obligation to

Figure 1 maps the changes in mil- Figure 1 itary recruitment across Europe IN EUROPE, 1993–2011 since 1993 (after the -up of Czechoslovakia and Yugoslavia). Initially only , Luxem- bourg, and the United Kingdom relied on all-voluntary service. In 2011, most countries do. The only EU countries still using conscription are , , (in which it is quite easy to avoid the service in practice), , and .

* University of Munich and Ifo Institute for Economic Research. ** University of Hannover and Ifo Insti- tute for Economic Research. We thank Max Ostermayer for helping us Sources: Own compilation, based on Wikipedia (2011), “Conscription” and selected gov- with the figures. ernment sources.

CESifo DICE Report 2/2011 36 Research Reports

Figure 2 tarism, human rights violations, MILITARY RECRUITMENT AROUND THE WORLD economic mismanagement and corruption. In the wake of demo- cratization, armed forces suffered considerable losses in prestige, leading civilian goverments in quite a number of countries to abolish or suspend conscription.

Typically, also countries without conscription during peacetime re- tain the option to re-introduce conscription in case of war – when it might be infeasible to mobilize Sources: Own compilation, based on CIA (2011), “World Factbook”, Wikipedia (2011) “Conscription” and selected government sources. the necessary manpower by vol- unteers or through fiscal taxes perform some service at the call of the state, the alone. Similar arguments can explain the use of con- decline in the use of conscription followed the scription in countries like , where the military end of the Cold War. After the collapse of the doctrine relies on the ability to mobilize most citi- and the end of the Warsaw Pact, zens to fight in case of a large-scale conflict. European governments no longer saw the need to Generally, there is a positive correlation between the prepare for large-scale warfare in Europe. This military threat perceived by countries and their use allowed many countries to seek budgetary savings of conscription (Mjoset and van Holde 2002). The by reducing the sizes of and reserves. At recent abolishment of military draft in several the same time, military capabilities were increas- European countries can then be explained by (the ingly aimed at and international perception of) reduced military threat after the fall missions for which conscripts are ill-suited and, in of the Iron Curtain. most countries, inadmissible by law. Military tech- nology moved away from territorial mass forces to smaller, mobile units equipped with sophisti- The economic case against conscription cated weaponry, increasing the necessity of pro- fessionalization. The smaller intake of draftees Economists traditionally were and are unambiguous raised equity concerns as conscription was no in favoring a volunteer (for surveys see Sandler longer hitting the full (male) age cohort but only and Hartley 1995, chapter 6; Warner and Asch 2001 those unlucky enough to be called up to service. or Poutvaara and Wagener 2007a, 2011), echoing Adam Smith’s verdict of the “irresistible superiority • For the Baltic, central and eastern European coun- which a well-regulated standing [i.e., all-volunteer] tries, which all started out with conscript armies, army has over a [i.e., temporary conscrip- the prospect of NATO membership meant the tion]” (Smith 1976, 701). Smith’s arguments focus on necessity of downsizing and restructuring their comparative advantage and the benefits from spe- armed forces. The role model of the alliance lead- cialization. Military conscription violates the princi- ers, increasing levels of draft avoidance, and the ple of comparative advantage, which demands that public opinion that conscript forces were a vestige be assigned to those who are relatively most of the Cold War or Soviet totalitarianism promot- productive in doing them, by forcing everybody into ed the eventual abolition of conscription in all a , irrespective of relative pro- countries except for Estonia (Williams 2005). ductivities. In consequence, the match between peo- ple and jobs will be inefficient. Benefits from spe- • Many Latin American countries were ruled by cialization arise when individuals become more pro- military dictatorships until the 1980s. The juntas ductive due to experience and frequent practice. relied heavily on conscription for military reasons Effective military operations require a considerable (in border disputes and inner conflicts with Mar- degree of training and mastery in handling complex xist movements) as well as for sake of indoctrina- systems. Drafted short-term soldiers are in- tion and social control. During their regimes these ferior to long-term professionals. In combat, the use armies had amassed dismal records of reckless mili- of less advanced , lack of training,

