Why Borders Cannot Be Open

David A. Coleman is professor of demography at the University of Oxford and adviser to the think tank MigrationWatch UK. This article is based on a speech given to the twenty-fourth General Population Conference, held in Brazil in August 2001.

Let us start with something on which we can all agree. That is, that some degree of international migration, in both directions, variable over time and space, is normal for any modern society in peacetime. In free societies, borders are never totally closed. Neither have they ever been in peacetime; the caricature of “Fortress Europe” exists only in the imagination of migration enthusiasts. No fortress lets out a million people per year, or lets in up to two million each year, as western Europe has done over the last decade.

People wish to move, for their own benefit, for all sorts of reasons: for better pay, to join a spouse or a parent, to enjoy more welfare or security. Receiving societies can benefit. For example, demand for labour may indeed be more easily satisfied if people can be recruited from abroad. At other times, however, the same country will export labour rather than import it. Most developed countries do both at once. That is natural in view of the ups and downs of economic cycles, international differences in economies and labour markets, and frictional forces in economic systems.

Whether migrants always benefit the population that is expected to receive them is quite another matter, however. The only unequivocal beneficiary of migration is the migrant. Whether their movement benefits the people in the country of destination all depends on circumstances. That is why borders cannot be fully open, just as in peacetime they cannot be fully closed.

Most legal international migrants both to the United States and to Europe in recent decades have not been labour migrants but spouses, dependants, relatives, students and asylum claimants, for whom demand is less clear. Many migrants enter illegally or on various short-term pretexts, seeking to stay and improve their position by any available means. And even the economic benefits of labour migration, once uncritically acclaimed, are less clear than they were. In some cases the consequences may well be negative when all costs are considered. Too easy an access to immigrant labour can create distortion and dependency in an economy. Any large modern state which finds that it in some way “needs” constant flows of immigrants, always inwards, over a long time, is suffering from problems with its society or labour market or economy which it ought to rectify by reforming itself, not by depending on the rest of the world.

More generally, the idea of fully open borders is hardly compatible with the principle of the nation-state or the coherent identity of its society, the existence of which depends on controlling its membership and its boundaries. The level of migration must be for each society and its electorate to decide, rather than some uniform international migration order—for social and political reasons even more than economic ones. In this respect it is like any other major measure affecting a society and its wellbeing. Let those who want open borders open theirs. Others who take a different view can then see what happens.

Free Trade in People?

“Open borders” are urged as a parallel to “free trade”, as though people were goods. But goods do not go where they are unwanted, goods have no rights or feelings, goods do not reproduce or vote, goods can be sent back or scrapped when no longer needed. Immigration concerns people, not objects, and consequently (depending on its scale) its political and social importance is potentially much greater than any economic effects it may have. The inability of economic models to accommodate political, social and cultural effects underlines the fact that immigration policy is far too important a matter to be left to economists.

All areas of human activity have safeguards and regulations because markets are imperfect. It would be a very harsh world without such protections. Absolutely free movement of people is no more possible than is absolute free trade. Trade is never free, only free-er, and “free trade” always depends on negotiated conditions. “Free markets” in goods are very far from a free-for-all. All trade and open markets are limited by conditions on labour, quality and safety in order to protect citizens, workers and customers: health and safety regulations, environmental regulations, consumer protection legislation. Throughout society, freedom of action is constrained by considerations of welfare and the rights of others in order to protect citizens from the imperfections of markets. That is why we have insurance and welfare arrangements to protect people from old age, illness, hazards and misfortune, and forces to maintain the law and security. Similar conditions must apply to migration to protect citizens and society in general from its adverse effects.

