In the United States and Europe alike, immigration policy isn't working – and the failure is most evident at the crossing-points of the rich and poor worlds, from the Mexican border to the Canary Islands, says Saskia Sassen.

It may look like one step forward and two back, but the European Union has actually accumulated a series of innovations that move it towards governing, rather than controlling, immigration inside the EU. This move towards governing is gaining strength even as national governments in the EU continue to speak the language of control.

At the same time, strengthening control is what the European Union is gearing up to do when it comes to immigration from outside its borders. Its first effort was to construct the equivalent of the Maginot line on its southern and eastern perimeter. It did not fare much better than that older defence line. Now it is moving towards the construction of a sort of Berlin wall across the Mediterranean and into the Atlantic – with the Islas Canarias (Canary Islands) the emblematic object of control-lust. History has given its verdict on the Berlin wall – and it was built on solid ground. The EU is trying to construct a Berlin wall on water.

But the difference is actually not where the limits of walls make themselves visible. The wall and the weaponised border function in a vaster ecology. That larger ecology helps explain the failures of government attempts to stop unauthorised migration via border controls. It might be argued that the Berlin wall "worked" on its own terms. True enough, but those terms were far too narrow to keep a larger ecology from making it unworkable in the long term. It was not US tanks that brought the wall down. It was that larger ecology.

Another common argument is that militarising the border in more sophisticated ways by using new surveillance technologies will work if a full commitment to the effort is made. Well, fifteen years of precisely such an effort has been made at the United States-Mexico border; today, it is the most militarised border in the world between two countries not at war.

This example shows how an exclusive focus on border control fails to do the job of controlling the border. The results of weaponising the US-Mexico border might, then, carry clues as to the limits of current EU efforts to move from governing immigration into the EU to controlling it, including a possible militarisation at the EU perimeter. There might be more promising alternatives to these projects, but they would be centred in inventing the administrative, legal, and political instruments for governing rather than merely controlling entry.

Saskia Sassen is professor in the department of sociology at the University of Chicago and at the London School of Economics. Her latest book is Territory, Authority, and Rights: From Medieval to Global Assemblages (Princeton University Press, 2006), based on a five-year project on governance and accountability in a global economy

Also by Saskia Sassen in openDemocracy:

-"A universal harm: making criminals of migrants"
(August 2003)
-"Fear and camouflage: the end of the liberal state?" – part of openDemocracy's worldwide symposium, "What does 2006 have in store?" (December 2005)
-"Free speech in the frontier-zone"
(February 2006)
-"A state of decay"
(May 2006)

The US-Mexico border: immigration as natural experiment

In comparison with the United States, the European Union remains far more concerned with the rights of immigrants and the unacceptability of people who are attempting to enter dying in the attempt. Yet it is still oriented overall towards controlling rather than governing immigration. This means leaving behind the serious and decades-long work the EU has done on governing migration inside the EU, often against its member-states's own insistence on unilateral control.

In the early 1990s, the US government began to escalate its effort to control the Mexico-US border (see Peter Andreas. "The Escalation of U.S. Immigration Control in the Post-NAFTA Era", Political Science Quarterly 113/4, 1998-89). This escalation has continued since then. Much attention has gone to the astounding array of control technologies and the vast military material deployed on that border. But after fifteen years of strengthening and weaponising border controls aimed at unauthorised immigration coming to the US from and through Mexico, what has been achieved is: a growth in the unauthorised immigrant population, a sharp increase of the cost of each arrest, and fewer arrests.

A first critical component in the escalation of border control was an increase in the annual budget of the Immigration and Nationality Service (INS) which rose from $200 million in 1996 to $1.6 billion in 2005. The number of border-patrol officers increased from around 2,500 in the early 1980s to around 12,000 today, making it the largest arms-bearing branch of the US government except for the military itself.

The most comprehensive study to date of the matter, Douglas S. Massey's Backfire at the Border: Why Enforcement without Legalization Cannot Stop Illegal Immigration reports two simple facts. First, there was a sharp increase in the costs per arrest and a sharp decline in the rate of apprehensions. Before 1992, the cost of making one arrest along the US-Mexico border stood at $300; by 2002, that cost had grown by 467% to $1,700 and the probability of apprehension had fallen to a forty-year low. In the 1980s, the probability that an undocumented migrant would be detained while crossing was 33%; by 2000 it was 10%, despite massive increases in spending on border enforcement.