37 CESifo DICE Report 2/2011 Research Reports

and the easy availability of apparently expendable Similar studies are, to our knowledge, not available soldiers lead to higher casualties and “cannon-fod- for the more recent abolitions of military conscrip- der”-type battlefield tactics.1 tion, say, in European NATO members. However, some summary indicators may also provide interest- ing insights. Compared to the 1980s, the “militariza- Consequences of the abolition of conscription tion” of society has declined across NATO, both in terms of the share of the labor force working for the Military expenditures, personnel and budgets military (a decline from an average of 2.7 to 1.1 per- cent between the late 1980s and the late 2000s) and, While quite a number of countries have abolished with the notable exception of the US over the past conscription over the past 20 years, a comparative decade, defense expenditures as a share of GDP, study of their experiences – military, fiscal or econo- which dropped from an average of 4.8 to 2.9 percent mic – is still missing. In parts, this is due to the fact that (NATO 2010). For more detailed comparisons, we the abolition of the draft generally runs parallel to group “old” NATO countries (i.e., members as of other changes (say, shifts in political regimes, geo- 1985) in Figures 3 and 4 according to their recruit- political developments or military technologies) that ment regimes. make it difficult to isolate the “pure” draft effects. Econometric analyses are not available, and compar- As Figure 3 shows, there are no obvious differences isons have to be based on rough summary indicators. in military expenditure trends between “old” NATO

For the government budget, oper- Figure 3 ating a draft system is generally 3 DEFENCE EXPENDITURES OF "OLD" NATO MEMBERS, 1985–2009 cheaper than a professional army. in 2000 prices % of GDP 7 Conscripts are not paid the mar- 2.5 conscription (a) 6 ket value of their labor service and abolition (b) are granted fringe benefits such as 5 UK and CA 2 USA health plans, family support, old- 4

age provisions, etc., on a much 3 smaller scale than professional sol- 1.5 2 diers (if at all). These savings are 1 partially offset by the generally 1 0 smaller size of all-volunteer forces 1985–89 1990–94 1995–99 2000–04 2005–09 5-year averages (allowed for due to increased pro- 0.5 (a) Unweighted averages across 1985 NATO members using conscription over full period, 1985–2009 (DK, ductivity). In an early study, Oneal GE, GR, NO, TK). (b) Unweighted averages across 1985 NATO members which abolished conscription after 1985 (BE, FR, IT, NL, PT, SP). (1992) found that budgetary sav- Luxembourg is omitted as an outlier. 0 ings from conscription in NATO Source: Own12345 calculations based on NATO (2010), p. 9. averaged around 9.2 percent of national military expenditures in Figure 4 1974, but decreased subsequently 3 MILITARY AND CIVILIAN PERSONNEL IN THE ARMED FORCES OF to only 5.7 percent in 1987.Warner "OLD" NATO MEMBERS, 1985–2009

and Asch (2001) report that the in % of labour force 2.54.0 budgetary costs of abolishing con- conscription (a) 3.5 scription in the US in 1973 came at abolition (b) 3.0 UK and CA 2 10 to 15 percent of the 1965 mili- USA 2.5

tary budget (which was chosen as a 2.0

reference point to exclude the 1.51.5

effect of the Vietnam War). 1.0

0.5 1 0.0 1985–89 1990–94 1995–99 2000–04 2005–09 1 th As early as the 19 century German 5-year averages economist J. H. von Thuenen (1875) ar- 0.5 gued that the carnage of ’s (a) Unweighted averages across 1985 NATO members using conscription over full period, 1985–2009 (DK, poorly prepared winter campaigns in GE, GR, NO, TK). (b) Unweighted averages across 1985 NATO members which abolished conscription after 1985 (BE, FR, IT, NL, PT, SP). could escalate only because sol- Luxembourg is omitted as an outlier. diers were easily available through the 0 system of conscription. Source: Own12345 calculations based on NATO (2010), p. 4.

