Sunday, November 27, 2005

Ethnic Segregation in Britain: Part 2   posted by DavidB @ 11/27/2005 03:33:00 AM

In a recent post I commented on some research by Ludi Simpson and colleagues. I said I would return to the claim by Simpson that the main factor in the growth of the ethnic minorities in Britain is the age structure of their population, and not continuing immigration or higher fertility.

Simpson's press releases say, inter alia:

After a couple of generations... the population growth of these groups [Black and Asian] will have slowed and probably stopped... Fertility [of ethnic minorities] has reduced rapidly from the high levels associated with immigrant families. It is the youthfulness of immigrant workers and therefore their low mortality which has caused population growth, not high fertility, and not further immigration.

A report in the Guardian says:

Immigration is not the reason for increased numbers of non-white Britons over the past decade... the increase in the number of non-white Britons is due to demographics rather than immigration. Ethnic minority populations are younger and have fewer elderly people than white communities. The number of Asian and black people is increasing because fewer die from old age and they have more women of childbearing age relative to white people. The author of the study, Ludi Simpson, said: "The common myth is that the growth of the ethnic minority population is due to immigration. That's not true - it is more due to the growth of [ethnic minority] people born in Britain."

These statements are unclear as to whether the growth of the ethnic minority population is wholly or just mainly due to the age structure of the population, and whether fertility of ethnic minorities has already fallen to average levels, or whether this is a prediction for the future. But taking all the statements together, it seems that Simpson's position is as follows:

- the main factor in the recent growth of ethnic minorities ['over the last decade'] has been the age structure of the ethnic minority population, and in particular its relative youthfulness

- immigration has been only a minor factor

- fertility of ethnic minorities has already fallen substantially and can reasonably be expected to fall to replacement level.

Are these claims true?

First we need to establish the size of the increase in the ethnic minority population. In England and Wales (where most non-white ethnic minorities live), the non-white population increased by about 50% between the 1991 and 2001 Censuses, from 3.2 million (6% of the total) to 4.7 million (9% of the total). This is a very rapid increase for a single decade. Some small proportion of the increase may be due to the availability of a new 'mixed' category in the 2001 Census. On the other hand, it is likely that the 2001 Census understates the true increase, since there was substantial illegal immigration during the decade, and the Office for National Statistics has admitted that the Census total for 2001 is too low. The increase is also likely to have continued, if not accelerated, since then.

The increase can be attributed to three factors:

a. age structure

b. immigration


c. higher lifetime fertility.

Simpson's statements imply that (a) is the most important factor, and that (b) and (c) are relatively minor.

British immigration data are poor. However, figures for grants of settlement suggest that in the first half of the 1990s immigration from Africa and Asia was relatively low and stable, at around 50,000 per year. But this still represents an inflow of more than 1% of the ethnic minority population in a single year, and an increase of cumulatively around 15% in a decade, which would be a significant proportion of a 50% total increase. Moreover, immigration from Africa and Asia increased rapidly after 1997 (see pp. 14-15 of the ONS Social Trends for 2005.) By 2003 it had increased to around 100,000 a year, representing more than 2% of the current non-white population per year. There is no sign of any slackening. One relevant factor has been abolition of the 'primary purpose' rule for the admission of spouses since 1997. Under the primary purpose rule, spouses of existing residents were not permitted to settle in the UK if the primary purpose of the marriage was assessed by the Immigration Service as being to obtain entry to the UK. The abolition of this rule (which admittedly was difficult to apply) has increased the number of arranged marriages with spouses from Pakistan or Bangladesh. Another factor in rising immigration is the large increase in immigrants from Africa, often in the guise of 'asylum seekers' from Nigeria or Somalia. It should however be noted that there has also been a large increase in white immigration from Eastern Europe in the last few years.

Overall, it seems likely that about one third of the increase in the non-white population over the last decade can be attributed directly to immigration. This is hardly negligible.

The third element to be considered is lifetime fertility, usually expressed as the Total Fertility Rate (TFR) for women of child-bearing age. Ludi Simpson claims that fertility has already reduced rapidly from the high levels associated with immigrant families, and he suggests that it is likely to fall further.

I have not been able to find reliable recent ethnic fertility data for Britain (or England) as a whole, but there are some useful data for London in a study of 'Fertility of Ethnic Groups in London' by the Data Management and Analysis Group of the Greater London Authority in September 2003. This gives the following TFRs for women in London:

Black Caribbean.........1.70
Black African...........2.28
Black Other.............1.51

It will be seen that most groups are below the replacement rate (just over 2 children per woman), the exceptions being Black African, Pakistani, and Bangladeshi. It is noteworthy that the Indian TFR is among the lowest, though not as low as White. It is also clear that the TFR for Bangladeshis and Pakistanis is still substantially above the replacement rate, and more than twice the level for Whites. This may reflect the fact that a larger proportion of women in these groups are first-generation immigrants (see the comments above on arranged marriages). However, Black Caribbeans and Indians, with low TFRs, are among the largest minority groups, so it may well be that the average TFR for all non-whites in London is around the replacement rate.

London is not entirely representative of Britain as a whole. It is difficult for young married couples to afford housing in London, so there tends to be a high proportion of single people, including single parents in subsidised housing. But a large proportion of ethnic minorities in Britain live in London, so the London data on ethnic minorities cannot be badly misleading. If the average TFR for all non-whites is around the replacement rate, then Ludi Simpson may well be right in claiming that the overall increase in the ethnic minority population in Britain in the last decade has had little to do with higher fertility. I did not expect to find this result when I started digging, so I am obliged to draw attention to it. On the downside, the Pakistani and Bangladeshi groups, which are still increasing due to higher fertility, as well as other factors, are among those with the lowest employment rates and the worst educational performance. They are also those with the lowest rates of intermarriage with other groups. So the prospect of increasing concentrations of discontented, idle, Muslim youths, a la Francaise, cannot yet be entirely discounted.