Paul Collier
Page 11
But if this is correct, then it introduces a further feedback mechanism into our model. Putnam finds that diversity reduces trust among the indigenous population: people hunker down. Translated into our framework, the larger the unabsorbed diaspora, the
lower its trust. But now we must add the feedback effect of this reduction in trust onto the rate at which the diaspora is absorbed.
The effect implies that the larger the diaspora is, the slower the
106 HOST SOCIETIES: WELCOME OR RESENTMENT?
rate of absorption. Absorption is reflected in the slope of the diaspora schedule; the slower the absorption is, the flatter it is. Building in this effect twists the schedule clockwise. I show the possible implications in Figure 3.2.
In the first panel the implication is a larger diaspora and a faster rate of migration. In the second, there is no longer a natural equilibrium: without migration controls, the diaspora and the rate of migration keep increasing. In the final panel the feedback effects of the diaspora on trust and back from trust onto absorption are sufficiently strong that beyond a certain size of diaspora, the number of people absorbed from it actually falls. If this happens, then there is a ceiling to the rate of migration. If migration controls exceed this ceiling, the diaspora keeps on expanding indefinitely.
Absorption and Host-Government Policies
The policies adopted by the host country government can, to an extent, affect the attitudes of both the indigenous population and migrants. Where multiculturalism defined as the maintenance of distinct migrant cultures is official policy, culturally specific social networks among immigrants are accepted and encouraged. Diasporas can become concentrated in a few cities, and some of the schools in these cities may have an overwhelmingly diasporic intake. The DS
DS
DS
DS’
M’
M’
on
on
on
M’
igrati
igrati
DS’
igrati
M
M
M –
M
DS’
Rate of
Rate of
Rate of M
M
M
Size of Diaspora
Size of Diaspora
Size of Diaspora
Figure 3.2 Trust and the Absorption of the Diaspora
THE SOCIAL CONSEQUENCES 107
encouragement of single-ethnicity immigrant schools would have been viewed with a horrified incredulity by the progressives
who promoted busing policies for American schoolchildren in
the 1960s.
However, while multicultural policies permit and encourage
immigrant groups to preserve their cultural and social distinctive-ness, policy toward the indigenous population is necessarily different. Well-founded fears of the potential and reality of anti-immigrant discrimination require government strongly to oppose the equivalent networks among the indigenous population. Prior to immigration, the social networks that exist in a country are, inevitably, exclusively indigenous. Antidiscrimination policy essentially for-bids such networks: quite properly, they have to become inclusive.
Recent research by Ruud Koopmans finds that the rate of
integration is indeed affected by these policy choices. 35 Integration is slower with multicultural policies. Multicultural policies have measurable effects such as a reduced aptitude of migrants in the national language, which we know reduces willingness to cooperate in public goods provision, and increased spatial segregation.
Koopmans also finds that generous welfare systems slow integration by tempting migrants into remaining at the bottom of the social ladder. Of course, they also tempt the indigenous population, but they appear to be more tempting for migrants because they are accustomed to radically lower living standards. Even the modest income provided by welfare systems appears attractive, and so the incentive to get a yet higher income by getting a job is weaker. Between them, multiculturalism and generous welfare systems slow integration at home and at work. On Koopmans’s
figures, both their effects are substantial.
It is easier to build social networks within groups—what Robert Putnam calls “bonding” social capital—than between them—“bridging”
108 HOST SOCIETIES: WELCOME OR RESENTMENT?
social capital. It is also easier to build social networks in small groups than in large. Hence, the conjunction of multiculturalism and antidiscrimination laws can inadvertently give rise to a paradox: immigrants may be better placed to build bonding social capital than the indigenous population. Immigrants are permitted and encouraged to form tightly knit communities that sustain their culture of origin. Indeed, the term “community” becomes routinely affixed to any people who have emigrated from the same country of origin: as in “Bangladeshi community” or “Somali community.” In contrast, by force of law all indigenous social networks are required to convert from bonding social capital to bridging. As a result, despite suffering the wrenching social upheaval of migration, the typical immigrant belongs to a denser social network than the typical indigene. Perhaps this is why Putnam finds that the indigenous population fragments. People are less bonded into social networks—they “hunker down,” in his phrase. The conjunction of policies of multicultural separatism applied to migrants, and antidiscrimination laws applied to the indigenous population, breaches the golden rule. One group receives treatment that cannot be conceded to the other. But quite evidently, the indigenous population cannot be permitted to maintain exclusive networks: here the integrationist agenda is essential.
