182 THOSE LEFT BEHIND
ignore them or even turn them into scapegoats. A current likely instance of the safety-valve effect is Zimbabwe: a million Zimba-bweans have fled to South Africa because of gross mismanagement by the Mugabe regime. In South Africa they have little influence either on political developments back in Zimbabwe or on the attitude of the South African government and its people toward President Mugabe. Quite possibly, had they remained in Zimbabwe,
highly disaffected, vocal, and numerous, the regime’s forces of repression would not have been able to contain them.
So improved governance is critical to whether the people left behind achieve prosperity. Migration has both favorable effects on governance and adverse effects, so a reliable estimate of the net effect would be decisive for an overall assessment of the impact of migration on countries of origin. When I turned to the technical economic literature on migration, which is now extensive, I was astonished to discover that this particular issue was virtually terra incognita, so I set to work, trying to get reliable research answers. I must confess that I now understand why this particular question has attracted so little credible research: it verges on being unan-swerable given the current availability of data.
Here, in a nutshell, is the nature of the problems. Governance is a somewhat slippery concept. Within reason we know both good
and bad governance when we see it, but smallish changes are hard to measure. Although there are now several data sets that purport to measure it in various dimensions, very few provide long time series with comprehensive international coverage. Further, migration can feed back on the governance of countries of origin in a variety of ways with opposing effects, so it is insufficient to investigate only one or two of them; it is their totality that matters. But the most acute difficulty is to sort which is the chicken and which the egg.
While migration may affect the quality of governance, the quality of
THE POLITICAL CONSEQUENCES 183
governance most surely affects migration. A badly governed country is likely to experience a lot of emigration: people who cannot vote with a ballot slip vote with their feet. Compounding the potential for confusion, many characteristics of a society are liable to affect both migration and governance. A country is poor so people leave, and also government becomes more difficult. So a simple empirical association between migration and governance cannot be interpreted. Is migration causing governance to deteriorate, is bad governance driving people out, or is poverty causing both? Economics often encounters such situations, and in principle this one has solutions. But a solution to this problem depends upon finding something that clearly affects migration but is independent of governance. Unfortunately, so many things could potentially influence governance that in practice this approach has not yet generated convincing solutions.
Researchers have, however, recently made a beginning. There
are two broad ways to proceed: macro and micro. Macroanalysis depends upon looking at country-level data, investigating differences between countries and over time. Microanalysis depends
upon building ingenious experiments in which ordinary people
take part to investigate particular channels through which migration might have effects. In the end, the questions are macro, but the most reliable approach is currently micro.
The macroanalysis is in its infancy and may remain there. One long-established measure of governance is the degree to which a country is democratic, measured year by year for many countries.
The degree of democracy is a very crude metric for the quality of governance: incumbent rulers are often able to manipulate elections so that they meet the appearance of legitimacy without threatening their power. Or politics becomes so corrupted by money that voters face a pointless choice between rival crooks. China, which
184 THOSE LEFT BEHIND
avoids elections, is better governed than the Democratic Republic of the Congo, which, despite its name and its contested elections, is mired in corrupt and ineffective rule. But nevertheless, other things being equal, more democracy is likely to trump less democracy.
The analysis of democracy evidently suffers from the same problems as any other aspect of governance. But according to the best macroanalysis currently available the net effect of migration is ambiguous, depending upon its composition and the scale of the
brain drain. 2 The migration of the unskilled unambiguously makes countries somewhat more democratic. However, given the trends in the migration policies of high-income countries, the more pertinent issue is the migration of the skilled. Unfortunately, the emigration of the skilled has two potentially opposing effects on the pressure for democracy. Although migration builds a diaspora that brings external pressure, it may drain the stock of educated people.
This matters because the greater the proportion of the population that is educated, the stronger the pressure for democracy. Where the brain drain predominates, which unfortunately is what happens in most of the small, poor countries, although migrants bring external pressure for political reform, they have depleted the resident pressure brought by the educated. Research has as yet been unable to resolve this ambiguity: the macro approach leaves us in the dark.
The microanalysis is also in its infancy but is growing. To my knowledge the first serious experiment was by my colleagues Pedro Vicente and Catia Batista. Pedro works on governance and has chosen to do his fieldwork in two small formerly Portuguese island colonies, Cape Verde and São Tomé. I reported on some of Pedro’s ingenious work in The Plundered Planet. Meanwhile, his wife, Catia, was working on labor market effects of migration. I suggested that they might extend their marriage to the intellectual domain and try a partnership investigating the effect of migration on governance.
