Can public spending reduce mortality disparities? Findings from East Germany after reunification

The Journal of the Economics of Ageing xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

The Journal of the Economics of Ageing journal homepage: www.elsevier.com/locate/jeoa

Full Length Article

Can public spending reduce mortality disparities? Findings from East Germany after reunification Tobias C. Vogt ⇑, Fanny A. Kluge Max-Planck-Institute for Demographic Research, Rostock, Germany

a r t i c l e

i n f o

Article history: Available online xxxx Keywords: Mortality differentials Pension income Health care spending Natural experiment Germany

a b s t r a c t Following the reunification of Germany, eastern Germans experienced large increases in life expectancy. The introduction of the West German welfare system was accompanied by mortality reductions, particularly for older East Germans. We use this natural experiment to investigate to what extent increased public social security transfers contributed to the rise in life expectancy. We find that every euro invested in East German pensions and health care yielded on average three hours of additional life expectancy. The rise in public spending was particularly beneficial for older age groups. Investments in health care were more beneficial for reducing all-cause mortality in the East. Still, investments in pensions were important for closing the life expectancy gap to West Germany. Our results suggest that public policy measures geared toward equalizing living standards can also help to narrow mortality differentials. Ó 2014 Elsevier B.V. All rights reserved.

Background There is an ongoing discussion about the relationship between increases in public spending and increases in life expectancy. Most researchers have emphasized the costs associated with rising life expectancy, mainly due to the expenses incurred by older age groups. Some observers have warned that changing age structures within a population and increasing proportions of older age groups will lead to higher public pension costs (Bongaarts, 2004; Jimeno et al., 2008; Börsch-Supan et al., 2003) or health care expenditures (Schulz et al., 2004; McGrail et al., 2000; Anderson and Hussey, 2000). Fewer researchers have examined the impact of public expenditures on rising life expectancy (Murphy and Topel, 2005; Cutler et al., 2006a; Mackenbach et al., 2011). Although these studies have shown that investments in health care can result in lower mortality at the individual level (Farahani et al., 2010), this was found within countries, but not between countries. Several studies have shown that health care expenditures account for only a fraction of inter-country health differentials (Filmer and Pritchett, 1999; Musgrove, 1996). These researchers asserted that income or poverty determine individual health and mortality to a larger extent than public health expenditures. Indeed, a correlation between income and mortality was found in many studies (Smith, 1999). Preston (1975) showed that countries with higher per capita income have lower mortality than those with lower ⇑ Corresponding author. Tel.: +49 381 2081 262. E-mail address: [email protected] (T.C. Vogt).

income, which implies that the level of life expectancy of a country is based on its national income. However, this relationship holds true only up to a certain income threshold; beyond this threshold, additional income does not yield further gains in life expectancy (Deaton, 2003). Thus, some scholars have argued that it is not only the income gradient between countries but also the degree of income inequality within countries that lead to the development of different mortality patterns in comparable countries (Subramanian and Kawachi, 2004; Kitagawa and Hauser, 1973; Wilkinson and Pickett, 2006; Wilkinson, 1996; Lynch et al., 2000). The evidence also suggests, however, that income levels cannot fully explain rising life expectancy levels, and that health care spending and investments in health care technology should also be taken into account (Cutler et al., 2006b). We seek to add to this discussion by addressing the question of whether converging social spending levels lead to converging life expectancy levels by looking at trends in public health care and pension spending. We make use of the division and reunification of Germany, a large natural experiment in which a single population was artificially separated and experienced very different social, economic, and political conditions for four decades. After the Fall of the Berlin Wall and the subsequent adoption of the West German social security system, public spending rose significantly in the East and converged to the West German level. Elderly East Germans benefited most from this process, as their retirement benefits rose sharply, and they started receiving care through the modern health care system imported from the West. At almost the same time, life expectancy in the East, which had been falling

http://dx.doi.org/10.1016/j.jeoa.2014.09.001 2212-828X/Ó 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

2

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx

behind that of the West since the 1970s, started to increase, and eventually caught up to West German levels. By focusing on the elderly, we avoid factors such as migration or education to affect our results. In order to quantify the impact of public spending, especially of increasing health care and pension spending, we proceed as follows. First, we present the development of life expectancy in the two parts of Germany since the 1970s. We calculate a Lee-Carter forecast for East Germany to determine how much life expectancy was gained due to reunification. Second, we describe the role of health care and pension spending in East Germany before and after reunification. Third, using the projected age-specific mortality rates, we calculate a difference-in-difference model in which we estimate the relative elasticity of mortality to public spending. Fourth, we refine our analysis by estimating the effect of pension and health care spending on eastern German mortality. We find some evidence that the significant investment in pensions and health care in the East facilitates convergence in mortality rates between the two regions and a gain in life expectancy. We find some indication that investments in health care in the East had a larger effect on mortality than pension payments. In addition, we find that pension payments also contributed to increase life expectancy to reach the West German level.

