Immigration: wishful thinking by well-meaning people

This is the first of four blog posts by Professor James Wickham on migration

James Wickham: In an article last December in the Irish Times (‘Extensive research contradicts the UKIP view of immigration to the UK', Irish Times 16.12.2014) John FitzGerald repeats the traditional economic arguments in favour of immigration. It is a good summary of the economic arguments used by the pro-immigration lobby. It also a good example of wishful thinking - people who oppose racism and discrimination seem to feel the need to oppose further restrictions on immigration. They also seem to assume that if it can be shown that there are economic benefits from immigration, this in itself means that immigration is desirable. The argument is thus a mixture of explicit ‘economic’ arguments and usually implicit sociological arguments and, even more deeply buried, moral arguments.

In this series of four blog posts, I try to sort out this muddle. Firstly, I suggest that the arguments for immigration are not so straightforward as FitzGerald claims, largely because it often depends on who actually benefits. Secondly I suggest that FitzGerald’s claims about the benefits of ‘diversity’ are actually extremely dubious. Finally, I argue that there really are moral arguments in favour of certain forms of immigration, and these should be made as such. There are occasions when we should argue for more immigration, without making spurious claims about its benefits.

1. Background: Mass immigration is not inevitable
It is sometimes assumed that mass immigration is inevitable. This is nonsense: immigration rates vary enormously between countries and over time. One major determinant is simply immigration policy – governments can influence immigration rates (emigration rates in a free country are more difficult). The UK case demonstrates this point. The UK had essentially zero or even negative net migration for much of the post WW2 period (emigration exceeded immigration). Indeed, as late as 1994 emigration from the UK exceeded immigration. The major change has been the dramatic increase in immigration post 1996, the result of a series of decisions by New Labour, in particular:
1) relaxing restrictions on family reunification, generating increased flow especially from Pakistan and Bangladesh;
2) facilitating foreign students who worked part-time; and above all
3) along with Ireland and Sweden, opening the labour market immediately to immigration from the New Member States of the EU.

According to most recent figures from the ONS, in the year ending June 2014 net immigration was 260,000, having peaked at fully 320,000 for year ending June 2005 (ONS release, 27 November 2014). These notes therefore refer largely to the UK case.

2. The economic arguments are more complex
Problem #1 An increase in total GDP does not necessarily benefit natives at all
It is sometimes claimed that immigration increases GDP and is therefore ipso facto beneficial. To the extent that any immigrants work in the market economy, this is obvious but surely irrelevant. For members of the existing society the economy of which is being discussed, what matters is not the total size of the economy (the aggregate GDP) but the GDP per capita:

Overall GDP, which the Government has persistently emphasised, is an irrelevant and misleading criterion for assessing the economic impacts of immigration on the UK. The total size of an economy is not an index of prosperity. The focus of analysis should rather be on the effects of immigration on income per head of the resident population (House of Lords 2008: abstract).

Most of FitzGerald’s article does explicitly focus on the ‘per-capita’ increase in GDP and this is what needs discussing.

Problem #2 A gain in average (per capita) GDP may be unequally distributed: the better off may gain, but the poor may actually loose
An increase in the mathematical average GDP does not mean that everyone benefits to the same extent. Some may benefit more than others, and for some there may even be losses. So what is crucial is the distributional consequences.

It is quite clear that the early mass immigration to Western Europe in the ‘trente glorieuse’ of the long post World War II ‘Fordist’ boom occurred when overall living standards were rising and income differentials were narrowing. Whether immigration accentuated or retarded these processes could be discussed, but what really matters is that the situation today is very different. Since the 1970s income inequality has been increasing in virtually all western societies (see for one overview OECD 2011). This is the crucial context for discussion of the economic consequences of immigration.

Initially at least, immigrants compare their wages and their working conditions with what they could have received ‘at home’ rather than with wages and conditions of native workers in their new country. Furthermore, they often see their stay in the new country as temporary, so issues like job security, promotion etc. are of less importance. This ‘dual frame of reference’ (Waldinger and Lichter, 2003) means that immigrants are prepared to accept wages and conditions that might be unacceptable to natives (Ruhs and Anderson, 2010). For this reason, employers often see new immigrants as ‘good workers’ and contrast their ‘work ethic’ favourably to that of the natives (MacKenzie and Forde 2009; Gomberg-Munoz 2010).

In the current context the mass immigration of unskilled workers (or more precisely, of immigrants taking unskilled jobs) undermines wages at the bottom of the income distribution. In the short term this form of mass immigration will, ceteris paribus, therefore make the society more unequal [1].

