MAKING MIGRATION POLICIES FIT FOR PURPOSE 265
“the impossible trinity”: a government that permits the free movement of capital and sets its own monetary policy cannot also set the exchange rate. In consequence, the free movement of capital has belatedly been recognized by the International Monetary Fund as inappropriate for some countries. There may, perhaps, be an equivalent impossible trinity arising from the free movement of people. It may prove unsustainable to combine rapid migration with multicultural policies that keep absorption rates low and welfare systems that are generous. The evidence pointing to such an impossible trinity is sketchy, but be wary of outraged dismissals: social scientists are not immune from systematically biased reasoning.
Legalizing Illegal Immigration
All controls inevitably induce evasion. Currently, those who successfully evade migration controls become illegal residents, and this illegality gives rise to serious problems such as crime and the black economy. Debates on what to do about illegal immigrants have
been as damagingly polarized as the larger migration debate. Social liberals want a one-off granting of full legal status; social conservatives oppose this on the grounds that rewarding evasion would encourage more of it. The result has been deadlock: nothing has been done and meanwhile illegal immigrants have accumulated: in America twelve million of them, in Britain nobody even knows. As I write, the Obama administration is beginning to wrestle with the problem.
The policy package offers an effective and straightforward
approach that meets the reasonable concerns of both camps but will presumably outrage the fundamentalists in both. To meet the reasonable concerns of social liberals, it recognizes that evasion is unavoidably a continuing process, so that future flows of illegal
266 RETHINKING MIGRATION POLICIES
immigrants need to be addressed as well as the accumulated stocks.
Any granting of rights that claims to be once-and-for-all is a piece of political deception. The package also recognizes that once border controls have been evaded, so that people have succeeded in entering the country illegally, all such migrants must be granted sufficient legal status to be able to work within the official economy.
Otherwise, illegal immigrants are a source of further illegality. To meet the reasonable concerns of social conservatives, it involves a penalty for evasion relative to legal entry, does not increase overall migration, and tightens the process for dealing with migrants who choose to remain illegal.
The approach is to maintain and indeed perhaps upgrade border controls, but to grant all those who despite these controls enter the country an initial status of guest workers. This status permits them to work and automatically places them in a queue to become permanent, fully legal immigrants. While guest workers, they would have an obligation to pay taxes but would not be entitled to social benefits: in using public services they would have the same rights as tourists. The slots to convert them into fully legal immigrants would count toward the overall ceiling on legal migration, so that illegal immigration would reduce legal migration rather than be supplementary to it. This would give the pro-migration lobby a strong incentive to support effective border controls. Finally, to strengthen the incentive to register, those illegal immigrants who chose not to do so would be subject to deportation without appeal
if detected. 6
Would such an approach dangerously increase the incentives for illegal migration? I think not. We can straightforwardly deduce that, despite the large stock of illegal migrants in many countries, existing controls are largely effective. The economic incentives to migrate
MAKING MIGRATION POLICIES FIT FOR PURPOSE 267
from poor countries are so substantial, and diasporas already sufficiently well established, that were the controls not effective, migration flows would have been far greater. Consequently, the flow of illegal migration is likely to be fairly insensitive to minor changes in incentives such as those I have proposed. The road to the status of a fully legal migrant would still be hard and long, typically requiring many years of taxation without benefits. If governments wanted to make the status of guest worker less attractive, those convicted of crimes could be subject to deportation without appeal. Would the proposed approach breach human rights? Only if the controls on migration themselves are judged to do so. If the controls are legitimate, then any policies that are forgiving of migrants who evade them are more humane than leaving them without any legal status.
How the Package Works
This package of ceilings, selection, integration, and legalization can be evaluated using our workhorse model. In may be worth flipping back to Figure 5.1, which depicts the political economy of panic that responds so damagingly to the initial absence of equilibrium.
Figure 12.1 starts from exactly the same position: as in Figure 5.1, there is initially no equilibrium.
But now the policies of ceilings combined with selective migration flatten the migration function, twisting it clockwise. Meanwhile, the policies of accelerated integration steepen the diaspora schedule, twisting it counterclockwise. As a result, the two lines now intersect: equilibrium is restored. With this package, migration initially accelerates but then stabilizes; similarly the diaspora initially grows but then stabilizes. The result of the package is superior to the political economy of panic in four important
268 RETHINKING MIGRATION POLICIES
M
Migration Policies
M’
DS’
Absorption Policies
f Migration
Rate o
DS
Size of Diaspora
Figure 12.1 The Political Economy of Selection and Integration respects. In the long run the migration-diaspora combination is better. Comparing Figures 12.1 and 5.1, for a common size of the diaspora in equilibrium, the rate of migration is higher, and conversely, for a common rate of migration, the size of the unabsorbed diaspora is smaller. Thus, the host society can choose to have both a higher rate of migration and a smaller diaspora. This is an improvement because the economic gains are generated by labor migration, while the social costs are generated by the unabsorbed diaspora. We also get to equilibrium rapidly, whereas the panic sequence might take a century.
