Could Immigration Prevent Population Decline? The Demographic Prospects of Germany Revisited Could Immigration Prevent Population Decline? The Demographic Prospects of Germany Revisited Hannes Weber Abstract: Germany has a record of more than 40 years of below-replacement fertility and annual death surplus. Hence, it is commonly accepted that Germany’s population will decline considerably in the coming decades. Recent increases in immigration may, however, challenge the offi cial long-term demographic projections for Germany. This paper assesses the impact of a permanent higher-than-expected level of net immigratio n to Germany as in the past three years on the projections for population, age structure and ethnic makeup by mid-century. The paper adds a higher immigration variant to the Federal Statistical Offi ce’s latest Coordinated Population Projection and two variants of a (a) constant or (b) de- creasing fertility rate among migrant women. It can be shown that with permanent net migration as high as in recent years (around 300,000 per annum), Germany’s population would not signifi cantly decrease in the coming decades but would rather remain at 80 million until 2050. On the other hand, the sharp rise in the old-age de- pendency ratio is only mildly weakened by increased immigration rates. This issue is therefore probably best addressed by other (or additional) means. The increase in retirees will level off after 2035 in any case. The ethnic makeup of society would be affected to a greater degree than its age composition: The share of fi rst- and second-generation immigrants among the total population is projected to rise to about 35 percent in this scenario (and to above 40 percent if the third generation is also counted). Keywords: Immigration · Germany · Population decline · Ethnic minorities Comparative Population Studies Vol. 40, 2 (2015): 165-190 (Date of release: 17.04.2015) © Federal Institute for Population Research 2015 URL: www.comparativepopulationstudies.de DOI: 10.12765/CPoS-2015-05en URN: urn:nbn:de:bib-cpos-2015-05en6 • Hannes Weber166 1 Introduction Every year since 1972, deaths have outnumbered births in Germany.1 Since then, the total fertility rate (TFR) has remained constant at around 1.4 children per wom- an. While fertility decline is witnessed by all societies in the course of demograph- ic transition (Dyson 2010), the case of Germany is exceptional with respect to the length and consistency of this period of below-replacement fertility and excess of deaths over births. In most other developed countries, the onset of natural popula- tion decline has occurred signifi cantly later, if at all. The fi rst post-war year in which natural population growth was negative in Russia was 1992, in Italy 1993, and in Ja- pan 2007. Some countries, such as Sweden or Denmark, have returned to birth sur- plus after brief episodes of negative natural population growth while others, such as France or the United Kingdom, have not had notable periods of post-war death surplus years. In a European comparison, most northern and western European countries have signifi cantly higher birth rates than Germany, while many central- eastern and southern European countries are characterised by comparatively low fertility rates, but have witnessed the onset of fertility decline much more recently (see e.g. Goldstein et al. 2009). Moreover, some of the latter countries are currently experiencing a recovering fertility rate after a period of “lowest-low” fertility as a result of birth postponements, an effect that appears to have slowed down since the late 1990s (Goldstein et al. 2009; Sobotka et al. 2011; Bongaarts/Sobotka 2012). Thus, in an international comparison, Germany stands out as the country with a his- tory of more than 40 years of negative natural population growth and a total fertility rate that is consistently below 1.5 children per woman. It is beyond the scope of this article to investigate the reasons for this exceptional process and the apparent futility of all pronatalist governmental policies. Rather, the purpose of this study is to investigate whether a permanently high level of net immigration, as in the past three years, may prevent population decline until mid-century despite this quite spectacular record of a continuous birth defi cit. So far, the impact of fertility decline on society and the economy has arguably been rather low. In 2012, both the absolute and relative sizes of the workforce were at record highs in Germany (Statistisches Bundesamt 2013c: 115). This indicates that Germany is still at a point in demographic transition where the economic bene- fi ts arising from the decline in the young dependent population are not yet fully can- celled out by the growing retirement-age population (the so-called “demographic dividend”, see Bloom et al. 2003): The dependency ratio (aged 0-14 and 65 or older relative to the working age population aged 15-64) is currently (0.51) still lower than in 1970 (0.57), before the onset of negative natural population growth. However, this will dramatically change in the coming two decades, when the baby boomer 1 This and the following statement refer to West Germany before and unifi ed Germany after 1990. The fertility rate in East Germany was higher prior to 1990 (Kreyenfeld 2004), but in total, East and West Germany still displayed a consistent death surplus since 1972. Data sources for this and the following statements are Statistisches Bundesamt (2013d) and United Nations (yearly). Could Immigration Prevent Population Decline? • 167 cohorts born in the 1950s and 1960s retire. Between 2025 and 2035, 13.4 million people now aged 45 to 55 will drop out of the labour force (assuming a retirement age of 67 and ignoring future migration and pre-retirement age mortality), while only 6.9 million – about half that number – of currently 0 to 10 year olds will enter the workforce. This obviously poses a great challenge to the pay-as-you-go pension system and Germany’s export-oriented economy. The number of taxpayers will shrink disproportionately compared to a sharp increase in retirees, and companies may struggle to replace their retiring employees. Furthermore, population decline will affect some geographical regions and branches of the economy more than oth- ers. The number of college graduates in the fi elds of social sciences, law and eco- nomics, for instance, has more than doubled between 1995 and 2011 (from 50,400 to 114,200) in Germany, while the number of engineering graduates has risen only mildly during that same period (from 50,600 to 65,100; see Statistisches Bundesamt 2012a: 44-45). It can thus be assumed that a shortage of labour will more likely hit sectors recruiting from the latter group as opposed to the former (which is already the case to some extent, see Bundesagentur für Arbeit 2013).2 It should be noted, however, that population and labour force decline do not necessarily have only negative consequences in the eyes of public opinion (see, e.g., Coleman/Rowthorn 2011). All other things being equal, population decline may lead to, among others, a decrease in housing prices, less traffi c, and less pollution. Moreover, if the demand for labour does not decrease in equal proportions, a de- cline in the labour force may lead to an increase in wages or a falling unemployment rate. While we cannot elaborate in greater detail on the likelihood or the desirability of such outcomes here, it should be emphasized that not every aspect of population decline and aging needs to be perceived as negative by a majority of the population. Offi cial projections estimate the population will drop from currently more than 80 million to between 65 and 70 million by 2060 (medium variant, lower and up- per limit, see Statistisches Bundesamt 2009). The prospects