THE POLITICAL CONSEQUENCES 185
Indeed, Cape Verde would be the perfect location for a field experiment because it had the highest rate of emigration in Africa. Pedro and Catia duly transformed that suggestion into a revealing piece of
research.3 Their idea was to see whether the exposure to democratic ideas provided by migration increases the pressure for political accountability. By presenting people with a chance to lobby for better governance, they show that households with a migrant are more likely to take part. Cape Verde might be dismissed as just an exceptional little island, but the same transfer of political engagement via migrants back to home populations has recently been shown for Mexico. 4 How do migrants influence the political behavior of their
families back home? There is nothing very mysterious about this, but researchers are now onto it. During the Senegalese elections of 2012, Senegalese migrants living in the United States and France were surveyed. Through phone calls, daily or weekly, most of these migrants were urging their relatives to register to vote, and nearly
half were recommending whom to support. 5
While Pedro and Catia focused on the influence of migrants who were still abroad, complementary studies investigate the impact of migrants who have returned to their countries of origin. A particularly convincing new study is by my colleagues Lisa Chauvet and
Marion Mercier. Their choice of country was Mali. 6 Mali may seem to be the ultimate little country far away, but in 2012 it was thrust onto the front pages on the international press by a succession of increasingly catastrophic political events. In the last days of Colo-nel Gaddafi, the regime hired mercenaries from the nomads of
northern Mali. Libya had accumulated huge stockpiles of sophisticated, money-no-object weapons that these mercenaries were able to loot as the regime collapsed. While as mercenaries they had little interest in fighting for Gaddafi, back home in Mali they had long-standing grievances and aspirations to separatism: fancy weapons
186 THOSE LEFT BEHIND
were just what they wanted. All that stood between the rebels and power was the Malian army. The army was under democratic control: Mali was an established democracy. Indeed, it was sufficiently democratic that the incumbent president
had decided to retire: the rebel invasion coincided with the approach of an election and the withdrawal of the president into lame-duck inactivity. Mali was under the standard donor pressure to minimize its military expenditure, so while the rebels had all the technology that a deluded and oil-enriched military dictator had been able to lay his hands on, the Malian army was threadbare. The army lobbied the president to increase the military budget, but the president prevaricated.
Facing defeat in the field, the army mutinied and overthrew the government. Since Mali was promptly ostracized by the international community, this did nothing to improve the military situation, but it did plunge the society into political chaos as refugees flooded south to escape the rebels, and the coup leaders partially handed back power, but to whom? Meanwhile, the rebel movement was infiltrated and overpowered by incoming al-Qaeda fighters who smelled a promising opportunity to create a haven for terrorism. As I write, the French military has dramatically intervened at the request of the Malian regime and is pressing it to return power to civilians. So politics in Mali suddenly matters.
Lisa and Marion investigated whether the political exposure provided by a period of emigration affected political participation and electoral competition once migrants had returned home: specifi-cally, did people turn out to vote? They found three effects in ascending order of practical importance. Least important: returning migrants are significantly more likely to vote than nonmigrants.
More remarkable: this behavior gets copied by nonmigrants. Those living in the vicinity of migrants are also more likely to vote. This result is not dependent only upon whether people tell interviewers
THE POLITICAL CONSEQUENCES 187
that they voted. Economists are suspicious of self-reported information because it might be subject to bias. The higher voter turnout was there to see in the election returns. Now for the really remarkable: among nonmigrants, the ones most inclined to copy the
behavior of returning migrants were the less educated. This is really encouraging. Not only do return migrants bring back new norms of democratic participation learned in the high-income societies, but they are also catalysts for change among the uneducated, who are otherwise hardest to reach. Is Mali exceptional? A very recent study for Moldova finds the same result. 7 The latest research is also revealing that it matters where migrants have gained their exposure to foreign political norms. The better governed and more democratic the host society is, the more significant the transfer of the norms of democracy: France and the United States are better seedbeds than Russia and Africa.
This recent evidence is a skimpy basis on which to answer what is potentially the most important question on migration. Although migrants themselves do well from migration, it can only be truly significant in addressing hardcore global poverty if it accelerates transformation in countries of origin. In turn, that transformation is at base a political and social, rather than an economic, process. So the potential for migration to affect the political process for those left behind really matters. These studies provide straws in the wind.
Political values nest into a larger set of values about relations with other people in society that, as I discussed in part 2, differ markedly between host countries and countries of origin. On average, the social norms of high-income counties are more conducive to prosperity, and so in this restricted but important sense they are superior.
After all, it is the prospect of higher income that induces migration.
So do functional social norms diffuse back to countries of origin in the same way as norms of democratic political participation? A new
188 THOSE LEFT BEHIND
study of fertility choices finds precisely this result. Desired family size is one of the stark social differences between rich and poor societies. The experience of living in a high-income society not only reduces the preferred family size of migrants themselves, but feeds
back to the attitudes of those back home. 8 Evidently, this benign
process of norm transfer depends upon migrants themselves being sufficiently integrated into their host society to absorb the new norms in the first place.
