Social Mix Policies in the French Eco-Districts: Discourses, Policies and Social Impacts

In the 1960s, France built large high-rise developments to house poor and immigrant populations. This policy led to the rise of crime and violent unrest in those developments. Responding to that failure, France has tried, especially since the eighties, to promote a social mix policy in its new housing developments. In the first decade of the twenty first century, France elaborated an eco-district (eco-quartier) program whose guidelines emphasize the goals of this social mix policy together with affordability in public social housing. In light of these developments, this paper focuses on the socio-economic aspects of French eco-districts, especially with respect to low-income populations. The eco-quartier housing distribution has shown that social mix goals are barely reached. In affluent cities, where property prices are high (such as Paris, its middle-class suburbs and some large cities), the municipalities build eco-quartiers in substandard neighborhoods, to attract middle class families. In average cities, some municipalities have implemented more social housing than planned, to provide developers with access to State subsidies and loans – but can still privilege the middle-class in the allocation of the resulting housing. In the poorest French towns, eco-quartiers can improve living conditions for local residents but do not effectively promote social mixing.


Introduction
Large housing projects were built in France for the working class from 1956 on (mostly in the 1960s and 70s), to deal with the serious housing crisis that the country was facing after the Second World War (Blanc, 2004). The construction of three million dwelling units in large high rises within ten years was expected to solve the crisis. In the 1960s, these French suburban social housing estates accommodated their first residents. In the seventies the estates turned into vertical slums. Since the late 1970s, French middle-class families have been deserting them, usually to move into a single-family house in a more distant suburb. Those who cannot afford to move out are mainly second-generation black and Arab immigrants from French ex-colonies. The vertical slums are characterized by high rates of poverty and unemployment (Lapeyronnie, 2008, p. 416).
Whereas discussions about ethnic and class segregation have a long history in America, France has tried to overlook or deny its own segregation practices (Massey, 1981), as they expose its failure to live up to the republican ideals of Equality and Fraternity. Severe riots in the fall of 2005 dramatized ethnic segregation, stunning public opinion.
In addition to building's design and construction (which directly affect the use of energy, water, and materials), green building takes into account other environmental issues, ranging from ongoing building operation to urban planning aimed at reducing the reliance on private cars. The introduction of neighborhood-level green building standards, has further extended the scope of sustainability topics, as well as the range of stakeholders, including municipal authorities, whose role is likely to expand in the future. While national policies strongly affect local energy consumption patterns, as well as the generation and disposal of pollutants, city planning and policies can help reduce GHG emissions, promoting sustainable development (Bulkeley et al., 2011).
In 2008, the European Commission launched the Covenant of Mayors (Note 2). Signatories agreed to meet and exceed the EU target of a 20 percent reduction in GHG emissions (from 1990 levels) by 2020. With more than 5000 signatories, the Covenant is the leading movement in Europe supporting local and regional authorities in achieving this goal. Some European cities started focusing on climate change well before signing the Covenant, making significant inroads in GHG reduction: the achievements of Stockholm, Malmö, Freiburg, Copenhagen, and others have been widely extolled (Note 3). One practice these cities have in common is the establishment of eco-districts-neighborhood scale developments addressing climate mitigation, energy savings, and the management of waste, transport and water resources. The establishment of such eco-districts may raise sustainable development standards throughout a city. There is a growing literature on eco-districts, but little systematic analysis of their effectiveness in attracting affluent populations or enhancing affordability. In 2008, the French Ministry of Ecology, Sustainable Development and Energy launched a national program of Eco-Districts (Éco-quartiers) as part of the ministerial plan -Sustainable City‖ (Plan Ville Durable). According to the eco-quartier framework requirements (2009), there should be a socio-economic, generational and cultural population mix within the neighborhoodin sharp contrast to the large post-war housing projects built exclusively for the poor (Blanc, 2004). However, the 186 certified eco-quartiers (2017) represent only 55,000 housing units (French Ministry of Housing, 2018)less than 2% of the three million public housing units built in the 1950s and 60s.
