New data on migrants earnings

And the importance of comparing by age

I have been meaning for a while to write an update on the economics of migration, following up on some previous pieces.

I have some new data from an FOI request I made to HMRC, which gives us more up-to-date and granular information on how much migrants in the UK are earning.

The long and the short of it is that the decline in non-EU migrant average earnings that we have seen since the new immigration system was introduced has remained.

This matters, because the number of EU nationals working in the UK has declined over recent years, while the number of non-EU nationals has increased rapidly (see annex below), nearly doubling since 2020.

The new data allows us to compare migrants’ earnings to UK nationals of the same age. It shows us that median earnings for migrants of working age (aged 22-64) fell back relative to the earnings of UK nationals after the introduction of the new migration system, and they remain lower:

Obviously the migrants who were already here and earning more in 2021 didn’t suddenly all disappear - so to drag down the average in this way, the new arrivals since 2021 have to have been earning significantly less.

This is a very bad result because:

• Census data (the last good data on employment rates) showed that employment rates for people born outside the UK were also lower.

• Given the UK problems with housing and infrastructure, we really need migrants to be earning a large premium on natives to offset the negative effects of adding more people on productivity and living standards. I have written before about this - adding more people pushes up housing costs, and further congests public services and infrastructure). So having migrants on average earning less than UK nationals of the same age is really not great for living standards.

HMRC normally produce the data without any age breakdown. And some people will say we should not compare like-for-like - a key argument that migration advocates make is that migrants are more likely to be working age, so more likely to be earning.

But there are several things to say about this.

First, there is just something a bit strange about comparing completely different people. For example. In the published HMRC data 9% of the UK nationals they have data on are under age 22, compared to just 3.3% of the non-EU nationals and 0.1% of the EU nationals.

Young people under age 22 earn far less (about 40% of what prime age people earn) and so drag the UK average down. We can of course make this comparison and ignore this difference in composition, but doing so it conceals as much as it reveals.

Second, the argument that migrants were more likely to be working age was probably more meaningful for policy in the past than it is now. During the peak period of EU migration from the mid-2000s to the late 2010s we did for a time have what seemed to be a ‘circular’ model of migration - there were a lot of single Polish guys who came, worked hard, didn’t go to the doctor much, and then left again.

But the mix under the new system is very different - people are now coming from much poorer countries and are much more likely to stay for good. So we should now be more interested in their likely long term earnings trajectory rather than just a snapshot at one moment in time.

If non-EU migrants are earning less than similar people of the same age and if they are then going to stay for good and ultimately retire here, that has big (negative) implications for their fiscal impact over the whole life cycle.

53% of households pay in less in tax than they tax out in public services and benefits, according to ONS. So we want to make sure that people who come are high earners if we want them to be net contributors.

Staying on

I have written before about changing staying-on rates for students. As the ONS has noted:

More non-EU+ students and their dependants have been staying longer, and nearly 1 in 2 transitioned to a different visa type after three years from YE June 2021; an increase from 1 in 10 after three years from YE June 2019.

The same growth in staying on is also true more broadly.

As a piece for the Migration Observatory notes: “Attrition rates are considerably lower among non-EU employees who entered the workforce between 2020 and 2023, reflecting the change in non-EU migration patterns post-Brexit. In other words, non-EU employees became more likely to stay in the employee workforce in this period.”

Of those who came to the UK and started working here in 2017, 33% had gone within three years. Of those who arrived in 2021 the exit rate was half that - just 16%. So we should be more interested in the likely long term trajectory of there migrants earnings.

Catching up?

The Migration Observatory piece I linked to above makes the point that migrants earnings increase relative to the (all-age) average the longer they are in the country. And part of the story about the recent decline in the average is indeed surely about having lots of new low earning new migrants.

But this more optimistic story about migrants’ earnings “catching up” still doesn’t try to compare migrants’ earnings to people of a similar age - instead they compare to a whole economy average, which includes much younger people. We would expect their earnings to increase relatively as they get older, as well as being in the country longer.

But the Migration Observatory data also shows that none of the cohorts who have arrived each year since 2014 have seen their earnings catch up with established UK workers who were already present when the HMRC data starts in 2014 - this is also a disappointing result, which suggests to me that we should aim for a higher earning mix.

Conclusions

This newer FOI data shows how off-track migration policy has been. We want any people we allow to come here to be the sorts of people who are very highly skilled and so earning significantly more than UK peers. But at present the median non-EU employee is actually earning less than people of the same age.

That means existing citizens are getting the downside from further pressure on housing, infrastructure, public services and the capital stock, but are not (currently) on track to get a massive fiscal gain to offset or justify that.

Previously you could argue that the within age-group comparison I have shown here mattered less, because more migration was temporary. But as the mix has shifted to more people staying for good, so making a like-for-like comparison matters more.

Even doing a “snapshot” analysis of the fiscal impact of migrants is tricky enough. For what it’s worth, all the historic studies listed by the Migration Observatory suggest non-EU migration has been a net fiscal negative. The Migration Advisory Committee have promised to return to this tricky issue of “life cycle” fiscal analysis in future publications, and having seen the data above I think looking at migrants’ earnings by age must be an essential part of that.

Everything I have written above is looking at the median earnings too. If even the median non-EU migrant is earning less than UK nationals of the same age, that implies half of people are earning even less than that. That in turn suggests to me that there is plenty of scope to reduce migration by making it more selective, and far from being negative for the economy, the overall effects on living standards would be positive.

While I expect migration numbers to fall over the coming years (partly because of the 2023 reforms), the policy moves the government are making are mainly in the wrong direction.

Abandoning plans to tighten the family route will mean more people coming with with low or no earnings. Plans for a EU mobility deal would by definition allow people to come who currently wouldn’t earn enough to meet the requirements for a work visa. The abandonment of Rwanda has left the government with no plans to reduce asylum, and claims have increased to a record high - this is the worst form of migration from an economic point of view, and many of those currently being granted asylum are really mainly motivated by the desire to move to a higher income country.

I have said it before and I will say it again: a more selective approach would be better for living standards.

Annex -

Number of employments held by EU and non-EU nationals (source: HMRC)