Economic effects of an aging population

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

1. Introduction

Aging population is one of the main challenges that the European Union will have to face in the year to come. The European Commission has examined the possibilities for Europeans to confront the demographic challenges. The European Commission stated four interactive demographic trends which effect the aging of the European Union.

The current fertility rate is low in all EU-27; with 1.5 children it is well below the replacement rate of 2.1 - required to stabilize the population size.

Nowadays, the baby boom generation constitutes the bulge in the size of population aged 45 to 65 years. This will lead to a substantial increase in the population of old people and retirees, who will need to be supported financially by a reduced working-age population.

Since 1960, life expectancy at birth has been rising by eight years and it is projected to continue to rise by at least a further five years by 2050. Europeans reaching the age of 65 in 2050 can expect to live four to five years longer then those reaching the same age today. This will lead to an increase in costs caused by old age population that is going to spend several years longer in retirement then previous generations.

Migration from third countries is going to increase in the next decades. In 2004, the EU received 1.8 million immigrants and projection shows that by 2050, around 40 million more people will emigrate. Migrants tend to bring down the average age of the population, but could only partially compensate the age distribution of the European population.

(European Commission 2006, p.3-4)

European population is aging rapidly and life expectation is growing. Working participation of population has not been changed along with changes in life expectancy. Trends have even been contrary to that: populations are getting retired earlier than former ones. The aging population doesn't create a problem as such, but problems are caused by the different sizes of various age groups. During the next 30 years, younger generations will have to take greater responsibility for the welfare of a larger number of people that is now slowly getting closer to retirement age. In most developed countries, the population is getting retired even earlier than they would according to the legal retirement age of 65 years. Working force participation in Europe, especially among older male workers, is declining as people exit their working life earlier than in previous decades. Therefore, working participation of aging workers is lower in European countries than, for example, in the United States and Japan.

Aging population and low work participation creates higher numbers of pensioners in comparison to working forces during the next decades. The European Commission estimates that old age dependency ratio will double by the year 2050. This creates great challenges regarding the sustainability of the European pension system. The commonly used pension system Pay-As-You-Go (when the working-age generation finances the current pensioners through social security payments) will cause a problem of financing retired population when over the next decades the working-age population is going to decline while the population aged 65+ will grow.

Pension system financing in the future can be solved by applying various strategies. Examples involve increasing social security payments, cutting pensions or taking a loan to cover cost of retirees. Yet these options will increase financial responsibilities and costs borne by younger generations or else leave future pensioners of worse than the ones today. A more pragmatic approach is to postpone legal retirement age, which will lead to lower old age dependency ratio and pension costs. This approach is logical as it will increase working-age population and at the same time decrease the amount of retirees.

To analyze retirement age postponing, it is necessary to know which factor are influencing the decision of timing of retirement. There are direct influences of retirement policies on workers' decisions to exit working life. In many countries the structure of the respective retirement system is advantageous for the employer. The main trend is that people retire as soon as they reached early retirement. In other words, workers benefit financially when starting their retirement time as soon as the early retirement age is reached. Benefits of extra working years do not seem to be high enough to compensate the benefits of early retirement. This so-called implicit tax is relatively high in all European countries. In some countries it is also possible to begin retirement before the actual legal retirement age through some additional early exit programs. Usually, this kind of option in the retirement system does not reduce the retirement benefits a lot. Yet does attract more people to start their retirement before the legal retirement age. In countries where legislation on early retirement possibilities is softer, increased legal retirement age might not necessarily increase the effective retirement age.

Furthermore, postponing retirement seems reasonable because it will reduce other expenses of the society as for example taxes and productivity. In some countries these expenses could reach more then 1/10 of their GDP. That is why in many countries retirement systems have been reformed to be less flexible to strategically impede early retirement. This was the aim of Finnish reforms of its retirement system in 2005 which aimed at increasing legal retirement age by 2 to 3 years. In that context, incentives had to be given to make early retirement less profitable than continuation of work. In this paper the core focus will be on analyzing the effect of postponing the average effective retirement age. Also the consequences of an aging population and the decrease of participation rate of older population will be a main focus. Finally, the Finnish reform of the respective pension system from 2005 will give the general considerations a concrete outlook.

1.1 Aim and structure of the paper

This paper examines the impact of old-age pension systems and other social transfer programs using Finnish retirement decisions as an example to have a closer look at. The aim is to examine countries' recent reforms on retirement systems and how effectively these changes will contribute to a stable future economy.