Social mix in the eco-quartiers is meant to be ensured through housing diversity, i.e. in price, dwelling size, typology (individual vs. collective), social/private housing or modality of access (purchase or rental). The actualization of such a housing mix requires, however, exceptional funding methods to control home prices. Indeed, the construction costs of high ecological performance apartments have been estimated to be10 to 20% higher than those of regular apartments (Guinand, 2007) (Note 4). Souami confirms that such funding is often necessary to make these neighborhoods affordable to all (2009, p. 80). According to his analysis, for the winners of the Eco-quartier call for proposals in 2009, public subsidies complementing ordinary funding represent from 10 to 95% of the gains (22% on average). These benefits are funded by regional and general councils (for example, through specific EcoFaur² schemes for the Brittany region), Land companies, ADEME, or the European Commission (calls for projects make the funding possible for exemplary projects, as was the case for the De Bonne eco-quartier in Grenoble, with Concerto subsidies (Note 5)). The European Commission allocated 3.9 million euros to the Concerto program for energy saving measures.
Projects for -sustainable districts‖ can also help implement some requirements, through special subsidies. For example, the call for projects of New Urban Neighborhoods promises subsidies of more than 3000 euros per eer.ccsenet.org Energy and Environment Research Vol. 10,No.

The French Eco-District Scheme
While the French ‗green' building label (BBC) only sets a standard for energy consumption, the eco-quartier scheme involves social issues as well. According to program requirements, the local authority must ascertain -that housing location and prices are appropriate for household income‖ Certification file (2015), Second dimension (Living environment and use)-Commitment 7 (Social Diversity) (Note 6) (see the Eco-quartier guidelines in annex).
The initiators of the eco-quartiers emphasized that the inclusion of planning measures promoting social mix, is a decisive criterion in assessing project applications. According to the Head of the Sustainable Planning Unit of the Ministry of Housing, Equality of Territories and Rurality (2015): -The social aspect is the French peculiarity; a project without social mix, not taking social issues into account, cannot be an eco-quartier. In any case, it will not be an eco-quartier as perceived by the Ministry [...] and it is incompatible with how we consider sustainable development in France! [...] Our aim is to promote the idea of social mix. We want to give all people the opportunity to live in an eco-quartier. The aim is not to sell the apartments at a higher price‖. De Chastenet et al. (2016) argue that -we are far from the edge cities, which are fully funded by private groups, aiming to produce luxury neighborhoods for urban gentrification‖.
Jean-Louis Borloo, Minister of Ecology, Energy, Sustainable Development and City and Country Planning in 2009, insisted that social issues be taken into account in the design of eco-quartiers: -Eco-quartiers and eco-cities also have a social and economic role to play [...]: they must make a strong contribution to developing social cohesion and reducing territorial inequalities.‖ (Note 7).
In general, the eco-quartier call for proposals seeks above all to avoid social segregation. The social dimension remains restricted to the search for population diversity, both -to reduce socio-spatial segregation‖ within the neighborhood and -to improve social cohesion‖ by -strengthening social ties‖ (MoESDTH (Note 8), 2009).
The definition given by the Regional Agency for Environment and Energy of Ile de France (ARENE), explains that social effects are a central objective: -Sustainable neighborhood development takes into account all effects, at the environmental level as well as at the economic and social levels, during both the preparation of the project and its operational phase‖ (Note 9). This distinguishes the eco-quartiers from ‗sustainable' districts in Europe, criticized for having become -ghettos for bobos (Bohemian Bourgeois (Note 10))‖ (Emelianoff, 2011, p. 3), and claiming a -French approach‖ to generating sustainable neighborhoods: without gentrification. -The eco-quartier is not the refuge of a few, like the American condominiums. It must be open (accessible to all). The eco-quartier therefore aims at several kinds of mixing: social (by socio-professional category and activity), generational (by age), and by family structure (nature and size of the household).‖ (Charlot-Valdieu & Outrequin, 2009, p. 60). For example, the -eco-quartier‖ charter of Lille (a city at the northern tip of France, in French Flanders) emphasizes social aspects, most of all social mix: -One often hears that an eco-quartier is reserved for the ‗bobos'. On the contrary, a fundamental objective of the eco-quartier is promoting social mix, by offering a variety of housing types capable of meeting the needs of all: large and small dwellings, housing for the elderly and young people as well as families, and private dwellings, together with social housing. In addition, an eco-quartier should be a well-connected neighborhood, close to everything, with more nature and less pollution ... in a word: a more pleasant neighborhood! Who wants to live in a neighborhood like this? Everybody!‖ (Note 11).
