• Retirement-income regimes are diverse and often involve a number of different programmes. Classifying pension systems and different retirement-income schemes is consequently difficult. The taxonomy of pensions used here consists of two mandatory “tiers”: a redistributive part and a savings part. Voluntary provision, be it individual or employer-provided, makes up a third tier.

  • Retirement-income programmes designed to ensure adequacy of old-age incomes make up the first tier of the OECD’s taxonomy of pension systems, which was set out in the previous indicator of the architecture of national pension schemes. Safety-net retirement benefits are worth 21.6% of economy-wide mean earnings on average. Eleven countries provide a minimum pension above this safety-net level. For full-career workers, the average retirement income – including these contributory minimum pensions – is 24.4% of economy-wide average earnings. About a third of older people receive some support from basic, targeted or minimum pensions on average.

  • The second tier of the OECD’s taxonomy of retirement-income provision comprises income-replacement pensions. The summary here shows the key parameters and rules of these schemes that determine the value of entitlements, including the long-term effect of pension reforms that have already been legislated.

  • The rules for eligibility to retire and draw a pension are very complex, often reflecting conflicting government objectives. On the one hand, encouraging people to work longer as the population ages has been a major feature of many pension reforms. On the other hand, government have often been concerned to protect workers perceived as vulnerable and unable to continue their jobs to an older age.

  • The indicators of pension entitlements that follow here in Part II.2 and the analysis of pension “savings gaps” in Part II.6 use the OECD pension models. The methodology and assumptions are common to the analysis of all countries, allowing the design of pension systems to be compared directly. Future entitlements under today’s parameter and rules.

  • The gross replacement rate shows the level of pensions in retirement relative to earnings when working. For workers with average earnings, the gross replacement rate averages 57% in the 34 OECD countries. But there is significant cross-country variation. At the bottom of the range, Ireland, Japan, Mexico and the United Kingdom offer future replacement rates of less than 35% to people starting work today. Iceland and Greece, at the top of the range, offer replacement rates of more than 95%. Other countries with high projected replacement rates (between 70% and 90%) are Austria, Denmark, Hungary, Luxembourg, the Netherlands and Spain.

  • Private pensions play a large and growing role in providing for old age. This is illustrated with calculations of gross pension replacement rates that have been separated out between public and private sectors. The OECD average for replacement rates of an average earner from public schemes alone is 42%, compared with 57% with mandatory private pensions included. When voluntary private pensions, under typical rules, are added, the average replacement rate is 64% for an average earner.

  • The personal tax system plays an important role in old-age support. Pensioners often do not pay social security contributions. Personal income taxes are progressive and pension entitlements are usually lower than earnings before retirement, so the average tax rate on pension income is typically less than the tax rate on earned income. In addition, most income tax systems give preferential treatment either to pension incomes or to pensioners, by giving additional allowances or credits to older people.

  • For average earners, the net replacement rate across OECD averages 69%, which is 12 percentage points higher than the gross replacement rate. This reflects the higher taxes and contributions that people paid on their earnings when working than they pay on their pensions in retirement. Net replacement rates again vary across a large range, from under 40% in Mexico, Ireland and Japan to well over 100% in Greece for average earners. For low earners (with half of mean earnings), the average net replacement rate across OECD countries is 83%. For high earners (150% of mean earnings) the average net replacement rate is 63%, lower than for low earners. As with gross replacement rates, the differences with earnings reflect progressive features of pension systems, such as minimum benefits and ceilings on pensionable earnings.

  • The OECD average for net replacement rates of an average earner from public schemes alone is 50%, compared with 68% with mandatory private pensions included. When voluntary private pensions, under typical rules, are added, the average net replacement rate is 77% for an average earner.

  • Most of the indicators of pension entitlements in this report are based on analysis of a single person. In many countries, pension systems are effectively “individualised”: the position of a married couples is the same as that of two single people with the same level of earnings. In others, however, marriage has an effect on pension entitlements.

  • The financial and economic crisis of 2008 has meant that investment risk has been at the forefront of policy makers minds when thinking about pensions. Private pension funds in OECD countries lost 24% of their value on average, worth USD 5.4 trillion. However, it is important to bear in mind that private pensions are only a part of the overall retirement-income package: a major part of retirement income is generally not affected by investment risk. In some countries, means-tested pensions protect low-income workers from much investment risk and the tax system can also act as an “automatic stabiliser” of retirement incomes.

