Migration and political reaction in Italy: The fortunes of the Northern League

Transcription

1 Migration and political reaction in Italy: The fortunes of the Northern League University of Colorado at Boulder Department of Political Science Paper prepared for presentation at the annual meeting of the Midwest Political Science Conference (Chicago, April 2-5, 2009) Abstract: This paper investigates the relationship between immigration and the political fortunes of the ethno-regionalist party Lega Nord (Northern League) in Italy in the elections of the last decade. The Northern League has made immigration of the main themes of its last political campaigns in Italy, by taking a strict anti-immigration position. First, this research will show how the League has occupied in Italy a political position usually reserved in Europe to nationalist parties; second, it will show how the party has exploited this position for political gain in reaction to recent waves of immigration by looking at electoral and demographic data of the last decade.

2 01. Introduction Italy has come a long way. Italian migrants provided much of the low-skilled labor force in the USA at the turn of the 19th century, and as late as the 1950s, in Germany. Now the country has experienced the same economic success and demographic change that have swept much of the developed world. Italy has become rich, literate, and old. The stereotypical image of the large Italian family, filled with many kids, exists only in the imagination of non-european foreigners. Italy now has one of the lowest birth rates in the world, and its society is aging rapidly. Those changes have also marked Italy s transition in the migration camp from sending country to receiving country. Immigration is a new phenomenon in Italy, and it brought to the forefront of the political debate all the issues that surround it in the developing countries. The inflow of foreigners in the country responded to the job opportunities that called for cheap labor in various sectors of the Italian market. Italy s improved education left many jobs open in the lower end of the market, both in the manufacturing and in the service sector. The country s aging population, coupled with its lackluster public service, opened a market for personal attendants for the elderly. Italy s tourism hotspots are filled with foreign (and mostly illegal) street vendors. Scenes of boat people trying to get to Italy, first from the coasts of Albania and now from Northern Africa, are a staple of the daily newscast in the summertime. This mounting tide of immigration, unprecedented by Italian standards, was bound to provoke a political reaction, and so it did. The immigration issue jumped to the forefront of the public s attention, first in the form of humanitarian rescue, because of the almost daily attempts of Albanian refugees to cross the sea to land in Italy, then in the form of concerns for citizens safety and unfair competition in the job market. Concerns about identity and integration followed suit very quickly. If the demographic situation related to the immigration issue in Italy is peculiar of modern times in Europe, the evolution of the politics of immigration has followed patterns that match other historical flows of migration around the world. That pattern has been familiar in many countries that became immigrants destinations, included the USA, an immigrants country par excellence. Newcomers have responded to job opportunities and demand (especially for low-skills, low-pay jobs). The inflow of migrants, especially the poorest and most different ones, caused a rejection of the society they moved in (see Stella 2006). The discontent with immigration then moved to politics, causing either the restriction of immigration rules (USA in the 1920s) or the rise of parties with an antiimmigration agenda (Fronte National in France or Lega Nord in Italy, for instance). The public does not seem to have any historical memory. Italian immigrants abroad experienced, in fact, much of what immigrants now face. One has only to read a few 2

3 articles about migrants at the turn of the 20th century to find a replica (in much harsher terms, for the sake of precision) of today s rhetoric (Stella 2006). The enlargement of the EU in 2004 and in 2007 has provided new fuel to the immigration debate and to antiimmigration feelings in Italy. Italy was amongst the EU countries that decided to put a moratorium on the free entry of new member states citizens in the country (with full access granted at the latest in 2011 and 2014 respectively). The inflow started nonetheless, and Romanian immigrants found themselves at the forefront of the immigration debate. This research wants to answer some classic questions about migration and migrants in Italy, and some new ones. How does migration influence politics in Italy? Are the electoral fortunes of the Lega Nord directly linked to the immigration phenomenon? Are all immigrants the same in that respect? Are all foreigners the same? My research is structured in the following way: section two will describe the situation of migration in Italy; section three will describe how the Northern League gained the center stage on the politics of migration in italy and it will introduce my theory; section four will describe my research design; section five will show my the results of my data analysis, both in my general sample and region-by-region, and finally section six will draw some conclusions and policy implications from my research Italy from sending to host country Italians have been on the move for centuries. The phenomenon of bel paese s citizens not moving more than a few miles from their birthplace is a relatively recent one. Since the middle ages, Italian merchants could be found in ports all over the world. Because of diminished weight of Italian commercial interests after the Renaissance, Italian international merchants left room to other Europeans for a few centuries. The presence of Italians abroad then revived with the demographic boom of the 19th Century. As Gian Antonio Stella (1996) reminds us, Italians from every region started to flood receiving countries everywhere in the world, from next door neighbors such as France to countries on the other side of the world, such as Australia and New Zealand. Italians were numerous, poor and illiterate. They kept migrating to neighboring countries until the 1960s, when the economic boom delivered what now is regarded as the country of La Dolce Vita. In the meantime, Italy s society started changing rapidly. Italians became prosperous, stopped moving even to other Italian cities, and birthrate dropped precipitously, leading one of the most prolific amongst Western Countries to attain the lowest fertility spots in the world. Since the beginning of the 1990s, Italy s fertility 3

4 rate has been consistently lower than other EU member states, sliding below 1.2 kids per couple in 1995 and climbing back to just below 1.4 in 2005 (Source: OECD). The Table below reports the last official figures of migration from Istat (The Italian National Institute for Statistics) and estimates from the 2008 Caritas Immigration Dossier. Both official figures and estimates refer to the end of [Table #1 about here] Since the 1990s, many factors have contributed to transform Italy from a sending to a host country. As we noted in the introduction, Italy s prosperity created the main determining factor: wage differentials. Italy now was an attractive destination for migrants in search of a fortune. With the increasing wealth of Italians came another change that made immigration more attractive: some jobs lost attractive for the locals, so immigrants were ready to take their place. Filipino migrants taking domestic personnel jobs in wealthy households were the first example of that trend, but many more professions were to follow: nurses, nannies, caretakers, etc. The particular condition of the Italian economy was also a determining factor. Italy has been slower than other EU countries to abandon the manufacturing sector. With the rise of outsourcing the third world countries, especially in the textile sector, the wage differential with Italy would have made Italian goods too expensive. In the pre-euro era, that was solved through the Lira s competitive devaluation. After the entry into the Euro area, that remedy was obviously out of the pool, so big producers started outsourcing their manufacture, while smaller ones had no such option. That created demand for immigrants that could take low-pay manufacturing jobs that the natives found less and less attractive (IStat, Rapporto Annuale - Yearly Report ). These flows, in practice, substituted internal migration, a.k.a. mobility. The highdensity production areas of the Center-North, especially in the NorthEast, offered higher wages than the South, historically poorer and depressed. That should have caused a flow of internal migrants from the South to satisfy the supply of jobs in the North, as it happened in the 1950s and 1960s, when hundreds of thousands of Southern Italians moved North to participate to the country s economic boom. That did not happen. Demographic and socio-economic factors prevented this internal re-balancement. On one hand, decreased fertility meant that the available supply of Italian workers was much scarcer than before. On the other hand, relocation to get a manufacturing job looked much less appealing to the South s youth, now much better educated than their older relatives. Plus, the differential for the cost of life, much higher in the North, would have eaten away much of the economic incentive for their relocation, given the inflexibility of wages in the peninsula. What s more, the enormous expansion of the public sector in the South was partially compensating for the lower supply of jobs in the private sector in the South (see Stella and Rizzo 2007). 4