Liberal arguments in favour of uncontrolled migration tend to give first place to business interests and economic criteria, even though on other matters liberal critics are usually in the forefront of attempts to control the excesses of capitalism. In no other area of public policy are we urged to believe without doubting whatever business has revealed about its supposed labour needs, and to give it all it wants. Business interests, however, are short term. Easy immediate access to labour will always be preferred to the costs of training and capital investment for the longer term. In the nature of economic cycles, yesterday’s essential labour can often become, as the defunct factories and mills of Europe have shown, today’s unemployed. Employers who demanded immigrant labour are not held to account for this or required to contribute to the subsequent cost of their unemployed former workers. Few things are more permanent than a temporary worker from a poor country. If businesses were made responsible for the lifetime costs of their migrant labour in the same way as they must now deal with the lifetime environmental costs of their products, perhaps enthusiasm for labour migration might be moderated and make way for longer-term investment in capital-intensive restructuring.

Countries are, after all, not just a set of economic levers. We no longer accept the argument that “what is good for General Motors is good for the United States” in any other area of policy and should not in respect of migration either. Societies are not workshops where anyone can drop in, do some work if they like, or do nothing if they don’t like. They are, however loosely, communities of people with shared values, commitments, identities, origins and aspirations.

The Economics of a Borderless World

Let us set aside these broader issues for a moment and consider the economic arguments. According to elementary economic theory, uncontrolled migration is always beneficial because labour is then enabled to flow from countries with abundant cheap labour and little capital to high-wage areas where labour is scarce but capital plentiful. Free migration is expected to level the ratio of capital to labour everywhere, until an equilibrium is reached where wages have equalised and capital efficiency is maximised. Net migration then ceases. In the process, wage inflation has been checked, output maximised and global average income raised.

However these simple assumptions are seldom satisfied. Poor counties with population to spare greatly outweigh destination countries. Their populations, compared with those of the latter, are effectively infinitely large. The equalisation of wages expected from migration means lower wages in the receiving countries. Elementary political theory and practice tells us the wage reductions so welcome to economists and employers are distinctly unattractive to employees and electorates. Most migrants do not bring capital with them; moreover, many move for reasons little connected with the labour market. So instead, the votaries of migration now expend much effort assuring us that the theoretically “desirable” macro-economic deflationary consequences of migration (i.e., reduced wages) cannot actually arise, but that all can benefit from higher incomes. The latter argument is looking increasingly threadbare as evidence mounts that the effect of migration is divisive. Previous immigrants and the poorer sections of society suffer adverse consequences while the middle class may enjoy cheaper services from migrant labour.

The “segmented labour market”—that some jobs will not be done by locals and must be done by immigrants—provides another escape clause. However, one reason why locals find some jobs unattractive is because it is mostly immigrants who perform them. If employers can pay immigrant, not local wages, they thereby become dependent on perpetual immigrant labour, in some cases illegal. The concept of segmented labour markets finds little empirical support on a large scale. Where such segmentation does exist, it tends to be a function of excessively low wages, insufficient capitalisation, or excessive levels of employment protection in the regular economy, all running hand-in-hand with illegal immigrant labour. The suggestion that some unattractive jobs must in future be done by foreigners implies a permanent ethnically distinct underclass. That notion should be contrary to the principles of any society which favours equality of opportunity and opposes ethnic or racial discrimination.

Theory apart, what can empirical analyses of the economic effects of migration tell us? Quite often these analyses come to favourable conclusions about the economic consequences of large-scale migration, or at least indicate no serious deleterious effects. Benefits are particularly obvious in specific new growing sectors of the economy, usually involving high-skill professional or managerial workers. But the overall net effect of all migration is usually judged to be relatively weak compared with other economic factors. For example, a recent careful but limited British Home Office analysis of the fiscal effect of immigration in the United Kingdom concluded that the costs were £28.8 billion, the benefits £31.2 billion and the net benefit was approximately the difference between these two large numbers, namely £2.5 billion.¹ This marginal benefit, however, explicitly excluded various costs which are known to be higher than average for immigrants: the additional costs of education, crime and prisons, the fiscal drain of remittances and the sums (well over £1 billion) spent on asylum claimants. A recent Danish study distinguished between Western and non-Western immigrants, estimating a net cost for the latter.²

