[T]here can be little doubt that no responsible Government can regard with equanimity a rapidly shrinking population, a dangerous strain on the adaptability of the industrial system, and a community containing an ever-declining proportion of young people and an ever-increasing burden of the old – not to speak of the possible threat of enemies at the gate with a far greater man-power than our own and with similar war-technique.
Eva M. Hubback (Reference Hubback1937: 272–3)
The more, in fact, that the waging of war has come to require a total effort by the nation the more have the dependant needs of the family been recognized and accepted as a social responsibility.
Richard M. Titmuss (Reference Titmuss and Titmuss1958: 84)
Introduction
Compared to social security branches such as old age, health, and unemployment insurance, family allowances were introduced at a much later moment in time. In most countries, they were only created in the 1930s and 1940s (Perrin Reference Perrin1969; Abbot and De Viney Reference Abbott and DeViney.1992; Gauthier Reference Gauthier1996; Kuhnle and Sander Reference Kuhnle, Sander, Béland, Morgan, Obinger and Pierson2021). Declining birth rates, poverty in families with many children and the associated struggle for a social wage are seen in the literature as driving forces for the introduction of cash benefits for families (Glass Reference Glass1967; Wennemo Reference Wennemo1992; Gauthier Reference Gauthier1996). However, this does not explain the concrete timing of program adoption or cross-national differences in benefit generosity. While the comparative literature has already examined the generosity of family benefits in 18 Western countries (Wennemo Reference Wennemo1992; Montanari Reference Montanari2000), this paper is interested in the timing of the introduction of family allowances in this group of nations. What explains the adoption of universal family allowances in so many countries within a rather short period of time?
We put pronatalist policies related to war preparation and the dreadful socioeconomic and demographic ramifications during and after the Second World War at the center of our explanation. Although pronatalism has long been seen as an important driver of state intervention in family policy (Glass Reference Glass1967; Berelson Reference Berelson1974; Pedersen Reference Pedersen1993; Gauthier Reference Gauthier1996), the connection with war and war preparation efforts has been widely neglected in explaining the timing of reforms in family policy. Using regression analysis and four brief case studies, we illustrate that war and the negative economic and demographic effects caused by war prompted the introduction of family allowances in many countries. As the quotes above demonstrate, such a warfare-welfare nexus was already recognized by many contemporary scholars (see also Waggaman Reference Waggaman1939: 1027). However, apart from case studies (e.g. Pedersen Reference Pedersen1993), the impact of war on the introduction of family allowances has never been examined for a larger number of countries.
Our argument as to why warfare led to the introduction of family allowances from a theoretical perspective is twofold. First, in the context of declining birth rates as well as fundamental changes in military organization and the nature of warfare, pronatalist population policies increasingly gained military importance in the 20th century. We argue that family allowances were used as an instrument to increase both the birth rate and, consequentially, military power. Second, the horrors and ravages of war had dramatic negative effects on the economic situation of families, the demographic structure of the population, and, in consequence, the future labor supply. Many countries responded by introducing family allowances to alleviate poverty in families and to increase the birth rate to offset the negative demographic effects caused by war and its aftershocks.
To test the effect of the Second World War on the introduction of family allowances in 18 Western countries, we first use regression analyses and find a significantly higher likelihood for the introduction of family allowances in wartime and especially immediately after the end of military conflict. Furthermore, we conduct brief case studies on four belligerent countries (France, Germany, Italy, and Japan) that deviate from this pattern, as they introduced family allowances either before the outbreak of war or long after armistice. The country outlines show that war also matters in these cases. In Italy, Germany, and France, war preparation swayed the introduction of family allowances in the 1930s, while the late adoption in Japan resulted from war-induced peculiarities, which postponed the introduction of family allowances for decades.
The paper is organized as follows. We start with theoretical considerations on the nexus between warfare and family allowances. Next, we present the findings of the regression analyses and, subsequently, of the four brief case studies. The final section concludes and outlines avenues for future research.
War and family allowances: theoretical considerations
Table 1 shows the timing of the introduction of family allowances in 18 countries. We argue in this section that war and war preparation are important causal factors behind the clustering of program adoption around the Second World War.
Table 1. Introduction of family allowances 1900–2000
Source: Tonelli et al. (Reference Tonelli, Böger, Son, Buhr, Drobnič and Huinink.2021) used the following coding criteria for the year of introduction: the program must be the first policy established by a national or federal law with a nationwide scope, and the benefit must be given to the recipients in the form of a cash transfer. The benefits can be universal, aimed either at the residents or citizens of a country, or employment-based, aimed at people employed in specific productive sectors of the economy, such as agriculture, industry, or commerce.
The second half of the 19th century witnessed fundamental changes in military technology and organization. Advances in military technology brought about a huge increase in the firepower and destructiveness of weapon systems, while more and more countries on the European continent introduced universal conscription, mostly in response to military defeats and rising international tensions (Obinger et al. Reference Obinger, Petersen and Starke2018). As a result of both developments, the nature of warfare changed fundamentally. The old cabinet wars, dubbed the “sport of kings” by Titmuss (Reference Titmuss and Titmuss1958: 78), were replaced by industrialized mass warfare.