of a declining German population are widely considered to be beyond question and are communicated as such in the media (see, e.g., Spiegel 2008; FAZ 2010; New York Times 2013). In recent years, however, immigration to Germany has risen to a considerably higher level than the offi cial projections had assumed. In 2013, deaths outnumbered births in Germany by around 200,000; yet in the same year, immigrants outnumbered emigrants by 437,000, constituting the largest net immigration rate since the early 1990s and a remarkable increase since the year 2009, which had seen more emi- grants from than immigrants to Germany (see Fig. 1). That year, the Federal Statisti- cal Offi ce (Statistisches Bundesamt 2009) estimated the latest Coordinated Popu- lation Projection. The models included two assumptions with regard to external migration: one variant assumed a net immigration of 100,000 from 2014 on, while 2 A shortage of skilled labour may be even more severe in branches that traditionally recruit from the German dual system of vocational education (as already reported in Bundesagentur für Ar- beit 2013), given the recent sharp increase in tertiary education: The share of college entrants within each cohort rose from 29 percent to 43 percent between 2000 and 2011 (Statistisches Bundesamt 2012a: 15). • Hannes Weber168 the other variant calculated 200,000 from 2020 on. Probably unforeseeable in 2009, the actual annual net immigration to Germany in the years 2011 to 2013 by far sur- passed the rate assumed in the higher immigration variant of the offi cial projection. It is uncertain whether migration infl ows will continue at this level. On the one hand, improving economic conditions in southern Europe may lead to less labour migrants especially from Greece, Spain, and Italy as compared to the recent years since the onset of the fi nancial crisis. Moreover, the countries in central-eastern and southeastern Europe – from where Germany currently receives the most migrants – are themselves characterised by aging populations and their future emigration potential may therefore decrease drastically. On the other hand, unrestricted access to the German labour market for Romanians and Bulgarians since 2014 may at least in the near future give rise to labour migration from these countries, as has been the case with Poland and other EU-10 countries since 2011, when labour market restric- Fig. 1: Net immigration and natural population growth in Germany, 1991-2013 -300,000 -100,000 100,000 300,000 500,000 700,000 900,000 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 Year Natural population growth Net immigration Source: data Federal Statistical Offi ce, own illustration Could Immigration Prevent Population Decline? • 169 tions for these citizens were dropped.3 Furthermore, the repeated refugee ship trag- edies off the Italian coastlines and the civil war in Syria, among other events, have led to debates about less restrictive asylum regulations. Finally, while the migration potential may decrease in central-eastern Europe, the opposite is likely to be true for Africa and the Middle East in the coming decades, given the prospects for popula- tion growth – especially in the age group 20 to 35 – in these regions, which has been shown to infl uence migration rates (Cohen et al. 2008; Mayda 2010). In any case, we should consider the possibility that the offi cial projections underestimate the long- term immigration rates to Germany. Therefore, this paper aims to investigate how ongoing higher net immigration rates would affect the demographic prospects of Germany and its ethnic makeup. 2 Research design and methodology The purpose of this study is to show how the offi cial population forecasts for Ger- many would change were there permanent higher-than-expected future immigra- tion rates of a magnitude as in recent years (2011 to 2013). It therefore follows the Federal Statistical Offi ce’s (Statistisches Bundesamt 2009) 12th Coordinated Popula- tion Projection, adopting all its major assumptions but adding a higher immigration variant and two variants with regard to fertility among migrant women. 2.1 Defi ning ethnic minorities Apart from the consequences for the total population size, these projections also seek to assess the impact of immigration on the ethnic makeup of Germany’s popu- lation. One of the biggest diffi culties of this task is defi ning ethnic groups and the rules for assigning them to individuals. There are different approaches to this topic based on differences in underlying concepts of national or ethnic identity as well as on national laws and customs of data collection. In Germany, at least two his- torical factors have infl uenced the defi nition of racial and ethnic minorities. On the one hand, there was a long tradition of an ius sanguinis nationality law, such that children born to foreigners in Germany were not granted German citizenship by birth unless they had some ethnic German ancestry (see, e.g., Brubaker 1992). On the other hand, after the racially motivated persecution in the Nazi era, German authorities usually refrain from using concepts such as race or ethnic ancestry in offi cial censuses or surveys. As a result, until recently ethnic minorities in post- 3 Yearly infl ows from Poland had been quite constant in the years before the 2004 enlargement, when they increased to 125,000, up from 88,000 in 2003. 2011 saw another signifi cant increase, when 163,000 people entered the country from Poland, as compared to 115,000 in 2010. The most recent value (189,000 in 2013) is twice the fi gure of 2003 (cf. Statistisches Bundesamt 2014b). Needless to say, there is no guarantee that this trend will be lasting, nor that a similar increase will occur with regard to Romanian and Bulgarian migrants from 2014 on, especially since a large number of Romanians had already migrated to Spain and Italy in the previous decade. • Hannes Weber170 war Germany have been almost exclusively analysed by looking at residents with foreign citizenship. This worked well for a long time, given the fact that even most second- or third-generation native-born residents of foreign origin did not hold Ger- man passports. This has changed, however, since the introduction of ius soli-type elements, i.e. German citizenship by birth to children of foreign parents under some conditions, to German nationality law in the year 2000. Among 25-35-year-old Ger- man-born residents with at least one foreign parent, for example, only a minority (41 percent) were German citizens in 2012. However, this fi gure rises to 88 percent among the age group of 0 to 10 years (Statistisches Bundesamt 2013a). The Federal Statistical Offi ce introduced the concept of “migration backgrounds” in 2005, which covers both foreign- as well as native-born residents of Germany with at least one foreign or foreign-born parent (excluding all persons who themselves or whose par- ents came to Germany before 1950, e.g., as expellees from central-eastern Europe in the aftermath of World War II or in previous migration waves, which were appar- ently not assumed to be relevant for analysing the current ethnic diversity in Germa- ny). Data on residents with “migration backgrounds” come from the micro-census, a household survey with a sample size of around 1 percent of the total population. Survey questions on “migration backgrounds” were fi rst included in 2005. Since offi cial statistics on ethnic affi liation based on self-ascription as used in the U.S. census do not exist in Germany, this study works with the concept of “migra- tion backgrounds” and uses this defi nition as a proxy variable for “ethnic minority status.” Obviously, no assertions can be made regarding subjective ethnic identity or cultural assimilation among people of migrant or mixed origin in the