Especially those who migrated at older ages and intra-European migrants suffer from migration in terms of wealth, finds a new MEA-study. January 2017 - Wealth is generally considered a long-run indicator of well-being. Particularly older families may rely on their wealth and on pensions as a major resource. For migrant households these resources might substantially differ from those of natives. This is due to differences in, for example, savings behaviour, rates of return or access to social welfare and social security. Given the surge of reforms aimed at reducing the generosity of the social security system all around Europe, immigrants form a large group that is potentially at risk of poverty in retirement. In order to face this issue, it is fundamental to assess whether there is a wealth gap and, if yes, what its main drivers are. A new study by Irene Ferrari estimates the gap between native, mixed and migrant households. Furthermore, the study goes beyond the average gap and explores whether some of the immigrant groups are better off than others. Potential reasons for a wealth gap between immigrants and natives Researchers expect individuals with a migration history to be worse off than natives for various reasons; the following three are among them: First, in terms of wages, immigrants face a relative earnings gap at arrival. This is caused by a lack of information about the host countries and barriers such as language skills or an uncertain legal status. Second, recent studies have shown that immigrants receive less private transfers such as gifts and inheritances, and they are less likely to be home owners. Finally, countries' regulations covering immigrant welfare eligibility may also contribute to a wealth gap. The access to social welfare programs is often limited for immigrants and some social security or pension rules require a minimum number of contribution years. Even when immigrants are able to meet eligibility criteria, they may only reach lower benefits because of lower earnings or fewer contribution years. Wealth gap varies over wealth distribution The study finds evidence that immigrant and mixed households are on average worse off than comparable natives. However, the wealth gap varies along the wealth distribution: The gap is higher for lower wealth levels, then decreases and finally becomes big and negative for the top 20% of wealth distribution. In other words: immigrants in the top percentiles exhibit higher wealth values than natives. Hence, households in the lower part of the wealth distribution especially suffer from this wealth gap. It is particularly pensions which drive this bigger gap of migrant households with a low wealth level. This reflects the difficulties of immigrants to meet the required minimum of contribution years and the lower benefits due to lower earnings. Migrant households in the upper part of wealth distribution, on the other hand, are significantly better off than natives. Socioeconomic factors play an important role The analysis further suggests that immigrant households which are worse-off are more likely to have migrated from countries outside of Europe. Additionally they are more likely to have a lower income, are less healthy, less educated, and are more likely to come from poorer families. Furthermore, the analysis indicates that the gap at lower wealth levels is driven by early childhood conditions, such as the size of the house. Better-off immigrant households, on the contrary, are more likely to have migrated from other European countries, have a higher income, are better educated, are healthier and come from richer families. For this reason, the study conducts an additional analysis that considers the heterogeneity of the migrant population. Heterogeneity of migrant population in terms of age at migration and origin For this analysis, the study distinguishes between individuals who migrated from another European country or from a non-European country, and between individuals who migrated before or after the age of 18. This distinction makes sense considering that intra-European immigrants should face lower migration costs and that young immigrants should benefit from schooling in the host country and have more time to settle. The results of the analysis are surprising: Whereas the assumption regarding younger immigrants can be confirmed, intra-European immigrants seem to suffer more from migration in terms of wealth than non-European immigrants. With a median wealth of 327,000 Euro, they are on average still better off than non-European immigrants with a median wealth of 158,000, but they exhibit a bigger gap when compared to a peer group of natives. Costs of migration are better absorbed by non-European households A likely explanation for the bigger gap between intra-European immigrants and natives with comparable characteristics is that highly-educated individuals moving to a foreign country may have more difficulties in finding a job that fits their education level than lower-educated individuals. The well-documented gap in wages that migrants experience at their arrival may thus be bigger for better-educated migrants, who may consequently suffer a permanent wealth loss. Further research is needed in order to understand the origin of the gap. The full MEA Discussion Paper can be read here .