Whether net cost or benefit, the effects amount to only a small percentage of GDP either way, which must be discounted on a per caput basis by the increase in population occasioned by immigration. For example, even if the results of the UK study noted above are taken at face value, the £2.5 billion fiscal gain from immigration constitutes just 0.25 per cent of the 2000 GDP of £944.7 billion. Furthermore, this is actually less than the contribution of net immigration to UK population growth (0.31 per cent), so that per caput GDP is apparently made lower as a consequence of immigration. The US National Research Council concluded that all immigration (legal and illegal) added between $1 billion and $10 billion per year to a US economy growing at $400 billion per year, while of course also adding about 0.5 per cent per year to population growth.³ All studies point to the differential impact of immigration on the poorer and richer elements of the host society, and the benefits arising from highly educated immigrants compared to the costs of those with little education.

It seems perverse to imagine that the economic dynamism of the United States is somehow mostly due to immigration, as enthusiasts for migration to Europe have claimed. The reverse seems closer to the truth: immigration to the United States is high because it has one of the most dynamic economies in the world, with low levels of welfare and regulation, a high tolerance of inequality, plenty of space, and a relative openness to immigration (most of which is non-economic).

External Costs

On the whole, economic analysis considers only those factors which economists can easily measure. Externalities which may be difficult to measure, politically embarrassing or politically incorrect tend to be ignored: the British Home Office report cited above is a typical case. The more easily measurable elements are the level of income and taxes paid set against welfare, unemployment and pension costs. The more difficult elements include the costs of education (often complicated by language difficulties and novel cultural needs), and the costs of regenerating urban areas or constructing new buildings for immigrant populations.

In general in Europe, crime levels and public order problems are more severe among the ethnic minority populations of recent immigrant origin.⁴ This is particularly true as regards street crime and syndicates involved in drugs, prostitution and the trafficking of illegal immigrants, in which asylum claimants appear to figure prominently. Statistics on offences and incarceration make this reasonably clear, but the associated costs are never included in cost–benefit analyses. In the English-speaking world especially, a large and pervasive “race relations industry” has grown up, employing in the United Kingdom tens of thousands of people and consuming the time of many others. This industry is concerned with integration and retraining programmes, with equal opportunities and ethnic target enforcement. It involves ethnic monitoring and numerous legal proceedings. These activities, too, have seldom if ever been costed, yet the “race relations industry” is a counterpart of the even bigger “migration industry”.⁵

More strategically, migration distorts economies and creates dependence on further migration. It allows obsolete, low-wage, low-productivity enterprises to continue, which otherwise would have to raise their workers’ wages, introduce more capital-intensive processes or export the job to countries where it could be performed more cheaply for everyone’s benefit. For example, the textile mill towns and foundry towns of northern England, unmodernised and failing in the 1960s, were able to struggle on for a further decade before finally and inevitably closing, thanks to the availability of immigrant labour. Those towns now have substantial and fast-growing, badly integrated ethnic minority populations with high levels of unemployment and segregation; in 2001, they suffered serious race riots. Britain’s inadequate nationalised National Health Service has depended for decades, uniquely in Europe, on foreign doctors and nurses. Their availability has permitted it to survive as Europe’s most under-funded and inadequate health service, providing poor conditions for its staff and a poor service to its patients. Because of this dependence on medical personnel from overseas, UK medical training has fallen to a level quite inadequate to supply medical personnel from domestic resources, and necessary radical reform has been deferred.

A Mixed Record

In Europe, the long-term record of immigration is not very encouraging. High-skill workers recruited through work permits or moved by inter-company transfers bring undoubted benefits. Otherwise, the record of mass migration is very mixed. Most immigrants to Europe or the United States are not workers; this is true of up to 80 per cent of immigrants to Europe in recent years and of about 75 per cent to the United States. Many who do not enter as workers may nonetheless work, of course, but they do not do so as a result of any evaluation of the needs of the labour market and their skills are usually of a low level. Consequently, the level of unemployment of foreign populations (continental Europe) or ethnic minority populations (United Kingdom), in the second as well as in the first generation, is usually at least double the national average. Furthermore, workforce participation rates are typically lower than the national average: at the extreme, just 20 per cent and 30 per cent respectively among Bangladeshi and Pakistani women in the United Kingdom.⁶ Although there are a number of successful ethnic minority and foreign populations, in general their skill and workforce situation lead to a concentration of poverty and other negative measures among these groups. Traditional cultural norms—including large family sizes, non-working wives, lone-parent households and high teenage fertility—can make the solution of poverty more intractable.