Parallel to these developments, the birth rate in Europe declined by about 50% between 1870 and 1940 (Teitelbaum and Winter Reference Teitelbaum and Winter1985: 14). This decline was of military relevance because the number of mobilizable soldiers was equated with military strength and a nation’s world standing. However, the combat power of an army not only depended on the size of a country and its past births rates but also on the physical fitness of the young male population. Moreover, the changes in weapons and communications technology required higher skills in the armed forces (Aghion et al. Reference Aghion, Jaravel, Persson and Rouzet.2019). In other words, with the rise of industrialized mass warfare, the quantity and quality of the population gained significant military importance (Titmuss Reference Titmuss and Titmuss1958).
With regard to the quantity of the population, we argue that the military interest in pronatalist population policies contributed to the adoption of family allowance. Along with measures such as better health care for mothers and children, tax deductions for families with children or banning abortion and contraceptives, providing cash benefits to families is a possible vehicle to raise birth rates because cash benefits lower the costs of family formation and maintenance. Countries planning a war of expansion or fearing a military attack are therefore likely to have implemented such a pronatalist population policy even before the outbreak of war in an effort to increase military power.
The horrors of industrialized mass warfare massively exacerbated the pressure of the demographic problem. Acts of war on the home territory, rampant diseases, poverty, malnutrition, and famine led to a dramatic decline in fertility and growing infant mortality. Moreover, the death of millions of young men of reproductive age in combat exacerbated the unfavorable demographic situation and nurtured concerns about an insufficient labor supply in the future. War and its aftershocks also worsened the socioeconomic situation of families with many children. Apart from the fact that wages typically did not take the number of children into account (ILO 1924), inflation, housing shortages, unemployment, and the death of the family breadwinner led to precarious economic conditions for many families in war-torn countries. The dire situation of families did not disappear with the end of the war, but continued for several years after armistice. All these factors should make the introduction of family allowances in wartime and especially in the crisis-ridden immediate post-war period more likely.
By contrast, the debates on the quality of the population had ambivalent consequences for social policy. On the one hand, there is empirical evidence that the growing military interest in improving combat power motivated reforms in labor protection legislation, primary education, public hygiene, and nutrition, since armies needed not only soldiers who were literate and could calculate but also healthy and physically strong recruits (e.g. Titmuss Reference Titmuss and Titmuss1958; Dwork Reference Dwork1987; Aghion et al. Reference Aghion, Jaravel, Persson and Rouzet.2019; Dörr and Grawe Reference Dörr and Grawe.2020). On the other hand, the growing concerns about the quality of the population contributed to the rise and radicalization of eugenics, which either rejected social protection of the weak for Social Darwinist reasons or, in its most extreme variant in Nazi Germany, paved the way for forced sterilizations and the mass murder of people with disabilities.
In sum, we argue that pronatalist population policies in the context of war preparation and the devastating social, economic, and demographic effects caused by industrialized mass warfare provided an important impetus for the introduction of family allowances in many Western countries. However, we contend for three reasons that the Second World War, and not the Great War, was the decisive game changer for the introduction of universal family allowances. First, demographic pressure was higher in the 1940s as fertility rates strongly declined in the interwar period. Second, the welfare state in many countries was still in its infancy in 1918. Hence, the introduction of other social protection schemes such as health insurance, old age pensions, unemployment insurance (related to military demobilization and lay-offs in the war industries), and provisions for millions of disabled veterans and the survivors of killed servicemen was a political priority. Third, international platforms, or international organizations (IOs) promoting and supporting policy adoption in this field, had a much greater importance at the time of World War 2 than World War 1 (Béland et al. Reference Béland, Marchildon, Mioni and Petersen.2022). However, and in line with our theoretical argument, there is ample empirical evidence that the Great War was already an important turning point for state intervention in other family policies, such as child and youth welfare, maternal leave, and child care (e.g. Montanari Reference Montanari2000; Böger et al. Reference Böger, Son, Tonelli, Windzio, Mossig, Besche-Truthe and Seitzer2022; Son Reference Son2023). Moreover, cash benefits were paid to the families of conscripted soldiers to compensate for the loss of income caused by war service. Finally, and more important for the sake of this paper, several countries introduced family allowances for specific professions for the first time, notably for employees in public servicesFootnote 1 . This policy shift was clearly a response to war-induced social needs: The “chief cause of the adoption of a system of family allowances in various countries during the war was the increase in the cost of living, which led to a reduction of real wages in a number of industries; this caused considerable hardships, especially to workers with large families” (ILO 1924: 7). When a second and even more brutal world war set off shock waves across the globe, again causing tremendous socioeconomic and demographic problems in belligerent and non-belligerent countries alike, the time was ripe for the introduction of universal family allowances within a relatively short period of time.
In the next two sections, we subject this hypothesis to an empirical test using a combination of quantitative and qualitative methods. First, we apply regression analyses across 18 countries to investigate whether the introduction of family allowances is associated with World War II. We then examine country outlines of four belligerent nations (Germany, Italy, France, and Japan) where the introduction of family allowances appears at first glance to be unrelated to the Second World War given the relatively higher time lag between war exposure of these nations and program adoption.