CESifo DICE Report 2/2011 38 Research Reports members that never had, always had or abolished and . The interpretation also changes with the conscription, except for different starting levels. definition of personnel expenditure. Using a narrow- There is no clear correlation between military ex- er definition that only accounts for military (but not penditure and the use of conscription; if anything, for civilian) personnel, Buch (2010) reports that the countries with conscription seem to afford larger share of personnel expenditure rose upon the suspen- military budgets. As Figure 4 (quite unsurprisingly) sion of conscription from 41 to 50 percent in , indicates, the share of people working for the mili- from 30 to 37 percent in , from 43 to 58 percent tary is higher in conscription countries. Remarkably, in – but decreased from 47 to 36 percent in the the reduction in army personnel has been largest in . Again, there seems to be no general rule. conscription countries, both in absolute and in rela- tive terms (which may indicate a problem of unfair- This picture is mirrored by changes in the share of ness due to selective draft calls). A bit more surpris- investment and other expenditures: where there was ingly, there is no indication that the abolition of con- a relative increase in personnel expenditure, there scription led to disruptions in army sizes or military was a relative cut in investments – and vice versa. budgets that would not, in similar orders of magni- tude, occur also in countries that did not change mil- Intangible and indirect effects itary regimes. Although budgetary costs of conscription are small- Conscription involves a distorted factor-price ratio er, its total social costs are likely to be substantially between labor and other military production factors, larger than with an all-volunteer army. The use of leading to an excessive staffing of armies and too lit- compulsion in itself suggests that the real costs of tle investment. This is indeed reflected in military conscription are higher than its budgetary ones, and budgets: as Figure 5 illustrates, military budgets in abolishing the draft will save society these opportu- countries with all-volunteer forces are less person- nity costs. Using empirical methods that were sophis- nel-intensive than in conscription countries. ticated for their time, Oi (1967, 59) conservatively estimated the opportunity costs of the US military Remarkably, the share of personnel expenditures draft in the late to be around USD 5.3 billion has always been highest in the group of countries (in 1960 values), when budgetary personnel expendi- that decided, in the 1990s and early 2000s, to abolish ture amounted to USD 12.7 billion. Kerstens and conscription. The relatively flat trend of these coun- Meyermans (1993) estimate that the social cost of tries in Figure 5 should be interpreted with some the (abolished) draft system in Belgium amounted to caution; the averaging across countries flattens di- twice its budgetary cost. vergent trends in single countries. Upon of conscription, the costs of personnel in the total Economically, military draft is an (in-kind) tax, and defense budget decreased in the Netherlands, Spain shares with all other taxes the feature that it is not and but increased considerably in Belgium neutral but rather induces substantial avoidance ac- tivities and, thus causes economic Figure 5 distortions and deadweight los-

3 SHARE OF PERSONNEL EXPENDITURES, 1985–2009 ses. For example, conscription goes along with various ways of in % of defence expenditures 80 “dodging”, inefficient , 2.5 preemptive emigration, pretended 60 schooling, hasty marriages, bribing 2 recruitment officers, faking med- 40 ical certificates, etc. These hard-

1.5 to-be-measured costs would be conscription (a) 20 abolition (b) saved in case of abolition. UK and CA

1 USA 0 By postponing, interrupting or 1985–89 1990–94 1995–99 2000–04 2005–09 even discouraging higher educa- 5-year averages 0.5 tion and entry into the labor mar- (a) Unweighted averages across 1985 NATO members using conscription over full period, 1985–2009 (DK, GE, GR, NO, TK). (b) Unweighted averages across 1985 NATO members which abolished conscription ket, conscription also has a nega- after 1985 (BE, FR, IT, NL, PT, SP). Until 1995 without France, until 1990 without Spain. Luxembourg is omitted as an outlier. tive effect on the accumulation of 0 Source: Own12345 calculations based on NATO (2010), p. 6. human capital (see Keller et al.