Life expectancy since the 1970s The changing institutional arrangements in the East after reunification led to a steep increase in life expectancy among East Germans. Prior to reunification, the gap in life expectancy between East and West Germany was large and increasing. In 1970, the average life expectancy for women in the East and the West was about the same, at around 73.5 years; and the average life expectancy for men in the East was actually 0.8 years higher than the 67.3 years for men in the West. However, between 1970 and the fall of the Berlin Wall, the gap in life expectancy rose substantially, partly because of the inability of the East to keep up with West German spending on health care and pensions (Volpp, 1991; Mielck, 1991; Hockerts, 1994; Schmidt, 1999; Gjonça et al., 2000). This changed rapidly after reunification with the age groups above age 65 benefiting most. In the two decades following the reunification of the country, eastern Germans experienced marked increases in life expectancy. Between 1990 and 2009, women in the East added 6.3 years and men in the East added 7.4 years to their average life expectancies, compared with increases of just 3.5 years for women and 5.1 years for men in the West. Thus, the gap in life expectancy between the East and West narrowed from 2.7 years for women and 3.4 years for men in 1990, to 0.09 years for women and 1.2 years for men in 2011 (Human Mortality Database, 2013). Fig. 1 shows the partial life expectancy below age 651 and the remaining life expectancy at age 65 in contrast to other central and eastern European countries that benefited from the Fall of the Iron Curtain. The purpose of this illustration is to show that the Czech Republic and East Germany prevented death at younger ages relatively well already before 1990. In contrast, remaining life expectancy at age 65 lagged significantly behind. Interestingly, while all countries witnessed increases after 1990, East Germany was the only region that converged to the West German level. This suggests that, apart from changing living standards, the massive transfers from West Germany helped to improve old-age survival more in East Germany than in other transformational countries. Fig. 1 suggests that the convergence of East German life expectancy resulted from mor1 The partial life expectancy at age 65 estimates how many years of life on average a newborn expects to live up to the age of 65.

tality improvements among the ages above 60. This becomes more apparent if we decompose the improvements in life expectancy at birth (Arriaga, 1984). Fig. 2 shows the contribution of different age groups to the gain in life expectancy since the Fall of the Berlin Wall. Mortality improvements among women in the age groups above age 60 were responsible for 75% of the increase in life expectancy at birth. Males in the same age groups contributed 64% on average per year since 1995. This marked increase in life expectancy since reunification was accompanied by soaring public spending on social security and social infrastructure designed to eliminate the disparities between the West and the East. Germany before reunification East German social policy was geared toward enabling citizens to contribute to the productivity and prosperity of the socialist economic system (Schmidt, 1999). It was intended not only to increase (female) labor market participation, but also to minimize the risk that people would drop out of the labor market. Thus, East German policy was largely focused on current or future workers. The decline in fertility rates during the 1970s coupled with out-migration led to an increased emphasis on policies designed to maintain or even increase the East German labor force. Thus, the highest political and fiscal priorities of the East German government were developing family policies, providing occupational medicine, and creating special benefit programs for individuals who were deemed particularly valuable for the East German economy (Schmidt, 1999). Meanwhile, policy makers showed considerably less interest in older East Germans who had left the labor market (Mientus, 2006). At the beginning of the 1970s, the average East German pension income was only 26.1% of the average labor income (Schmidt, 1999). Whereas the status-preserving West German pension system granted beneficiaries 43% of their gross wages, which corresponded to a net replacement rate of 63% (EU Report 2006,), the East German system caused a majority of pensioners to spend their retirement in a precarious social and financial situation (Queisser, 1664). A voluntary supplemental pension scheme designed to alleviate this problem was established in 1971. In 1990, those retirees who had contributed to the scheme over the whole period were receiving 70 East German marks in addition to the average East German monthly pension of 380 East German marks (Meinhardt, 2000). The social and political disregard for the elderly was also implicit in the availability of health care resources. Older East Germans suffered the most from the underfunded and technically outdated health care system. During the 1970s and 1980s, public investments in the health care system, 93% of which was owned by the state, declined to under 3% of state expenditures (GDR Statistical Yearbook 1990). In contrast, total health expenditures amounted to 9% in West Germany in 1985 (OECD, 2013). In 1990, it was estimated that East German health care technologies and pharmaceuticals were lagging 15 to 20 years behind Western standards (Mielck, 1991). At the same time, the shortage of medical personnel and the poorly maintained medical infrastructure led to an undersupply of health care (Busse and Riesberg, 2004). This failure to keep up with Western standards was reflected in the rise in the number of deaths from treatable causes, especially circulatory diseases (Nolte et al., 2002). The reunification of Germany Immediately after the Wall fell, large amounts of money started flowing from West to East. As part of the economic, social, and cur-

