CIVIC LEGACIES AND THE DEVELOPMENT OF URBAN INTEGRATION
POLICY IN FRANCE: MIGRANT ROMA HOUSING POLICY
Chapter 1. Introduction
“’Laudato si’ means to welcome, protect, promote and integrate.”3
Why was Toulouse a leader in housing policy for migrant Roma while Lille was a
follower? In 2015, the City of Toulouse announced it would pivot from a logic of
shortterm accommodation to a logic of long-term accommodation.4 This long-term
accommodation policy was comprised of, “autonomous housing units with social support
in view of assuming the rights and duties of being a tenant able to cover the cost of rent
in the long term.” Alternatively, Lille did not adapt their housing policies from shortterm
accommodation until 2017 when a National Housing First policy was launched.
Toulouse was a leader in introducing autonomous housing policy with social support for
EU migrant Roma while Lille was a follower and only adapted its policy after national
3In 2019, at a council meeting in Toulouse, Dr. Daniel Rougé used this reference to, Laudato Si, to
describe the values which underpinned the Toulouse model of housing integration for precarious European
nationals and refugees. Toulouse Municipal Council, « Conseil Municipal, Séance du vendredi 10 Juillet
2020 N4, » pg 24, 10 July 2020, Toulouse Municipal Online Archive.
4In 2015, Toulouse had around 1,200 EU migrant Roma mainly from Bulgaria and Romania living in
15 camps (Frederic Dessort, “A Toulouse, l’insertion des Roms passe par les toits, » 2019,
Libération,https://-linsertion-des-roms-passe-par-les-toits_1721996/). Toulouse Municipal Council,
“Dispositif d'hébergement accompagné pour les personnes en situation de précarité : modification et
harmonisation du cadre normatif, » pg 1, 22 October 2021, Délibération n 1.4, Toulouse Online Municipal
Archives. Toulouse Municipal Council, « Terrains et sites d’accueil gères par la ville pour les personnes en
situation de précarité : modification et harmonisation du cadre normatif, » pg 1,2, 26 October 2015,
Déliberation n 4.2, Toulouse Online Municipal Archives. “15 cities learn from Toulouse how to support
Roma to access adequate housing,” 2019, Eurocities. http://www.eurocities.eu/eurocities/news/15 - cities -
learn - from Toulouse - how - to - support - Roma - to - access- adequate-housing-WSPO-BBDLD3.
1
level policy changes. Understanding the divergence in these cases explains why some
cities lead while others follow in their policy-making concerning stigmatized migrants
and how that integration policy changes over time.
Furthermore, Roma are an important group for several reasons (not all discussed
here). Roma are the largest minority in Europe and live in every country in Western
Europe in addition to Canada and the United States. At the same time, non-Roma
Europeans harbor some of the highest levels of prejudice and racism toward Roma. So,
boiled down, the question of housing integration policy for EU migrant Roma is a
question of integration policy and racism broadly. How do urban governments across
Europe house people who face such a high level of prejudice?
The literature on civic legacies suggests that policy entrepreneurs draw upon
existing networks (NGOs) and ideas (policy frames) developed in previous integration
efforts to shape integration policy (Arrington 2016; Chung 2020; Ireland 2004). I suggest
that Toulouse developed networks and ideas during the integration of migrant Roma in
the 1990s which set a precedent for its city council to shape new housing policy in 2015.
Alternatively, Lille lacked a civic legacy in 1990 to set precedent for 2015 and thus did
not change their policy until there was a national level change in policy.
This dissertation develops a theory of how civic legacies influence integration
policy. The evidence demonstrates that the legacy of local grassroots NGOs, policy
entrepreneurs and policy frames results in policy change. By “legacy,” this dissertation
refers to the NGOs, policy entrepreneurs and policy frames that existed historically and
set a precedent for the current contemporary housing integration policy.
2
While civic legacies are a necessary condition, they are not sufficient for policy
change. Policy entrepreneurs are required as the agents of policy change. Therefore,
while policy entrepreneurs are part of the civic legacy which sets a precedent for policy
change, there must also be a policy entrepreneur acting to change policy in the present.
The rest of this introductory chapter explains variation in housing policy for EU migrant
Roma in Lille and Toulouse, the argument in brief, a short discussion of methods,
important context concerning policy on Roma and the cities of interest, empirical
expectations, the significance of the inquiry and lastly a roadmap for the rest of the
dissertation.
Variation in Housing Policy for EU migrant Roma in France
Housing Policy in France
The French government has a rich history of introducing social policy for
socioeconomically disadvantaged populations. Magda Maaoui (2020) describes the
housing market in France as, “characterized by an interventionist state-led urban regime
where national programs in place since the postwar era guaranteed a consistent
production of social housing units for the country” (Maaoui 2021, 1-2). In the early 2000s
the French government introduced the Solidarity and Urban Renewal Law (Loi Solidarité
Et Renouvellement Urbain (SRU)) in an attempt to “rebalance the geographic location of
social housing units in response to growing trends of segregation and concentration of
poverty in French cities” (Maauoi 2021, 2). The SRU law was designed to promote “la
mixité social,” directly translated as “social mixing,” or social diversity and thus
integration outcomes. France has introduced the SRU law that required selected
municipalities to devote 25% of their local housing stock to social housing. For
3
comparison, most American inclusionary housing programs (e.g., in Illinois, Connecticut,
Rhode Island, and Massachusetts) set a fair share of social housing at about 10% (Silver
2019). Social housing stock (social rental housing with capped rents) is relatively large in
France. Therefore, while France does not collect data on ethnicity or race, French
national law and policy has a disproportionate focus on the integration on social housing
stock.
That said, while France has a relatively large stock of social housing and a
deliberate focus on the housing integration in France, while the state has the legal duty to
integrate or deport EU migrant Roma from its territory, it is up to city governments to
introduce housing policies for different groups. Therefore even when the administrative
status of a group is clear and the responsibility to provide housing clearly lies with the
government, as a stalling tactic city and state representatives will contest whose
responsibility (legal and otherwise) it is to accommodate migrant Roma. Therefore,
integration policies change depending on how the city frames its sense of responsibility
toward a target group—such as a humanitarian responsibility versus a legal responsibility.
Thus, housing policy depends partially on the administrative status of the migrants, but it
is also influenced by the policy frame articulated by the city regarding the target group.
In practice, French policy makers refer to integration policies that adopt the
national-level legal responsibility to house migrants as “volontariste” policies, or
proactive policies. The adoption of housing integration policies for migrant Roma is
perceived as the city taking on extra responsibility for these groups and developing
political will beyond its legal competencies to introduce policy that legally should be
4
handled at the state-level. Typically, it is the political parties on the left of the ideological
spectrum that take on these proactive policies, but in the case of Toulouse, for example,
we observe the center-right adopting proactive policies on behalf of the state in order to
house their local EU migrant Roma.
Policy for EU Migrant Roma
While the policies analyzed for this dissertation focus on housing for one specific
group “EU migrant Roma,” the broader ethnic group, “Roma” is comprised of a
multifarious people that city bureaucracy and NGOs tend to homogenize into one
administrative group for the purpose of service delivery. The name, “Roma,” is an
umbrella term, first adopted at the World Congress in London in 1971. “Roma” refers to
Roma, Sinti, Kale and related groups in Europe, including Travelers and the Eastern
groups (Dom and Lom), and covers the wide diversity of the groups concerned,
including persons who identify themselves as Gypsies.12 Some Roma share common
geographic and linguistic origins. At the end of the 18th century, linguists found
evidence that Roma migrated from the north of India as early as the 9th century and
concluded that Roma share a language, Romani, derived from Sanskrit.13 Historical
evidence suggests Roma have been present in Europe since the end of the 14th century
(Council of Europe 2012).
Roma have a long history in France specifically. In France, Roma have had many
different policy names over the last two centuries: “nomades,” (nomads) “gens du
voyage,” (French travellers) “tsiganes,” (gypsies) and finally “Roms” (Roma). These
exonyms refer to different groups of Roma over time, some migrants, some French.
5
Today, 10 to 12 million Roma live in the EU, the majority of whom are sedentary. This
number is a conjecture due to the “…varying definition and classification systems for
race/ethnicity, ethical and legal limitations, and reluctance to identify as Roma for fear
of stigmatization” (Warwick-Booth et al. 2017, 670). Nevertheless, Roma are a
nonnegligeable population as Europe’s largest minority. In 2019, policy centers and
NGOs estimated there were between 12,000 and 15,000 migrant Roma in France. At the
High-Level Conference on the EU Framework for National Roma Integration Strategies
in March 2019, the French Minister for Housing (DIHAL), Mr. Sylvain Mathieu stated
13 The 2012 “Glossary of Terms Related to Roma issues” addresses the challenges in defining who counts
as Roma, “There are, in addition, other groups, particularly in the Balkans, who are regarded or who regard
themselves as Roma, but who do not speak Romani and who do not have the same North Indian origin.
These include the Boyash (also known as Beash, Bayash, Banyash, Baieşi or Rudari, depending onthe
country) whose language derives from Moeso-Romanian, and some members of the Ashkali, who speak
Albanian. Other groups, who resemble the Roma in certain respects, such as the Albanian-speaking
“Egyptians” (so-called because they reputedly came from Egypt) and certain Ashkali, insist on their ethnic
difference. Both groups are Albanian-speaking In western Europe, a distinction should be made between
the Roma/Sinti/Kale and other groups who sometimes share their lifestyle, or their plight (difficulties in
integrating, discrimination, etc.) but who do not have the same ethnic origin or the same language. These
include the Yenish (found in Switzerland and some neighbouring countries) and the Irish Travelers (see
below).” Council of Europe, "Fighting Housing Discrimination in Europe," https://rm.coe.int/1680088eab.
that throughout France, “(t)here are around 500 slum areas populated by 16,000 people,
mainly Romanian and Bulgarian nationals (Mathieu 2019, 2). The question of housing
for EU migrant Roma is a pivotal one for France.
Migrant Roma from Post-Soviet Romania and Bulgaria
For this dissertation specifically, “migrant Roma” refer to Roma migrants from
Romania and Bulgaria who began their migration west after 1990. These migrants were
fleeing discrimination and seeking economic opportunities. In Romania and Bulgaria,
after the fall of communism in 1990, while Roma had their rights formally reinstated,
discrimination was still widespread across education, labor, health and housing sectors.
For example, in February 1990, the leading political coalition in Romania, the National
6
Salvation Front (with the support of national television stations) used Roma as scapegoats
for the violence of Romania’s Revolution in December 1989. In response to this
heightened discrimination, Roma travelled west to countries such as France for new
economic opportunities. French cities struggled to house these new immigrants, but some
cities were more successful than others. For example, in Toulouse social workers
collected surveys of migrant Roma families and attempted to find them suitable housing.
In Lille, there is no archival record that local NGOs conducted systematic surveys or
aided migrant Roma families from Romania or Bulgaria during this time.
Between 1990 and 2007, Toulouse continued its efforts to re-house migrant Roma
living in informal camps. Whereas Lille policy focused solely on providing parking sites
for nomadic Roma with French citizenship, called “French travelers,” or gens du
voyages, not migrant Roma. Most French cities, included Lille and Toulouse, provided
and still provide French travelers with designated parking lots with basic water and
sanitation services. Lille did not introduce policy for migrant Roma until 2007 when
Romania joined the EU and migrant Roma were legally allowed to work, travel and live
in France (with some restrictions). In this way, variation in housing policy depends
partially on the administrative status of migrants (whether they are EU citizens, refugees,
asylum seekers, etc.). In 2016, for example, migrant Roma with Romanian citizenship
had a right to live and work in France while migrant Roma with Albanian citizenship did
not. Therefore, variation in housing policy partially depends on citizenship and requisite
administrative status of migrants.
7
In 2007, Lille and Toulouse relied on a patchwork of housing policies regarding
emergency accommodation (like hotels), short-term and long-term housing for migrant
Roma. However, these policies targeted unhoused and marginalized residents generally,
not migrant Roma specifically. Then, in 2007, cities in the north of France developed a
new type of short-term accommodation specifically for migrant Roma. The housing
consisted of gated parking lots with varying levels of resources and services provided by
the city. Some parking lots were equipped with only the basics such as water and
sanitation services, while others were further equipped with mobile homes. In any case,
the defining feature of these parking lots was that they were gated. These parking lots for
migrant Roma were referred to as “integration villages,” or villages d’insertion. Migrant
Roma families applied to reside in these integration villages for a designated amount of
time, usually to avoid being deported from France. By 2007, integration villages were
explicitly funded by the EU.
However by 2010, French academics and activists were concerned that integration
villages segregated migrant Roma from the city, policed their movements in and out of
the sites and did not in fact lead to broader integration outcomes (Bessone et al. 2013;
Legros 2010; Picker 2017). Despite this backlash, both Lille and Toulouse considered
integration villages a valid policy option for housing until 2015. In 2015, Toulouse
publicly pivoted from a logic of short-term housing (integration villages) to autonomous
long-term housing with social support. Lille, however, did not make the same pivot away
from integration villages. Between 2015-2018 Toulouse worked to eliminate integration
villages while Lille maintained a stock of integration villages. In September 2017, France
introduced a national Housing First plan similar to Toulouse’s 2015 policy that focused
8
on long-term autonomous housing with social support. This 2015 policy divergence had
lingering effects years after which can be identified in 2020 policy documents. By 2020
Toulouse no longer considered integration villages a valid housing policy, whereas in
2020 Lille still mentioned integration villages as a valid housing option. This background
on housing policy for migrant Roma in Toulouse and Lille—summarized in chart 1 below
—provides context for the puzzle of why Toulouse led the way and pivoted from
integration village policy in 2015 and Lille did not.
Anti-Roma Discrimination/Racism in France
In 2014, 66 percent of French had an unfavorable view of Roma (whereas 27
percent of French had an unfavorable view of Muslims, for context).20 In France,
antiRoma prejudice is stronger than general anti-immigrant prejudice. In France, anti-
Roma prejudice is exacerbated by the presence of informal encampments. If migrants
cannot get access to housing, some join the informal camps already set up by family or
friends. These camps are not serviced by the city; detritus piles up creating a safety
hazard and negatively impacts the surrounding areas. Often, these neglected camps are
the most visible presence migrant Roma have in cities like Toulouse and Lille. This anti-
Roma prejudice presents an additional challenge for city councils. The challenge of anti-
Roma racism, sometimes referred to as antigypsyism, is well-documented and well-
theorized. However, this research is focused on how city councils introduce new policies
that provide enough support for migrant Roma and end up assuaging the resentment that
fuels anti-Roma sentiment (racism and prejudice), public opinion and voting preferences.
9
20 Richard Wike, Jacob Poushter, Laura Silver, Kat Devlin, Janell Fetterolf, Alexandra Castillo, and
Christine Huang, "European Public Opinion Three Decades After the Fall of Communism," 2019, Pew
Research Center, https://www.pewresearch.org/global/2014/05/12/chapter-4-views-of-roma-muslims-jews/
Chart (1): Housing Policy Eras in Toulouse and Lille
Policy Eras Toulouse Policy Lille Policy
1990-2007: Policy Development.
Rehousing efforts and social
support in Toulouse but not Lille.
Some focus on migrant
Roma in re-housing
efforts. Surveys of
migrant Roma
conducted by social
services network.
No focus on migrant
Roma. Focus on
parking sites French
travelers, “gens du
voyage. » No surveys
of migrant Roma
conducted by social
services network.
2007-2015: Policy similarity.
Integration villages in both cities.
Integration villages in
addition to other types
of housing, moratorium
on evictions
(20102012), an increase
in integration villages
after 2010.
Integration villages in
addition to other types
of housing, moratorium
on evictions
(20102012), an increase
in integration villages
after 2010.
2015-2017: Policy Divergence.
Toulouse pivots from integration
village policy and Lille continues
with integration village policy.
City council phases out
integration villages,
relies primarily on
autonomous housing
with social support and
preempts national-level
policy. One integration
village remains.
Despite closure of a
few integration villages
(from 10-7) continued
reliance upon them.
2017-2020: Policy Continuity.
National Housing First policy is
introduced.
Long-term autonomous
housing with social
support is referred to as
a success and the
model for future
integration of refugees
in Toulouse City
Council.
Under national
Housing First policy,
autonomous long-term
housing becomes
policy while different
types of integration
villages still used.
10
The difference between Toulouse and Lille demonstrates why certain cities lead
misunderstands the power structures and incentives that bolster anti-Roma racism. However, I am claiming
that certain housing policies as opposed to others can make progress toward a more humane and fair
housing process at the local level and can potentially address the anti-Roma attitudes present at the local
level that influence housing policy.
while others follow in their policymaking concerning stigmatized migrants. Toulouse led
the change from integration villages to long-term autonomous housing with social
support. Furthermore, Toulouse’s long-term housing policy served as a model for the
integration of future refugee populations in Toulouse and was recommended to different
cities in France and beyond as a policy solution for migrant Roma integration.23 Lille
implemented long-term autonomous housing with social support after it was mandated by
the federal government and then continued to use integration villages years after they
were considered dubious housing policy at best by academics and policymakers alike.
Argument in Brief
Broadly, civic legacies are the networks (including NGOs and social workers
(chapter 4)) ideas (policy frames (chapter 5)) and actors (policy entrepreneurs (chapter 5
and 6)) developed in previous integration efforts that set a precedent and influence
contemporary integration policy. This dissertation focuses specifically on the civic
legacies that influence migrant Roma housing policy and distinguish cities as leaders or
followers. So, this dissertation is focused on the networks and ideas developed during
previous integration efforts that set a precedent for policy entrepreneurs to influence
contemporary integration housing policy.
Networks create and solidify connections amongst civil society and government
actors. They also generate information-sharing and trust between the target group, civil
society actors and city council that is necessary for housing policy. The civic society
11
network can play a role in shaping integration policy by holding national and local
political actors accountable, giving a voice to migrant concerns, raising public awareness
23 15 cities learn from Toulouse how to support Roma to access adequate housing,” 2019, Eurocities.
http://www.eurocities.eu/eurocities/news/15 - cities - learn - from - Toulouse - how - to - support - Roma - to - access-
adequate-housing-WSPO-BBDLD3.
of policies and advocating for certain policies and reforms (Chung 2020). The civic
legacy of a network is developed over time. The civic legacy of the network is
established by the vanguard group. This vanguard group has the institutional power to
advocate for the new migrants based on its historical experience with integration. The
vanguard group develops the network during previous efforts to integrate the same or
similarly stigmatized migrant groups that require more support from the government. The
role of the network is explored further in chapter 4.
An important actor in the network is the policy entrepreneur (introduced in
chapter 5). Policy entrepreneurs draw upon existing networks (NGOs) and ideas (policy
frames) developed in previous integration efforts to shape contemporary integration
policy. A successful policy entrepreneur convincingly articulates why their policy is the
preferable policy. Chapters 5 and 6 of this dissertation demonstrate how a policy
entrepreneur converses in city council meetings in order to convince opposition parties
and mayors of other municipalities that their housing policy is preferred. In order to
convince others, the policy entrepreneur can introduce policy frames with a civic legacy.
When a policy entrepreneur introduces a frame, she is introducing or constraining
possibilities to change a policy. Successful policy entrepreneurs invoke a policy frame
that develops the political will for a particular policy. Policy frames shape how
civilsociety actors negotiate housing integration policy. Policy frames structure and
12
contextualize debates over integration policy and provide tools for migrant advocates.
Policy frames used in previous struggles for integration have a civic legacy. The civic
legacy of policy frames provides a repertoire of policy frames that have specific
resonance in the community. These policy frames have historical significance and a
different semiotic weight because of that significance. As such, if actors invoke a specific
frame with a civic legacy, this frame may have more resonance in the community, which
might garner the necessary political will to introduce one policy over another. In this way,
the policy frame is part of the city’s civic legacy. Sometimes these frames are
strategically introduced and sometimes not. In Toulouse, the earlier development of
policy frames in reaction to re-housing efforts in the 1990s contributed to the civic legacy
which made Toulouse a leader in housing policy for migrant Roma, while the latent
development of policy frames in reaction to integration villages in 2008 in Lille
contributed to the lack of a civic legacy in Lille and its follower status. The influence of
policy frames on housing policy is explained further in chapter 5.
This dissertation finds evidence that civic legacy of policy frames (ideas) and
NGOs (networks) is activated by policy entrepreneurs to influence contemporary housing
policy. In this way this dissertation describes the transformation of integration policies
over time and the distinction of cities as leaders or followers of migrant integration
policy.
Research Design
Case Selection
In picking the city case studies, Toulouse and Lille, the research design aims to
achieve a similar systems design, or Mill’s “method of difference,” to clarify the role of
13
civic legacies in each city. This paired case comparison holds country-level explanations
(national immigration, integration, welfare and capitalist regimes) constant,
as well as metropolitan area population sizes and the estimated number of migrants in
each city. In 2015, at the time of the key policy divergence at the center of this study,
Toulouse and Lille had the fourth and sixth largest metropolitan areas by population at
1,360,829 and 1,187,820 people, respectively. While there is no census of migrant Roma
in Toulouse and Lille, in 2016 the Interministerial Delegation for Accommodation and
Access to Housing (DIHAL) did track the number of EU migrants evicted from slums
and enrolled in each city’s integration programs. We can assume that most of these
migrants were EU migrant Roma from Romania and Bulgaria at this time. So, in 2016 – a
year after policy divergence—Toulouse reported 817 EU migrants enrolled in their
integration program and Lille reported 1,415 EU migrants enrolled in their integration
program, the second and third highest number of EU migrants enrolled after the Paris
metropolitan area.
Therefore, while it is true that Toulouse and Lille are often sequential in their
population counts, and while a lot of the national level context is the same for Toulouse
and Lille, it is important to note how these cities are different. Eurostat reports that in
Toulouse migrants comprised 18 percent of the population with 124,066 migrants in
2015, while in Lille migrants comprised 13 percent of the population with 118,651
migrants in 2015.27 One could claim that Toulouse has a larger share of immigrants and a
stronger culture of welcoming immigrants. This strong history and culture of immigration
14
would not doubt influence Toulouse’s ability to lead change in integration policy for
migrant Roma.
Toulouse’s geographic location in the southwest of France near the border of
Spain and on the Garonne River which connects to the Mediterranean Sea has historically
made it a hub for immigrants and refugees. In the 1920s and 30’s Toulouse welcomed
Spanish and Italian refugees and migrants fleeing their respective fascist regimes.
Thousands of Jewish migrants traveled to Toulouse during the Holocaust.29 French and
Algerian immigrants traveled to Toulouse after Algeria’s 1962 independence.
However, Lille also has a strong history of migration. In the early 20th century,
Lille attracted migrants (mainly from Poland and Italy) through its burgeoning coal
mining and textiles industries. Between 1960 and 1970’s the first wave of immigrants
from North Africa came to the region, specifically from Algeria and Morocco. However,
immigration slowed down after the 1974 recession and national immigration policies that
restricted the number of permanent workers in France. But in 1994 with the arrival of the
TGV and then Eurostar rails, Lille found itself at the crossroads of all major European
capitals. While Toulouse sits in the southwest and Lille in the northeast, both cities have a
historical reputation as crossroads for the rest of Europe. So, while Toulouse does have a
strong history and culture of immigration so does Lille. Since the late 20th century, Lille
has modernized to accommodate the changes in demographics and is clearly up to the
challenge of welcoming migrants.
By 2015, both Toulouse and Lille were considered global cities with strong
economies, progressive residents, and a history of immigration suggesting either one had
15
the potential to be a leader in housing policies for EU migrant Roma. Keeping these
contexts similar helps to identify the influence of civic legacies. While both Toulouse and
Lille make big efforts to house migrant Roma and deserve credit for generating the
political will to introduce housing policy for the largest minority and most discriminated
group in modern-day Europe, the way they introduce policy is different, with Toulouse
leading the way and Lille following.
Methods
A quick note on methodology before it is covered in depth in chapter 3. First, to
identify relevant networks and policy entrepreneurs, a process-trace was applied to
municipal and metropole level archives, online city council meeting minutes and
interviews gathered in Lille and Toulouse. All content gathered and analyzed concerned
migrant Roma housing policies between the 1990s and 2020.
Next, to identify policy frames, a qualitative analysis was performed in NVIVO.
Policy concepts that repeated themselves (sometimes referred to as “themes”) were
inductively coded across the documents. A pattern was established by tracking which
concepts repeated themselves, why, when and by whom. These patterns delineated policy
frames. Lastly, the chronology of the frames was described in a timeline. This dissertation
does not claim that these are the only policy frames that are always found in urban
integration policy. But, it is likely that these policy frames are present concerning housing
integration policy that require extra social supports.
Empirical Expectations
In Toulouse we should expect to observe mobilized NGOs, policy entrepreneurs
and policy frames (a civic legacy) regarding housing for migrant Roma in the 1990s that
16
set precedent for the policy change of interest in 2015. However, in the Lille case, we
should expect to observe an absence of mobilized NGOs, policy entrepreneurs and policy
frames regarding housing for migrant Roma in the 1990s and thus no change in housing
policy for migrant Roma in 2015 (See chart 2).