39 CESifo DICE Report 2/2011 Research Reports

2010 and the references therein). At the macro and of decent quality, pay and working conditions level, the disruption of human capital investments have to be competitive with the qualification and by military conscription translates into lower stocks outside options of those one wishes to hire. As of human capital, reduced labor productivity and Warner and Asch (2001) demonstrate for the US, substantial losses in GDP (Lau et al. 2004). From eventually the economists’ predictions have borne 1960 to 2000, GDP growth rates in OECD countries out: efficiency gains of the all-volunteer force were with conscription were lower by up to a quarter per- reaped with the additional benefit that the bud- centage point than in countries with all-volunteer getary costs for reflect its true forces (Keller et al. 2009). This is remarkably large economic costs. given that military expenditures or the size of the military labor force per se do not seem to exert any systematic effect on GDP and its growth (Dunne et Myths about the draft al. 2005). Proponents of the military draft argue that the oppor- Transition problems tunity costs of conscription are tolerable since they are compensated for by additional societal, military or The experiences of countries that switched from con- political advantages of conscription over all-volunteer script to all-volunteer forces show that the transition forces. However, a closer look at these alleged extra is rocky (Warner and Asch 2001; Williams 2005; benefits of conscription reveals that they are largely Rostker 2006; Buch 2010). Initially, all countries seem mythical. to have problems in meeting their requirements for staffing and quality in the military. Enticing new Myth 1: A conscript military is more “representative” recruits to join the forces is difficult, turnover rates of society than a professional army. are high, imbalances across occupational specialties arise, with too few people with the cognitive apti- All-volunteer forces are thought to prey dispropor- tudes, skills, technical abilities and work experiences tionately on the poorly educated, the lower classes, useful in professional armies. Projected savings may ethnic minorities or otherwise marginalized strata of only materialize later than anticipated since, e.g., the society. By contrast, conscription is considered to be military pay, post-service benefits and working and more egalitarian since all are included in universal living conditions to attract new turn out to be service. more costly than foreseen, the extant staff needs re- training, redundant military bases cannot be closed In fact, there is hardly any reason to believe that down immediately, etc. conscription makes the military (more) representa- tive. First, a genuine cross-section of the population Such difficulties may lead some to call into question in the army was never the aim in conscription coun- the efficiency gains attributed to all-volunteer forces. tries: conscription covered substantially less than 50 Thus, the share of the military budget devoted to percent of the population; it excluded women, personnel actually increased during the first few migrants and often certain religious groups, fathers years after the 1973 abolition of conscription in the or gays. Second, even for its main targets (young US, in spite of a reduction in army size (Williams males), the military draft is de facto biased, typical- 2005). Physical or intellectual recruitment standards ly favoring individuals of high socio-economic or for soldiers have been lowered in several countries educational status with shorter terms of service, (Williams 2005; Buch 2010). In the Netherlands, some complete exemptions, legal and illegal buyout options or privileged work conditions (e.g., doctors army units remained understaffed, and France or athletes). Third, the claim that the social compo- appears to have down-scaled its ambitions for out-of- sition in all-volunteer forces is more biased towards area deployability of its armed forces (Buch 2010). the disadvantaged than in conscript armies is still open to debate.2 These non-negligible transitory problems do, how- ever, not qualify as arguments against all-volunteer forces. Rather, they merely reflect to what dramat- ic extent conscription insulated the military from a most elementary feature of standard labor mar- 2 Segal and Segal (2004) report that the US all- is more female, less white, more married, better educated and more kets: if one wishes to hire staff in sufficient number middle-class than the draft-era military.