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

3

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx

Poland Czech Republic Hungary

East Germany West Germany

Females Age 0−65

Poland Czech Republic Hungary

Females Age 65

East Germany West Germany

Males Age 0−65

Males Age 65 seq(1, 10)

64

20 10

15

Remaining Life Expectancy

62 60

Partial Life Expectancy

1970

1980

1990

2000

Calendar Years

2010

5

56 5

56

10

58

15

Remaining Life Expectancy

60 58

Partial Life Expectancy

62

20

64

25

25

seq(1, 10)

1970

1980

1990

2000

2010

Calendar Years

1970

1980

1990

2000

Calendar Years

2010

1970

1980

1990

2000

2010

Calendar Years

Fig. 1. Partial life expectancy and remaining life expectancy for females (left) and males (right), 1970–2010, East and West Germany, Poland, Czech Republic and Hungary. Source: Human Mortality Database (2013), own calculations

Fig. 2. Contribution of different age groups to the increase of East German life expectancy at birth. Source: Human Mortality Database 2011

rency union between the two parts of Germany, the West German government channeled large sums of money into the East in an effort to equalize the infrastructure and living standards across the country. In the early years, more than 160 billion marks were transferred to the East (Wagner, 2001). The social union led to the introduction of the West German social insurance system. Thus, older East Germans became beneficiaries of the generous West German public pension system. Retirement benefits were recalculated under the West German pension system, and individual pension benefits rose significantly. Fig. 3 shows the

convergence of public spending as West-East ratios of average pension levels and per capita total health expenditures. Before 1990, the average East German retiree received a pension of about 40% of the average West German pension, even assuming the overvalued exchange rate of 1:1. Estimations of a more accurate exchange rate between the East and West German currencies range from 1:5 to 1:10, but even the 1:1 exchange rate brought a two-and-a-half-fold increase in nominal income. At the beginning of the 1990s, the average size of pensions in the East already exceeded western levels. Still, women in the East reached the

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

4

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx

Both Sexes 1:1 Both Sexes 5:1

●

Females Males Health Spending Per Capita

reunification

10 5

reunification

10 5

West/East Ratios

15

Average Pensions

● ● ● ● ● ● ● ● ● ●

0

0

●

1980

1986

1992

1998

1980

1986

1992

1998

Calendar Years Fig. 3. West/East ratios of average pension benefits and per capita health expenditures. Source: Statistical Yearbooks FRG and GDR 1980–2000, own calculations

average pension level of women in the West more quickly than men.2 This was mainly due to the fact that eastern German pensioners had longer working histories in general and considerably higher rates of female labor force participation, even for mothers. In 1989, the female labor force participation rate was 89% in the East and only 56% in the West (Rosenfeld et al., 2004). At the same time, average health care expenditures began to rise. Before reunification, total health expenditures per capita were up to three times higher in the West than in the East. This ratio changed quickly after 1990. Within just three years, per capita health expenditures in the East had reached the western level, and exceeded it thereafter. The additional health expenditures during the mid-1990s are attributable to investments in infrastructure and the underlying social and political goal of establishing comparable health care standards in East and West. Data and Methodology We analyze public pension and public health care data by single years of age for all-cause mortality. The detailed estimates on individual consumption of public expenditures are obtained from the National Transfer Accounts (NTA) database.3 The NTAs provide detailed information about public transfers to the elderly, including health or long-term care services and payments from public pension schemes. For this part of the analysis, the mean for each age group by single years of age provides useful information, as public transfer flows do not depend on income, but on age utilization patterns. We cover the period from 1980 to 2000, and we distinguish between the former GDR and the FRG in the analysis. Micro-data for West Germany are available for 1978, 1993, and 2003. In addition to the age-specific means of NTA, we calculate age- and sex-specific estimates. The years in between the years of observation are interpolated. However, all of the profiles are matched to the health care and pension expenditures of the respective year from the National Accounts. Thus, while a bias can occur within age groups, the age 2 Average pension benefits for East Germany by sex were not available for the years before reunification. 3 The theoretical framework of NTA builds upon Samuelson (1958), Diamond (1965), and Lee (1994). A detailed overview of the construction of NTAs is available at www.ntaccounts.org.