Chart (2): Presence of Civic Legacies in 1990s and Outcome
1990s Policy
Frames
(Ideas)
1990s
Successful
Policy
Entrepreneur
(Actors)
1990s NGO
Network
(Networks)
2015 Policy
Change
(Outcome)
Toulouse Yes Yes Yes Yes
Lille No No No No
In Toulouse, the associative network mobilized in the 1990s was a precedent to
the housing policy change in 2015. Toulouse also had a policy entrepreneur that was part
of the civic legacy in Toulouse. Dr. Jean-Claude Guiraud ran a pivotal organization for
nearly 30 years between 1971-1998. Dr. Guiraud established the trust, mediation and
expertise for the social support of migrant Roma families. These policy frames and social
networks were activated again by a new policy entrepreneur when new leadership took
control in 2015. Toulouse had a policy entrepreneur in Mr. Daniel Rougé who worked at
the city council and echoed specific policy frames that were used by Dr. Guiraud in the
1990s in his role as an Associate Director of his NGO.
Lille, in comparison, lacked a mobilized associative network, policy entrepreneurs
and policy frames in the 1990s. Lille did not have a historical associative network that
supported migrant Roma or a policy entrepreneur who articulated policy frames for
migrant Roma housing in the 1990s. In this way, Lille did not have a civic legacy that set
precedent for policy change in 2015. Instead, policy change in Lille occurred later in
17
2017 after a national level policy was introduced.
In 1990 Toulouse, we expect to identify policy entrepreneurs articulating specific
policy frames on the topic of housing for migrant Roma. This would further signal the
presence of a civic legacy. And then in 2015, we again expect to observe policy frames
articulated by particular policy entrepreneurs. For Lille in 1990, we do not expect the
presence of policy frames on the topic of housing for migrant Roma articulated by policy
entrepreneurs. In 2015, we expect to observe policy frames regarding migrant Roma, but
we do not expect to observe a particular policy entrepreneur articulating these policy
frames. In this way, the presence of policy frames is necessary but not sufficient for
policy change. Policy frames must be articulated by a policy entrepreneur to result in
policy change. The substantive messages of the policy frames, not just the presence, are
further explored in chapter 5.
Significance
While the European Commission and member states have done important work on
EU Roma integration over the past decades, policy development and implementation are
left to the cities. In France, this has resulted in some cities leading and others following in
policymaking for EU migrant Roma. This dissertation is interested in what drives that
divergence at the city level. However, the international and national levels are not
irrelevant.
The EU and member states have set broad priorities and limited funding that
encourages housing integration but does not mandate or implement policy change at the
national or local levels. There has been some EU policy introduced that has created
incentives for cities to introduce housing policies. The “Decade of Roma Inclusion,”
launched in 2005, was the first international effort in Europe to improve the
18
socioeconomic conditions of EU Roma. As part of this decade of inclusion, in 2008 the
European Commission drafted the Framework for Roma Integration and mandated that
all EU member countries draft National Strategies for Roma Integration (European
Commission 2020). Between 2014 and 2020, the European Commission earmarked
around 1.5 billion euros for the “socio-economic integration of marginalized communities
such as Roma,” with most investment allocated for education and housing (European
Court of Auditors 2016 10). By 2018, it became clear that the 2008 standards for
inclusion were not sufficiently met (European Commission 2019). There were some
improvements to education and health outcomes, but housing and employment outcomes
stagnated or worsened. Member states described vague, aspirational integration efforts,
leaving the implementation to municipalities. Some cities, like Toulouse, took the
opportunity to lead in its housing policy, while others, like Lille, followed national-level
policy.
To review, a seemingly small and local policy change in Toulouse in 2015 matters
because it demonstrates the difference between a city that is willing to lead policy
change concerning stigmatized migrants versus a city that follows status quo policy.
Toulouse led the change from integration villages to long-term autonomous housing
with social support. Toulouse’s innovative policy was broadcast to different cities in
France and beyond as a policy response for migrant Roma integration.35 Furthermore,
Toulouse’s long-term housing policy served as a model for the integration of future
refugee populations in Toulouse. In contrast, Lille was a follower in its policymaking.
Lille City Council developed autonomous long-term housing policy after it was
19
mandated by the federal government under its Housing First initiative. Lille continued
to use integration villages beyond 2020, years after they were considered dubious
housing policy at best by academics and policymakers alike.
Most importantly, the question of housing integration policy for EU migrant
Roma is a question of integration policy broadly. Non-Roma Europeans harbor some of
the highest levels of prejudice toward Roma and this anti-Roma racism is pervasive
across EU countries. Furthermore, Roma are the largest minority in Europe and live in
every country in Western Europe in addition to Canada and the United States. How do
urban governments across Europe house people who face such a high level of prejudice?
Finally, housing is not only a challenge for EU migrant Roma, but also for other
populations that face a similar level of stigma and/or need access to a similar level of
social support to integrate successfully. As such, this project explores a puzzle relevant
to all urban democratic societies that welcome stigmatized newcomers such as refugees,
asylum seekers and migrants with fewer employment and housing options.
Chapter Overview/Road Map
Chapter 1 began with the puzzle, why did Toulouse lead the change in housing
policy for migrant Roma in 2015, and why did Lille follow status quo? Toulouse
developed its own policy change from integration villages to autonomous housing with
social support, while Lille adopted national level policy later. While the policy change in
question is focused on 2015, the evidence points to a historical process of over twenty
years that set the precedent for policy change. This dissertation finds evidence that policy
20
entrepreneurs (actors) draw on existing NGOs (networks) and policy frames (ideas)
developed in previous integration efforts—civic legacies—to shape housing policy
(Arrington 2016; Chung 2020; Ireland 2004). Toulouse developed a civic legacy in the
1990s that influenced a housing policy shift in 2015, while Lille did not have the aspects
of a civic legacy to influence a policy change at the time. In this way, civic legacies
impact how much political will cities can generate to change policy, and how much they
are led by national level integration policy. Chapter 2 will develop the concept of civic
legacies through a review of the literature on civic legacies and integration policy.
Chapter 3 will explain how to empirically track evidence of civic legacies using process
tracing and content analysis. Chapter 4 will lay out the variation in networks over time.
Specifically, Chapter 4 will establish what institutions and networks were developed in
the 1990s and activated again in 2015. Chapter 5 will lay out the variation in policy
frames and how policy entrepreneurs leverage them. Chapter 6 will focus on the 2015
housing policy change in Toulouse and the lack of policy change in Lille at this time.
This chapter will demonstrate the link between historical policy frames, NGO networks
and political entrepreneurs—civic legacies—and contemporary policies. Finally, Chapter
7 will conclude and describe the theoretical and policy significance of the findings. In
this way this dissertation demonstrates how civic legacies—policy frames, NGO
networks and policy entrepreneurs—influence migrant integration policy over time and
make leaders versus followers out of cities.
21
Chapter 2. Literature Review/Conceptualizing Civic Legacies
The central puzzle of this dissertation asks why Toulouse was a leader in
integration policy for migrant Roma while Lille was a follower. This dissertation
demonstrates how civic legacies shape integration policy (Chung 2020; de Graauw 2016;
Good 2016). Civic legacies are the historical ideas, networks and actors developed during
previous integration efforts (sometimes for different groups) that set precedent for
contemporary integration efforts (Chung 2020). While there is a rich history of literature
on the role of civic society in democracies, the focus on legacies, or the historical
influence of civic society is underexplored.
Civic Legacy Concept
First, it is necessary to define civic legacies versus civil society. Erin Chung
(2020) defines civil society using Gellner’s 1995 definition, “civil society as ‘that set of
diverse non-governmental institutions, which is strong enough to counterbalance the
state, and, whilst not preventing the state from fulfilling its role of keeper of the peace
22
and arbitrator between major interests, can nevertheless prevent the state from
dominating and atomizing the rest of society” (Chung, pg 7). Chung defines civil society
as non-governmental institutions. Chung (2020) then defines civic legacies as the
“interactions…between actors in the state and civil society and between civil-society
actors” (Chung, pg 25). For Chung, civic legacies include both non-governmental and
governmental actors. Chung does not define civic society, and for the purposes of this
dissertation, civil and civic society can be used interchangeably.
Earlier studies on civic legacies include Robert Putnam’s Making Democracy
Work (1993). Putnam developed the concept of civic traditions to explain why political
institutions in North-Central Italy are more democratic than Southern Italy. He claimed
that civic traditions formed through “associationalism,”—or spontaneous interactions
between private citizens—encourage civic engagement, more involvement in local
organizations and ultimately more horizontal governance structures and democracy. Sheri
Berman (1997) characterized Putnam’s argument as such,
“Like the mass society theorists, recent neo-Tocquevillean analyses stress the way
individuals relate to each other and their society when explaining why individuals
relate to each other and their society when explaining why democratic regimes
function well. To measure and explain the success of democracy, Putnam, for
example, uses the concepts of civic community and social capital; for both of
these the key indicator is what might be termed associationism, the propensity of
individuals to form and join a wide range of organizations spontaneously.”
This dissertation does not touch on why residents join organizations. Furthermore,
unlike Putnam (1993) and Chung (2020), this dissertation is focused specifically on
housing integration and not democratization broadly. Therefore, while the significance of
integration outcomes likely has bearing on democratization this is only surmised in the
concluding chapter of the dissertation when considering the significance of the project.
Berman’s (1997) main critique of Making Democracy Work was that
23
“associationism [which underpins civic society] should be considered a politically neutral
multiplier – neither inherently good or bad, but rather dependent for its effects on the
wider political context.” (Berman 427; emphasis added). Chung (2020) agrees with this
critique when she argues that for the disaggregation of civic legacies into strategies,
networks and ideas; “rather than assume that civil-society actors are unified in their goals,
worldviews, or strategies for democratic inclusion, or that civil-society actors necessarily
seek to incorporate immigrants in their local societies.” Instead, she “introduce(s) the
concept of civic legacies to refer to the ideas, networks, and strategies applied in previous
struggles for democratic inclusion that shape how civil-society actors make claims to the
state, mobilize public opinion, and build networks.” In this way,
civic legacies do not determine immigrant incorporation patterns, they inform the
discourse on citizenship and democracy within civil society. The language,
symbols, and strategies used in past struggles for democratic inclusion, then,
provide a blueprint for how civil society actors—including immigrants
themselves—negotiate exclusionary policies, understand democratic inclusion
and political empowerment, and engage with the state (Chung 2020; 25).
This dissertation supports Chung’s civic legacy theory by demonstrating that the
legacy of local NGO networks, policy entrepreneurs and specific policy frames provide a
precedent (both produce and constrain opportunities) for integration policy change. The
main point of difference between Chung’s theory and this dissertation is her focus on
historical democratization movements as the precedent for immigrant incorporation and
political engagement. This dissertation focuses on how previous integration efforts set
precedent for local integration policies. Instead of democratization movements
concerning other groups, this dissertation focuses on integration efforts for a similar
group, 30 years earlier. Therefore, the networks, ideas and actors involved should bear
even more similarity to each other and set an even stronger precedent than those than
24
Chung investigated. Those components of the argument are unpacked below.
Networks
While there has not been much focus on civic legacies in the integration literature,
there is work on how civil society actors affect a patchwork of integration outcomes (not
just housing), typically through activism and advocacy efforts. These studies do identify
which actors, networks and ideas matter for integration outcomes and why. However, this
literature does not focus on the process of path dependency—they either assume it,
include a descriptive historical explanation, or take a snapshot instead of a longitudinal
inquiry. For example, De Graauw (2016) finds NGOs act as brokers between migrants
and the city by supporting goods provisioning and mitigating nativist backlash. De
Graauw (2016) finds that NGO networks in San Francisco ensure access to integration
policies for migrants such as job opportunities and municipal ID cards. Kristin Good
(2016) in her study on Canadian cities does not focus on civic legacies but she does find
that integration outcomes vary depending on key stakeholders’ public/private partnerships
which determine how resources are pooled to implement policy (Good 2016, 13; Ireland
2004; Dancygier 2010). However, her argument does not have a strong historical
explanation. She mentions other scholars who argue that demographic shifts in the
ethnocultural composition of urban and suburban societies shape local governance. For
example, Kymlicka (2008) argues that multicultural societies are partially conditioned by
the ethnoracial configuration of historical flows of migrants,
If multiculturalism in Canada had initially been demanded by groups who were
perceived as having strong religious/cultural commitments to illiberal practices –
say by Somalis or Saudis rather than Ukrainians and Italians – and if their demand
for multiculturalism had been perceived as a demand that such practices be
tolerated and accommodated, then multiculturalism would probably not have
arisen’ (ibid., 5).
25
Good (2016) takes history into consideration by describing the different
multicultural periods or stages in Canada based on policies and initiatives associated with
multiculturalism. These are useful ways of organizing multicultural regimes but she never
argues for path dependence or how a preceding era conditions the next. Good’s argument
is not a historical one, instead she focuses on resources and argues that cities that have
more resources and can better coordinate between private and public relations to use
those resources tend to be the leaders of urban regimes; eg Toronto. This dissertation
focuses on one type of integration policy, housing for one group, EU migrant Roma so
that the theory of civic legacies can focus on the historical influence that networks play to
explain contemporary policies.
Chung (2020) does explain the historical legacies of networks in her book.
Networks are comprised of the “’organizations, networks, traditions, and solidarities that
include policy networks and informal networks between civil-society and state actors”
(Chung, 28). Networks mediate immigrant incorporation patterns in policy and practice.
Civil-society actors can play a significant role in shaping immigrant incorporation by
interpreting, negotiating, and enforcing policies pertaining to immigrants, holding
national and local state actors accountable, giving voice to migrant concerns, raising
public awareness and pushing for policy reforms…”(Chung 30). Through their
relationships, or network, with city council, civil society actors attempt to hold
government actors accountable, give voice to migrant concerns and push for service
provisioning.
According to Chung (2020), the civic legacy of the network is established by the
vanguard group, which developed the network during “previous struggles for democratic
26
inclusion…” (Chung, 30). This vanguard group has the institutional power to advocate
for the new migrants based on its historical experience with integration. In Toulouse, the
key organization that worked with Roma migrants in the 90’s was still a lead organization
in integration for EU migrant Roma through to 2019. Therefore, this organization
comprised and developed the vanguard network that influenced outcomes later on. Chung
explains the role of vanguard civil-society organizations may be to “provide guidance and
logistical support to migrant activists, open or rebrand their campaigns to provide migrant
advocacy, help migrant advocates establish new allies, and/or retool themselves as
migrant advocates.” (Chung, 30). The main role of the NGOs working in Toulouse and
Lille was not focused on activism as much as service provisioning. The role of NGOs was
more broadly to generate trust and information sharing between the immigrants, social
workers and city council.
But how does the vanguard group within civil society have the greatest effect on
integration? Chung finds that first, “The reach of influence of the vanguard group is
based on, “its capacity to provide symbolic significance and legitimacy to a new
movement” (Chung 30). This is the civic legacy of policy frames. And second, “the
vanguard group’s established connections with other civil-society actors and state
officials at the local, national and international levels (Chung 30).” This is the role of
policy entrepreneurs who mediate the vanguard’s established connections with state
officials. First the civic legacy of policy frames will be explained and then the civic
legacy of policy entrepreneurs.
Policy Frames
One way the NGO network provides symbolic significance and legitimacy to
housing integration efforts for migrant Roma is through policy frames. In her analysis of
27
integration rights and policies for migrants in San Francisco, De Graauw (2016) finds that
policy frames are customized and introduced by NGOs to connect with city officials,
voters and the media, in order to successfully navigate “a highly politicized environment
in which their objectives remain a difficult political sell” (De Graauw, 17). De Graauw
argues that advocates for migrant integration use certain policy frames to downplay how
integration policies directly benefited immigrants. Instead, they framed these policies as
benefitting all city residents’ civil and labor rights,
They used different advocacy frames to establish this linked fate in each of the
campaigns. They framed language access legislation as a civil rights issue and a
much-needed local attempt to enforce existing state and federal language access
provisions (chapter 3). In contrast, they used the lens of economic justice to frame
labor rights legislation (chapter 4), and they framed municipal ID cards legislation
as a public safety and civic integration issue (chapter 5). By strategically
downplaying that these policies primarily benefited immigrants, immigrant-
serving nonprofits successfully navigated a highly politicized environment in
which their objectives remained a difficult political sell.
By customizing policy frames, De Graauw argues that NGO advocates were able to
create political will for policies that were initially unpopular. This dissertation supports
De Graauw’s findings that policy framing can help develop the political will for
integration policy for stigmatized and discriminated migrant groups. How the housing
policy for migrant Roma is framed (e.g. through the lens of humanitarian rights, human
rights, feminism, civic integration) influences how the policy is perceived by residents
and representatives alike and whether there is enough political will at the city council
level to introduce the policy.
Chung (2020) defines policy frames as “the ideational frames through which
debates and policies on immigration and citizenship operate” (Chung, 26, 27). She goes
on to explain that debates about immigration and citizenship are “closely tied to
collective understandings of nationhood,” and that some collective understandings of
28
nationhood are “rooted in ideas about race and racial inequality, class domination,
patriarchy, caste and colonial classifications, for example” (Chung 26, 27). While it is
true that the policy frames at the focus of this inquiry could and often do implicate
collective understandings of nationhood (rooted in ideas about race, class, patriarchy,
colonialism, etc), at the local level other frames are frequently invoked that concern
collective understandings of the city, the urban identity. Policy frames in Toulouse and
Lille considered topics such as integration, social support, proactive policy, precarity,
crime and human rights generally. The heterogeneity of the local level policy frames are
more eclectic than the hegemonic and homogenized policy frames at the national-level.
According to Chung (2020), policy frames structure debates over integration
policy and provide tools for migrant advocates, “while ideational frames may constrain
immigration policy debates and reforms, they also provide the tools for making claims
legible to the state and the larger public” (Chung, 27). Because of the lessening of the
nationalism policy frame and the diversity of the policy frames at the local level, city
actors and networks are freer to be creative with how they use which policy frames to
further their agenda. Therefore, the most successful policy frames are not necessarily the
most obvious, they can be place-based and historically informed, which is where the idea
of the civic legacy of a policy frame is useful.
Chung (2020) then defines the civic legacy of policy frames as the frames used in
previous struggles for integration which can be used as contemporary housing policy
frames and shape how civil-society actors negotiate contemporary challenges to housing
integration. “Struggles over citizenship status, identity, and rights…inform later debates
and movements for democratic inclusion...ideational frames used in previous struggles
29
for democratic inclusion shape how (1) civil-society actors negotiate exclusionary
policies in their attempts to gain rights for migrants” (Chung, pg 27). The civic legacy of
policy frames provides a repertoire of policy frames that have precedence in the
community. These policy frames have historical significance and a different semiotic
weight. As such, if actors invoke a specific frame with a civic legacy, this frame may
have more resonance in the community, which might garner the necessary political will to
introduce one policy over another.
This dissertation considers policy frames regarding housing policy for EU migrant
Roma. There is a lot of important work on how policy frames regarding specific migrant
communities (such as Roma) inform housing policy, but this is a different research goal.
Policy frames regarding migrant Roma securitization, mobility, economic deservingness,
morality, ethnicity, culture, lifestyle, citizenship, European-ness, and governmentality—
all of these frames inform integration policy (Van Baar 2014, Manca & Vergnano from
Magazzini and Piemontese 2019, Kostka 2015, Chapter 3 of Triandafyllidou and Spencer
2020; Chauvin and Garcés-Mascarñas, Foucault 1991, Scott 1998, Vermeersch 2012,
Vermeersch 2012, Guy, 2003; Molnar Diop, 2013, Vašečka & Vašečka, 2003, Benedik,
2010, Kóczé & Trehan, 2009, Kocze 2018, 460). However, this project does not ask how
ideas regarding migrant Roma influence housing policy, instead this project looks at how
policy frames regarding housing differentially influence housing policy. The difference is
nuanced but important, because it is a major contribution of this dissertation. For this
dissertation, the policy frames of interest concern the policies themselves, not the groups
who are targets of the policies. Next, the role of the policy entrepreneurs who articulate
the policy frames will be explained.
30
Policy Entrepreneurs
The literature on discursive institutionalism is useful in understanding how policy
entrepreneurs articulate policy frames and coordinate actors (Keck and Sikkink 1998;
Schmidt 2008). Policy entrepreneurs articulate policy frames and demonstrate a
sophisticated awareness of the discursive field in which they are operating (Schmidt
2008; 310). The literature on social movements that uses the term political entrepreneur
suggests that policy entrepreneurs can draw from their own personal experience in
government or activist movements. Political entrepreneurs, as opposed to policy
entrepreneurs, mobilize resources like information and membership and show a
sophisticated awareness of the political opportunity structures in which they are
operating. Policy entrepreneurs demonstrate this same savvy and level of coordination
across networks and institutions except at the local level instead of the trans-national
level studied in Keck and Sikkink’s work (2005). Policy entrepreneurs are also
specifically focused on policy change as opposed to political change more broadly.
Furthermore, this project explores how policy entrepreneurs can draw upon the civic
legacies of networks and policy frames coordinated by previous policy entrepreneurs to
convince political opposition of particular policies.
Chung (2020) does not include policy entrepreneurs in her conceptualization of
civic legacies because she is not interested in “elite-based approaches (that) focus
exclusively on policies and policymakers,” because “they tend to overlook how policies
are implemented, negotiated, and contested on the ground” (Chung, 27). However, policy
entrepreneurs do not have to be elites and this explanation does not focus only on
policymakers. This explanation includes the role of individuals, yes, however a policy
31
entrepreneur can be positioned in civil society, outside of government, as well as inside
government as either bureaucrats or representatives. Furthermore, at the local level as
opposed to the national, individuals have a greater effect on political outcomes, so it
makes intuitive sense to focus on the agency of individuals to a certain extent.
When a policy entrepreneur introduces a policy frame into the political arena; she
is introducing possibilities. Policy entrepreneurs can introduce an idea rooted in a
different time or place and force it into the present; sometimes this creates a rupture or a
discontinuation within the predominant policy frame. As suggested above, some policy
frames have historical significance and a different semiotic weight. As such, if actors
invoke a specific frame with a civic legacy, this frame may have more resonance in the
community. Policy entrepreneurs that are aware of which policy frames have more
resonance are better able to garner the necessary political will and “provide symbolic
significance and legitimacy,” to introduce one policy over another (Chung 30). This is the
mediating and coordinating effect that a policy entrepreneur can wield in order to
convince political opposition of the merits of a specific policy.
Policy entrepreneurs articulate policy frames with a civic legacy, but they can also
be part of a civic legacy. Policy entrepreneurs with a civic legacy had a role in
perpetuating the policy frames sometimes within the vanguard network that set a
precedent for the policy frames, policy entrepreneurs and networks that would support
integration policy for migrant groups later. These actors are not necessarily leading
national level social movements for democracy, but they are playing a necessary role in
developing the political will to introduce integration policy for stigmatized groups in
societies with high levels of racism and discrimination.
32
The contribution of this research is an innovative focus on the historical
precedents of integration policy at the urban level. Within integration literature, the focus
is on national-level policy, as if the city-level does not have the most effect on policy-
outcomes. The integration literature also overlooks the historical legacies of cities,
including NGOs, policies actors and frames. This focus on civic legacies contributes
something slightly different than research on policy legacies, policy feedbacks or path
dependence provides. An inquiry into civic legacies is an inquiry into the potential for
policy change and transformation, which is different from the process of continuity that
path dependency suggests. The theoretical motivation is to understand how civic legacies
—including policy frames, networks and entrepreneurs—could shape contemporary
integration policies of a city and sets precedent for some cities to lead while others follow
in their housing policy for stigmatized groups. The next chapter details the methods and
methodology that is used to operationalize and measure civic legacies’ influence on
integration policy.
33
Chapter 3. Methodology/Measuring Civic Legacies
How to measure a civic legacy? To measure the influence of networks and policy
entrepreneurs this project will use process tracing and map the networks and actors
involved in the housing process using metropolitan and municipal level archival
documents and interview data between (1990-2022). To measure the influence of policy
frames this project will conduct a content analysis of the same data (1990-2022). This
project will also create a timeline of housing policy for migrant Roma using the same
data to trace how the policies change over time. This project employs an NVIVO content
analysis of municipal and metropolitan archives, online municipal and metropolitan
meetings and interviews with municipal and metropolitan stakeholders concerning Roma
migrant housing policy in Toulouse and Lille between 1990-2022.
Measuring Policy Eras
Chart 1 reflects the four relevant policy eras in Toulouse and Lille. I analyzed
municipal and metropolitan archives and online metropolitan and municipal level
meetings minutes to demarcate the different policy eras. Each new era represents a shift
in housing policy for migrant Roma. The process tracing began in 1990s—when NGOs in
Toulouse began to conduct surveys of the migrant Roma population in their city—
through to 2022. As I documented potential policy eras, I noted potential critical policy
34
junctures (moments in policymaking when counterfactual policy outcomes could be
imagined) such as when surveys and re-housing efforts were conducted, when camp
evictions were halted and when new housing policies were introduced. In Toulouse,
rehousing efforts of migrant Roma in 1998 was an important juncture during which the
social support network was activated. The next critical policy juncture was the
introduction of integration villages in 2007, and Toulouse’s pivot to long-term
autonomous housing with social support in 2015, followed by the national decision to
implement Housing First in 2017. I noted these critical policy junctures on a timeline of
policies, actors, networks and events from 1990-2022. At these critical junctures I noted
which actors and organizations were involved and the extent of their involvement.