This consideration of social aspects was, however, seen as complementary to environmental objectives: the main focus was on the technical ‗green' aspects, to ensure the projects' ecological performance. The awarding of credits in the Lille eco-quartier call for projects indicates a priority for balance: the thematic ranking of projects eer.ccsenet.org Energy and Environment Research Vol. 10, No. 1;(of Eco-quartier 2009 includes categories related to water, urban biodiversity/nature in the city, waste, energy savings, mobility, and ‗green' building. The only category pertaining to the projects' social dimension is -density and urban form‖. The initial approach developed during the first Eco-quartier call, emphasizes the need to fight -social risks‖ in these neighborhoods, especially segregation. Thus, while the main thrust was to promote projects whose approach was ‗green' in environmental terms, the inclusion of social responses was to prevent negative side-effects, such as socio-spatial segregation. French local authorities have complex set of regulations, each of which has some influence on the planning and implementation of Eco-quartiers, mainly thermal regulations and the law on Solidarity and Urban Renewal (SRU, 2000). This web of actors, and their interactions determine green building funding in Eco-quartiers. The Eco-quartiers have provided affordable housing thanks to subsidies at the European, State, regional, departmental and municipal levels, as well as loans from the CDC (-Caisse des Dé pôts et Consignations -Deposits and Consignments Fund‖), a public financial institution established in 1916, under the control of the Parliament. Caisse des Dé pôts has supported 15 eco-quartier projects, to strengthen their coherence throughout their progress rather than simply labeling them. In 2008, the Caisse des Dé pôts group also launched a research and development strategy on the sustainable city, together with its real estate subsidiaries. The method of the Caisse des Dé pôts was to do follow-ups on eco-quartier projects, selected according to their ambition and to the pragmatism they display in the pursuit of sustainable development.
Other public funds (national or European) support the thermal refurbishment of social housing. For example, the European fund -Concerto‖ has assisted the Eco-quartiers of Grenoble and Reims. The State Agency for the Improvement of Housing (ANAH) allocates subsidies for the rehabilitation of buildings older than 15 years. The goal is helping privately owned buildings which are owner-occupied or rented as primary residences. The target group of owner-occupiers is limited to those whose economic resources are beyond a certain limit. Buildings and housing units eligible for subsidies are those with insufficiencies regarding safety, impacts on health, appliances and housing quality (ANAH, 2006;ADEME, 2006a).
An older subsidy program called PALULOS was launched in 1977 to address the lack of maintenance in rental social housing owned or managed by social entities. As with ANAH, eligible projects include improvements in safety and comfort, upgrades for compliance with regulations and housing norms, and energy savings (Note 12). Buildings must be at least 15 years old, but for energy conservation improvements they must hail back to July 1981 or earlier. The subsidies are attributed as a percentage of the renovation costs, up to a maximum amount, usually 13 000 Euros per housing unit (Note 13).
The Ministry decided in 2014 to grant a 30,000 Euro subsidy to each local authority that obtained the Eco-quartier certification by implementing a monitoring and evaluation method. Thanks to this funding system, even poor municipalities can design and monitor an eco-quartier including ‗green' and affordable housing.

