In the recession of 1979-81, unemployment was more concentrated in the Midlands and the North. This time, every professional family in the South-east knows someone with a tale of woe. People are much less inclined to blame the unemployed for their own fate.
Unemployment is overwhelmingly involuntary. People do not choose redundancy. The idea that high benefit levels have encouraged the work-shy is even more absurd today than it was in 1979. The latest calculations by the Organisation for Economic Co-operation and Development show that the number of people who could expect to get benefit worth 70 per cent or more of their previous income if they became unemployed has fallen from 1,870,000 in 1985 to 425,000 in 1991/2.
It is also wide of the mark to blame the rise in unemployment on technology. Machines have always replaced labour, with the result that we are able to move to other occupations and increase our living standards. There can be problems during the transition. But if new technology caused unemployment in the long run, the jobless total would have been on a rising curve ever since Watt invented the steam engine.
The main cause of the recent rise in unemployment is the collapse in spending during this recession. Once demand and output begin to grow rapidly, unemployment will turn round. The first requirement is therefore for policies to encourage spending and output.
The sharp devaluation of sterling will help exports. But the impact of lower interest rates is likely to be muted, because the burden of debt inherited from the 1980s is so great. People will prefer to repay debt than to spend freely, which is why the Government's own spending is so important. An increase in taxes or a cut in public spending would risk aborting the recovery.
Demand management, though, can only take us so far. With good luck, it might allow us to cut unemployment to some 8 per cent. But below that figure, the leading labour market specialists (Professors Stephen Nickell of Oxford and Richard Layard of the London School of Economics) estimate that wage pressures would begin to rise. Wages would steadily outstrip the potential growth of the economy, resulting in accelerating prices.
Classical economists have traditionally called this figure the 'natural' rate of unemployment, but there is nothing natural about it. A better, if inelegant, name is the 'Non-accelerating inflation rate of unemployment' (Nairu). Let's call it core unemployment. It is caused by supply-side problems such as the lack of trained people for particular vacancies, so that employers have to bid higher wages to poach a scarce supply. Public policy can address these problems as effectively as it can address a deficiency of demand.
For example, wage pressure at any level of unemployment has certainly been reduced by the trade union reforms, the fall in benefits compared with earnings, and the Restart scheme, whereby the long-term unemployed have been pressured by the threat of losing benefits into reviving their job-hunting efforts. This was why unemployment fell so sharply in 1987 and 1988.
A new initiative will be necessary now that long-term unemployment is rising again. Employers are more reluctant to hire people who have been out of work for a long time, and they in turn become demoralised. Like unsold flowers, they are moved further back in the florist's shop, each time reducing their chances of sale. They fail to compete with those in work, so that there is a rise in the amount of unemployment needed to contain wages.
In extremis, the Government could introduce workfare, which deprives people of any benefit if they refuse even menial jobs. But Sweden's active labour market policy is more humane and effective. It also contains an element of compulsion, since it is heavily inspired by the Lutheran work ethic: benefits are withdrawn if people refuse offers. But there are incentives to employers who hire the long-term jobless, and the training genuinely prepares the unemployed for useful jobs, raising their long-run incomes.
Such training would reduce the mismatch between the unemployed and available vacancies, which remains a problem in Britain. More than half of the unemployed have no qualifications, and many vacancies for people with skills remain unfilled. Nevertheless, the recent newspaper comment implying that this mismatch has grown worse is almost certainly wrong. Skill shortages are a reason for a continuing high level of unemployment even in the late 1980s, but they have not caused the recent rises in unemployment.
The most telling piece of evidence simply compares the number of unfilled vacancies now with the number when unemployment was last this high in 1985: reported vacancies are now a mere 105,000 compared with 170,000 then. In other words, the degree of mismatch has fallen sharply because so many skilled people have been thrown out of work. It is wrong to say rising mismatch has caused unemployment; in fact, rising unemployment has reduced mismatch.
Another way of looking at the mismatch is to examine the increases in unemployment rates by occupational groups since the trough of unemployment in 1989. If computer operators are all in work, and the unskilled are out of work, this is clearly telling us that shifting patterns of job demand may be responsible for the rise in unemployment. However, the Labour Force Survey (available from Quantime) shows that unemployment has increased in all groups. You still stand a better chance of surviving the recession unscathed if you are educated, white- collar and skilled. The only oddity in the figures, shown in the table, is for the completely unskilled.
The survey results appear to show that the rise in unemployment for the unskilled is actually smaller than it is for the manual skilled. If true, this would torpedo much of the conventional wisdom about the benefits of skills. But it is almost certainly a statistical quirk.
In delving deep into the LFS, Jonathan Wadsworth of the LSE has found out that 18 per cent of those formerly in unskilled work appear to have withdrawn entirely from the labour force. We can only speculate why. They may have despaired of finding work. Perhaps two-thirds of them are drawing invalidity benefit, which has soared, rather than unemployment benefit.
This is a subject that deserves more study. My tentative conclusion is that there are as many reasons for supposing that core unemployment has fallen as there are for supposing that it has risen. A Swedish-style policy is still the best way of reducing core unemployment. In time, there is no reason why we should not reduce unemployment to the 2-3 per cent levels of the 1950s and 1960s. But the first priority must be to revive demand, output and jobs growth.