Further, the path to equilibrium avoids a prolonged detour involving wild swings in both the rate of migration and the size of the diaspora.
Finally, the pool of illegal (and therefore unabsorbed) migrants that accumulates during the political economy of panic is entirely avoided.
MAKING MIGRATION POLICIES FIT FOR PURPOSE 269
Two lessons can be drawn from this straightforward application of the model. One is that the pertinent array of policies for successful migration is quite wide. If some desired ceiling on diversity is the objective, then the rate of absorption as well as migration matters.
The other is that appropriate policies need to be set early in the migration process with a view to the long term. Climate change is not the only policy that needs long-term thinking. In Britain, the Office of Budgetary Responsibility has recently put out an analysis suggesting that if Britain adopted a higher rate of net immigration, the per capita growth rate of GDP during the next three years might increase by around 0.3 percentage points. With due respect to the team that produced this forecast, it is categorically how not to think through migration policy.
How might such a policy package affect the groups that matter for migration?
There is no reason to expect that the migration rate that the policy package would generate would be ideal for those left behind in poor countries of origin. Indeed, we do not currently have the evidence even to estimate what such a rate of migration would be.
But we know that for many poor countries even the current rate is excessive: a somewhat slower rate of emigration would probably benefit them. It also seems likely that the savage reduction in migration that would be risked by the political economy of panic would switch it to being inadequate. Hence, since
selection and integration would result in a faster rate of migration than that, it would likely be an improvement from the perspective of the poorest societies.
From the perspective of the indigenous population of the host society, the policy package is considerably superior. The sustainable rate of migration is higher, enabling the economy to continue
270 RETHINKING MIGRATION POLICIES
to benefit from the modest economic gains thereby implied, and the social costs of an excessive, unabsorbed diaspora are avoided.
From the perspective of the existing stock of migrants, the political economy of panic is unattractive in both economic and social respects. In economic terms existing migrants are the big losers from further migration, and so during the anxiety phase of accelerating migration they would be squeezed by competition from new entrants. Socially, during the ugly phase of tightening restrictions and mounting social costs, they would be the ones at risk of xenophobia. The package of selection and integration does, however, place demands upon migrants: they are discouraged from remaining in a comfort zone of cultural separation. They are required to learn the indigenous language and send their children to integrated schools, and their right to bring in relatives is limited.
No migration policy can benefit everyone. In the package I have proposed the losers are those potential migrants who in the absence of the proposed policies would in the near future have migrated.
The policies of selection and integration indeed imply that the sustainable rate of migration is higher, so that would-be migrants benefit eventually, but the package avoids the phase during which migration temporarily exceeds that rate. Why is this justified?
Although potential migrants have interests like anyone else, there is no reason for their interest to trump those of others, which is what happens in the absence of a fit-for-purpose policy. The indigenous populations of host countries have a right to control entry, taking into account not only their own interest but also a sense of charity to others. But in exercising charity, their chief concern should be the vast group of poor people left behind in countries of origin, rather than the relatively tiny group of fortunate people who get dramatic increases in their income through being permitted to migrate.
MAKING MIGRATION POLICIES FIT FOR PURPOSE 271
Conclusion: Converging Economies, Diverging Societies
Migration is a large topic and this is a short book. But few areas of public policy are more in need of accessible and dispassionate analysis. I have attempted to shake the polarized positions: the hostility to migrants, tinged by xenophobia and racism, that is widespread among ordinary citizens, opposed by the contemptuous refrain
from the business and liberal elites, supported by social science academics, that open doors will continue to confer large benefits and are ethically imperative.
Mass international migration is a response to extreme global
inequality. As never before, young people in the poorest countries are aware of opportunities elsewhere. That inequality opened up over the past two centuries and will close during the coming century. Most developing countries are now rapidly converging on the high-income countries: this is the great story of our time. Mass migration is therefore not a permanent feature of globalization.
Quite the contrary, it is a temporary response to an ugly phase in which prosperity has not yet globalized. A century from now, the world will be far more integrated than now in respect to trade, information, and finance, but the net flow of migration will have diminished.
Although international migration responds to global inequality, it does not significantly change it. What is driving economic convergence is the transformation of the social models prevailing in poor societies. Gradually, their institutions are becoming more inclusive and less the preserve of extractive elites. Their economic narratives are shifting from the zero-sum mentality of grievance, to recognition of the scope for positive-sum cooperation. Loyalties are gradually expanding from clans to nations. Organizations are learning how to make workers more productive by combining scale with
272 RETHINKING MIGRATION POLICIES
motivation. These profound changes are being achieved through adapting global ideas to local contexts. As social models strengthen and economies grow, migration from rural poverty indeed matters, but the journey is to Lagos and Mumbai, not London and Madrid.