Methodology
Case studies are appropriate when the researcher has little control over events and wants to find out how a phenomenon has happened, or why (Yin, 2009). The case study method is used in ethnography, surveys, quantitative modelling etc. (Yin, 2011), and values the use of multiple sources of evidence, to foster reliability and improve understanding. Case studies can be used to test theories, or deepen our understanding (de Vaus, 2001), as we have attempted to do in the present study. We have selected three eco-quartiers for intensive data collection: • An eco-quartier in Paris (Claude Bernard Eco-quartier) that received the national label and where the apartment buildings are certified ‗green' (BBC-effinergie certification) • The -Clause Bois Badeau‖ Eco-quartier in Bretigny Sur Orge, that also includes BBC-effinergie housing.
• The -Pays de France‖ eco-quartier in Reims, where the housing also received the BBC-effinergie label. We selected three (though not geographically) representative case studies according to the municipality's socio-economics data as detailed in the following table (we did not include villages and the overseas cities (Note 14)).  France is one of the more affluent countries in the OECD, but exhibits significant poverty and meaningful inequalities of income, wealth, and life outcomes. The past several years of rising unemployment have worsened these circumstances. This variation often occurs across the lines of ethnicity and national origins; immigrant communities in France tend to have significantly higher levels of poverty, unemployment, and health disparities, often concentrated in specific suburbs of major cities. To be considered to be living in poverty an individual must survive on €977 per month, after taxes and including government subsidies. That sum is 60 percent of the national average income (Maurin, 2018).
Our three eco-quartier case studies are representative of three of the detailed categories of municipalities detailed above: (1): The Parisian metropolis with a high average income (though a few poor neighborhoods remain, mostly in eastern Paris, 18, 19, and 20 th district); (2) Parisian remote suburbs like Bretigny Sur Orge with high share of low-income social housing; (3) Municipalities like Reims with high rates of poverty and large income disparities between poor and rich neighborhoods.
The logic of case study selection was to analyze how the green building concept was developed, considering the municipality's location (core-periphery), housing policy and socio-economic status. We considered the major city (Note 16) (Paris), one of its remote suburbs (Bretigny Sur Orge), and a low-income French eastern peripheral city (Reims). The aim was to find out whether the location and the socio-economic status of the municipality affected the input of ‗green' and affordable housing.
Fieldwork included interviews and participant observation in key professional events, as well as analysis of documents from a variety of sources (policy-making bodies, NGOs, interest groups, experts, the media).
The analysis aimed to (1) identify the key policy tools designed to promote green affordable housing; (2) gather evaluations of their application through analysis of documents, including policy reports, central and local government action plans, professional NGO reports, statements of opinion, newspaper articles, protocols of municipality meetings, and academic research reports.
Interviews were conducted with key figures in green building, selected by the positions they hold in the field (Scott, 1991 During our second fieldwork period in France, in April 2016, interviews were conducted with stakeholder representatives in the local authorities corresponding to the three case studies (in Paris, Bretigny sur Orge and Reims), and a second interview was conducted with the head of the Eco-quartier program at the Ministry of Housing, aiming to understand the evolution of the label's requirements. An internet survey was carried out in 2016 among municipalities including at least one eco-quartier. We aimed to collect data on housing distribution (private apartments/social housing for low-income/middle-class/upper middle-class) in all certified French eco-districts.