It may be the case that that over time there is both an overall increase in GDP and that enough of this increase is at the bottom of the income distribution to cancel out the natives’ loss of earnings, but even this does not necessarily cancel out the impact on inequality. In other words, it is possible that everyone may end up better off, but the gap between poor and rich has nonetheless increased. However, in the current context, even this seems unlikely. In the UK and the USA in particular, low paid unskilled work has expanded.

In the UK there are now forms of agricultural production (e.g. strawberry growing) which would be impossible without the new low wage immigrant labour force (Ruhs and Anderson 2010). In the US immigrants fill most manual jobs in agriculture and sections of the food processing industry. Employers – sometimes joined by immigration advocates - argue that American food production can only occur if there is (cheap) immigrant labour. Yet as one researcher comments: ‘History is littered with predictions that there are no alternatives to slaves or guest workers to produce food and fiber’ (Martin 2009: 139). At the most extreme, the new availability of cheap immigrant labour with limited labour rights allows technological regression: the (re-)introduction of less sophisticated technologies, so that the workplaces begin to resemble the work camps of the soviet gulag (Wickham 2011) [2].

Any empirical study of the impact of immigration that focuses only on the impact on average wages will be misleading. Thus an initial review concluded that:

The available evidence suggests that immigration has had a small negative impact on the lowest-paid workers in the UK, and a small positive impact on the earnings of higher-paid workers. Resident workers whose wages have been adversely affected by immigration are likely to include a significant proportion of previous immigrants and workers from ethnic minority groups (House of Lords 2008: 28)

More recent empirical work in economics has disaggregated the impact of immigration and clearly suggests that immigration has different impacts on different groups of wage and salary earners. For the UK, Dustmann et al (2013) show that in the period 1997-2005 immigration led to an overall increase in average wages, but had a negative impact on wages of the lower paid. As they say, immigration has a ‘sizeable negative impact of immigration on the lower wage quantiles’ (160) and again ‘Overall, these results suggest that immigration tends to stretch the wage distribution, particularly below the median' (161). Equally, a recent report by the UK Migration Advisory Committee concludes that ‘Wages for the low-paid may be lowered as a result of [low skill] migration, although…this effect is moderate at the national level but possibly larger in London’ (MAC 2014: 31f).

The claim that immigration brings economic benefits to everybody is therefore no longer universally accepted. Although there is considerable debate in the economics literature, it is clear that those who benefit least from the immigration of more low paid workers are those (including previous immigrants) who are already in low paid jobs.

Problem #3 The fiscal benefits of immigration depend on continuing immigration
New immigrants are almost bound to make a net contribution to the state’s finances (the ‘fiscal benefit’). Because of their age they are more likely to be at work than the host population, so they contribute more in taxes, while they are make fewer claims on the health and other social services (it is those past retirement age that are a bigger ‘burden’).

While this has always been so, now new immigrants bring a further benefit: overall immigrants of working age are better educated than the host population (OECD 2008). Because of their education they are likely to be more productive than the host population and so overall make a bigger economic contribution; because they are better paid they contribute more in taxes and place less demand on social services (fiscal benefit). While the education/productivity argument is often disputed (especially because it assumes that the immigrants are employed at their qualification level), the age/employment argument is universally accepted (see for the UK especially Lisenkova et al, 2013) [3]. In this vein, the recent and frequently cited paper by Dustmann and Fratini (2014) reports that recent immigrants to the UK have been net financial contributors.

However, these gains depend on a continuing inflow of new immigrants. If they stay, immigrants will get older and place their own demands on health and other services. The gains only continue if new young and healthy immigrants arrive and take up jobs. To the extent that older and longer resident immigrants are concentrated in low skill jobs, they are disproportionately at risk of unemployment; some minority ethnic groups also have lower levels of labour market participation than the native population. Thus Dustmann and Fratini also show that those born outside the UK who arrived earlier (‘non-recent immigrants’) are now - like the native born population - a net fiscal charge (Dustmann and Fratini 2014; also MAC 2014:31).

NOTES
[1] There are possible countervailing factors. For example, during the Irish ‘Celtic Tiger’ boom mass immigration was not restricted to low-skilled jobs. Many immigrants, especially returning Irish emigrants, took well paid jobs, so reducing wage pressure in this area.
[2] The comparison is not fanciful. Ethnographic accounts of Polish workers in low wage British manufacturing report the workers themselves saying ‘This is like a work camp’ – and work camp for Poles has fairly obvious historical connotations (e.g., Wilczek 2012).
[3] Obviously different immigrant groups will make different levels of contribution and incur different levels of cost. A crucial issue is the extent to which immigrants are actually employed and in what sort of jobs. In the UK A10 immigrants have a high level of employment but are concentrated in low skilled – and low paid – jobs. Recent immigrants in low skilled jobs are also especially likely to send remittances home.

Click here for Part 2