Financial literacy and private old-age provision Private old-age provision is growing increasingly important in times of demographic change and mounting strains on the public pension system. For many individuals in Germany the “need” to save for old-age in addition to the state pension is new and households seem to face difficulties saving for old age due to the high complexity and the large variety of old-age savings contracts. Therefore, it is important to evaluate who saves and who does not and whether this decision is related to financial knowledge of individuals. How much do individuals know? Financial literacy is measured on the basis of three simple questions: 1. Understanding of Interest Rate (Numeracy) “Suppose you had 100€ in a savings account and the interest rate was 2% per year. After 5 years, how much do you think you would have in the account if you left the money to grow: more than 102€, exactly 102€, less than 102€?” 2. Understanding of Inflation “Imagine that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, would you be able to buy more than, exactly the same as, or less than today with the money in this account?” 3. Understanding of Risk and Diversification “Do you think that the following statement is true or false? “Buying a single company stock usually provides a safer return than a stock mutual fund.” About 85% of the households know the correct answer to the interest question, and more than 86% answer the inflation question correctly. Taken together around 82 % of the individuals comprehend both, the functioning of interest and inflation. And 60 % of the individuals understand the relationship of risk and diversification. Overall around 52 % of the individuals give correct answers to all three considered questions of financial literacy. Who knows a lot and who knows little? The next question is: who is able to give three correct answers and who is not? Financial literacy relates to higher levels of wealth, higher income and higher education. Moreover, men are more likely to know all the answers than women and individuals older than 55 are less literate than younger individuals. Individuals in East and West are equally literate, when controlling for differences in income, wealth and education. How is financial literacy related to retirement savings? A positive correlation of financial literacy and financial decision making is identified: more literate households are more likely to save privately for their old-age and at the same time households saving privately for their old-age acquire financial knowledge to improve their investment decisions. Interestingly, the possession of a state subsidised Riester contract is related to lower levels of financial literacy than the possession of other non-subsidised forms of private old-age provision. On the one hand this indicates that Riester subsidies to some extent successfully encourage individuals with lower financial knowledge to save privately for old-age. Nevertheless, individuals in the lowest income quintile still have low levels of private coverage despite the high subsidies. At the same time they show the lowest levels of financial literacy. On the other hand, if the causal relationship runs in the opposite direction, individuals signing a Riester seem to acquire some financial knowledge in the course of this process. However, they acquire less financial knowledge than individuals with non-subsidised old-age savings contracts. Thus, the analysis above suggests that the subsidies provided by Riester successfully encourage certain groups in the middle of the income distribution and with moderate levels of financial literacy. However, the poorest 20% of the households still do not respond to the policy measures. Thus, more effort is needed here. It would probably be efficient to target specific groups at risk with financial information. more information: Financial Literacy and Private Old-age Provision in Germany MEA Discussion Paper: 192-09 Tabea Bucher-Koenen Picture: FrankU - Fotolia.com

Wie die Deutschen ihr Vermögen mehren- Anlageverhalten im Wandel Im internationalen Vergleich unterscheiden sich die Finanzportfolios deutscher Haushalte noch immer deutlich. Der Anteil von Anlagen bei Banken und Sparkassen liegt in Deutschland vergleichsweise hoch, während Aktien und Investmentfonds eine eher untergeordnete Rolle spielen. Diese erfreuten sich im Laufe der letzten 15 Jahre zwar eines starken Wachstums, schaffen es aber nicht, sich dem durchschnittlichen internationalen Anteil anzunähern. Insbesondere Lebensversicherungsprodukte nehmen in Deutschland weiterhin eine herausgehobene Stellung im Portfolio ein. Durchschnittlich 30 Prozent des Finanzvermögens deutscher Haushalte werden in diese investiert, während exakt vergleichbare Produkte in vielen anderen Ländern gar nicht existieren. Speziell ist hier die Kombination aus Garantieverzinsung und steuerlicher