Results
The Major City: Paris The map in Figure 2 represents eco-quartier distribution among the twenty administrative districts (arrondissements) of Paris: Boucicaut (15th), Gare de Rungis (13th), Clichy-Batignolles (17th), Claude Bernard (19th), Frequel-Fontarabie (20th), and Pajol (18th). So far, they include only 0.6% of the Paris population (Paris municipality, 2015) and half of their housing units are social housing. Thus, in Claude Bernard eco-quartier, half of the dwellings are public housing, 25% private rental housing (controlled rent) and 25% private apartments of various sizes and kind (such as a home for elderly, student dormitory…) to ensure social mix. It was mentioned in the eco-quartier guidelines that public housing for the upper middle-class should compensate for the high share of low-income households in the neighborhood (see the socio-economic classification in France, Annex Table 2). In 2009, the private apartments were sold for €6,250/sq.m. (according to the developer's representative), well above the 19 th arrondissement average of €4,900/sq.m., but slightly below the new apartment average price at the time (€ 6,500/sq.m.). According to the neighborhood developer representative (personal communication, 2016), the promoter had to decrease apartment prices to sell them, because -no one would have bought an apartment in that neighborhoodnext to the highway, without public transportation, and with so many prostitutes and drug dealers‖.  Figure 3 shows that the share of social housing units for low income families is minimal in all eco-quartiers: 5% in Claude Bernard, 10% in Clichy-Batignolles, 3% in Gare de Rungis and 5% in Frequel Fontarabie. According to Machline et al. (2016), all eco-quartiers are located in areas that already include a high number of social housing units (90% or more such units). Moreover, no eco-quartier was planned in areas lacking social housing (Machline et al., 2016), to increase its supply.
According to Machline et al. (2016), four out of six eco-quartiers were planned in the lowest socio-economic status block areas and arrondissements (Note 17) (Category 3 -13th and 18th-20th). According to a Paris Municipality Planning Unit representative (personal communication, 13 April 2016), the eco-quartiers aimed to increase the share of private dwellings and upper middle-class social housing, to attract higher income residents.
The Claude Bernard eco-quartier included 50% private dwellings and 25% middle-class social housing, and was thus designed primarily for middle-class residents. In fact, the neighborhood's share of low-income social housing is only 5%, whereas in the statistical block areas surrounding it the share exceeds 90%. The 19th arrondissement as a whole includes 35% social housing while the average annual household income is less than €30,000very low compared to the Parisian average (Machline et al., 2016). Thus, at least 75% of the housing units in the eco-quartier will be unaffordable to the local (19 th arrondissement) population. According to the social housing company representative (personal communication, 2015) -the shares of social housing categories were decided upon by the municipality‖.
eer.ccsenet.org Dwellings in Claude Bernard, public or private fail to provide ‗affordable housing prices' (as stipulated in the certification file) to pre-existing residents of the 19th arrondissement. According to the Paris Municipality Planning Unit and housing authority (2015), the aim was to attract a wealthiest population: -The aim is to balance the socio-economic level of the population‖ (Head of the Housing Authority of Paris, 2015). According to a representative of the developer (2015), veteran residents were strongly encouraged to move out.
The private dwellings in Claude Bernard were sold at 4% less than the average 19th arrondissement prices for new housing. Moreover, according to the manager of the planning company (2015): -At the beginning it was not easy to sell the apartments. The first home buyers were very brave to move into a neighborhood where they were exposed to prostitution and drugs on a daily basis!‖ The situation is similar for the Gare de Rungis eco-quartier (13 th arrondissement), where the share of social housing is 65% but the upper middle-class public housing units constitute 63%, while the low-income units are only 4.4% of the social housing stock (Figure 3). In the Frequel-Fontarabie eco-quartier, which included 27 of the 1030 Parisian substandard buildings (Observatory for Building Degradation Prevention in Paris, 2012), 95% of the social housing is meant for the middle-class (Figure 3). However, before the establishment of the eco-quartier, all of the housing stock was for low income people (Paris Housing Authority, 2014). Thus, 38% of the social housing built in Paris since 2001 was intended for the upper-middle-class, which did not match the demand: among the 100,000 applicants for social housing in Paris, 75% earned below the low-income housing limit, while only 4% could apply for upper middle-class housing (APUR, 2007). Nowadays, low-income public dwellings represent only 15% of the social housing built in Paris (Clerval, 2010). At the city scale, between 2001 and 2013, the newly built social housing consisted of 25% for low income and 37% for middle income, while the share for the upper-middle-class was even slightly larger, at 38%. This city scale ratio is substantially lower than that of the eco-quartiers (Gare de Rungis: 63%, Claude Bernard: 50%)and to the extent that such social housing promotes gentrification in the poorest districts of Paris, the eco-quartiers themselves foster eer.ccsenet.org Energy and Environment Research Vol. 10, No. 1; 2020 ‗eco-gentrification', at the expense of the poor. While working-class households face growing difficulties in finding public or private dwellings in Paris, public policies have only improved housing conditions for those who can afford to stay in the city or move into it (Clerval & Fleury, 2012).