future, or the possible changes in future public discourse with regard to those labelled “minority members.” This paper addresses the quantitative question of how many future resi- dents will be fi rst- or second-generation immigrants, depending on future levels of immigration and fertility among migrant women, but we do not intend to speculate here whether this will be of any sociological signifi cance in the future. The study thus looks at: • fi rst-generation immigrants, defi ned as future newcomers to Germany in ad- dition to foreign-born residents already living in the country, • second-generation immigrants, defi ned as native-born residents with at least one foreign or foreign-born parent, and • persons with migration backgrounds, defi ned as being either of the above. The migration background concept as used by the Federal Statistical Offi ce also applies to the third generation in cases where at least one of the parents, despite being native-born, is a foreign (or naturalised) citizen. As the percentage of sec- ond-generation immigrants with (exclusively) foreign nationality is rapidly declining (from 59 percent to 12 percent among children aged 0 to 10 in the past 20 years), the proportion of third-generation immigrants covered by the offi cial defi nition will di- minish as well. For the sake of simplicity, all children born to second-generation im- migrants in the future are therefore treated as natives in this projection. Moreover, the impact of future unions between natives and immigrants is neglected. Mixed unions may lead to a numerically higher proportion of migrant-origin residents. Could Immigration Prevent Population Decline? • 171 2.2 Data and methods The offi cial population projections are based on traditional point forecast models which vary several key parameters and designate the resulting variants as “high,” “medium,” “low,” etc. It has long been argued that a stochastic approach, indicating confi dence intervals and probability estimates, is preferable for long-term popula- tion forecasts (e.g., Lutz et al. 1998; Keilman et al. 2002). However, estimating long- term prediction intervals for future immigration rates – which are the main scope of this article – is more diffi cult than projecting parameters such as the number of births or the old-age dependency ratio, as these are path-dependent factors that in most cases do not change abruptly, while immigration rates are highly sensitive to unpredictable changes in push- and pull factors such as economic growth, political violence and war, or migration-related policies in various countries. Projections for future immigration rates can be based on expert estimates or surveys among potential migrants, or they can use empirical data on past migration fl ows to develop forecasting models based on demographic, geographic, economic (such as wage and employment differences) and other factors (e.g., linguistic simi- larities, past colonial ties, civil war or human rights violations, stock of migrants al- ready living in the country as an indicator of migration networks; see, e.g., Pedersen et al. 2008; Brücker/Siliverstovs 2006; Cohen et al. 2008). These factors themselves must be projected into the future by certain routines or expert judgments that are associated with some degree of uncertainty, whether quantifi able or not. For in- stance, Brücker/Siliverstovs (2006: 49) assume constant unemployment and GDP growth rates for sending and receiving countries in their projection of future migra- tion fl ows to Germany. Future empirical deviations from such assumptions add to the amount of unexplained variance in the various econometric models (39-50 per- cent in the case of the aforementioned study) due to unobserved factors, which inevitably exist despite the vast number of variables and model specifi cations that the authors consider. It is therefore not surprising that the recent substantial increase in immigration to Germany has not been predicted by these models. For instance, Brücker/Siliver- stovs (2006: 50) projected a sharply declining net migration potential from the ten central-eastern EU accession countries (i.e., the new member states of 2004/2007 except Malta and Cyprus) to Germany from 98,000 in 2010 to 29,000 in 2015. The actual fi gure for 2010 was even below the projected number (80,000), but the re- cent increase to 189,000 in net infl ows from these countries in 2013 (Statistisches Bundesamt 2014b) stands in marked contrast to the projections. Obviously, among other factors, the fi nancial and fi scal crisis starting in 2008, which affected Germany less than other EU member states, could not have been taken into account in these projections (but this and other factors may well change in the near future). Studies that quantify the degree of uncertainty with regard to projected migration fl ows in- clude Alders et al. (2007). Their projections of international migration rely on expert judgments, the extrapolation of past trends, and ad-hoc assumptions about which countries are likely to receive below- or above-average levels of net migration (Al- ders et al. 2007: 60). The limits of their 80 percent prediction interval for yearly net • Hannes Weber172 migration to Germany until mid-century are –80,000 and 640,000, respectively. This illustrates why Brunborg and Cappelen (2010: 323) write that “with rapidly chang- ing migration fl ows, the confi dence intervals easily become so large that they are of little or no value for users of the projections.” Given the diffi culties in predicting migration fl ows with a quantifi able degree of uncertainty, we consider it appropriate to make a point forecast based on fi xed parameters of immigration as a “what-if” scenario to complement the existing offi cial scenarios. The starting point of the projection is the actual age pyramid of Germany as of December 31, 2012. Data from the 2011 census, as projected in Statistisches Bun- desamt (2014a), have been combined with estimates on “migration backgrounds” among 5-year age groups from 2012 (Statistisches Bundesamt 2013a). The data on age cohort size for the total population were multiplied by two vectors contain- ing the percentage values for foreign- and native-born residents of migration back- ground, respectively, by age cohort. Missing values due to age cohort grouping were linearly interpolated. Hence, three separate matrices were obtained contain- ing data for residents without “migration backgrounds”, foreign-born residents, and second-generation immigrants, respectively. Newcomers are added to the foreign- born group; their children are members of the second-generation group, while their offspring in turn count as native (see 2.3). It should be noted that the population estimates based on the census of 2011 are signifi cantly (around 1.5 million) lower than the projections based on the census of 1987, which prevailed in offi cial publica- tions and projections until mid-2013. As described above, the model differentiates between residents without “migration backgrounds,” immigrants (i.e. foreign-born residents) and second-generation immigrants (i.e. native-born residents with at least one foreign-born parent). Distinguishing between the two sexes within each of these groups resulted in six separate matrices. For reasons of space, the projec- tion algorithms are given for females only in the following. The size of a future age cohort from the fi rst group is estimated as Na,t = Na,t-1 * Sa,t, for a > 0, where Na,t is the number of native females without “migration backgrounds” of age a in year t, Sa,t is the age-specifi c survival rate for females (from Statistisches Bundesamt 2013e), varying by year as it has been linearly extrapolated to account for a rising life expectancy (see 2.2). The number of female newborns without “mi- gration