profiles for these specific items are rather robust and do not change from year to year (Kluge, 2011). Data for East Germany are available for the period immediately after reunification and for the year 2003, and could be used to show the trend of expenditures. For the period before 1990, information is scarce. Several sources are combined to calculate age- and sex-specific expenditures for individuals in the East before reunification. The estimates are based on our knowledge that the retirement age was 60 for women and 65 for men, and that there was relatively little variation in pensions among individuals. The constructed health care profile takes into account the fact that more attention was given to people of younger or working ages in the East than in the West. Nevertheless, as we mentioned above, the expenditures do not change considerably over time, and the results are quite robust to moderate changes in the profiles. In addition, as all of the estimations are scaled to fit the National Accounts, our main concern is the potential for within-age distortions. In addition to the profiles of social spending by age, sex, and region before and after reunification, we use mortality rates obtained from the Human Mortality Database for single years of age (Human Mortality Database, 2013). As a first step, we estimate a simple difference-in-difference model that uses the reunification of Germany as a structural break to determine the amount of money that was invested for each additional year of life. Here, we have two groups of eastern German individuals. The first is the real observed group with the real investment in per capita pensions and health care, and their observed mortality rates. Then, to provide a contrast, we project a second hypothetical population using the Lee-Carter method (Lee and Carter, 1992) and social security expenditures with a linear extrapolation of the expenditures observed before reunification. The baseline expenditures and hypothetical increase in life expectancy are then compared to the real observed scenario. This allows us to estimate how much additional life expectancy can be attributed to the massive transfers to the East. In a more refined analysis, we use a linear model to estimate the importance of changing levels in public spending on the rising life expectancy in East Germany after the reunification. Human mortality at older ages is modeled best using a Gompertz model (Beard, 1959; Vaupel et al., 1979). We use the Gompertz approach in the form of l^x ¼ abx to model mortality rates as a function of age. We assume that b, also called the rate of aging, remains unaffected by the societal transformation. The changes in the environment, in our case changes in social spending, have rather an effect on the level of mortality, a. The regression equation of our linear model is given by

  t þb ^0 þ b ^1 x þ b ^2 Pens ^3 Health ^ ^ x;t ¼ b ln l t þ b4 Sex þ x;t

ð1Þ

^ x;t denotes the log of the mortality hazard by age and time. where l ^0 and b ^1 refer to the parameters of the Gompertz model. Fig. 4 b shows the log mortality for East Germans aged 60–90 for the years

Fig. 4. Mortality for ages 60 to 90 for East Germany, 1980–2000, R-Squared 90.0. Source: Human Mortality Database 2014, own calculations

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

5

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx

1980 and 2000. It reflects that mortality rates decline almost linearly for all ages during the two decades. The ’pure’ Gompertz model that only takes account for mortality over age explains 90% of the variance for the given ages and years. Thus, we assume that the rest of the variation or decline in mortality is explained by changes over time, t. To capture the change over time we use a twofold strategy. First, we estimate the timing effect by including calendar years as a predictor for lx;t . In a second, step we seek to identify if the effect of time itself represents the effect of changing social expenditure levels over time. Therefore, we introduce pension and health care payments by sex as yearly averages. We run the model separately for East and West Germany.

Table 1 Regression results for East Germany, reduced model with only three covariates.

(Intercept) Age Year SexMale

Estimate

Std. error

P-value

40.0332 0.10502 0.02563 0.48689

1.099024 0.000374 0.000552 0.006686

<2e16 <2e16 <2e16 <2e16

⁄⁄⁄ ⁄⁄⁄ ⁄⁄⁄ ⁄⁄⁄

Significance levels: ⁄ p < 0.05; ⁄⁄p < 0.01. Source: NTA data for East Germany, Human Mortality Database, own calculations. ⁄⁄⁄ p < 0.001

Table 2 Regression results for East Germany, model substituting year with expenditures on pensions and health.