Förderung. Ungeachtet der weiterhin bestehenden Unterschiede, besonders zu den Portfolios angelsächsischer Haushalte, hat sich das Anlageverhalten der Deutschen in den vergangenen Jahrzehnten deutlich gewandelt (siehe Abb. 1). Anlagen in Aktien und Wertpapierfonds machten zu Beginn der 80er Jahre gerade einmal 6 Prozent des Bruttofinanzvermögens aus. 1993 erreichte der Anteil 10 Prozent und konnte sich, nicht zuletzt durch den Aktienboom der 90er, auf etwa 20 Prozent im Jahr 2003 nochmals verdoppeln. Die Aktienmarktbaisse der vergangenen Jahre beeinträchtigte zwar den Portfolioanteil von Direktanlagen in Aktien, jedoch blieb der Anteil der Aktienbesitzer nahezu unverändert. Investmentfonds erfreuten sich derweil unverminderter Popularität, sodass der gemeinsame Anteil beider Anlageformen zwischen 1998 und 2003 nicht zurückging. Im Gegensatz dazu haben Anlagen bei Banken und Sparkassen seit Beginn der 80er Jahre ca. ein Drittel ihrer früheren Bedeutung eingebüßt: der Portfolioanteil sank von 27 Prozent im Jahr 1983 auf unter 20 Prozent im Jahr 2003. Neben der Analyse der Portfolioveränderung über die Zeit widmete sich der Forscher auch den Anteilsunterschieden zwischen verschiedenen Jahrgängen und Altersgruppen. Dazu wurden die Jahrgänge zu Kohorten (Gruppen) zusammengefasst und ihr Anlageverhalten über Jahre hinweg beobachtet. Die nach Altersklassen getrennte Untersuchung zeigt, dass Anlagen am Kapitalmarkt in allen Altersgruppen an Bedeutung gewonnen haben. Ebenso lässt sich ein Rückgang der Anlagen bei Banken und Sparkassen bei nahezu allen Altersgruppen feststellen. Lebensversicherungen haben bei den jüngsten und den ältesten Altersgruppen verloren. Abbildung 2 stellt die Entwicklung des Anteils der Haushalte dar, die eine Lebensversicherung besitzen. Für jeden Geburtenjahrgang zeigt eine Linie in entsprechender Farbe die Entwicklung über seinen Lebenszyklus auf. Der Grund für den Rückgang des Anteils der Lebensversicherungsbesitzer bei den jüngsten Altersgruppen bedarf einer weitergehenden Untersuchung. Zum einen sind diese Geburtenjahrgänge von den Rentenreformen der letzten Jahre am stärksten betroffen und sind daher künftig verstärkt auf eine zusätzliche Altersvorsorge angewiesen. Zum anderen mögen andere Einflussfaktoren wie die Absenkung des Garantiezinses oder Änderungen in der steuerlichen Behandlung von Lebensversicherungen eine Rolle spielen. Mehr Informationen Trends in German households' portfolio behavior - assessing the importance of age- and cohort-effects MEA Discussion Paper: 082-05, Mathias Sommer.

Who Lost the Most? Financial Literacy, Cognitive Abilities and the Financial Crisis The recent financial downturn and economic crisis provided a major challenge for financial institutions, politicians, and companies around the world. In this context, it is of major importance to analyse how private households were affected by the crisis, how they reacted to such a shock and what the long-term consequences will be. One central question is whether households with higher financial literacy were better at protecting themselves from the effects of the financial and economic crisis. This question, however, is not easily answered because more financially literate households are also more prone to own risky assets. Therefore, these households are more likely affected by financial losses as a consequence of the crisis. Hence, the aim of the analysis by Tabea Bucher-Koenen and Michael Ziegelmeyer is not only to examine who was hit by financial losses, but also to investigate how households reacted to the shock and whether they realized their losses for sure by selling assets which lost in value. The analysis reveals that little more than 20% of households in Germany report to have suffered financial losses due to the financial crisis 1 . On average, households report a loss of € 2,560 or 3.6% of their financial assets. A comparison of simulated losses, based on households' portfolio composition at the end of 2007 and average returns of these assets during 2008, illustrates that the self-reported measures are relatively close to the simulated ones. These findings give rise to the assumption that, on average, households have a reasonable understanding of the effects of the financial crisis on their financial assets. Ex ante, the relation between financial literacy and financial losses is not clear. On the one hand individuals with lower literacy and cognitive abilities are more prone to make mistakes; on the other hand they are more likely to stay out of risky asset markets. The analysis reveals that the second effect outweighs the first. Individuals with lower levels of financial knowledge and otherwise similar socio-demographic characteristics are less likely to have invested in the stock market and are therefore in general less likely to report losses in wealth due to the financial crisis. In contrast to their