The main focus lies on the implicit tax rate connected to extending work. This comprises various dimensions of retirement incentives such as the pension accrual rate but also, to a smaller extent, the availability and generosity of benefits. On the basis of estimated coefficients, past changes in implicit tax rates and standard retirement ages are found to explain about a third (31%) of the trend decline in older males' labor force participation in OECD countries. Though this is more than in previous comparable studies, it remains plausible that other determinants, such as preferences for leisure or "demand-side" factors, may have also played a role in driving down participation rates (Duval 2004 p.1).

Duval introduces four main reasons why it is necessary to evaluate early retirement incentives and social transfer programs:

In well-functioning labor and product markets, supply-side factors such as productivity should be the major long-run determinants of labor force participation.

Reducing early retirement incentives would increase labor participation and ease labor demand constraints, for example reducing taxes by lengthening the tax paying period.

Other supply-side factors like living standards and/or demand for leisure, cannot account as a main reasons to low effective retirement ages. For instance, even the countries which are ranked as one of the highest in terms of GDP have significantly above-average participation rates for older workers. Also demand for leisure can not explain low participation rates, as countries which are economically and socio-culturally well integrated have still significant differences in effective retirement ages.

Adjustments of early retirement incentives is main policy target in most of the countries and is the main instrument available to rise effective retirement ages of labor force.

(Duval 2004 p. 3)

Potential sustainability of European pension systems has been undermined by two main trends. The first one concerns the population structure and, above all, the increasing share of older people. The second trend is the declining or low labor force participation in general and that of the elderly in particular. Both trends combine to a rising dependency ratio. That effect is due to retirees receiving pension benefits for a longer period and there being fewer workers per retiree to finance the pension system. Another indictor demonstrating the problematic participation behavior of the European elderly is the positive gap between the normal retirement age and the average exit age of labor forces (Piekkola 2004 p.1).

After defining some key termins, and general European trends shall be demonstrated quite in detail to lay the base for the final examination of the exemplary case of Finland and its pension-related situation after major reforms in 2005.

1.2 Definitions

Accrual = An increase in the pension benefit for an individual who has not yet retired

Accrual rate = The multiple that is used to calculate an increase in the pension benefit that is due to an additional year of work.

Actuarially neutral pension (Actuarial neutrality) = Which requires that the present value of accrued pension benefits for working an additional year is the same as in the year before (meaning that benefits increase only by the additional entitlement earned in that year).

Actuarially fair pension (Actuarial fairness) = Which requires that the present value of lifetime contributions equals the present value of lifetime benefits. Actuarial fairness relates to the entire lifetime of contributions and benefits

Defined benefit (DB) = The amount of the pension benefit is pre-determined, for example, by the years of employment. The worker bears no risk from the financial markets

Defined contribution (DC) = The amount of the pension benefit depends on how

much the worker has contributed to the system and possibly on the yieldy of the assets

Dependancy ratio = The ratio of non-workers to workers

Earnings related pension (Employment pension) = Pension system where the

amount of pension benefit depend on the career of the individual

Fully funded = Pension plan where future liabilities are collected in a full amount beforehand (invested often in the capital market)

Labor force participation rate = The percentage of working-age persons in an economy who are employed or unemployed but looking for a job

Old-age dependency ratio = Ratio of the population over 65 years of age to the

population of working age

Pay-As-You-Go = The financing system where current social contributions are used

to pay the current beneficiaries.

Replacement rate = Share of the replaced income when the labor market status changes, pension benefit over wage

2. Old-age dependency in industrial countries

Over the next decades one common development can be observed in all industrial countries while the working age generation is getting older. Future trends are not only alerting because of many people retiring, but also because more and more of them retire earlier than according to the official retirement age. Working participation of older generations has been falling in most OECD countries after the 1960's. This trend will result in declining participation- and employment rates as well as increasing old-age dependency ratios. Smaller working-age population will put pressure on the fiscal sustainability of social expenditures and living standards. By rising the effective retirement age and increasing the employment of older working population, we could smoothen the problem of an ageing population. This would balance spendings on pensions and also increase labor supply, which would additionally generate higher tax revenues.

