Raising retirement age, increasing inequalities

67 is to become the “new 65”, we are told. Life expectancy is increasing, but it is not for all and not in the same proportions.

Last December, the OECD released the 2015 edition of Pensions at a Glance, a +350p report on pension reforms undertaken by OECD and G20 countries over the last two years. As with previous editions, the report provides with a comprehensive overview of the pension landscape.

The report confirms that raising statutory retirement age remains the much preferred route for governments when it comes to maintaining pension systems on a financially “sustainable” path. Population is ageing, life expectancy is increasing. “Retirement ages have risen substantially, with retirement at 67 becoming the new 65 in many countries. Several countries are planning to move towards 70, including the Czech Republic, Denmark, Ireland, Italy and the United Kingdom”.

Clearly, if retirement age is to increase and if “67 is becoming the new 65” as the OECD experts put it, then governments should make sure employment rates in older age follow suit. Otherwise workers wont contribute enough toward their pensions and the burden of old age dependency will simply be shifted from pension to unemployment and/or to disability insurance schemes. On that, the OECD press release accompanying the report offers a quiet optimistic view: employment rates of the 55–64 years have “increased sharply in many countries”, and from 45 to 66% in the case of Germany, 31 to 46% in Italy and from 52 to 57% on average across the OECD.

But a closer look at the report shows a different picture. For individuals of age 55 to 59, the OECD average employment rate is 67%. Fine. But it is falling at 44% for those aged 60–64 and at 20% for those aged 65–69. And these are averages. It is in fact below 30% for the 60–64 in one third of OECD countries, and below 10% for the 65–69 in two thirds of OECD countries.

Realising the obvious, the OECD report suggests that there is “significant room for improvement in the vast majority of countries” when it comes to employment rates of older workers and if indeed the new normal, 67 year, is to be met. The question is how governments are to help fill the gap. The report highlights the many measures to extend working lives: tightening of early retirement provisions, higher financial and tax incentives to work beyond the pensionable age, greater possibilities to combine work and pensions, etc. More fundamentally there is the hope that “changing the statutory retirement age serves as a signal on how individuals are expected to modify behaviours when planning for retirement, thereby influencing social norms”.

In the current economic context and with such low levels of employment rates for the +60 years, it might require more than a signal and more than influencing social norms, for older workers to find a job on the market — assuming for a first that they are fit for and willing to take on a job. At a Ministerial meeting on employment earlier in January, the OECD adopted a new Recommendation on Ageing and Employment Policies. The text offers a number of government commitments to improve labour market participation at an older age. In particular, the text recognises the central role that social partners (trade unions and employers) should play.

All this is welcome, but a fundamental concern remains about the sequencing and the order of priority of reform between (i) raising statutory retirement age and (ii) boosting old-age employment. Raising retirement age has a direct, tangible impact on workers and their households. It will happen, whatever the economic and labour market context. Incentives for old age employment are nice, assuming that older workers are in good shape and able to work, but their effectiveness is far more unpredictable and intangible — particularly in the post-crisis context.

Another concern with raising retirement age is with the implications in terms of income and social inequality. Pensions at a Glance 2015 addresses in part the impact of pension reforms on inequality (in chapters dealing with the erosion of first-tier social safety nets, and with the impact of incomplete careers). It is silent however on the impact of raising retirement age.

What matters here is life expectancy: for most pension systems (at least those that cover the longevity risk), the longer you live after retirement, the more you earn “pension wealth” (the accumulation of pension benefits during the retirement period). Yet it is widely assumed that life expectancies are different, if not very different, when looking not at the population as a whole, but by socio-economic groups: it is lower, if not substantially lower for low-income earners and low-skilled workers than for high-income earners and high-skilled workers.

Available data is limited however. On-going research by the OECD is trying to redress that. According to preliminary findings “the significant differences in life expectancies at age 65 for different socioeconomic groups are clear across all measures of socioeconomic status: educational attainment, relative income and occupational level. Furthermore, these disparities are increasing over time for most countries”.

That is particularly true regarding differences in education level. Looking at a sample of 18 OECD countries, the difference for men ranges from 1.5 year (Nordic countries) to 7 years (Czech Rep.) between the lowest and highest educated. Differences are less important for women, but they are there. And over the past decades, the higher educated (male or female) have enjoyed higher gains in life expectancy than the lower educated.

The OECD preliminary findings come with similar conclusions when looking at differences in income and in occupational levels (managerial positions versus employees).

In sum, raising retirement age across the board in a way that ignores life expectancy inequalities is likely to lead to an implicit transfer of pension wealth from the poor to the rich, from the less to the higher educated, from workers with routine functions to those with managerial responsibilities. Accordingly it would fuel, not reduce, inequalities — a good reminder as “inclusive growth” is supposed to become a broad policy priority for the OECD.