A remote suburb: Bretigny Sur Orge
The Clause Bois Badeau eco-quartier is located in the small municipality of Bré tigny sur Orge, 35 kilometers south of Paris, in the department of Essonne, in the Ile De France region (Note 18), on line C of the RER.
In the second half of the 20th century, Bré tigny-sur-Orge, like most municipalities in the Paris region, experienced rapid population growth, going from 3,673 residents in 1946 to more than 12,000 in 1968 and 21,650 in 1999. According to the development company's project manager, the reason for building 46% social housingrather than 30% as originally plannedwas the 2008 real estate crisis. Some promoters had to cancel their participation, and thus, to implement the project, the socialist-led municipality decided to plan more social dwellings, as social housing companies are funded by the State. The project, which was to start with private apartments, began with social housing units. However, according to the Head of the Planning Unit of Bretigny sur Orge, the new neighborhood's start with social housing gave it a bad image in the town.
Bretigny is mainly composed of low-density neighborhoods; the population was scared that Clause Bois Badeau would become a -dangerous vertical slum‖, like others in the Essonne department, for example in Corbeil-Essonnes, Evry and Grigny. There was a feeling that the entire municipal budget would be spent on this new neighborhood. The loss of the main green and open spaces of Bretigny was also hard to accept. According to the Development Company Manager: -The owner of the private rental buildings encountered difficulties in finding rentersfor instance, the cheapest four room unit (100 sq.m. on average) costs 2,400 euro/month and should not be above 1,800 to be affordable for a family.‖ Apartments in the new neighborhood (about 3,500 euro/sq.m.) are cheaper than those around it (3,800 euro/ sq.m.). Thus, the building of a ‗green' neighborhood in Bretigny Sur Orge had no gentrifying effect. The newly elected right-wing municipal team (in 2014) canceled a social housing operation before its implementation and, since the election, only private dwellings have been designed. The aim is to attract a more affluent population and even plan single-family house projects (Development company project manager, personal communication 21/04/2016).
According to the development company's project manager, most home buyers are young families from Bretigny or surrounding cities, purchasing their first apartment. The municipality wanted at least 60% of the homebuyers to be residents (to limit the share of outside investors). Thus, the real estate companies selling the apartments were asked to prevent that share from rising above 40% (Head of the Urbanism Unit, Municipality of Bretigny, personal communication 18/04/2016).
Within the eco-quartier, a central urban heating network was installed, consuming wood pellets (considered cleaner than other heating fuels). However, according to the Head of the Urbanism Unit, the residents complain of thermal discomfort and construction defects. Most buildings fail to attain the planned energy efficiency, and even if overall energy consumption is lower than for conventional buildings, the residents' monthly bills are not lower, as the maintenance cost of the ‗greener' heating system is higher than that of conventional heating based on natural gas. To avoid construction defects, the municipality has tried to ensure that the promoters improve their inspection skills and the Bretigny Mayor makes it a point to be present for each building inauguration and inform the promoters that -if there are construction defects, they will hear from him‖ (Head of the Urbanism unit, municipality of Bretigny, personal communication 18/04/2016).

