Retiring later makes sense

Economics Department

Click to enlarge.

Ageing poses a serious challenge to OECD countries, in particular, how to pay for future public pension liabilities. And early retirement places an unsustainable burden on pension financing. There is no easy solution, but delaying retirement could help.

Retiring early has become so ingrained in OECD countries that these days it is almost an individual professional goal. It is even possible to drum up economic arguments in its favour: retirees spend money rather than save it, for instance, and are valuable for such sectors as tourism. Retiring early, some argue, frees up jobs for younger recruits and helps to boost productivity. To many, retirement is a hard-earned right, and so the earlier the better. Indeed, futurists have been telling us for years that early retirement would become the norm, that thanks to labour-saving technology we would soon be spending more of our lives at play and less at work.

All very well, but the reality is quite different. Early retirement may seem like a worthy individual goal, but it is a socially expensive one, and, as far as public pensions are concerned, quite unsustainable. The essential reason is that more people are retiring early and living longer. That means more retirees depending on the funding of those in work for their income. The outlook is worrying. In the next 50 years, low fertility rates and rising life expectancy in OECD countries will cause this old-age dependency rate to roughly double in size. Public pension payments, which pay 30-80% of total retirement incomes in OECD countries, are expected to rise, on average, by over three percentage points in GDP and by as much as eight percentage points in some countries. Such is the pressure on pension funds that there is a danger of today’s workers not getting the pensions they expected or felt they paid for.

Action is needed, but simply aiming to reduce the generosity (and cost) of public pensions, or trying to augment the role of privately funded pensions within the system, though necessary steps, may be insufficient to deal with the dependency challenge. After years of advancing early retirement schemes to avoid redundancies and higher unemployment, many governments are now looking at persuading people to stay in work until they are older. Surely, the thinking goes, if we are healthier now and jobs are physically less strenuous and unemployment is down, then perhaps the present rate should rise anew. In fact, increasing the participation rate of persons aged 55 to 64 years is one of the main objectives of social policy within the European Union under the Lisbon and Amsterdam Treaties.

The approach makes economic sense. For a start, as long as the extra labour resources from delayed retirement are put to work, then in theory the level of GDP will rise, thereby increasing the resources available for consumption. This is simplistic of course: having more old people at work is not enough to improve productivity. Indeed, some contend that the level of GDP could fall, since retiring early acts as an incentive to work hard and save more and so boosts productivity, whereas delaying it could dampen the morale and productivity of would-be retirees. However, these negative effects appear to be small, so on the whole retiring later would increase GDP in the longer term. Working people certainly pay more income taxes and social security contributions than retired people, so a later effective retirement age would generate more funds to pay for pensions. Likewise, there would be less pressure on those funds as delayed retirement means people start drawing their pensions later. Working longer also helps people to stay out of poverty, which is particularly important where pensions risk falling to low levels.

Still, people have been retiring at younger and younger ages for decades. They have been enticed by generously high net pensions and other benefits, as well as the lure of more travel and leisure; but they have also been pushed by employers anxious to cut costs (it is often easier to shed older staff, particularly in hard times) or boost productivity by replacing them with younger recruits. Apart from the cost savings of later retirement, there is an equity issue at play here: while early retirement is an option for many workers, labour markets tend to prevent late retirement. Healthy people may wish to work longer, rather than leave work or accept low pensions. Not everyone wishes to be “retired” early. Equally, companies may wish to retain older employees for their experience, but cannot, often due to inflexible labour market rules that, in the end, hurt workers. Also, labour taxes may be too high for employers, or part-time jobs inaccessible to older workers. If retirement is delayed, these market conditions would have to be improved. Otherwise, as some economists argue, later retirement could simply fuel older unemployment.

But this does not alter the key message: early retirement places an unsustainable burden on pension financing. Governments must reconsider their policies with respect to retirement age. Comparing effective retirement ages and labour participation of older workers across countries may be instructive.

Different old folks 

The standard official retirement age to qualify for a public pension in most OECD countries is currently 65. The chief exceptions to this are France and Korea, where it is 60, and Norway, where it is 67. Several countries offer early retirement pensions allowing people to retire two to five years before the standard age. And in a number of countries, there are relatively generous eligibility criteria for disability pensions and unemployment benefits for older workers. Severance packages, including early occupational pensions, also help some older workers to make the jump at relatively low personal cost.

It is therefore not surprising that the average age of actual retirement is often three to five years earlier than the standard official age. Only in the United States does the average actual retirement age correspond to the current standard age (65). Even so, the United States is gradually raising the retirement age to 67, and is debating the merits of later retirement. The average worker in Japan and Korea retires at 69 and 67, respectively four and seven years later than the official standard age. But these are the exceptions. In Europe, less than half of the male population aged 55 to 64 is currently working (see table).