5 That opened the way for migrants, both to the North and to the South. In the North, they were mainly filling job openings in small-to-medium firms in the manufacturing sector, whereas in the South, they were employed in seasonal jobs in the farming sector. Even though mobility is slightly on the rise as of recent (Istat, Rapporto Annuale - Yearly Report & 2006), that pattern helped rebalancing the Italian economy in the absence of incentives strong enough for the natives to move. All those factors together explain the transformation of Italy from a sending country to a host country for migrants. In the next section I will concentrate on the differences between the pre-eu enlargement and the post-eu enlargement era Pre- and post-eu enlargement The EU enlargement has opened new doors to immigration toward Italy, and with that, new issues, or at least old ones with a new packaging. Before the expansion of the EU, there was a clear distinction between intra-eu and extra-eu migrants. Even lexicon reflected that difference. Whereas EU migrants were simply referred as expatriates or just as citizens of their country of origin, extra-eu migrants were dubbed as extracomunitari in the popular lexicon (Ginsborg 2001). It is important to notice that this term in the popular usage, was utilized in a specific way to refer to migrants from poorer countries. For instance, an American citizen would not be an extracomunitario, even though she/he would legally belong to the category. A typical example of extracomunitari would be the Philippino community that moved to Italy to get jobs in the family services sector in the 1980s. Eastern Europeans, whose flow started from the early 1990s, were therefore included in the group of extracomunitari ; in a sense, legally and culturally alien to EU member states citizens. The 2004 enlargement re-shuffled the deck, enclosing many extracomunitari in the Union. As we remind the reader, Italy was amongst the countries that decided to put a moratorium on the application of full EU citizen s rights for new member states citizens, following the rule of the EU. That rule gives older members the possibility of assessing the progresses of migration from new member states and reviewing them first after two years, than after another three, and then at the end of a seven year period, new member citizens will enjoy full EU citizenship rights no matter what. Italy decided to follow the procedure too for the 2007 enlargement, which brought Bulgaria and Romania into the EU. The inflow of Eastern European citizens has been followed with apprehension on the Italian press (Lewis/BBC 2009a). Romanian citizens especially became frequent protagonist of criminal episodes and target of resentment, so much that future Prime Minister Silvio 5

6 Berlusconi in 2007 menaced to ban Romanian workers from Italy (The International Herald Tribune, November 4, 2007) and then, once in power, to call for expulsion of groups of migrants (The Economist, January 29, 2009). That gained the Italian government a reprimand from the EU, but the move was very popular amongst the Italian public (Lewis/BBC 2009c). A number of egregious criminal cases involving Romanian migrants made the first pages of Italian newspapers and the opening titles of the evening newscasts (Lewis/BBC 2009a). That has allowed the NL to steal the spotlight in advocating restrictions of rights of migrants and to push the now Prime Minister Berlusconi to take a tougher stance on migration. As it is easy to infer from the relentless coverage on the media, migration and migrants-related security issues got to the forefront of the political agenda in Italy, and the Northern League kept the pedal to the metal on the topic, even against the stance of the Catholic Church, which in Italy has taken a very soft approach on immigration (Corriere della Sera, January 15, 2009). The next part will describe the situation of the political far right in Italy and will highlight how the Northern League came to capture the political salience of the immigration issue The Far Right in Italy Traditionally, far right parties have led the charge against immigration, legal or illegal, in Western Europe. The Front Nationale in France, the British National Party in the UK, and so on, have taken the issue of open borders as a menace against national integrity, values and identity (see, amongst others, Ignazi 2005). Italy, in theory, should have one of the most extreme far-right parties, since its history traces back directly to the Partito Fascista Italiano, the Italian Fascist Party that ruled the peninsula from 1922 to In fact, Alleanza Nazionale (National Alliance) is the successor of the Movimento Sociale Italiano (MSI - Italian Social Movement), which was in practice the neo-fascist party in post- WWII Italy. To the casual observer, it would seem obvious to locate Alleanza Nazionale on the farthest spot on the right in the Italian political spectrum. That, however, is not the case. As Ignazi (2005) shows, that spot is now property of the Lega Nord (Northern League), which has taken the role that nationalistic parties usually play in the political arena of other European countries. Italy s nationalistic party, Alleanza Nazionale, has taken a much softer tone on immigration than the Lega Nord (Ignazi 2005). Under the leadership of Gianfranco Fini, AN has assumed the connotation of a classic European conservative party such as the 6

7 French Gaullists, proposing itself as a law-and-order party, but never proposing xenophobic stances in its political positions (Tarchi 2003). This is consistsent, actually, with all the post-fascist right, which has never adopted a xenophobic agenda even under the flag of the previous (and less moderate) incarnation of the party, the MSI. That is peculiar to the Italian situation and quite dissimilar to the rest of other European far-right parties (Veugelers & Chiarini 2002, Cole 2005, Ignazi 2005; for a review of the early 1990s far-right populist parties in Europe, see also Betz 1993). Even though AN voters have expressed xenophobic stances in surveys (1990, 1995, 2001), party cadres have kept the moderate stance that marked the evolution of the party since its change of name in In fact, though AN s leader Gianfranco Fini has signed a bill on stricter immigration regulation together with the Northern League leader Umberto Bossi in 2003, he also made the surprise proposal to provide regular migrants with the right to vote at local elections, and 70 percent of AN voters approved (Corriere della Sera, 1 December 2004). Ignazi (2005) concludes that [ ] the voters are far from xenophobic in all their stances and that some milieux of the party are slow in keeping pace with the modernization and openness promoted by the leadership. In recent outings, Gianfranco Fini, speaker of the Lower House (Camera) in the 2008-present Berlusconi government, condemned the association between criminality and immigration as hateful and declared that there is no alternative to integration, the future is going to be more and more multiethnic (La Repubblica, February 20, 2009). In sum, AN in Italy does not deal with the immigration issue in the typical fashion of the other nationalistic parties in the EU. The Lega Nord has taken this role in the Italian political arena. This fact is even more peculiar given that the Lega was born as a regionalist movement, which had as a negative target not immigrants, but the South of Italy in general (see also Spektorowski 2003). The Northern League has been the most successful amongst the ethno-regionalist movements in Europe, achieving both electoral success and participation to government coalitions in Italy. What s more, after a period of electoral decline in , its fortunes are on the rise again, on the wave of Italians discontent with the continuing waves of migrants, both from outside and from inside the recently enlarged EU. I will now briefly introduce the path of the Lega Nord (Northern League) in the Italian political arena so far The Rise of the League It is not a mystery that Italy has had deep regional differences for a long time. Besides the recent unification of the country, that dates back to 1870, the peninsula has a history 7