A remote periphery: Reims
Reims, in the Marne department of the Grand Est region of France (previously in the Champagne-Ardenne region), lies 129 km east-northeast of Paris. The 2013 census recorded 182,592 dwellers in the city proper and 317,611 inhabitants in its metropolitan area. Historically, the Notre Dame de Reims cathedral has been the place of coronation of French kings. The bubbling Champagne, produced in the Reims region since the end of the 17th century, has become one of its economic assets. The city of Reims has undergone rapid economic growth eer.ccsenet.org Energy and Environment Research Vol. 10, No. 1;between 1960and 1975 to the establishment of new industries (some of them owned by IKEA). The period was marked by notable spatial expansion and a 54% demographic growth rate (from 131,000 in 1954 to 200,000 in 1975). Like most French cities, Reims subsequently experienced both economic and demographic stagnation.
The arrival of the TGV high-speed train in 2007 provided a new boost to growth, putting Reims at just 45 minutes from Paris, 35 minutes from Marne-la-Vallé e (Note 20) and 30 minutes from the Roissy-Charles de Gaulle International Airport. Reims would like to host Parisian companies wishing to relocate to lower operating costs. A new tramway line was built in 2011, running through the Reims city center and serving its two most important neighborhoods, the Orgeval district in the north and the Red Cross district in the south. Notwithstanding these growth enhancing steps, Reims still belongs to the poorest third of French municipalities.
The construction of large housing projects began in the 1960s. One of the most important HLM neighborhoods is the Croix Rouge district (divided into three sectors: Nord housing units were built and remained empty, as their rent is even higher than in the private sector. Furthermore, the image of the neighborhood detracted from the apartments' attractiveness. -We made upper middle-class apartments, but it was no success. Still, we believe that the neighborhood's great location will attract new populations. We are optimistic. At present, there is no population renewal. Until we renovate the entire district (Croix Rouge), its bad reputation will linger on. It will probably take an entire generation to change the image. Even if the eco-quartier buildings are high standard, they are surrounded by very low-quality buildings. The social mix aspect was one of the weaknesses of our application. We could not satisfy all requirements, given that we first planned the projects and only then applied for the label.‖ (Head of Reims City and Housing Policy Department, personal communication, 2016). All buildings received the BBC certification. However, to pay for the refurbishment, the social housing company raised rents by about 50 euro/month. According to the Head of the Social and Urban Development Department of the Foyer Remois social housing company, -thanks to the BBC certification requirements, the final energy consumption decreased by 30-50%. Thus, even if the rent increased, the residents pay little more than before, 5 euros more on average). According to the Head of Reims City and Housing Policy Department, when the rent increase is too high (as it has been for a few families in Pays de France), the municipality has subsidized it during the first four years (20 Euros/month during the first year, then dropping to 15, 10 and 5). In Reims, the eco-district had no gentrifying effect and brought ‗green' housing to the poor.

Discussion
In France, which remains one of Europe's most centralized states, national regulations have played a decisive role in the implementation of ‗green' building. Even the locally implemented urban eco-districts may be seen as an initiative of the State, which viewed the eco-quartier label as a way to regulate local ‗green' neighborhood initiatives (Head of Sustainable Planning Unit, French Ministry of Housing, the Equality of Territories and Rurality, personal communication, 2015). Each of the three main stakeholders involved in eco-quartier design, promotes his own agenda: (1) the E.U. aims to reduce greenhouse gas emissions (through ‗green' building), (2) The French state implements its social housing regulations and, (3) the municipalities wish for ‗gentrification (to collect more taxes and reduce social expenses) but often also pursue wider construction activities and population growth. These divergent agendas may explain the disjunction between policy discourse and implementation. There is a national funding system for subsidizing public housing construction (the SRU law), and more than half the environmentally-certified apartments are defined as public social housing under that scheme. Additionally, as an EU country, France benefits from European funds earmarked for improving the energy performance of public social housing, as part of the European Energy Performance of Buildings Directive-EPBD (2002;. One way to implement the EPBD was to establish a national framework for eco-districts, whereby municipalities received budgets from the European Union (through the State) subsidizing the construction of ‗green' and affordable apartments. In France, the Eco-quartier framework stipulates social diversity. As mentioned, its