Results The difference-in-difference approach is not meant to provide an exact measure how much money has to be invested to increase an individual’s life expectancy by one year. In fact, we would prefer to show the potential impact public transfer could have on survival. Therefore, we use this approach as a first step, to show how much money was invested per capita and the corresponding gains in life expectancy. The results from the difference-in-difference model indicate that public spending contributed to closing the gap in life expectancy. Fig. 5 depicts the expenditures and life expectancy 10 years before and after reunification. Our comparison of the means of the hypothetical and real population reveals that per capita social security spending after reunification amounted to 5077 euros for the real population, but just about one-tenth of this spending for the hypothetical eastern population, assuming an exchange rate of 5:1. As the two groups faced identical expenditures and mortality risks prior to reunification, this translates into a 4500-euro difference in per capita spending on eastern German individuals. This enormous increase in per capita spending was accompanied by an increase in life expectancy to 75.2 years (the Lee-Carter forecast predicted a mean life expectancy for the hypothetical population of 73.4 years (Vogt, 2013)). In this simple scenario, one euro invested per capita per year in pensions or health care was associated with an increase in life expectancy of three hours per year. We are, of course, aware that a variety of mortality reducing factors changed: pollution levels declined, nutrition and health behavior improved, life style factors changed, and living standard rose. Yet, changes in life style like reducing smoking affect mortality in the long run (Preston et al., 2010). Pollution is assumed to play only a minor role as more polluted areas caught up as quickly as less polluted areas (Bobak and Feachem, 1995). The rising living standard can also be an expression of higher public transfers as it includes better health facilities or the individual’s possibilities to

Without Reunification

Reunification Total Social Expenditure

reunification

5500 per capita in Euros 2800 3700 4600

70

1000

71

72

1900

73

Life Expectancy at Birth 74 75 76 77 78

reunification

79

80

6400

Life Expectancy

1980

1990

2000

Years

1980

1990

2000

Fig. 5. Real and projected life expectancies and social security expenditures, East Germany 1980–2000. Source: Human Mortality Database 2012, Statistical Yearbook GDR 1990, NTA 2003

(Intercept) Age AvPens AvHealth SexMale ⁄

Estimate

Std. error

P-value

10.75 0.105 3.9E05 9.7E05 0.4808

0.02991 0.000376 5.11E06 1.56E05 0.008153

<2e16 <2e16 3.84E14 8.24E10 <2e16

⁄⁄⁄ ⁄⁄⁄ ⁄⁄⁄ ⁄⁄⁄ ⁄⁄⁄

⁄⁄

Significance levels: p < 0.05; p < 0.01; Source: NTA data for East Germany, Human Mortality Database, own calculations. ⁄⁄⁄ p < 0.001

invest in better housing, nutrition or goods and services that became available only after reunification. Also in other studies, pensions and health care are deemed the main factors that contributed to increasing life expectancy (Bobak and Marmot, 1996). This finding is in line with results which showed that pollution or migration are of only minor importance in the convergence of mortality levels, and that better health care and rising income are the main drivers of increasing life expectancy (Diehl, 2008). This is especially true for individuals of retirement age. Eastern pensioners received higher per capita public pension payments than their counterparts in the West starting in the early 1990s. The linear model is estimated separately for East Germany in the period 1980–2000. We use a two step strategy to estimate the importance of public pensions and public health care expenditures on the declining mortality levels in the East. About 98 percent of the changes in mortality are explained with only three variables: age, year, and sex. Now, we assume that the explanatory power of the model should result in similar values if we substitute the variable year with pension and health care expenditures. In case we are able to achieve a similar explanatory power in the variance, the annual changes could be driven by the changes in our two covariates and could indicate that additional life expectancy can be ’bought’ by investments in health infrastructure or public pensions. The regression results are given in Table 1 and 2. The models show the inverse relation between increasing public spending on pensions and health care and decreasing mortality rates. Mortality naturally increases by age and men have relative higher mortality as compared to females. Being a man increases mortality by about 0.48. Furthermore, we find that health investment play an important role in reducing mortality for older East Germans. In general, one euro invested in health care in the East was about 2.5 times as beneficial in reducing log mortality as one euro invested in pensions. This hints to the fact that pension payments were less beneficial for declining mortality in East Germany but important for the mortality convergence to the West German mortality level. Our model implies that other potential factors for the mortality convergence only play a minor role. The significant impact of both, public pensions and health expenditures, hints to the crucial role of public spending in leveling mortality differences.

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

6

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx

The second model is able to explain a comparable share of the variance in mortality rates as the model only controlling for time. Thus, we assume that public pension payments and health care are important drivers of the mortality increase in East Germany after reunification.

