A third step to a more flexible regulation of migration flows is the implementation of compensation mechanisms. The net positive effect of migration does not mean that everyone benefits from a more liberalised process. Just as when trade is opened up, workers, typically those with lower skills and little experience, face greater competition in the labour market with more immigration (OECD, 2004). But compensating losers is more difficult for immigration than it is for trade and capital (Felbermay and Kohler, 2007).
Beyond reasons of equity and fairness, compensating losers is essential to facilitate political support for policies aiming at fully benefiting from globalisation. But it is difficult to find a Pareto efficient equilibrium: a change in policy that makes at least one person better off and no one worse off. One possible way is through a redistribution of the gains derived from migration, for instance by subsidising directly the work of the lowest-skilled (Sinn, 2005).
The difficulty lies in defining the winners and losers. If by winners and losers, we mean having a migrant status vs. being locally born, the division can easily be made. But, if policy makers aim at compensating losers based on skill level, the distinction is blurry. Instead of skills, compensation should be based on income (Safi, 2010), as income can act as a proxy for skill level.
Several types of compensation have been discussed, but contain flaws in their efficiency in redistributing the benefits or in their implementation. Countries of origin benefit from welfare-improving effects such as remittances, skill-enhancing return migration, entrepreneurship and technology spillovers. There has been discussion on asking major immigrant source countries to offer concessions. These concessions can take the shape of more favourable trade relationships, such as less stringent quotas.
The problem is that this approach does not directly compensate the losers. Instead, the economy gets boosted in two ways: businesses import much needed labour and then enjoy trade concessions with other countries. While the overall gain may be large, it may be viewed as favouring the rich and neglecting those really losing from migration. This approach also discriminates against very poor countries and people, in effect eliminating them from the market and reducing their chances of reducing poverty through migration.
Another instrument proposed is instead allowing immigrants fully to enjoy the economic benefits of their labour but limiting their access to the welfare state, at least for a certain period of time. Immigrants with long-term tenure could then see access granted, as long as they pay taxes. Access could be given after a certain number of years, or gradually from year 1, according to the principle of selectively delayed integration (Richter, 2004). The downside here relates to human rights and minimum welfare. Disconnecting workers from the right to subsidised healthcare, day care, housing and child education can increase inequality and damage social cohesion. In addition, removing a significant amount of benefits may turn away too many immigrants willing to work in the country - a direct loss of needed labour.
Immigration countries can also anticipate problems and deal with them head-on: that is, by ensuring that competition in the labour market is not an issue beforehand, through an efficient balance of high and low-skilled migrants. One possible way is to improve bilateral matching systems between home and host country. Through a visa auction system, for instance, host countries can increase the selection process of immigration by ensuring that only those qualified, even low-skilled workers, enter the country at the right price - the proceeds being used to alleviate tax burdens or train those without work (Freeman, 2006; Richter, 2004).
The most logical instrument is probably a system of tax-based compensation (Hatton and Williamson, 2005). All other things being equal, immigrants would pay higher tax rates than native workers. The revenue collected would then compensate those affected by immigration. Not only is this ideal in terms of limiting the negative pressures of immigration, but the proceeds could also be used to alleviate the tax burden on native-born workers. Since a rising number of immigrants implies more public health and education expenditure (Boeri, 2010),3 the locally born might indeed feel concerned by the fiscal costs associated with higher immigration rates in welfare states. Another option is to tax instead employers (Hatton and Williamson, 2005).
In the end, the aim remains the same: to tax the benefits of immigration from those benefiting and use the proceeds to help those who lose or come under stress. Side payments have thus the potential to be Pareto efficient: immigrants still earn more than they would have in their home country, employers can still access much-needed labour (to a certain point) and local workers pay less tax. This can then be combined with incentives for immigrants to move to cities and regions within the country where labour is in most need.
The proceeds can go to training programmes to enhance the human capital of workers losing their jobs. This would imply efficiently pinpointing the sectors most affected and the type of workers and training needed in the market. An example of this mechanism is the extension of the U S Trade Adjustment Assistance Program to those losing their jobs because of immigration (Trachtman, 2009).4