Second, the escalation of border control has raised the risks and costs of illegal crossing, which in turn has changed a seasonal circulatory migration – with workers leaving their families behind – into family migration and long-term stays. The border study established that in the early 1980s, about half of all undocumented Mexicans returned home within a year of entry. By 2000 the rate of return migration stood at just 25%. Thus, the results were the opposite of what the government aimed at: border militarisation did not reduce the probability of illegal crossings on the US-Mexico border, and it raised the likelihood of a longer-term stay with families.

Billions of citizens' tax dollars spent on militarising the border have put apprehensions at an all-time low and forced unauthorised immigrants to stay longer than they want and bring their families even when they would rather not. These facts are overlooked by much of the media, by politicians, by "experts", and even by well-intentioned public intellectuals who want to help immigrants.

A three-part vacuum

There are three peculiar absences in the enforcement effort in the United States: they regard the visa-application process, workplace inspections, and control of visa overstayers. These are less relevant to the European Union effort to control entries at the perimeter of the EU. But they show clearly how misguided an immigration policy can be, and, perhaps more importantly, how the built-in interests of particular powerful actors can prevail.

First, it is now known that a high proportion of the undocumented are actually also lawful applicants who have crossed the border illegally because (often for family reasons) they cannot wait the ten years it can take the INS to process applications. Thus, speeding up INS processing would be a critical arena for the larger enforcement effort.

Second, while one administration after another since 1990 has seemingly been happy to triple the budget for buying military equipment to secure the US-Mexico border, such largesse has not been bestowed on workplace inspections. Only about 2% of the INS budget is devoted to enforcement of sanctions on employers; in any case, almost no sanctions have been imposed since the passing of this legislation as part of the Immigration Reform and Control Act in 1986.

Workplace inspections would be particularly feasible and effective with large corporate workplaces – agribusiness, meatpacking houses, poultry farms, and the Wal-Marts of this world. It is this type of firm that employs undocumented workers, and has enough staff to check on workers' documents. Such firms are ready-made cases for employers' sanctions. It has been remarked many times that all the deployment at the border stops at a certain geographic distance – the large farms close to the border employing all these workers are not inspected.

Third, little if anything has been done about visa overstayers. It is well known that an estimated 150,000 of the annual growth in the undocumented population over the last decade are people who enter with proper papers (student visas, tourist visas) and then simply stay on after the document expires. Thus, while the US militarises the border – to the great delight of armaments-makers who had no war to count on in the 1990s (Bosnia partly excluded) – only 20% of the work of INS inspectors focuses on visa overstayers.

These are the elements that begin to point to a larger ecology within which border control in the United States functions, an ecology that undermines the objectives of that border control. In the case of the European Union, that ecology will have very different features. But it can still contribute to alter the outcome that might be predicted.

The ecology of border control

The experience of border controls in the last two decades in the United States is a matter of history that goes way beyond party politics. The purpose of recalling it is not to draw parallels with the European Union, for the EU is foundationally different from the US in its political culture. Yet, some of the conditions described below may have equivalent emergent forms. More important is the notion itself of a larger ecology within which border controls function, the need to understand its features, and – in my perspective – to argue the necessity for governing rather than controlling immigration.

In the case of the US, at least part of the explanation for that larger ecology is a straightforward matter of policy. Immigration is a complex process, but much of the resort in attempting to control rather than govern it derives from flawed policy rather than this complexity. At the same time, some of what looks like failure from the perspective of controlling entry is actually delivering results that particular sectors inside the US want from immigration. The development of a reasonable immigration policy must begin by clarifying the record on these two issues.

There are three critical differences between the investments in border control since the early 1990s and the other three options (concerning workplace inspections, visa applications, and visa overstayers) discussed in the preceding section. They concern jobs, lobbies, and propaganda dressed as politics.