Cultural diversity created by post-war immigration in formerly more homogeneous societies used to be regarded as a problem requiring a difficult process of acculturation or assimilation on the part of immigrants. Now, at least in the English-speaking world, multicultural policy requires such diversity to be (officially) “celebrated” as a permanent cultural asset to which the host society must adapt, although its benefits beyond a wider range of ethnic restaurants for the middle classes, and new kinds of pop music for youth, appear rather hard to specify. Critics of multicultural policy, however, claim that it helps to preserve the isolation and segregation of immigrant populations and to perpetuate new social divisions in Western society which may be associated with serious conflicts of interest and loyalty.

Problems of security have been highlighted since the atrocities of 11 September 2001, with the painful realisation from opinion surveys in some European countries that substantial proportions of immigrant populations, including the young of the second generation, do not side with their host society in recent conflicts and do not disapprove of the actions of the al-Qaeda terrorist group; a minority even applaud them. Even before 11 September threw problems of security and national solidarity into such sharp focus, mass migration had already imported into Western countries conflicts from other parts of the world, with separatist and revolutionary movements (Kurds against Turks, Sikhs against Indians) complicating domestic and foreign policy in new ways.

Immigrants as Demographic Salvation

What about the future labour force and the problems of population decline and population ageing in a developed world where birth rates are low and personal survival long? Surely it is convenient that immigration pressures should be so high just when, it is claimed, Europe is beginning to need more immigrants to rescue it from population ageing and preserve the “potential support ratio” of working-age people to pensioners, as a United Nations report has recently suggested?⁷

In fact, population ageing is inevitable; it has no “solution”, although with suitable reforms it can be managed provided the birth rate does not fall too low. Longer lives and fewer babies will eventually give all mankind an older population structure, forever. Today’s relatively favourable support ratios in the developed world are a passing inheritance of the twentieth century.⁸ To preserve the present ratio of working-age people to pensioners in the European Union (about 4:1) up to 2050 would require an average of 13 million additional immigrants per year every year to 2050, by which time the population would have more than tripled to over 1,228 million (about that of China), with a further doubling before 2100. Even a rise in births to the replacement level of about 2.1 children, a more desirable outcome which would not inflate population, could only restore the ratio to about 3:1. However, given a reasonable birth rate, much can be done to ameliorate the effects of an unavoidable decline in the support ratio by non-demographic means such as improving workforce participation, reforming pension schemes and productivity, and above all by moving retirement age gradually upwards.

Turning to the labour force, reductions in the size of the working-age population are projected for the medium term for most European countries, and much sooner for some. However, the European Union has the lowest level of employment of working-age people of any major economic bloc worldwide—scarcely 62 per cent. In Spain and Italy, only about 53 per cent of the working-age population is actually in work. In those countries, rigid job protection, high unemployment and high levels of illegal immigration have co-existed for years. Reforming Europe’s rigid labour markets, with their excessive levels of “social protection”, damaging pattern of early retirement and vulnerable pay-as-you-go pension schemes, is clearly politically very difficult. The future of the European Union depends on becoming more productive, and easy access to the short-term expedient of additional migrant labour will only delay and further complicate essential reform. Given the moderate rises in workforce participation expected by Eurostat, the European Union’s statistical information service, only Italy will show a smaller working-age population in 2020 than in 2000.⁹ After that, other measures will need to be taken. Changing conditions so that women will feel able to produce the number of children they consistently say they want (at least two) is the first priority on grounds of welfare, gender equity and the long-term future of the population.

Immigration can “preserve” population size if that is thought to be important; indeed, at the moment immigration is so high that it has promoted population growth in some European countries to a level they find undesirable, because of pressure on housing and land.