Macro-quantitative analysis
In the following, we test whether the economic and demographic effects caused by World War II affected the introduction of family allowances in a macro-quantitative framework.
Data and measurement
Our dependent variable is the introduction of a national family allowance scheme in 18 countries.
The dependent variable is coded 1 when a country introduced a family allowance program in a given year and 0 otherwise. To analyze the influence of World War II on the timing of program adoption, we estimate random effect logit models using a standard maximum likelihood procedure. The countries are considered until family allowances have been introduced. Once this event has occurred, the country is excluded from the analysis. Ordinary probit or logit regressions rest on the assumption that the observations are temporally independent. However, the likelihood of adopting family allowance programs should not be time independent but rather increase over time. Ordinary logit would be misleading, and the standard errors would be biased in that case. We therefore include a linear trend variable to control for the time dependency of our dependent variable.Footnote 2 The period of observation ranges from the 1920s, when the first family allowance program was introduced, until 1970, i.e., shortly after the last family allowance scheme was introduced in Japan.
To test the hypothesis whether World War II affected the likelihood of introducing family allowances, we use two different indicators as key independent variables. First, in Table 2 model 1, we include a wartime variable (WW II) equalling 1 during wartime and the three years after the end of war and 0 otherwise.Footnote 3 Note that this is a conservative test strategy as we have argued in the theory section that war may not only exert an influence during wartime itself and the immediate post-war period but also in the period of war preparation, i.e., before the outbreak of war. To test period-specific wartime effects, we split the WW II variable and include a dummy for the war period and a separate dummy for the immediate three-year post-war period in model 2.Footnote 4 As the second key independent variable, we use an index of war intensity developed by Obinger and Schmitt (Reference Obinger and Schmitt2018) in models 3 and 4. The index measures the extent to which a country has been exposed to war. This allows us to test whether the degree to which a country has been affected by violence and combat activities on its home territory makes a difference to introducing family allowances. The index is the unweighted sum of three standardized indicators which capture the different aspects of mass warfare, namely (i) the duration of war, (ii) the military and civilian casualties as a percentage of the pre-war population, and (iii) the presence of combat activities and violent occupation on home territory. All subindicators are standardized, ranging from 0 to 1 and subsequently summed. In a final step, we constructed a composite index of war intensity which is the average of the three standardized indicators and therefore reflects several dimensions of warfare (WW II Intensity). Hence, our index ranges from zero to one and shows high values if a country was heavily affected by World War II and low values otherwise (see Table A1 in the appendix). During the war period and the three immediate post-war years, the index measures the intensity to which a country has been affected by war. In all other peacetime years, the index equals zero. We hypothesize that the likelihood of introducing family allowances is higher in countries highly involved in war activities.
Table 2. Determinants of introducing family allowances – regression results
Notes: Odds ratio are reported; standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1, note that standard errors for odds ratio are calculated as follows: se(OR) = exp(_b[_var])*_se[_var].
In addition to our key independent variables, we control for alternative explanatory factors discussed in the literature as important determinants of welfare legislation. Our basic models in Table 2 include the following variables: GDP per capita is the key variable of a functionalist welfare state theory (Wilensky Reference Wilensky1975). Data on GDP per capita (in 2011 US Dollars) is provided by the Maddison Project Database (see Bolt et al. Reference Bolt, Inklaar, de Jong and van Zanden2018). The level of economic development is an indicator of socioeconomic modernization and a proxy for governments’ capacity to finance expensive social programs. According to the “logic of industrialization” we should expect a positive impact of economic affluence on the introduction of family allowance programs. Moreover, we include growth of GDP, also using data from the Maddison Project Database (2018). We hypothesize that in times of economic prosperity the financial leeway for introducing family allowances should be greater. Moreover, we include the fertility rate to capture the demographic pressure outlined in the theoretical section. We would expect that countries with a comparatively low fertility rate were more likely to adopt family allowance programs than countries with high fertility rates. Additionally, we include a dummy variable capturing government ideology. The dummy variable equals 1 in the case of a right-wing head of government and 0 otherwise (Right-wing government). The data is taken from Brambor and Lindvall (Reference Brambor and Lindvall.2018) and we assume that right-wing governmentsFootnote 5 are more nationalist and support pronatalism and the military more strongly than left-wing ones. Lastly, governments’ capability to initiate family policies might depend on the institutional setting of their political system. To measure institutional constraints that might impede governments’ capacity to implement policy change, we compiled an additive and time-varying index of institutional veto points (ranging from 0 to 5) which reflects the existence of the following five institutions: (i) judicial review by constitutional courts, (ii) proportional representation, (iii) referenda at the national level, (iv) federalism, and (v) presidentialism. Table 2 reports our main empirical findings.