8 of deeply divided regions that were (and, largely, still are) different for wealth, culture, and even language, if dialect counts as such. It is noticeable that Italy did not have strong separatist movements with the exception of South Tirol, which was a post-wwi acquisition, and Sicily. Even, so Italian regions maintained their strong identity up to these days, even though it is not as strong as cities. The Northern League has given the political answer to deep geographical differences in Italy. The Northern League (Lega Nord) was born in 1991 through the alliance between the Lombard League, a rising regional party, and other six regional leagues. Since then, the League acquired national political ambitions, ambitions that were crowned with the League s participation to the first Berlusconi government in The electoral weight of the party remained concentrated in the North, and the party polled only a 8.4 percent of votes at the national level, nonetheless that allowed it to be essential for the formation of the government. The first political issue on which the League focused was the North-South divide. That has been ingrained in Italian politics for a long time under the name of questione meridionale (South issue). The League s message rode Northern voters discontent against the State under the banner Roma ladrona. The League wanted to provide much more fiscal and administrative autonomy to the regions of the North, in order for them to keep their money instead of sending it to Rome to be wasted for political patronage in the South. The judicial scandal of the early 1990s provided the political opportunity for the League to appropriate the electorate of the former Christian Democrats, in the North: workers and employees in small businesses; small entrepreneurs; non-regular practicing catholics; residents of small and medium-sized municipalities (Giordano 1999, Ignazi 2005). The existing regional leagues consolidated into the Northern League (Lega Nord) in 1991, under the leadership of Umberto Bossi, who has been so far instrumental to the success of the party in a larger arena (see Agnew et al. 2002, Ignazi 2005 and Agnew & Shin 2007). The change of the Italian electoral system to a single member district majoritarian rule favored the League, allowing it to conquer its strongholds in the North while practically ignoring Italian voters south of the Po river. The League entered the first Berlusconi government in 1994 and it brought it down shortly thereafter. That notwithstanding, the League re-entered the alliance with Berlusconi s Forza Italia and Fini s Alleanza Nazionale to join the second (2001) and the third (2008) Berlusconi government. It is now necessary to cast some light on the message of the Northern League, from a political and from a geographical standpoint. Further on, I will focus on the League s position on migration and its exploitation of migration s issues for political gain. 8

9 The League movement started out as a rejection of the geographical balance of power of Italian politics. The League(s) electorate was in origin part of the Christian Democrats electorate in the North. Those voters rejected the notion that the productive North should have supported and helped the underdeveloped South. If the North was the economic engine of the peninsula, why should have Northern citizens wasted their taxes to pay for patronage in the South? The original message of the Leagues, in fact, focused on the economic differences between Italian regions and how the State (mis)managed them. As Agnew (1995) pointed out, the League can be seen as representing an attempt at a clean start for Italy [ ] with its focus in the productivity of northern business (big and small) weighed down by corrupt (and Southern) Roman bureaucracy and Roman political parties. The Lombard League and its allies (which will merge in 1991) started as ethnic movements whose early goals were to preserve dialect and identify regional cultural differences (Agnew 1995). Following their electoral success, the various Leagues rose their political target and started talking explicitly about secession from the corrupt and inefficient South. They fabricated ex novo, in their rethoric, a regional entity called Padania, which would encompass the regions that touch the plain of the Po river, and stated that their goal was to split it from the Italian State. The achievement of national prominence brought first the consolidation of the movement into one single party in 1991, but also a progressive moderation of tones, necessary to enter the coalition of the first Berlusconi government. Since then, the argument about secession has left the main stage to the issues of institutional reforms that would give regions much more administrative and fiscal autonomy than before. Federalismo fiscale (fiscal federalism) then became the League s battle cry (Ignazi 2005). The League(s) took off also thanks to a charismatic leader, Umberto Bossi, who, as in many cases of populist movements, presented himself and his followers as antipoliticians. The League s leader, and Bossi in primis, devised a language that plugged into the popular discourse and put it in direct antagonism with traditional Roman politicians, which was the precise goal of the movement. We ll see that the League was and still is extremely effective at devising powerful messages to nail citizens perception of various issues, amongst which migration is prominent. In sum, there were two main themes determining the rise of the League s phenomenon: a political identity rooted in the territory and an anti-system stance and image (Giordano 2003). The emphasis on those two themes, as we ll see, helped the League bounce back to electoral prominence in 2008 after the disappointing electoral performances of 2001 and

10 [Figure #1: Northern League 2006 & 2008 election results about here] 03.3 The League & Migrants The League, both before and after its consolidation in a single movement, has always had a message deeply rooted in the territory. That does not mean that its message about the North s identity comes from centuries of common history that developed a strong sense of belonging to the Padania macro-region. As it happens many times for populist political movements, that identity is largely fabricated to serve political goals (see Agnew 1995, Giordano 2003 and Ignazi 2005). The opposition against the other, though, is not. The League has used this juxtaposition to determine and reinforce the identity of Northern Italian citizens (the League would omit the Italian part). As Spektorowski (2003) points out, the invention of the Padanian identity was designed to oppose three forces: a) the existing Roman bureaucracy and politicians; b) big capitalist interests and the welfare state and c) those who are different either southerners or immigrant workers, depicted as either welfare-mantained parasites or menaces to Northern values and culture. Now we ll focus on that last factor. The first other of the League was the Italian South, Rome included. The Center s Red Regions (Emilia-Romagna, Tuscany, Umbria, Abruzzi and Marche) were not included in the League s rhetoric against the unproductive South. The League(s) message was that the South was a dead weight for the productive North, since Roman politicians would use the North s taxes to maintain their political patronage in the South. As Poche (1992, cited in Spektorowski 2003) points out, Unlike other regional or separatist movements in Europe, the Lega does not protest against long-standing foreign rule but against what could be defined as the common interests of the Italian bureaucratic state, the political class, the underdeveloped South, immigrants and big business. The League has convinced Northern voters that the centralization of political power in Italy has harmed the interests of the North, effectively transferring its wealth to the parasitic and clientelistic South, through the leadership of Roman politicians (Spektorowski 2003). In the 1990s, until the country s entry in the Euro area, the League was a strong supporter of the European Union. Its leaders were envisioning a Europe of regions, where the Italian North could have left the dead weight of the South behind and join the productive regions of Central and Northern Europe. In that sense, being closer to Brussels meant being farther away from Rome (Agnew and Brusa 1999, Giordano and Roller 2001, Giordano 2003). After the entry of Italy in the EMU, the League started to change its attitude toward the Union, also because of the vacuity of the promise of the empowerment 10

