About a million of these are among Britain's ethnic minorities, many working in retailing, hotels, catering, services and the making of clothing and footwear. These industries, in addition to low pay, offer low union membership, irregular employment, and often poor working conditions. The over-representation of ethnic minorities in these sectors reflects discrimination, poor qualifications, concentration in cities and a relatively low age profile.
Research during the 1960s and 1970s painted racial disadvantage in the labour market as a simple contrast between well-off 'whites' and poor 'blacks', a category into which all the ethnic minorities were conveniently subsumed. But changes during the 1980s have rendered this description largely obsolete. Discrimination and disadvantage remain, but it has become increasingly misleading to talk of the ethnic minorities as a single group with common interests and experiences.
In 1990 ethnic minorities made up about 2.6 million of Britain's 52 million population. They included 584,000 Indians, 563,000 Pakistanis and Bangladeshis, 474,000 Afro-Caribbeans, 256,000 African Asians, 171,000 Africans and 136,000 Chinese. The ethnic minorities have grown by about a quarter during the 1980s, while the population as a whole grew by about 1 per cent.
Last winter's Labour Force Survey put the unemployment rate among the ethnic minorities at 20 per cent, twice the rate among the white population. The jobless total for the ethnic minorities tends to fall faster than that for whites during upturns and rises faster during recessions. They are more vulnerable to job-shedding, but tend to be re-hired more quickly.
The rise in ethnic minority unemployment in the last recession may have been tempered by the fact that big cities suffered lightly in comparison to most downturns. The upturn may also be more beneficial because ethnic minorities are concentrated in younger age groups, while demographic trends point to a shortage of young workers.
The graphic below shows that trends in unemployment have differed widely among the ethnic groups since the mid-1980s. The Pakistani and Bangladeshi communities have consistently suffered worst, with a jobless rate this winter of 33 per cent. In marked contrast, the unemployment rate for Indians is 13 per cent, not too far above the figure for whites. The rates for Chinese and African Asians are even lower. The Afro-Caribbeans lie between the two extremes, with a jobless rate of about 25 per cent.
Early studies assumed that ethnic minority unemployment rates would move towards those of the indigenous population as each immigrant group adapted to conditions in Britain. But this process is very slow. The biggest contrast in unemployment rates is between two of the groups who have arrived most recently - the Bangladeshis and African Asians. Differences between unemployment rates are also largest among 16-24 year olds, and in all ethnic groups most people in this age group will have spent their formative years in Britain.
The differences in jobless rates are partly explained by the types of jobs in which each community is concentrated, which in turn affects their vulnerability to redundancy. A recent study by the Policy Studies Institute concluded: 'Certain groups - the African Asians and Indians, in particular - have come to occupy a labour market position hardly inferior to that of whites. Other groups - especially the Bangladeshis, and to a lesser extent the Pakistanis - continue to occupy a very much poorer position.' *
The job levels of some ethnic groups are much more polarised than those of whites. The proportion of Indian, Chinese and African Asian men in professional and managerial jobs is as high as or higher than the proportion of white men, but a higher proportion also have low-grade jobs. Interestingly, even groups with large numbers in top jobs tend to be under-represented as managers in big companies.
But even after removing the effect of differences in job levels, disparities in unemployment rates remain. The same is true after account is taken of differences in qualifications, age, gender and location. Earnings show a similar pattern. The chart based on work by Warwick University's Centre for Research in Ethnic Relations suggests the average household income of Indians was above that of whites through the late 1980s. Afro-Caribbean incomes rose towards the national average during the boom but fell back as the economy turned.
The Policy Studies Institute argued that it was unlikely the remaining differences in earnings and unemployment between ethnic groups were explained by differences in the degree of discrimination they faced, but rather that some groups were better placed to develop ways of overcoming the resulting constraints.
African Asians, for example, brought skills, qualifications and capital with them when they fled persecution in Kenya and Uganda, while the Bangladeshis had little capital or skills when they came to Britain to escape poverty at home.
This suggests that the impact of the abolition of the Wages Councils will differ widely between ethnic groups. The Government argues, correctly, that Wages Councils reduce employment. But this effect will be reversed only if their abolition results in the reduction of already low wages in these industries. Those ethnic groups, particularly Pakistanis and Bangladeshis, concentrated in them will find one form of disadvantage worsening and another improving. Overall, their lot is unlikely to improve.
Perhaps the most hopeful route to reducing racial disadvantage lies in improving the quality of post-16 education. Young people from even the most disadvantaged groups are much more likely to stay on in education or training after school than their white equivalents. Improving education is no quick solution, but alternatives may be little more effective.
* Britain's Ethnic Minorities, Policy Studies Institute, 1993.
Christopher Huhne is on holiday.Reuse content