In welfare states, the economic benefits from immigration may be eroded by the additional costs of government handouts and public services. This problem has been particularly evident with some refugee populations. For example, according to one estimate, just one in ten working-age Somalis in the UK is in full-time employment, while the vast majority are dependent on subsidised housing, much of it in high-cost London. Indeed, the welfare system may be largely to blame for such poor outcomes, due to the high effective marginal tax rates imposed on those moving into work.

Outcomes for EU migrants have tended to be more positive, with northern and central Europeans in particular exhibiting high employment rates. Nevertheless, the large numbers in low-paid work that pay relatively little tax may still impose significant costs, particularly households with children that receive child benefit, child tax credits, possibly housing benefit and/or social housing, childcare subsidies, and state education, which costs taxpayers around £5,000 per pupil.

In addition, there may be significant ‘externalities’ from migration, such as increased congestion on transport networks, crime and anti-social behaviour, disruption to settled communities and difficult-to-predict long-term effects on culture (both positive and negative). The strong tendency for BME groups to support the Labour Party is an example of a long-term impact on the UK’s political environment.

However, the costs identified are to a large extent not the consequence of immigration per se, but result from its juxtaposition with an extensive welfare state and government provision of services. In a truly free society, by contrast, support for low-income households and services such as education would be provided voluntarily. Taxpayers would not be forced to pay for them. At the same time, property owners, whether individually or in voluntary communities, would be far freer to decide how their land was used and would also enjoy freedom of association or dissociation. In other words, they could decide who would be allowed to enter their property, which, as well as residences and workplaces, might also include roads and other ‘public spaces’ currently owned by governments.

Various rules of entry could be adopted (see these case studies). For example, existing residents might vote on whether to allow individual applicants to move into their community. Alternatively, restricted covenants could require residents or workers to meet certain requirements before gaining entry. In some cases, simple rules of thumb might be used in order to minimise transaction costs. A completely open policy would of course be another option.

Rather than a one-size-fits-all policy imposed by government, different models would compete with one another, allowing market segmentation. Such voluntary associations could therefore be tailored to individual preferences. Cosmopolitans preferring a diverse cultural mix would be free to choose a community with an open approach. Conservatives placing a high value on their own traditions might prefer a model with far more restrictive rules of entry. The latter approach could prove popular with religious and cultural minorities wishing to preserve practices under threat from the influence of wider society.

Under such a system, property owners and voluntary associations would bear the lion’s share of the costs of their policies towards incomers. A market discovery process would ensue, with successful models attracting more business and unsuccessful ones declining. In this way, rules of entry would gradually evolve, tending to increase the benefits of migration and reduce the costs, while adjusting to changing conditions.

Contrast this with the one-size-fits-all policies imposed by governments. Politicians cannot obtain the knowledge required to set efficient quotas or entry requirements (such as points-based schemes or charging immigrants a fee), and such measures cannot be tailored to suit the wide variation of preferences on immigration. The transaction costs of state systems may also be high, with poor incentives to reduce them. Moreover, immigration policy will tend to be influenced by concentrated interests, for example ‘key’ sectors seeking special favours.

Despite the obvious flaws of immigration policies based on central planning by governments, the prospects for a voluntary system are slim. In the UK, there are very strict state controls over land-use, most transport infrastructure is government-owned and in both the UK and EU there are strict prohibitions on freedom of association/dissociation. Given the dominant political culture, it is difficult to envisage that these constraints will be removed in the near future.

This raises the question of which immigration policies should be adopted post-Brexit if a free-market model is off the table. The most straightforward way of increasing benefits and reducing costs would probably be to reduce substantially migrants’ entitlements to welfare benefits, social housing and government services such as childcare and education, while at the same time removing barriers to low-cost private provision, which eventually could be adopted by the whole population. In addition, market pricing could be introduced on transport networks to address congestion issues. Other things being equal, this approach would be likely to cut numbers significantly, while addressing directly the issue of costs imposed on taxpayers and pressure on public services.

It would avoid the central planning problems, special-interest capture and high administration burden of points-based rationing. But it would also contravene current European Economic Area rules on equal treatment, with implications for the deal between the UK and the EU. Nevertheless, because it would maintain freedom of movement, EU institutions and member states might consider it less objectionable than the alternatives.

Dr Richard Wellings is the IEA’s Deputy Academic and Research Director, as well as Head of Transport.