The steep decline in the effective retirement age has been slightly reversed during the 1990's in few OECD countries. Effective retirement ages nowadays are still significantly lower then in the early 1970's and variation across the countries remain significant for the current level of participation rates of older workers. Participation rates of older male workers in 2001 ranged from a percentage of 40 (Belgium, Hungary and Luxembourg) to over 80 (Iceland, Japan and Mexico). Settg off at comparable levels in the late 1960's, the participation rates declined by 40 percentage points in Finland and the Netherlands, while in Japan, for instance, it remained broadly stable (Duval 2003, p.2).

2.1 Elderly population's participation in working life

In most developed economies, the participation of old-age generations, especially male workers, in working life has been declining during the last four decades. Working participation of 60 to 64 year old males, for instance, has dropped dramatically. Also aged female participation was declining in the past, but this development has been softened by the general increase of female work participation in the last decades (Casey et al. 2003 p.12). Even though work participation rates have been declining in all developed economies changes are most visible Europe. Figure 1 depicts the development of elderly male work participation during the last decades in some European countries, the United States, and Japan.

Figure 1 clearly demonstrates that the work participation of European elderly workers has declined more rapidly during the last decades than, for instance, in the United States. In France, the participation of elderly male workers aged 60 to 64 eas nearly 70 per cent in the 1970's but declined to dramatic 15,5 per cent by the year 2000. The United States shows similar development though smoothened: working participation among elderly males declined from 80 per cent in 1965 to roughly 50 per cent in 1995 (after which it starts to rise again). Most European countries have not been able to reach the 50 per cent participation rate for many decades.

Figure 1: Labor force participation of age 60-64 years male

Source: OECD 2010

Despite the similar trends regarding countries' working male participation rates, differences are still remarkable. Not only that the United States and Japan are showing better results of elderly male working participation, but significant differences can also be observed among European countries. The French elderly male participation rate from age 60 to 64 was tremendously low in 2005 with only 18,5 per cent. In Germany, the same rates were close to Finnish ones, namely 40%. Sweden, which has traditionally been showing good results in elderly male work participation, displayed even higher results then the United States.

Participation of elderly population to working life seems to decline even before average legal retirement age is reached which is set at approximately 65 years in Europe. Legal retirement age differs among European countries and ranges from 60 years in France to 67 in Island and Norway. In some countries, the legal retirement age is lower for females than for males, even though a recent trend is that male and female legal retirement ages are starting to converge. Legal retirement ages can even vary within one country according to between different sectors and professions. Usually, the legal retirement age is lower in the public-sector. Often, more demanding professions have a lower legal retirement age (Casey et al. 2003 p.1; Blöndal and Scarpetta 1999 p.15).

In all European countries the average elderly worker is retiring earlier than his respective legal retirement age. Table 1 features the average worker's age when exiting work compared to the standard and early retirement ages in European countries. In all countries, the average retirement age is lower then the standard retirement age and in some countries even lower then early retirement age.

Table 1: Average exit age of the labor force - weighted by the probability of withdrawal from the labor market

Source: OECD 2009; Eurostat 2010

2.2 Aging of population

Life expectation and birthrates are main aggregates in the aging process of a population. To make this more evident, the world's nations can be divided into two dimensions: life expectancy and birthrates. The results of that combination divide nations into four categories (Table 2). Europe belongs to the left upper quarter, life expectancy being high and birthrate low. From the perspective of financing a pension system this quarter is most critical, because of the growing gap between an aging population and younger generations. The United States belong to the right upper quarter, life expectancy also being high yet birthrates being higher than in the left quarter. This shows the difference between the United States and Europe, meaning that Europe will face bigger problems in the future as birthrates are much lower than the ones of older generations. The lower left quarter features both low life expectancy and birth rate. This trend can be observed in most states of the former Soviet Union, including Russia. The fourth quarter shows life expectancy and birthrate of the countries most developing, where life expectancy is low but birthrate is high.