  • Pension wealth measures the total value of the lifetime flow of retirement incomes. For average earners, pension wealth is 9.6 times annual earnings on average in OECD countries. The figure is higher for women – 11.1 times individual earnings – because of their longer life expectancy.

  • Net pension wealth, like the equivalent indicator in gross terms, shows the present value of the lifetime flow of pension benefits. But it also takes account of taxes and contribution paid on retirement incomes. Both figures for pension wealth are expressed as a multiple of individual gross earnings. For average earners, net pension wealth for OECD countries averages 8.2 times gross individual earnings for men and 9.6 for women. Val

  • The progressivity index is designed to summarise the relationship between pension in retirement and earnings when working in a single number. The results show variation from 100 in pure basic schemes (such as Ireland and New Zealand), through zero in Hungary to a negative result in Sweden, indicating that the retirement-income system overall is regressive. The average index across OECD countries is 37. Regional differences are striking, with the index averaging 80 in the Anglophone countries: public pensions are strongly progressive. In southern European countries, by contrast, it averages just 8, indicating a very strong link between earnings and pension benefits.

  • In some countries, such as Hungary, Italy and the Slovak Republic, there is a very strong link between pension entitlements and pre-retirement earnings. In contrast, flat-rate benefits in Ireland and New Zealand mean that there is no link between pension and earnings.

  • The indicators so far have shown replacement rates, relative pension levels and pension wealth for people at different levels of earnings. By taking a weighted average of these indicators over the earnings range, the measures presented here show the average for the pension level at the time of retirement and pension wealth, the lifetime value of pension payments. The first of these is designed to show the level of the average retirement income, taking account of the different treatment of workers with different incomes. The average pension level is 55.3% of economy-wide average earnings across the OECD34 countries. The second aims to summarise the total cost of providing old-age incomes. Weighted average pension wealth is an average of 10.3 times annual economy-wide average earnings for men and 12.0 for women.

  • Incomes of older people are generally lower than those of the population, even when differences in household size are taken into account. On average in OECD countries, over 65s had incomes of 82% of the population as a whole in the mid-2000s. Older people’s incomes grew faster than the population’s between the mid-1980s and the mid-2000s in 13 out of the 25 countries where data is available. In most OECD countries, public transfers provide the bulk of income in old age.

  • On average, 13.5% of over 65s in OECD countries live in income poverty, defined as an income below half the national median. There is large variation between countries, from two with practically no old-age poverty to four with poverty rates double the OECD average. Poverty rates are higher for older people than for the population as whole, which averages 10.6%. A greater proportion of older women live in poverty than older men and old-age poverty rates increase with age.

  • On average, 13.5% of over 65s in OECD countries live in income poverty, defined as an income below half the national median. There is large variation between countries, from two with practically no old-age poverty to four with poverty rates double the OECD average. Poverty rates are higher for older people than for the population as whole, which averages 10.6%. A greater proportion of older women live in poverty than older men and old-age poverty rates increase with age.

  • Public spending on cash old-age pensions and survivors’ benefits in the OECD increased 15% faster than the growth in national income between 1990 and 2007, from an average of 6.1% of gross domestic product (GDP) to 7.0%. Public pensions are often the largest single item of government expenditure, accounting for 17% of total government spending on average.

  • Pension contribution rates have remained broadly stable since the mid-1990s. The average contribution rate in the 25 OECD countries that levy separate public contributions increased from 19.2% in 1994 to 19.6% in 2009, reaching a high of 20.0% in 2004. This probably reflects governments’ concerns over the effect on employment of high labour taxes. Indeed, these concerns seem to have taken precedence over the pressure on pension-system finances from aging populations and maturing of schemes. In the 23 countries for which data are available, revenues from these contributions were worth an average of 5.1% of national income, representing 14.2% of total government revenues raised from taxes and contributions.

  • Public spending on pensions has been on this rise in most OECD countries for the past two decades, as shown by the previous two indicators. Long-term projections show that pension spending is expected to go on growing in 25 out of 29 OECD countries where data are available. On average pension expenditure is forecast to grow from 8.4% of gross domestic product (GDP) in 2010 to 11.4% of GDP in 2050.