Table 2 yields several interesting results. With respect to our main independent variables, the results clearly show that World War II has driven the introduction of family allowances. Coefficients larger than 1 indicate such an effect while coefficients smaller than 1 indicate that the associated variable decreases the likelihood of introducing family allowance programs. The coefficients for all dummy variables capturing the wartime and the post-war period of World War II are above 1 and highly statistically significant. The probability of adopting a family allowance program during and in the immediate aftermath of World War II is more than six times higher than during peacetime years not directly following war. This is a remarkable result considering the size and statistical significance of the coefficients. When comparing the results for wartime and the immediate post-war period (model 2), it becomes clear that the introduction of family allowances was triggered in the short period after the military conflict rather than during wartime. The coefficient of our war variable (WW II post-war period) for the three years after World War II is statistically significant at the 1% level and estimated to be almost three times higher than that for the wartime period (WW II wartime), which is statistically significant at the 5% level.
In models 3 and 4, we test whether countries highly affected by war are more likely to introduce family allowance programs compared to, for example, neutral countries or whether the introduction of family allowances is a more general phenomenon induced by the transnational socio-economic shock waves set off by war. The empirical evidence points to the importance of wartime rather than war intensity for explaining the introduction of family allowances. Even though the coefficient of our war intensity index (WW II Intensity) in model 3 is statistically significant, the effect disappears when taking the time effect into account. This is not surprising, considering the fact that, for example, neutral countries such as Switzerland were surrounded by belligerent nations. Socioeconomic shock waves therefore also affected neutral countries and war-induced inflation worsened the living situation of large families in particular in almost all European countries.
The results for the control variables also reveal some interesting findings. Most importantly, the ideological orientation of the head of government is statistically significant. The likelihood of introducing a family allowance program under a right-wing head of government is in most of the models five times higher than under a center or left-wing head of government.Footnote 6 This result fits very well to findings showing that (radical) right-wing and right populist parties strongly support the expansion of family benefits (Chueri Reference Chueri2022; Akkerman Reference Akkerman2015; Ennser-Jedenastik Reference Ennser-Jedenastik2017, Reference Ennser-Jedenastik2022). Even more than Christian-democratic parties, right-wing parties hold pro-natalist positions and promote policies that “support and reward having children” (Ennser-Jedenastik Reference Ennser-Jedenastik2022: 158; Akkerman Reference Akkerman2015; Fenger Reference Fenger2018).
Besides government ideology, we only find statistically significant results for institutional constraints in models 2 and 3. In line with the theoretical expectations, the higher the number of institutional constraints, the less likely the introduction of family allowances. Interestingly, the fertility rate does not seem to play a role for the introduction of family cash benefits. One reason might be that in the first half of the 20th century, declining birth rates were a common phenomenon in many of the 18 countries and were considered a problem independently of the specific fertility rate. All other variables included in our models are not statistically significant.
In a next step, we test the robustness of our findings by including alternative explanatory variables (see Table 3). In model 1, we include the regime type and the age of the welfare state as control variables. Regime type is measured with an index on the level of democracy provided by the “Polity IV Project” (Marshall et al. Reference Marshall, Gurr and Jaggers.2014). This indicator ranges from –10 (autocracy) to +10 (full democracy). We assume that democracies are more likely to provide family allowances due to their greater responsiveness to social needs. The age of the welfare state is captured by the year when the first social security law was introduced (see Schmidt Reference Schmidt2005: 188). The adoption of family allowances should be more likely and happen earlier in mature welfare states. Model 2 includes the number of neighboring countries that have already introduced family cash benefits (own coding) to capture the possibility of policy diffusion processes. Additionally, we include a dummy measuring whether female suffrage exists in each country. We hypothesize that countries with female suffrage are more likely to introduce family allowances than countries without female suffrage. Model 3 tests whether ILO membership affects the promotion and adoption of family allowance schemes as argued by Gauthier (Reference Gauthier1996: 69). In line with previous research we expect ILO membership to have an accelerating effect (Schmitt et al. Reference Schmitt, Lierse, Obinger and Seelkopf2015). We integrate a dummy variable that equals 1 if a country is a member of the ILO and 0 otherwise (ILO membership). Additionally, we include the cabinet seats held by Christian democratic parties (Parlgov)Footnote 7 to account for the family-affirmative Christian ideology that should push the introduction of family allowances in model 3. In the final model 4, we test the influence of urbanization on the introduction of family allowance schemes measured by the percentage of people living in urban areas. In urban centers, the need for family allowances should be greater due to the absence of other more traditional networks of family support such as the extended family. Data is taken from Vdem (see Coppedge et al. Reference Coppedge2023). Lastly, we include an alternative measurement of fertility rates to avoid bias due to short-term fluctuations. In model 4, the fertility rate is therefore calculated as a five-year moving average.
Table 3. Determinants of introducing family allowances – robustness checks
Notes: Odds ratio are reported; standard errors in parentheses. ***p < 0.01, ** p < 0.05, *p < 0.1, note that standard errors for odds ratio are calculated as follows: se(OR) = exp(_b[_var])*_se[_var].
The results presented in Table 3 support our main findings reported in Table 2. The wartime variable including the immediate post-war period is substantive in size and statistically significant at the 1% level in all models. This again underpins the effect of war on the introduction of family allowance programs. As in Table 2, the results show that right-leaning governments are more likely than other governments to introduce family allowances. Apart from the mentioned variables and the time variable, only institutional constraints are statistically significant in one model. A high number of institutional veto points inhibits the introduction of family allowance programs. All other control variables included in the robustness section are not statistically significant.
