Gross flows

This prompted by a piece in today's FT about knowing about migration. That is to say, reference 1.

Fifty years ago when I was deeply involved in what was then called the Mid Year Estimates, we used to try estimate the population of every local authority in the country - all 400 or so of them - excluding Scotland and Northern Ireland that is - every mid year. Estimates which were an important part of all kinds of policy and planning activities and which also figured in the computation of how much money was to be given to the local authorities by the central authorities, it being thought then, as now, that local authorities were not competent to raise their own money - although they were allowed to spend it, subject, of course, to central supervision and interference.

The drill in those days was that population in year 2 equalled population in year 1 plus births minus deaths plus net inward migration. A ten year drill which rolled us forward from one decennial census, more or less authoritative, to the next. We had a pretty good grip on births and deaths - but we did not have a good grip at all on migration - and sometimes the accumulated error by the time we got to the next census was large.

The first step was to analyse net inward migration into migration in from elsewhere in the UK, immigration from abroad, migration out to elsewhere in the UK and emigration. I forget whether for these purposes Ireland was included with the UK or with abroad. The difficulty of estimating net inward migration being illustrated by the fact that this net figure was small compared with the gross figures. Relatively small errors in the gross figures could result in large errors in the net figures. And in the case of abroad, huge numbers of people were moving in and out of the UK all the time, only a very small proportion of whom were migrants. Quite a lot of whom would not know that they were to become migrants at their time of entry or exit. So how on earth could we know? And in this mobile age, what exactly do we mean by a migrant? What do we do with all those people who keep moving backwards and forwards all the time? 

Various wheezes were used to try to estimate migration within the UK, wheezes which mainly involved milking administrative records, perhaps health records, collected for some other purpose. Some people thought that that a national register and identity cards was the answer, but this idea never really took off, although various consultants made money out of it. And it still seems that we don't mind Amazon, Facebook and Google knowing all about us, but that we do mind the government knowing. Unless, that is, we want to extract some money out of it. While I now wonder how much easier it might have been to control the virus if we had had such a register to build on - a register with 70 million records being much less of a engineering challenge than it would have been back in the 1970's.

While for abroad, we mainly relied on something called the International Passenger Survey, an interview (sample) survey conducted at places of entry and exit, places like Heathrow, a survey seemingly still alive and well, although suspended for the virus, and supplemented by information collected by the Home Office. With a lot of this last being collected in my day on small pieces of white cardboard called landing cards, this being well before the days when the people manning desks at places like Heathrow had computers. I don't remember whether people travelling within the EU, then newly joined, were excused the landing cards; quite possibly. As I recall, the Home Office was rather asymmetrical, being much more interested in counting people coming in - immigration being a big issue at the time - than counting people out. And it counted far more people coming here to live from the Indian sub-continent than the Passenger Survey was picking up in the crowded arrival halls at Heathrow - where conditions were far from ideal for statistical surveys on a shoestring. Not least the facts that some did not speak much English and that some, not many, were extremely rude. One of the mysteries of human nature being that despite having sat on an aeroplane for hours, when they get off, every second counts. Cruise liners were just as bad, if not worse.

The impression given in the article in the FT is that things have not moved on all that much, although the Labour Force Survey (of reference 2) is now added to the mix, only just getting under way in my day - and which I would have thought rather a blunt instrument for trying to measure migration. And I was surprised to find no mention of further exploitation of the extensive computer records which must be held by both airlines and the Home Office, bearing rather more directly on peoples' movements. I don't suppose I will ever get to know why not.

Reference 1: 'Massive uncertainty’ over UK migration data amid pandemic: Britain is ‘flying blind’, warns expert, after suspension of immigration survey as post-Brexit system is rolled out - Robert Wright/FT - 2021. 5th February 2021.

Reference 2: https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/methodologies/labourforcesurveylfsqmi.