A few weeks back I saw this chart in a speech about the UK labour market by Michael Saunders. The chart shows the evolution of total employment in the UK by country of birth (not by nationality) from 2000Q1. The data is not seasonally adjusted.
I was intrigued, so I decided to reproduce the chart, and quite easily I found the relevant ONS data.
A few things have to be kept in mind here.
One, changes in employment for those born outside of the UK reflect immigration, more than anything else, since we are talking about millions of people, not employment rates.
Two, employment in the UK has been on the rise, regardless of the country of birth. The increase has been most pronounced for those born outside of the EU.
Three, as quoted by the BoE, more than half of the growth in employment for the past 10ish years occurred because of those born outside of the UK. It can be nicely seen when the same data is presented like this:
This is largely unsurprising. The stock of those born, living, and working in Britain did not expand considerably in the past 20 years. Hence, one wouldn’t expect huge employment changes (unless people started joining the labour force from a previous state of inactivity en mass, which was not the case).
What I found most intriguing in this chart is the resilience of employment for those born outside of the UK during the recession, and the considerable (a good couple of hundred thousand people) drop in employment of those born in the UK in the aftermath of the recession.
It’s difficult to form a coherent narrative with the absence of other data. I found this chart pretty scary, though, because the most straightforward explanation is that foreign workers – willing to accept lower wages – displaced UK-born citizens in the labour market. Hence, it was the British-born who felt the effects of the crisis.
Of course, that’s not the only possible story. Another story would be that the types of jobs performed by migrants have been less affected by the crisis, but it’s difficult to test this theory in the absence of reliable data. There is some data on skill levels by country of origin, yet there is likely to be a considerable amount of downgrading going on, with migrants willing to accept jobs below their skill levels (all the stories of cleaners in the UK coming from Eastern Europe with university degrees).
One more alternative is outmigration. Migrants who lost their jobs in Britain could have simply left the UK, and were replaced with other migrants, so job losses are not reflected in employment statistics. Therefore, the composition of the group of those foreign-born is changing over the period considered, and is likely to include only the most motivated and resilient. I’m also not aware of any reliable outmigration statistics, though.
Let’s move over to employment rates.
What stands out here is that the employment rate declined for all groups considered during the crisis. Nevertheless, it is still higher now than it was pre-crisis for all three groups.
ONS provides total employment and employment rates data, so I figured I can infer the size of the labour force by country of birth.
I got another surprising chart in return.
The chart suggests that increases in the employment rate after the crisis (reflected in Chart 3) for those born outside of the UK are different than the reasons for the increase in the employment rate for those born in the UK.
The employment rate is made up of total employment (numerator) and size of the labour force (denominator). The labour force is made up of those in employment and those classified as unemployed.
Following the crisis the numerator is increasing in the group of the foreign born. As suggested by Chart 4, the denominator is also increasing, albeit by less, since the employment rate is increasing.
But the same is not the case for the British born, where the labour force is seen to increase in the run up to the crisis (particularly sharply between 2009 and 2011), and decrease following the crisis. In fact, the labour force shrank by 700 000 people born in Britain following the crisis!
There are two stories which could explain the decline in the labour force, however neither of them withstands scrutiny. The first would be that the population of working-age British-born people declined markedly following the crisis. This is very unlikely. The second would be that the British were dropping out of the labour force, en masse. This makes more sense, and it would reflect a general sense of dissatisfaction with economic conditions, potentially suggesting other negative effects of migration on the labour market.
Another explanation for the chart has to do with the data and employment reporting. It could be that more people were registering as unemployed around the time of the crisis than it would have been the case in the absence of a recession. The subsequent decline would then reflect some sort of adjustment (as the size of the labour force reverts back to its pre-crisis levels).
I was really puzzled by this, so I tried to look at other sources of data to gain support for one of the theories.
The only data on labour force participation by country of origin which I could find comes from the OECD:
The chart in fact shows a decrease in labour force participation in the run up to the crisis, and an increase in labour force participation for the UK-born! The opposite of what would be suggested above.
ONS data comes from the labour force survey, while the OECD also cites labour force surveys as their source. Nevertheless, the charts – in the absence of large jumps in the population of those born in Britain – seem to be incompatible with each other.
This is where I stopped, since I can’t really get hold of other data. The story of the macroeconomic impacts of migration on the UK labour market did not seem easy to begin with, but now it’s even more difficult than before.
This post greatly benefited from discussions with JMR.