CESifo DICE Report 2/2011 40 Research Reports

Myth 2: Conscription promotes peace of society in any or all sectors of the economy have so far been unheard of. But even if the alienation It is sometimes argued that a conscript army, which from the rest of society were particularly trouble- draws from all quarters of society, provides a check some in the military, conscription is not a solution. on military adventurism since it cannot be deployed Praetorian tendencies are most likely to emerge without a fair measure of public support. Empirical- from the officers’ corps, which has always consisted ly, this “peacemaker” argument is dubious. In fact, of professionals. Moreover, the democratic controls the military draft may even contribute to a milita- arising from a draft are open to debate. In democra- rization of society by instilling in conscripts the view tic countries like Argentina, , Chile, Greece or that killing for the home country is a patriotic duty. conscription did not help to prevent military The possible legitimization for the production and coups in the past. Conversely, recent experiences use of violence may even raise the likelihood and from Latin America demonstrate that democratiza- severity of armed conflicts. Studying violent inter- tion and the abolition of conscription may be inter- state conflicts from 1886 to 1992, Choi and James twined. Exemplarily democratic countries (like the (2003) find that a military manpower system based UK, the US, , Canada or ) on conscripted soldiers is associated with more mili- have traditionally run all-volunteer armies without tary disputes than professional or voluntary armies. ever facing the risk of regime change or military Similarly,Anderson et al. (1996) conclude that “war- coups. All these observations as well as the econo- like” states are more likely to rely on conscription. metric evidence established by Mulligan and Shleifer Linking conscription with changes this (2005) and Pfaffenzeller (2010) indicate that no picture somewhat (Vasquez 2005); still, there is no causality in whatever direction exists between the convincing evidence that the military draft has any form of government and the structure of armed mediating effect on the likelihood or severity of in- forces in a country. ternational military disputes. Myth 4: Conscription provides better reserves Within societies, conscription may even contribute to brutalization: Using data on Argentine draft lotter- A precautionary argument in favor of military con- ies, Galiani et al. (2011) show that having been con- scription is that it provides manpower reserves to scripted increases the likelihood of developing a augment the regular army in the case of a military criminal record; this holds also for those who served emergency. If this argument should imply that re- during peacetime. serves cannot be maintained with all-volunteer forces, then it is empirically false, as the examples of all coun- Myth 3: Conscription is a better match with democra- tries with professional armies show. Moreover, the cy than all-volunteer armies reserve argument loses validity if are not appropriately prepared for their assignments in case Military conscription is often attributed with a of . greater affinity with democracy than an all-volunteer force. Army structures, which operate on the basis of A key issue for maintaining militarily meaningful re- order and command rather than on voting, are inher- serves is that reservists are paid sufficient compensa- ently non-democratic. Still conscripts may act as tion for their participation in regular exercises. Such mediators between a society and its army, while a contracted (as contrasted to conscripted) reservists professional military tends to alienate from society would make the true opportunity costs of alternative and form a “state within a state”. Advocates of con- military strategies visible and help to allocate re- scription appeal to the soldat citoyen, referring back sources efficiently between personnel and material. to the origins of modern draft in France and Prussia, as a figure that bridges the gap between army and civil society. Obstacles for abolition

Economically, the “isolation” of the military from The inefficiency of conscription results to a great ex- the rest of society is just an example of an increased tent from ignoring comparative advantage and spe- division of labor. In a sense, every specialized work- cialization, which results in higher social costs than er is “alienated” in his work from the rest of society, with a voluntary army. At the same time, there is no but calls for compulsory of all members empirical support for the claim that the use of con-

41 CESifo DICE Report 2/2011 Research Reports

scription would help to protect democracy, promote their populations by forced labor and the desire to social cohesion or tame belligerence. Given that hard- maintain numerically large armies seem to be attrac- ly any argument in favor of the draft survives under tive features of military draft. For developing countries, closer scrutiny, the question arises why so many coun- with their inability to raise sufficient fiscal revenues tries still have a hard time in eliminating conscription and their generally lower opportunity costs of labor, or do not even consider it. A number of obstacles to the in-kind tax of military draft could even be econom- the abolition of the military draft are conceivable: ically efficient from an optimal-tax perspective.