11 of European regions. Bossi started attacking the EU as a Stalinist super-state. From the 2001 elections on, the Lega was able to position itself as the defender of the North against globalization and the European Union. (Betz 2002, as cited in Albertazzi and McDonnell 2005). This shift coincided with the accentuation of his rhetoric about the transferral of more and more powers to regions in Italy. That was necessary because virtually all other political party in Italy had embraced the argument of federalism: the League was becoming a victim of its own success. It also coincided with a radicalization of the party s discourse about immigrants (See for example, La Repubblica, August 9, 2008, : the NL-elected Euro MP Mauro Borghezio swore in Genoa to defend Christianity from Islamic profanation. ). Diamanti (La Repubblica, 12 October 2003) describes this shift as the transformation of the League from an innovative political movement (the issue of federalism, for instance) into one that primarily reflects and reproduces fear (ibid.). Albertazzi and McDonnell (2005) maintain that the League in Italy has become the mirror of the fear of the perceived effects of economic and cultural globalization: immigration, multinational corporations, secularism, international free trade, pension reform, multiculturalism, the blurring of identities, etc. (Ibid.) Following the poor electoral performances of the League from 1996 to 2001, They also state that the League s anti-immigration stance has alienated part of its traditional voting base: small and medium business owners, who need migrants and their low wages to keep their companies competitive in the market (Pastore 2004, Albertazzi and McDonnell 2005). The 2008 elections results seem to contradict that analysis. Apparently, the League has managed to take a tough stance on immigration, as its campaign and its distinctive posters show, while at the same time regaining its electorate of small and medium business owners in the North. The League is now the only openly anti-immigration Italian party, and its political position matches closely its electorate. In a 2001 ISPO (Institute for the Study of Public Opinion) survey, the League s electorate achieved the highest scores in several dimensions of hostile sentiments against migrants, with significant distances from the mean of Italian voters. Ignazi (2005) reports that League s voters were the ones that agreed the most when asked the question of being annoyed by the presence of too many immigrants : 67 percent of them agreed, compared with a national average of 42, and with Alleanza Nazionale s voters coming distant second with 51 percent and Forza Italia s third with 49 percent. In sum, another confirmation that the League is solidly in command of the xenophobic electorate in Italy, role that is traditionally in the hands of nationalistic parties elsewhere in Europe, but which Alleanza Nazionale does not play in Italy. If this is the case, the electoral fortunes of the League, all else being equal, should go hand in hand with migration. Also, since the League pushes hard on the issue of culture 11

12 homogeneity, they should be linked to the presence of non-european migrants such as Northern Africans or Chinese, rather than to new EU member states citizens such as Romanians or Bulgarians. We should therefore notice differences in the performances of the League according to the presence (or flow) of different groups of migrants. [Figure #2: NL Electoral Poster about here] In the 2008 election campaign, even more than before, the League has highlighted the alienness of migrants to Italian values and culture. Some of the electoral posters (see Figure #2) were explicitly designed to evoke strong feelings about immigration.we can consider the approach of the League toward migrants as twofold: on one side, the League opposes all kind of migration because outsiders are alien to Northern Italian culture and values; on the other side, while integration allows some migrants to blend into the social fabric of the territory, that process is much harder (according to the League) for some group of migrants, especially Muslims. The League s approach toward muslim migrants has always been hostile (see Allievi 2003, Ignazi 2005). A July 2008 fight about the (denied) authorization for the construction of a 3,000 seat mosque in the Milan area was just one of the many indications of the strategy of the League to oppose the penetration of alien Muslim elements into the Italian social fabric [Corriere della Sera, July 5, 2008]. The analysis of Muslim migration in Italy would confirm the high degree of alienness of migrants to Italian society. Allievi (2003) highlights several factors that make muslim immigration in Italy different than what happened in other European countries that already experienced flows of muslim migrants. Differently from other countries, Italy s muslim migrants do not come from formers colonies, so they do not share any residual cultural trait (e.g. Language) in common with Italians like, for instance, Pakistani muslims, most of which knew English before they moved to Britain. They do not come mostly from a single country like Germany s Turks, which makes it impossible for the Italian government to treat the issue as a foreign policy one. Also, right now Muslim migrants are coming from countries where Islam is culturally and socially on top of the agenda (Allievi 2003). Previous waves of Muslim migrants, 40, 30 or even 20 years ago, moved when cultural and political priorities wer Arab nationalism or socialism, and when political leaders were not openly religious. In a nutshell, the political climate of many sending countries veered toward Islam, and recent migrants reflect that changed cultural and political reality. Italy, therefore, is hosting much more pious migrants than it would have happened a few decades ago. We should also remember that Islamic-inspired terrorism has been in the forefront of news only in the last years, and the fears of Western public reflect that. 12