Measuring Networks
The literature review on networks established that networks create and solidify
connections amongst civil society and government actors. Therefore, using the
comparison described above, I used inductive reasoning to identify important
organizations and actors, which comprise the civic legacy of the network. After creating
the timeline in NVIVO based relevant policies, events, actors and networks, I noticed
which organizations and actors were relevant to these changes. The number of mentions
of an organization or individual was the greatest indicator of the relevancy of that actor in
the network of actors involved in housing policy for EU migrant Roma. Another metric of
relevancy was mentions over time. I identified instances when the same organization or
individual kept appearing in the timeline of critical policy junctures even as years passed.
For example, if an organization was mentioned in the 1990s and mid 2010s, this was of
interest to me because it demonstrated an institutional and historical continuity, and
potentially a civic legacy. When certain actors and organizations began to distinguish
35
themselves in these ways, I then kept an eye out for these in the archival and online
documents.
I would cross reference my timeline of actors and policies with the archival and
online documents and interviews to get a sense of the substantive decisions and roles the
actors played in the process of housing for migrant Roma. Through this process I
understood which organizations were mediating between civil society and the local
government, advocating on behalf of migrant Roma, holding government accountable,
giving voice to migrant concerns and raising public awareness for policy reforms and in
turn shaping integration policies; all key aspect of civic legacies of networks found in the
literature (Chung pg 30).
From this, I understood how the relevant actors and organization interacted with
each other within the network over time and how they established and solidified
connections amongst civil society actors and government. Next, I gleaned who were
potential policy entrepreneurs, and which policy frames were most relevant over time.
Measuring Policy Entrepreneurs
Policy entrepreneurs articulate policy frames with a civic legacy, but they can also
be part of a civic legacy. The policy entrepreneur has the power to change policy. In this
analysis, actors with a level of political influence who articulated a specific policy frame
consistently over time were identified as a policy entrepreneurs. However, an effective
policy entrepreneur can draw upon the civic legacies of networks and policy frames and
convince the political opposition of particular policies. It is unlikely that the analysis
would have missed a relevant effective policy entrepreneur due to the triangulation of
meeting minutes, archives and interviews. It is unlikely that a relevant policy
36
entrepreneur was influencing policy at the time but was not mentioned in any of these
data sources.
Measuring Policy Frames
The literature on policy frames expects policy frames to structure debate over
integration policy and provide tools for migrant advocates to push for policy for migrant
Roma (Chung 2020). The literature defines policy frames as the ideational frames
through which debates and policies on integration operate (Chung 2020). These frames
should be customized and introduced by actors to connect with city officials, voters and
the media to navigate a “highly politized environment,” and develop the political will for
a policy (De Graauw 2016).
To measure policy frames, this project will use a policy frame analysis to analyze
the archival, online and interview data and generate a list of relevant policy frames.
Policy frame analysis is useful for analyzing policies beyond the level of factual disputes.
Policy frame analysis defines what is a stake in policy debates, the differences in
worldviews and policy values (Schmidt from Yanow and Schwartz-Shea pg 323). as “a
system of representation that has developed socially in order to make and Through the
process of defining what is at stake, the differences in worldviews and values, policy
frames generate associated meaning, or ideational frames, attached to a policy. In this
way, the policy frame acts circulate a coherent set of meanings” (Weldes from
SchwartzShea and Yanow 2014, 230). The policy frames are articulated by specific actors
in specific contexts. Who articulates the policy frame should give us a sense of the
relevant network and policy entrepreneurs as well.
The literature also defines the civic legacy of policy frames as frames that have
been used in previous struggles for integration that can be used as contemporary housing
37
policy frames and shape how civil-society actors negotiate contemporary challenges to
housing integration. The civic legacy of policy frames provides a repertoire policy frames
that could have precedence in the community. In this way, a policy frame analysis
contextualizes a policy frame within the historical repertoire of policy frames. These
policy frames have a historical significance and a different semiotic weight because of
that significance. This semiotic weight helps develop the political will for particular
policies. Furthermore, a policy frame analysis method avoids implicitly reifying policy
debates and policy beliefs.
The benefits of a policy frame analysis are multifold. Applying a longitudinal and
comparative policy frame analysis provides an explanation that is less taken-for-granted
and more contextualized. I can use a model that historically considers how policy frames
define debates and issues at play and, in turn, shape policy. I can also identify the absence
of a policy frame – or the presence of marginalized or subjugated policy frames—at the
same time as focusing on the hegemonic policy frames articulated by policy
entrepreneurs. Therefore, while policy frame analysis is time intensive and the findings
are descriptive and not often causally determinative or parsimonious, it is a detailed and
historically contextualized analysis, lending a richness of understanding to the subject
matter. The next section delves further into the method of policy frame analysis.
Content Analysis in NVivo
To measure the policy frames, I conducted a content analysis in the program
NVivo. There were three coding steps to this content analysis: 1) code policy frames 2)
describe the context and the actors involved 3) code arguments and core values/themes.
The first step was to code basic policy frames that arose inductively from the archival
documents and city council meeting minutes. During this first step I asked myself, “What
38
are the categories being used in this policy issue?” During this step it was important to
generate as many categories of analysis as possible. So, I coded any concepts that were
relevant to “housing for EU migrant Roma,” and I took an expansive historically-based
definition of “Roma.” Therefore, any concepts regarding housing for “tsiganes,”
“travelers,” “gens du voyages,” “nomades,” “Roma,” and “migrant Roma,” were all
included (explained further in chapter 4). In the program NVIVO the policy frames in
this first step are referred to as “codes.” To “code,” means I highlighted the relevant
group of words that reflected a specific policy frame, such as “humanitarianism.” I knew
it was better to create “too many,” codes at the beginning and then pare down the relevant
codes later. I created a new code every time I came across a concept that reflected a
policy frame in the corpus. After I highlighted and identified the group of words as one
instance of a code, that group of words became a “reference” and was filed automatically
under a code such as “humanitarian.” After I had gone back into the qualitative data to
contextualize the codes and references, I pared down which codes were relevant for
which city and when I went back over the data to look for saturation, I paid more
attention to these codes and looked for references that I may have overlooked in my first
round of coding.
The second step of the content analysis is to identify the actors who articulate
policy frames, or policy entrepreneurs, sometimes referred to as protagonists. In order to
identify the main policy entrepreneurs, I asked, “What do they want? Why do they care
about this issue? What is their view of the social context.” It is also important to identify
the “backgrounds, motivations, key positions in the debate,” of each actor. Then, ask
39
“what is at stake for whom,” “what kind of arguments are actors making repeatedly?”
This step analyzes the relationship between communities and actors and can also help
identify relevant networks.
The third and last step is the identification of policy frame values. During this
third step I identified the policy arguments and core value disagreements. In other words,
in this step I attached contextualized meaning to the phrases, actors and concepts that
were coded in steps one and two. Instead of “categories,” some scholars refer to “threads,
patterns, themes, regularities, topics, constructs, actions, events and repeating ideas”
(Packer 2010). These arguments are unpacked in chapter 5. Relevant questions for this
step are, “what are the central arguments?” and “what are the core values that underlie
these arguments?”
Next, I identify the “silences.” I ask what arguments are not being used, which
policy frames are not referenced, and which actors’ voices are not present (JC. Scott
1990; Spivak 2009). I also ask what assumptions are being made by actors when they use
a specific argument. This questioning gets at the “taken for grantedness,” or logic of
appropriateness in the themes (HS Becker 1998) (167), Schutz (1967, 74).
I stop the process of coding when codes, actors, values and policy frames begin to
repeat themselves across texts. In her methodology on interpretive social science Milliken
states that, “An analysis can be said to be complete (validated) when upon adding new
texts and comparing their object spaces, the researcher finds consistently that the
theoretical categories she has generated work for those texts (Milliken 1999, 234).
Internal validity is achieved when policy frames repeat themselves across interview,
archival and council meeting minutes documents.
40
The content analysis resulted in a ranking of the most common codes; a snapshot
of the top 20 policy frames in both cities (Chart 4). After I identify the timeline of
housing policies for migrant Roma and completed the policy frame analysis, I place the
results of the policy process tracing and the policy frame analysis side by side and
produced Figure 1. Using this figure I can check if any changes in the policies coincide
with an uptick or drop in aggregate number of code references per year. If there is an
uptick in references, I check the policy timeline to see whether there was a critical policy
juncture. For example, in Toulouse in 1990 there was an uptick in code references with
87 references counted. This is reflective of the policy development that was introduced
by a specific policy entrepreneur at a specific NGO. In this way, the policy frame analysis
triangulates the findings from the process tracing and can further substantiate findings
concerning the relevant networks and actors involved in civic legacies.
Chart (3) Policy Frame Analysis
Policy Frame Analysis
1. Identify the issue and
protagonists
- “What is the problem or situation being
argued about?
- How do the competing camps in the
policy identify themselves in relation
to each other?
2. Describe the context and the
protagonists’ policy proposals
- Make clear what is at stake for whom,
this step is essentially articulating the
relationship between the policy
entrepreneur and the policy frame.
- Identify, “backgrounds, motivations,
key positions in the debate.”
- Identify subject-positions, interpretive
communities and networks
- “What do actors want? Why do they
care about this issue? What is their
view of the social context.”
- Acknowledge resources that go toward
sponsoring certain frames, the prestige,
41
reputation and networks associate with
a specific subjectposition.
3. Describe the Arguments and
Core Values
-Identify relevant concepts or
“relevant theoretical abstractions,” -
“What are the categories being used
in this policy issue?”
- “Threads, patterns, themes,
regularities, topics, constructs, actions,
events, repeating ideas”
- What are the factual disputes versus
value disagreements?
Corpus
The corpus is a representative sample of government documentation on housing
policy for migrant Roma in both metropolitan regions, Toulouse and Lille. The corpus
used for the content analysis to identify policy frames includes metropolitan-level
(inclusive of municipal level as well) government documentation concerning housing
policy for EU migrant Roma (and earlier populations categorized as “nomade” or
“tzigane), inclusive of the years 1990s-2022. The corpus represents a history of housing
policy for migrant Roma in the metropolitan region. The archival documentation included
municipal and metropolitan policy announcements, police reports, letters and emails
between departments and between citizens and the city, local and national newspaper
articles, and surveys and reports produced by local NGOs. The corpus includes
supplementary materials from non-governmental organizations (civil society), such as
NGOs, housing associations, as well as national-level policy memos and reports. These
documents were included in municipal and metropolitan archives and were included in
the corpus, as these archives were curated by the city. To be clear, the corpus does not
include a representative sample of newspapers or journalism regarding housing policy.
The only newspaper articles included in the analysis were ones that were archived by the
city.
42
To construct the corpus, in the archives (online and in-person) I searched the key
words: “Roms,” (Roma) “gen(s) du voyage,” (french traveller(s)) “migrant(s),”
“tsiganes,” (“gypsy/gypsies”) “nomade(s),” “bidonvilles,” (slums) and
“logement(s) ”(housing) and pulled whatever was available. I analyzed a total of 1344
archival items. I collected 340 items from the Lille municipal and metropolitan archives
and 293 items from the online archives to bring a total of 633 items analyzed for Lille in
NVIVO. I collected 514 items from the online Toulouse archives and 165 items from the
Toulouse municipal and metropolitan archives.
The Role of Interviews
Throughout the dissertation, I mention that interviews informed some of my
insights. I conducted 36 stakeholder interviews. These included interviews with a policy
analyst at the European Commission, a journalist and lawyer, municipal and
metropolitan-level employees in housing and integration sectors, NGO members, and
migrant Roma families living in public and private housing. These interviews helped me
to scope this project. For example, when I began research on housing for migrant Roma, I
began at the level of the EU Commission since at the beginning of my fieldwork it had
just evaluated the EU Framework for National Integration Strategies up to 2020. Between
2014 and 2020 alone, the European Commission earmarked around 1.5 billion euros for
the “socio-economic integration of marginalized communities such as Roma,” with most
investment allocated for education and housing (European Court of Auditors 2016 10).
In other words, when I was conducting interviews in 2020, 10 years of political
will and funding had just been focused on Roma integration in both Western and Central
European countries. When I asked an EU policy analyst who worked on social inclusion
43
for Roma at the European Commission, “to what extent the EU has influenced the
housing process for EU migrant Roma in specific cities?” The interviewee suggested that
while the EU had clear legal responsibility when it came to policy development, outside
of this, housing for Roma is not an EU competence.i The most he suggested they
can do is coordinate and help member states if the member states solicit their help, as a
mutual exercise between both member states and the EU.ii After this interview, I began to
scope my attention to the city level. I interviewed metropolitan housing officers and NGO
workers who worked on housing for migrant Roma. It became clear I was speaking to the
right people as they were able to speak to policy changes that affected housing outcomes
in the city and they could identify the actors and timelines directly involved in my
outcome of interest, housing policy over time. Therefore, based on these preliminary
interviews, I decided to scope the dissertation to the urban (municipal and metropolitan)
level.
From the interviews with migrant Roma families I understood that there is a great
heterogeneity to the types of housing that migrant Roma families are interested in. It was
not the case that the migrant Roma families in Toulouse had systematically different
preferences in housing than those in Lille. From some of my interviews, I began to get a
sense that the matriarchs of families might have a strong housing preference if there was
a personal history of domestic violence, however I did not gather enough data for
systematic inferences. Furthermore, when I researched the effects of civic legacies on the
housing integration of migrant Roma, I did not find any patterns regarding domestic
violence and housing policy for migrant Roma families. Overall, housing preferences
44
were heterogenous depending on the families' life circumstances. Some preferred housing
near their families, some preferred housing on their own, others preferred housing in the
urban center, others on the periphery or the countryside. Most wanted stable housing and
would take what they could get. If anything, this was the commonality that I found in my
interviews.
This might be partially due to the selection bias of the group of families who I
spoke to. In Toulouse and Lille, I was reliant on NGO members’ pre-established contacts
with families to get interviews. In other words, the families that I could interview were
the ones who were seeking and receiving help from NGOs. Some were grateful for the
situation they were in and had lived there for years, others were seeking improved
accommodation options. While most families expressed they were grateful, there was a
clear power dynamic at play. I always conducted interviews with NGO members in the
room with me, so it was unlikely they would express something other than gratitude for
their help navigating the housing sector while the NGO member was present.
While I enjoyed conducting interviews immensely, COVID impeded my ability to
conduct as many interviews as I would like since my time in the field reduced from a year
to three months. Furthermore, translating through a mask was very challenging but I did
not want to put either myself or the families at risk. Mostly, I used the interviews to direct
me to relevant actors, events and policy-developments at the city level, as explained
above.
Conclusion
The process tracing identifies networks, actors and policies, while the content
analysis identifies the policy frames and associated actors and networks. From both of
these analyses, I am able to infer when, how and which policy frames, networks and
45
actors shaped housing policies over time for migrant Roma at the urban level. Therefore,
I can answer the question: why did Toulouse lead housing policy for EU migrant Roma
while Lille followed? My research finds the civic legacies developed in Toulouse set the
precedent for leadership in housing policy; while the lack of a strong civic legacy in Lille
set the precedent for it to be a follower.
The following chapters will lay out what aspects of a civic legacy were developed
when and how they influenced contemporary housing policy. Chapter 4 will lay out the
variation in the networks, policy entrepreneurs and policies regarding housing integration
for migrant Roma in Lille and Toulouse identified through process tracing. Chapter 5 will
lay out the variation in policy frames in the civic legacy in Toulouse and Lille using the
content analysis. Chapter 6 will investigate the specific policy change (in the case of
Toulouse) in 2015 to demonstrate how a legacy of local NGOs, policy entrepreneurs and
specific policy frames can result in independent policy change while the absence of this
legacy results in delayed policy change through the adoption of national policies (in the
case of Lille). Chapter 4, 5, and 6 present the empirical evidence that civic legacies
developed in previous integration efforts set the precedent of whether policy
entrepreneurs can innovate and lead housing policy change in the future.
46
Chapter 4. Establishing a Civic Legacy: Divergent Networks in Toulouse v. Lille
Why did Toulouse pivot from integration villages toward long-term housing in 2015
and emerge a leader in housing policy for migrant Roma while Lille was a follower? This
chapter compares the role of networks in Toulouse and Lille. The evidence suggests that
Toulouse developed networks during the integration of migrant Roma in the 1990s which
set a precedent for its city council to shape new housing policy in 2015 and empowered it
to emerge as a leader in housing policy for migrant Roma. Alternatively, Lille lacked the
development of civic legacy networks in 1990 to set precedent for 2015 and thus did not
change their policy until there was a national level change in policy and emerged as a
follower.
Using process tracing, this chapter identifies important organizations and actors,
which comprise the civic legacy of the network. As a reminder, the literature on networks
established that networks create and solidify connections amongst civil society and
government actors. The network also generates information-sharing and trust between the
target group, social workers and city council that is necessary for successful housing
policy. The civic society network can play a role in shaping integration policy by holding
national and local political actors accountable, giving a voice to migrant concerns, raising
public awareness of policies and advocating for certain policies and reforms (Chung
2020). The civic legacy of a network is developed over time by the vanguard group. This
vanguard group has the institutional power to advocate for the new migrants based on its
historical experience with integration. The vanguard group develops the network during
previous efforts to integrate the same or similarly discriminated migrant groups that can
require more support from the government.
47
This chapter organizes the evidence which demonstrates how the civic legacies of
networks influence integration policy in Lille and Toulouse. The evidence demonstrates
that through the process of providing housing options for the group, a civic legacy was
developed in Toulouse. In Toulouse, the associative network mobilized in the 1990s
during an effort to re-house migrant Roma set the precedent for housing policy change in
2015. Dr. Jean-Claude Guiraud ran the pivotal organization, “Association Tsiganes
Solidarité” (ATS), which was involved in the re-housing process and solidified
connections amongst civil society and government actors for nearly 30 years between
1971-1998. Dr. Guiraud established the trust, mediation, and expertise for the social
support of migrant Roma families. This vanguard group in Toulouse, ATS, had the
institutional power to advocate for the new migrants in 2015 based on its historical
experience with integration in 1990. This role of this network was to generate the
information and trust between the target group, social workers and city council that was
necessary for housing policy. The civic legacy of this network that was established by
ATS in 1990 during re-housing efforts was activated again when new leadership took
control in 2015. Eurocities, a transnational NGO, reported that in 2015, Toulouse had
changed its approach from re-housing migrant Roma in integration villages to re-housing
migrant Roma in long-term autonomous housing.43 Toulouse was a leader in making this
policy switch to long-term autonomous housing from integration villages, and continued
this policy through 2020.
43 15 cities learn from Toulouse how to support Roma to access adequate housing,” 2019, Eurocities.
48
http://www.eurocities.eu/eurocities/news/15 - cities - learn - from - Toulouse - how - to - support - Roma - to - access-
adequate-housing-WSPO-BBDLD3.
Lille, in comparison, lacked the civic legacy of a network in the 1990s. Instead of
a network with a civic legacy and a vanguard group that was interested in the rights of
migrant Roma, there is little evidence that Lille had a civic legacy of networks which
could have set precedent for policy change in 2015. This does not mean that Lille did not
have a civic society interested in housing policy change for migrant Roma in 2015, nor
does it mean that Lille did not have some evidence of a civic legacy. However, Lille’s
civic legacy of networks was not developed as early as the one in Toulouse and did not
have the same level of institutional continuity over time(the organization ATS lasted
nearly 30 years in Toulouse), which makes it more difficult to establish the path
dependency necessary to make the historical argument for Lille. By 2007, it appears that
Lille introduced integration villages because there was EU funding for integration
villages available. Then, later in 2017, Lille changed its policy on integration villages
only after a national level policy was introduced. In these ways, Lille emerged as a policy
follower and not a policy leader.
This chapter will organize the evidence around the four housing policy eras
reflected below in chart 1. The first era is policy development, during which Toulouse
develops its civic legacy of networks through its rehousing policy for migrant Roma and
Lille does not develop its civic legacy, instead focusing on policy only for French
travelers. The second era is a period of policy similarity, during which Toulouse and Lille
both introduce integration villages because of EU funding. The third period is the era of
policy divergence, when the civic legacy of Toulouse empowers it to switch from the
integration village policy to long-term autonomous housing policy while Lille continues
49
its integration village policy without a network with a civic legacy to encourage policy
change. Finally, the last period is the era of policy continuity, when Toulouse continues to
rely on its new model of long-term autonomous housing and Lille continues to rely on
integration villages in addition to introducing the national level policy on long-term
autonomous housing. This chapter will establish the civic legacy of networks in Toulouse
which empowered it to lead housing policy for EU migrant Roma, while the lack of a
civic legacy of networks in Lille compelled it to follow national-level policy. The next
chapter, chapter 5, will focus on the influence of the civic legacy of policy frames.
Chart (1): Housing Policy Eras in Toulouse and Lille
Policy Eras Toulouse Policy Lille Policy
1990-2007: Policy Development.
Rehousing efforts and social
support in Toulouse but not Lille.
Some focus on migrant
Roma in re-housing
efforts. Surveys of
migrant Roma conducted
by social services
network.
No focus on migrant
Roma. Focus on
parking sites French
travelers, “gens du
voyage. » No surveys
of migrant Roma
conducted by social
services network.
2007-2015: Policy similarity.
Integration villages in both cities.
Integration Villages in
addition to other types of
housing, moratorium on
evictions (2010-2012), an
increase in integration
villages after 2010.
Integration Villages in
addition to other types
of housing,
moratorium on
evictions (20102012),
an increase in
integration villages
after 2010.
2015-2017: Policy Divergence.
Toulouse pivots from integration
village policy and Lille continues
with integration village policy.
City council phases out
integration villages,
relies primarily on
autonomous housing
with social support and
preempts national-level
policy. One integration
village remains.
Despite closure of a
few integration
villages (from 10-7)
continued reliance
upon them.
50
2017-2020: Policy Continuity.
National Housing First policy is
introduced.
In City Council, longterm
autonomous housing with
social support is referred
to as a
Under national
Housing First policy,
autonomous longterm
housing
success and the model for
future integration of
refugees in Toulouse.
becomes policy while
different types of
integration villages
still used.
First Housing Policy Era (1990s): Policy development, Rehousing efforts in Toulouse but
not in Lille.
Toulouse
In the 1990’s Toulouse began re-housing migrant Roma living in camps with the
aid of NGOs such as A.T.S. while Lille did not. CCPS was part of ATS which was
founded in 1971 by Dr. Jean-Claude Guiraud, who remained its President until 2001. This
NGO support would generate the civic legacy of networks that the city would later draw
upon. By 1991 Toulouse, with the help of ATS, had conducted a census of families in
stopping sites and camps and described 17 clans of travelers around Toulouse and its
metropolitan region. Toulouse managed several camps (Ginestous, Martres and
Mounède for example) for migrant Roma, while Lille mainly managed stopping sites for
French Roma. Toulouse counted 160 families between Camp de Martres and Camp de
Ginestous.
By 1999 Toulouse had recorded the demographics of families in their camps. In
1998 there were around 600 families with a French background and 25 families (around
100-150 people) who were Bosniac, Romanian or Bulgarian. The camp Ginestous was
the largest and had the most diversity with 242 Roma people from “French, Spanish,
German, Bosniac, Romanian” backgrounds. This level of recorded diversity within Roma
families is one way in which the Toulouse integration effort differed from Lille early on.
51
A.T.S. conducted all of these surveys and reported them to the city council. To be clear,
while this is the first instance of recorded diversity within Roma families settling in
Toulouse, this is not the first time Roma (migrant or otherwise) were put on a registry in
France. It is illegal to collect information on ethnicity and race in the French census
because of this history.49
In 2001, ATS turned into the « Coordinating Committee for the Promotion and
Solidarity of Migrant/Gypsy Communities in Difficulty (CCPS).” The CCPS NGO
played a central role in supporting families in their housing efforts and other integration
efforts as well as mediating between the city and the families. This is the origin of the
evidenced presence of a civic legacy of networks, particularly that of the vanguard group
A.T.S.
Lille
In Lille, in 1991, the city was working with the NGO Friends of French
Travellers, “Amitie aux Gens du Voyage,” to provide parking sites for French Roma. In
2001, Martine Aubry (PS)(center-left) was elected the Mayor of Lille, taking over from a
previous center-left Mayor. Philippe Douste-Blazy of the center-right (UMP) was elected
Mayor of Toulouse, taking over from a previous UMP Mayor. In 2004, the UMP was led
by Jean-Luc Moudenc who was elected Mayor.
Second Housing Policy Era (2007-2015): Period of Policy Similarity, Integration Villages
In 2007, the new policy of integration villages was introduced in both Toulouse
and Lille. The EU identified integration villages as effective housing options for EU
migrant Roma and funded efforts across France to introduce integration villages. Between
52
January 1, 2007 and December 31, 2013, the EU funded integration villages for the
“support and housing of marginalized populations,” set at 332,862,73 Euros. Unlike other
accommodation policies, integration villages was housing for EU migrant Roma
exclusively, not French travelers. This made the integration village policy stand out.