Discussion In this study, we investigate how social security spending in the form of public pension and health care expenditures affects differences in mortality. For the analysis, we use the natural experiment setting provided by the division and reunification of Germany to illustrate how different institutional settings influence observed mortality patterns. We focus on the years 1980 to 2000 because these decades include a period when the life expectancy gap between the two parts of Germany was growing, as well as the period after 1990 when the catch-up process occurred. Our results indicate that both health and pension payments substantially contributed to the reduction in mortality differences between the East and the West after reunification. In addition, we find some evidence that, in the East, the investment in health infrastructure had a greater effect on all-cause mortality than pension expenditures. This could mean that, when there are two systems with major differences in health care infrastructure, the greatest gain in life expectancy can be achieved by closing this gap; but when there are two systems with comparable levels of health care provision, personal income (e.g., public pensions) is of greater importance. If income and health investment would not have played an important role, then these factors could not explain the variance over time. We should also bear in mind that individuals have to be healthy in the first place in order to reach a certain age and become eligible to receive pension payments. In this study we focus explicitly on pensions and health care. Still, the mechanism behind rising pensions/income to foster survival is not clear. It relates to a number of factors such as better living standard or housing that can be bought with higher income. This is a major shortcoming of our study as the macro perspective cannot take into account how public spending influences mortality at the individual level. Studies that focus on the relationship between changing social spending levels and mortality may also face a causality problem as longer lives are not only the result of increasing investments but also the trigger. This could be certainly the case in West Germany for the years studied. However, we assume that dropping mortality levels in East Germany are rather based on the introduction of the West German social security system. Another important limitation of our study is that we cannot control for every aspect that changed in the lives of elderly East Germans. Unlike an ideal natural experiment the German reunification altered more than one mortality condition. Not only health care and pensions became available but living conditions in general improved. Thus, we believe that increases in pension income helped older East Germans to improve their living standard. After the Fall of the Berlin Wall, western goods and services became available and, most importantly, affordable with the adoption of the West German pension system. This may have helped to improve nutrition or the housing situation of the elderly and keep up to the western living standard. The income convergence may also be a reason why the mortality of older East Germans converged to the West German level while other central and eastern European countries are still lagging behind. Likewise, eastern Europeans or younger East Germans faced high economic uncertainty during the economic turmoil at the beginning of the 1990s. One third of the East German labor force lost its job and was forced into migration, long phases of unemployment and general job insecu-

rity (Dornbusch et al., 1992). While these younger age groups may have suffered, East German pensioners were economically protected from the stressful transformation of their society. Thus, the results seem to capture an important cause of reduced mortality. East Germany was only one of the countries that benefited from the fall of the Iron Curtain, but it benefited the most in terms of the pace of increasing life expectancy. Nolte et al. (2002) found that, in the 1990s, improvements in health care had less of an effect in Poland than in East Germany, and that life expectancy did not increase as quickly in Poland as in East Germany. Thus, it appears that the massive transfers of money from West to East Germany played a major role in lowering mortality. These results may also have important implications for other countries, such as the U.S., eastern Europe, or developing countries. They show that investments in public health infrastructure and pension programs are important drivers of increased life expectancy. Public spending geared towards improving health care or income levels seems to ‘buy’ increases in life expectancy at moderate costs. The investments in life saving in the East were relatively cheap compared to results from studies which estimate the costs of gained years of life based on expenses for a variety of life saving measures ranging from smoking bans to traffic controls. Tengs et al. (1995) estimated the costs of 500 interventions for saving one year of life. They found that the median intervention cost for a life year saved was $42,000. In comparison, a year of life saved in East Germany was as inexpensive as €2920 on average per year. This seems to correspond to results from the Netherlands that quantified the effect of increasing health care spending on reduced old age mortality (Mackenbach et al., 2011). Additionally, our results indicate that reducing income disparities within a single country also reduces differences in mortality (Kitagawa and Hauser, 1973). The formerly disadvantaged older East Germans experienced a convergence of their living standards with those of their West German compatriots. Consequently, their life expectancy began to catch up to the life expectancy of West Germans. At the same time, modern health care facilities became available in the East, which led to a reduction in mortality, mainly due to better treatment of circulatory disease. The gains in life expectancy also reflect the gains that older West Germans experienced during the 1970s and 1980s. Reunification and the subsequent increase in health care spending led to a cardiovascular revolution, a process that affluent western societies had undergone two decades earlier (Vallin and Meslé, 2004). We therefore conclude that life expectancy is highly plastic to changing living conditions, even at the oldest ages (Vaupel et al., 2003). Our analysis suggests that public spending that seeks to reduce disparities in living conditions may lead to a convergence in mortality levels within a country. Acknowledgments The authors are grateful to James W. Vaupel and James E. Oeppen for valuable comments and suggestions and to Jutta Gampe for statistical guidance. We also appreciate the helpful comments by two anonymous referees and by Mauricio Avendano Pabon on an earlier version of the manuscript. References Anderson, G., Hussey, P., 2000. Population aging: a comparison among industrialized countries. Health Aff. 19 (3), 191–203. Arriaga, E.E., 1984. Measuring and explaining the change in life expectancies. Demography 21 (1), 83–96. Beard, R., 1959. Note on Some Mathematical Mortality Models. In: Woolstenholme G.E.W., O’Connor M. (Eds.), Little, Brown and Company, Boston, pp. 302–311. Bobak, M., Feachem, R.G.A., 1995. Air pollution and mortality in Central and Eastern Europe. Eur. J. Public Health 5, 82–86.