Life expectancy at the average effective retirement age can be as high as 18-20 years, about a third longer than it was 30 years ago (see graph, “Living longer). It is projected to increase further, so the retirement period will lengthen unless retirement itself is delayed.

Delaying retirement looks like the only option. Increases in the standard retirement age of women to match that of men are being phased in in Australia and Germany, and for both men and women in Hungary, Italy, Japan, Korea and the United States. Pension systems are also being adjusted so that if people retire earlier their pension level will be reduced accordingly. This seems fair–experts refer to it as being actuarially neutral–since it reduces pressure on pension funds by ensuring that benefits are more in line with contribution payments. Australia, Finland, Germany, Iceland, Italy, Netherlands, Norway, Sweden and the United States are all moving in this direction.

Other ways of discouraging early retirement include reducing pension levels directly, as Germany is doing, or prolonging the contribution period needed to qualify for a full pension, the approach France and Hungary are adopting. Some countries are tightening access to disability pensions and unemployment benefits (Finland, Germany, Netherlands, United Kingdom). The risk of early retirement policies leading to unemployment has to be minimised and countries are also acting to improve employment opportunities for older workers by outlawing age discrimination (Australia, Netherlands, United Kingdom), or providing wage subsidies for older workers (France, Germany, Korea). But while these measures go in the right direction, significant incentives for early retirement still remain.

Reducing incentives for early retirement 

Raising the effective retirement age cannot, of course, be achieved unless early retirement incentives are reduced. The OECD has calculated two measures of early retirement incentives in public pension schemes. The first is the replacement rate–a person’s pension as a percentage of his or her working income prior to retirement; the higher the replacement rate, the higher the incentive to retire. The second measure is the change in net pension wealth from working an additional year; the principle here is that the incentive to retire early would rise if working an extra year implied paying additional contributions with little or no increase in future pension gains. Using this measure for 15 countries, it is evident that there are incentives to retire early in the regular old-age pension system, though not before the age of 60. In fact, early retirement is generally not permitted before this age. The only exceptions are Italy (where the earliest retirement age is 57 and the replacement rate of pension income is above 50% of previous earnings) and Australia (where individuals can draw on their mandatory savings from 55).

But beyond regular pension systems, incentives do exist. In a number of countries, like Germany, Netherlands, Finland, Norway and France, disability pensions and unemployment benefits can be used as de facto early retirement benefits. There are also incentives for “normal” workers to retire after 60 but before 65 when pensions are offered with relatively high replacement rates. Sometimes, as in the UK and Canada, complementary occupational pension schemes also provide strong incentives for early retirement–the retirement age in some UK companies that have their own private pension schemes, is 60, and not the UK standard of 65.

Clearly, any attempt to push up retirement ages must be bolstered by policies aiming to increase both the supply and demand of older workers. Eliminating incentives for early retirement would help to solve the supply problem, as fewer workers will retire early. But it is obviously not enough that labour supply increases; demand should also be there. In practice, this should not be a major concern, since countries with high participation rates also tend to have high employment rates. Nonetheless, policy measures could help demand to meet supply. Making labour markets more flexible by allowing wages to better match labour productivity would increase demand for older workers. Strict employment protection also reduces the chances of older workers to find work. Also, retraining older workers would help supply and demand, and it would also be more cost-effective if workers were kept employed for more years.

Ageing will reduce the relative labour supply significantly over the coming decades, and governments should not reduce it further by providing incentives for early withdrawal from the labour market and penalising those who continue to work. Eliminating such distortions would encourage people to work longer and retire later so that effective retirement ages automatically adjust with rising life expectancy. People have a right to retire decently and confidently. Right now, the pressure on pension funds is such that, without action, this right is being eroded.


Dang et al (2001), “Fiscal Implications of Ageing: Projections of Age-related Spending”, OECD Economics Department Working Paper No. 305.

OECD (forthcoming), “Policies for an Ageing Society: Recent Measures and Areas for Further Reform”, Economics Department Working Paper.

©OECD Observer No 234, Ocober 2002

Economic data

GDP growth: -9.8% Q2/Q1 2020 2020
Consumer price inflation: 1.3% Sep 2020 annual
Trade (G20): -17.7% exp, -16.7% imp, Q2/Q1 2020
Unemployment: 7.3% Sep 2020
Last update: 10 Nov 2020

OECD Observer Newsletter

Stay up-to-date with the latest news from the OECD by signing up for our e-newsletter :

Twitter feed

Digital Editions

Don't miss

Most Popular Articles

NOTE: All signed articles in the OECD Observer express the opinions of the authors
and do not necessarily represent the official views of OECD member countries.

All rights reserved. OECD 2020