First, US governments, regardless of political party, have repeatedly shown a strong reluctance to allocate funds and create jobs to inspect workplaces – irrespective of the fact that the incidence of fatal workplace accidents is higher compared to all other advanced countries.

Second, the presence or absence of lobbies able to make one's case in Congress can make an enormous difference in US politics. Armament-makers and large corporate employers in agri-business, meatpacking, and other sectors known to employ significant numbers of undocumented workers have powerful lobbies. The inspectors and green-card processors employed by the INS, as well as large sectors of the US workforce, do not.

Third, the electoral and public-opinion machinery is well attuned – especially in visual-media terms – to weaponising a border. Hiring more INS inspectors and green-card processors just does not make for such good visual effects and soundbites.

It is also the case that immigration policy comprises many different, specialised regimes. Many of these regimes are working fine – visas for international business persons being one example. Further, many of the failures of immigration "control" actually work well for agri-business, meatpackers and countless others.

But there are costs attached to some of these "successes" of immigration policy: the higher costs of crossing the border irregularly, more deaths, and in the end, more desperate poverty among the undocumented families that can become the object of hate campaigns in their countries of residence.

The reality of the border

There is a strong contrast (possibly a contradiction) between the project of militarising border control and the reality of the border-zone. The Border Study reports these figures for 2004: 175,000 legal immigrants entered the US from Mexico, along with 3.8 million visitors for pleasure; 433,000 visitors for business; 118,000 temporary workers and dependents; 25,000 intra-company transferees and dependents; 21,000 students and dependents; 8,400 exchange visitors and dependents; and 6,200 traders and investors.

On the other hand, one million Americans live in Mexico, 19 million travel there each year as visitors, and US foreign direct investment in Mexico now totals $62 billion annually. Trade with Mexico grew by a factor of eight from 1986 to 2005.

More difficult to measure, but still very real, are the multiple cross-border networks connecting people from both sides of the border, which go beyond physical border crossings. Similarly, this social and economic reality underlines the impossibility of rounding up 12 million undocumented workers and deporting them. It may well be that the only way to handle the matter is with a general amnesty.

Furthermore, when it comes to specific immigration channels, the government has had no trouble designing workable policies (in relation to specialised visas for high-level professionals, for example). Nafta contains mini-migration policies that cover the cross-border movement and multi-year residencies of foreign professionals. This is not presented as immigration policy but rather submerged into each of the treaty's main chapters – on finance, specialised services, or telecommunications.

This obscures a critical feature of today's global economy: that cross-border trade and investment require mobile workers. By recoding this migration in the language of investment and trade, the built-in necessity for this migration is lost in the debate about immigration.

But the permeability is also there with lower-level workers, from cleaners and nannies to gardeners and restaurant workers, many of whom cross every day, or come for the workweek. The permeability of the border even with fairly low-wage workers can be illustrated through one particular case among many. The American Hospital Association reports large numbers of vacancies for nurses every year, and the US imports large numbers of nurses.

But there is another side to the story: every year nursing-education institutions in the US have to reject 150,000 students in the US who apply for nursing school. The reason is insufficient nursing teachers, in part due to the low pay of these teachers. The American Nurses Association, a professional trade association that represents 155,000 registered nurses, opposes the import of nurses from developing countries and calls for the expansion of teaching personnel.

These circumstances underline once again the fact that borders are institutions, and as institutions they are undergoing change and stress. This is clear in the tension between building a European Union-level economic space for capital, goods, and information flows, and the ongoing effort to control specific population flows. Further, the borders inside the EU are very different types of institutions than the borders at the perimeter.

In my research, I have tried to track the formation of a whole range of novel forms of bordering capabilities – involving types of controls that are not embedded in the notion of "borders" that is part of the historic process of nation-state formation. These are highly technical capacities which have multiple institutional locations beyond the "border" per se.

For instance, in global finance, there are multiple such bordering capacities (many inside financial institutions rather than in government offices) that have replaced the older traditional national borders. Free-trade agreements certainly open up countries, but they also multiply processes of certification at the point of production, which again is not the typical border.

When "borders" are looked at through these lenses, one can see a wide range of possibilities of how to move from control to governing borders. Can this approach be extended to immigration?