13 Therefore, on one hand we see migrants that are much more likely to be observantly religious than a few decades ago and on the other hand we have a public that is much more sensitive to the Islamic issue no matter whether its fears are justified or not. The next section will introduce my theory Theory Empirical observation of the flow of migrants in Italy and of the resurgence of the Northern League drive my theoretical expectations. The goal of this research is to show whether the rise in legal and illegal migrants, both from the new members of the EU and from third countries, has an effect on the electoral fortunes of an explicitly anti-immigration party such as the Northern League. Indirectly this is a test of the public s attitudes about immigration. It is, of course an indirect test because the Northern League s campaign was not only about migration, but the party was indeed the most explicit in its anti-migration message, whereas Alleanza Nazionale, for example, highlighted law & order issues without necessarily focusing on migrants. An explanation is necessary at this point. Those different attitudes of right-wing parties reflect their different geographical strength. Whereas Alleanza Nazionale is stronger in the South, where organized crime is the main concern, the Northern League focuses almost exclusively on the North, where other kinds of criminal activities received more attention (Wilkes et al. 2007, Semyonov 2007). Whereas organized crime in the form of Mafia activities is traditionally perceived as a domestic-bred problem, the rise of small crime in the North is, at least in the media, associated with the recent increase in immigrants population. In the end, both logic and empirical observation point to the Northern League as the most direct electoral outlet for discontent with the presence of migrants and migrant-driven increases in crime. The above considerations lead us to formulate the following hypothesis: H1: everything else equal, higher presence of migrants will have a positive effect on the electoral performances of the anti-migration party, namely, the Northern League. This hypothesis calls for further specification. All migrants are not the same in the eyes of the public. Each receiving country has attached some stereotypes to different group of migrants. Also, those perceptions and stereotypes vary with time. One need only recall the successive waves of migration in the US, UK, Germany, etc., and the labels attached to each group of newcomer. The status and therefore the mainstream public perception of national/ethnic groups changes over time: for example, Italian migrants were 13

14 considered at the bottom of the social hierarchy in the US for a long time, but it is easy to say that that perception does not apply anymore. Similarly, some national or ethnic groups have received focused negative attention in the media and in the eyes of the general public in Italy (Stella 2002). Sometimes that depends on their nationality, as it happened for Albanian migrants (Bonifazi & Sabatino 2003, King & Mai 2009) or recently for Romanian, sometimes on their ethnicity (Northern Africans). This leads us to a further specification of H1: H1a: everything else equal, specific nationality or ethnic groups will have different effects on the electoral performances of the Northern League. Since it is hard or impossible for the public to precisely identify migrants according to their nationality on sight, I have proceeded to merge migrants data by nationality into container groups. I will explain this grouping process more into details in the research design section. Another part of my research relates to the recent new status of Italy as a receiving country of migration. This new phenomenon comes in correspondence with the aging of Italy s population, and its decline in fertility rates. Point in fact, Italy s age pyramid has shifted toward the gray ages (OECD data 2007). An older population, coupled with the country s novelty of hosting a more diverse population should make the public less receptive to migrants, especially those belonging to groups that most look different from the natives. At the local level then, a higher percentage of elderly population should increase the performance of the party that campaigns against migration. The above considerations lead me to formulate the following hypothesis, which I will test through my control variables: H2: everything else equal, in older municipalities the presence of migrants will have a stronger effect on the performances of the anti-migrants party. Since the Northern League has appointed itself as the protest party in Italy, or at least the party that fights the hardest against the system (See Betz 1993 & 2001, Agnew & Brusa 1999, Zaslove 2004 amongst others), we have to see if its consent comes from its exploitation of the migration issue or from the deterioration of economic conditions that the public blames on the inaction of Italian governments. This considerations produce the following hypothesis: H3: everything else equal, worse economic conditions will have a positive effect on the electoral performance of the protest party. In the next section I will describe my research design and the data I am using to test my hypotheses. 14

15 04. Research Design The design of my research comes from the focus of my dependent variable. Since the goal is to investigate how the inflow of migrants in Italy influenced electoral results, the focus is on the Lega Nord (Northern League) performances. Specifically, I look at the difference in the performance of the party between the political elections (i.e. The elections for the Italian Parliament) of 2006 and the ones of The next sections will explain the geographic scope of my research and the reasons why I considered my dependent variable in the way I did. The main dependent variable is the presence of migrants. The underlying assumption behind my research, as reported above, is that host country citizens do not see migrants in the same way. The simple criterion behind the public s perception of migrants is that the more different they look and speak from the native population, the more they stand out. Standing out by itself does not mean a necessarily negative perception. There is, however, a historical correlation between looks, language and perception. Another important side of public perception of migrants is their relative poverty, generally speaking compared to the native population. Public s perception of migrants of different ethnic minorities comes also from media coverage of crime stories in which migrants are involved. The mass media today is definitely more politically correct than, say, the press in the period between the 19th and the 20th century. That notwithstanding, crime stories involving migrants are frequently sensationalized on the print press anon TV [Lewis/BBC 2009c). This introduces to the topic of grouping and simplification. In countries like the UK, census data report the ethnic (self-)identification of citizens, which makes it easy to group them according to ethnicity. The Italian census data does not; it reports nationality instead. It does not sound like a stretch to assume that most citizens are not able to distinguish immigrants by nationality by just looking at them. They make generalizations based on the look of migrants. Except for the case of massive inflows of migrants from a single country (we ll deal with that later), the public is not likely to consider migrants in term of nationality, but it will do that in terms of ethnicity and area of origin. For this reason, I decided to group migrants into groups that roughly reflect their ethnicity. I say roughly because this grouping is not (and it cannot be) 100% accurate, since nationality per se is not an indicator of ethnicity. In my analysis, I used the following grouping: African / African Origin: this groups includes sub-saharian African countries (but not Northern African countries such as Egypt or Libia) and Caribbean Island whose population os mostly of African descent, such as Haiti or Jamaica. 15

16 Asian: contains East Asian nationals, with the exception of the Philippines, which I grouped with South Asian for reasons of looks and legacy (the Philippino community in Italy pre-dates other Asian migrant groups). Eastern European: contains all Eastern European countries, from Russia to the new members of the EU (following the 2004 and 2007 enlargement rounds). Those include Balkan countries but not Greece. Western: includes all EU-15 members plus Switzerland, Norway, Anglo-Saxon countries. Latin America: all Central and Latin American countries, from Mexico to Argentina. Middle East: this groups includes all Northern African countries, from Morocco to Egypt, all Middle Eastern countries, plus central Asian countries (the former-ussr central Asian states), plus Pakistan. Other Asian: all the countries of the Indian sub-continent with the exception of Pakistan, plus the Philippinos. The following parts will describe the area of my research, my data, and the variables that I am using for my model Places The particularity of the Northern League as a national party is that it does not run like one. In fact, the party concentrates its efforts on the Northern Regions of Italy, and it loosely federates with similar movement in other non-northern regions. That comes from the initial message of the League that, recall, called for a separation of the North from the parasitic areas of Southern Italy and the creation of a separate political entity called Padania (from the denomination of the Po valley). The result of this legacy is the political absence of the party in the regions South of the Emilia-Romagna region. That is the main reason for the geographic limitation of this research. Since the League did not run in those regions, I excluded them from my sample. The party situation situation is also peculiar in two special-statute regions that enjoy greater administrative autonomy than other regular regions, Valle d Aosta, which borders with France, and Trentino-Alto Adige, which borders with Austria. In Valle d Aosta, local politics already register the presence of an alternative regionalist party, and that kept the Northern League from running there. For that reason, I have excluded that region from my 16