Furthermore, in 2007 Romania became a member in the EU, this allowed migrant Roma
from Romania the right to travel, work and stay in France. However, despite Romania’s
EU membership, there were still labor market restrictions which made it difficult for
migrant Roma to find employment (Sardelik 2021). In 2007, due to the influence of EU
funds for integration villages and Romania’s EU membership, Lille finally began to
recognize migrant Roma as a different population from French Roma, with distinguished
rights and needs.
Lille
By 2008, evidence of a civic legacy of networks in Lille emerged through
AFEJI’s support of integration villages. By December 2008, on the initiative of Mayor
Martine Aubry, Lille Metropole had set up 4 integration villages, housing 87 Roma
migrants. In Lille, AFEJI supported the families during their transition to integration
villages, “this work of integration will be based on voluntary work by the families, which
will commit themselves by contract with the AFEJI, since November 1, 2007, the State
has entrusted AFEJI with an inter-associative coordination mission within the framework
of this integration objective.” The President of the Metropolis requested on December 24,
2008 that the towns of the Lille metropolitan area provide land, which would be serviced
53
to accommodate the Roma populations in mobile homes. Particularly, “those families
likely to adhere to a social and professional integration program.” In 2008,
Lille was introducing the support network that Toulouse introduced in the 1990s.
Social support was provided in the integration villages by NGOs that had and still
do provide support to French Roma, such as AREAS and AFEJI,
…access to social cover, schooling, training, employment, housing. This
integration work will be carried out through the on-the-ground expertise acquired
by the NGOs, at the forefront of which will be the AFEJI and the Regional
Association for Study and Social Action (AREAS). This work of integration will
be based on voluntary work by the families, which will commit themselves by
contract with the AFEJI.
Lille introduced integration villages with their own conditions and duties for the families
to respect, and the city provided social support. Lille enthusiastically worked with EU
funds to introduce integration villages while Toulouse did not demonstrate as much
enthusiasm.
Integration villages encouraged assimilation and rewarded families that adhered
best to the conditionalities of the integration villages and were thus deserving of a place
in the housing system. “This proactive accommodation action will be aimed primarily at
those households likely to adhere to a social and professional integration program.” There
was a strong conditionality to living in integration villages, families had to either have an
income or be actively working toward earning an income. Families’ social lives were
surveilled, classified, and controlled, usually by NGO members. On the other hand,
stakeholders “specified that ‘the criteria are neither exhaustive nor scientifically
evaluated.’ It will above all be a question for the families ‘of expressing a real desire to
54
settle permanently in France, to learn French and to look for a job.’" So, while there were
conditionalities to receiving and maintaining one’s place in an integration village, the
families’ participation was technically voluntary, the conditions could be interpreted
as the “bare-minimum” to integration, and the process involved significant support from
social workers,
This integration work will be carried out through the on-the-ground expertise
acquired by the NGOs, at the forefront of which will be the AFEJI and the
Regional Association for Study and Social Action (AREAS). This work of
integration will be based on voluntary work by the families, which will commit
themselves by contract with the AFEJI. Since November 1, 2007, the State has
entrusted AFEJI with an inter-associative coordination mission within the
framework of this integration objective.
While NGOs provided experiential knowledge, acted as a mediator between the
families and the city and were a huge source of trust for the families, they were also in
charge of “observing” families,
The main means of intervention remains the regular presence at the various
parking places. Going to meet families where they live makes it possible to
identify and get to know the people, their needs and to list their requests. It is also
a main tool for observing the connections between family groups and their
evolution. It is also the only way for families to identify us and identify our
mission.
While migrant Roma were considered a “marginal group” – by 2009, only 600-
1000 Roma individuals lived in Lille Metropolis—their presence was discussed a
disproportionate amount.
…for the Roma who wish to settle permanently in our territory - and these are
marginal - integration villages have been created. In the form of an airlock, they
make it possible to verify the continuation of this integration in the best
conditions. The results today are both very positive for some families and
negative for others.”
Toulouse
In Toulouse, the Socialist Party (PS) was voted into office in 2008, led by Mayor
Pierre Cohen. France had just elected President Nicolas Sarkozy on the center-right, and
55
in 2008 Toulouse elected its first center-left Mayor since 1971. Claude Touchefeu was
deputy mayor in charge of urban politics, which included Roma integration. However,
this new political alignment did not seem to influence policy for EU migrant Roma. In
the city council deliberations there is no mention of Roma migrants as separate from
French Roma, or related to the topic of housing until 2010. Then, in 2010, the federal
government did something that ratcheted up anti-Roma rhetoric in all of France and so
“Roma” were finally mentioned in city council meetings.
2010-2012: National and International Influence
Between 2010 and 2012, national and supranational level politics heavily
influenced integration policy for migrant Roma families in France. As a result, in 2010
both Toulouse and Lille city councils distinguished French travelers from migrant Roma,
and Lille introduced six more integration villages. The impetus was the 2010 evictions of
slums and camps in the northern city of Grenoble, France. On July 30, 2010, in the wake
of violence in Grenoble, President Sarkozy made a speech that “lumped together ‘gens de
voyage’ (travelers), Roma immigrants and French citizens ‘of foreign origin’ and linked,
at the same time, crime and migration” (McGarry 2017, 238). Then, in September 2010, a
secret memo from the French Ministry of Interior signed by Interior Minister Brice
Hortefeux, dated August 2010, was released to the public. The memo asked local police
authorities to evict 300 Roma encampments within the following three months. In
response, an estimated 50,000 protestors took to the streets in early September 2010 in
Paris, Bordeaux, Marseilles and Toulouse, as well as French embassies in London,
Brussels, Madrid, Barcelona, Serbia and Hungary. Protestors held placards that read, “Je
suis Européen,” (I am European) and “Ne nous expulsez pas,” (Do not deport us)
(McGarry 2017, 239).
56
The deportations themselves were legal (most Roma that were subsequently
deported from France had been residents for at least three months and had not secured
employment), but the targeting of a specific group for evictions was against the EU’s
Racial Equality Directive (2000). Viviane Reding, the VP of the EU Commission at the
time, publicly denounced the French government and threatened legal proceedings as
“discrimination on the basis of ethnic origin or race has no place in Europe” (Reding
2010). EU justice chief Viviane Reding dropped the threat of legal action against
France and “expressed regret” for the “wartime comparison,” by October 2010, however
“that summer period was synonymous with exclusion, precariousness and moral and
physical violence.”
Toulouse and Lille both denounced these evictions along with the rest of the
international community.67 Lille responded to President Sarkozy’s Grenoble speech by
introducing a moratorium on evictions, building six more integration villages and
conducting social surveys of the families living in the camps to get a better sense of what
families in the informal encampments needed, “In this sense, our establishment voted on
November 5, 2010 an action plan relating to ‘provisional and emergency reception and
accommodation, in particular calling for the creation of additional integration villages for
a global reception of 200 people, in the long term.”68 The metropolitan government had a
goal of 10 integration villages by the end of 2011.
Toulouse
In Toulouse, 2010 is the first year the subject of “Roma” appeared in the Toulouse
City Council deliberations by name. The Toulouse City Council condemned the leaked
memo and also called for an eviction moratorium “on the free movement and settlement
57
of European citizens who do not represent an ‘unreasonable burden’ for society.” Lastly,
Toulouse asked that the state respected the “one month period which must be granted to
any person subject to an eviction order on the principle of ‘proportionality.’” However,
the center-right accused the incumbent government of leveraging the “discriminatory
policy of the Sarkozy government” to avoid “clearing ‘dangerous’ spaces.” In other
words, the center right accused the center left of using national-level events as cover to
avoid accountability at the local level, this is a similar accusation that the center-right
wields against the center-left in Lille at the time.73
68Lille Métropole, FEDER, « Création de Villages d'lnsertion pour populations marginalisées sur le
territoire de Lille Metropole (phase 2), » pg 2, 1345W7, 2013, Lille Metropole Archives, 2 Rue de l'Europe,
59320 Sequedin, France.
Lille Métropole, “Accueil et hébergement provisoire et d’urgence. Propositions de Lille Métropole pour un
plan d’actions, » pg 6, November 12, 2010, 847W47, Lille Metropole Archives, 2 Rue de l'Europe, 59320
Sequedin, France.
At the same time, in 2010, EU cities were incentivized by the European Regional
Development Fund (ERDF) to provide more integrated accommodation generally,
“Opportunities for Lille Metropolis: rehabilitation of existing integrated housing,
demolition of precarious housing, construction of new housing or shared housing,
integration villages, short-term housing,” Lille described integration villages as a success,
while some scholars and policymakers condemned them.
Then, 2 years after the Sarkozy scandal, in August 26, 2012, Interior Minister
Manuel Valls of the new center-left national government issued a memo on, “anticipating
and supporting eviction operations of unauthorized settlements,” signed by seven
ministries. This ended the moratorium on evictions and between June 2013 and
58
November 2014, nine camps and around 2,000 people were evicted from their homes
(including a camp of 550 people in Lille).
As such, 2013 was a period of evictions for Romani migrants throughout France,
“French authorities forcibly evicted more than 21,537 Romani migrants in 2013, more
than double the total for 2012. Law enforcement officers carried out 165 evictions
affecting almost 19,380 people. There were 22 evictions due to fire, affecting 2,157
Roma.” Toulouse reported 139 people evicted and Lille reported 1878 people evicted.
In light of these evictions, by 2013, the center-left government in Toulouse
explained how difficult it was for migrant Roma families from Romania and Bulgaria to
secure housing, and identified integration villages as an appropriate policy response,
Indeed, since the entry of Romania and Bulgaria into the European Union in
2007, nationals are subject to a restrictive transitional legal regime. This status
and the material precariousness they experience both in their country of origin
and in France creates a particular context which represents a challenge both for
social work and for the logic of public order.
In 2013, the center-left government began a last-ditch effort to provide housing
for migrant Roma. The center-left identified integration villages as an appropriate
housing option due to France’s restrictive transitional legal regime regarding migrant
Roma from Romania and Bulgaria. “Roma families who are in a regular situation on the
territory…nevertheless cannot claim social housing today, in particular because the
conditions of access to work are quite restrictive.”81 At this point, French cities were
anticipating Romania’s full accession to the EU in 2014, and the requisite lifting of
transitional measures such as work and residence permits. So, by the end of 2013, the
center-left in Toulouse acknowledged that, “Despite this initial work undertaken by the
59
City, illegal occupations and the phenomenon of over-occupation in certain degraded
condominiums continue to exist,” and proceeded to ask the federal government for
100,000 euros to support an Urban and Social Project Management program (MOUS)
that is focused on providing, “stable” autonomous housing, as opposed to integration
villages. This was a last-ditch effort of the center-left leadership to successfully introduce
housing options for migrant Roma; nevertheless, a new city council was elected in 2014.
Toulouse
In 2014 the Toulouse City Council changed hands from the PS (center-left) to the
UMP (center-right) and Mr. Moundenc was elected Mayor (for a second time) under the
premise that he would decrease the number of people living in informal camps and
effectively offer housing options. In an interview with a pivotal NGO member in
Toulouse, they reported that they predicted that the center-left would lose power because
they were no longer responsive to what the NGO network wanted in regards to housing
policy for migrant Roma, while the center-right party was responsive.iii
In 2014, Mr. Daniel Rougé was elected as the new deputy mayor in charge of
social housing, solidarity policies and social affairs. He is credited by many respondents
as being one of the reasons that Toulouse has had such success with housing efforts. In
Toulouse in 2015, once the center-right was re-elected, it was revealed in the city council
meeting minutes why the center-right leadership thought the center-left was not
successful in housing policy for migrant Roma,
… the debates that are agitating French society and the French Left at the moment
clearly show the immense problem of the Socialist Party and the gap that exists
between what it says and what it does. What it says when it is in opposition, like
here, and what it does when it is in charge.
60
Mr. Moundenc called the center-left cynical, performative, and callously strategic in the
face of complex social questions. A new administration with a new agenda for integration
was part of the pivot away from integration villages as a primary policy option for
housing integration.
Lille
In Lille on the other hand, in 2013 Lille implemented EU-funded policy (EDRF
and ESF funds) that focused on integrated housing options for Roma,
This operation allows Lille Metropolitan area to continue its work with
marginalized communities, including the Roma community, from a new
perspective through integrated housing and social support operations. Indeed, for
the first time, to support the development of the experimentation of integration
villages carried out by Lille Metropolitan area, the action of the ERDF funds will
be accompanied by the action of the European social fund.
The integration villages were a cooperation between local, metropolitan, state and
international bodies, the highest level of cooperation for integration policy thus far,
Following the modification in May 2010 of the European regulation relating to
the support by the ERDF of integrated housing operations in favor of
marginalized populations and the creation of the Roma task force by the European
Commission in September 2010…In July 2011, the Nord-Pas de Calais regional
program monitoring committee approved the creation of a priority…support for
access to housing for marginalized communities.
Throughout 2014, NGOs continued to support families in integration villages and
other housing. NGOs reported a lot of success and increased their focus on trust-building
and mediating between the families, their neighbors and the city, “in fact, taking problems
into account soothes households, which feel less alone and more supported. Of course,
building trust is a daily job. But despite occasional tensions, the accompaniments are
carried out serenely.” Between 2012 and 2014, in an effort to provide integration villages,
there was an enhanced level of stakeholder cooperation between NGOs, city council and
61
the state in regards to integration villages and EU migrant Roma integration more
generally.
Third Housing Policy Era: Policy Divergence between Toulouse and Lille
2015: Toulouse Pivots from Integration Village Policy and Leads Integration Policy
Toulouse
In 2015, instead of introducing more integration villages, on the advice of the
“Collectif d’Entraide et d’Innovation Sociale” (CEDIS), a mutual aid NGO with ties to
CCPS, the city pivoted from a logic of short-term accommodation to a logic of long-term
housing.iv The city coordinated with the prefect and 13 social housing landlords and
found 150 accommodations that comprised low-cost social housing.v Eurocities, a
transnational NGO, reported that in 2015, Toulouse shifted from a logic of providing
“short-term housing,” to a focus on long-term housing— “autonomous housing units with
social support in view of assuming the rights and duties of being a tenant able to cover
the cost of rent in the long term.” Other policies included subsidizing rent costs for
families interested in long-term housing, rather than re-housing these families in
integration villages or short-term accommodation. This shift in policy in Toulouse was at
least one year after Lille declared their integration village policy a success and decided to
increase the number of integration villages. It appeared that this policy change was the
outcome of the change in administration which was advised by a network of NGOs and
other civil society organizations to move in the direction of long-term autonomous
housing for migrant Roma. This network included the vanguard group with prior
experience integrating migrant Roma in the 1990s.
62
The network that had established itself back in the 1990s, led by CCPS, was still
involved in housing for migrant Roma in 2015 when the government switched from
center-left to center-right. The NGO networks established in Toulouse in the1990s that
had continuously developed through to 2015 helped empower the center-right to
introduce new policy for migrant Roma that focused on long-term housing. Organizations
such as CCPS and CEDIS31 created the networks and solidified connections, trust,
mediation and expertise between civil society organizations and migrant Roma families
through housing provisioning support efforts. The development of these networks
culminated in a new government in 2014 that was more responsive to the will of civic
society and subsequent
policy change in favor of long-term autonomous housing in 2015. As such, the evidence
suggests that Toulouse developed networks during the integration of migrant Roma in the
1990s which set a precedent for its city council to shape new housing policy in 2015 and
empowered it to emerge as a leader in housing policy for migrant Roma.
2015: Lille continues reliance on integration village policy
In Lille instead of policy influenced by a vanguard group within a network that
was interested in the rights of migrant Roma, Lille introduced policies that had the most
funding available. In Lille there is no evidence of a local civic legacy of an NGO network
with a vanguard group advocating for housing for migrant Roma similar to the one
Toulouse has which empowers its government to be responsive to the rights of migrant
Roma.
63
Instead in Lille, policy change for long-term housing occurred later in 2017 after a
national-level policy was introduced. As stated above, this does not mean that Lille did
not have a civic society interested in housing policy change for migrant Roma in 2015,
nor does it mean that Lille did not have some evidence of a civic legacy. In fact, the NGO
network in Lille advocated for policies including integration villages and international
cooperation programs with Romania.vi However, Lille’s civic legacy of networks was not
developed as early as the one in Toulouse and did not have the same level of institutional
continuity (the organization ATS lasted nearly 30 years in Toulouse) over time. Evidence
suggests that Lille was more influenced by international and EU funding than local
advocacy interests which were developed through civic legacy networks.
Fourth Housing Policy Era (2017-2020): Policy Continuity, National Housing Policy
Introduced
Lille
In September 2017, the National government launched a ‘Housing First’ Plan on
“the observation that direct access to housing, or staying in housing, with support, are
generally better suited to people and less costly for the community, than emergency
accommodation.” In 2020, Housing First was considered a success in Lille, and an
important process in moving people out of slums and into housing.
Housing First Update: ‘The territory of the MEL (Metropolitan Region) is
confronted with the phenomena of poverty and exclusion which particularly
affects people confronted with poor housing and homelessness. The MEL is
resolutely committed alongside the State as part of the "housing first" plan, of
which it became in 2019 one of the 23 territories for accelerated implementation.
The first results of this partnership have been positive in several areas, including
the number of people living in emergency accommodation accommodated in
social housing or the strengthening of systems aimed at comprehensive support
(housing, health, integration, etc.) key to success of the “Housing First” policy.
Similarly, the mobilization of the MEL alongside the State has made it possible to
evacuate several slums by relocating their inhabitants and providing them with
appropriate social care.
64
The support from the EU for integration villages came to an end in 2019 because
the EU had only agreed to fund the integration villages between 2014-2019,
The commissioning of the integration village in Toucoing having been financed as
an investment by European funds for a period of 5 years: from the final payment of the
operation made on 02/12/2014, it is advisable to extend the availability of the building at
the MEL until 03/12/2019 in order to be in line with the financing methods, in particular
European, that can be mobilized for the operation. By 2020, the city mentioned
integration villages alongside other “mechanisms for access to housing,”
…mobilizing its sovereign functions alongside social policies of integration and
access to housing. It will involve both the activation of mechanisms for access to
housing (e.g. housing first) but also the construction of temporary reception
structures such as SAS or integration villages.
The Metropolis recognized that the Covid-19 crisis increased the urgency for
adequate housing and offered housing first and integration villages as policy options,
The contract should also make it possible to intensify efforts to clear slums. The
covid crisis and the period of confinement that followed have only increased the
urgent need to remedy an undignified situation for the occupants of the camps, …
It will involve both the activation of mechanisms for access to housing (e.g.
housing first) but also the construction of temporary reception structures such as
SAS or integration villages.
In 2021, the center-right coalition outlined the Metropolis’ current strategy. In the
Metropolitan online archives, VP and Mayor of Loos, Madame Anne Voituriez described
instances of inter-municipal cooperation, the use of different types of integration villages,
work with Housing first, and NGOs such as Sauvegarde de Nord, CANAL, SOLIHA,
AFEJI, and with the Prefect of the Nord Department. While the Lille Metropolitan area
committed to providing autonomous housing options, still in 2021 integration villages
were mentioned as a valid option in city council meetings. Lille’s previous reliance on
integration villages made it difficult to entirely transition from the integration village
policy toward autonomous long-term housing, but the fact that the city council and
65
metropolitan representatives weren’t more critical of the integration village policy is
evidence that the civic legacy of the NGO network wasn’t as strong in Lille.
Toulouse
In contrast, in the Toulouse City Council meeting minutes there is evidence that
representatives did critique previous housing policy. During a city council meeting in
2017, Deputy Mayor Daniel Rougé reflected on the policy change in 2015. Rougé
pointed out the continuities and discontinuities with the previous administration. Deputy
Mayor Daniel Rougé acknowledges that integration villages can be helpful but that it was
necessary to move on from this policy if the city was going to provide the level of
support that the families needed. Rougé argued that if the center-left government had
remained in power, the city would have continued with the integration village policy.
Rougé acknowledged that the city will continue with the Urban and Social Project
Management (MOUS) contract with the State decided on January 1, 2014 under the
previous administration and then he acknowledged the great strides taken with the
integration village Grynfogel but then announces that due to the number of families that
needed support have decided not to continue with the integration village method.99 And
then Rougé talks about the discontinuities,
There is, in my opinion, great progress and I can provide you with the facts: since
we took on this issue, we have evicted 105 squats in the city of Toulouse. That
doesn't mean there aren't any left, of course. On the other question of the method,
where it is different, it is that, concerning precarious European nationals, we have
used methods of temporary emergency accommodation, with social support to
bring about the integration of people who go to social housing. But this
accommodation being integrated, we did not use camps or an integration village,
like that of Grynfogel, even if the results of Grynfogel are very interesting. The
great methodology is the mobilization of the assets of the City and of EPFL, but
also, the efforts of the City also for this purpose, and also the great mobilization
of social landlords in this question. Very great mobilization of social landlords on
66
this issue, who have, by the method, been secured in the use of their property, for
which neighborhood issues do not cause major problems.
He mentions that the new administration has evicted 105 squats, this is their main talking
point. However, he goes a step further to mention the social support and integrated
housing that the city is advocating for and supplying, instead of the segregated Grynfogel
integration village. Here, Rougé establishes that the pivot from the 2015 policy of
integration villages to more integrated housing with increased social support was in fact
the right move, even a few years in retrospect.
By 2018, Toulouse still maintained one integration village, but when asked about
the topic of integration and housing, policymakers spoke of their efforts for long-term
housing accommodation and no longer mentioned integration villages as a policy option.
Toulouse focused more on the “Five-year plan for Housing First (2018-2022) and the
fight against homelessness.” This plan was the result of cooperation between Toulouse
Metropolis and the Haute-Garonne Prefect. In 2018 during a city council deliberation,
Mayor Moudenc announced that in 2015, there were 1500 people in informal camps and
that by 2018 they had around 750 people in informal camps. Additionally, they had
offered short-term (hébergement) accommodation to 1090 people in 2018. The Mayor
also reported that the reduction of slums in 2017 was praised by
CEDIS13, a mutual-aid NGO. A policymaker in Toulouse municipality urged that
67
while Toulouse is understood as a model of housing through integration villages, in
practice, they have surpassed this policy and believed that it is long-term housing that
leads to integration.
In October 2018, the Mayor of Toulouse announced that the work on Housing
First is a direct result of Mr. Daniel Rougé’s work with “precarious populations,”
particularly “migrants,”
It is the recognition of the work that has been done under the impetus of Daniel
ROUGÉ with the intervention of Franck BIASOTTO and other elected officials,
in connection with the associative world concerned, in connection with the
prefecture. So all the work we are doing on this difficult issue of housing the most
precarious and so I salute this good decision.
The Mayor attributed Toulouse’s “success” to the work done by Mr. Rougé, other
elected officials, in connection with the “associative world,” as well as the relationship
with the prefect. The “associative world” is the network of NGOs and other civil society
organizations that work with the city council governments to advocate for long-term
housing.
By 2018 Toulouse had a specific interest in integrated housing, “what is
interesting in these questions is that for social inclusion, it is still much better to achieve
social diversity in integrated housing than to do so with a much more closed (or
segregated) design of the system…” The closed design system is a reference to
integration villages which were enclosed, gated spaces with curfews for its inhabitants.
This move away from integration villages and toward integrated housing is reflected in
the city budget under the fight against exclusion policy, “1 integration village, 2 reception
sites, 144 housing units spread over the whole of Toulouse to meet the accommodation
needs of precarious migrant populations.» By 2018, Toulouse is explicitly referring to
68
integration villages as a policy of the past, and that integrated housing options are
preferable and a result of the work of Mr. Daniel Rougé, with his associative network that
allows for this model to be understood as a success. The municipality has great hopes that
this model of integration could be applied to the integration process for other groups as
well.
By 2018, Toulouse welcomed more non-EU migrant populations, and
distinguished this state-led effort to the city council’s “proactive approach” of housing
EU migrants, “This action falls within a field of competence quite distinct from that of
the Town Hall, whose proactive approach consists of offering precarious European people
living in slums accommodation for the purpose of integration.” By 2018, the city council
articulated the integration of EU migrants as a template for the integration of refugees.
They have gone so far as to develop the “Toulouse method,” which is “a model for the
reduction of camps, with social inclusion through integrated accommodation.”
Whilst articulating their successful policy, they explained how they “pushed” the state to
work in concert with them, acknowledged the “fluidity” of the policies, the importance of
remaining humble and also critiqued the previous center-left leadership for its “militant
cynicism” toward integration issues.
In Toulouse City Council meeting minutes from 2017-2020 there is a lot more
discussion about the success of Toulouse City Council housing policy. The policies are
referred to as a model for other groups’ integration and other cities in France looking for
successful integration policy. There is a lot more discussion about housing options for
69
migrant Roma as a whole, compared to the city council meeting minutes in Lille. In Lille
between 2017-2020 there is mainly discussion about the patchwork of policy options
available to migrant Roma, the national level policy focused on autonomous long-term
housing and then the emergency efforts related to COVID-19 housing policies.
