Simulating an Ageing Population: A Microsimulation Approach Applied to Sweden: Volume 285


Table of contents

(21 chapters)

Edited by

Page iii
Content available

This series consists of a number of hitherto unpublished studies, which are introduced by the editors in the belief that they represent fresh contributions to economic science.

Thus, this volume reports on our empirical studies, our model development, and the results of our scenario simulations. It further presents our experience from the micro-simulation approach and discusses the results of our simulations.

Content available

List of Figures

Pages xxi-xxvi
Content available

List of Tables

Pages xxvii-xxxiii
Content available

The challenge of an ageing population is a major concern to policymakers and researchers all over the world. As evident in Figure 1, the percentage of people aged 60 and above will increase substantially between 2000 and 2050 in all parts of the world. Europe has the highest proportion; only Japan has a similar age structure. The already high proportion of older people in Europe is expected to rise to an even higher level by 2050, from currently 19 percent to an estimated 34 percent.

Microsimulation is a technique that uses the capacity of modern computers to make microunits act and interact in such a way that it is possible to aggregate to the level of interest. A microsimulation model can be seen as a set of rules, which operates on a sample of microunits such as individuals, households, and firms. Each microunit is defined and characterized by a set of properties (variables) and as the model is simulated these properties are updated for each and every microunit. The model might simply be a set of deterministic rules such as the income tax rules of a country operating on a sample of taxpayers, and used to compute the distribution of after-tax income, the aggregate income tax revenue, or other fiscal entities of interest. But the model could also include behavioral assumptions usually formulated as stochastic models. Examples are fertility models, models for household formation and dissolution, labor supply, and mobility.

In 1997, SESIM was developed as a tool at the Swedish ministry of finance to evaluate the Swedish system to finance higher education. Part of that work was documented in Ericson and Hussénius (2000). We refer to this as version I of SESIM. Focus then shifted from education to pensions. SESIM was used to evaluate the financial sustainability of the new Swedish pension system. This new application implied that SESIM was developed into a general micro-simulation model (MSM) that can be used for a broad set of issues. We refer to this as the second version of SESIM and the documentation is presented in Flood (2003). The present version, SESIM III, maintains the focus on pensions but extends the analyses to include health issues, regional mobility, and wealth.

As summarized in our introductory Chapter 1, the trend toward ever-healthier elderly seems to have been broken (Figures 8 and 9). The share of young and middle-aged Swedish men and women, reporting very good or good health status to the Survey of Living Conditions, started to decline already in the 1980s. As a consequence, as the cohorts are graying, the share of elderly people, reporting very good or good health status, has also begun to decline. Increasing health problems among Swedish oldest old have also been reported from the SWEOLD (SWEdish panel of living conditions of oldest OLD) study (Parker et al., 2004). Similar trends have been reported for the United States and for the entire EU-15. Part of the explanation appears to be the growth at young ages in allergy, asthma, diabetes, other long-standing illness, and health problems associated with obesity. In the time perspective of our simulations, these trends in long-standing health problems might have less impact on the health of the elderly (and their demand for healthcare and old-age care or their life expectancy) than on the health of people in their middle ages but still be important. In this section, we will present some additional information on the development of health status during the last 20 years or so for the Swedish population.

There are many factors that may explain the number of spells and the number of days of absence from work reported as due to sickness. Health problems seem to be the most natural candidate to include among the explanatory factors, but individual health behavior could enter the scene in several ways. A day of reported sickness might primarily be due to the fact that a person's capacity to produce market goods and household commodities is so heavily reduced so the day is just spent at home with very little or no household commodities produced. It might also be a day when the person actively produces a restoration of his or her health, combining own time and healthcare of some kind. It might be a day when the person waits for a hospital treatment, for instance, a hip replacement, but his or her condition is an obstacle for taking part in market production (very much depending on the kind of job, in which the person would normally be involved).

In the 1990s, individuals aged 18–64 were eligible for disability insurance, if their work capacity was reduced by at least 25 percent (50 percent before 1993). In the beginning of the period, before 1991, disability insurance could also be granted for labor market reasons (i.e., if unemployed had been compensated long enough to exhaust their benefits – obtained benefits for 300 days). This possibility was gradually phased out after 1991. In 1995, the enforcement of the rules was tightened. When evaluating applications for disability pensions, local insurance offices now had to request a medical certificate and a work-related test of the applicant's degree of work capacity. Local offices also had to consult the applicant's employer, physician, or other qualified personnel, and even pay personal visits to the applicant. The possibilities for rehabilitating the applicant should also be investigated. From 1997, work incapacity should be evaluated in relation to all possible employment opportunities. Potential income changes resulting from changes in employment should not affect the evaluation4 (National Social Insurance Board, 2005).