17 sample. The situation is similar in Trentino-Alto Adige, which has a strong regionalist German-language party, the SudTiroler VolksPartei (SVP), formerly a separatist movement that wanted to re-join the Alto Adige area (formerly known as South Tirol) with Austria. The party has such a large consensus in the Northern part of the region that it is actually able to access the national parliament. That notwithstanding, the Northern League decided to run in the region, and that is the reason why I have included it in my sample. That leaves us with the following regions: Piemonte (Piedmont), Lombardia (the region of Milan), Liguria (the region of Genoa), Veneto (the region of Venice), Trentino- Alto Adige (see above), Friuli-Venezia Giulia, and Emilia-Romagna. In the above regions, we register 4,405 municipalities, which constitute our unit of analysis. [Map#1 About here] 04.2 Data The data for my analysis comes from three sources. The demographic and employment data comes from the Italian National Institute of Statistics (IStat) data base. The demographic data about foreigners were the latest available at the municipal level at the time of compiling my data set. IStat will be releasing further data at the province every year, and every two year at the municipal level. For that reason, my foreigners presence data lags behind the election data. If anything, however, that will have the effect of dampening my results, given the further increase in migratory flows to Italy in the last two years [CITATION]. Also, there is a theoretical reason that favors the use of pre-dating data: the public does not react instantly to the change in the demographic of their municipality, but needs some time to react the presence of newcomers. The income data, both at the municipality level and at the province level, come from the Treasury Ministry data base, updated at the last fiscal year available, Finally, the electoral result data comes from the database of the Ministry of Interior, which I compiled for each municipality. I decided to focus on the results of the Northern League alone, even though the party ran as part of a coalition that included also future Prime Minister s Berlusconi s party Forza Italia and the other mainstream right-wing party Alleanza Nazionale 1. There are several reasons to consider the NL separately. First of all, Forza Italia and Alleanza Nazionale ran as a single list, under the label Partito della Liberta (Party of Freedom - PdL). The NL instead decided to keep its identity separate and 1 For a discussion on the stance of Alleanza Nazionale on the issue, see Ter Wal 2000 and Ignazi

18 run under its own label. That makes it straightforward to assess that its electoral results did not get a boost from the association with other parties labels, but only from the association with the larger center-right coalition. For Forza Italia and Alleanza Nazionale, which ran with a unified list, it would not be possible to assess the results of each single party. Second, even though all parties of the center-right coalition emphasized law-and-order issues in their campaigns, the NL was the only party that focused explicitly on immigration. As mentioned above, Italian citizens who wanted to send a clear message of opposition to immigration, had a clear outlet in voting for the NL. This attitude of NL voters to attribute increase of crime to migrants is not a recent development: a 2001 ISPO (Istituto per la Pubblica Opinione - Public Opinion Institute) opinion survey found that a total of 67% of NL voters agree that immigrants cause crime (43% agreed, 24% much agreed). The next section will describe into details the variables that I utilize in my model, whose explanation comes in the section after that Variables & Model My dependent variable is the electoral performance of the Northern League in 2008 Italian general elections at the municipal level. Specifically, the relative difference in percentage points (not the mere subtraction between the two electoral rounds. My main dependent variables are the following container variables (also at the municipal level), which gather different nationality into categories of migrants 2. I also describe the expectations of their effect on the dependent variable: African or African origin: the expectation on this group is to have a positive impact on the dependent variable. Immigrants of African origin are the easiest to identify as migrants in Italy (even though that might not be true, given a small group of Eritrean and Somalians that have lived in Italy for a long time). Asian: positive impact. Asian migration should capture the worries about the growth of Chinatowns in Italian cities or metro areas. Eastern European: positive impact. Eastern European migration should capture the negative perception of migrants such Albanians and Romanians, which have been associated in the public perception with a rise in criminal activity. Middle Eastern: positive impact. Middle Eastern migration should capture the perception of the menace of Muslim immigration, both from a security standpoint and 18

19 from a cultural standpoint (see Allievi 2003). It doesn t matter that not all migrants belonging to this group are actually Muslim or aren t necessarily very religious: they are the most likely to be identified as Muslim. Other Asian: positive impact. Given that I have included Philippinos in this group, the impact of this group might not be as straightforward as the other ones above. In fact, Philippinos are one of the first numerically relevant and highly visible group of migrants in Italy. Many Philippino migrants moved to Italy in the 1980s to work as service personnel, and they have become relatively integrated in Italian society. At the very least, their presence did not elicit any relevant backlash. Westerner : neutral. All the above come from 2007 demographic data, directly from the Italian Institute for Statistics (IStat), the government agency that also compiles Censuses in Italy. I have compiled the nationality count at the municipal level This categorization is of course a simplification, as stated before. It is necessary to try to assess the impact of migration on the public, which also simplifies migration through its perception, though. In addition to that, I will also specific nationalities in successive specifications of the model, in order to see if some nationalities actually stand out, even relatively to other foreigners that belong to the same group as specified above. The control variables of my model belong to demographic and economic categories. As reported above, unfortunately not all the data is time-consistent, meaning that it does not all belongs to the same year. That is due to the collection and reporting of data at the municipal level, compared to the regional or provincial level. Some of the variables I utilized were simply not available at the municipal level except for the year of the national Census (2001). My controls are the following 3, all at the municipal level: 2006 NL performance: the performance of the Northern League in the 2006 general elections. I expect this variable to be of course positively related to my dependent variable. As Keel & Kelly (2006) remarked, the mere probability that the process is at least weakly dynamic could lead to a misled specification of the model without the inclusion of a lag of 2 Please look at the appendix for the full list of nationalities included in each group. 3 In previous specifications of my models, I had included also youth unemployment (15-25), the ration of foreign-inmovers (2007), the total migration balance (2007), and household income (2007). Preliminary analysis showed me that those variables did not have a relevant effect on my main independent variables or were statistically relevant. Also, 19