Special interests: Several societal groups benefit from conscription – and therefore favor it. These include Conclusions labor unions (that favor conscription as it keeps potential competitors off the labor market; Conscription is an inefficient form of recruiting sol- see Anderson et al. (1996) for empirical support), diers for an army. As the strikingly positive experi- bureaucracies (for reasons of inertia and fear of or- ences of the UK or the US exemplify, its abolition ganizational change), the military itself (which might therefore is worthwhile – in spite of transition prob- fear a loss in their social status, importance and visi- lems and potentially negative effect on government bility), or regional interests (in areas where military budgets. All-volunteer forces deliver modern, high- bases would become redundant by the absence of technology defense capabilities at lower and more conscripts). Special mention goes to the various or- transparent instantaneous and dynamic costs than ganizations and firms in the welfare industry that their conscript counterparts. benefit from the cheap labor provided by conscien- tious objectors to . Needless to say – Each country’s decision to retain or eliminate con- we hope – all economic arguments against forced scription is motivated by a unique array of different labor in the military also apply mutatis mutandis in factors (although, as in France or , con- the social sector. scription may happen to be abolished in an almost cavalier way). Economic arguments unfortunately Intergenerational issues: From an intergenerational appear to focus primarily on the state in terms of perspective, military conscription is similar to a pay-as- budgetary expenditures and military capabilities. you-go pension scheme: its introduction is a (tempo- The adverse effects of conscription on the well-being rary) way around higher fiscal taxes, the static in- of individual citizens and the substantial excess bur- efficiencies remain largely unnoticed and its dynamic den of the draft tax rarely enter into the discussion. costs only surface with considerable time lags. As for More empirical research on these issues may con- an unfunded pension scheme, starting a draft scheme tribute to the final elimination of peacetime con- means making a “gift” (in the form of a reduced fiscal scription and related forms of forced labor. tax burden) to the cohorts that are beyond draft age at that moment. Such a gift may be revolved from cohort to cohort, but it can never be undone in a Pareto- References improving manner (Poutvaara and Wagener 2007b). Anderson, G. M., D. Halcoussis and R. D. Tollison (1996),“Drafting The abolition of the draft would impose an extra fiscal the Competition. Labor Unions and Military Conscription”, burden on age cohorts beyond the draft age. Since Defence and Peace Economics 7, 189–202. these largely outnumber younger cohorts at or below Buch, D. (2010),“Die Zukunft der deutschen Wehrpflicht”, Working paper FG03-AP01, Stiftung Wissenschaft und Politik, Berlin. the draft age, retaining military draft garners wide- Choi, S.-W. and P. James (2003), ”No Professional Soldiers, No spread political support. The casual observation that Militarized Interstate Disputes? A New Question for Neo- the staunchest advocates of conscription usually come Kantianism”, Journal of Conflict Resolution 47, 796–816. from age groups well above draft age provides support Dunne, J. P., R. P. Smith and D. Willenbockel (2005), “Models of Military Expenditure and Growth: A Critical Review”, Defence and for this view. Peace Economics 16, 449–61.

Friedman, M. (1974), “Why Not a Voluntary Army?”, in J. O’Sullivan Non-democratic and developing countries: In non- and A. M. Meckler, eds., The Draft and Its Enemies: A Documentary History, University of Illinois Press, Chicago/Urbana, 253–63. democratic regimes – which are currently the dominant Galiani, S., M. A. Rossi and E. Schargrodsky (2011), “Conscription users of conscription – popular support for conscrip- and Crime: Evidence from the Argentine Draft Lottery”, American tion is less politically relevant. In these cases, aspects of Economic Journal: Applied Economics 3, 119–36. indoctrination, intimidation and social control, the eco- Keller, K., P. Poutvaara and A. Wagener (2010), “Does Military Draft Discourage Enrollment in Higher Education? Evidence from nomic benefits to the ruling cliques from exploiting OECD Countries”, Finanzarchiv 66, 97–120.