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001

T.C. Vogt, F.A. Kluge / The Journal of the Economics of Ageing xxx (2014) xxx–xxx Bobak, M., Marmot, M., 1996. East-West mortality divide and its potential explanations: proposed research agenda. Br. Med. J. 312 (7028), 421–425. Bongaarts, J., 2004. Population aging and the rising cost of public pensions. Popul. Dev. Rev. 30 (1), 1–23. Börsch-Supan, A., Heiss, F., Ludwig, A., Winter, J., 2003. Pension reform, capital markets and the rate of return. German Econ. Rev. 4 (2), 151–181. Busse, R., Riesberg, A., 2004. Health Care Systems in Transition: Germany. WHO Regional Office for Europe on behalf of the European Observatory on Health Systems and Policies, Copenhagen. Cutler, D.M., Deaton, A.S., Lleras-Muney, A., 2006a. The Determinants of Mortality. NBER Working Papers, No. 11963:1–44. Cutler, D.M., Rosen, A.B., Vijan, S., 2006b. The value of medical spending in the United States, 1960–2000. New Engl. J. Med. 355, 920–927. Deaton, A., 2003. Health, inequality, and economic development. J. Econ. Lit. 41, 113–158. Diamond, P., 1965. National debt in a neoclassical growth model. Am. Econ. Rev. 55 (5), 1126–1150. Diehl, K., 2008. Mögliche Ursachen für die rasche Reduktion der ostdeutschen Übersterblichkeit nach der Wiedervereinigung. Zeitschrift für Bevölkerungswissenschaft 33, 89–110. Dornbusch, R., Wolf, H., Alexander, L., 1992. Economic transition in Eastern Germany. Brookings Pap. Econ. Act. 1992 (1), 235–272. EU Report 2006, 2006. Current and Prospective Theoretical Pension Replacement Rates. Technical Report, Report by the Indicators Sub-Group (ISG) of the Social Protection Committee (SPC). Farahani, M., Subramanian, S., Canning, D., 2010. Effects of state-level public spending on health on the mortality probability in India. Health Econ. 19 (11), 1361–1376. Filmer, D., Pritchett, L., 1999. The impact of public spending on health: does money matter? Soc. Sci. Med. 49, 24. Gjonça, A., Brockmann, H., Maier, H., 2000. Old-age mortality in Germany prior to and after reunification. Demogr. Res. 3 (1). Hockerts, H.G., 1994. Grundlinien und soziale Folgen der Sozialpolitik in der DDR. In: Kaelble, H., Kocka, J., Zwahr, H. (Eds.), Sozialgeschichte der DDR, Stuttgart, Klett- Cotta, pp. 519–544. Human Mortality Database, 2013. Human Mortality Database. University of California, Berkeley (USA) and Max- Planck- Institut for Demographic Research, Rostock (Germany). Data downloaded: March 11, 2012. Jimeno, J., Rojas, J., Puente, S., 2008. Modelling the impact of aging on social security expenditures. Econ. Model. 25 (2), 201–224. Kitagawa, E.M., Hauser, P.M., 1973. Differential Mortality in the United States: A Study of Socioeconomic Epidemiology. Harvard University Press, Cambridge. Kluge, F.A., 2011. The Individual Economic Lifecycle and Its Fiscal Implications in an Aging Germany – Findings from National Transfer Accounts. PhD thesis. University of Rostock. Lee, R., 1994. The formal demography of population aging, transfers, and the economic life cycle. In: Martin, L.G., Preston, S.H. (Eds.), Demography of Aging. National Academy Press, Washington, DC, pp. 8–49. Lee, R., Carter, L., 1992. Modeling and forecasting US mortality. J. Am. Stat. Assoc. 87 (419), 659–671. Lynch, J., Smith, G., Kaplan, G., House, J., 2000. Income inequality and mortality: importance to health of individual income, psychosocial environment, or material conditions. Br. Med. J. 320 (7243), 1200–1204. Mackenbach, J.P., Slobbe, L., Looman, C.W.N., vander Heide, A., Polder, J., Garssen, J., 2011. Sharp upturn of life expectancy in the Netherlands: effect of more health care for the elderly? Eur. J. Epidemiol. 26, 903–914.