The contrast between French and British policies toward immi-
grant cultural practices as exemplified by veil wearing illustrates the lack of coherence. Veil wearing quite literally destroys mutual regard. In France this was widely recognized as being incompatible with fraternity and so veil wearing was banned. The ban was supported by both communists and the mainstream Right. In Britain, while some politicians across the political spectrum lamented the increase in veil wearing, all parties defined it as an issue of liberty from government interference. But, as the French decision indicates, the liberty to destroy fraternity need not be considered a
THE SOCIAL CONSEQUENCES 109
Integrationist DS
Integrationist DS
Multicultural DS
M’
M’
Multicultural DS
Rate of Migration M
Rate of Migration M
Size of Diaspora
Size of Diaspora
Figure 3.3 Integrationist versus Multicultural Policies in Equilibrium human right. A consequence of these differences in policy choices is that although Britain has a much smaller Muslim population than France, veil wearing is increasingly common, whereas in France it is nonexistent.
Once again our model can be used to investigate how the choice between integrationist and multicultural policies eventually plays out if migration is allowed to accelerate. Their effect is to alter the absorption rate: integrationist policies raise it; multicultural policies lower it. The slower the rate of absorption, the flatter the diaspora schedule. Slower absorption can play out in two different ways: I show them in Figure 3.3. In the left-hand panel, by slowing absorption, multicultural policies eventually increase both the diaspora and the rate of migration. The right-hand panel depicts another possibility: slower absorption removes the possibility of equilibrium. In the absence of controls, the diaspora and migration both keep increasing.
You may be starting to see the scope for policy blunders. But first it is time to turn to the economic consequences of migration for host populations.
This page intentionally left blank
CHAPTER 4
The Economic Consequences
ECONOMICS PROVIDES TWO CLEAR PREDICTIONS about
the effects of immigration on host populations. Inevitably, t
hese predictions turn out to be too simplistic and sometimes utterly wrong, but before we get complicated it is well to start simple.
The economic well-being of indigenous households comes partly from private income and partly from government services. As to income, from the first principles of economics, the immigration of workers would be expected to reduce wages and increase the
returns on capital. As a result, indigenous workers would be worse off and indigenous wealth owners would be better off. As to
government-provided services, the existing stock of public capital—
schools, hospitals, roads—would be shared among more people
and so per capita provision would deteriorate. Poorer people receive more of their income from work and less from wealth and more of their overall well-being from government-provided services. Hence, the prediction from first principles of economics is that immigration
112 HOST SOCIETIES: WELCOME OR RESENTMENT?
benefits those indigenous people who are wealthy but makes poorer indigenous people worse off. In parody, this already oversimple analysis amounts to the assessment that the middle classes benefit from the archetypal immigrant staff such as cleaners and nannies, but that the working classes lose from competition with workers willing to accept lower pay and competition with immigrant families using social services.
Effects on Wages
It is time to turn to some evidence. Fortunately there is a highly credible new study of the effect of immigration on wages in Britain
covering the phase of high immigration. 1 The study investigated not just the average effect on wages, but the changes along the entire spectrum from high to low. It found that at the bottom of the wage spectrum immigration indeed reduces wages, as predicted by the elementary principles of economics. However, along the rest of the spectrum it increased them. Further, the increases were larger and more extensive than the reduction: most indigenous workers gained from migration. While the decrease in wages at the bottom of the spectrum is consistent with elementary economics, the increase higher up the range can only be explained by introducing effects that simple analysis ignores. The researchers themselves speculate that the fluidity introduced by immigrant workers improved the efficiency of the labor market—immigrants concentrated in the cities and niches with the greatest potential for new jobs—in other words the expanding service economy of South East England. The greater ease of expansion enabled by immigration helped entrepreneurs to increase productivity and so pay higher wages.
Another new study of the effects of immigration on the labor mar-
ket looks at evidence across Europe. 2 It also finds that immigration
THE ECONOMIC CONSEQUENCES 113
increases the wages of indigenous workers. However, the mechanism whereby this happens is itself revealing: on average in Europe, immigrants are more skilled than indigenous workers, although part of this is simply the churning of skilled workers around Europe.
Skilled workers complement unskilled workers rather than compete with them and so raise the productivity of the unskilled. Of course, this effect depends upon immigration being sufficiently selective to raise overall skill levels.
A standard question economists pose when there are both win-
ners and losers is whether the winners could afford fully to compensate the losers and still be better off. In respect of the impact on wages the better-off indigenous households gain much more
than the poorest lose and so could afford to compensate them.
However, what actually matters is not just whether compensation could be afforded but whether it actually happens. This returns us to our previous discussion of mutual regard and the willingness of the fortunate to help the less fortunate. While migration increases the need for such transfers, it may reduce the willingness to make them.