CESifo DICE Report 2/2011 42 Research Reports

Keller, K., P.Poutvaara and A. Wagener (2009), “Military Draft and Economic Growth in OECD Countries”, Defence and Peace Econo- mics 20, 373–93.

Kerstens, K. and E. Meyermans (1993), “The Draft versus an All- Volunteer Force: Issues of Efficiency and Equity in the Belgian Draft”, Defence Economics 4, 271–84.

Lau, M. I., P. Poutvaara and A. Wagener (2004), “Dynamic Costs of the Draft”, German Economic Review 5, 381–406.

Mjoset, L. and S. van Holde (2002), “Killing for the State, Dying for the Nation: An Introductory Essay on the Life Cycle of Conscrip- tion into Europe’s Armed Forces”, in L. Mjoset and S. van Holde, eds., The Comparative Study of Conscription in the Armed Forces, JAI Press, Amsterdam, 4–98.

Mulligan, C. and A. Shleifer (2005), “Conscription as Regulation”, American Law and Economics Review 7, 85–111.

NATO (2010), “Financial and Economic Data Relating to NATO Defence”, Press Release PR/CP(2010)078, NATO Public Diplom- acy Division, Brussels.

Oi,W.Y.(1967),“The Economic Costs of the Draft”, American Eco- nomic Review (Papers and Proceedings) 57, 39–62.

Oneal, J. R. (1962), “Budgetary Savings from Conscription and Burden Sharing in NATO”, Defence Economics 3, 113–25.

Pfaffenzeller, S. (2010), “Conscription and Democracy: The Mythology of Civil-Military Relations”, Armed Forces & Society 36, 481–506.

Poutvaara, P. and A. Wagener (2011), “The Political Economy of Conscription”, in C. Coyne and R. Mathers, eds., Handbook on the Political Economy of War, Edward Elgar, London, 154–75.

Poutvaara, P. and A. Wagener (2007a), “Conscription: Economic Costs and Political Allure”, Economics of Peace and Security Jour- nal 2, 5–15.

Poutvaara, P. and A. Wagener (2007b), “To Draft or Not to Draft: Inefficiency, Intergenerational Incidence, and Political Economy of Military Conscription”, European Journal of Political Economy 23, 975–87.

Rostker, B. (2006), I Want You! The Evolution of the All-Volunteer Force, The RAND Corporation.

Sandler, T. and K. Hartley (1995), The Economics of Defense, Cambridge University Press, Cambridge.

Segal, D. R. and M. W. Segal (2004), “America’s Military Popula- tion”, Population Bulletin 59, 1–40.

Smith, A. (1776), An Inquiry into the Nature and Causes of the Wealth of Nations, reprinted in W. B. Todd, ed., (1976), Glasgow Edition of the Works and Correspondence of Adam Smith, vol. I, Oxford University Press, Oxford.

Thuenen, J. H. von (1875), Der isolierte Staat in Beziehung auf Landwirtschaft und Nationalökonomie, 3rd ed., Wiegardt, Hempel & Parey, Berlin (English edition: Isolated State (1966), Pergamon Press, Oxford/New ).

Vasquez, J. P. (2005), “Shouldering the Soldiering: Democracy, Conscription, and Military Casualties”, Journal of Conflict Resolu- tion 49, 849–73.

Warner, J T. and B. J.Asch (2001),“The Record and Prospects of the All-Volunteer Military in the United States”, Journal of Economic Perspectives 15, 169–92.

Williams, C. (2005), “From Conscripts to Volunteers. NATO’s Transitions to All-Volunteer Forces”, Naval War College Review 58, 35–62.

43 CESifo DICE Report 2/2011