expectations, the two MEA researchers found that among households reporting losses those with higher financial literacy did not lose a smaller fraction of their assets. However, households with lower levels of financial literacy sold their assets which lost in value with a higher likelihood. This reaction of households with low financial literacy to short-term losses may have substantial long-term consequences for the distribution of wealth. It is very likely that these households did not participate in the profits during the recovery of the markets in 2009 and 2010. Additionally, these households will face lower rates of return in the long-term - especially if the negative experiences discourage them from future risky investments. According to estimations, the average annual consumption expenditures of stockholders are about 1.5 - 2% above those of households not participating at the stock market (cf. Cocco et al., 2005 and Mankiw and Zeldes, 1991). 1 In SAVE 2009 a special module of questions was added to the questionnaire, which investigates the impact of the financial and economic crisis on households. The analyses are based on the evaluation of answers of 2,012 households representative for Germany. Further Information: Who Lost the Most? Financial Literacy, Cognitive Abilities and the Financial Crisis MEA Discussion Paper 234-11 Bucher-Koenen, Tabea; Ziegelmeyer, Michael Foto: Sven Hoppe - Fotolia.com

The „Asset Meltdown“: A Popular Hypothesis Put into Perspective While correct, the arguments underlying the pessimistic scenario are incomplete. According to Börsch-Supan et al. mechanisms exist, which may attenuate or even reverse the negative effects of aging. On the one hand, the analysis by Börsch-Supan et al. verifies the main effects underlying the asset meltdown hypothesis: household savings will indeed decline when the baby boomers retire. On the other hand, the authors give several arguments against a massive decline of capital returns. The effects of aging on capital returns are noticeable, but they are within ranges that do not justify the dramatic catchword “Asset Meltdown”. As a consequence of demographic change, returns to capital will fall by roughly 0.8 percent points until 2035 if capital flows freely within the OECD. Based on the long-term average annual return on productive capital over the last 50 years, returns are lowered from 7.7 to 6.9 percent. Under the counterfactual scenario of a closed capital market, the decline of the rate of return in Germany would be about 0.4 percentage points higher. ? The pure demographic decline of capital returns will be aggravated when asset accumulation of retirement provisions covered by capital will rise because the increased demand leads to an additional downward pressure on the rate of return to capital. This effect will be in the order of magnitude of about 0.4 percent percentage points. ? However, the size of the additional effects of fundamental pension reforms on the rate of return also depends on labour supply reactions. As a fundamental pension reform decreases the level of distortionary taxation and since such a reform leads to incentives to work longer, age-specific labour supply shares will rise. As a consequence, capital productivity will increase – relative to a scenario without changing labour supply ? which dampens – or even reverses ? the projected additional decline of the rate of return.? An aging society needs more not less capital because it has to substitute labor for capital. The rising demand for real capital increases capital returns during exactly the time period for which pessimists expect the “Asset Meltdown” to occur.? The “Asset Meltdown” does not take place, because the demographically induced adjustment on capital markets will neither be a sudden nor an unexpected event. The retirement of the baby boomers will last for a period of about 20 years. Because of the fact that demographic developments are largely foreseen, capital markets will anticipate this development. Therefore the decline of returns will spread over 20 years and can hardly be perceived in single periods.? In a globalized world capital returns are not dominated by the demographic developments of individual countries. The international diversification helps especially the strongly aging countries (Germany, Italy and Japan) to reduce the magnitude of declines in rates of return to capital. Despite population aging, countries like Great Britain, France and the U.S. do not experience such a strong decrease of working age population ratios because of higher fertility rates. Capital productivity will therefore not shrink as much in these countries. For more information please have a look at: Börsch-Supan, Axel; Heiss, Florian; Ludwig, Alexander; Winter, Joachim (2003): Pension Reform, Capital Markets, and the Rate of Return, German Economic Review, Vol. 4, Issue 2, May 2003, 151-181.

