Wanted: a warts-and-all tally of UK's jobless

New Year 1997, the last Hogmanay before the hustings. We all know politicians are lousy at keeping promises but they can at least make resolutions. And what better than to resolve to reform Britain's approach to measuring unemployment. To some this may seem a rather arcane suggestion for the festive season, the kind of thing that provides a turn-on for none save a few sad anorak-wearers.

However, when one reflects on the extent to which key macro-economic and employment policy decisions are based on job statistics, the importance of a clear picture of unemployment becomes self-evident. Yet, at present, the picture is anything but clear.

The Government may applaud the downward path in unemployment, which has taken the monthly count of people claiming Jobseeker's Allowance below the politically sensitive 2 million mark. But the claimant count is widely mistrusted.

In 1995,the Royal Statistical Society called for a new monthly count to be derived from the 60,000 household based Labour Force Survey (LFS) which provides a measure of unemployment based on an internationally agreed definition of what constitutes an unemployed jobseeker. But the LFS is conducted on a quarterly basis and the Treasury has ruled out the extra pounds 8m a year needed to upgrade it (the Chancellor's fiscal rectitude on this matter being backed up by ministers at the Department for Education and Employment who, with the general election on their minds, are concerned by the fact that the level of unemployment as measured by the LFS is around 200,000 higher than the claimant count).

However, even if the Treasury were to relent, it is not clear whether the LFS measure alone is a totally reliable indicator of the extent of slack in the jobs market. For example, the standard LFS measure excludes people on the margins of the workforce who tell the LFS that they want jobs and might enter the market in good times. Moreover, the standard LFS measure may well understate the social distress caused by unemployment. Although derived from a household survey, it measures unemployed individuals, whereas the social distress caused by unemployment may well depend upon the distribution of joblessness across households.

What is required, therefore, is a range or spectrum of unemployment measures that, when considered together, can provide a truly rounded account of Britain's unemployment problem.

The US Bureau of Labor Statistics adopted such an approach in 1976 and, following a modification made in 1995, publishes six measures of unemployment based on its monthly equivalent of the LFS.

Now that the need for a range of inflation measures has been accepted in Britain - with the Treasury publishing RPI, RPIX and RPIY - why not also adopt the US-style spectrum approach to measuring unemployment?

This was precisely the question posed earlier this year by the House of Commons Select Committee on Employment, which called for a new approach to compiling the jobless figures. But the Government rejected this call, merely responding that it would be "keeping under review the need for more informative reporting of unemployment measures".

Faced with this official reticence, the Employment Policy Institute (EPI) has decided to construct its own range of unemployment measures in conjunction with the Centre for Economic Performance at the London School of Economics. These are published quarterly as part of a regular and wide-ranging Employment Audit of government jobs figures.

The EPI joblessness indicators - U1 to U5, available for the summer 1996 quarter - are shown in the chart. Although US thinking has influenced the EPI's approach the EPI indicators have been designed with the British labour market in mind. The first three indicators focus on labour market slack. U1 is the standard measure of unemployment provided by the LFS and covers all economically active jobseekers. U2 includes in addition so-called "discouraged workers", ie people who do not seek jobs because they feel there are no jobs available.

U3 adds to the numbers in U1 and U2 those whose response to the LFS is merely that they want a job. This indicator includes all the people marginally attached to the labour market who might enter the market in buoyant times. It could be said to approximate to the number of jobs needed to create "full employment" in Britain.

The last three indicators are more akin to measures of social distress. U4 refers to the duration of job search and measures people who have been unemployed for more than six months. U5 moves away from an individually- based joblessness indicator to a household-based one.

U5a measures individuals living in households where no adult has a job, while U5b measures households where no adult has a job. The indicator excludes households where the head of household is beyond retirement age and those containing only students.

What do these indicators tell us about current levels of joblessness? Well, for a start, U3 shows that the number of jobless people who say they want a job is not the 1.9 million registered by the claimant count, but well over 4 million. Of course, caveats abound. There are undoubtedly people measured by U3 whose desire for work is not matched by any form of realism about the type of work they are likely to get. But U3 none the less points to a considerable "job shortfall" in Britain.

U1-U4 do broadly reflect the trend fall in unemployment registered by the claimant count. However, with regard to the "social distress" indicator U5, there has been no sign of any corresponding reduction in the number of workless households. These account for almost one in five of all British households - up from well below one in ten 20 years ago. Between the summer of 1994 and the summer of 1995 - the latest quarter for which U5 can be constructed - the number of such households increased by 250,000.

The EPI has an open mind on whether U1 to U5 are the most appropriate indicators and how they might be developed. Their purpose is as much to stimulate debate on a new approach as it is to offer the final word on measuring unemployment.

What is clear from these indicators, however, is that large numbers of jobless people, particularly those living in workless households, cannot find work even after four years of recovery in the labour market.

There is an urgent need for further policies to help them; and good policy- making requires a complete picture of the underlying jobs scene. In the short term, of course, it will be politically expedient for the present government to focus on the "good news" from the claimant count.

And there is as yet no guarantee that a Labour government would necessarily want to provide the British people with a warts-and-all account of the jobs market.

However, sooner or later the penny will drop and politicians will realise that what the electorate wants is openness, not illusion. For now, let's raise a toast and wish a happy New Year and good job-hunting to Britain's 1.9 million unemployed. Or should that be 4 million?

The writer is director of the Employment Policy Institute, an independent think-tank. The EPI Employment Audit is available as a quarterly subscription publication, priced p.a. from Employment Policy Institute, Southbank House, Black Prince Road, London SE1 7SJ.