  • The total fertility rate is below the replacement level – the number of children needed to keep the total population constant – in 29 out of 34 OECD countries for 2005-10. The only exceptions are Israel and Mexico (with 2.8 and 2.2 children per woman, respectively) and Iceland, Turkey and the United States (at replacement level of 2.1). However in more than two-thirds of OECD countries there has been a moderate increase in fertility rates over the last decade. Fertility rates have a profound implication for pension systems because they, along with life expectancy, are the drivers of population ageing.

  • The remarkable increase in life expectancy is one of the greatest achievements of the last century. Lives continue to get longer, and this trend is predicted to continue. In 2005-10, life expectancy at birth averaged 76.1 years for men and 81.8 years for women. Among women, the figure was highest in Japan (86.2 years), followed by France, Switzerland, Italy and Spain. For men, life expectancy at birth was highest in Iceland (80.2 years) followed by Switzerland, Australia, Japan and Sweden.

  • Population ageing is one of the main driving forces behind the wave of pension reforms in recent years. The old-age support ratio is an important indicator of the pressures that demographics pose for pension systems. It measures how many people there are of working age (20-64) relative to the number of retirement age (65+). At the moment, there are just over four people of working age for every one of pension age on average. OECD countries have been ageing for some time: between 1950 and 1980, the average support ratio decreased from 7.2 to 5.1. However, the decline in the more recent period has been slower, with the fall from 5.1 to 4.1 taking 30 years. From 2010, population ageing is expected to accelerate. By 2025, the support ratio is projected to reach three and fall further to just over two in 2050.

  • “Average earnings” are an important metric underlying the presentation of system parameters and the results of pension modelling. The distribution of earnings is used to calculate composite indicators, such as the progressivity of pension systems, the structure of the retirement-income package and weighted averages.

  • Private pension arrangements have been growing in importance in recent years as pension reforms have reduced public pension entitlements. In 17 OECD countries, private pensions are mandatory or quasimandatory (that is, they achieve near-universal coverage of employees through collective bargaining agreements). In a further six OECD countries, voluntary private pensions (occupational and personal) cover a significant part of the work age population: more than 40%.

  • Private pension plans can be funded through various financing vehicles. In 2009, for OECD countries for which data were available, on average, 74% of OECD private pension assets were held by pension funds, 19% were held in pension insurance contracts run by life and pension insurance companies, 4% were held in retirement products provided by banks or investment management companies, and 3% were book reserves. Within pension funds, DC plans are playing an increasing role, even if DB plans still dominate pension fund assets in some countries, largely due to their historical prominence as the favoured arrangement for occupational (workplace) pensions in many countries.

  • There are 18 countries with a mandatory pension scheme giving a replacement rate below the average for the 34 OECD countries. This “pension gap” is over 28% of pay for an average earner in Ireland and for women in Mexico. It also exceeds 25% in the United Kingdom and for men in Mexico. Pension contributions required to fill the pension gap and bring the overall replacement rate up to the OECD average can be up to 7.5% of earnings if contributions are made for the full career. However, most workers do not start paying into a voluntary private pension until well into their careers. As a result, contribution rates of 10-15% would be required in six countries for workers with 20 years missing from their contribution records.

  • Substantial assets have been accumulated in most OECD countries to help meet future pension liabilities. Total pension funds’ assets were the equivalent to nearly 68% of gross domestic product (GDP) in 2009. Half of OECD countries have built up public pension reserves to help pay for state pensions. In these countries, public pension reserves were worth nearly 20% of GDP.

  • At the end of 2009, traditional asset classes (primarily bonds and equities) were still the most common kind of investment in pension fund and public pension reserve fund portfolios. Proportions of equities and bonds vary considerably across countries but there is, generally, a greater preference for bonds.

  • During 2009, pension funds experienced a positive real investment rate of return of 6.5% on average. Despite this recovery, by 31 December 2009 their asset values were still on average 9% below their December 2007 levels. In 2009, public pension reserve funds regained the ground lost during the 2008 crisis. By the end of 2009, the total amount of PPRF assets was on average 7.3% higher than at the end of 2008, and 13.9% higher than in December 2007.

  • Private pension systems efficiency, as measured by the total operating costs in relation to assets managed, varies considerably between countries, ranking from 0.1% of assets under management to 1.2% Fees charged to plan members to cover these costs vary considerably in structure and level across countries.

  • Funding ratios of exchange-listed companies’ defined-benefit plans were still significantly lower at the end of 2009 as compared to end 2007.