Brief country outlines
In this section, we focus on those countries where war exposure appears to be least related to the timing of family allowances. By taking a closer look at the rationales for introducing family allowances in France, Germany, Italy, and Japan, we use primary sources (e.g., archival material and speeches of policymakers) and secondary literature to show that even in these countries, war plays a crucial role for implementing family allowances in the period of war preparation. All these nations planned a war of expansion in the 1930s or, as in the case of France, feared a military attack. Relying on primary and secondary sources, we show that three of these great powers introduced family cash benefits largely for military reasons and as part of a wide range of pronatalist policies. To understand the broader context of program adoption, we need to look back to the First World War and, in the case of France, even to the last quarter of the 19th century.
France
Compared to all other European countries, France’s birth rate started to decline very early on. This phenomenon, known as dénatalité, was seen as a threat to national security in the context of increasing military competition with Prussia: “In a country whose fertility had begun to fall decades before the rest of Europe, the belief that a state’s power depended on its birth rate had emerged in the 1860s in the face of the military threat from Prussia” (Rosental Reference Rosental2012: 540). The traumatic military defeat in 1870/71 and the resulting loss of Alsace-Lorraine heightened concerns about a prolonged period of military weakness and not only triggered reforms in primary education (Aghion et al. Reference Aghion, Jaravel, Persson and Rouzet.2019: 380–82), but also fuelled pronatalist ideas. Numerous pronatalist interest groups such as the alliance nationale contre la depopulation (founded in 1896) emerged and a cross-party Parliamentary Group for the Protection of the Birth Rate was formed, to which a majority of MPs belonged by 1914 (Dörr Reference Dörr2020: 75). The pronatalists’ numerous proposals to increase the birth rate also included reforms in family policy. In 1913, the state provided financial aid to families with at least three children and financial support for pregnant women (Tomlinson Reference Teitelbaum and Winter1985: 407; Dörr Reference Dörr2020: 89). The First World War further exacerbated the demographic problem in light of a comparatively high number of war casualties and a significant decline in the birth rate. During the First World War, family allowances were introduced for civil servants (Gauthier Reference Gauthier1996: 43). Fiscal problems after the end of the war prevented financial support for families (Tomlinson Reference Tomlinson1985). Instead, for pronatalist reasons, a very restrictive abortion law was passed in 1920, and the distribution of contraceptives was made a punishable offence. As a symbolic measure, a Mother’s Day (fête des mères) and the awarding of medals to mothers with many children were introduced in 1920 (Huss Reference Huss1990: 43). A significant development in terms of family policy was the introduction of the Social Security Act, passed in 1928, which improved social protection of families and mothers (Dörr Reference Dörr2020: 94–5).
When it came to family allowances for employees in the private sector, philanthropic and Catholic employers, who shared the pronatalist ideas of the Alliance Nationale, initially took the lead (Huss Reference Huss1990: 55; Pedersen Reference Pedersen1993; Gauthier Reference Gauthier1996: 42–3; Dörr Reference Dörr2020: 75–6). In the 1930s, however, the state increasingly began to get involved as family allowances in the private sector were only provided by large companies. The so-called Loi Landry, enacted in 1932, generalized the principle of family surcharges for all employees in industry and commerce with at least two pre-school aged children and employers were required to join a compensation fund (caisses de compensation). State intervention was still limited, however. Rearmament and pronatalism in Nazi Germany and fascist Italy brought population policy even more into focus for military reasons: “The Alliance Nationale worried incessantly about the belligerent intentions of France’s more populous neighbour, especially after the Nazis inaugurated comprehensive pronatalist policies of their own” (Pedersen Reference Pedersen1993: 359). Moreover, the massive decline in births caused by World War I had a strong impact on troop strength in the 1930s. The decimated birth cohorts of the Great War now reached mustering age, leading to a thinning out of the army – a scenario that Alliance Nationale and military planners had always warned against (Huss Reference Huss1990: 56). In 1937, the number of births during peacetime reached an all-time low (Huss Reference Huss1990: 62). The impact on the military was drastically described by the later prime minister and member of the Alliance Nationale, Paul Reynaud, in 1937: “There is a single factor which dominates everything: the demographic factor. Forty-one million Frenchmen face sixty-seven million Germans and forty-three million Italians, these last two linked by the Berlin-Rome axis … As far as numbers are concerned we are beaten” (quoted from Tomlinson Reference Tomlinson1985: 412; cf. Dörr Reference Dörr2020). These military concerns and the ubiquitous demographic panic fuelled by pronatalist pressure groups forced the government to improve family benefits (Pedersen Reference Pedersen1993: 378, 388). In 1938, family allowances were extended to the agricultural sector and one year later, the décret relatif à la famille et à la natalité françaises (code de la famille) was enacted. The French Family Code contained various measures with a clear pronatalist thrust, such as marriage loans for couples in rural areas (to prevent a rural exodus), better care for pregnant women, measures to curb infant mortality, a tightening of abortion law, and a reorganization of family allowances. A birth bonus was now paid for the first child, and all workers with two or more children up to the age of 14 (17 if in education or apprenticeship) received family allowances (Monthly Labor Review 1939: 917–9; Doublet Reference Doublet1948: 222). This program was financed through higher taxes paid by single people and childless couples (Monthly Labor Review 1939; Tomlinson Reference Tomlinson1985). The government justified the law with national and colonial security interests: “Among the incalculable consequences of the low birth rate in France, the heightening danger of attacks from without is the most pressing: How are we, a country with a declining workforce and reduced fighting population, able to respond to the threat to our metropolitan areas and imperial borders, posed by nations whose increase in number favors their ambition? We risk letting our military forces and capacity for economic armament wane; the country is ruined little by little” (Journal officiel de la République française, no. 178, 30 juillet 1939: 9607; our translation).