backgrounds” in a given year is estimated as where Fa is the age-specifi c fertility rate and Ma,t is the number of second-gener- ation migrant women in a respective age group in year t, as their children (i.e. the third generation) are classifi ed as natives. This group, in turn, is equallyforward- projected by Ma,t = Ma,t-1 * Sa,t, for a > 0, (1) (2)N0,t = , (3) Could Immigration Prevent Population Decline? • 173 while is the number of births to foreign-born women Ia,t who count as second-generation migrants, given the differing age-specifi c fertility rate Ga for immigrant women. Fi- nally, the foreign-born group grows by a yearly constant number of newcomers Ca,v, dependent on variant v and distributed across age groups: Ia,t = Ia,t-1 * Sa,t + Ca,v, for a > 0, I0,t = C0,v. 2.2 Underlying assumptions of the model The parameters of the model have been mainly adjusted according to the medium or basic variants of the offi cial projections: The total fertility rate is assumed to remain constant at the present level of roughly 1.4 children per woman, following the Federal Statistical Offi ce’s medium variant. Certainly, a TFR of 1.4 is not considered an inevitable fate and no policy rec- ommendation is being made here to conceive it as such. Yet, existing (point) fore- casts expect Germany’s TFR to remain at 1.4 until mid-century (Alders et al. 2007) or increase only marginally (to 1.47 by 2050 according to Eurostat, see Scherbov et al. 2008).4 Given the fact that the TFR “seems to have stalled at a level of 1.4 children since the 1970s” (Goldstein/Kreyenfeld 2011: 454), it appears reasonable to fi x this parameter for the present analysis. Age group-specifi c fertility rates are taken from the Human Fertility Database (2013) and are assumed to be distributed as in 2010. Further fertility postponement may occur as a result of, for example, the continu- ing growth of participation in higher education (Ní Brolchaín/Beaujouan 2012). On the other hand, the pace of fertility postponement is declining and tempo effects are projected to disappear in the near future (Bongaarts/Sobotka 2012). The offi cial projections (Statistisches Bundesamt 2009) assume that the number of births to women aged 30 and over will rise, while the age-specifi c fertility rate for younger women is either assumed to decrease proportionately or over-proportionately or remain constant (which would result in a higher period TFR). As there is currently no coherent theory for the prediction of the future development of fertility, with (sociological, economic, or biological) arguments available in favor of both a rise as well as further decline (for an overview see Lutz 2006), for the sake of simplicity both period and cohort TFR are held constant in the present study. M0,t = (4) (5) (6) 4 Exceptions include the United Nations (2013: 75) who assume a rise in Germany’s TFR to 1.64 by 2050, similar to the Federal Statistical Offi ce’s “higher variant.” • Hannes Weber174 Fertility patterns among native and migrant women are assumed to either re- main constant (variant A), or fertility among foreign-born women will decline to 1.4 children per woman (variant B). According to Statistisches Bundesamt (2012b: 39), the total number of births to foreign-born women until the age of 39 was 1.9 children on average. As the difference between native and foreign-born women has remained quite constant over the past decades (Statistisches Bundesamt 2012b), Variant A of the model assumes that this trend will continue. However, there are some reasons to expect immigrant women’s fertility rates to decline (see Schmid/ Kohls 2009 for an overview of the determinants of migrants’ reproductive behavior in Germany). Apart from a general fertility decline in most countries of origin and an expectable approximation of immigrant women to host country family customs, the growing share of central-eastern and south-eastern European immigrants, whose fertility rates are comparable to those among Germans, might lead to a convergence of birth rates among natives and immigrants (Statistisches Bundesamt 2012b: 23). For example, the completed fertility rate among migrant women aged 40 or older was 3.47 children for Turkish women compared to 1.53 children for Polish women (Schmid/Kohls 2009: 49). While Turks make up the largest group of foreign residents in Germany, Poland has been the number one country of origin for new arrivers in the past years. Fertility among migrants is thus likely to decline through this com- position effect, in addition to other effects (e.g. socialisation). Therefore, the second variant assumes a fertility rate among the foreign-born of 1.4 children per woman, as for the current total population, with age-specifi c fertility rates equaling those of native women. It should be noted that the TFR for native women without “migrant backgrounds” might be below the threshold of 1.4, but no reliable data are avail- able to distinguish native women with and without migration backgrounds. Second- generation immigrant women (whose children count as natives without “migration backgrounds”) are assumed not to differ from native women with regard to period and cohort TFR due to a lack of reliable data on fertility among native-born women with “migration backgrounds.” Life expectancy is assumed to rise to about 87 years (85 for males and 89 for females) by 2050 according to the Federal Statistical Offi ce’s base assumption (Al- ders et al. 2007 as well as Bijak et al. 2007, among others, project similar values). In the past, reality has disproved projections of a decreasing slope in the rise of life expectancy and the trend has instead been rather linear (Oeppen/Vaupel 2002). For the lack of a more plausible alternative assumption, the model specifi es a continu- ously linear trend. The probability of surviving from birth to age 85 is thus assumed to be 50 percent in 2050 (up from 30.1 percent in 2011) for males and rises linearly until then. The same is true for females until the age of 89. Age-specifi c survival rates have therefore been multiplied by an interpolated factor such that survival rates linearly increase within each age group until reaching the target value for 2050. Obviously, a rise in life expectancy will affect the age-specifi c survival rates of older age groups more than those of younger cohorts, which are already close to 1. Therefore, the infl ating factor has been weighted by an exponential weight variable to affect older age groups more than younger ones. As a result, for instance, the probability of surviving the next year for a 30-year-old male rises from 99.94 per- Could Immigration Prevent Population Decline? • 175 cent to 99.96 percent, whereas the same probability for an 85-year-old male rises from 89.1 percent to 92.1 percent. Four different variants are calculated for external migration rates. Two of them correspond to the variants included in the Federal Statistical Offi ce’s offi cial projec- tion, assuming annual net migration infl ows of 100,000 and 200,000, respectively. A high-migration variant assumes an annual net immigration of 300,000, based on recent trends in migration to Germany, whereas a zero variant assumes balanced external migration and is included for comparison. The 10th Coordinated Popula- tion Projection, published in 2003 (Statistisches Bundesamt 2003), still included a variant assuming 300,000 immigrants per year, in light of an average annual net migration of 335,000 in the period of 1991 to 2000. The follow-up projections have dropped this variant given low immigration rates in the 2000s. Recent increases in net migration (see fi g. 1) suggest reconsidering the 300,000 variant. Changes in co- hort size and age structure as a result of net immigration are supposed to be limited to foreign nationals, while external