Table 2: The four quadrates of demography

Source: Kotlikoff 2004, p.17

It follows that another important issue in addition to the participation in working life of the elderly is the aging population. In all European countries, younger generations will decrease in amount compared to increasing older generations during following decades. Population aging is mainly caused by the mentioned development of birth and life expectations. Birthrates in Europe are generally on a very low level while, on the other hand, life expectancy is growing. The European Commission (2006, p.3) filters four main reasons of population aging in Europe:

Low birthrates

Baby Boom (post-war generation)

Life expectancy growth

Amount of immigrants

Low birthrates are the main reason for an aging population. Birthrates are below the so-called replacement rate in all European countries. Replacement rates shows how many babies should be born to keep a population level stable. In 2005, the average birthrate was 1,5 children per female, while the replacement rate is slightly above two (European Commission 2006 p.3). Booming birth rates after the Second World War, the so-called Baby Boom, had significant impact on demographic development in many European countries. Because of this development, distribution in population structure figures is narrowing from the bottom (youth) and widening from the top (elderly) in many countries. Figure 2 represents forecast of Finnish population structure by the year 2030. It can be observed that the age groups from 50 to 59 years are particularly big. Members of this age group are going to retire within the next few years. Finnish Baby Boomers are born earlier than in most other European countries. That is why Finland is facing aging of population earlier then the rest of the Europe.Other European countries will follow Finnish demographic developments in their own time.

Figure 2: a) Population by age and gender 2010, projection 2009

b) Population by age and gender 2030, projection 2009

Source: Statistics Finland 2010

3. Consequencies of growing older generations

The increase of old-age dependency ratio can lead to higher public retirement costs and other pension related costs, for instance in the healthcare system. One of the main difficulties related to this is the finance system of future retirement. Most European countries make wide use of the Pay-As-You-Go (PAYG) financing system, in which current social contributions are used to pay current beneficiaries (Hakola 2002 p.34). This might evoke negative consequences, unless some changes will take place, because the aging population rises the number of retirees in relation to that of workers. Therefore, the viability of PAYG public pension systems is threatened in the long run (Lacomba and Francisco 2006, p.1).

Different approaches have been used to move from defined benefit PAYG systems towards contribution-based pension systems.

Some countries, including Hungary, Poland, Sweden, and Finland have shifted from a defined-benefit PAYG system to a mixed public-private system which includes a PAYG tier and a privately-managed fully-funded compulsory tier. In this system, mandatory contributions finance the two pillars (private and public) in different proportions depending on the country (Blöndal and Scarpetta 1999, p.25). In the following chapter, an overview of retirement systems will be given to deduce what steps countries should take to face the challenges of aging population.

3.1 Financing of pension system

There are two main pension system categories: PAYG-system and Fully-Funded system (FF). As mentioned before, PAYG-system financing is when current social contributions are used to pay the current retirees. In Fully-Funded systems each generation saves for its own pensions in pension funds.

The PAYG pension system can be described by a simple equation:

amount of workers in the labor force x average wages x social system contribution rate


amount of retirees x benefit per retiree

Because there is no saving or borrowing, the pension system is financed by collected social security charges of the population that currently works. The amount that can be collected depends on the level of salaries, number of individuals in labor force and the level of social security contribution.

Previous equation can be re-organized as follows:

social security contribution rate


old-age dependency ratio x replacement rate

The social contribution (tax rate) is dependent on the old-age dependency ratio and the replacement rate. The old-age dependency ratio comprises the amount of retirees per worker. The replacement rate is defined by the level of benefits per wage. Most pension systems are schemes with Defined Benefits (DB), where pension level for the insured is pre-specified and guaranteed by the pension provider, often the state itself. In this case the replacement rate is unchangeable also. It is then fixed by pre-specified rules that do not account for any changes in the circumstances. In equation 2 we can observe that if there is a shock to the old-age dependency ratio while applying Defined Benefits the social security contribution rate will also have to increase. Social security contribution rates are nowadays very high in most developed countries already, therefore growth in old-age dependency ratio can cause significant problems (Hakola 2002, p.3-4).

In PAYG pension financing systems the main problem is characterized by the future development of old-age dependency ratio rendering the pension system unsustainable. A number of economists are suggesting to move from PAYG systems to Fully-Funded systems, as the PAYG-system is less flexible and can lead to the old-age crisis (Cremer and Pestieau 2000 p.978). In FF systems each generation saves for its own pension funds, privately or through collective gathering. These funds can then be invested, often in the capital market to gain higher rate of return. FF systems have higher rate of return, that phenomenon even increasing, than PAYG-systems. Benefits of the Fully-Funded system particularly consist in rate of return of savings. Incomes from invested funds can be used to reduce social security contribution. Funds can be invested for many decades and, therefore, gain higher rates of return to lower social security contributions.