Germany
Due to Germany’s military successes in the wars of unification and high birth rates, depopulation fears did not play a major role in public and military debates until 1914 (Dörr and Grawe Reference Dörr and Grawe.2020). The outbreak and long duration of the First World War marked a turning point. The disastrous supply situation and enormous losses of the war alarmed not only social reformers but also the German military leadership in the late war period. Social policy experts such as Friedrich Zahn pleaded for an expansion of family policy after the war, because the social damages of war would endanger the “quantity and quality of our future people” (Zahn Reference Zahn1916: 449–50). At the center of his proposals was the promotion and support of families with many children. In addition to child allowances and transfer payments to large families, he advocated for tax relief for large families and higher taxes for bachelors and also proposed a housing offensive. For pronatalist reasons, Zahn recommended reducing married women’s employment and restricting abortions and the distribution of contraceptives (Zahn Reference Zahn1916: 457, 474). The 3rd Supreme Army Command under Hindenburg and Ludendorff was also increasingly concerned about the high war losses. In a letter to the Minister of War, Ludendorff noted in 1916 that the enemy had “an almost inexhaustible supply of human material,” while “our human supplies are limited” (Ludendorff Reference Ludendorff1922: 63). Ludendorff commissioned a memorandum on the “German People’s and Military Strength” (Federal Archive-Military Archive PH 3/446), which is a remarkable document in that it propagates comprehensive welfare reforms to increase the birth rate and to strengthen military power. The proposed measures largely coincide with Zahn’s recommendations and ranged from housing and urban development, hygiene and nutrition issues, to occupational health and safety, family policy to tax policy (Dörr and Grawe Reference Dörr and Grawe.2020). While during both the late war period and the crisis-ridden years of the Weimar Republic, financial means for such extensive reforms were lacking, the Nazis took up several of the aforementioned proposals. They were obsessed with issues of population policy and Friedrich Burgdörfer, a former student of Zahn, became the leading demographer of the Nazi regime. The thrust of Nazi population policy is clearly spelled out in the justification for the “Law for the Prevention of Hereditary Diseases” enacted in 1933: “Since the national uprising,Footnote 8 the public has been increasingly preoccupied with questions of population policy and the ever-increasing decline of the birth rate. However, it is not only the decline of the size of the population that gives rise to the most serious concerns, but to the same extent the increasingly apparent quality of the hereditary constitution of our people. While the hereditarily healthy families have for the most part switched to the one-child and no-child system, countless inferior and hereditarily burdened people are reproducing without restraint – their sick and antisocial offspring are a burden to society as a whole” (Federal Archive RMdI. II A 1079/6.7; emphasis in the original, our translation). This quote shows that for the Nazis and their racial ideology, the quantity and quality of the population were equally important. In terms of quantity, the Nazis feared that a continuous decline in the birth rate would in the long run lead to the “death of the people” (Volkstod). Burgdörfer noted in 1932: “It is not defeat and political oppression that ultimately determine the future of a people, but the strength of its biological will to live. […] A people can only be extinguished and wiped out by itself, by its own infertility. This is the most dangerous enemy of every people. No people actually dies out, it is “born out.” This is the danger facing our German people” (Burgdörfer Reference Burgdörfer1932: XVI, our translation). Ludendorff, who had commissioned the memorandum of the Supreme Army Command during World War I and later became a Nazi sympathiser, viewed declining birth rates as an “immeasurable danger” for the German military. He therefore advocated for health-promoting and pronatalist measures imbued with eugenic principles with a view to creating “a healthy, reproducing population, which would strengthen the army and be able to wage and endure a total war” (Ludendorff Reference Ludendorff1937[1935]: 23).
Apart from a restrictive regulation of abortion and pronatalist propaganda such as the Cross of Honour of the German Mother and celebration of a Mother’s Day, the Nazis relied on cash benefits and tax relief for large families to encourage a higher birth rate. An income tax reform in 1934 relieved the burden on large families (Glass Reference Glass1967[1940]: 300). Marriage loans were introduced in 1933. Married couples could apply for an interest-free loan, amounting to between 600 and 1,000 Reichsmark, provided the wife gave up her job. In order to increase birth rates, the amount that had to be repaid was reduced by a quarter for each child born. Child allowances were also introduced for pronatalist reasons. In October 1935, one-time child allowances were granted for large families, which were finally converted into a regular child allowance in 1936 (Neumaier Reference Neumaier2019: 210–11). The benefit (10 Reichsmark per month) was initially only paid from the fifth child under 16 years of age onward. However, the group of beneficiaries was soon expanded. At the end of 1940, child benefit was paid from the third child under the age of 21 (Recker Reference Recker and Pohl1991: 259). Moreover, education allowances (covering school fees, learning materials, school transportation) were introduced for families with more than four children in 1938 (Ruhl Reference Ruhl1991: 483).
