Economics: If it's good enough for everyone else, it's good enough for us

Labour has promised a minimum wage, but how high should it be and should it vary? Alan Manning weighs up the arguments

One of the few commitments made by the Labour Party in the general election campaign was the establishment of a Low Pay Commission charged with the responsibility of introducing a National Minimum Wage at a "sensible" level.

Although Labour was firm on the principal they were less precise on the detail. Now the election is won, decisions need to be made about what form the minimum wage should take. Two basic questions need to be answered: what would be a "sensible" level for the minimum wage, and should there be a single rate or a variety of different rates for different workers?

In setting the level of the minimum wage, the trade-off is a simple one. Set it too low and its effect will be so small as to be pointless, but set it too high and jobs will almost certainly be lost. While there is a broad agreement about this, there is considerable disagreement about the levels implied. The TUC Conference passed resolutions in favour of a general formula (half male median hourly earnings) and a more specific figure, namely pounds 4.26 per hour. But certain employers mutter about how something around pounds 3 per hour would be barely tolerable, and anything higher risks seriously damaging the economy. However, the CBI hinted recently that pounds 3.50 may be acceptable. But how should we decide on an appropriate minimum wage level?

A good place to begin is the level set by the old Wages Councils which, prior to their abolition in 1983, set minimum wages in a number of low- paying industries like retail, catering and clothing. Abolition of the Wages Councils was justified by the then government on the grounds that minimum wage levels destroyed jobs - an allegation also levelled at the proposed National Minimum Wage. Conspicuously absent from the Conservatives' arguments was any evidence that abolition of the Wages Councils had created a single job.

There was a good reason for this: abolition of the Wages Councils seems to have had absolutely no effect on employment; therefore, the levels of minimum wage set by the Wages Councils would be a good place to start in setting a minimum wage. At abolition these ranged from pounds 2.72 an hour in clothing to pounds 3.18 in retail, with an average of something like pounds 3 per hour. Uprating in line with average earnings growth would lead to rates today of pounds 3.10-pounds 3.60, with an average of pounds 3.40.

Setting a minimum wage much beyond this level pushes us into the unknown. It may be that we can have a substantially higher minimum wage without destroying jobs, but we have no previous experience on which to base such an opinion. There are probably more dangers in setting a minimum wage at too high a level and then trying to back-track than in setting it at a more moderate level and then raising it if the effects are very small.

A minimum wage of pounds 3.40 per hour might seem much lower than the demands of the unions, but the difference is not as big as it appears at first sight. The formula of half male average earnings may sound clear-cut but can lead to big differences according to the data set used and the precise method of calculation. The pounds 4.26 figure is virtually the highest among all the possible ways in which one might compute it (curiously it involves dividing weekly earnings including overtime by average hours excluding overtime), and others lead to much lower figures. For example, the Labour Force Survey (the largest representative sample we have) would have half male median hourly earnings as something like pounds 3.75.

The impact of the minimum wage is likely to depend as much on the variation in it as on the headline rate. The argument for variation is that the wage at which job losses will occur is different for different groups of workers. Having a single minimum wage risks it being far too high in some parts of the country and too low in others. The main argument against extensive variation in minimum wages is one of excessive complexity: prior to 1983, the old Wages Councils used to set so many different rates that it often required a book to summarise them.

The strongest case for variation in the minimum wage is by age. The gap in earnings between young and old workers is very substantial. Average hourly earnings in the Labour Force Survey are around pounds 7.80 an hour but teenagers make only pounds 3.10. A minimum wage of pounds 3.50 would directly affect as many as two-thirds of teenagers as opposed to about 10 per cent of the whole population.

The case for varying the minimum wage by region is much weaker as regional wage differentials are not that large in the UK. The graph shows average wages by region for workers who left school by the age of 16. Wages are noticeably higher in London and the South-East and lower in Northern Ireland, but otherwise there is little variation. So the only possible case for regional variation is a higher rate for London, but this is administratively difficult as one cannot draw a line around London inside of which wages are high and outside of which they are low. So, expect to see a single National Minimum Wage.

Trainees are another potential exception to the minimum wage. Although most low-wage earners do not receive any training, a minority do and it would make sense for trainees working for an accredited qualification to have a lower minimum.

One often hears claims from certain industries and types of businesses that they should have a lower minimum because they cannot afford to pay any more than they do at the moment. These appeals should be ignored. Low-wage sectors are the least productive sectors of the economy and to allow a lower minimum wage would be effectively to subsidise them (a bad idea) out of the pockets of their workers (an even worse idea).

At the moment, the UK is the only industrialised country without any form of minimum wages (except in agriculture). Although the Conservatives always tried to associate the minimum wage with the high unemployment European countries, it is an institution that is far more widespread than that. The United States with its dynamic labour force recently decided to raise its minimum by 20 per cent, a rise that was supported by Democrats and Republicans alike.

But one should not get too excited about the prospect of a minimum wage. Its impact on the whole economy will be small, at even at the highest levels being proposed it will have a smaller impact on the wage bill than did the Equal Pay Act in the 1970s. But to the minority of people who are affected it is very important in protecting them from the worst excesses of the labour market. If Tony Blair is looking for "practical measures in pursuit of a noble cause" a minimum wage could help to protect the most vulnerable in our society.

Alan Manning is a lecturer in economics at the Centre for Economic Performance, London School of Economics. He is also a Research Fellow in the Centre for Economic Policy Research's Human Resources programme. Further research conducted by Dr Manning on the minimum wage: ''Minimum Wages - The European Experience, Economic Policy No. 23'', published by Blackwells Publishers for CEPR. Further details on CEPR on 0171 878 2900.