Shifting to a FF scheme might be difficult, in cases where the transition generation will have to contribute to the current retirees in addition to making their own savings for retirement. Cremer and Pestieau suggest that shifting the current system to a FF setting could be neutral if the transition generation is compensated for abandoning PAYG through the state taking loan to cover discrepancies. Such a shift would not effect disposable incomes and the wealth of individuals belonging to the transition generation. In this scenario, the government will run an increased deficit, but it will be outweighed by the increase in private savings of the new pension plan.

Relatively seen, the PAYG system is usually more beneficial for early generations as well. In most cases, early generations contributed less to pensions in comparison to the pension they finally receive, this system have been even more then actuarially fair for them. Younger generations might have to pay higher social contributions and receive smaller pension benefits in comparison to previous generations. (Disney and Johanson 2001 p.17-18). For instance, by the year 2030 British retirees will get only half of the benefits that retirees from the late 90's got. Nevertheless, younger generations will have to contribute as much as older generations. Börsch-Supra states that the PAYG system is due to its structure impossible to be turned into actuarially neutral which would allow different generations to have the same present value of contributions and benefits (Börsch-Supra 2000 p.26).

3.2 How to respond to pension system challanges

For countries still using the PAYG system as their only pension system, the increase in old-age dependency ratio will bring financing problems. The old-age dependency ratio problem could be solved by increasing social security payments, cutting pension benefits, taking more loans, or increasing the legal retirement age. Social contribution rates are already pretty high in most developed countries, therefore further increases could be very difficult. On the other hand, pension systems are developed to secure the population's welfare after a working career rendering the option of cutting pension benefits not very attractive either. Cremer and Pestieau argue that we should not increase the financial weight for younger generations more. Going along with that, increasing social secure contributions, taking additional loans and cutting retirement benefits should not be considered as options. The only option left would then be to increase retirement age to keep old-age dependency ratio on a sustainable level. (Cremer and Pestieau 2003 p.1)

Prolonging working time by increasing participation rates and average retirement age are important not only to avoid financing problems. The withdrawal of older workers from the labor force causes a reduced tax base, an increase in unused production capacity, and an increased burden on the pension and fiscal system. These costs are already now very high in developed countries. Table 3 shows results of the cost of early retirement in some developed countries as a share of potential GDP over the last two decades. In countries like Japan, Norway, Sweden, and the United States, costs seem to level off reasonably. In these countries, the participation of older workers in working life is higher than in other developed countries (see Figure 1). Continental and Eastern Europe tend to have higher costs than the rest of the OECD, in some countries even more than 15 per cent. Finland is one of the countries where cost of early retirement reaches more then 15 per cent.

Early retirement costs have been increasing since the 1980's and are predicted to increase even further. The projected rise in costs of early retirement over the course of the next decade is largely due to population aging. Rising costs over the past twenty years were primarily due to lower labor force participation of older workers. If elderly worker participation stayed in the same level, costs of early retirement would increase in the future even more due to the sizes of new retiree generations. To keep early retirement costs in nowadays' levels would require changes of average retirement ages (Herbertsson and Orszag 2003 p.1, 10-11).

Table 3: Cost of early retirement in the OECD as a share of potential GDP

Source: Herbertsson and Orszag 2003 p.11

Most developed countries have significantly reformed their pension systems over the past two decades. These reforms were mainly undergone for fiscal purposes, namely to improve the sustainability of public pension systems. Upgrading work incentives has also been a key aspect (OECD 2005 p.15,55). Aging population comes hand in hand with early exit from the labor force and increasing life expectancy which accelerates the growth of pension expenditures. Measures for reaching financial sustainability of pension systems include increasing actuarial neutrality of pension systems, pre-funding for future pensions, controlling the increase of contribution level, restricting early retirement, and making it more desirable to continue work. The Finnish pension reform from 2005 includes all these measures (Forma et al. 2006 p.7).

3.3 Why do workers retire earlier?

To avoid a crisis regarding retirement system finances and to keep all other social costs reasonably leveled, it would be necessary to increase the exit age from labor force in the next decades. First of all, it is important to analyze the reasons of labor force leaving working life early and the factors influencing this decision.

Since intensified early retirement has been witnessed in most member states of OECD, many analyses and extensive studies have been done on the consequences of early retirement systems. Most studies conclude that generosity of pension systems has an influence on early retirement (An, Chong-Bum 2009 p.7). That is why in many European countries various forms of pension reforms have been started, just as Finnish pension reforms were implemented in 2005. Nevertheless, most of the studies are mainly focusing on the influence of pension systems on retirement decisions. Few studies have been taking other factors into account, such as personal health and retirement timing of spouses. Also, quality of working life and motivational factors have been proved to be effective on retirement timing decisions (among others: Pelkonen 2005 & Hyrkkönen 2006).