Concerning migration on a national level, two phenomena emerge: people migrating from one region to another and people moving from the countryside to the cities. The geographical shift of the population between regions in a country is a slow process. In Sweden, only a few percent of the population migrate yearly. Nevertheless, migration has caused and still causes considerable redistribution of the population toward the metropolitan regions in Sweden. This section will emphasize general trends in population concentration through urbanization and migration in Sweden and compare these trends with changes in other countries.

Since SESIM is of a fundamental importance for this analysis, we also give a short presentation of the income-generating mechanisms in the model, focusing on earnings and income from capital.

It is not easy to get a long perspective on the distribution of wealth in Sweden because there is no single data source that gives a consistent view for a long period of time. The early estimates of the distribution of wealth were based on the concept of tax-assessed wealth which is the basis of the wealth tax. This definition has the disadvantage of not including assets that were not taxed, and no or very unreliable data were given for the majority of the tax payers who were below the taxation threshold. Furthermore, this variable was defined for individuals and for jointly taxed individuals, but no economically meaningful household concept was available. Register data have since then improved, in particular after the late 1990s when data became available directly from banks, brokers, and insurance companies without the filtering of the tax payers. The problem with the household definition remains, but in SESIM we have made corrections to get a useful definition (see Chapter 3). A relatively large survey (HEK) run by Statistics Sweden which combines survey information about the household with register data on assets estimates the median household wealth to 156000 SEK in 1999 and 197000 SEK in 2003.2 The latter estimate is in the 1999 price level.3 These estimates apply to all households independent of age. As will be shown below, the level of wealth depends very much on age.

The Swedish health care system is commonly characterized as a national health-service (or Beveridge) model (Freeman, 2000; Blank and Burau, 2004). It is certainly both financed by taxes and organized as a government responsibility, but it has developed over time as a decentralized rather than a national system (Lindgren, 1995). In Europe, only Finland seems to have a more decentralized system (Häkkinen, 2005). Most political decisions on health and health care in Sweden are made at the level of its presently 20 county councils and 290 local municipalities, which are empowered to put proportional income taxes on their citizens in order to finance their activities. Central government has a more passive role. Apart from supervising the fulfilment of the overall objectives of the health care legislation, which has a strong emphasis on equity,1 its influence is primarily manifested through indirect measures such as general and targeted subsidies. It can also impose ceilings on county council and municipality taxes.

In Sweden, responsibility for the public care of the frail elderly rests with three authorities acting at different levels. At national level, the Riksdag and the Government realize policy goals through legislation and financial control measures. At regional level, 18 county councils and two regions are responsible for the provision of health and medical care. At local level, Sweden's 290 municipalities have a statutory duty to meet the social service and housing needs of the elderly. Sweden's municipalities and county council have a high level of autonomy by international standards. Activities in caring services are ultimately controlled by politicians appointed to policy-making assemblies in municipalities and county councils through general elections. The decentralization of responsibility for elderly care makes it possible for local and regional conditions to be taken into account when policies for the elderly are formulated. The national authorities – the National Board of Health and Welfare and the 20 county/region administrative boards – are responsible for supervision, follow-up, and evaluation of municipal and county council caring services.

To allow for an improved health progression for the elderly, we adjusted the health index for those aged 40–90 proportionally to their age minus 40 and the calendar year minus 2000 in such a way that a 90-year-old person in 2040 will have the same health as an 80 years old in the base scenario. This implies that the improvement in health comes gradually and is largest for the oldest.

The predicted increase in the population share of elderly in Sweden is rather modest compared to some of the central and south European countries. The share of 65+ will, in our base scenario, increase from 17.5 to 23.9 percent in the period 2000–2040. Yet, this implies a major increase in the number of elderly. The number of 65+ will increase by 58 percent. The very old and care intensive group 80+ will increase even more, by 75 percent, and their share of the population will increase from 5.1 to 7.9 percent in 40 years. This is likely to put an increased pressure on the political system to match the expected increased demand for health care and social care by an increased supply. In some countries, the increase in the number of elderly will become balanced by a decrease in the number of children, and thus a natural reallocation of resources from children to elderly is possible. This is, according to our simulations, not the case in Sweden. The population share of those below 18 will stay rather stable between 21 and 22 percent.

Subject Index

Pages 421-429
Content available
Publication date
Book series
Contributions to Economic Analysis
Series copyright holder
Emerald Publishing Limited
Book series ISSN