Germany Learns to Save for Old Age The governmental subsidized old-age provision, the Riester-pension scheme, is gaining an enormous increase in popularity. Although not having found much approval by savers in the first few years when introduced, it has now emerged as saver’s absolute favourite amongst private old-age provision models. But despite the growing importance of the Riester pension very few detailed studies have examined these developments so far. Many questions remain to be answered: What sort of people take out Riester pensions and profit from state subsidies? Has the recent growth in the pension’s popularity also reached those segments of the population who have not had a Riester pension so far? Why do some people not open a Riester pension plan? Answers to these questions are provided by MEA’s evaluations of the SAVE data - a unique survey of savings behaviour and old age provision in Germany which MEA has regularly undertaken since 2001. The data confirms the dynamic development and reveals a general rising tendency of private pension provisions: 17per cent of all working households held a Riester policy by the end of 2005, almost thrice as many as at the end of 2002. In addition, German savers also entered increasingly into other private pension policies, thus causing a doubling in the section during this period. The Riester pension is especially popular with large families with at least four children. Growth was strongest in this section since 2002 and more than a fifth of households in this section had taken out a Riester policy by the end of 2005. Lower income groups however, which could also profit of a high level of state subsidy, are much less likely to have a Riester pension than people in middle and higher income brackets. Nevertheless, Riester pensions are already the most popular form of saving for old age among the people in lower income groups, thereby superseding occupational and other forms of private pension. Consequently it is satisfying to see that this trend has begun to reach people in the bottom 40 per cent income group, thereby adjusting the distribution to that of higher income brackets. Finally the SAVE data also reveals a crowding-out effect between old-age provision and other forms of saving. People for whom an important savings motive is to buy their own home are less likely to take out a Riester policy. And likewise does the desire to bequeath assets prevent Riester pension. As always in life, it is only possible to spend a euro once. Thus the conclusion for policymakers is to set decided priorities with the use of state support. For more information please have a look at Das Sparverhalten deutscher Haushalte: Erste Erfahrungen mit der Riester Rente. MEA Discussion Paper: 114-06, Axel Börsch-Supan, Anette Reil-Held, Daniel Schunk.

Births, Economic Growth and Population Aging The gradually accelerating demographic change is one of the key factors shaping the future development of our society. In an aging population a shrinking working age population is accompanied by a rising number of pensioners. In the future this means that the financial burden of supporting ever more pensioners will fall on ever fewer shoulders and will exercise increasing pressure on social security systems and on the economy as a whole. To date discussion has focused primarily on the consequences of aging, and on the financing and design of the public pension system. However, it would also make sense to investigate the causes of the aging phenomenon, focusing in particular on the continuing decrease of the fertility rate. The obvious question to ask is whether the aging problem can be solved by raising the fertility rate? Economic theory offers inconclusive guidance, even if the idea that if we have too many old people, we need more children to balance the effects out, appears plausible enough. The quantitative study discussed here also comes to more differentiated conclusions: A long-term boost in per capita gross national income will only result from a higher fertility rate if the additional children born are also better educated and trained. This means that the formation of human capital and not a higher fertility rate itself is decisive for long-term growth. The three most important economic policy conclusions consequently relate to the formation of human capital, the role of tax financed family transfers ("Familienlastenausgleich") and the priority of further reforms of our social security systems: An aging Germany needs better trained and educated - and consequently highly productive - children. Following international comparative studies - such as studies of educational standards like the "TIMMS Study" or the "PISA Study" in which Germany scored conspicuously poorly - there is a need for farreaching reforms in Germany's vocational and continuing professional train- ing sector. In a period of demographic change the engine of future growth - training and education in the context of the family, school, university, and continuing professional training - merits special attention and support. As the number of newborn children has very little influence on per capita GNI, there are no particularly obvious reasons on economic grounds for encouraging higher fertility rates. If higher fertility rates are desirable, the corresponding rationale will have to be obtained from other scientific disciplines. The study considered here does, however, generalize on the basis of the problems of current tax financed family transfers. It would be beyond the scope of this paper to discuss whether such transfers are adequate. However, this paper does support the conclusion that all that economists can really call for is compensation