Macroeconomic studies based on aggregate data proved the existence of retirement systems being effective. Feldstein (1977) observes the data of 12 countries' labor participation of age 65+ including their average lifetime wage per pension income, and concludes that higher retirement rates arise out of higher wage levels. Also Modigliani and Sterling (1983) prove that the social security system has a great influence on ones retirement decision. Hurd (1990) and Ruhm (1995) conclude that retirement rates increase rapidly around the age 62 and that this is influenced by social security wealth only, not by any other institutional or economical causes (Chong-Bum An 2009 p.7-9).

Some well established microeconomic studies [1] conclude that male working forces aged 55-61 have a lower implicit tax rate than the ones between 65-69 years of age. This means that older workers have larger tax burdens if they continue to work. Implicit taxes act as disincentive for workers who continue to work after the normal retirement age. On the basis of the option value model study of Stock and Wise (1990), the results of Gruber and Wise (2004) suggest that all countries where the eligible age is extended for three years, male workers aged 56-65 show higher participation rates (An, Chong-Bum 2009 p.7-9). The implicit tax shall be a topic for the following chapters.

In the majority of OECD countries, legal and - to a lesser extent - early retirement ages have remained constant since the late 1960's. In countries where retirement age cuts have occurred in the 1970's to the 1980's, such as Canada, Finland, Germany, Ireland, Netherlands, Norway, Spain, and Sweden, participation rates of workers above 65 years have been dropping constantly. Though in the 1990's increases of legal and early retirement ages in Finland, Italy, New Zealand, and Sweden have again increased participation rates in these countries (see Figure 1; Duval 2004 p.5). It would, therefore, be reasonable to assume that structure and rules connected to a pension system and its incentives towards early retirement effect work participation of elderly people. In the next chapters it shall be explained in what ways pension systems can influence retirement timing.

3.4 What causes early withdrawal from working life?

When it comes to evaluating or even reforming pension systems, it is necessary to consider potential factors influencing the decision of retirement timing. Those may include labor-market demand or individual situations. A study of Lee, for example, indicates that fluctuation in the demand for labor was an important caus of long-term unemployment of old workers in the late 19th and early 20th century in the United States (Lee 2003 p.14). Piekkola argues that early retirement has been very much a supply-driven phenomenon. But he also applies that development in retirement and labor force participation of the elderly is rather driven by the evolution of unemployment than by early retirement incentives (Piekkola and Deschreyvere 2004 p.6). Börsch-Supan discovered that the retirement age and the unemployment rate of Germany have a fairly low time-series correlation between 1960 and 1995, and that the German pension system provided strong incentives to retire early. That again led to old age labour supply being relatively robust (Börsch-Supan 2000 p.33-35). For the case of Finland, Piekkola and Böckerman explain that the low participation rate of older employees was not a phenomenon of the early 1990's economic crisis in Finland but has actually been continuing throughout the whole of the 1990's (Piekkola amd Böckermann 2001 p.6).

Disney explains that high participation rates in periods just before the legal retirement age proves that individuals decide on their retirement timing themselves (Disney 1996 p.202-3). He illustrates this decision making process with a simple static model (Figure 3) where the margin value product (MVP) represents benefit from wages and marginal utility of leisure (MUL) stands for benefitting from leisure. The period of chosen retirement age is identical with MVP and MUL variables intersecting to ensure maximum benefit.

Figure 3: Static model of the retirement age decided on, with MVP representing margin value product and MUL marginal utility of leisure.

Source: Disney 1996 p.203

In the previous figure it can be observed that retirement was settled on the age of 60. This proves that the individual decision plays a more important role in timing than actual demand. It is also important to notice that retirements usually do not take place when both variables are at their maximum.

Burniaux argues that the decision to participate in working life is of interest in its own right as increasing participation might be driven by raising the degree of social cohesion (Burniaux et al. 2003 p. 5-7). Differences in participation of specific groups like elderly workers may partially depend on non-economic (e.g. cultural or social) factors. On the other hand, divergences in participation rates among OECD countries suggest that policies play an important role, too. Figure 4 proves that there is a close and positive correlation between participation and employment rates, across