20 the dependent variable. On the other hand, the larger dimension of the sample makes bias of the estimation more probable. To obviate to this issue, I also perform smaller-sample region-based analysis. Unemployment: unemployment rate from the Census 2001 data. Since the NL ran against the government in charge and as a protest party to protect Italian workers from foreigners competition, the expectation is that higher levels of unemployment should boost the party s performances. Population age index: population aged 65 + divided by the population aged 0-14, times 100, from the Census 2001 data. The influence of this control could potentially go both ways: if older citizens are the ones who worry the most about migration, higher scores of this index should have a positive influence on the NL electoral performances. On the other hand, if the NL captures youth s discontent with the socio-economic situation in Italy, the influence on the dependent variable could be the opposite. Education index: percentage of high-school diploma attainment, from the Census 2001 data. Given the self-appointed label of the NL as a protest party, we would expect higher scores on this control to have a negative effect on the dependent variable Italian in-movers: ratio if Italians that moved into the municipality at the end of This control variable should have an opposite effect compared to the one above, with Italian in-movers balancing the inflow of foreigners. Income - variation 99-07: average household income variation in percentage between 1999 and 2007, from the Italian Fiscal Agency. My expectations are similar to the above control: worsening of economic conditions should favor a protest party such as the NL. Density: habitative density per square kilometer, from the 2001 data. Lower density should make migrants stand out. On the other hand, higher density municipalities also tend to attract more migrants and those tend to concentrate in specific neighborhood. Therefore, I keep neutral expectations about this control variable. After the description of the variables in my data set, it is time to introduce my model: NL electoral performance = (Foreign Group[s] + previous performance + controls + error) The following section will discuss the results of my general model and of my regionby-region analysis. the foreign in-movers variable obviously created problems of autocorrelation with my main explanatory variables. 20

21 05.1 General Results My analysis uses OLS regression with robust standard errors, to account for heteroskedasticity of observations at the local level. Of the 4460 municipalities, we are left with 4405 valid observations, because of the absence of income data for some of the smallest municipalities. That data loss comes from a) the exclusion of the results of Valle d Aosta and b) the Lombardia region, where it derives from the change in the administrative division of the province of Milan, part of which split to form a new one. If anything this loss of data is likely to drive my results down more than anything, since the Lombardia region has always been a stronghold of the Northern League (whose original name, recall, was Lega Lombarda - the Lombard League). The table below reports the total results from my general sample: [Table #2 about here] In my general analysis, the presence of immigrants groups had a strong effect on the NL electoral performance. Not all groups follow the expected pattern though. The African origin group has a positive effect on the dependent variable, with a t-score of 4.70 and 99% statistical relevance. That is not surprising, since migrants of African origin stand out from the general population, so they serve as a quick sign of an increased presence of migrants. The same applies to the Asian group, which shows a positive sign, a t-score of 5.5 and a statistical relevance of 99%. That captures the worries about growing presence of Chinese population in Italian largest cities or in their immediate surroundings. The Eastern European group shows the same results, with a positive sign, a t-score of 6.76 and 99% statistical relevance. The NL campaign seems to have yielded its results here. So far, the general results follow what we expected. Surprisingly though (and contrary to my expectations), given the stress that the NL put on the issue of cultural menace, the Middle Eastern group is actually negatively associated with better NL performances. The variable s sign is in fact negative, with a t- score of and a 99% statistical relevance. We will see in the regional results that the variation of that variable across regions is considerable. It looks like that, at the very least, not all Italian regions share that kind of concerns. The Southern Asian group variable behaves as expected, therefore showing a negative sign (t-score of and 99% statistical relevance). That possibly reflects the standing of the Philippino community, one of the oldest immigrant groups in Italy, which 21

22 has received widespread acceptance, or at the very hasn t provoked any strong reaction of negative nature. The Westerners group variable is statistically significant (at the 95% level), with a negative sign. My speculation about it is that Westerners are more likely to move to Italian municipalities that by their very own characteristics are less likely to lean toward the Northern League. In absence of other pieces of evidence, I would maintain that that is a more likely explanation than Westerner migrants having a dampening effect on their neighbors attitudes to other groups of foreigners. The same explanation could apply to the Latin American group, which shows a negative sign, a t-score of (relevant at the 99% level). It is hard to figure that out without more precise socio-demographic information about the group itself. My control variables hold some interesting insights too. The Italian in-movers variable, which is aimed at measuring how dynamic a municipality is, has a negative sign and is statistically relevant at the 95% level (t-score of -2.41). Apparently, turnover of citizens dampens the NL electoral performances. The story about unemployment is more interesting: the variable is negative and statistically relevant at the 95% level (t-score of ), suggesting that this is not an economic discontent story. We should take this result carefully, since that data comes from 2001 census, so we have some time inconsistency. On the other hand, descriptive data by region show (see Figure #2) that unemployment levels had been on a downward trend until They started rising from mid-2008, but the general elections were held in April. Also, income variation is also negative which would lead some credibility to economic reasons, but the variable is far from being statistically relevant (t-score of -1.61). This means that we cannot reject the null hypothesis for H3. It seems that immigration is the star of the story here, not a supporting character. [Figure #2 about here]. A few words about the remaining controls: the population age index shows a negative sign and it is statistically relevant at the 99% level (t-score of -4.76). The reason for that might be that older citizens do not appreciate the radical stance of the NL and its opposition to the Catholic Church on matters of migration. An alternative explanation would be their attachment to older parties or at least parties that have more established ancestors like the Democratic Party, but this explanation would seem to be weaker just because of the continuous changes in labels in Italy from the 1990s on. With these considerations in mind, we cannot reject the null hypothesis for H2. 22

23 Finally, education levels seem to have no effect, but we should be aware that we are talking about a measurement that is likely to be updated, since it measures high-school graduation rates, rather than BA attainment rates. When data becomes available at the municipal level, that should become part of a revision of the present analysis. In sum, my general analysis leads me to reject the null hypothesis for H1 and H1a. Migrants seem to have an impact on the performance of the Northern League, and different groups impact it in a different fashion. The general analysis tells only one part of the story though. As theories of migrations have explained, migrants cluster into networks of the same nationality or at least of the same area. That network effect would get smoothed out in a general analysis. For that reason I decided to perform a regio-by region analysis. That will highlight regional patterns that might have disappeared in the general analysis Results by Region The electoral results of the NL in 2008 were strong across the board in our sample. In fact, the average gain by municipality was strong in all regions. Table #3 shows the party s relative average gain by municipality, its relative gain, and its absolute gain, all compared to the 2006 elections. [Table #3 about here] We can see from the descriptive data above that the NL relative gains were strong in all regions, even though the most striking result was in Emilia-Romagna, a traditional stronghold of the left since the post-war II era. Our general results, summarized in Table #4, show that migrants of African origin have helped the electoral results of the Northern League. However, we know from qualitative reports (Leiden/BBC 2009 amongst others) that public attitude toward migrants is not the same across our geographic sample. [Table #4 about here] First of all some considerations on our group of foreigners. The Lombardia region (Milan as its capital), with its 1490 municipalities is the main driving force for the African Origin group variable. There, it is positive and statistically significant at the 99% level (tscore of 3.98). In all other regions the variable is statistically insignificant. Interestingly enough, in Veneto and Liguria, its direction is the opposite of the general sample. The Asian group variable is not statistically relevant in any of the regions, even though it is 23