migration of native Germans is assumed to be balanced and without impact on the age structure. Note, however, that net migra- tion of German nationals has been negative since 2005, but the data do not show a coherent trend. Migrants are assumed to continuously have the same age and sex distribution as the new arrivers in 2012 did (as reported in Statistisches Bundesamt 2014c). Most migrants are males aged 20 to 24 (16 percent of all male immigrants), the modal value being 22 years. In total, 60.3 percent of all newcomers in 2012 were reported to be male. 3 Results 3.1 Total population size Figure 2 shows the results of the projections, with the upper graph displaying the constant fertility variants (1A to 4A) while in the lower graph migrant women’s birth rates are assumed to decline to the present TFR among the total population in the future (variants 1B to 4B). As the graphs show, the model reproduces the Federal Statistical Offi ce’s forecasts for the annual net immigration assumptions of 100.000 and 200.000, respectively (variants 2 and 3). Similar to the offi cial projections, a de- Tab. 1: Overview of variants included in the projection model TFR among migrant women remains at 1.9 children declines to 1.4 children External migration zero net migration Variant 1A 1B 100,000 per year 2A 2B 200,000 per year 3A 3B 300,000 per year 4A 4B Source: own design • Hannes Weber176 Fig. 2: Projections for Germany’s population (in millions) by mid-century 40 45 50 55 60 65 70 75 80 85 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net migration 200,000 annual net migration 100,000 annual net migration Balanced external migration Year Projections for total population (Variants A: Constant Fertility) 40 45 50 55 60 65 70 75 80 85 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net migration 200,000 annual net migration 100,000 annual net migration Balanced external migration Year Projections for total population (Variants B: Declining fertility among migrant women) Millions Millions Source: own calculations Could Immigration Prevent Population Decline? • 177 clining population to between 67 and 72 million in 2050 is forecasted for these two scenarios. If Germany were to experience permanently higher immigration rates of around 300,000 per year, however, according to the projections the population would not signifi cantly decline by mid-century and would continue to be around 80 million until 2050. After 2050, population decline is probably inevitable if fertility rates do not rise. Until then, Germany may achieve stationarity through immigration (cf. Schmertmann 1992), despite a then 80-year pattern of birth defi cit and death surplus. The lower graph in Figure 2 shows that the mid-century population can be expected to be slightly lower if fertility among native and migrant women con- verges as opposed to the constant fertility variants. In the scenario where fertility among migrant women declines and external emigration is balanced, Germany’s population drops by more than 17 million to around 63 million by 2050. 3.2 Old-age dependency ratio With or without immigration, the percentage of retirees among the total population will greatly increase in the coming decades. As Figure 3 shows, the old-age de- pendency ratio (the number of people aged 65 or older relative to the working age population of 15-64 years) is projected to rise from currently 0.31 to between 0.54 (Variant 4A) and 0.66 (Variant 1B) by 2050. We can thus conclude that while high im- migration rates could prevent population decline until around 2050, they would only mitigate the challenge of population ageing to a relatively small extent, even though new immigrants are signifi cantly younger than the resident population. This fi nd- ing is consistent with the United Nations Population Division’s (2000) report about “replacement migration,” which in the year 2000 estimated that Germany needed 344,000 immigrants per year to keep the population at a constant level until 2050, while an absurdly high annual infl ow of 3.6 million (or 181 million in total until 2050) would be needed to maintain the old-age dependency ratio at the 2000 level (for the development and public reception of this study, see Teitelbaum 2004; similar results have been obtained by Bijak et al. 2008). An annual net immigration rate of 300,000 instead of 200,000 would result in six million additional people living in Germany, yet would lower the old-age dependency ratio by only 3-4 percent (see Appendix for details). From around 2035 on, the slope will be less steep in any case, as the cohorts reaching retirement age from then on are already post-baby boom. Starting in the late 2030s, all retirees will already have been born in the low-fertility period, so that the gap between the number of retirees and the number of young people en- tering the workforce each year will be much smaller than in the period from 2025 to 2035. The most challenging period of the demographic transition therefore certainly lies ahead for Germany, specifi cally in the next two decades. • Hannes Weber178 Fig. 3: Projected number of people aged 65 or older per resident in age group 15 to 64 0 10 20 30 40 50 60 70 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net immigration 200,000 annual net immigration 100,000 annual net immigration Balanced external migration Year Projections for old-age dependency ratio (Variants A: Constant Fertility) 0 10 20 30 40 50 60 70 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net immigration 200,000 annual net immigration 100,000 annual net immigration Balanced external migration Year Projections for old-age dependency ratio (Variants B: Declining fertility among migrant women) per 100 persons per 100 persons Source: own calculations Could Immigration Prevent Population Decline? • 179 3.3 Share of ethnic minorities Figure 4 displays the projected impact of different scenarios of immigration and fertility rates on the ethnic makeup of Germany. While immigration can only mod- erately affect Germany’s future age structure, the infl uence on the percentage of ethnic minorities is signifi cantly greater. In the high immigration and constant fertil- ity scenario, fi rst- and second-generation migrants will make up 35 percent of the total population in 2050. With zero immigration and migrants’ fertility as low as natives’, the share of residents with “migration backgrounds” would stagnate at around 20 percent until mid-century. With future immigration rates as assumed in the offi cial projections (100,000 or 200,000), the share of fi rst- and second-genera- tion migrants is projected to rise to 24 percent in the lowest (2B) and 29 percent in the highest variant (3A).5 Whether birth rates among migrant women remain at 1.9 or decline to 1.4 children per woman has only a marginal impact on the projection results with regard to this indicator. If the third generation of migrants is included in the defi nition of “migration background” as well, then the share of the latter group rises above 40 percent in the high immigration scenario. We must emphasize again that it is diffi cult to defi ne who belongs to an ethnic minority group and that patterns of subjective ethnic identifi cation may develop dif- ferently. In case of ongoing net migration of a magnitude as in recent years, howev- er, we can project that the share of the population with a foreign background will rise signifi cantly in the future, even if the third and later generations of immigrants or mi- norities are defi ned as natives. Moreover, the distribution of immigrants and ethnic minorities will probably continue to be uneven across the country. The percentage of residents with “migration backgrounds” currently ranges from only around 1 per- cent in some east German districts to more than 50 percent in the city of Offenbach near Frankfurt am Main. In many west and south German urban agglomerations such as Frankfurt am Main, Munich or Stuttgart, the majority of children today are already being born into migrant families.6 The possible economic (e.g., Rowthorn 2008; Glitz 2012) or social (e.g., Schlueter/Wagner 2008; Eger 2012) implications of this trend are beyond the scope of this study. Yet from a demographic point of view, we can project that variance in future immigration rates will have a greater impact on the ethnic composition of society (as visualised by the diverging curves in Fig. 4) than on its age structure (and thus the dependency ratio). 