While the change in government from the center-left to the center-right in 2014 is
an important change that resulted in a change of housing policy, the role of the NGO
network with a civic legacy in Toulouse is also important. The network solidifies the
connections amongst civil society and government institutions and actors that allow the
experience, information and trust to influence policy decisions. This network was
developed during prior attempts at integration in the 1990s and early 2000’s, by decades
of social support already provided to migrants generally and migrant Roma specifically,
which developed the trust, expertise and institutional channels necessary for good
governance and leadership in policymaking.
Conclusion
The evidence suggests that Toulouse developed networks during the integration of
migrant Roma in the 1990s which set a precedent for its city council to shape new
housing policy in 2015 and empowered it to emerge as a leader in housing policy for
migrant Roma. Alternatively, Lille lacked the development of civic legacy networks in
1990 to set precedent for 2015 and thus did not change their policy until there was a
national level change in policy and emerged as a follower.
The civic legacy of networks in Toulouse had established itself back in the 1990s,
led by CCPS, continually developed and was still involved in housing for migrant Roma
in 2015 when the government switched from center-left to center-right. Organizations
70
such as CCPS and CEDIS31 created the networks and solidified connections, trust,
mediation and expertise between civil society organizations and migrant Roma families
through housing provisioning support efforts. The development of this civic legacy
culminated in 2015 when this network helped empower the newly elected center-right—
which appeared more responsive to the will of civic society—to introduce new policy for
migrant Roma that focused on long-term housing. As such, the evidence suggests that
Toulouse developed networks during the integration of migrant Roma in the 1990s which
set a precedent for its city council to shape new housing policy in 2015 and empowered it
to emerge as a leader in housing policy for migrant Roma.
This chapter set up the argument that the vanguard group in Toulouse, ATS, had
the institutional power to advocate for migrant Roma in 2015 based on its historical
experience with integration in 1990. This role of this network was to generate the
information and trust between the target group, social workers and city council that is
necessary for housing policy. The civic legacy of this network that was established by
ATS in 1990 during re-housing efforts was activated again when new leadership took
control in 2015. This network was able to hold local political actors accountable, give
voice to migrant concerns and advocate for certain policies over others. Toulouse was a
leader in making this policy switch and continues with this policy through 2020.
Lille on the other hand, did not introduce long-term housing until 2017 with a
change in national policy. Lille lacked a strong vanguard group like CCPS in Toulouse
with an institutional legacy developed over decades that could solidify the necessary
connections, trust, mediation and expertise between civil society organizations, migrant
Roma families and city council. A network was developed later in 2007 when the civil
71
society network supported migrant Roma families in their move from encampments to
integration villages. However, this network did not have the historical legacy like the one
developed in Toulouse. Without a robust civic legacy, Lille did not change its policy from
integration villages until there was a national level change in policy in 2017.
Instead, policy change in Lille was lead by national policy and EU funding opportunities.
In this way, the evidence suggests that the development of a civic legacy of a city
can influence contemporary integration policy. This chapter described the relevant
networks that influence integration policy, however this is not the full picture of the role
of a civic legacy. Civic legacies include networks, ideas and actors. The next chapter will
focus on the role of an actor, the policy entrepreneur, and how they engage with policy
frames to influence policy. Specifically, chapter 5 will focus on the role of an actor
introduced in this chapter, the role of Daniel Rougé and how he used policy frames to
lead Toulouse in housing policy. Alternatively, Chapter 5 will also describe how Lille’s
lack of a successful policy entrepreneur impeded its ability to lead policy change for EU
migrant Roma.
Chapter 5. Establishing a Civic Legacy: Policy Frames and Entrepreneurs (1990-
2007)
Chapter 4 established how the civic legacy in Toulouse shaped integration policy
for migrant Roma in 2015 when the city switched to long-term autonomous housing and
why Lille did not develop a civic legacy and did not switch its housing policy until 2017
when a national policy was introduced. The integration efforts in the 1990s developed a
network of civil society actors who established the trust, information-sharing between
72
civil society and government to influence housing policy later in 2015. This network was
able to hold local political actors accountable, give a voice to migrant concerns and
advocate for certain policies over others. Since Lille did not have that historical network
development, the information-sharing and trust between civil society actors and
government was lacking and the vanguard group was not able to encourage policy change
like Toulouse and instead it followed national level policy.
But the network is not the only aspect of a civic legacy potentially at work
influencing integration policy. Why else did Toulouse pivot from short-term housing to
long-term housing in 2015 and emerge a leader in housing policy for migrant Roma while
Lille was a follower? Using a content analysis, this chapter presents the evidence from
the content analysis and identifies which resonant policy frames comprise the potential
civic legacy of each city. As a reminder, the literature stipulates that when a policy
entrepreneur introduces a frame, she is introducing or constraining possibilities to change
a policy. Policy frames shape how civil-society actors negotiate housing integration
policy. Policy frames structure and contextualize debates over integration policy and
provide tools for migrant advocates. Policy frames used in previous struggles for
integration have a civic legacy. The civic legacy of policy frames provides a repertoire of
policy frames that have specific resonance in the community. These policy frames have
historical significance and a different semiotic weight because of that significance. As
such, if actors invoke a specific frame with a civic legacy, this frame may have more
resonance in the community, which might garner the necessary political will to introduce
one policy over another. In this way, the policy frame is part of the city’s civic legacy.
Sometimes these frames are strategically introduced and sometimes not.
73
This chapter presents the evidence from the content analysis to compare the role of
policy frames in the civic legacies of Toulouse and Lille (see Chart (4)). Between 1990
through to 2020, the “humanitarianism” and “hospitality” frames distinguish Toulouse
from Lille, as do the “balancing act” and “policy critique” frames. However, the
humanitarianism policy frame was used both historically and in 2015 in Toulouse and not
in Lille to the same extent. This specific “humanitarianism” frame had a resonance in
Toulouse that appeared to reflect a civic legacy and helped through the network and
wielded by the policy entrepreneur to develop the political will for the policy switch from
short term to long-term housing for migrant Roma.
In Lille, there is not one policy frame that has resonance and appears both in the
1990s and later on during the period of policy change in 2015. Lille’s frames focus a lot
on “history” of “French travelers,” not migrant Roma. Like in Toulouse there is a focus
on “stakeholder coordination,” “social support” “dialogue and mediation” and “dignity.”
All policy frames which develop the political will for housing policy for migrant Roma.
However, these are also found in the top 20 frames in Toulouse and do not distinguish
Lille from Toulouse. Lille has a focus on “pedagogy” that Toulouse does not. The
development of this policy frame appears prominently after 2012, three years before the
policy switch in 2015, but it is not reflected earlier in Lille’s history. So there does not
appear to be a civic legacy of the frame “pedagogy” to the same extent that there is to the
humanitarianism frame. The meaning of these frames will be explored more throughout
the chapter.
Chart (4): Top 20 Policy Frames in Toulouse and Lille (1990-2020)
Toulouse Lille
Stakeholder
Coordination History
74
Humanitariani
sm French Travelers
Residency Status Stakeholder
Coordination
Social
Support Social
Support
Hospitali
ty Neighbo
r
s
Precarity Dialogue/
Mediation
Dignit
y Hygien
e
Securi
t
y Precarity
Social Relations Pedago
gy
Way of Life/Against the
Grain Securit
y
Dialogue/
Mediation Proactive
Policy
Human
Rights Eviction
s
Identity Dignit
y
Balancing
Act Residency Status
Crime/
Delinquency Coordination with Other Countries
Policy
Critique Surveillan
ce
Housing Needs State
Responsibility
Healt
h Emergency
Conditions
History Shared public intervention
Diversit
y Re-housing
policy frames develop political will for housing.
policy frames block political will for housing.
Crossed out policy frames indicate that the frame was present in both Toulouse and Lille. The
remaining policy frames that are not crossed out were identified in one city but not the other, emphasizing
the variation of frames in each city.
Themes without highlight are ambivalent policy frames
However, from the evidence it appears that in addition to the question of which policy
frames were developed, is also the question of when were policy frames developed? The
evidence suggests that Toulouse developed a civic legacy of policy frames concerning
migrant Roma in the 1990s (during the re-housing efforts explained in the last chapter)
75
Blue
Red
which set a repertoire of frames to influence a policy change in 2015. Alternatively, Lille
lacked the development of a civic legacy of policy frames in 1990. Instead, most of
Lille’s policy frame development occurred later in 2008 in response to the integration
village policy. This development set a different repertoire of policy frames and thus Lille
did not change its policy until there was a national level change in policy and emerged as
a follower. In these ways, the earlier development of policy frames in reaction to
rehousing efforts in Toulouse contributed to the civic legacy which made Toulouse a
leader in housing policy for migrant Roma, while the latent development of policy frames
in reaction to integration villages in Lille contributed to the lack of a civic legacy in Lille
and its follower status.
Figure 1: Civic Legacy Represented by Number of Policy Frame References (1990-2021)
Integration Policy Frames in France
Conventional understanding from the literature is that policy actors invoke,
“collective understandings of nationhood,” or nationalist policy frames to develop the
political will for or against housing policies for migrant Roma (Chung 26, 27).
“Nationalism” is a dominant policy frame in ideological debates concerning immigration
and integration policy. Furthermore, French Civic and Republican ideals are deeply
interwoven with French nationalism and are constitutive of French identity (Brubaker
1992). So, it is perplexing that “Nationalism” is not a prominent policy frame in the
content analysis regarding migrant Roma integration. The next section unpacks different
76
types of conventional “Nationalist” policy frames and concludes by previewing and
comparing which policy frames were identified in Toulouse and Lille.
In his work on Nationalism in France Compared to Germany, Brubaker (1992)
argues that France’s modern-day nationalism is state-centric, assimilationist and values
Republicanism and civic integration. He posits that France’s nationalism evokes the
dominance of political citizenship over ethnocultural conceptions of nationhood. France’s
nationalism was born in the 18th century and reified by the French Revolution and its
aftermath including the “mission civilisatrice” carried out by the Third Republic’s
schoolteachers whose mission was to “institute” the nation (Brubaker 1992, 6). In
addition to the imperialist and colonial mission to “liberalize and civilize,” overseas
populations and create “la plus grande France,” French republicans pursued an
assimilationist, civilizing and nationalizing mission inside France as well. “In the 1880s
this assimilationist internal nationalism, linked to reforms of primary education and
military conscription, formed the backdrop of an expansive, assimilationist reform of
citizenship law whose central provisions have endured to this day” (Brubaker 1992, 11).
Later, the Dreyfus affair (1894-1906) “challenged” the state-centric and assimilationist
French nationalism in favor of an ethnoculturalism, as represented by the party that Jean-
Marie Le Pen’s founded, The National Front. This is the understanding of French
nationalism as articulated by Brubaker (1992). Brubaker’s analysis gives us a good
understanding of a traditional French nationalism, while Beaman (2017) provides a more
contemporary analysis of French nationalism that takes racist discourses into
consideration.
France has a contemporary history of wielding its dominant Republican values
77
(egalité, fraternité, and laicité) like a discursive cudgel, forcing groups on one side or the
other of a discursive debate (e.g. liberal v. illiberal; Christian v. non-christian). This is
partially because France’s nationalist ideology is a Republican and “colorblind ideology.”
Since France is a liberal republican country, “per French Republican ideology, (one) is as
culturally French as any other French person, so long as they adhere to the values of the
Republic” (Beaman 2017, 41). A perverse implication of this is that in national-level
debate, France has a rich history of putting its “Republicanisme,” value which includes,
“Liberté, Egalité, Fraternité,” in contestation with the presumed “values” of groups trying
to integrate into French society.
The modern national debate concerning immigration in France began in the late
1980s with the rise of Jean-Marie Le Pen’s National Front Party and through the 1990s
(Morgan 2017, 892). At the time, the policy focus was on the first generation of North
African/Maghrebin migrants that had migrated first in the post-war years and through the
1970s. “Anxiety over the integration of these immigrants reflected the sense that
nonEuropean populations would have difficulty fitting into French society, and many of
these fears concerned Muslims and the place of Islam in French society” (Morgan 2017,
892). This modern policy frame regarding “Muslim integration” claims to be predicated
upon liberal v. illiberal democratic French values. Within this policy frame, immigrants
are responsible for their ability to integrate or assimilate and if they are unable to then it
is due to a values clash. Given enough time and pressure, this innocuous clash of values
soon morphs into an amalgamation of cultural, religious, post-colonial and finally racial
78
(racist) differences (Beaman 2017, 18, 19). This pervasive racist frame sometimes
motivates and culminates in violence such as the Charlie Hebdo massacre in 2015 and the
Bayonne mosque shooting in 2019. While this liberalism frame is popular at the national
level, it was not a common policy frame in Toulouse and Lille.
Another policy frame that is popular at the national level and has racist undertones
is the “welfare-chauvinist” argument. Welfare-chauvinism is the ideology that immigrants
unjustly benefit from a welfare system meant to support citizens and that this support is a
drag on the country’s economy and citizen’s pocketbooks. Dancygier (2011) finds that
“welfare-chauvinism” was a 90’s a popular policy frame of the far-right as a
means to garner votes in municipal elections,
According to a survey of fifty-eight of the party’s manifestos during the 1995
municipal contests, the FN’s rhetoric focused predominantly on the costs that
immigrants imposed on France’s welfare state and the party’s proposed solution –
Preference Nationale – which envisioned a system whereby French nationals
would receive preferential treatment in the allocation of all state-provided
services. Preference Nationale was mentioned fifty-four times, compared to
“Identité et immigrés” or “Déliquance immigrée,” themes that appeared only
seven and two times, respectively. (Dancygier 2011, 11)
However, welfare-chauvinism was not a common policy frame in the content analysis of
Toulouse and Lille.
Then in 2005, as the FN party platform was gaining influence in local and
national elections, the “merit-based” or “deservingness” policy frames against Muslim
and North African second-generation migrants were also gaining traction. Beaman (2017)
describes the “merit” or “deserving-ness” policy frames introduced in 2005 in reaction to
the riots (in reaction to the deaths of two young people from migrant communities that
lasted for three weeks in the suburbs of Paris, and other cities, including Toulouse and
Lille) which discriminated against immigrants and specifically children of immigrants,
79
It was they (second generation migrants) who were the problem, not France. They
cultivated an oppositional identity and did not see themselves as French.
Marginalization in France was based on class or socioeconomic status, not race
and ethnicity. The issue was that this population (second generation migrants
predominately from North Africa) did not work hard enough to make it in French
society. (Beaman 2017, 46)
This “merit-based” policy frame was not as common in Toulouse and Lille as predicted
by conventional understandings of anti-immigrant policy frames.
In 2010, Marine Le Pen took over her father’s party and attempted to pull the
party further toward the center by invoking “feminist” policy frames to bolster an
antiimmigrant discourse, “she and others in the party have linked immigration to
delinquency and violence against women…the party line on violence against women is
that it is largely a consequence of ‘the massive and continuous arrival of non-European
foreigners who do not share the culture, mores, or customs of most French people’”
(Morgan 2017, 900). In 2012 she placed third in the presidential election with 17.9% of
the vote, in 2017 she placed second with 33% of the vote, and in 2022 she placed second
again with 41.45% of the vote. However, in 2022 turn-out was lowest since 1969,
signaling that these two parties’ political discourses were not the most appealing to
voters. During the 2022 election Le Pen had to contend with competition such as Eric
Zemmour. Zemmour replaced “welfare-chauvinist” arguments with blatant conspiracy
theories underpinned by a threat discourse such as the “great replacement” rhetoric. As Le
Pen moved further towards the center with her discourse, a gap was filled by other
discourse entrepreneurs,
like Zemmour, who were more than willing to provide extremist anti-immigrant
discourse.
80
There is an empirical difference between the national policy frames regarding
integration and local policy frames regarding integration. In order to weaken political
will, I expected some actors to invoke French ideals such as, “republicanism”
“nationalism,” “color-blindness,” “meritocracy,” or to scape-goat others with claims of
“welfare-chauvinism,” “illiberalism” and “anti-feminism.” Sometimes local
representatives and NGOs did invoke the Republican ideal, or the French identity, but
they did not do this as often as expected. While these policy frames have a strong history
in French national politics, these policy frames were not commonly used in Toulouse and
Lille between 1990-2020.
However, three national-level policy frames, “crime and “delinquency,”
“security” and “values-clash” (but not necessarily illiberalism) policy frames did appear
in local level policy debates. In 2010, national level events regarding migrant Roma
explicitly influenced these policy frames. As mentioned in the chapter on networks, in
July 30, 2010, President at the time, Nicolas Sarkozy made a speech in the wake of
violence in Grenoble that associated French travelers and migrant Roma with crime and
security threats. The interior ministry then introduced a policy targeting Roma informal
encampments. The “crime and delinquency” and “security” frame was identified in both
Toulouse and Lille and was one of the most common policy frames. These frames were
articulated in order to weaken political will for integration policies.
In this way, this project explores the independence of local policy frames from
national-level policy frames. While the notion of French identity and its Republican ideal
appears in the cities’ integration policy frames, this project finds that a city can develop
its own policy frames regarding housing policy (frames such as “humanitarianism,”
81
“human dignity,” “mediation,” and “social support”) based on its civic legacy—the
NGOs, policy entrepreneurs and frames—established during previous periods of
integration, such as the 1990 re-housing of migrant Roma in Toulouse.
Toulouse and Lille Top 20 Policy Frames
Table 3 lists the policy frames present in Toulouse and Lille. Table 3 is a
comparison of the top 20 most common policy frames in Toulouse and Lille (1990-2020),
by aggregate counts of policy frame references from the corpus. The policy frames that
are crossed out appear in the top 20 policy frames in both cities. The policy frames that
are left are unique to each city’s top 20 policy frames. The most common policy frame
unique to Toulouse is the “humanitarian” frame, followed by “hospitality.” The most
common policy frame unique to Lille is “history” and “French travelers,” which
emphasizes that most policy in Lille was focused on travelers. “Neighbors,” and
“hygiene” are unique policy frames to Lille that weaken political will. While “pedagogy”
and “proactive policy” are unique to Lille.
Chart (4): Top 20 Policy Frames in Toulouse and Lille (1990-2020)
Toulouse Lille
1. Stakeholder
Coordination History
2. Humanitariani
sm French Travelers
3. Residency Status Stakeholder
Coordination
4. Social
Support Social
Support
5. Hospitali
ty Neighbo
r
s
6. Precarity Dialogue/
Mediation
7. Dignit
y Hygien
e
8. Securi
t
y Precarity
9. Social Relations Pedago
gy
82
10. Way of Life/Against the
Grain Securit
y
11. Dialogue/
Mediation Proactive
Policy
12. Human
Rights Eviction
s
13. Identity Dignit
y
14. Balancing
Act Residency Status
15. Crime/
Delinquency Coordination with Other Countries
16. Policy
Critique Surveillan
ce
17. Housing Needs State
Responsibility
18. Healt
h Emergency
Conditions
19. History Shared public intervention
20. Diversit
y Re-housing
policy frames develop political will for housing
policy frames weaken political will for housing
Crossed out policy frames indicate that the frame was present in both Toulouse and Lille. The
remaining policy frames that are not crossed out were identified in one city but not the other, emphasizing
the variation of frames in each city.
Themes without highlight are ambivalent policy frames
Policy Frames Timeline
Figure 1 is the result of the qualitative content analysis of policy frames. Every
time a policy frame was identified in the corpus of archival material, city council meeting
minutes and interviews, it was noted as a “reference.” The number of “references,” coded
per year were visualized in the Figure 1 timeline. The x-axis is by year and the y-axis is
the number of references (coded in NVIVO). The chart is a visual representation of the
number of policy frame references to migrant Roma housing per year in the corpus.
Higher peaks reflect more references. So, for example, taking a zoomed-out view of the
timeline we can observe more policy frame references in the 1990s, with the highest peak
in 1998, in Toulouse. This indicates a period of civic legacy development. In Lille, the
highest peak in policy frames is observed in 2010. This figure supports the claim that
83
Blue
Red
Toulouse’s civic legacy was developed earlier than Lille’s. Throughout the analysis,
Figure 1 is used to broadly indicate the development of a civic legacy through the number
of policy references over time.
Figure 1: Civic Legacy Represented by Number of Policy Frame References (1990-2021)
First Policy Era (1990s): Policy development, Rehousing efforts in Toulouse but not
in Lille.
Toulouse
In Toulouse, this first policy era was a period of policy development. There was
some focus on migrant Roma due to the city’s re-housing efforts which included surveys
that were conducted by NGOs. These efforts resulted in most of policy frames in
Toulouse between 1990-2007. In 1988, ATS and the city council identified that the
Ginestous camp was a safety risk for its inhabitants. The Mayor proclaimed, “it is my
duty to call and draw attention to the urgency of the work, because the 40 families of
Ginestous representing some 300 people live in a veritable ‘ghetto’ whose law could be
that of the ‘jungle.’” At the same time the AUAT agency conducted another social survey
on “nomades” in Toulouse in 1988. The AUAT presented the study as an openended
question as opposed to a list of solutions or recommendations, which reflected a critical
perspective,
84
The purpose of this present file is not to find an answer or a logic to these
different situations, but rather, based on the observation of the current way of
living of the
Gypsy families in Toulouse and its Metropolitan area, to determine the problems
and the housing needs of these families according to their ethnic groups, their
customs and their professional activities.
There was robust discussion regarding housing for migrant Roma in Toulouse in
the 1990s. Toulouse developed the civic legacy of networks and policy frames during the
administration of diagnostic surveys. The policy frames generated from these surveys
were “dialogue/mediation,” “policy critique,” “balance,” “hospitality” and “diversity.”
By the 1990’s, Toulouse NGOs used a “responsibility” frame for migrant Roma
policy. NGOs and city council portrayed integration as a “balancing act,” as opposed to a
unilateral process with all the onus on the migrant’s ability to integrate. For example, the
housing commission in 1990 admitted that an eviction was a result of a failure of
intervening actors, not the result of intrinsic characteristics of the evicted individuals. At
one of the meetings for the intercommunal syndicate of Toulouse, the group
acknowledged a moral commitment to “nomades” during the development of the
Metropolitan area. The AUAT held a meeting and demonstrated that it was aware of the
importance of educating non-Roma; it emphasized the “sensitization” of “elected
officials, union members, residents, administrators and institutions,” to the “lifestyles, the
different ethnic groups, professional activities, parking problems, experiences of
reception sites, citizenship, the rights of communities etc.” The AUAT and the
intercommunal syndicate were aware of the importance of educating non-Roma about
85
their fellow residents and the integration process more generally. This was a cutting-edge
policy frame regarding integration policy for 1990. In Toulouse, 1990s housing policy
frames were articulated by local NGOs, social workers and city council with references to
“diversity”, “hospitality,” “cultural mediation” and “humanitarianism.”
By the 1990s, due to the work of ATS, Toulouse had conducted enough social
surveys that they framed housing integration for migrant Roma around concepts such as
“cultural sensitization” of election officials, the “responsibility” that Toulouse had to
provide housing options, and integration as a “balancing act.” The city, with the help of
NGOs, distributed short, accessible public service announcements and cartoon strips that
addressed common challenges for migrant Roma families trying to integrate.
By 1998, Toulouse understood the importance of representation and of civic
engagement in the integration process, “the improvement of the participation of the
inhabitants could also be a support to revitalize and regenerate the associative life on the
districts. Like previous social surveys, the one on contemporary populations of
“Bosniacs and Romanians,” was detailed to the point that ATS and the city knew how
many families lived in which areas, their ethnicities and their immigration statuses.
By 1999, ATS had a description of the history of “Tsiganes” in Toulouse from 1613
through the 1990s. In this way, the local NGOs were fully engaged in policy framing
regarding migrant Roma by 1999.
In 2000 the “issues” with Ginestous camp worsened and the city had to figure out
how to resolve them. ATS reminded the city that the problem of “delinquency,” was not
86
developed inside the camp and that the city was responsible for the conditions that made
the camp possible,
…it should above all be remembered that, according to the NGOs themselves, the
residents of Ginestous want to live in Ginestous and that the current problems do
not come from the occupation but from the fact that delinquency has been able to
develop there with complete impunity and that its origin is partly outside the
camp itself.
This is a more holistic view of responsibility and what it means to host migrants.
This interpretation of violence in the camp re-framed the issue. It emphasized Ginestous
as a place that residents value and where they feel like they belong. ATS suggested that in
this case it might not be in the city’s interest to take that security away from the residents.
In 2000, NGOs did a lot of work to mediate and develop the will to act on behalf of the
families did a lot of work in making sure families felt acknowledged, “this allowed us to
intervene in several municipalities this year. Dialogue and goodwill made it possible to
resolve the problems thanks to good mediation by each party. We intervene at the request
of families, town halls and the state.” The policy frame remained localized,
the city did not reference international or national rights, norms or laws to rationalize its
integration efforts.
It was at this time that the Toulouse metropolitan government acknowledged the
influence of policy framing on the integration process. In 2001 the intermunicipal
communication committee was focused on four objectives, “establish(ing) ‘a discourse’
of intermunicipal cooperation, promot(ing) the experience of S.I.E.A.N.A.T., shar(ing)
know-how; learn(ing) from the experiences of other French cities or departments.” The
articulation of these values is very nuanced for the time; particularly its outright
87
acknowledgment of the importance of discourse. Specifically, the committee was focused
on establishing a cohesive discourse that centered on the importance of intermunicipal
cooperation, promoted the “experience” of the intermunicipal institution, as well as the
expertise and knowledge generated by the city and other organizations and lastly, to
generate knowledge from the experience of other cities and departments. At this point,
Toulouse city government had already developed a self-awareness of the policy frames it
used and the influence of its networks. In 2001, the UMP was elected and the city—with
the help of NGOs A.T.S. and Goutte d’Eau—undertook rehousing of the residents of
Ginestous. While they conducted a needs assessment, they simultaneously collected the
demographic information of the residents of Ginestous, as well as their living conditions
and their wishes for their re-housing (which they categorized as “difficult,”
“conceivable,” and “desirable” options).
