What effect does immigration have on wages?
By Ian Preston
Concern about the economic impact on UK-born workers motivates much opposition to high levels of immigration. Immigration over the past decade and a half has brought new workers into the country who are younger and better educated both than British born workers and earlier immigrants. For these newly arrived workers to be employed requires either that they displace employed workers already resident or that the labour market adjusts somehow so as to absorb them. Fears that such adjustment may harm the interests of the existing UK labour force reappear frequently in political argument.
The evidence that immigration adds to UK unemployment is thin so how does the labour market adjust to incorporate them? There are many ways in which the labour market can adjust so as to accommodate the higher number of new workers. It can change the types of goods produced, expanding production of goods which use newly abundant skill types. It can adopt different technologies which are more appropriate to the new mix of skills. It can import more of other inputs, particularly internationally mobile sorts of capital, raising demand for workers by raising their productivity. Any or all of these may happen and it is not implausible that in the long run they may be all that is needed to absorb the immigrant inflow into the labour market. However adjustment can also come through wages. Rates of remuneration of different inputs to production can change so as to make employment of the newly arrived labour more attractive. To the extent that such changes can happen quickly and to the extent that such changes may be a necessary intermediate step that provides signals which prompt other adjustments to occur, these effects may be the main ones to occur in the short run.
If changes in wages are to occur then they may not be simple. Immigrants may compete strongly with workers performing certain tasks, while complementing the labour of those doing others. Evidence published in the latest Review of Economic Studies and drawing on evidence for the period 1997-2005 looks to see at what points in the wage distribution of UK-born workers changes are most strongly correlated with immigrant inflows. Estimates of the wage effects of immigration at points across the whole distribution present a clear picture suggestive of strongest negative effects at the bottom end of the distribution of wages and wage rises higher up. These effects are modest; for each 1% increase in the ratio of immigrants to the UK born working-age population the fall in wages is about 0.5% at the lowest decile and the gain about 0.6% at the median.
Given the observation that immigrants are typically better educated than those already working in the UK labour market the location of these effects is perhaps surprising. Assuming that immigrants compete most strongly with those similarly qualified would lead one to expect depressing effects on wages higher up the distribution. However if you look at the wages earned by recent immigrants a different picture emerges. Immigrants, especially the most recently arrived, work for wages much lower than you would expect for a UK born worker with similar qualifications. There is indeed a striking inverted similarity in shape between pictures showing where in the wage distribution immigrants are found within 2 years of arrival and where wage effects are strongest.
Looking at the wages of more and less recent immigrants shows that over time immigrants move up the distribution towards jobs providing wages more suited to their qualifications as they integrate into job-finding networks and acquire host-country-specific skills, especially language ability, which enhance productivity in the British labour market. As that happens the downward pressure on wages at the lower end of the labour market should alleviate.
Are these effects on wages pure redistribution or is there also an imbalance between losses and gains at different points in the distribution? Taking the effects at all points in the distribution together, the evidence seems to point to a slight average gain. Maybe immigration “greases the wheels” of the labour market, overcoming problems caused by immobility of local labour and addressing specific labour shortages; perhaps immigrants are paid less than the value of what they add to production because of downgrading generating a surplus that benefits other workers.
Whatever the reason, the gains in wages of those higher up appear to be at least enough to compensate the losers at the bottom. There is little reason to see immigration as harming the returns to labour overall. What it may do is aggravate wage inequality in the short term as newly arrived immigrants enter the labour market at the lower end, holding back wages of the lowest paid, even if only modestly so, while benefiting those higher up. If a liberal immigration policy could be combined with progressive policies to effectively redistribute the gains then the benefits could be enjoyed generally. Meanwhile, labour market institutions, such as the national minimum wage, which protect wages at the bottom remain important safeguards for the lowest paid.
Ian Preston is Professor of Economics at University College London, Deputy Research Director of the Centre for Research and Analysis of Migration and a Research Fellow of the Institute for Fiscal Studies. He is on Twitter at @ianppreston. For more information on this topic, the Review of Economic Studies has made the paper The Effect of Immigration on the Distribution of Wages, by Christian Dustmann, Tommaso Frattini, and Ian Preston free for a limited time.