initiators emphasize that proposing measures promoting social mix is a prerequisite for project approval. According to eco-district program requirements, the local authority had to verify (until 2015) -whether housing location and prices are appropriate to household's income‖ (Certification file, 2015) (Note 24). In contrast to the LEED ND certification, in the French label social mix and housing affordability are mandatory eer.ccsenet.org Energy and Environment Research Vol. 10, No. 1; elements. Thus, the eco-quartiers are mandated to have at least 40% public social housing. One might say that the French state authorities want to build ‗green' and affordable housing. However, local decision-makers manipulate funding methods, with the status (subsidized housing for low, middle, upper middle classes) of the accommodation and the size of the apartments, while attempting to generate some social diversity (some apartments are rental social housing). Moreover, regulations specify that municipalities in deficit must include ‗HLM' construction in their urban development projects. Thus, more than half the BBC apartments are social housing. However, as the law fails to specify the shares of social housing types, the numbers of units for low-income families are very small in attractive locations, like Paris (5% average). More generally, in affluent cities, where property prices are high (like in Paris, its well-to-do suburbs and some large cities), municipalities build eco-quartiers in substandard neighborhoods, to attract middle class families. In average cities, like Bretigny sur Orge, municipalities may implement more social housing than planned, especially if they are socialist municipalities and thus less reluctant to host low-income populations in eco-quartiers. In the poorest French towns, like Reims, the eco-quartiers will improve living conditions, but remain irrelevant to the social mix. In other words, poor and middle-income municipalities build a large share of social housing in their eco-quartiers, due to a lack of private investors willing and able to invest their own capital in such municipalities. Attracting well-to-do residents is only possible in Paris, its closer suburbs and other affluent cities, where the share of low-income populations in eco-quartiers is minimal.
In the last revision of the Eco-quartier label guidelines (2016) (Note 25), the ministry deleted the requirement that the local authority verify -whether housing location and prices are appropriate to household's income‖ (Certification file, 2015), as it was considered -too complicated to fulfill for the municipalities‖.
Most eco-quartiers are located in average income cities and none in the wealthiest ones (which need no eco-district to attract middle class residents). However, we find some eco-quartiers in Paris and in some close suburban localities (such as Boulogne) where the annual income is well above the median of 20.150 euro/year (see Table 2 in Appendix).

Conclusion
France entered the Eco-Quartier planning process in 2008, after the Northern European Countries, but since the early 2000s, more than 1,000 municipalities have applied for the eco-quartier label. The multiplication of ‗green' neighborhood projects shows a growing concern of French authorities for the environment, as well as a drive among municipalities to obtain funds for construction purposes.
The specificity of the French eco-districts remains the social mix requirement, countering socio-spatial segregation. Avoiding the repetition of past ‗mistakes' (such as concentrating poor people in large HLM projects) is essential in the view of French planning authorities. The eco-quartiers are usually located in substandard districts of large cities or average income municipalities, and their role is to attract middle class people, seen as guarantors of social cohesion. However, social mix goals are barely reached. In affluent cities where property prices are high (like Paris, its suburbs and some large cities), the municipalities build eco-quartiers in substandard neighborhoods to attract middle class families. In average cities (such as Bretigny Sur Orge), some municipalities have implemented more social housing than planned, to obtain State subsidies and loans for developers (thus the greening was motivated by budget considerations)but can still privilege the middle class in that housing's allocation. The high share of social housing is due to these municipalities' difficulties in attracting middle-class customers to buy the private apartments and to their willingness to accept some poor residents, if that's what it takes to encourage further construction. In the poorest French towns, like Reims, the eco-quartiers will improve living conditions in substandard neighborhoods but remain irrelevant to social mix.
Attracting well-to-do populations is only possible in Paris, including some of its closer suburbs like Levallois Perret and Boulogne, and other affluent cities (for example Bordeaux), where the share of low-income populations in the eco-quartiers will be minimal. In the wealthiest large cities, ‗green' housing has a gentrifying effect. However, in the municipalities of middle and lower-income (such as Mulhouse or Tours), eco-districts cause no gentrification.