The sharpest and most developed example of these new bordering capabilities and people flows is at present the portable rights of the new transnational professional class, acquired through free-trade agreements, the International Monetary Fund and other such supranational institutions deeply engaged in global processes. There is no similar regime for working-class migrations. But I see that as one possible regime in the future – perhaps as part of a move to create flexible migration flows which, inter alia, would enable return and circular migrations.

Also in openDemocracy, a major debate on "People Flow" in Europe, and more analyses of migration around the world, including:
- Caroline Moorehead, "The peopling of London: how 'they' became 'we'" (24 April 2003)
- Nigel Harris, "Open borders: a future for Europe, migrants, and the world economy"
(12 June 2003)
- Ash Amin, "From ethnicity to empathy: a new idea of Europe"
(24 July 2003)
- Hank Heifetz, "Looking north: Mexicans in migration"
( 11 April 2006)
-Michele Wucker, "Don't get immigration wrong – again" (20 June 2006)

The larger picture

United States immigration policymakers have chosen to concentrate resources on particular aspects (the border with Mexico) and not on others (workplace violations). If these features of immigration policy are considered, rather than focusing merely on the presence of immigrants in the country, there are some clear winners but overall many losers.

The winners include arms manufacturers; large corporate employers in particular sectors of the economy that tend to employ significant numbers of undocumented workers; various lobbies; employers of undocumented immigrants generally (insofar as employers' sanctions are not seriously enforced); and the growing numbers of smugglers whose fees and business have expanded as government policies make border-crossing more difficult and risky. (The US border patrol with its increased numbers and weaponry might be included among the winners – although many patrol officers don't like what is happening).

The losers include citizens whose taxes are paying for a far larger and costlier border-control operation that is not even reducing illegal crossings (the intended policy outcome); the migrants themselves whose crossings have become far more difficult, dangerous, and sometimes deadly (as well as costly, given the greater need to use a smuggler); the INS inspectors who have not seen sharp increases in their numbers and resources to enforce employers' sanctions; and the INS's overworked and understaffed processing units.

If there is a contradiction in excessively concentrating on militarising the border while deciding to overlook workplace regulation, it is less between citizens and immigrants than between those with resources and those without – or those who have powerful lobbies versus those with none.

This overall tally might suggest that a peculiar mix of enforcement and laxity can be seen as good policy for those who can afford lobbies to fight for their interests and who can profit from control weaponry or armies of low-age workers (including underpaid technical specialists such as imported nurses); and as bad policy for low-wage workers themselves (not only undocumented workers, but also regular low-wage workers whose pay is stagnating, including INS workers). It has also not worked well for the larger societal project of creating new jobs.

But the problems go deeper. Some of the direct effects of border-control policy are unacceptable from normative perspectives – whether these are social-justice norms, human-rights norms, or religious values. A consequence of the militarised US-Mexico border is a rising number of deaths among illegal crossers; almost 500 were reported in 2005 alone. These are people in search of work to support their families.

Such deaths are unacceptable. They become even more tragic in view of the fact that there are other, better, simpler ways of governing migration. But these involve creating vast numbers of new jobs for workers to inspect workplaces and to process legal immigration applications. And they involve raising salaries of farm-workers, meatpackers, the cleaners of hotel chains, and many others. All of this would level the field among workers but also for governing immigration.

Instead, the current policy creates monstrous distortions – from using billions of tax dollars to buy products from weapons-procurers to the growing degradation of large numbers of low-wage workers who are legal immigrants and citizens. Relocating the sites for enforcement away from the border and the immigrant body and on to workplaces and employers would be a good start.

Beyond controlling immigration, beyond reducing unauthorised entry, is the question of governing immigration. Immigrants will keep coming. The rich countries will keep needing them: indeed under current fertility and legal immigration levels, the United States by the end of the 21st century will have almost 35 million fewer people; the equivalent estimated figure in the European Union is 70 million.

After all these issues are considered, there remains, at least for some of us, yet another step in the work of shaping an immigration policy that is fair and just to all involved. Beyond immigration policy and its possibilities lie questions of social justice and human rights. There is a need to forge a substantive conception of political membership for immigrants as well as for marginalised citizens and poor people in the countries of the rich world.