Italy
Fascist Italy also fought a “battle for births” in the interwar period (Forucci Reference Forucci2010) and resorted to similar measures as France and Nazi Germany, namely punitive taxes on the childless, pronatalist propaganda, restrictive abortion regulations, and social benefits. And, as in the other two countries, military motives played an important role in the creation of family allowances: “Mussolini praised fecundity, marriage, and procreation. In reality, though, he was pushing for larger families to meet the manpower needs of his military machine. The fascists would need a steady supply of men to fulfill Mussolini’s dreams of creating a new Rome in the twentieth century” (Forucci Reference Forucci2010: 7).
In 1926, a graduated tax on bachelors was introduced with a view to increasing the birth rate. The revenue collected from the tax was given to the National Agency for the Protection of Maternity and Infant Welfare (Opera Nazionale per la Maternità e Infanzia – ONMI) established one year previously (Glass Reference Glass1967[1940]: 236). Focussed on women living in rural areas, ONMI disseminated information on public health, hygiene, and parenting, so as to contain maternal and infant mortality, and was engaged in pronatalist propaganda. OMNI also organized leisure activities for children and provided welfare services for needy mothers and their children. Ferrera (Reference Ferrera2018: 115) notes that this “Fascist policy on mother- and child care was part of a larger strategy aimed at restoring the glories of imperial Rome and restoring the central role that, according to Mussolini, Italy merited in the international arena.” Mussolini openly expressed this focus in a famous speech given in May 1927, in which he also justified the measures already taken to increase the birth rate: “We must be vigilant […] in watching over the destiny of our race, we must take care of our race, beginning with maternity and infancy. This is the purpose of the National Agency for the Protection of Maternity and Infant Welfare […]: In this country there are 5,700 institutions devoted to maternity and infancy, but they do not have sufficient funds. Hence the tax on bachelors, and perhaps in the future there will be a tax on childless marriages […]. I introduced this tax to give the Nation a demographic boom.” And looking to the future, the dictator stated: “I affirm that the most fundamental if not essential element in the political power and therefore economic and moral power of nations, is their demographic strength. Let’s speak frankly: What are 40 million Italians compared to 90 million Germans and 200 million Slavs? Let’s turn to the West: what are 40 million Italians compared to 40 million Frenchmen, in addition to 90 million colonial inhabitants, or compared to 46 million Englishmen, in addition to the 450 million colonials? Gentlemen, if Italy is to count for anything in the world, then she must reach a population of no less than 60 million inhabitants by the middle of this century” (Mussolini Reference Mussolini1927: 16–7).
Against this background, further pronatalist measures were introduced in the following years. Income tax for bachelors and childless couples was raised in 1928, while large families were relieved. In the same year, abortion laws were tightened and advertising for contraception was criminalized. In 1929, a law was enacted that provided for preferential hiring of people with many children in the civil service. This was later extended to private companies and preference was also given to large families in the allocation of housing (Glass Reference Glass1936: 107–8). Other measures included a nuptiality bonus for public employees (1930), an expansion of maternity insurance (1930), and loans for young couples in 1937 (Ferrera Reference Ferrera2018: 115).
One important pillar of fascist family policy was the provision of family cash benefits. In 1929, family allowances were introduced for state employees, which were extended to industry and commerce in 1934. This regulation was based on an agreement between the Fascist Confederation of Manufacturers and the Fascist Confederation of Industrial Workers (Glass Reference Glass1936: 110). Benefits were provided for workers with children under 14 years of age (beginning with the second child) and whose working time did not exceed 40 hours per week. Under the auspices of a National Family Allowance Fund, the benefit was financed by equal contributions from employers and employees. Domestic workers were, however, excluded (Monthly Labor Review 1935: 653). The first child became entitled to benefits in 1935 and since 1936, family allowances were paid regardless of the hours worked per week. In addition, the state contributed to the financing and family allowances were extended to new sectors. In 1937 und 1938, coverage was expanded further (Waggaman Reference Waggaman1939: 1036–7; Glass Reference Glass1967[1940]: 251–4).
Japan
At first glance, a warfare-welfare nexus in family policy is absent in Japan. Even though the country planned and launched a war of aggression and was heavily struck by acts of war, family allowances were introduced very late. However, this belated program adoption is also causally related to the Second World War, but in a very peculiar way.