5 Lanzieri (2011) has somewhat higher results (32-39 percent residents of foreign origin in Ger- many by 2051) with a number of yearly net infl ows fi xed at around 155,000 on average per year. The differences can probably be attributed to the divergent defi nitions of “foreign origin.” In the present study, around 4-6 percent of the projected population in 2050 is made up of third- generation immigrants who are not included in the defi nition of “migration background” in Fig- ure 4. 6 For the share of foreign nationals on local (NUTS-3) level, see Statistische Ämter des Bundes und der Länder (2013b). For the share of people and children with “migration backgrounds” in selected cities and regions, see Statistische Ämter des Bundes und der Länder (2013a). The latter source is based on the micro-census survey, but several cities also use special software such as “MigraPro” to determine the number of residents with “migration backgrounds” from their register data. • Hannes Weber180 Fig. 4: Projections for the share of people with “migration backgrounds” 0 5 10 15 20 25 30 35 40 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net immigration 200,000 annual net immigration 100,000 annual net immigration Balanced external migration Year Projections for share of ethnic minorities (Variants A: Constant Fertility) 0 5 10 15 20 25 30 35 40 2012 2017 2022 2027 2032 2037 2042 2047 300,000 annual net immigration 200,000 annual net immigration 100,000 annual net immigration Balanced external migration Year Projections for share of ethnic minorities (Variants B: Declining fertility among migrant women) Percent Percent Source: own calculations Could Immigration Prevent Population Decline? • 181 4 Conclusions Contrary to the widely held assumption, Germany’s population may not decline sig- nifi cantly by mid-century. Ceteris paribus, an annual net immigration of 300,000 per year could prevent population decline until 2050 and cause the number of inhabit- ants in Germany to stay roughly around 80 million. This would be the case if annual infl ows would be as high as in 2011-2013 (and during the 1990s). The share of fi rst- and second-generation immigrants would rise to around 35 percent by mid-century (or above 40 percent including the third generation). The economic and social con- sequences of such an increase are disputed and cannot be assessed here, nor can we discuss the question whether it is ethically justifi able to draw on working-age immigrants to mitigate one’s own demographic problems, thereby potentially caus- ing diffi culties in the countries of origin where these people may be needed as well. It is furthermore impossible to estimate the numerical likelihood of this higher net-infl ows scenario; but given the recent increase in net immigration to more than 400.000 in 2013, the possibility should be considered. With less immigration, Ger- many’s population would indeed decline. In any case, increased immigration rates could only slightly ease the problem of the sharp increase in the old-age dependent population in the coming decades. This increase is an obvious result of the baby boomer cohorts retiring between 2025 and 2035. After this period, the gap between the annual number of retirees and the num- ber of young people entering the workforce will abruptly decrease due to the fact that the retirees will themselves have been born in the era of below-replacement fertility. Increasing the annual number of immigrants from 200.000 (as in the Federal Statistical Offi ce’s high assumption) to 300.000 would result in a mere 3-4 percent decrease in the old-age dependency ratio by 2050, despite adding six million people to the total population. If a stationary total population by mid-century is considered desirable, immigra- tion rates as high as in the 2011-2013 period are probably the only practicable means of achieving this goal. However, the main challenge arising from the demographic transition is arguably the rise in pensioners alongside the decline in the working- age population rather than a reduction in the total population (which may, after all, involve some desirable effects as well in the eyes of the public). This challenge is probably best met by means other than (or in addition to) immigration, as the latter only affects the old-age dependency ratio to a limited degree. Such means may in- clude increasing the labour market participation of older workers, as only about one in two men aged 55 to 64 in Germany is active in the labour force – a sharp decline since the 1970s and a considerably lower value than in other developed countries (Cooke 2006). Policy changes to reverse this trend have already been implemented, but with disputable effects so far (Buchholz et al. 2013). In total, the Federal Statisti- cal Offi ce (Statistisches Bundesamt 2013c: 117) estimates an unused potential of almost seven million unemployed, underemployed and other non-active members of the labour force seeking work. This number is larger than the gap in the labour force that will occur when the baby boomer cohorts retire between 2025 and 2035. And since many of them are women, better opportunities to combine career and • Hannes Weber182 children are an obvious measure that policymakers may want to pursue. Germany is still enjoying the “demographic dividend,” the concurrence of few children and a so far moderate number of old-age dependent persons. This will naturally change in the coming decades, but the major challenge will diminish after 2035. Until then, there appear to be suffi cient options to address the issues associated with a shrinking and aging population, including, but not limited to, a higher future net immigration. References Alders, Marteen; Keilman, Nico; Cruijsen, Harri 2007: Assumptions for long-term sto- chastic population forecasts in 18 European countries. In: European Journal of Popu- lation 23: 33-69 [doi: 10.1007/s10680-006-9104-4]. Bijak, Jakub; Kupiszewska, Dorota; Kupiszewski, Marek 2008: Replacement Migration Revisited: Simulations of the Effects of Selected Population and Labor Market Strate- gies for the Aging Europe, 2002–2052. In: Population Research and Policy Review 27: 321-342 [doi: 10.1007/s11113-007-9065-2]. Bijak, Jakub et al. 2007: Population and labour force projections for 27 European coun- tries, 2002–2052: impact of international migration on population ageing. In: European Journal of Population 23: 1-31 [doi: 10.1007/s10680-006-9110-6]. Bloom, David E.; Canning, David; Sevilla, Jaypee 2003: The Demographic Dividend. A New Perspective on the Economic Consequences of Population Change. Santa Mon- ica et al.: RAND. Bongaarts, John; Sobotka, Tomáš 2012: A Demographic Explanation for the Recent Rise in European Fertility. In: Population and Development Review 38,1: 83-120 [doi: 10.1111/j.1728-4457.2012.00473.x]. Brubaker, Rogers 1992: Citizenship and Nationhood in France and Germany. Cambridge: Harvard University Press. Brücker, Herbert; Siliverstovs, Boris 2006: Estimating and forecasting European migra- tion: methods, problems and results. In: Zeitschrift für Arbeitsmarktforschung – Jour- nal of Labour Market Reseach 29,1: 35-56. Brunborg,Helge; Cappelen, Adne 