Labouring under delusions about cost of old age

The rise in German unemployment in January was another shocking illustration of how badly Europe's labour markets perform. The immediate social cost of such wasteful levels of unemployment is high and obvious. But the full economic costs are still not fully appreciated. High unemployment contributes to the European pensions crisis - the two problems are largely one and the same.

If European unemployment can be brought down, and employment boosted, the economic burden imposed by the need to provide adequate old-age pensions will automatically become much lighter. Indeed, if continental unemployment and labour force participation rates move closer to British and US levels, old-age pensions might even be funded, in an economic sense, on a pay- as-you-go basis indefinitely. Europe has a labour market problem, not a pension problem. This view is markedly at odds with received wisdom. A sharp rise in the number of European pensioners, relative to the number of adults of working age, is a demographic inevitability. And most continental pension provision is funded and paid from ongoing tax receipts.

As a result, it is taken for granted that Europe will find it difficult to support its pensioner population because the tax base will shrink. The report of the UK Social Security Select Committee in October on unfunded pension liabilities in the EU provides a recent illustration of consensus thinking on the subject. However, the crude demographic arithmetic is misleading. The age profile of the population is only one of the many factors influencing the dependency burden. Most importantly, adults in work support not just the elderly, but also non-working adults - those who have chosen not to participate in the workforce.

The extent of adult dependency varies considerably across countries, and through time. In the US and the UK, participation rates are high, and unemployment is low, partly reflecting the flexibility of labour supply and demand. And in both countries, participation rates have risen in the last 20 years as household habits have changed. In Europe, participation rates are relatively low, and unemployment is high. As a result, labour as a factor of production is under-utilised. But relatively small changes in the employed portion of the population of working age can have a big effect on the dependency arithmetic.

The point is easily illustrated. Over the next quarter-century, the proportion of the French population aged 65 and above will rise from roughly 15 to 21 per cent, while the proportion of working age will fall slightly, from 65 to 63 per cent: as a result, the ratio of pensioners to potential workers will rise from 23 to 33 per cent, a proportionate increase of more than two-fifths. This is the conventional arithmetic: it suggests a sharp, potentially worrying rise in dependency.

However, only 88 per cent of the French workforce is currently employed; and with a participation rate of roughly 67 per cent this in turn probably represents just 59 per cent of the population of working age, or just 38 per cent of the total population. Meanwhile, if we add the non-working portion of the non-retired adult population to those who are retired, the proportion of adult "dependents" rises from 15 per cent of the total population to 42 per cent.

When the denominator is adjusted downwards accordingly, and the numerator upwards, the current dependency ratio rises sharply, to more than 100 per cent. Thus in France there are already more adult dependents than there are workers. In itself, this simple adjustment makes the problem look dramatically different. If the pattern of participation and employment remains the same, but the population's age structure evolves as expected over the next quarter-century, the resultant rise in the ratio of pensioners to potential workers turns out to be one-fifth, not two-fifths, a much smaller increase. Non-working adults age alongside the workers.

But the arithmetic really becomes interesting if we suppose that in the quarter-century ahead, France is able, via a combination of more flexible working practices and changes in household preferences, to approach the sort of unemployment and participation rates seen in the UK and the US. Then, instead of rising, adult dependency might actually fall by the year 2020, perhaps by as much as one-fifth.

This is a sensational possibility. Of course, it takes no account of details such as the extent of part-time employment and low pay, but it illustrates the potential importance of changes in labour market practice. Other continental economies are in a similar position. In principle, today's pay-as-you-go intra-family transfers (housekeeping), unemployment benefits and student grants could provide tomorrow's pay-as-you-go-pensions - if European labour markets reform.

This also understates the potential good news. Economies that employ a bigger proportion of their population will be more productive, and the size of the economic cake available for redistribution could be much bigger than is currently predicted on the basis of past growth trends. Thus at one end of the spectrum is the current spectre of longer working lives and lowered pension entitlements; at the other lies the tantalising prospect of a longer and wealthier retirement.

A less wasteful usage of European labour need not prevent the long-awaited shift from public to private-sector pension schemes on the Continent. In an increasingly unified, flexible labour market in which workers are able and willing to change jobs and location more often than in the past, private savings schemes may become more attractive in their own right. Meanwhile, the growing interest in equities in the European financial markets is likely to give a further boost to the process. But a more productive Europe will be better able to meet the claims on future output which these schemes represent than a Europe that continues to squander its valuable human resources.

The arithmetical illustration above, while extreme, is not beyond the realms of possibility: the UK unemployment rate has moved in a 9-point range in the last quarter-century, the participation rate in a 5-point range. The political pressure to do something about European unemployment is rising. Is it not inconsistent to be worrying about both a prospective glut of labour (unemployment) and a shortage (too many pensioners) at the same time?

Kevin Gardiner is a senior economist at Morgan Stanley International: the views expressed are his own.

The elderly are not the only dependants

% population, 1995 France UK

Below working age 20 19

In work 38 44

Unemployed 5 4

Non-participating 21 16

Retired 15 16

Ratio of non-working

to working population 1.6x 1.3x