As in Germany and Italy, a right-wing nationalist government had pushed pronatalist policies before and during the Pacific War since the military needed a large number of soldiers for the conquest of the Asian mainland. The culmination was a policy adopted in 1941 which “urged strongly that all the efforts be directed toward increasing the birth rate in order to carry out the war and set a target of 100 million population in Japan proper by 1960” (Muramatsu and Kuroda Reference Muramatsu, Kuroda and Berelson1974: 707). A broad range of maternal and child services was proposed to meet this target but this did not include the introduction of family allowances. The government rather focused on the expansion of other welfare programs as Japan was clearly a welfare state laggard in the 1930s. A welfare ministry was founded in 1937 (which soon was directed by a high ranking member of the military), health insurance was expanded significantly in wartime to improve the health status of (prospective) soldiers and an old age pension scheme for workers was introduced in 1942 which, however, mainly served to finance the war (Kasza Reference Kasza2002: 423–5). However, Japan’s military defeat and the related huge territorial losses on the Asian mainland led to a repatriation of 6.25 million civilians and soldiers (Steiner Reference Steiner1953: 245). In addition, Japan witnessed a baby boom in the late 1940s. Given a total fertility rate of 4.63 in 1947 (Gauthier Reference Gauthier1996, 61), the fear of overpopulation on a massively shrunken national territory caused an anti-natalist turnaround immediately after the end of war. A eugenics law passed in 1948 legalized abortions, contraception was liberalized, and the two-child family was declared the model of societal progress (Nennstiel Reference Nennstiel, Chiavacci and Wieczorek2012). In consequence, the birth rate declined quickly and eventually fell below the net reproduction rate in the late 1950s. Fertility then was one of the lowest in the world. Labor shortages related to the economic boom in the 1960s and increasing concerns about an unfavorable demographic structure in the future fuelled discussions on how to increase the birth rate. One proposal was to improve the welfare of families, which eventually paved the way for the introduction of family allowances, effective since 1972. The new scheme represented “a new direction in the government’s post-war population policy: from one of limiting population growth to one of encouraging it” (Kreitler Kirkpatrick Reference Kreitler Kirkpatrick.1972: 43).
Conclusion
Inspired by Titmuss’ (Reference Titmuss and Titmuss1958) classic essay on war and social policy, many recent studies have shown that world wars led both to the adoption of new welfare programs and, in consequence, rising social expenditure (e.g. Obinger and Schmitt Reference Obinger, Petersen and Starke2018, Reference Obinger and Schmitt2020). Family policy is no exception. While family cash benefits were paid to public sector employees for the first time during the Great War, almost everywhere in the Western world, universal family allowances were only introduced around the Second World War. This clustering of program adoption is not coincidental. This paper has argued and empirically shown for 18 Western countries that the Second World War and its socioeconomic aftershocks have been an important causal factor in the introduction of universal family allowances, notably under the auspices of right-wing governments.Footnote 9 According to the findings of the regression analysis, the probability of adopting family allowances during and in the immediate aftermath of World War II is more than six times higher than during peacetime. It was not so much the destructiveness of and the involvement in the war that played a role, but rather a general wartime crisis that affected belligerent and non-belligerent countries in similar ways. Families with many children were particularly vulnerable to the horrors and socio-economic repercussions of war and many of them lived in poverty. Total war and its aftershocks, therefore, reinforced a long-lasting debate about a social wage or what the ILO (1924) described as “remuneration of labor according to need”. In addition, war had several negative demographic effects that exacerbated the general decline in fertility rates. This fuelled concerns about a lack of manpower required for economic reconstruction and prompted pronatalist population policies in the post-war period. In Austria, for example, the introduction of family allowances in 1948 was not only motivated by the hardship of families but was also seen as a vehicle to restore higher birth rates (Obinger Reference Obinger and Obinger2018: 94). In some great powers, militarily motivated pronatalism was already a key impetus for the introduction of family allowances in the run-up to war. As our brief case studies suggest, Germany, Italy, and France introduced pronatalist policies, including family allowances in the 1930s, as these nations were either planning a war of expansion (fascist Italy and Germany) or saw themselves as militarily threatened (France).
If we look beyond the 18 Western countries studied in this paper, we also find a close temporal overlap between the Second World War and the adoption of family allowances. Many war-torn countries in Eastern Europe such as Bulgaria (1942), Czechoslovakia (1945), Romania (1944), and the Soviet Union (1944) introduced family allowances in wartime. In these war-torn countries, a causal nexus between warfare and welfare reform is very likely but was to date not systematically examined. But also Portugal (1942) and, outside Europe, Algeria (1941), Morocco (1942), Tunisia (1944), Brazil (1941), Uruguay (1943), and Vietnam (1944) created such a program during wartime (Perrin Reference Perrin1969: 285–7; Kuhnle and Sander Reference Kuhnle, Sander, Béland, Morgan, Obinger and Pierson2021: 78–81). The extent to which military interests and the ramifications of war influenced the introduction of family allowances in these countries should be the subject of future research.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0143814X24000035
Data availability statement
Replication materials are available in the Journal of Public Policy Dataverse at doi:10.7910/DVN/RHB1M9
Acknowledgements
We thank the participants of the Comparative Political Economy Workshop held at the Max Planck Institute for the Study of Societies in Cologne in November 2022 for their valuable comments. We are particularly grateful to Christine Trampusch, Nikolas Dörr and Klaus Petersen and thank three reviewers for their helpful comments. This research was funded by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Projektnummer 374666841 –SFB 1342.
Open access