2010: Forecasting migration fl ows to and from Nor- way using an econometric model. In: Eurostat work session on demographic projec- tions. Lisbon 28-30 April 2010: 321-346. Buchholz, Sandra; Rinklake, Annika; Blossfeld, Hans-Peter 2013: Reversing Early Retire- ment in Germany. A Longitudinal Analysis of the Effects of Recent Pension Reforms on the Timing of the Transition to Retirement and on Pension Incomes. In: Compara- tive Population Studies – Zeitschrift für Bevölkerungswissenschaft 38,4: 881-906 [doi: 10.12765/CPoS-2013-23en]. Bundesagentur für Arbeit 2013: Der Arbeitsmarkt in Deutschland – Fachkräfteengpass- analyse Juni 2013. Nürnberg: Bundesagentur für Arbeit. Cohen, Joel E.; Roig, Marta; Reuman, Daniel C.; GoGwilt, Cai 2008: International mi- gration beyond gravity: A statistical model for use in population projections. In: Pro- ceedings of the National Academy of Sciences of the USA 105,40: 15269-15274 [doi: 10.1073/pnas.0808185105]. Coleman, David; Rowthorn, Robert 2011: Who’s Afraid of Population Decline? A Critical Examination of Its Consequences. In: Population and Development Review 37 (Sup- plement): 217-248 [doi: 10.1111/j.1728-4457.2011.00385.x]. Could Immigration Prevent Population Decline? • 183 Cooke, Martin 2006: Policy Changes and the Labour Force Participation of Older Work- ers: Evidence from Six Countries. In: Canadian Journal of Aging 25,4: 387-400 [doi: 10.1353/cja.2007.0015]. Dyson, Tim 2010: Population and Development: The Demographic Transition. London/ New York: Zed Books. Eger, Maureen A. 2010: Even in Sweden: The Effect of Immigration on Support for Wel- fare State Spending. In: European Sociological Review 26,2: 203-217 [doi: 10.1093/ esr/jcp017]. FAZ 2010: Demographie: Deutschland wird trotz Familienpolitik schrumpfen. In: Frankfurter Allgemeine Zeitung 11.10.2010 [http://www.faz.net/aktuell/wirtschaft/ wirtschaf tspolitik /demographie-deutschland-wird-trotz-familienpolitik-schrump- fen-11056047.html, 27.12.2013]. Goldstein, Joshua R.; Kreyenfeld, Michaela 2011: Has East Germany Overtaken West Germany? Recent Trend in Order-Specifi c Fertility. In: Population and Development Review 37,3: 453-472 [doi: 10.1111/j.1728-4457.2011.00430.x]. Goldstein, Joshua R.; Sobotka, Tomáš; Jasilioniene, Aiva 2009: The End of “Lowest-Low” Fertility? In: Population and Development Review 35,4: 663-699 [doi: 10.1111/j.1728- 4457.2009.00304.x]. Glitz, Albrecht 2012: The Labor Market Impact of Immigration: A Quasi-Experiment Ex- ploiting Immigrant Location Rules in Germany. In: Journal of Labor Economics 30,1: 175-213 [doi: 10.1086/662143]. Human Fertility Database 2013: Max Planck Institute for Demographic Research (Ger- many) and Vienna Institute of Demography (Austria) [www.humanfertility.org, 27.12.2013]. Keilman, Nico; Quang Pham, Dinh; Hetland, Arve 2002: Why population forecasts should be probabilistic – illustrated by the case of Norway. In: Demographic Research 6,15: 409-454 [doi: 10.4054/DemRes.2002.6.15]. Kreyenfeld, Michaela 2004: Fertility Decisions in the FRG and GDR: An Analysis with Data from the German Fertility and Family Survey. In: Demographic Research, Special Collection 3,11: 275-318 [doi: 10.4054/DemRes.2004.S3.11]. Lanzieri, Giovanni 2011: Fewer, older and multicultural? Projections of the EU popula- tions by foreign/national background. In: Eurostat Methodologies and Working Pa- pers. Luxembourg: Publications Offi ce of the European Union. Lutz, Wolfgang 2006: Fertility rates and future population trends: will Europe’s birth rate recover or continue to decline? In: International Journal of Andrology 29: 25-33 [doi: 10.1111/j.1365-2605.2005.00639.x]. Lutz, Wolfgang; Sanderson, Warren C.; Scherbov, Sergei 1998: Expert-Based Proba- bilistic Population Projections. In: Population and Development Review 24: 139-155. Mayda, Anna Maria 2010: International migration: a panel data analysis of the deter- minants of bilateral fl ows. In: Journal of Population Economics 23: 1249-1274 [doi: 10.1007/s00148-009-0251-x]. Ní Bhrolcháin, Máire; Beaujouan, Eva 2012: Fertility postponement is largely due to ris- ing educational enrolment. In: Population Studies: A Journal of Demography 66,3: 311-327 [doi: 10.1080/00324728.2012.697569]. NY Times 2013: Germany fi ghts population drop. In: New York Times 13.08.2013 [http:// www.nytimes.com/2013/08/14/world/europe/germany-fi ghts-population-drop.html, 27.12.2013]. • Hannes Weber184 Oeppen, Jim; Vaupel, James W. 2002: Broken Limits to Life Expectancy. In: Science 295,5570: 1029-1031 [doi: 10.1126/science.1069675]. Pedersen, Peder J.; Pytlikova, Mariola; Smith, Nina 2008: Selection and network effects – Migration fl ows into OECD Countries 1990-2000. In: European Economic Review 52,7: 1160-1186 [doi: 10.1016/j.euroecorev.2007.12.002]. Rowthorn, Robert 2008: The fi scal impact of immigration on the advanced economies. In: Oxford Review of Economic Policy 24,3: 560-580 [doi: 10.1093/oxrep/grn025]. Scherbov, Sergei; Mamolo, Marija; Lutz, Wolfgang 2008: Probabilistic Population Pro- jections for the 27 EU Member States Based on Eurostat Assumptions. In: European Demographic Research Papers 2. Vienna, Austria: Vienna Institute of Demography. Schlueter, Elmar; Wagner, Ulrich 2008: Regional Differences Matter: Examining the Dual Infl uence of the Regional Size of the Immigrant Population on Derogation of Im- migrants in Europe. In: International Journal of Comparative Sociology 49,2-3: 153- 173 [doi: 10.1177/0020815207088910]. Schmertmann, Carl P. 1992: Immigrants’ Ages and the Structure of Stationary Popula- tions with Below-Replacement Fertility. In: Demography 29,4: 595-612. Schmid, Susanne; Kohls, Martin 2009: Reproductive behaviour of migrant women in Germany: Data, patterns and determinants. In: Vienna Yearbook of population re- search: 39-61 [doi: 10.2307/23025523]. Sobotka, Tomáš et al. 2011: Postponement and Recuperation in Cohort Fertility: Austria, Germany and Switzerland in a European Context. In: Comparative Population Stud- ies – Zeitschrift für Bevölkerungswissenschaft 36,2-3: 417-452 [doi: 10.4232/10.CPoS- 2011-16de]. SPIEGEL 2008: Letter from Berlin: Defusing Germany’s Demographic Timebomb. In: Spiegel Online 30.04.2008 [http://www.spiegel.de/international/germany/letter-from- berlin-defusing-germany-s-demographic-timebomb-a-550506.html, 27.12.2013]. Statistische Ämter des Bundes und der Länder 2013a: Bevölkerung nach Migrationssta- tus regional. Ergebnisse des Mikrozensus 2011. Wiesbaden: Hessisches Statistisches Landesamt. Statistische Ämter des Bundes und der Länder 2013b: Bevölkerungsstand: Bevölkerung nach Geschlecht, Nationalität und Altersgruppen – Stichtag 31.12. – regionale Tiefe: Kreise und krfr. Städte [https://www.regionalstatistik.de/genesis/online/logon, 27.12.2013]. Statistisches Bundesamt 2003: Bevölkerung Deutschlands bis 2050. 10. koordinierte Bevölkerungsvorausberechnung. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt 2009: Germany’s Population by 2060. Results of the 12th Coor- dinated Population Projection. Wiesbaden: Federal Statistical Offi ce. Statistisches Bundesamt 2012a: Bildung und Kultur. Nichtmonetäre hochschulstatis- tische Kennzahlen 1980-2011. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt 2012b: Geburten in Deutschland. Ausgabe 2012. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt 2013a: Bevölkerung und Erwerbstätigkeit. Bevölkerung mit Migrationshintergrund. Ergebnisse des Mikrozensus 2012. