Recent headlines about how average salaries in hospitality and retail have risen pose some interesting questions. On Twitter I was asked: ‘How do economists reconcile with the idea EU immigration was not impacting wages? Timing? Sector? Nothing to do with Brexit? Or were the models wrong?’
These are fair questions, in light of widespread reports of labour shortages resulting from the exodus of EU-origin workers during the pandemic.
For the last 15 years – at least since the accession of central and Eastern European countries to the EU in 2004 – economic research has consistently shown that migration had no significant impacts on the employment prospects of UK-born workers; and that wage impacts, while observable, are relatively small.
The most well-publicised estimate suggests that migration might have depressed wages for lower-skilled service sector workers by perhaps one per cent over a period of almost a decade.
As a consequence, most economists argued that the ending of free movement after Brexit would be unlikely to do much to boost wages for low-paid British workers. Are we now being proved wrong? Were our earlier estimates (not really ‘models’, since little sophisticated analysis or theory was involved – it was just what the data said) incorrect, or is it simply that circumstances are very different?
As you’d expect from an academic, my short answer is ‘more research is needed’. But that doesn’t mean that there aren’t some useful things we can say now.
First, take the headlines with a very large pinch of salt, and fact-free politically motivated commentary with an even larger one. Remember we’ve been here before.
In 2017 and 2018, there were widespread reports that a post-Brexit reduction in EU migration was leading to staff shortages and wage increases, supposedly of 10% or more in the hospitality sector. The usual suspects seized on this: ‘This is precisely why a lot of people voted for Brexit,’ said Fraser Nelson. This is ‘more evidence that free movement resulted in lower wages’, claimed Paul Embery.
Except that’s not what the actual hard data showed; in fact wages in the sectors most dependent on relatively low-paid service workers, particularly those from the EU, rose about 2.5% a year in 2017 and 2018 – mediocre by historical standards (and even those rises were in large part driven not by changes in migration but by rises in the national minimum wage).
Overall, Brexit reduced, rather than increased, real wage growth between the referendum and the pandemic, much as I and other economists predicted. So rather than taking excitable press releases as fact (still less Twitter commentary driven more by ideology than analysis), we’ll need to wait for official data to find out what’s happening to wages.
So far – while the data is hard to interpret, given the swings driven by the pandemic and the furlough scheme – there’s no sign of any wage explosion in the relevant sectors, with pay in wholesale/retail and restaurants/hotels up less than four per cent in May of this year, compared to two years earlier.
Until we have at least a few months’ more data that tells us what actually happened after most restrictions were lifted, we should be very cautious indeed about making general claims about the impact – or not – of changes to migration patterns.
That said, there are very clear and consistent reports of staff shortages as businesses – especially in the hospitality sector – reopen. And there is clear evidence that there was indeed a large exodus (even if we remain uncertain as to its magnitude) of EU-origin workers during the pandemic. It’s entirely reasonable to suggest that the two are connected.
But it’s also important to recognise that these are very special circumstances indeed.
The hospitality sector has been hit by successive shocks to labour demand (driven first by the imposition of restrictions, and now by their removal) and to labour supply (resulting primarily from emigration, but also perhaps from some workers leaving the sector permanently). It’s hardly surprising that this should result in substantial labour market ‘mismatches’ – where the supply and demand for labour don’t match up.
What does that tell us about the impact of migration on wages (or other outcomes we care about, like unemployment, working conditions, productivity and investment)?
Frankly, not a lot, because current conditions will, almost by definition, be temporary and transitory. That could work either way: maybe businesses will put up wages to attract scarce workers now into hospitality and other sectors, even if they know they can’t afford to do so long-term, hoping to lower them again in future when more workers are available.
Or maybe they’ll just leave vacancies unfilled and wait it out. But either way, their behaviour during this period of transition doesn’t tell us that much.
That doesn’t mean, however, that the pandemic, the departure of many EU workers last year, and the introduction of the post-Brexit migration system won’t have lasting impacts on the labour market; these are not just transitory impacts but, potentially, big long-term shocks. But how?
First, the pandemic may, or may not, affect labour demand. If there is a big shift towards working from home some or all of the time for office workers, that will have an impact on the demand for many businesses in the hospitality sector and other customer-facing service businesses in city centres; which are in turn big employers of migrants.
Second, to the extent that those who left the UK do not return, and the new migration system affects the ability or willingness of new migrants to come here, that will affect labour supply.
The net effect? If – while it does seem the most likely outcome, it’s still not certain – labour supply is indeed reduced more than demand, then there are a variety of possible responses: higher employment for UK-origin workers, higher wages (which in turn would likely mean higher prices), higher productivity, or lower output/fewer businesses.
All of these are possible in theory; but it’s here that the evidence of the recent past does at least enable us to make some predictions.
These are that the main impact will be through lower overall employment and output; that there will be no significant increase in employment for the UK-born; there will be some, but probably rather small, increase in relative real wages for workers in the most affected sectors (offset by an even smaller fall in relative real wages for other workers who consume the outputs of those sectors); and that there will be no offsetting increase in productivity (indeed, productivity, which in the UK at least seems to be positively impacted by migration, may even fall).
These predictions are far from the overheated rhetoric of some commentators – both those who think ending free movement will be an economic disaster and those who think it will transform the prospects of British workers.
And, based as they are on the evidence accumulated over two decades of high net immigration to the UK, they may not be valid for a period of much lower migration, possibly accompanied by major structural change driven by the pandemic.
But so far they are still the best guide we have to what is likely to happen.