for the burdens borne by families which simultaneously represent benefits for others. The mere existence of more children does not in itself present a longterm solution to the demographic-driven problems of the future. A higher fertility rate does not represent an alternative to a reform of the social security system aimed at solving the immediate aging problem which, if no reform is forthcoming, results in a crisis in the public pension and health systems in the period between 2020 and 2040. This also applies if the impact of human capital is taken into account. The transition period after which a higher fertility rate would result in a larger and better trained labor force able to pay contributions to the pension and other social insurance systems is far too long. Further reforms of our social security systems must have top priority. Specifically, further pension reforms are needed which go well beyond the steps taken by the "Riester reform" and which tackle the problems which will arise after 2015. Furthermore, a reform of the health system, which even faces more pressing problems, is needed. Investments in human capital and reforms of social security systems are investments in the future which initially impose painful costs. However, it would be futile to hope for a painless cure to the problems associated with demographic change. The happy circumstance that we are living longer on average at the same time involves the need to finance this longevity. The financing burden must be mainly borne by the generation which will itself enjoy a longer life. The option of postponing urgent social reforms and shifting the burden to later (and possibly larger) generations will, as this paper has demonstrated, ultimately prove to be an economic nonstarter. More Information Berkel, Barbara; Börsch-Supan, Axel; Ludwig, Alexander; Winter, Joachim (2004): Sind die Probleme der Bevölkerungsalterung durch eine höhere Geburtenrate lösbar?, Perspektiven der Wirtschaftspolitik, Band 5, Heft 1, 71-90 Sind die Probleme der Bevölkerungsalterung durch eine höhere Geburtenrate lösbar? MEA Discussion Paper: 025-02 Barbara Berkel, Axel Börsch-Supan, Alexander Ludwig und Joachim Winter

Aging and Productivity: State of Research Munich, February 2017. There is a wide-spread assumption that older workers are less productive – which would result in a negative effect on overall productivity due to the increasing number of elderly employees. This assumption is contradicted by MEA studies concerning productivity and age. Studies in gerontology and sociology show, that experience and general education increase with age, while wit and the ability to combine decrease with age. But the individual aging process and therefore the performance depend also on a variety of additional factors, among them socio-economic status and social environment belong to them. For example people with higher income show better physical fitness. Occupational medics have also examined the relationship between aging and productivity. They found out, that physical and psychic capacity, especially the speed of perceiving signals, decrease with age. Nevertheless one finds clear differences among people in the same age group. These differences can be explained by varying working conditions and training of the human information processing system. Thus, performance and productivity cannot generally be defined through age. They depend definitely on individual working conditions. This causes the varying productivity among individuals during their life cycle. Ergonomists can therefore not prove a common deficit-hypothesis of aging, the assumption of the degradation of abilities over lifetime. Economic research of aging studies age dependent productivity on the firm level Even the aggregated perspective shows differing correlations of age and productivity. Studies on the wage structure even unravel a positive correlation between wages and age, which can be interpreted as the seniority specific productivity. Employees in public service for example are paid according to their age disregarding their actual productivity. In contrast, among sales workers, wages rise only to the age of 55 and decrease afterwards. Age and Productivity Out of a general analysis of detailed research fields, the researchers conclude, that productivity in a modern working society is more realised in collective cooperation than on the individual level. Especially invisible contributions of older employees to value added like experience and balance in conflicts are realised only in the group result, but are not measurable in individual measures of exposure and cognition. Because of that, one should not focus on individual measurements and instead consider the average age and the age structure of a work group to conclude to productivity. This was done in a MEA study by Axel Börsch-Supan and Matthias Weiss (2016), which combines data on errors occurring in the production process of a truck assembly plant with sociodemographic information about the workers. The results show that the average productivity of a worker continuously increases until the age of retirement (65 years), even in a work environment that requires physical fitness. This is – among other things – due to the experience of older workers. They do not make fewer mistakes than younger ones, but less serious mistakes. Studying the productivity of work teams shows that older workers can use their skills to handle difficult situations more effective and support their younger colleagues. For more information please have a look at: Börsch-Supan, Axel; Weiss, Matthias (2016): " Productivity and age: Evidence from work teams at the assembly line ". In: The Journal of the Economics of Ageing, online first (doi:10.1016/j.jeoa.2015.12.001) Trends in German households' portfolio behavior - assessing the importance of age- and cohort-effects MEA Discussion Paper: 073-05, Axel Börsch-Supan, Ismail Düzgün, Matthias Weiss