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt 2013b: Bevölkerung und Erwerbstätigkeit. Vorläufi ge Ergeb- nisse der Bevölkerungsfortschreibung auf Grundlage des Zensus 2011. Wiesbaden: Statistisches Bundesamt. Could Immigration Prevent Population Decline? • 185 Statistisches Bundesamt 2013c: Datenreport 2013. Ein Sozialbericht für die Bun- desrepublik Deutschland. Berlin: Bundeszentrale für politische Bildung. Statistisches Bundesamt 2013d: Geborene, Gestorbene, Geburten-/Sterbeüberschuss (ab 1951) [ht tps://w w w.destatis.de/DE/ ZahlenFak ten/Indikatoren/LangeReihen/ Bevoelkerung/lrbev04.html, 27.12.2013]. Statistisches Bundesamt 2013e: Sterbetafel Deutschland 2009/11 [https://www.desta- tis.de/DE/ZahlenFakten/GesellschaftStaat/Bevoelkerung/Sterbefaelle/Tabellen/Ster- betafelDeutschland.xls?__blob=publicationFile, 27.12.2013]. Statistisches Bundesamt 2014a: Bevölkerung: Deutschland, Stichtag, Altersjahre, Na- tionalität/Geschlecht/Familienstand [https://www-genesis.destatis.de/genesis/online/ link/tabellen/12411*, 28.05.2014]. Statistisches Bundesamt 2014b: Bevölkerung und Erwerbstätigkeit. Vorläufi ge Wan- derungsergebnisse [ht t ps:// w w w.des t atis.de/ DE / Publikationen / Thematisch / Bevoelkerung / Wanderungen / vorlaeu f igeWanderungen51271011370 0 4.pd f ?_ _ blob=publicationFile, 22.05.2014]. Statistisches Bundesamt 2014c: Bevölkerung und Erwerbstätigkeit. Wanderungen 2012 [https://www.destatis.de/DE/Publikationen/ Thematisch/Bevoelkerung/ Wanderun- gen/Wanderungen2010120127004.pdf?__blob=publicationFile, 28.05.2014]. Teitelbaum, Michael T. 2004: The Media Marketplace for Garbled Demography. In: Popu- lation and Development Review 30,2: 317-327 [doi: 10.1111/j.1728-4457.2004.015_1.x]. United Nations yearly: Demographic Yearbook. New York: United Nations. United Nations 2013: World Population Projections. The 2012 Revision. Volume I: Com- prehensive Tables. New York: United Nations. United Nations Population Division 2000: Replacement Migration. Is It a Solution to De- clining and Ageing Populations? [http://www.un.org/esa/population/publications/mi- gration/migration.htm, 02.01.2014]. Date of submission: 13.01.2014 Date of acceptance: 12.09.2014 Hannes Weber (). Universität Stuttgart. Stuttgart, Germany. E-Mail: hweber@startmail.com URL: http://www.uni-stuttgart.de/soz/avps/mitarbeiter/index.html • Hannes Weber186 T a b . A 1: P ro je c ti o n s fo r to ta l p o p u la ti o n s iz e o f G e rm an y ( V a ri an ts 1 A t o 4 B ) T a b . A 2 : P ro je c ti o n s fo r th e s h a re o f re si d e n ts w it h “ m ig ra ti o n b a ck g ro u n d ” in % 2 01 2 2 01 5 2 0 2 0 2 0 2 5 2 0 3 0 2 0 3 5 2 0 4 0 2 0 4 5 2 0 5 0 1A 2 0 .0 5 2 0 .5 1 21 .1 2 21 .5 0 21 .6 6 21 .6 1 21 .4 2 21 .1 0 2 0 .6 6 2 A 2 0 .0 5 2 0 .7 7 21 .8 4 2 2 .7 3 2 3 .4 4 2 3 .9 8 2 4 .4 3 2 4 .8 0 2 5 .1 0 3 A 2 0 .0 5 21 .0 3 2 2 .5 5 2 3 .9 2 2 5 .1 4 2 6 .2 2 2 7. 2 2 2 8 .1 6 2 9 .0 7 4 A 2 0 .0 5 21 .4 2 2 3 .6 2 2 5 .7 3 2 7. 7 3 2 9 .6 1 31 .4 1 3 3 .1 7 3 4 .8 8 1B 2 0 .0 5 2 0 .3 8 2 0 .7 9 21 .0 4 21 .1 0 2 0 .9 9 2 0 .7 6 2 0 .4 2 19 .9 6 2 B 2 0 .0 5 2 0 .6 4 21 .5 1 2 2 .2 5 2 2 .8 5 2 3 .3 1 2 3 .6 9 2 4 .0 1 2 4 .2 8 3 B 2 0 .0 5 2 0 .9 0 2 2 .2 1 2 3 .4 2 2 4 .5 2 2 5 .5 0 2 6 .4 1 2 7. 3 0 2 7. 8 8 4 B 2 0 .0 5 21 .2 8 2 3 .2 6 2 5 .1 8 2 7. 0 2 2 8 .7 5 3 0 .4 3 3 2 .0 8 3 3 .7 1 A p p e n d ix 2 01 2 2 01 5 2 0 2 0 2 0 2 5 2 0 3 0 2 0 3 5 2 0 4 0 2 0 4 5 2 0 5 0 1A 8 0 ,5 2 3 ,7 4 6 7 9 ,9 4 3 ,6 0 6 7 8 ,7 12 ,7 5 4 7 6 ,9 5 6 ,4 4 2 74 ,8 2 0 ,4 31 7 2 ,6 0 4 ,3 6 7 7 0 ,0 9 3 ,6 12 6 7, 31 3 ,7 4 7 6 4 ,2 7 5 ,9 10 2 A 8 0 ,5 2 3 ,7 4 6 8 0 ,2 0 6 ,2 3 8 7 9 ,4 3 9 ,6 8 7 7 8 ,1 7 9 ,4 6 8 7 6 ,5 6 0 ,7 8 6 74 ,8 7 0 ,5 6 2 7 2 ,8 8 8 ,2 91 7 0 ,6 3 9 ,5 14 6 8 ,1 3 6 ,3 4 4 3 A 8 0 ,5 2 3 ,7 4 6 8 0 ,4 6 8 ,8 71 8 0 ,1 6 6 ,6 2 0 7 9 ,4 0 2 ,4 9 3 7 8 ,3 01 ,1 41 7 7, 13 6 ,7 5 6 7 5 ,6 8 2 ,9 7 0 7 3 ,9 6 5 ,2 8 0 71 ,9 9 6 ,7 7 9 4 A 8 0 ,5 2 3 ,7 4 6 8 0 ,8 61 ,3 81 81 ,2 8 5 ,4 4 3 81 ,3 3 5 ,2 0 7 81 ,1 0 9 ,8 4 0 8 0 ,8 5 2 ,8 9 7 8 0 ,3 2 5 ,5 8 5 7 9 ,5 5 6 ,8 0 7 7 8 ,5 6 6 ,3 6 4 1B 8 0 ,5 2 3 ,7 4 6 7 9 ,8 0 6 ,4 9 4 7 8 ,3 8 7, 19 6 7 6 ,5 0 3 ,3 74 74 ,2 9 2 ,6 41 7 2 ,0 2 9 ,3 2 5 6 9 ,4 6 9 ,5 6 3 6 6 ,6 16 ,6 12 6 3 ,4 8 3 ,3 4 0 2 B 8 0 ,5 2 3 ,7 4 6 8 0 ,0 6 7, 9 8 9 7 9 ,1 0 3 ,0 5 4 7 7, 6 9 5 ,5 0 8 7 5 ,9 74 ,9 3 9 74 ,2 0 5 ,9 8 2 7 2 ,1 3 9 ,9 5 4 6 9 ,7 7 9 ,3 4 9 6 7, 13 6 ,8 6 6 3 B 8 0 ,5 2 3 ,7 4 6 8 0 ,3 2 9 ,8 7 5 7 9 ,8 19 ,8 2 8 7 8 ,8 8 9 ,2 4 0 7 7, 6 5 9 ,5 7 6 7 6 ,3 8 5 ,4 8 3 74 ,8 14 ,0 6 7 7 2 ,9 4 6 ,9 8 3 7 0 ,5 2 2 ,1 16 4 B 8 0 ,5 2 3 ,7 4 6 8 0 ,7 19 ,0 74 8 0 ,9 0 9 ,3 2 7 8 0 ,7 41 ,1 10 8 0 ,3 17 ,0 0 3 7 9 ,8 6 9 ,0 9 2 7 9 ,1 3 4 ,1 2 7 7 8 ,1 15 ,4 0 0 7 6 ,8 2 9 ,2 0 2 Could Immigration Prevent Population Decline? • 187 T a b . A 3: P ro je c ti o n s fo r o ld -a g e d e p e n d e n cy r a ti o ( p e o p le a g e d 6 5 o r o ld e r/ p e o p le a g e d 1 5 -6 4 , p e r 10 0 p e rs o n s) S o u rc e : o w n p ro je ct io n s V a ri a n te 2 01 2 2 01 5 2 0 2 0 2 0 2 5 2 0 3 0 2 0 3 5 2 0 4 0 2 0 4 5 2 0 5 0 1A 31 ,1 8 3 4 ,9 5 3 9 ,5 3 4 5 ,8 9 5 4 ,1 1 6 0 ,3 3 61 ,2 0 6 2 ,4 9 6 3 ,8 8 2 A 31 ,1 8 3 4 ,6 8 3 8 ,9 0 4 4 ,7 3 5 2 ,1 5 5 7, 5 3 5 7, 8 4 5 8 ,5 4 5 9 ,3 0 3 A 31 ,1 8 3 4 ,4 2 3 8 ,3 1 4 3 ,6 8 5 0 ,4 3 5 5 ,1 5 5 5 ,1 3 5 5 ,5 0 5 5 ,9 8 4 A 31 ,1 8 3 4 ,1 6 3 7, 7 3 4 2 ,6 7 4 8 ,8 3 5 3 ,0 0 5 2 ,7 2 5 2 ,8 7 5 3 ,1 6 1B 31 ,1 8 3 4 ,9 5 3 9 ,5 2 4 5 ,8 5 5 4 ,0 9 6 0 ,5 1 61 ,7 4 6 3 ,5 5 6 5 ,5 6 2 B 31 ,1 8 3 4 ,6 8 3 8 ,9 0 4 4 ,7 3 5 2 ,2 2 5 7, 8 5 5 8 ,6 2 5 9 ,9 4 61 ,4 6 3 B 31 ,1 8 3 4 ,4 2 3 8 ,3 1 4 3 ,6 8 5 0 ,5 0 5 5 ,4 6 5 5 ,8 7 5 6 ,8 5 5 8 ,0 4 4 B 31 ,1 8 3 4 ,1 6 3 7, 7 3 4 2 ,6 7 4 8 ,8 9 5 3 ,2 9 5 3 ,4 3 5 4 ,1 6 5 5 ,1 4 • Hannes Weber188 Fig. A1: Population pyramid of Germany in 2050 (High immigration variant 4A) Note: The sudden increase in natives below the age of 38 is due to the projection defi ning second-generation migrants’ children as natives. See text for details. Source: own projections 0 200 400 600 800 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100+ Without migration background 1st generation immigrants 2nd generation immigrants Females in 1,000 800 600 400 200 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100+ Males in 1,000 Could Immigration Prevent Population Decline? • 189 Fig. A2: Population pyramid of Germany in 2050 (No immigration variant 1B) Source: own projections 0 200 400 600 800 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100+ Without migration background 1st geneneration immigrants 2nd geneneratgion immigrants Females in 1,000 800 600 400 200 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100+ Males in 1,000 Published by / Herausgegeben von Prof. Dr. Norbert F. Schneider Federal Institute for Population Research D-65180 Wiesbaden / Germany Managing Editor / Verantwortlicher Redakteur Frank Swiaczny Assistant Managing Editor / Stellvertretende Redakteurin Katrin Schiefer Copy Editor (German) / Lektorat (deutsch) Dr. Evelyn Grünheid Layout / Satz Beatriz Feiler-Fuchs E-mail: cpos@bib.bund.de Scientifi c Advisory Board / Wissenschaftlicher Beirat Paul Gans (Mannheim) Johannes Huinink (Bremen) Michaela Kreyenfeld (Rostock) Marc Luy (Wien) Clara H. Mulder (Groningen) Notburga Ott (Bochum) Peter Preisendörfer (Mainz) Zsolt Spéder (Budapest) Comparative Population Studies www.comparativepopulationstudies.de ISSN: 1869-8980 (Print) – 1869-8999 (Internet) Board of Reviewers / Gutachterbeirat Martin Abraham (Erlangen) Laura Bernardi (Lausanne) Hansjörg Bucher (Bonn) Claudia Diehl (Konstanz) Andreas Diekmann (Zürich) Gabriele Doblhammer-Reiter (Rostock) Jürgen Dorbritz (Wiesbaden) Anette Eva Fasang (Berlin) E.-Jürgen Flöthmann (Bielefeld) Alexia Fürnkranz-Prskawetz (Wien) Beat Fux (Salzburg) Joshua Goldstein (Berkeley) Karsten Hank (Köln) Sonja Haug (Regensburg) Hill Kulu (Liverpool) Aart C. Liefbroer (Den Haag) Kurt Lüscher (Konstanz) Emma Lundholm (Umeå) Nadja Milewski (Rostock) Dimiter Philipov (Wien) Roland Rau (Rostock) Tomáš Sobotka (Wien) Jeroen Spijker (Barcelona) Olivier Thévenon (Paris) Helga de Valk (Brussel) Heike Trappe (Rostock) Michael Wagner (Köln) © Federal Institute for Population Research 2015 – All rights reserved