Cross-country comparisons of pensioners' incomes
A hard copy of this report summary can be obtained by contacting Paul Noakes [E-Mail: Paul.Noakes@dwp.gsi.gov.uk] or by writing to him at the 'Social Research Division, Department for Work and Pensions, 4th Floor, Adelphi, 1-11 John Adam Street, London WC2N 6HT'.
Research Report No. 142
By Richard Disney & Edward Whitehouse
International comparisons of older people's incomes provide important information about how different pension systems contribute to the overall incomes of pensioners. Comparing the position of UK pensioners with their OECD country counterparts puts the operation of public and private pensions in the UK into a broader context.
The purpose of the research was to examine the position of the UK relative to other EU and OECD countries; to review the methods and approaches used in international comparisons; and, to update an earlier study commissioned by the Department which used data from the mid- to late 1980s. The study described below drew on some 12 comparative studies based on data up to the mid-1990s.
The report examined:
- How pensioners' incomes compare with those of the whole population and how they have changed over time
- measures and levels of income poverty, and how these vary between younger and older pensioners and over time
- the relationship between publicly and privately funded sources of pensioner income in different countries
- the position of widows
- the effect on income comparisons of including housing subsidies
- the effect of including other public spending programmes - in particular health and education - when assessing pensioners' incomes and the difficulties with this approach.
The key findings are:
- When comparing the incomes of UK pensioners with those of pensioners in other countries the relative position of UK pensioners is highly sensitive to:
- the definition of income
- whether the household or family is used as the measurement unit
- whether single pensioners and pensioner couples are treated separately
- the choice of equivalence scale
- what reference group is used (e.g. all pensioners compared with all working age people or younger pensioners compared with older workers).
- The most recent of the studies reviewed - from the OECD secretariat and based on data from the mid 1990s - expresses average pensioner incomes as a percentage of the average incomes of the population. This produced a replacement rate for the UK of 78 per cent, slightly below the mean of 83 per cent for the 15 countries in the study. Using the same data and comparing pensioners incomes with those of older workers produces a lower mean replacement rate for all 15 countries - 74 per cent. In this comparison the UK's replacement rate of 72 per cent fell closer to the mean.
- Examining absolute living standards of pensioners is problematic and needs to take account of purchasing power parities in different countries, adjusted with an equivalence scale. Without the equivalence scale pensioners' incomes in Italy, the Netherlands and Sweden are broadly similar to those of the UK. Using the equivalence scale adjustment alters the position and ranking of the UK significantly: for example Swedish pensioners become less well off, and Italian pensioners better off, than their UK counterparts.
- The choice of equivalence scale used is important as it makes assumptions about the sharing of resources within households and economies of scale that result. The average size of households with pensioners differs across countries, making comparisons difficult. In a large household, per-capita income may be low, but equivalised income per person could be high.
- Compared with the whole population, pensioners are under-represented in the bottom one and two income deciles and over-represented in the third to fifth deciles. Deciding where to set the `poverty line' (a level of income below which a household can be considered to be in poverty) has an important effect on poverty comparisons.
- The UK is generally in the middle of the distribution of pensioner poverty rates but there are indications that those rates are lower in more recent studies. An alternative measure of income poverty in cross-country comparisons is the proportion of poor who are elderly (as opposed to elderly who are poor). On this measure the UK is again at the middle to lower end of the ranking order compared with other countries.
- When looking at Income inequality the UK is in the middle of the OECD distribution. Countries with a `flat' element in public pension schemes (such as the UK's basic state pension) have greater equality of incomes (i.e. the flat, universally paid, component contributes to reducing income inequalities) when compared with earnings related public pensions such as in Germany. This finding holds even when private sources of income are included.
- In the late 1970s UK pensioner incomes increased at the same, or a slightly slower rate than the population average. However, between the mid 1980s and mid 1990s UK pensioner incomes increased more rapidly than the population as a whole, and more rapidly than the OECD 14 country average. However, this was mainly among younger pensioners (age 65-74).
- Pensioner poverty in the UK fell sharply in 1970s and early 1990s using an income measure. Using an expenditure measure implies greater poverty, a trend similar to that found in other countries.
- In all countries state benefits make up almost all the income of pensioners in the bottom fifth of the income distribution. Income sources in the upper fifths vary widely across countries.
- Measures of poverty don't take account of all wealth - e.g. appreciation of housing wealth and run-down of private pension wealth. Taking housing wealth into account could affect the ordering of countries - e.g. on this basis Australian pensioners have below average incomes but above average wealth
- Housing wealth is an important component of imputed income but many elderly households are reluctant to realise this asset, for example by trading down to a smaller home or by taking out a `reverse annuity mortgage'.
- Indirect taxes vary widely between OECD countries and taking them into account would also re-order the relative position of pensioner incomes. UK evidence suggests pensioners pay around 2 per cent less of their income in indirect taxes than working age people. Because there is no data that combines information on income and expenditure consistently in different countries the impact of indirect taxation is difficult to use for systematic comparisons.
- While widowhood is associated with a significant drop in income, using an equivalisation scale suggests that standards of living may not always fall. However, because of the relationship between mortality rates and income (in general, higher mortality is associated with lower income), widows are more likely to be poor.
- Benefits from the public services pensioners make use of cannot be measured in a straightforward way. For example, health services may be better in a country where taxation is higher but it is only the tax on the income, not the benefit from better healthcare that can be recorded.
- The research looked at two major public spending programmes - healthcare and education. Older people benefit more from the former, while working age people gain more from the latter. Taken together, the net benefit is slightly greater for older people than those of working age. Taking account of health and education puts the UK roughly in the centre of the distribution of comparison countries. The authors conclude that including this `in-kind' income would have little effect on the ranking of UK pensioners in terms of living standards.
Publication details:
Disney, R. and Whitehouse, E. (2001) “Cross-country comparisons of pensioners' incomes” (DSS Research Report No. 142), Leeds: CDS (£33.00) ISBN 1 84123 351 X.
Other relevant publications
Whiteford, P. and Kennedy, S. (1995)“ Incomes and living standards of older people: a comparative analysis ”(DSS Research Report no. 34), London: HMSO (£24.95).