High Income Pensioners Live Longer The more people earn, the longer they are likely to live. The relationship between income and health has been established for some time internationally and has now been confirmed empirically for pensioners in Germany in a MEA study. The research by Hans-Martin von Gaudecker from MEA and Rembrandt Scholz from the Max Planck Institute for Demographic Research involved analysing a broad swathe of data on several million pensioners. Owing to the lack of suitable data comparatively few studies have been carried out on this subject in Germany. The authors draw on data produced by the Pension Insurance Research Data Center which has been providing researchers access to high-quality data records since 2004. The MEA study was, for example, able to include everyone who was receiving a pension in 2002. The final analysis produced extremely precise results based on data on 3.8 million pensioners. The research findings confirm what has been known internationally for several decades. The key results of the study, which focused exclusively on male pensioners, are presented in Figure 1. The average life expectancy of a pensioner aged 65 is 15 years (right hand column in Figure 1). However, the mortality picture looks quite different if the life expectancy of people on different levels of pension is calculated. The life expectancy of a pensioner receiving less than one thousand Euros is under 14 years. Above the 1,800 Euros level a 65-year old pensioner can expect to live almost another 19 years. The 'benchmark pensioner', who has earned an average wage over a 45-year working life, received a pension in 2002 worth around 1,100 Euros and, at age 65, had a remaining life expectancy of fourteen and a half years. It is important to emphasise that these results do not describe a causal relationship between income and mortality but merely show that they are correlated. It is still not clear whether the more affluent are able to buy better health, for example, or whether people suffering from ill health simply have lower earnings capacities. Another possibility is that income and mortality may be linked through a third factor. Higher education, for example, is known to lead to higher income. It is also established that well-educated people follow the therapeutic advice given by their doctors much more closely than other patients. As things stand, however, there is still a great deal of work to be done before more light can be shed on the relevant causal relationships and the influence of separate factors can be combined into a fuller picture. more information: Lifetime Earnings and Life Expectancy MEA Discussion Paper: 102-06 Hans-Martin von Gaudecker, Rembrandt D. Scholz

Labor Market Participation, Home Production and the Demand for Unskilled Labor In the course of increasing labor market participation, especially among women, and rising wealth, one observes an increase in outsourcing of home production. Cleaning ladies and house maids, but also restaurants, garages and kindergartens, take over work that has formerly been done by the people themselves. The process creates new jobs for unskilled workers in the service sector. Against this background, Melanie Lührmann and Matthias Weiss have examined the effects of prolonging the number of working hours per week or the impact of increasing labour market participation on the demand for unskilled labor and unemployment. The researchers developed and tested a general equilibrium model in which an increase in labor supply leads to a fall in unemployment. They considered three margins at which labor supply can change: the number of working hours per week, retirement age, and labor force participation. An increase in labor supply at either margin has two direct effects: People have higher incomes and less (leisure) time. Higher incomes lead to higher consumption expenditures and they argued that this increase is not proportional across different categories of consumer demand. As consumers have less time, they will disproportionately increase demand for goods and services that they have so far produced on their own. Examples for this outsourcing of homeproducible goods and services are house cleaning, food preparation, child care, repairs at home, ironing, lawn mowing, fixing bicycles, etc. Consumer demand thus shifts towards goods and services that everyone can in principle produce on their own. These goods and services are mostly produced/supplied by unskilled workers. Consumer demand shifts towards unskill-intensive goods and therefore, labor demand shifts towards unskilled labor. Given the concentration of unemployment at unskilled labor, an increase in the relative demand for unskilled labor has positive employment effects. The theoretical model has several testable implications: In times and countries, where labor supply is high, unemployment should be low. Households that supply more labor should spend less time on "home production". Households that supply more labor should spend more on goods and services that substitute for "home production". We test conclusion 1 for 23 OECD countries for the time period 1980 - 2003. Conclusions 2 and 3 are tested using Time Use data of 4000 German households. The results corroborate all three conclusions from the theoretical model. more information: Market Work, Home Production, Consumer Demand and Unemployment among the Unskilled MEA Discussion Paper: 101-06 Melanie Lührmann, Matthias Weiss