By 2006, CCPS (previously ATS) was also rehousing families at a stopping site in
La Mounède and reported that they were still “searching for housing solutions,” speaking
with families and that “demands are emerging in terms of rehousing (and social support)
but remain to be expounded upon.” Simultaneously, Toulouse articulated integration as
cultural and social integration, “Globally, our working method consists of intercultural
mediation. This mediation is always done with respect for the cultures of individuals but
by placing them in our social context and our environment. We also take into account the
person received as a whole.” Each aspect of integration—employment, housing,
education, etc.— had a cultural component to it, but CCPS also framed the broader
process of integration through a sociological lens that focused on “culture,” “hospitality,”
88
“societal context” and notions of “belonging,” instead of a process that was only
measured through technocratic metrics such as job attainment, educational achievement,
etc. CCPS also emphasized the importance of mediation between several actors across
several sectors; they enumerated the importance of mediation with public representatives,
mediation between French travelers and their neighbors, mediation with common law
rights access, such as housing and domiciliation processes. All of these re-frame the
responsibility of integration away from the migrant, to the responsibility of the city as a
whole.
At the same time, this focus on the holistic process of integration made social
surveys susceptible to paternalistic discourses such as the following, “Gypsies have
certainly ‘advanced’ in forty years, but we must affirm that they are not yet completely
settled... It is on a daily basis in the actions with them that our NGO continues this
work...There is a lot to be done to bridge the chasms of exclusion from which the Gypsies
still suffer.” While Toulouse treated the integration process with relative nuance it is not
devoid of biased (and outright discriminatory) policy frames.
LILLE
Lille’s period of civic legacy development occurred in 2007, in response to the
integration villages policy (reflected in Figure 1). Between 1990-2008 in Lille there was a
strong focus on parking sites for French travelers, but no surveys conducted for migrant
Roma and no re-housing efforts. The few policy frames identified in this early period
concerned “flexibility,” “social support,” “right to housing” and “humanitarianism,” and
were used in talking about parking sites for French travelers. By 2007, Lille emerged
from this first policy era prepared to provide integration villages for migrant Roma and
89
little else.
In 1990, Lille published best practices for welcoming French travelers. These best
practices referenced “flexibility” and “dialogue and mediation” themes,
… Furthermore, planners must always bear in mind the heterogeneity of
populations. Varied and diversified parking projects are essential to meet the
needs of travelers and the requirements of municipalities. However, there is no
‘miracle and unique’ solution, so actions must be considered and considered on a
case-bycase basis, in conjunction with the populations concerned in order to work
towards an effective and coordinated reception.
This quote encapsulates an early understanding that the policy for French travelers had to
be flexible to respect the diversity of the people the department was serving.
By 1992, the Urban Community of Lille began to acknowledge that some French
travelers wanted to settle in Lille, a process that was referred to as “sedentarization;” this
provided the impetus for the city to consider an adaptation of the services that they
offered, “…the growing phenomenon of sedentarization has been accompanied by
various secondary effects: heterogeneity and plurality of needs, evolution in demand.”
This also presented an opportunity for the city to get to know these populations better. By
the 1990s Lille attempted to employ social surveys to get a better understanding of who
was arriving. This accounts for the steady uptick in references between 1992 and 1998,
with the peak in 1994 with 157 references. In addition to short-term stopping sites, the
city offered an increase in social support linked to housing, “sensitization work with
partners” and financial aid. The North Department and the Urban Community of Lille
conducted a study on welcoming French travelers in Lille, "Study prior to the
development of the Departmental Scheme of reception areas for Travelers: Welcoming
Travelers in the Lille District.” In the introduction of the report they acknowledged the
relevant stakeholders: residents, travelers, elected officials and police. The Urban
90
Community then acknowledged that all actors had duties as well as rights, and that to
deny a group their rights was to marginalize them. But the Urban Community did not
take as much responsibility as it could have,
Over time, the arrival of caravans in a municipality is unraveling in an
increasingly pronounced climate of suspicion, where everyone protects
themselves against the other and thus causes conflictual situations for all
(residents, travelers, elected
officials, police...). Today, our region must face a deficit of parking sites to foster
harmony between the Tsiganes and the local communities, where everyone would
have the time to assume their duties and obligations while respecting their rights.
To deny Tsiganes and Travelers their independence through their way of life
would be to…marginalize them.
The Urban Community acknowledged how the attitude of “flexibility” and respect for
differences was necessary for stopping-site policy-making,
The stopping place for travelers, whether temporary or not, precarious or
organized, obliged or decided, must be considered as a privileged meeting place
(between nomads, residents, elected officials and local officials...). On the other
hand, it should in no case be solely a parking lot for caravans but a set of
dwellings adapted to a way of life.
Lastly, the working group uncovered a need for “reflection,” greater
“intermunicipal coordination,” and “flexibility” in allowing certain caravans to stay for
longer. These were advanced observations for the time and provide evidence that this first
era of stopping-site policy for French travelers may have laid the groundwork for housing
policy for EU migrant Roma later on,
The reception of travelers raises many comments and questions on the part of
local authorities, local services, as well as the population, on the district of Lille.
The objective of this study was therefore, to observe the situation and to introduce
a different reflection. From the synthesis of these observations results the need for
intermunicipal groupings, and the need for impact studies, to better understand the
difficulties of designing parking lots in municipalities. In addition, it seems
important to take the regulatory provisions …to allow certain caravans to park for
a longer period.
91
Second Housing Policy Era (2007): Period of Policy similarity, integration villages in
both cities.
Lille
In Lille, the second housing policy era (2007-2010) concerned accommodation for
EU migrant Roma and was the beginning of Lille’s civic legacy development. Lille
residents started to mobilize against the installment of integration villages, this is
reflected in newspapers clippings from the archives. However, in response, some NGOs
used policy frames to refer to their own role building “experiential knowledge” and
“trust,” referred to social workers as “mediators,” and referred to the “European
Cosmopolitanism” of Roma migrants to build political will for integration villages for
migrant Roma. These policy frames were localized in their specific focus on integration
villages in Lille, and were supported by NGOs such as AREAS and AFEJI, and Mayors
like Martine Aubry.
In Lille, the policy frame timeline depicts a big spike in references to housing for
migrant Roma in 2008. Between 2008 and 2010, references spiked from 10 references in
2008 to 370 references in 2010. During this initial period of integration villages between
2008 and 2010, the most common policy frames referenced were “European
cosmopolitanism,” “experiential knowledge,” “trust” and “stakeholder coordination.”
Integration villages were described as an “airlock,” which evoked exclusionary
policy frames. By 2009, migrant Roma individuals who had citizenship in Romania were
EU citizens and so “European Cosmopolitan” policy frames were introduced. The
Mayor of Villaneuve d’Ascq, Gerard Caudron, “recalled that the Roma were citizens of
the European Union and expressed his refusal to ‘evict the slums.’ The European
Comsopolitian frame had a few interpretations. For example, a Lille resident described
92
being proud to have migrant Roma as neighbors, “I am proud to live in this neighborhood
and the arrival of these Europeans near my home.” At the same time, EU citizenship
could be an opportunity for cities to eschew their responsibilities toward migrant Roma
families. For example, a Lille Metropolis mayor Gerard Caudron declared, ‘it is up to the
state and Europe to face up to their responsibilities.’ In these cases, EU migrant Roma’s
EU identity only served as further evidence that the responsibility to accommodate
families was with the France or the EU, not Lille.
The mayor of Lille at the time, Martine Aubry, was a strong advocate for
integration villages. In May 2009, Mayor Martine Aubry blamed state budgets for the
city’s inability to welcome new families and asked for more funds for integration
villages. Then, on 19 December 2009, while Mayor Martine Aubry called on mayors to
welcome Roma families in the metropolitan region, but there was push back from some
Mayors, “this Roma village will add misery to misery and pull the neighborhood from
below.”
The municipal and metropolitan governments realized that during private
discussions there was more opportunity for dialogue about accommodation options for
migrant Roma families. This reflected an awareness that public policy frames and
discussions were not helpful for policy development. Lille established a working group to
“make it possible to reflect together and ask about available land." Closed to the public
and the press, this working group could make it possible to envisage local solutions, at
the very least to open a dialogue between the main actors concerned.
93
By 2010, Lille Metropolis considered the integration villages policy a success.
“The experiment is a success. Six programs of the same type will see the light of day by
the end of next year.”151 Once the city could count the new integration villages as a
successful policy, the city of Lille took more responsibility for Roma integration, “there
are others [integration villages] in Seine Saint- Denis, in Nantes, but it's in the North that
is doing it well. In the department, these are small integration villages, which
accommodate between two and five families at most. It blends more easily into the local
fabric.” Some recognized the implementation of the integration village policy as an
opportunity for the city to prove itself, to take responsibility. During the meetings of the
North district council, in particular, there was a lot of talk about one migrant Roma
integration village. “But the mobile homes are now there, and they will soon be inhabited.
It will therefore be necessary for everyone to do their part: after all, isn't this
the opportunity to show that the famous ‘living together,’ so pronounced on paper, can
find a happy concretization here?”
This is evidence that before Sarkozy’s Grenoble speech, integration villages were
a politicized phenomenon in Lille. Four months before the Sarkozy speech, the Socialist
leadership in Lille was defending its introduction of more integration villages despite
backlash from some residents and the political opposition represented by the center right.
There was evidence of the left exerting its ideological commitment to the policy of
integration villages, “I am the left, elected on a municipal program of the left, I am not
elected to apply the politics of the right.” Even before Sarkozy’s Grenoble speech, there
94
was the politics of the left and politics of the right in Lille and these political themes had
strong organizing policy frames in regard to migrant Roma accommodation. By
2010, both political parties were “using” the Roma integration issue to garner electoral
support.
Toulouse
In Toulouse, 2007 is the first year “Roma” were mentioned in a city council
meeting. “Roma” were mentioned for the first time because the metropolitan area of
Toulouse was participating in a meeting at the EU. “Indeed, the President of [the
metropolitan area], member of the steering committee of the European program
‘Development of local initiatives for the inclusion of Roma and Travelers,’ took part in
the first meetings of this committee in December 2006.” However, there was not much
95
talk about integration villages at the time. There was much more discussion in Lille in
2007 regarding the introduction of integration villages and it was not until 2010 that
policy frames proliferated once again.
In 2010 national level events influenced Toulouse’s policy frames. Sarkozy’s
speech at Grenoble in July 2010, followed by the leaked memo by the Minister of the
Interior Brice Hortefeux, dated August 2010, incited an ordering moment across France
in regard to policy for migrant Roma, as city representatives wanted to be on what they
deemed to be the “correct” side of the issue. 156157 Previously, policy frames that
referenced migrant Roma were Toulouse-specific, based on scenarios and actors at the
local level. After 2010, these same frames concerned international and supranationallevel
policies and actors. Toulouse City Council meetings demonstrated that the city’s
representatives had an enhanced awareness of the power of policy frames to
instrumentalize a political issue in the wake of Sarkozy’s comments. Toulouse City
Council released a statement that the speech given by the President Sarkozy in Grenoble
introduced an “openly discriminatory policy,” and that instead of “easing social tensions,
exacerbate(d) them, [and] it also exempt[ed] itself from any duty of mutual aid and
solidarity.”158
156 It is important to clarify that the young Romani who was killed and whose death sparked protest was a
French citizen, not a migrant. Ethel Brooks, "Stop this state persecution of Roma," The Guardian, August
19, 2010, https://www.theguardian.com/commentisfree/2010/aug/19/state-persecution- Kristi Severance,
“France’s Expulsion of Roma Migrants: A Test Case for Europe,” 2010, The Online Journal of the
Migration Policy Institute. https://www.migrationpolicy.org/article/frances-expulsion-roma-migrants-
testcase-europehttps://www.migrationpolicy.org/article/frances-expulsion-roma-migrants-test-case-europe.
157 While there were only 69 references observed in this spike, these references more often directly
Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard des »
pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
96
addressed “Roma migrants,” while previous references implicitly addressed “tsiganes,” and “migrants”
more generally.
158
Roms,
Toulouse Online Municipal Archives.
Toulouse City Council went so far as to criticize Sarkozy’s efforts at electoral
instrumentalization with his speech, “this policy, dictated by electoral motivations,
stigmatizes a population in the eyes of public opinion and in fact fuels prejudice.”159 A
city council Green Party representative pointed out the instrumentalization of the
Sarkozy speech at Grenoble, “This political crisis is also due, more broadly, to a political
practice implemented by Nicolas Sarkozy and his colleagues, that of the
instrumentalization of the event.”160 Even more than the stigmatization of an ethnic
group, it was the political instrumentalization—the effort to manipulate policy frames—
that was condemned, “It is this political practice that we must denounce and it is another
political practice that we must embody in order to live up to the political moment…”161
And finally, the city council criticized the discursive move that Sarkozy made by
conflating “delinquency” and “immigration” policy frames, “the political class should
claim, on the contrary, not to deal with the issue of delinquency at the same time as that
of immigration.”162 This move can be referred to as discursive amalgamation; when an
actor conflates two different concepts—either intentionally or unintentionally. This
engineered association between the two concepts then gives each policy frame a new (in
this case, discriminatory) meaning when mentioned in certain contexts. Toulouse City
Council members went a step further and warned against the ills of discursive
Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard des »
pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
97
159 Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard
des Roms, » pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
Toulouse Online Municipal Archives.
160 Toulouse Municipal Council, « Séance du vendredi 24 septembre 2010, » pg 9, 24 September
2010, Toulouse Online Municipal Archives.
161 Toulouse Municipal Council, « Séance du vendredi 24 septembre 2010, » pg 9, 24 September
2010, Toulouse Online Municipal Archives.
162
Roms,
Toulouse Online Municipal Archives.
amalgamation. For a municipal meeting, this was a nuanced discussion concerning
discourse, led by the opposition, Jean-Luc Moudenc of the center-left,
…so that means that you are in line with what Mr. Nicolas Sarkozy does, that is
to say talking about insecurity, immigration, and the Roma and that is exactly
what we accuse the Government of. It is not the fact that every once and a while
we have a Republican reminder of the occupations of places that are illegal, that
is normal. We do referrals. But the debate as it was launched is that it started on
security with Grenoble and bandits in a city; then it was discussed with
immigration. And finally it has been harped on by all European countries, it has
been stigmatized around a type of population called the Roma… It's true that
when we talk about Travelers, which has nothing to do with the Roma and even
less with insecurity, well, there are also rights and duties, and the municipalities
are not meeting their responsibilities vis-à-vis the areas. So all this is a debate
that must be conducted calmly and the Left is not angelic, but it is certainly more
Republican than the current (national) Government!163
Here, the opposition is pointing out not only the danger of discriminatory policy, but also
the danger of conflating or associating policy frames, particularly stigmatizing policy
frames, with a specific ethnic group, in this case Roma. This demonstrates that by 2010,
Toulouse’s political representatives were well aware of the influence of policy frames.
This city council meeting concluded the announcement with policy prescriptions
for the national and local level, “Instead of deporting people who, as targets of
discrimination in their country of origin, will keep coming back, it would be appropriate
to prepare for their integration...”164 They recommended reception areas, schooling for
children and literacy and professional training for adults.165 They mentioned the
Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard des »
pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
98
importance of “will” to support Roma populations for “more dignified living
conditions,” and then cited themselves as an example of a city that has implemented
“support and
163 Toulouse Municipal Council, « Séance du vendredi 24 septembre 2010, » pg 9, 24 September
2010, Toulouse Online Municipal Archives.
164 Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard
des Roms, » pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
Toulouse Online Municipal Archives.
165
Roms,
Toulouse Online Municipal Archives.
Toulouse City Council, “Vœu du Conseil Municipal Relatif a la Politique du gouvernment a égard des »
pg 1, 24 September 2010, Conseil Municipal N° 6 du 24 septembre 2010 Délibération n° 84,
99
integration schemes,” to achieve better living conditions. Finally, they asked European
member states to respect Roma as European citizens who have the right to enjoy the
freedom of movement throughout Europe and they named and shamed the Sarkozy
administration by declaring that, “the designation of a scapegoat dishonors the homeland
of human rights.”
In October 2010, the city council held another meeting and Roma integration
policy was heavily discussed. Not only was integration policy discussed but the nature of
the debate regarding the policy was discussed. A Socialist Party representative analyzed
the debate between the right and left as “Manichean;” as an unnecessarily binary debate.
In response, once again, the Mayor stressed the danger of amalgamations, not just
between ethnic groups, or “evictions” and “deportations,” but even broader concepts,
themes, and discourses such as “social situations” and “personal statuses,”
And this is where the President of the Republic and the Prime Minister moreover
overflowed this summer: it is that we must always be careful of amalgams
between social situations, or personal statuses, which are not necessarily or even
often security issues. That is to say, you have to be very careful not to mix things
up - so obviously I won't talk about Roma - I could talk about the homeless, for
example. There are enough sufficiently strong concerns about the problem of the
homeless, without falling into the amalgam of insecurity, because insecurity is
different, it is really a problem that is totally different.
The latter half of 2010 was an inflammatory year for political rhetoric concerning
Roma due to Sarkozy’s speech in Grenoble. However, in Toulouse, policy regarding
Roma remained in city council meetings beyond 2010. By 2012, the center-left criticized
the national policy introduced in 2012 by Manuel Valls, Minster of the Interior for
Sarkozy, to dismantle illegal settlements,
100
…we repeat it to Manuel VALLS: beware, the eviction of Roma in a systematic
way reproduces occupations of land. Toulouse is not Marseille, fortunately, since
in Marseille, in view of an experience of the summer, citizens declared
themselves police officers which the jurisdiction to go and evict Roma.
However, the Mayor also admitted that the city was struggling with the
governance of “violence,” in Toulouse, and the opposition did not hesitate to point this
out.
Manual Valls, Minister of the Interior in the Hollande government, closed out the
second housing policy era by claiming that, “Roma ‘do not want to integrate’ and that
their lifestyles are ‘clearly in confrontation’ with the French way of life. He added: ‘the
majority [of Roma] should be delivered back to the borders. We are not here to welcome
these people’” (van Baar 2014, 31). So between 2010 and 2013 at the national level
migrant Roma were depicted as “moral outcasts,” “unwilling to integrate and contribute
to society,” which left local level governments to undo that policy frame and introduce
alternative policy frames that encouraged integration policy instead of eviction and
deportation (Kostka from Magazzini and Piemontese 2019; 157).
Lille
In Lille in 2010, in the aftermath of the international condemnation of Sarkozy’s
speech and the leaked interior minister’s memo, the accommodation of EU migrant
Roma became an issue of “French values,” of “European values,” of “identity,” of
“human dignity” and “rights.” Whether Lille decide to dismantle camps and evict
migrants, or
not, became a much more politicized choice. Suddenly, accommodating Roma was an
issue of identity and “nationalism,” that made cities respond with policy changes,
“refusing the policy led by the government with regard to migrant populations, judged
101
by many in France and abroad to be 'contrary to the values of France', the urban
community has decided to accelerate its action intended to welcome these populations
with dignity and more generally Travelers and homeless people.”
In the aftermath of the 2010 Sarkozy speech, the policy frame “proactive policy,”
was emphasized. This policy frame is unique to French discourse and is a major policy
frame in Lille that distinguishes the extra responsibility the city takes on to introduce
policy that legally should be managed by the department or federal government.
Proactive policy refers to policies that the city implements that are outside of its official
legal mandate and are typically state-level responsibilities. In Lille, integration policy
intended for migrant Roma as well as other groups was referred to as “proactive policy,”
“the city of Lille, faithful to its social tradition, introduces a proactive policy to fight
against exclusion” (CCAS Report 2010). Through this proactive policy, Mayor Martine
Aubry claimed to promote dignity, health, security and tranquility,
at the heart of the roles of the State, this reception cannot be done without the
involvement of the municipalities and the responsibility of the mayors must be
supported. It is this intervention shared between the State, Lille Metropolis and
the municipalities that contributes to ensuring public dignity, health, security and
tranquility.
The municipalities that demonstrated the political will to house migrant Roma
implemented “proactive policy,” which supplemented the state’s legal jurisdiction in
housing integration.
As you know, Lille Metropolis has embarked on a proactive policy in matters of
welcoming Roma populations on its territory, even though this policy is not part
of the competences of the urban community, in order to guarantee these families
decent living conditions, and to facilitate good acceptance by local residents. This
policy is reflected in the development and management of several parking sites,
102
the creation of integration villages and the search for housing solutions in diverse
areas.
By 2012, proactive policy included the establishment of the parking sites,
integration villages and housing opportunities in diverse areas. In Lille’s report to the
national French strategy for the integration of Roma until 2020, Lille described that a
proactive policy required that local actors were dedicated to the objectives which in turn
required mayoral support and ultimately the political will to welcome migrant Roma at
the level of each municipality, “a proactive policy in this area will only have an effect if
the local actors, already at work on the ground, are dedicated to their objectives.” The
fact that proactive policy is such a common policy frame in Lille indicates that actors are
consistently debating whose responsibility it is to introduce policy for migrant Roma. In
the aftermath of the 2010 and 2012 federal-level controversies, the “proactive policy,”
“European identity,” “NGOs as mediator and source of trust,” and “stakeholder
cooperation between the state and city,” policy frames were strengthened along with the
integration villages policy.
By 2012, Interior Minister Manuel Valls called for public order and the reduction
of slums. There was a rhetorical focus on “flows of people” and “migrant flows.” Both
sides compared EU migrant Roma integration policies to the French republic’s values
and standards and to the French identity, however these frames were used to make
nationalist/nativist claims. By 2013, integration villages were promoted by Lille NGOs.
Lille NGOs mentioned their “expertise,” the “trust” they fostered between families and
103
the city, as well as the “flexibility” they provided in housing options, and sometimes they
articulated policy critiques and critiques of their own practices.
In response to the July 2010 deportations the political left, represented by the
Greens, critiqued the state representatives who did not actively support the Roma. Malik
Ifri, (Green party) in charge of the issue of travelers and Roma in the metropolitan
region, articulated the lack of political will on behalf of the State with regard to
integration policy for Roma, “I have never seen the prefect requisition land for the
reception of these populations. Moreover, it should not be forgotten that the Roma are
European citizens with specific status. France imposes very strong constraints on them
which do not favor their integration and their employability.”
However, some center-left mayors were unhappy with the moratorium on
evictions. During a city council meeting Mayor Gerard Caudron of the center-left called
again on the authorities responsible for Roma policy,
While the moratorium on evictions decided by the president of the urban
community inevitably suggests new encampments, Gérard Caudron believes for
his part that “the Roma file has resumed its spiral of failures...” Gérard Caudron
already draws the line not to cross, “I do not accept that this results in new
encampments…
almost everywhere, the day before winter, in Villeneuve d'Ascq. Encampments
that will put us back in serious trouble.” The alarm signal is sounded, will it be heard?
However, there is also evidence that Mayors also showed solidarity for one another by
acknowledging when the migrant Roma issue was exploited by political opponents, “I
know that as Mayor you share this concern of welcoming these families in good
conditions. I salute this commitment and I can only imagine the anger you feel in
104
witnessing this complex issue exploited by an opposition that calls for hatred and
rejection of others.”
In response to the international backlash to Sarkozy’s request to dismantle Roma
camps, Mr. Marc-Philippe Daubresse and Mayor Martine Aubry introduced six more
integration villages, and the newspapers portrayed integration villages as the “nordiste”
(Northern) example of Roma integration. Lille’s moratorium on camp evictions was
introduced because of Sarkozy’s comments, “In summary: Martine Aubry has frozen the
procedures for evacuating Roma since the expulsion measures taken by the (national)
government. If land has to be freed up for operational reasons, the city must find other
land in compensation to accommodate the Roma.” However, by the end of the year,
social workers were conducting social surveys with the families to get a sense of who
they were and what they wanted when they were evicted,
the summer period was synonymous with exclusion, precariousness and moral
and physical violence. Following [the metropolitan area’s] positioning to stop the
evictions of [Romania] Roma families on its land, things calmed down and the
stabilization of parking places facilitated the work of social support. Thus, at the
end of the year, work was carried out with [the metropolitan government] in
terms
105
of expertise and social survey in the field in order to find answers adapted to the
lifestyles and functioning of families.
While the EU had already agreed to fund new integration villages in 2009, in the
wake of the Sarkozy scandal in July 2010, the policy of integration villages was at the
forefront of the EU and Lille’s toolkit concerning housing integration of EU migrant
Roma. After 2010, Lille wanted 10 integration villages set up by the end of 2011.
However, the integration villages had a specific policy frame on the center-right.
