New Year, new Resolution
As everyone knows, the numbers working in the UK and the employment rate both continue at record-ish levels. I’ve pointed out on many occasions before that these headline numbers aren’t necessarily a very good guide to what’s actually happening in the labour market.
There are two main reasons. Firstly, simply counting the number of people in work doesn’t say anything about the kind of work they are doing. Secondly, increasing numbers of workers from abroad skew the headline figures and obscure what is happening to the UK-born population.
I’m reminded of this a new report from Resolution Foundation, looking at differences between employment and wages in the UK and the USA, and finding greater employment growth in the UK than in the USA contrasted with much higher wage growth (and productivity) in the USA than the UK. Among other interesting things it notes a large difference and divergence in the proportion of self-employed people.
Official statistics on employment come from the Labour Force Survey (LFS) carried out by the Office of National Statistics and there is big release of these labour market statistics every month. This includes all sorts of interesting detail about employment status, hours worked, industry sector, earnings, regional breakdowns etc etc. However, the release only has any information on country of birth and nationality every third month and this information is restricted to a simple count of numbers working and employment rate. So it isn’t possible to tell anything about whether workers from abroad are more or less likely to have become self-employed, or to work in particular industry sectors or in particular parts of the country. One might say “so what?”, but this also means that you can’t say anything similar about UK-born workers either, because it isn’t possible to disentangle the ‘migrant’ and UK-born components of any change in the official statistics from one period to another (apart from the reported headline employment numbers and employment rate). You can work it out for yourself as (most) of the LFS dataset is published, but not until a month or two after the official statistics come out. This isn’t very satisfactory because it means that up to five months can pass after the publication of ONS monthly labour market statistics before one can see what is really behind the headlines.
That’s enough preamble. The latest LFS dataset came out a few weeks ago, and it’s one I look out for as UK-born employment has a very seasonal pattern which peaks almost without exception in the third quarter of the year as a new tranche of people leave education and enter employment. I’ve taken the opportunity to compare where we are now (or were on the most recent Jul-Sep 2016 figures) with the same quarter in 2008 before employment tumbled post-recession.
The headline figures show that UK employment has increased by over 2 million, and the employment rate has gone up by 2 percentage points over the period. However, looking at the published figures for the UK-born, employment has gone up by only 400k and the employment rate has gone up by only 1.6 percentage points. That still looks quite good, half a loaf being better than none. But is it as good as it seems?
Firstly, as a further preamble, the published statistics for employment levels and rates by country of birth aren’t different views of the same thing. The employment level reported is for all ages but the employment rate reported is for people aged 16–64 only. If we look at employment levels for UK-born 16–64 year olds in the LFS dataset itself (as you won’t find them published), they have barely moved since 2008. In fact they show a tiny decline of around 50,000, so in bare numbers all of the UK-born increase in employment is down to more people working who are 65 and over.
But the lack of change in overall employment for UK-born 16–64 year olds hides change in the nature of the work done, with half a million fewer working-age people in full-time employee jobs and much the same number more in part-time jobs or self-employment.
Now you might wonder how the UK-born employment rate can have gone up if the number working has remained the same. The answer is of course that if the numerator is the same, then the denominator must have shrunk! And indeed the LFS shows the UK-born 16–64 population to have declined by over 500k since 2008. The two main reasons for this are demographics as a post-WW2 baby bulge hits 65 and is not fully replaced by a new cohort reaching working-age (as illustrated below) and a slow but steady loss to emigration.
Thus not only is the rise in employment levels for the UK-born entirely accounted for by older people, but the 1.6% increase in the reported employment rate for the UK-born of working age is more than entirely accounted for by work that is not full-time employee work, as shown in the picture below. It’s possible that some of this is preference but against that, lots of part-timers (about a million all told) say in the LFS that they would like a full-time job but can’t find one. The survey doesn’t ask the self-employed if they would prefer to be employees.
Now just as Resolution Foundation has framed their UK/USA comparison in the context of TRUMP! so the debate over jobs’n’wages in the UK is in a context of BREXIT! Over the 2008–2016 period there have been large changes in the number of people from the EU working in the UK. On the one hand it could be argued that along with people from outside the EU too they have made up for the drop in the working-age UK-born population (though together they comfortably exceed it) but on the other hand people from the EU have disproportionately ended up with full-time employee jobs over a period during which the percentage of working-age UK-born workers who have such jobs has fallen (as above), and the number of part-time workers who wanted but could not find full-time jobs has increased by 400,000.
In the TRUMP!/BREXIT! context, it is also interesting to note the differing extent of change in the amount of labour from abroad in the USA and the UK. I wrote a little observation about this a year ago, with this picture …
and noting …
UK-born employment levels didn’t drop as much as Native-born in the USA, but once levels in the USA had caught back up in mid-2011, they have evolved very much in the same way up to the present, to be a little bit higher than they were pre-recession. Over the entire period, foreign-born employment increased at a far greater rate in the UK than in the USA. To match the increase in the UK, the number of foreign-born workers in the USA would have had to have increased by over 7 million rather than the 2.5 million actually seen. Which they didn’t.
Thus while in both the UK and the USA, foreign-born workers have accounted for most of headline employment growth since pre-recession peaks, the greater employment growth in the UK derives from an even greater number of foreign-born workers. This isn’t the whole picture of course, as the political narrative in the USA is as much about jobs going to workers abroad rather than workers from abroad, in contrast to the narrative in the UK. However, it remains the case that growth in workers from abroad has been much more significant in the UK than in the USA. Whether this might be part of the explanation for lower wage growth and productivity in the UK over the period is of course a matter of further debate.
As always, this is intended merely to look behind the headlines and consider what ‘confounders’ might be there. No judgement as to the value of migration is implied or to be inferred!