Yet although international migration is a transient sideshow to convergence, it may leave permanent legacies. One sure legacy that is unambiguously benign is that the high-income societies have become multiracial. Given their past history of racism, the revolution in sentiments consequent upon intermarriage and coexistence has been profoundly liberating for all concerned.
But in the absence of effective migration policies, migration will continue to accelerate, and this could imply other possible legacies.
The currently high-income countries could become postnational, multicultural societies. On the hopeful new view of multiculturalism propounded by Western elites, this would also be benign: such societies would be stimulating and prosperous. But the track record of culturally diverse societies is not so encouraging that this is the only possible outcome from an unlimited increase in diversity. In most societies for most of history high diversity has been a handicap. Even within modern Europe, the relatively modest cultural difference between Germans and Greeks has stretched to breaking the limited institutional harmonization achieved by the European Union. It is possible that permanently rising cultural diversity would gradually undermine mutual regard and that unabsorbed
diasporas would hang onto dysfunctional aspects of the social models that prevailed in their countries of origin at the time of migration. A further possible legacy of a continuing acceleration in migration is that small, poor countries like Haiti that can offer little to their most talented people would suffer an accelerating hemorrhage of capabilities: an exodus. They are already beyond the point at which emigration is beneficial. While the fortunate would leave,
MAKING MIGRATION POLICIES FIT FOR PURPOSE 273
those left behind might be unable to catch up with the rest of mankind.
Meanwhile, the emerging high-income societies are likely to become less multicultural. As part of the gradual transformation of their social models, identities will have enlarged from the fragmentation of clans to the unifying sense of the nation. In embracing the benign uses of nationalism, they will come to resemble the old high-income countries prior to migration.
Periodically, over the centuries the fortunes of societies have reversed. North America overtook Latin America; Europe overtook China. The financial crisis, with its source and effects in the high-income societies, has dented the smug complacency by which their citizens took economic superiority for granted. That most societies will catch up with the West is now accepted. But convergence may not be the end of the story. Singapore, which in 1950 was much poorer than Europe, is now much richer. If social models really are the fundamental determinants of prosperity, the rise of multiculturalism in one part of the world, coincident with its decline elsewhere, could have surprising implications.
As I finish this book I look up again at Karl Hellenschmidt. He was, before his time, the archetypical modern migrant. Leaving a small, poor village and a large, poor family, he reaped the modest rewards afforded to a low-skilled migrant in a high-income city. But my eye travels on to another photograph, to another man in middle age, who bears a family resemblance. I realize that he, not my grandfather, is the true role model for this book. Karl Hellenschmidt Jr. faced the habitual second-generation choice. Should he cling to an affectation of difference or embrace a new identity? He took the leap. Which is why you have just finished a book by Paul Collier, not Paul Hellenschmidt.
This page intentionally left blank
Notes
Chapter 1
1. Haidt ( 2012).
2. Benabou and Tirole ( 2011).
3. Wente ( 2012).
4. Dustmann et al. ( 2003).
Chapter 2
1. Besley and Persson ( 2011); Acemoglu and Robinson ( 2012).
2. Jones and Olken ( 2005).
3. Kay ( 2012).
4. In a brilliant new study, Timothy Besley and Marta Reynal-Querol (2012) show that in Africa remembered conflicts from as far back as the fifteenth century still cause violent conflict today.
5. Greif and Bates ( 1995).
6. Pinker ( 2011).
7. Akerlof and Kranton ( 2011).
8. Beatty and Pritchett ( 2012).
9. Beine et al. ( 2011).
276 NOTES TO PAGES 38–59
10. Carrington et al. ( 1996).
11. The Dunbar constant proposes that there is a ceiling of around 150 to the number of people with whom we can maintain a meaningful relationship (Dunbar 1992).
12. For example, in a particularly careful study of the Turkish and Serbian diasporas in Germany, Koczan ( 2013) shows that the higher the proportion of diaspora children is in the class, the more likely a diaspora child will grow up with a strong sense of diaspora identity.
13. By convention, this point at which the two axes of the diagram meet is termed its origin.
14. To see this, suppose for a moment that the absorption rate did not depend on the diaspora: for example, every year 2 percent of the diaspora merged into the mainstream population regardless of its size. In that case, if the diaspora doubled, the number of people absorbed into the mainstream would also double. With twice as many people flowing out of the diaspora, there would be room for twice as many migrants to flow in: doubling the diaspora would double the rate of migration that kept the diaspora stable. Visually, the diaspora schedule would be a straight line coming out of the corner of the diagram. Now suppose, more plausibly, that the absorption rate declines as the diaspora increases. If there are 30,000 Tongalese in New Zealand, there are sufficient interactions with other members of society to support an absorption rate of 2 percent, but if there are 60,000, the typical Tongalese has fewer interactions outside the group and so the rate of absorption falls to 1.5 percent. As a result, if the diaspora doubles, the number of people being absorbed from it less than doubles.
Paul Collier Page 26