The center right was focused on the families’ identities as Romanian or Bulgarian, as
well as their specific immigration status that allowed them to travel but not stay in the
EU without a job for over three months,
The Roma of Romania or Bulgaria concerned must demonstrate a real
willingness to learn the language of Voltaire, educate their children and look for jobs,”
says Marc-Philippe Daubresse…We are ready to finance integration projects for people
in a regular situation who demonstrate a true willingness to integrate. But we will not
finance the structures for welcoming people in an irregular situation. Mr. Marc-Phillippe
Daubresse of the UMP (center-right) who was responsible for the opposition group at the
metropolitan level, reportedly justified the evictions the Sarkozy government conducted.
At the national level he served as the secretary for housing from March to November
2004 and from November 2004 to 2005 he was the deputy minister for housing and
cities. At different times Mr. Marc-Philippe Daubresse was the Vice-President of the
Urban Community of Lille Metropolis (2001-2008) as well as the mayor of Lambersart
between 1988 and 2017. In Lille, Mr. Daubresse supported a mix of policies that
reportedly confused the majority government in Lille. While he supported
the evictions that Sarkozy called for and did not support the policy to pay for hotel
rooms (emergency housing) for migrant Roma, Mr. Daubresse also supported the
Archives,
106
integration village policy. Mr. Daubresse called for the closure of hotel rooms for EU
migrant
Roma, referencing a welfare magnet policy frame,
The urban community's plan to find solutions for the Roma has found its funding.
But behind the State-[metropolitan] agreement, the communication war is
raging…Marc-Philippe Daubresse used the welfare magnet argument to justify
the closure of hotel rooms. The state did not intend to continue funding these
schemes indefinitely…we cannot welcome all the misery in the world.
The implicit logic of the welfare magnet frame is that of welfare chauvinism. The
idea is that if the city introduces welfare policies to support marginalized migrant
families that policy would attract more people that the city was not equipped to
welcome, so therefore cities should not introduce welfare policies. Mr. Daubresse has a
lot of institutional power over time and across institutions and so his opinion held sway.
At the very least, Mr. Daubresse’s support of the integration villages reveals that
the integration village policy was attractive to those who aligned themselves with a
specific policy frame. Particularly, representatives like Mr. Daubresse were sympathetic
to the policy that if a Roma family wanted to live in an integration village, they had to
demonstrate a “willingness to integrate,” under this policy. Taken to its logical
conclusion, this “deservingness” or “meritocracy” policy frame was underpinned by the
same policy frame as the welfare magnet policy frame. Therefore, while it is already
difficult for migrant families to meet the standards for integration, some political actors
reference the “deservingness” or “meritocracy” frame to rationalize why some families
are not “fit” to remain in France.
Archives,
107
Then, in August 26, 2012, Interior Minister Manuel Valls of the new Socialist
French government issued a circular on anticipating and supporting operations of
evictions of unauthorized settlements,” signed by seven ministries. This ended the
moratorium on evictions and between June 2013 and November 2014, nine camps and
around 2,000 people were evacuated (including a camp of 550 people in Lille). Lille
conducted social surveys in an attempt to give Romanian and Bulgarian migrant Roma
housing options before they were evicted, “in accordance with the ministry circular of
August 26, 2012 relating to the anticipation and support of operations to evict illegal
camps, this decision was taken on the basis of an in-depth health and social survey and a
proposal to stay given to each family.”
The integration village projects were emphasized as a policy for nationals of the
EU (not “simply” Roma families) and representatives emphasized the inclusion of
migrant Roma from Romania at the time, not only the welfare magnet policy frame. In
this way the “EU cosmopolitanism” theme was further referenced in the wake of the
Sarkozy speeches,
The exceptional character lies in the nature of the project, namely the hosting of a
few Roma families, which is part of a metropolitan plan to provide temporary
prefabricated accommodation to respond urgently to the situation of great distress
and great fragility of Roma families who are nationals of the European Union and
homeless.
This effort to provide integration villages was recognized and funded by the EU through
their ESF and EDRF programs, and was portrayed as a model integration program with
an emphasis on the work of NGOs,
Archives,
108
European credits will be requested for all initiatives. In particular, Europe has
opened up the possibility of funding housing projects for marginalized
communities…The ESF could also be mobilized for the actions carried out in
favor of the economic and social integration of the Roma, in particular through
the NGOs working in favor of these populations.
Political leaders discovered that neighbors and neighborhood problems were
reportedly the source of electoral backlash, so with the integration villages policy, the
metropolis kept the numbers of residents low, “These experiences take place without
neighborhood problems, because of their limited size, the willingness of families, and
the support they receive.” The city relied a lot on the “expertise” and “trust” that NGO
members built with the families to mediate the relationship between families in the
integration villages and their neighbors. NGOs such as CANAL offered social support
that “ensured good neighborly relations with local residents…CANAL offered
information and explanation upstream of an installation because for local residents, it is a
Archives,
109
subject of concern.” NGOs such as CANAL would speak to the neighborhood even
before an integration village was installed in order to dialogue with the families and help
them understand what was going on in their area. Municipal councils had a good
understanding of the pivotal role NGOs played in establishing trust between Roma and
their neighbors. By this point, Lille has finally understood the role that policy frames and
networks play in influencing integration policy.
By 2012, the NGOs CANAL, AREAS and AFEJI were working closely to
provide social support for migrant Roma which strengthened political will for
integration policies
CANAL is in regular, even constant, contact with the other associative structures
officially assigned to the field: the AREAS which is in charge of social support,
schooling and the health of Residents. The AFEJI which supports us in the
possibility of hiring residents of the integration village.
CANAL touted a level of “policy critique” that supposedly allowed them to support
Roma migrant families better, “without judgement or interpretation.” The CANAL NGO
endeavored to listen to all relevant stakeholders and to educate its members in order to
communicate their own understanding of Roma “culture” and “way of life,”
CANAL is listening, CANAL is constantly listening to elected officials and the
general public. The goal being to make available our knowledge of the culture and
the way of life of the residents without judgment or interpretation, in a completely
transparent way in the desire to silence certain prejudices, certain stigmatizing or
hateful ideas, but without obscuring the difficulties and nuisances caused by
informal camps. Thus, the CANAL teams regularly attend meetings, debates and
conferences on the subject of "Roma.”
Some of this language such as the “way of life” of the residents, can be alienating
and exclusionary, but in the context of educating social workers, listening to all
stakeholders and encouraging critical thought for the people providing the service and not
the migrant families is unique for the context.
It is around 2012 that CANAL realized its role as a mediator and communicator
between Roma residents and the city, “the NGO is committed to the establishment of
coordination, the establishment of mediation and communication, to bring a better life to
the Roma public…” The city put a lot of emphasis on pedagogy between municipalities
regarding best practices, “the urban community and the departmental union of the CCAS
will jointly bring together all the municipalities, in order to know and promote the
exchange of practices of solidarity with regard to the populations received.” This
emphasis on pedagogy continued into 2012 when the NGO CANAL joined the
stakeholder network.
As the March 2014 municipal elections grew closer, the center left searched for a
policy equilibrium which would “enable a certain number of Roma to meet the conditions
for integration within the Metropolis, but also prevent[ed] the development of informal
settlements which have consequences in terms of health, hygiene, security and
neighborhood.” Outside of the context of the elections there was still anti-Roma vitriol,
“a lifting of anti-Roma outcry...counterbalanced by the position taken by a collective of
humanist NGOs and left-wing political parties.” The left continued to call out
discriminatory comments made by the political opposition, “deploring a mountain of
irresponsibility in the remarks made about Roma families by the political opposition and
by all those who express themselves in this debate...”
At this time, it was reported that the topics of migrant Roma and encampments
were hot button issues, “in municipal elections, candidates are campaigning in particular
on the issue of the reception of migrant Roma and encampments. The subject becomes
very controversial and dividing. Communities then adopt a cautious attitude on this
subject” In Lille, the policy era of integration villages (during 2010-2014), was the most
incendiary period of time for the topic of housing for migrant Roma.
Conclusion
The evidence supports De Graauw’s findings that policy framing can help develop
the political will for integration policy for stigmatized and discriminated migrant groups.
How housing policy for migrant Roma is framed (e.g. through the lens of
humanitarianism, human rights, proactive policy, mediation) influences whether there is
enough political will at the city council level to introduce the policy.
In Toulouse, the civic legacy of housing policy frames included “mediation,”
“pedagogy,” “diversity,” “humanitarianism,” “critical policymaking” and “stakeholder
coordination.” These policy frames were initially articulated by local NGOs in their
efforts to re-house migrants in the 1990s. Later on, these policy frames were adopted by
city council members. However, policy frames were articulated most vocally by policy
entrepreneur Mr. Daniel Rougé. His role is theorized more broadly and contrasted to
Lille’s discourse entrepreneurs in chapter 7.
Lille’s policy frames included themes such as “pedagogy,” “social support,”
“dignity,” “mediation,” “flexibility,” and “proactive policy.” Lille did not have a similar
first housing era like in Toulouse in the 1990s that was triggered by NGOs’ management
of camps, instead it was the introduction of integration villages in 2008 that shaped
policy frames regarding migrant Roma housing to a greater extent in Lille. While
Toulouse main policy frame was “humanitarianism,” Lille’s main policy frame was
“proactive policy.” According to Chung (2020), policy frames structure debates over
integration policy and provide tools for migrant advocates, “while ideational frames may
constrain immigration policy debates and reforms, they also provide the tools for making
claims legible to the state and the larger public” (Chung, 27).This suggests that
Toulouse’s understanding of its responsibility was to the humanitarian needs of migrant
Roma, while Lille’s understanding of its responsibility was in relation to the state’s legal
responsibility to migrant Roma.
This next chapter will answer the question, how do policy entrepreneurs use civic
legacies to shape contemporary housing policy for EU migrant Roma? This chapter will
demonstrate the link between civic legacies of policy frames, networks and policy
entrepreneurs by looking at a specific critical juncture when a policy entrepreneur
changed policy through the use of a specific historical policy frame. This chapter will
describe how the policy entrepreneur convinces political opposition to follow his lead by
introducing specific policy frames and drawing on the civic legacy of the town regarding
housing integration.
Chapter 6. Policy Frames and Entrepreneurs (2007-2020)
Chapters 4 and 5 set out the policy frames and networks that comprise the civic
legacy in Toulouse and Lille. This next chapter will demonstrate how the civic legacy in
Toulouse was re-activated in 2015 when a policy entrepreneur (Mr. Daniel Rougé) used
historical policy frames to redefine the issue of housing and generated the political will to
introduce new housing policy. Meanwhile, while Lille may have had evidence of an NGO
network and policy frames by 2007, Lille did not have necessary contemporary policy
entrepreneur to make the pivot in 2014, instead financial incentives drove policy change
in Lille in 2017.
This chapter explores how policy entrepreneurs can draw upon the civic legacies
of networks and policy frames coordinated during groups’ previous attempts to integrate
to change contemporary policy. One way the NGO network provides symbolic
significance and legitimacy to housing integration efforts for migrant Roma is through
policy frames. In her analysis of integration rights and policies for migrants in San
Francisco, De Graauw (2016) finds that policy frames are customized and introduced by
NGOs to connect with city officials, voters and the media, in order to successfully
navigate “a highly politicized environment in which their objectives remain a difficult
political sell” (De Graauw, 17).
In Toulouse there is a history of individuals who directed the NGOs which
established the trust, mediation and expertise that provided accommodation options. One
individual, Dr. Guiraud ran a pivotal organization for nearly 30 years between 19711998.
This long-term commitment spearheaded by one person established a cohesive policy
frame around accommodation for “tsiganes” in Toulouse, which included migrant Roma.
The policy frames established in the 1990s were activated when new leadership took
control in 2015. The 2015 policy pivot from short to long-term housing policy was
spearheaded by a policy entrepreneur, Mr. Daniel Rougé. Lille, on the other hand, did not
have a historical policy entrepreneur with similar institutional longevity, or a policy
entrepreneur to change policy in 2015. This process of successfully leveraging policy
frames to convince multiple stakeholders (predominately municipal mayors, prefects and
political opposition) of a specific policy delimits the housing policy options available for
EU migrant Roma.
In Lille, while NGOs developed policy frames and networks through their
mediation and social support of migrant Roma, this development occurred later, in 2007
as opposed to 1990 in Toulouse. This could explain why Lille was slow to pivot from
integration villages to long-term housing. While the Mayors Pierre Mauroy and Martine
Aubry were vocal Mayors who acted as policy entrepreneurs, they could not convince the
center right parties, other mayors and NGOs that long-term housing policy was
preferable. Furthermore, policy entrepreneur, Mr. Marc-Philippe Daubresse, was heavily
involved in multiple Lille political institutions as well as national institutions and was not
interested in developing political will for EU migrant Roma housing. This evidence
supports the argument that in Lille, with a delayed civic legacy and without a policy
entrepreneur to develop the political will for housing, there was no independent policy
change from integration villages.
The Role of Policy Entrepreneurs
Civic legacies, historical policy frames, networks and policy entrepreneurs
influence housing policy. However, the evidence supports that the role of one
contemporary policy entrepreneur in Toulouse was necessary for policy change. Without
the political influence of one policy actor, Mr. Daniel Rougé, it is unclear whether the
presence of a civic legacy would be enough to result in policy change in Toulouse 2015.
One source of authority for the policy entrepreneur is how she persuades her
political opposition by articulating specific policy frames over time. This actor has the
power to convince political actors such as the center left and right party leadership as
well as national level representation such as prefects and NGO advocates. Anyone with
that level of persuasive power while articulating a specific policy frame over time (around
ten years or more) was identified as a policy entrepreneur. If they did not have that level
of influence, they probably would not have ended up in the archives.
In Toulouse, Dr. Jean-Claude Guiraud served as a policy entrepreneur for over 30
years during his service as the director of A.T.S./C.C.P.S. from 1971-2005. In total, Dr.
Guiraud was involved with Roma in Toulouse for 40 years (1965-2005). This long-term
commitment spearheaded by one person established the trust and expertise necessary to
communicate policy frames concerning migrant Roma integration.
In the Toulouse archive, most documents published between 1970-2000 mention
Dr. Guiraud as the president of A.T.S. and then C.C.P.S. Dr. Guiraud ran A.T.S. for 30
years, beginning with his involvement in the Ginestous camp in 1966, “Jean-Claude
Guiraud, a 5th year medical student, introduces a reception clinic, in the camp of
Ginestous...because the ‘official’ teams did not dare to go to the camps. The reception
clinic was a converted freight car.” In his interview with C.C.P.S., Guiraud reported that
he was motivated to visit the camp of his own volition, with nothing but an understanding
of “how much more this community was rejected,” and a concern for the infant mortality
rates of the Ginestous camp,
since I was in the 5th year of medicine when I went, voluntarily, no one sent me
there, I went to the Ginestous camp since I had already realized much more how much
this community was rejected, the Gypsy population in general. And from the outset there
was also a very high infant mortality compared to the national average of the time…very
quickly I thought that for these questions concerning the exclusion of the Gypsies it was
necessary to mobilize disciplines other than mine, medicine, and that it was necessary to
mobilize other wills than mine, hence the idea of creating the NGO which called Foyer
d'éducation populaire de Ginestous, then called Amitié Tsigane de Toulouse, then called
A.T.S. and then C.C.P.S. Dr. Guiraud continued to lead this group until 2005, so as not to
“lose” his “sense of reality,” in other words, it seems his work with migrants grounded his
medical career and sense of self.
Dr. Guiraud oversaw and supported an extended moment of hospitality toward
migrant Roma on the Toulouse University Rangueil Campus. After the fall of the
Ceausescu regime in 1989 Romania, Toulouse received migrant Roma from Romania and
Dr. Guiraud was part of that integration process, “at the time, the 1st major problem for
them (migrant Roma) was that of their regularization, we worked a lot with the social
service for aid to migrants, we worked with lawyers, we worked with Secours Catholique
to help their application file as asylum seekers.” On Rangueil campus there was a surge
of hospitality toward migrant Roma from Romania, what Dr. Guiraud referred to as
“benevolence,” to the extent that access to housing and social support was provided by
the University community,
on the Rangeuil campus, it was surreal. They (Roma migrants) arrived just before
the end of the Ceausescu era, so at a time when there was a great "optimism" of
citizens for all that was aid to Romania… So great benevolence, great
benevolence, to such an extent once again that we had put in place actions,
relating to protection and access to care, and some actions relating to housing,
they (Roma migrants) were very surrounded by the community of teachers and
students on campus.209
This hospitality lasted two years. Even though Toulouse had a policy entrepreneur in Dr.
Guiraud and a surge of hospitality in their civil society, the Metropolis failed to provide
public housing options for the first wave of Romanian Roma. Dr. Guiraud described his
attempt to involve state level stakeholders to re-house migrant Roma before they were
evicted,
We tried to put in place rehousing solutions, they did not arrive. I brought
successive sub-prefects, at least two to the site. We had considered a somewhat
lengthy relocation assistance procedure, which did not take place, and in fact it
was they themselves (migrant Roma) who relocated, in the private housing sector.
They had no help with social rehousing.”210
Dr. Guiraud was an important policy entrepreneur for Toulouse beginning in the
late 1960s. He established trust between prefects, local NGOs and the city, mediated
between Roma families and the city and had the medical and social expertise that was
necessary to provide opportunities for integration. This long-term commitment
spearheaded by one person established cohesive policy frames around accommodation
for “nomades” in Toulouse. These older policy frames were used again when new
leadership took control in 2015. Toulouse had a policy entrepreneur in Mr. Daniel Rougé
who worked at the city council and echoed specific policy frames that were used by Dr.
Guiraud in the 1990s in his role as NGO director.
Third Housing Policy Era (2015): Policy Divergence between Toulouse and Lille,
Toulouse Pivots from Integration Village Policy and Leads Integration Policy
Toulouse
The Toulouse City Council minutes revealed that Mr. Rougé was central to the
change in Roma integration policy in 2015. In the minutes, the Mayor often called upon
to Mr. Rougé whenever there was a question regarding migrant Roma. Stakeholders who
had worked in Toulouse on the topic of housing for migrant Roma prior to Mr. Rougé’s
appointment said that with Mr. Rougé’s involvement, the integration effort was
“progressed further” under his leadershipvii They described Mr. Rougé as sincere, “Mr.
Rougé, pushes for inclusion and solidarity. Inclusion—it is not a vain word for him. It's
really means something.”viii Also his specific medical authority resonates with Toulouse’s
civic legacy of “humanitarianism” policy frames. Mr. Rougé’s comfort with the
“humanitarianism” policy frame could also be attributed to his medical profession. As a
former hand surgeon and Dean of the University of Toulouse medical schoolix he
approaches the topic from a humanitarian and medical perspective,
Mr. Rougé was a doctor, he did research at the university, and he was very active
in humanitarianism…So he's very open minded. According to him, no one should
be excluded for any reason. So you have to find a way to support the dignity of
the person, to have a house, to have a job, to be able to raise a family. So he's very,
very attentive to all that.x
He is also interpreted as “apolitical,” he is placed outside politics but has a
legitimate and credible concern for integration policy. He is “’apolitical…Daniel Rougé
is someone who is not a politician at all… and he is someone who is there because he is a
humanist…He does not have a card in a political party. He was taken for doing good
things and things that made social and health sense.”xi While some say he is “apolitical,”
Mr. Rougé is heavily involved in politics as the deputy mayor in charge of social housing,
solidarity policies and social affairs since 2014. Mr. Rougé has developed ties with the
NGO network of Toulouse. A member of a mutual aid organization and NGO
commented, “he and I are very complementary and I am really in the field, in contact
with people. He is also much more in contact with institutions. But we understand each
other very well. We have the same priorities.”xii Mr. Rougé also works well with national
institutions such as DIHAL, “Daniel Rougé, works very well with certain ministries. For
example, the Interministerial Delegation for Accommodation, DIHAL, Sylvain Matthieu,
worked with Toulouse for a long time.”xiii While Mr. Rougé never explicitly invoked Dr.
Guiraud’s legacy, they both articulated a medical/humanitarian aid policy frame as the
rationale for their policies. Through these policy frames they were able to advocate
for specific policies at the municipal/metropolitan level while remaining “apolitical.”
This evidence does not mean that in order to introduce housing policy, a doctor
with a medical humanitarian perspective is necessary. I interpret this to mean that a city is
more effective in introducing housing integration policy when there is an authoritative
and institutionally established actor working to provide a convincing policy frame
(usually with historical roots relevant to the city) to develop the political will for housing
policy.
By 2015, Toulouse looked beyond the issue of migrant Roma to non-EU migrants
which suggests that Toulouse policy actors began to coalesce around its housing policy
for migrant Roma. A logic of appropriateness became the rationality for housing for
migrant Roma and other “precarious European people living in slums,” signaling that
policy frames for housing for migrant Roma had become normalized. By 2020, Toulouse
City Council had a model of housing for migrant Roma that it wanted to apply to
accommodation for non-EU migrants as well.
The citizens of Romania received full EU labor market access in January 2014,
however there was not a discussion of a marked change in “flows of migrants,” or how a
change in administrative status and access to the labor market might affect city policy, or
at least the experience of migrants. Perhaps this was because 2014 also happened to be
the year that municipal elections occurred across France. At the end of 2013, Mayor
Cohen described Toulouse’s legal responsibility to integrate migrant Roma as, “a
situation where republican law has motivated us and where we have considered that our
social responsibility is to provide solutions.” Drawing on policy frames (which he refers
to as logics) of “responsibility,” “reciprocity” “respect,” “humanitarianism,” and
“health concerns,” he requested financial support for the Urban and Social Project
Management (MOUS), which introduced long-term housing in comparison to the
shortterm integration village.
The reception on these grounds is contractualized in a logic of responsibility,
reciprocity, and respect for a certain number of principles such as the schooling of
children. Support for families has been set up in a humanitarian and health
framework. Despite this initial work undertaken by the City, illegal occupations
and the phenomenon of over-occupation in certain degraded condominiums
continue to exist. In order to optimize the possibilities of integrating these
families, the City must seek the means of the State within the framework of an
Urban and
Social Project Management (MOUS).
These policy changes were too little too late for the Toulouse public and they
ended up electing the center right option, Mr. Jean-Luc Moudenc in the 2014 municipal
elections. This new leadership included policy entrepreneur Mr. Daniel Rougé as the third
deputy mayor of Toulouse in charge of solidarity and inclusion policies in October 2013
(more on the role of policy entrepreneurs in chapter 6). Mayor Moundenc framed his
position on integration relative to the center-left early on in 2015,
…hosting for me is something other than adopting a political posture and making
it an instrument of combat in the pre-election period. Welcoming has another
meaning; welcoming is what we did a few weeks ago, by opening 500 rooms for
the most precarious in Toulouse.
Mayor Moudenc took it a step further and accused the previous administration of
being “cynical,” and “conducting petitions of principle, political performances and
electoral postures.”
In 2015, on the advice of the NGO “Collectif d’Entraide et d’Innovation Sociale”
(CEDIS), the city pivoted from a logic of short-term accommodation to a logic of
longterm housing. The city coordinated with the prefect and 13 social housing landlords
and found 150 low-cost social housing units.218 Eurocities, a transnational NGO, reported
that in 2015, Toulouse had changed its approach from re-housing migrant Roma in
integration villages to re-housing migrant Roma in housing through either the public or
private market.219 Toulouse provisioners shifted from a logic of providing “short-term
housing,” to a focus on long-term housing— “autonomous housing units with social
support in view of assuming the rights and duties of being a tenant able to cover the cost
of rent in the long term.”220
Lille
After the 2014 municipal elections the policy rhetoric regarding housing for
migrant Roma cooled down. There is a sharp decrease in policy frame references after the
peak in 2014.221 In the Lille archives, the figures with the requisite institutional longevity
and coordinative influence to be a policy entrepreneur for housing policy are Mayor
Pierre Mauroy and Mayor Martine Aubry. The evidence in chapter 5 demonstrates that
these Mayors articulated various policy frames, but they were not successful at
advocating for a policy change. Why? Because the network is an important aspect of
advocacy for housing policy and there is not evidence that the Mayors in Lille had the
same advocacy network that Toulouse had. Without a civil society network, it is difficult
218 15 cities learn from Toulouse how to support Roma to access adequate housing,” 2019, Eurocities,
http://www.eurocities.eu/eurocities/news/15-cities-learn-from-Toulouse-how-to-support-Roma-to- access-
adequate-housing-WSPO-BBDLD3.
219 15 cities learn from Toulouse how to support Roma to access adequate housing,” 2019, Eurocities,
http://www.eurocities.eu/eurocities/news/15-cities-learn-from-Toulouse-how-to-support-Roma-to- access-
adequate-housing-WSPO-BBDLD3.
220 Toulouse Municipal Council, “Dispositif d'hébergement accompagné pour les personnes en
situation de précarité : modification et harmonisation du cadre normatif, » pg 1, 22 October 2021,
Délibération n 1.4, Toulouse Online Municipal Archives. Toulouse Municipal Council, « Terrains et sites
d’accueil gères par la ville pour les personnes en situation de précarité : modification et harmonisation du
cadre normatif, » pg 1,2, 26 October 2015, Déliberation n 4.2, Toulouse Online Municipal Archives.
221 From the height of 156 references in 2014 to 34 references in 2015, the next peak was 166
references in 2018.
to generate information-sharing and trust between migrants, NGO workers and city
council that is necessary for housing policy.