7

McGrail, K., Green, B., Barer, M., Evans, R., Hertzman, C., Normand, C., 2000. Age, costs of acute and long-term care and proximity to death: evidence for 1987–88 and 1994–95 in British Columbia. Age Ageing 29 (3), 249–253. Meinhardt, V., 2000. Der Prozess der Angleichung im Bereich der sozialen Sicherung: 10 Jahre nach der Einführung der Währungs-, Wirtschafts-und Sozialunion. Vierteljahrshefte zur Wirtschaftsforschung/Q. J. Econ. Res. 69 (2), 225–248. Mielck, A., 1991. MortalitSt und Lebenserwartung: Unterschiede zwischen Ost- und Westdeutschland. Sozialer Fortschritt 9, 220–223. Mientus, F., 2006. Zur Quantifizierung wiedervereinigungsinduzierter Kosten und Transfers in der gesetzlichen Rentenversicherung. Mannheim Research Institute for the Economics of Aging, Study No. 4. Murphy, K., Topel, R., 2005. The Value of Health and Longevity. Technical Report, National Bureau of Economic Research. Musgrove, P., 1996. Public and Private Roles in Health, World Bank. Nolte, E., Scholz, R., Shkolnikov, V., McKee, M., 2002. The contribution of medical care to changing life expectancy in Germany and Poland. Soc. Sci. Med. 55, 1905–1921. OECD, 2013. OECD Health Data: Health Expenditure and Financing, OECD Health Statistics (database), last access Oct 29, 2013. Preston, S., 1975. The changing relation between mortality and level of economic development. Popul. Stud. 29 (2), 231–248. Preston, S.H., Glei, D.A., Wilmoth, J.R., 2010. Contribution of Smoking to International Differences in Life Expectancy. International Differences in Mortality at Older Ages: Dimensions and Sources, pp. 105–131. Queisser, M., 1996. Pensions in Germany, World Bank Policy Research Working Paper, 1664. Rosenfeld, R., Trappe, H., Gornick, J., 2004. Gender and work in Germany: before and after reunification. Annu. Rev. Sociol. 30, 103–124. Samuelson, P.A., 1958. An exact consumption-loan model of interest with or without the social contrivance of money. J. Polit. Econ. 66 (6), 467–482. Schmidt, M., 1999. Grundzüge der Sozialpolitik in der DDR. Springer. Schulz, E., Leidl, R., König, H., 2004. The impact of ageing on hospital care and longterm care – the example of Germany. Health Policy 67 (1), 57–74. Smith, J., 1999. Healthy bodies and thick wallets: the dual relation between health and economic status. J. Econ. Perspect. 13 (2), :145–166. Subramanian, S.V., Kawachi, I., 2004. Income inequality and health: what have we learned so far? Epidemiol. Rev. Environ. Health 26, 78–91. Tengs, T.O., Adams, M.E., Pliskin, J.S., Safran, D.G., Siegel, J.E., Weinstein, M.C., Graham, J.D., 1995. Five-hundred life-saving interventions and their costeffectiveness. Risk Anal. 15 (3), 369–390. Vallin, J., Meslé, F., 2004. Convergence and divergence in mortality. A new approach to health transition. Demogr. Res. Spec. Collection 2, 12–40. Vaupel, J., Manton, K., Stallard, E., 1979. The impact of heterogeneity in individual frailty on the dynamics of mortality. Demography 16 (3), 439–454. Vaupel, J., Carey, J., Christensen, K., 2003. It’s never too late. Science 301 (5640), 1679–1681. Vogt, T.C., 2013. How many years of life did the fall of the Berlin Wall add? A projection of East German life expectancy. Gerontology 59 (3), 276–282. Volpp, K.G., 1991. The structure of health care delivery in communist East Germany. Medizin Mensch Gesellschaft 16, 3–13. Wagner, I., 2001. Gegenstand und finanzielle Bedeutung des Solidarpakts. ifo Dresden berichtet 2/2001, 3/2001(3):46–53. Wilkinson, R.G., 1996. Unhealthy Societies: The Afflictions of Inequality. Routledge. Wilkinson, R., Pickett, K., et al., 2006. Income inequality and population health: a review and explanation of the evidence. Soc. Sci. Med. 62 (7), 1768 (1982).

Please cite this article in press as: Vogt, T.C., Kluge, F.A.. The Journal of the Economics of Ageing (2014), http://dx.doi.org/10.1016/j.jeoa.2014.09.001