So the most likely effects of past migration on wages are that most indigenous workers end up gaining, while the poorest end up losing. While these effects are supportive of migration, they are all quite modest. The effects of migration on the wages of indigenous workers are trivial relative to the fuss that has been made about them. However, the empirical studies can only analyze the effects of migration within the observed range of variation. They tell us little about what would happen if migration continued to accelerate. For that, we would be safer to retreat to the economic first principles with which I started: the wages of most indigenous workers would drop considerably and remain
lower for many years.
114 HOST SOCIETIES: WELCOME OR RESENTMENT?
Effects on Housing
In high-income countries housing is the single most important asset, accounting for around a half of the entire stock of tangible assets. So in addition to the effect of migration on the flow of income from work, its effect on access to housing is potentially important for the economic well-being of the indigenous population. Evidently, through various routes, migrants increase the pressure on the housing stock.
Potentially the most important effect is that those migrants who arrive poor and have families compete with the indigenous poor for social housing. Because migrants tend to be poorer and have larger families than the indigenous population, they have atypically high needs for social housing, but meeting these needs inevitably crowds out the indigenous poor. While the effects on the wages of low-income indigenous workers are tiny, competition for social housing has been much more substantial: migrants are not only poor but they concentrate in a few poor neighborhoods. Even past levels of migration are likely to have had significant crowding-out effects. A continued acceleration would potentially seriously reduce the access of the indigenous poor to social housing.
Whether migrants should have distinctive rights to social housing is an active area of policy debate and an ethically tangled issue.
While migrants are needy relative to the indigenous population of their host society, they have already benefited from a massive windfall relative to how their needs would have been met in their society of origin. Meeting the additional tranche of needs provided by social housing requires a transfer from indigenous people who are themselves needy relative to the standards of the host society.
Social housing is not the only such rationed public good: a particularly acute conflict is in the classroom. The children of those immigrants who do not speak the indigenous language need extra
THE ECONOMIC CONSEQUENCES 115
attention, but so also do the underperforming children of the indigenous poor. Assiduous targeting of budgets could to an extent address this concern, but in practice teachers will be faced by hard choices as to which need takes priority. Universalist utilitarians will nevertheless conclude that since migrants are needier than the indigenous people whom they displace, overall global well-being has been enhanced. But others may conclude that since migrants were already fortunate to have gained a windfall in their private income, there is little reason to transfer a disproportionate share of social housing to them.
The principle of equal treatment of migrants and the indigenous can be applied either to groups or to individuals. If a certain percentage of the indigenous population is provided with rationed access to social housing, then equality of treatment of groups implies that migrants should be entitled to the same percentage, overriding differences in individual characteristics. This has indeed evolved to be the practice in some localities. The impetus has in part been a sense of fairness as perceived by the local indigenous population, and in part a practical concern for integration.
The ethical case against equal rights for groups is that to avoid second-class status each individual immigrant should have precisely the same rights as an individual indigenous person. If migrants are needier than the indigenous, then needs-based criteria should indeed grant greater access to excludable social housing. However, the second-class-citizen argument applied at the individual level itself has limits. As I discussed in chapter 3, the provision of public goods such as social housing depends upon the
maintenance of a myriad of cooperative games. Although citizenship is a legal concept, to have moral force it must be grounded in some deeper notion of mutual regard. Citizenship is neither primarily about entitlements to government benefits nor about obligations to respect the law: it
116 HOST SOCIETIES: WELCOME OR RESENTMENT?
is about attitudes toward others. The continued provision of public goods depends upon both migrants and the indigenous population adopting the same attitude of mutual regard as the indigenous population has toward its own members. If the preservation of cultural difference is regarded as an individual right despite its potential threat to public goods, there is a tension between this right to difference and an individual right to the social housing that the indigenous culture has made possible. Whether this principle of group-based rights is seen as ethically reasonable turns out to be a matter of considerable moment, and I return to it in chapter 6.
In addition to competition for social housing, as migrants become established they will compete in the private housing markets, driving up rents and house prices. A recent estimate for Britain by the Office of Budgetary Responsibility is that house prices are around 10
percent higher due to migration. Again this effect via housing looks to be much larger than the effects on wages. Since the housing stock is disproportionately owned by older and richer people, the appreciation in house prices due to migration has implied a large regressive transfer from lower-income groups. Further, since the migration has been spatially highly concentrated, it will have affected regions very differently. The 10 percent increase in national house prices due to migration breaks down into a negligible effect in much of the country, and very sizable price increases in London, the South East, and a few other pockets of high immigration. Paradoxically, by severely widening the north-south divide in house prices, this has made it more difficult to move from the other regions of Britain to the South East. Immigration has increased the ability of firms in growth areas to recruit workers, but it has inadvertently reduced the internal mobility of the indigenous workforce. This introduces a further mechanism of income loss for indigenous workers: they are crowded out of moving to better-paying jobs in growth areas.