While bad governments in countries of origin may appear to
deserve all they get from disaffected diasporas, not all diaspora pressure is for the good. Indeed, diasporas are often regarded by governments as hotbeds of extremist political opposition that fuel conflict. These fears are not entirely fanciful: diasporas are disproportionately drawn from ethnic minorities that have been oppressed in their countries of origin and harbor lingering resentments. At their worst, diasporas are seriously out of touch with present-day realities in their countries of origin but continue to nurse grievances over long-past conditions as badges of differentiated identity within their host society. They finance and otherwise encourage the most extreme elements within their country of origin as symbols of solidarity with their imagined identity. A disastrous instance of this phenomenon is the support provided by the Tamil diaspora in
North America and Europe for the Tamil Tiger separatist rebellion in Sri Lanka. This has most surely left Sri Lankan Tamils worse off than had the diaspora been quiescent. Nor is the existence of safe havens from bad governments unambiguously beneficial. The
regime of the Russian czars epitomized misgovernance, but the return of Lenin from safe haven in Switzerland frustrated what might otherwise have been a transition to democracy. Similarly, the return of Ayatollah Khomeini to Iran from safe haven in France scarcely ushered in an era of sweetness and light. While in such
THE POLITICAL CONSEQUENCES 189
extreme instances governments are right to fear diasporas, more commonly policies of discouragement appear to be based on little more than resentment at success. For example, Haiti, with its huge latent diaspora asset, has denied migrants the right to dual citizenship. Governments are only slowly waking up to the need to manage this asset as carefully as a conventional sovereign wealth fund.
The potential is far greater: while placing substantial financial capital abroad at negligible interest rates makes little sense for a poor country, it will inevitably have a huge stock of human capital abroad and so should plan to use it well.
The diaspora as an asset is of particular importance in postconflict situations following civil wars. Typically, civil wars last many years, during which the educated young get out. Political instability
and religious divisions both fuel migration. 9 Wealth also flees, as an
alternative to being destroyed. So by the postconflict stage, much of the society’s human and financial capital is abroad. The challenge is to get both back, and the two are linked: if people return, they are more likely to bring back their wealth in order to build homes and establish businesses. The severity of the skill shortage in postconflict situations is often alarming. For example, during the brutal rule of Idi Amin in Uganda in which around half a million people were killed, the educated were systematically targeted. One of the priorities postconflict was to restore higher education. A search of the Ugandan diaspora found forty-seven doctorates just in the South Pacific: one was persuaded to return to run the country’s first think tank.
As indicated by Uganda, the government of the country of origin has some scope to encourage return. There is also scope for the migration policy of host countries to facilitate postconflict recovery.
High-income countries have a manifest interest in the success of postconflict situations: in recent decades the costs of trying to shore
190 THOSE LEFT BEHIND
them up have been stupendous. Historically, close to half of them have reverted to violence, so if migration policy can be helpful, it is sensible to make it so. However, if restrictions on immigration from the country are tightened once peace has been restored, those who fled during conflict may be less inclined to risk return: will they be able to get back if necessary? The right time to adopt migration policies that would be helpful in postconflict situations is during the conflict. Both from the perspective of
the duty of rescue, and to help preserve the country’s human capital from violence, during conflict a migration policy needs to be exceptionally generous. The conventional criteria of skills and family need to be overruled by one based on human needs and human rights. However, the right of residence in the host country could be linked to the duration of the conflict. If rights of residence lapsed following the restoration of peace, migrants would be encouraged to be psychologically and socially better prepared for return: for example, they would send more remittances. As rights of residence in host countries expired, they would provide postconflict societies with a substantial influx of skill and money.
Does Emigration Increase the Supply of Good Leaders?
While one route by which migration might affect the quality of governance is through political pressure, another is through the supply of capable and well-motivated people. Small, poor societies
hemorrhage educated emigrants. 10 At the apex of public policy the vital talent could be drained away. However, it could instead be enhanced because a few key people come back, having gained
vital experience while abroad. By the nature of their chosen path, instances in which people leave their societies who would otherwise have become valued leaders are unknowable. My own favorite
THE POLITICAL CONSEQUENCES 191
plausible case concerns Tidjane Thiam, a former minister of economic development for Côte d’Ivoire who left following a coup.
Once in Britain he revealed truly exceptional talents by rising in the highly competitive world of international business to become CEO
of the largest insurance company in Europe.
But while such credible cases of serious leadership loss can be found, what is much more striking in small, poor countries is the disproportionate number of competent presidents, finance ministers, and governors of central banks who have had periods living in host countries, whether as students or longer-term residents. President Sirleaf of Liberia, winner of a Nobel Prize; President Condé, the first democratic president of Guinea; President Outtarra, the skilled technocrat who is restoring Côte d’Ivoire, and the highly respected Nigerian finance minister Dr. Okonjo-Iweala, are all current instances of countries of origin calling on valuable experience built up elsewhere. Overall, as of 1990, over two-thirds of the heads
Paul Collier Page 18