On the other hand, in the aftermath of Sarkozy’s speech, Lille had a strong policy
entrepreneur who was against housing policy. Mr. Marc-Philippe Daubresse served as the
Mayor of Lambersart from 1988 to 2017. He also served as the Vice-President of the
Urban Community of Lille Metropole between 2001 and 2008. At the national level he
had served as the Deputy Minister for Housing and Cities and the Secretary of State for
Housing under Jacques Chirac between 2004 and 2005. He has been a Senator since
2017. He was responsible for articulating policy frames that diminished the political will
for housing integration for migrant Roma.
Alternatively, one expert in Lille, located in civil society, has worked at the
Sauvegarde de Nord (an NGO) for 16 years, but he is nowhere in the archive or in the
newspapers. He provides necessary support and expertise but there is no evidence that he
articulates convincing public policy frames to persuade political actors to adopt his
policies. He would not be considered a policy entrepreneur by this study’s standards.
It is unlikely that the corpus would overlook a relevant policy entrepreneur. The
corpus consists of municipal and metropolitan archives (which include municipal and
metropolitan policy announcements, police reports, letters and emails between
departments and between citizens and the city, local and national newspaper articles, and
surveys and reports produced by local NGOs), coupled with the online municipal and
metropolitan archives as well as the interviews conducted with relevant NGOs and
housing and solidarity deputy mayors. Part of the qualification of being a policy
entrepreneur is the ability to articulate a policy frame over time, across institutions; if this
influence is not picked up by the archives, meeting minutes and interviews, then it is
unlikely this overlooked actor is a policy entrepreneur.
Fourth Housing Policy Era (2017-2020): Policy Continuity, National Housing First
Policy is introduced.
Toulouse
By 2018, Toulouse was moving on from integration villages and toward providing
housing for other precarious groups such as non-EU migrants and refugees. By 2018,
when the NGO SOLIHA asked for Toulouse City Council to provide a subsidy for them,
the city council did not mention integration villages as a service that they supported. In
one city council meeting, Mr. Rougé commented that integrated accommodation is
preferred to closed design systems, alluding to integration villages, “What is interesting
in these questions is that for social inclusion, it is still much better to achieve social
diversity in integrated accommodation than to do so with a much more closed design of
the system...” By 2018 there was only 1 integration village left in Toulouse.
Furthermore, by 2018, the city council discussed the issue of EU migrant Roma
less and moved on to extra-Communautaire (non-EU citizens) populations. This signaled
a normalization of integration policy for EU migrant Roma. In addition, the city
acknowledged that while they could offer “precarious European people living in slums,”
housing options, they were not legally allowed to offer extra-Communautaire peoples the
same integration options, “(t)his action falls within a field of competence quite distinct
from that of the Town Hall, whose proactive approach consists of offering precarious
European people living in slums accommodation for the purpose of integration.”
However, Mr. Rougé goes on to explain how despite their status, the city will attempt to
apply the “great methodology” they used to house migrant Roma to other groups.
Mr. Rougé mentions the “great methodology,”—the Toulouse model that Mr.
Rougé refers to here—which resurfaces in 2019. The policy frames “mediation,”
“pedagogy,” “diversity,” “humanitarianism,” “critical policymaking” and “social
support,” are all laid out here in this one paragraph articulated by Mr. Rougé. This is a
response to the center-left asking how Mr. Rougé’s administration’s policies differed
from theirs in the past,
Here are the change points. Regarding the measures you had taken, we have
modified the monitoring of squats, which we started as you had done, and evolved
with the prefecture…This is, in my opinion, great progress and I can provide you
with the facts: since we took on this issue, we have evicted 105 squats in the city
of Toulouse. That doesn't mean there aren't any left, of course. On the other
question of the method, where it is different, it is that, concerning precarious
European nationals, we have used methods of temporary emergency
accommodation, with social support to bring about the integration of people who
go to social housing. But this accommodation is integrated, it is not a camp or an
integration village, like that of Grynfogel, even if the results of Grynfogel are very
interesting. The great methodology is the mobilization of assets of the City and of
the Metropolis, but also, and the efforts of the City also for this purpose, and also
the great mobilization of social landlords in this question. An enormous amount
of mobilization of social landlords on this issue, who have, by the method, been
secured in the use of their property, for which neighborhood issues do not cause
major problems.
Mr. Rougé is a policy entrepreneur because he effectively uses policy frames to
provide context and build support for his policy decisions. The integration policies for
migrant Roma are referred to as a “model,” for the integration of other groups. Mr. Rougé
explained to a local newspaper that “after four years of experimentation, I am convinced
that we cannot lift someone out of great poverty if we do not give them a real home and
basic security, without which an integration project is doomed to failure or will be
difficult.”
In 2019, the center-right government still referenced its “accommodation project”
for “precarious European nationals,” that was founded by the “principle of inclusion
when people have the right to stay.” Mr. Daniel Rougé identified that one of the most
valuable aspects of the accommodation project was the “interlayer culture,” the NGO
network, civil society, or the group of organizations that mediate the relationship between
residents and their local government.And finally, he re-emphasized Toulouse’s reliance
on the integration model which was developed through the accommodation of precarious
European nationals, which includes EU migrant Roma, “We are also going to do the same
thing, in the same proportions, for refugees, according to the model of social inclusion
through work as we have done for precarious European nationals.”
In a 2019 city council meeting Mr. Rougé explicitly referred to Toulouse’s efforts
to house migrant Roma and other European nationals as the “Toulouse method,” “a
model for the reduction of camps, with social inclusion through integrated
accommodation, known as the ‘Toulouse method.’” Toulouse paired up with DIHAL and
“imposed it largely in France for about half of the territory for precarious European
nationals and refugees.” Lastly, Mr. Rougé described the values that drive the model for
housing efforts for precarious European nationals that they had been proposing since
2014: “’laudato si’ means to welcome, protect, promote and integrate.” Here Mr.
Rougé refers to a defining ethos of the Catholic Laudato Si’ Movement. Pope Francis
defined Laudato Si in May 2015 as “care for the natural environment and all people, as
well as broader questions of the relationship between God, humans and the Earth.”
Cecilia Dall’Oglio, associate director of European programs of the Laudato
si’Movement, summarized it’s meaning as “hospitality expressed through a mission of
dialogue, a commitment to build a shelter for others.” Mr. Rougé had the skill to
introduce a new policy frame into the debate, one with historical foundations and
religious significance for the city.
Lille
By 2017, Lille and the rest of France, appeared to move on from a strategy of
integration villages to long-term housing through the national housing first strategy, “in
September 2017, the government launched a National ‘Housing First’ Plan on the
observation that direct access to housing, or staying in housing, with support, are
generally better suited to people and less costly for the community, as emergency
accommodation.”
By 2017, Lille had moved on from the integration policy to the Housing First
policy. However, in Lille the more radical left and right wing parties—such as the Greens
and National Rally party—were more vocal in city council meeting minutes than in
Toulouse, suggesting this policy debate was not as settled. Beginning in 2017 and
continuing through 2020, Lille’s two centrist parties were in policy alignment, while the
more extreme right and left parties took this opportunity to ramp up their ideological
discourse. In 2017, the vitriol of the far-right was still reaching new heights. Opponents
of a long-term housing policy claimed that the “proactive policy” that the Metropolis
implemented in 2008 was a threat to non-Roma residents. They claimed the structure of
the Metropolis itself was to blame because it introduced an unfair power dynamic
between the Metropolis and individual municipalities and that in response the Mayors
needed their own “lever[s] for action.” The National Rally representative argued the
Metropolis threatens the “rights” of its residents and “go[es] against the will of the
metropolitans who suffer” from the “pro-immigration policy of its leaders.”
The municipality is still a scale that allowed citizens to have control over their
direct environment, over their daily life, but what will it be tomorrow? Their
interlocutor, namely the mayor of the town, will no longer be anything more than
the legal representative of a community but without any lever for action. What
better example than the dilemma of rest areas for travelers, and particularly the
care of the Roma population, to demonstrate that this PLU can be set up without
Metropolitan residents? The policy that you conduct in the metropolis for the
reception of Roma populations goes against the will of the metropolitans who
suffer in all its forms from the pro-immigration policy of its leaders.
At this point housing policies had been distilled into pro versus anti-immigration policies.
This representative went on to articulate policy frames such as “meritocracy,” or “duty,”
and to question whether the EU citizenship of migrants even matters,
They are European, so they have rights on our territory? Have we ever spoken
once of the duties that this kind of privilege should automatically oppose? Never!
On the other hand, we oblige the French vis-à-vis them and impose these
populations while the vast majority of the French would like us to take care of the
distress of our own people above all.
In 2020, there was a “green wave,” across France as the Greens (Ecologiste) party
won more representation. The ecologists asked the center-left party to take more action to
find housing for migrant Roma before they were evicted from informal camps,
Allow me to come back to the eviction of the Roma camp last week on Napoleon
road in Hellemmes at the request of the Metropolitan government. Let's start
perhaps by no longer reproducing this kind of episode in which the housing
pathways were not prepared in advance with the people concerned. Let us respect
the dignity of these people as much as possible.
By 2020, the city mentioned integration villages only alongside other
“mechanisms for access to housing,”
…mobilizing its sovereign functions alongside social policies of integration and
access to housing. It will involve both the activation of mechanisms for access to
housing (e.g. housing first) but also the construction of temporary reception
structures such as SAS or integration villages.
The emergence of more ideologically extreme parties such as the Greens and the
National Rally started to introduce more polarizing policy frames and debates regarding
integration, such as the issue of “municipal autonomy” discussed above by the National
Rally representative. Unlike in Toulouse, the debate over housing for migrant Roma was
never normalized. Instead in Lille the debate became further polarized.
Conclusion
The policy entrepreneur, Dr. Guiraud was heavily involved in Toulouse in the
1990s and articulated specific policy frames for Roma which developed Toulouse’s civic
legacy for housing policy. In Lille, 2007 was the beginning of policy frame development
for migrant Roma with the introduction of integration village. By 2014, Toulouse had a
prominent policy entrepreneur in Mr. Rougé who articulated policy frames which helped
develop political will for long-term housing policy. Lille did not have a similar policy
entrepreneur in support of long-term housing policy for migrant Roma in 2014.
Toulouse’s civic legacy was re-activated in 2015 when a policy entrepreneur (Mr. Daniel
Rougé) used historical policy frames (such as a humanitarian frame) to redefine the issue
of housing and generated the political will to introduce new housing policy. Meanwhile,
while Lille had developed an important network and policy frames by 2015, they were
newer than Toulouse’s, and Lille did not have necessary contemporary policy
entrepreneur to make the pivot in 2014, instead national policy drove policy change in
Lille in 2017.
In Toulouse there was a history of individuals who directed the NGOs which
established the trust, mediation and expertise that provided accommodation options. One
individual, Dr. Guiraud ran a pivotal organization for nearly 30 years between 19711998.
The 2015 policy pivot from short to long-term housing policy was spearheaded by a
policy entrepreneur, Mr. Daniel Rougé. Lille, on the other hand, did not have a historical
policy entrepreneur with similar institutional longevity, or a policy entrepreneur to
change policy in 2015.
In Lille, while the Mayors Pierre Mauroy and Martine Aubry were vocal mayors
who could be considered policy entrepreneurs, neither launched persuasive policy
arguments to convince political opposition, other mayors and NGOs. Furthermore, policy
entrepreneur, Mr. Marc-Philippe Daubresse, was heavily involved in multiple Lille
political institutions as well as national institutions and was not interested in developing
political will for EU migrant Roma housing. This evidence supports the argument that in
Lille, with a delayed civic legacy and without a policy entrepreneur to develop the
political will for housing, there was no independent policy change away from integration
villages toward integrated long-term housing with social support. Therefore, Toulouse
was able to independently lead policy change toward long-term housing while Lille
followed national policy later on. In this way, the strong civic legacy of NGOs, policy
entrepreneurs and policy frames influenced housing policy for migrant Roma in 2015 in
Toulouse.
Chapter 7. Conclusion
With this dissertation, I make several contributions. Firstly, a contribution to
knowledge regarding the importance and influence of civic legacies. An inquiry into civic
legacies captures more than a focus on policy feedback loops or path dependence
provides. An inquiry into civic legacies is an inquiry into the potential for policy change
and transformation, which is different from the process of continuity that path
dependency suggests. The theoretical motivation is to understand how civic legacies—
including policy frames, networks and entrepreneurs—could shape contemporary
integration policies of a city and sets precedent for some cities to lead while others follow
in their housing policy for stigmatized groups. This dissertation suggests that the whole
of a civic legacy is greater than the sum of its parts, in this case policies, policy frames
and NGO mobilization. Furthermore, the civic legacy of a city can influence policy in a
way that a national-level legacy might not.
With this dissertation, I also demonstrate how to collect evidence and devise
theory when evidence is not readily available online or otherwise. My first step was to
immerse myself in interpretivist methodologies and explore other ways of collecting
evidence. I searched for cases where evidence on this group was successfully collected
and analyzed. After this, I was able to find evidence for civic legacies in multiple cities
regarding policies for a group that societies have historically undercounted, overlooked
and actively excluded. When I first approached this research topic, the most common
response I received from non-Roma academics was that this would be a difficult topic to
study for myriad reasons. First, the historical racism against this group results in
inaccurate data collection. For example, I knew that most social science scholars would
want statistics on the number of migrant Roma in a city in order to do a direct
comparison. However, comparable numbers are not available. When they are available,
they are tracked only recently, say, within the last 10 years. And finally, the definition of
the group changes so much that often the numbers are not direct apples to apples
comparisons, so to provide numbers would be more of a slight of hand than useful
information. But instead of giving up and changing topic, I used teachings from
interpretivism and tried to find unconventional types of evidence to address my question:
what are the various ways in which cities have provided housing for migrant Roma over
time? So, I put myself on the ground and found the evidence in the archives. What this
evidence consisted of was not statistics on the numbers of families year by year, city by
city, but instead a fractured record of policy-makers and NGOs mentioning the Roma
over time. This dissertation models how to create a cohesive historical narrative of
integration policy for a group, particularly when the historical record of a people seems
sparse. This is one of the greatest contributions of this dissertation; to provide a model of
how to collect and present evidence that others say is not easily collected, analyzed and
conveyed.
Third, my dissertation uses civic legacy theory to emphasize how history bears on
and presents opportunities for housing at the urban level. This topic is not merely a
theoretical inquiry. In recent history, cities have tried to provide adequate housing for
migrant Roma families for at least the past 30 years, in earnest since the COVID-19
pandemic. Beginning with the 2014 “refugee crisis,” the integration of stigmatized
groups has flummoxed city governments and was only exacerbated with the COVID-19
pandemic. But also, since racism is such a strong and obviously present explanation for
policies concerning migrant Roma, it is often the only explanation explored in policy
analysis. This dissertation on the other hand is focused on real opportunities for
integration policy, not only theoretical challenges and “problems.”
Lastly, I only know of one other (very well done) dissertation manuscript in
Political Science that was focused on the Roma and still the case of the Roma serves as
one case of three cases (Buzas 2014). With my focus on service provisioning for this one
group, I have demonstrated that focusing solely on one group does not have to be a
methodological problem but an opportunity to explore how governments represent and
interact with in-group diversity and how multiple city-governments differently engage
with the “same” group. These are the myriad contributions of the work.
An important question that one may be asking is: where are the Roma in this
story? Where is Roma voice? I conducted 11 interviews with migrant Roma families.246
However, from a methodological perspective I was not comfortable including my
interviews because I did not conduct enough interviews to achieve generalizability. On
the other hand, I did end up with enough archival evidence for generalizability, so I
followed the evidence, as I explained above. Additionally, there is already a lot of
important work on housing for Roma that has been authored or published by people with
Roma backgrounds.247 But in none of these will you find an analysis of urban housing
246 While I conducted 11 interviews there were often several family/friend members present at each
interview.
247 Here are some papers published in Roma academic journals that consider preferable or desirable
housing and housing policies for Roma (Cittadini, Critical Romani Studies, 2019) Here are some
publications written by Roma considering integration policy for Roma (Rostas, 2019). The journal Critical
Romani Studies has been publishing important work on Roma-related topics including integration since
2018. Here are some publications written by Roma considering racism toward Roma in Europe and the U.S.
(Kocze 2018; McGarry 2017, Picker 2018, Acton 2016, End 2012, Sergio et al 2017; Matache 2020).
There are also very good publications written by non-Roma that consider Roma integration in France and
Roma Racism in France and the EU (Buzas 2017; Sardelic 2019; Sigona and Vermeersch 2012; Van Baar;
Kostka 2015, Silver 2019, Van Baar 2014, Manca & Vergnano from Magazzini and Piemontese 2019,
policies for migrant Roma over the last 30 years in France. So, while political science as
a discipline may find the subject matter of this dissertation myopic, this dissertation
contains novel insights that concerns the largest minority in Europe.
I would like to suggest that asking the question, “well what type of
accommodation styles do migrant Roma want?” is not as useful as one would hope. Even
though it is well intended, asking about housing preferences for migrant Roma would
unfortunately, not change the reality of housing for migrant Roma. Firstly, because what
migrant Roma wanted/want is heterogenous and would require another dissertation to
adequately explore. Even within cities, families want different things, and within families
individuals prefer different types of housing. In my interviews, I did not get a sense that
the migrants in Lille v. Toulouse were different on average in their preferences. I did get
the sense that no matter the preferences of the families and individuals, the realistic
options for most of these families is to accept whatever help from the NGOs or leave/get
evicted from their encampments by the state. Secondly, the more I spoke with families,
the more I understood that—very simply—migrant Roma families want what all families
want: a stable housing situation, an income, healthcare and oftentimes nearby education
options for their children. Some families want to make money and send it back to their
homes abroad. This is their prerogative. It informs some decision-making of course, but
to focus too much on this question “well, what do migrant Roma want,” risks
essentializing and “othering” migrant Roma preferences and gets in the way of
investigating what cities can do to welcome this group. So, for readers interested in
Kostka 2015, Chapter 3 of Triandafyllidou and Spencer 2020; Chauvin and Garcés-Mascarñas, Vermeersch
2012, Vermeersch 2012, Guy, 2003; Molnar Diop, 2013, Vašečka & Vašečka, 2003, Benedik, 2010, Kóczé
& Trehan, 2009, Kocze 2018, 460).
furthering their understanding of housing integration process and policy for migrant
Roma, written by people with Roma and non-Roma backgrounds, I encourage you to do
the research into the large literature on integration policies for the heterogenous Roma
people.248 Below, I restate the specific arguments and findings laid out by chapters 1-6.
The role of civic legacies is not the only explanation for housing policy for
migrant Roma. The roles of racist attitudes and institutions, educational attainment,
political representation, political culture, and economics do matter for housing policy for
migrant Roma. But the role of civic legacies must be considered because it emphasizes
the historical role that civic society, including NGOs, policy entrepreneurs and their
public policy ideologies, plays in shaping integration policies. The role of civic legacies
contributes more than a historical illustrative account, it is a theory regarding how the
history of civic engagement in a place, in this case a city, shapes the potential for future
integration policies.
Therefore, why did Toulouse emerge as a leader and Lille a follower regarding
housing policy for EU migrant Roma? The evidence suggests that Toulouse developed a
civic legacy in the 1990s which allowed its city council to autonomously pivot away
from one housing policy to another in 2015, while Lille’s civic legacy was developed in
2007
and lacked a policy entrepreneur, thus Lille did not change its policy until there was a
248 Cittadini (2019) published a paper specifically on housing policy and human rights for Roma in the
Critical Romani Studies journal. For work that considers integration policy for Roma please refer to Rostas
(2019). The following are publications written by Roma considering racism toward Roma in Europe and
the U.S. (Kocze 2018; McGarry 2017, Picker 2018, Acton 2016, End 2012, Sergio et al 2017; Matache
2020). There are also very good publications written by non-Roma that consider Roma integration in
France and Roma Racism in France and the EU (Buzas 2017; Sardelic 2019; Sigona and Vermeersch 2012;
Van Baar; Kostka 2015, Silver 2019, Van Baar 2014, Manca & Vergnano from Magazzini and Piemontese
2019, Kostka 2015, Chapter 3 of Triandafyllidou and Spencer 2020; Chauvin and Garcés-Mascarñas,
Vermeersch 2012, Vermeersch 2012, Guy, 2003; Molnar Diop, 2013, Vašečka & Vašečka, 2003, Benedik,
2010, Kóczé & Trehan, 2009, Kocze 2018, 460).
national level change in policy. Therefore, to return to the research question, how do civic
legacies shape contemporary housing policy for EU migrant Roma? This dissertation
finds evidence that civic legacies—networks, policy frames and policy entrepreneurs—
developed in previous integration efforts shape housing policy. Thus, the legacy of local
NGO networks, and specific policy frames used by policy entrepreneurs can result in
independent policy change (in the case of Toulouse) while the absence of a developed
NGO network, policy frames and a policy entrepreneur that can activate this legacy
results in delayed policy change through other avenues such as national level policy (in
the case of Lille). However, civic legacies are a necessary condition, they are not
sufficient for policy change. I find that the role of one policy entrepreneur in
Toulouse was indispensable. Without the influence of the policy entrepreneur, Mr. Daniel
Rougé, it is unclear whether the presence of a civic legacy would be enough to result in
policy change.
This dissertation presents evidence that Toulouse developed networks during the
integration of migrant Roma in the 1990s which set a precedent for its city council to
shape new housing policy in 2015 and empowered it to emerge as a leader in housing
policy for migrant Roma. Chapter four described how the vanguard group in Toulouse,
ATS, had the institutional power to advocate for migrant Roma in 2015 based on its
historical experience with integration in 1990. This role of this network was to generate
the information and trust between the target group, social workers and city council that is
necessary for housing policy. The civic legacy of this network that was established by
ATS in 1990 during re-housing efforts was activated again when new leadership took
control in 2015. This network was able to hold local political actors accountable, give
voice to migrant concerns and advocate for certain policies over others. Whereas Lille
lacked a strong vanguard group like CCPS in Toulouse with an institutional legacy
developed over decades that could solidify the necessary connections, trust, mediation
and expertise between civil society organizations, migrant Roma families and city
council. A network was developed later in 2007 when the civil society network supported
migrant Roma families in their move from encampments to integration villages.
However, this network did not have the historical legacy like the one developed in
Toulouse. Without a robust civic legacy, Lille did not change its policy from integration
villages until there was a national level change in policy in 2017. Instead, policy change
in Lille was lead by national policy and EU funding opportunities.
Then chapter 5 and 6 described the roles of policy frames and policy
entrepreneurs. The historic policy frames developed by NGOs and policy entrepreneurs
during previous integration attempts establishes a repertoire of policy frames which
potentially resonate with a city in future. In Toulouse, policy entrepreneurs emphasized
the humanitarian policy frame which had historical roots, while in Lille the present policy
frames did not persuade multiple stakeholders (predominately municipal mayors, prefects
and political opposition) to change policy. In Toulouse, the policy entrepreneur was able
to coordinate policy change across the metropolis. In this way, the history of cities’
experience with integration—the policy frames, NGO networks and policy entrepreneurs
—does matter in current integration processes.
Practical and Theoretical Implications
The practical policy significance of these findings provides incentives for city
councils to research their own civic legacies regarding integration to achieve a better
understanding of how their past integration policies may influence present and future
integration policies. The theoretical significance of this project demonstrates how
historical civic legacies influence integration policy. In this way this dissertation will
contribute to migrant integration studies most directly; however, this project has
implications for integration policies introduced by cities for marginalized groups more
broadly, in addition to migrant groups, such as formerly incarcerated, autochthonous, and
unhoused individuals.
The policy frames that contextualize and categorize marginalized migrant groups
(which include stereotypes, biases etc.) are well understood within the literature.
However, policy frames (both historic and contemporary) regarding service provisioning
such as housing, healthcare, education and employment opportunities are equally
important for academics to focus on. Without an understanding of how cities frame
housing policy, policy analysts are missing half of the picture when it comes to welfare
policy. This could lead to the incorrect assumption that cities introduce objective policy
options that are not influenced by policy frames. Or that policy frames are generated out
of thin air, instead of through the development of civic legacies. Ultimately, this new
consideration provides city leadership with more agency and responsibility regarding
policy development and implementation, and a greater appreciation for its own civic
legacy.
While this dissertation analyzes housing for EU migrant Roma in specific cities in
democratic societies, this project has portability to other cases. This dissertation
interrogated housing integration policy for EU migrant Roma in select cities in France.
However, non-Roma Europeans harbor some of the highest levels of prejudice and
racism toward Roma. Furthermore, Roma are the largest minority in Europe and live in
every country in Western Europe in addition to Canada and the United States. Therefore,
this is a policy question regarding a group in diaspora across the world who have been
historically discriminated and persecuted. In this way, this specific policy question
applies to many contemporary societies. Migrants are not the only group subject to a
government-supported integration process. Formerly incarcerated, unhoused and
autochthonous individuals are also subject to governance policies that are heavily
influenced by civic legacies.250 In this way, this project has theoretical implications even
beyond the field of migrant integration and housing and applies to service provisioning
during (re)integration processes of marginalized groups generally.