Migration’s Long-Term Effects

In the end, any level of net immigration into a country with below-replacement fertility will eventually replace the original population with one of immigrant origin. Even in the United States with its high fertility, white non-Hispanics are officially projected to become the minority shortly after 2050. Populations of immigrant origin constituting about 30 per cent of the national total and rising are projected before that date in Denmark and Germany. How far this ethnic replacement is thought to be a problem must reside with the electorates concerned. But on present trends that is the indicated outcome, and much sooner for some major cities with extensive settlement by immigrants.

Ultimately, we are all going to have to learn to do without the short-term expedient of large-scale migration and address instead the more difficult long-term solutions to problems. In retrospect, the present situation will come to be regarded as highly transient: a minority of receiving countries taking in surplus population from poor sending countries with high population growth may be typical of the first part of the twenty-first century, but it will not long remain so.¹⁰ Towards the end of the century populations of more and more countries are likely to stabilise or even decline and the world will start running out of potential migrants. We will then have to learn to live on limited labour supply just as we will have to learn to live on limited natural resources; we will have to husband our own demographic resources, and not depend on those from elsewhere. The closeness of the analogy to sustainable development is clear. The world will not forgive today’s developed countries, in the vanguard of the human experiment of living without population growth, if they throw away the chance to reform their societies and productivity because of a preference for the short-term convenience of immigration.

Migration has its place in every civilised state and can bring benefits to individuals and society as a whole. But like all other human activities it must not become a free-for-all. Migration must be kept subject to law, and organised or limited according to the democratically expressed preferences of the society it affects. Free movement of goods and free movement of people are not parallel cases. It is regrettable that some economic liberals appear incapable of seeing the difference between goods and people, and that some political liberals cannot acknowledge a distinction between citizens and foreigners. Those who promote large-scale migration seldom have to live with its consequences. Even Adam Smith admitted that people were the most difficult baggage to transport over borders. Only those in favour of unbridled capitalism, those who put the short-term interests of employers before anything else, those who cannot see that a society’s boundaries must be protected, can be in favour of “open borders”.

endnotes

1. Ceri Gott and Karl Johnston, “The Migrant Population in the UK: Fiscal Effects”, occasional paper 77, Research Development and Statistics Directorate, Home Office, London, 2002.

2. M. L. Schultz-Nielsen, The Integration of Non-Western Immigrants in a Scandinavian Labour Market: The Danish Experience (Copenhagen: Statistics Denmark, 2001).

3. James P. Smith and Barry Edmonston, eds., The New Americans: Economic, Demographic and Fiscal Effects of Immigration (Washington, D.C.: National Academy Press, 1997).

4. See D. J. Smith, “Race, Crime and Criminal Justice”, in The Oxford Handbook of Criminology, ed. Mike Maguire, Rod Morgan and Robert Reiner (Oxford: Clarendon Press, 1994), pp. 1041–117.

5. See J. Salt, “The Business of International Migration”, in International Migration into the 21st Century: Essays in Honour of Reginald Appleyard, ed. M. A. B. Siddique (Cheltenham, UK: Edward Elgar, 2001), pp. 86–108.

6. See F. Sly, T. Thair and A. Risdon, “Labour Market Participation of Ethnic Groups”, Labour Market Trends, December 1998, pp. 631–9.

7. United Nations Population Division, Department of Economic and Social Affairs, Replacement Migration: Is It a Solution to Declining and Ageing Populations? (New York: United Nations, 2000).

8. See D. A. Coleman, “Replacement Migration, or Why Everyone Is Going to Have to Live in Korea: A Fable for Our Times from the United Nations”, Philosophical Transactions of the Royal Society, series B 357 (2002), pp. 583–98.

9. See S. Feld, “Active Population Growth and Immigration Hypotheses in Western Europe”, European Journal of Population 16 (2000), pp. 3–40.

10. See W. Lutz, W. Sanderson and S. Scherbov, “The End of World Population Growth”, Nature 412 (August 2001), pp. 543–5.