24 highly relevant in the general sample. That might derive from the concentration of Chinese migrants in specific enclaves, and therefore their relative invisibility outside of such areas. Data based on municipalities as the unit might not be able to capture their effects on electoral politics. The Other Asian variable largely followed our expectations as far as direction: it kept its sign negative in all regions, with the exception of Piedmont, where it was also statistically relevant (at the 95% level). It was also statistically relevant in Lombardia (at the 99% level) and in Veneto (at the 95% level), where it showed the expected negative sign. I do not have data to convincingly explain the Piedmont result, but that is definitely something that warrants more investigation, and probably a re-definition of my group in terms of the nationalities composing it. Another bigger surprise came from the Eastern European variable, which had a negative sign in both Piedmont and Lombardia, where it was also statistically relevant. That variable showed the expected positive sign in all other regions, and it was statistically relevant in Trentino/A.A. and in Emilia-Romagna. The analysis of the concentration of that group by regions does not reveal patterns that clearly explain this uneven reaction. To further explore this phenomenon, I need to unpack the group and explore its socioeconomic features more in depth at the regional level. What seems to come out of the regional result though is that this group had an effect in the traditionally leftist Emilia- Romagna, where concerns about security might have helped the NL. An interesting remark about Latin American group. In Emilia-Romagna the variable is positive and statistically significant (99% level). What changes from the other regions in our sample? Concentration seems to be the answer: the presence of migrants from Latin America by municipality in proportion to the local population is at least three times as high as in other regions. Proximity brought reaction in the (formerly) tolerant Emilia- Romagna too. The strongest driver of the Middle Eastern group scores is the Veneto region. That reflects the political battle for Muslim s worshipping space (La Repubblica, November 11, 2007), which reached new highs (or lows) with the NL-organized stunt of a pig led to pasture on the lot of land designated for a new mosque in Padua (Padova). The same results (only 95% statistical relevance though) apply to Veneto s Eastern neighbor, Friuli, where the concentration of Middle Easterners is much lower though. Emilia-Romagna, on the other hand, does not show the influence of such cultural fights: there the variable is negative and statistically significant. The next section will draw some conclusion from my research, and will describe some of its possible implications. 24

25 06. Conclusions My research has shown that migrants in Italy are giving new life to the Northern League. It is impossible now to tell whether immigration will continue to be on top of the agenda of the party, or it will ebb soon (see Cole 2005, Semyonov et al. 2007), but it is very likely that the NL will ride the issue until it can provide some political gain. This said, the panorama of migration is extremely variegated in Italy, and so is its influence on politics. One general lesson is that not migration matters, but also it matters who the migrants are. The concentration of specific groups of migrants such as Middle Eastern has provoked strong political reactions. The left in Italy has for long ignored the issue, or at the very least it has not addressed the worries of the general public. Migration from Eastern Europe has raised concerns about law and order, which made their way to the public even in a red stronghold such as Emilia-Romagna. Those concerns could subsume if and once Eastern European migrants, especially Romanian, return home because of improved economic opportunities, as it happened for Poles migrants in the UK. In the meantime, the NL has had an easy time ramping up its anti-eu rhetoric and blaming crime issues on migrants. The results of my research square with a recent CNEL (National Council on the Economy and Labor - a government agency) report on the integration of migrants in Italy (February 2009). Whereas Northern regions scored very high in attractiveness for migrants, and also high as far as integration in absolute terms, they did not score well in comparative terms. To compile comparative rankings, the CNEL report compared the living conditions of migrants with those of locals. Its findings highlighted that Southern regions scored much better than Northern ones. For example, Emilia-Romagna (the Bologna region), Lombardia (the Milan region) ranked respectively first and fourth in the absolute socio-economic integration ranking. However, in the comparative ranking they placed 14th and 18th respectively (out of 20 regions). The bottom line is that in Northern regions the inequality between the conditions of migrants and those of locals is extremely high. That might be one explanation for the difficulty of connection between locals and migrants in those areas, an issue that favors (for now) the electoral strategy of the Northern League. Another issue to point out is that such reports so far do not focus on the relative integration of different groups of migrants. My analysis would suggest that not all migrants are same as far as provoking locals reactions. In such respect, the position of migrants from Africa and from the Middle East deserves a specific comment. The NL had an easy capturing the fears of the public by targeting culturally foreign groups. The debates on the 25

26 building and location of mosques have provided political ammunition for the NL claim that those migrants represent a cultural menace. The public seem to have many doubts about the possibility of integration of culturally alien groups and so part of Italian intellectuals (See Fallaci 2002 and Sartori 2002). My findings and the CNEL report would highlight that the most productive part of Italy is also ground zero in the battle for integration. The North is the most attractive destination for migrants, but also where increased flows of foreigners have provoked the harshest political reaction. From an economic standpoint, there is no doubt that Italy needs migrants. From a cultural standpoint though, there is no doubt that the integration of specific groups of migrants looks, to say the least, problematic, otherwise the NL campaign could not have been as productive as we ve seen. It could be that worse economic conditions rein in the inflows of legal and illegal migrants, but if not, Italy could move to an increasingly conflictual society. To avoid that, Italian governments might try to seriously limit immigration from non-eu countries. Such a move could make it easier for the authorities and society alike to concentrate on the assimilation of Romanians and Eastern European in general. Unfortunately, the difficulty of monitoring Italy s borders might make that attractive option not practical. If that is Italy s future, the road to integration might prove very bumpy, if successful at all, which might open even more possibilities for a party like the Northern League. 26

27 Figures and Tables Table #1: Migrants in Italy, 2007 Country Legal Residents (IStat) Count % Of foreigners Total Estimate (Caritas estimate) Count % Of foreigners Romania 625, , Albania 401, , Morocco 365, , China PRC 156, , Ukraine 132, , Philippines 105, , Total 3,432,651 3,987,100 Figure #1: Northern League Election Results, 2006 and 2008, % 27

28 Figure #2: Northern League Electoral Poster, Piedmont Translation: Guess Who s Last? For rights on: housing, labor, health care. 28

29 Map #1: Northern Italy, Regions Table #2: General Sample results Dependent Variable: NL 2008 Electoral Performance Independent Variables Coefficient t P>t African Origin Asian Other Asian Eastern European Westerner Latin American Middle Eastern Northern League In-movers Income variation Unemployment Age index Education Density Constant