Refl ections on the economic aspects of multilingualism in Transylvania

Using an economic perspective, the study deals with fairness problems associated with language use in multi-ethnic Transylvania. It showcases empirical fi ndings to demonstrate that asymmetric bilingualism in its current state creates economic disadvantages for linguistic minorities. Among the theoretical solutions off ered by the economics of language, it examines the possibility of the introduction of the English, as a lingua franca and some of the institutional conditions for a symmetric bilingualism. It argues that these scenarios are highly improbable to be accepted by the linguistic majority especially if they are backed by confrontational ethical arguments used so far. As an alternative it analyses the promotion of multilingualism through the economy with reference to the extra revenues and aggregate welfare benefi ts ethno-linguistic diversity can have for the society as a whole.


Asymmetric bilingualism and chance inequalities
Th e main thesis of our analysis is that within this institutionalist perspective, in Transylvania, a greater part of the costs of economic integration is borne by the minorities, as they learn the language of the majority. If the effi cient functioning of the economic institutions and its coordination mechanisms are considered public goods, the members of the majority are freeriders, since they have to invest less in order to participate in the economic communication. 1 Due to this asymmetry, members of the linguistic majority, the Romanians, have advantages, because (Grin 2004): -Th ey save the learning costs of Romanian. According to the experts of the economics of language (Piron 1994), for a near-native language profi ciency, a total of 12,000 hours of learning, education and exposure is required. 2 Th e costs are infl uenced by many factors, beginning from individual competencies (e.g. talent for languages), through the degree of diffi culty of the language and the level of exposure, to the eff ectiveness of teaching methods. On the latter, most of the experts regret that in Romanian public education not functional, everyday Romanian is taught, but Romanian literature and grammar (Szilágyi 1998). Although in the last fi ve-ten years substantial changes have taken place in this respect (e.g. the introduction of manuals and curricula specially designed for minority pupils), the methods that are used are still outdated, because they usually neglect a communicative-interactive approach, but build on deductive logic used in the education of classical languages, so they are over-concerned with grammar and text-centred knowledge of literature, and off er little room for practising the language (Ben 2012). Th is means that the learning of the language of the majority requires additional eff ort and investment, which is not covered by public funds. Th us, the catching-up process is fi nanced by the individuals themselves. Th is generates further inequality because it puts a disproportionately higher burden especially on the shoulders of those parents who live in majority Hungarian 1 Th e same is true for the global economy, the English as a lingua franca is used by more and more people worldwide. Th e US alone is saving 16 billion dollars annually since they do not have to teach any other languages for their pupils in the elementary and primary school. (Grin 2004: 200). We have no similar estimates for our case in Romania. 2 It is important to note here that not everyone needs to know Romanian at a near-native level, in order to succeed in life in this country. But there is a much higher probability for a member of a minority language group to get confronted with disadvantageous consequences arising from his/her language diffi culties.

BELVEDERE M E R I DI ON A L E
. .
areas, have lower education and more modest fi nancial conditions to fi nance the realignment of their children.
-Th ey do not have to count on the alternative costs of learning Romanian. Instead of learning Romanian, they can spend more time to learn other foreign languages, get a deeper knowledge in other domains or simply just give more time for recreation etc. Th e same applies to those who assist in language learning. Th is aspect is discussed in more detail in Chapter 2.
-Th ey can save on communication costs. Th ese costs are present in all the formal, institutional interactions where people with diff erent mother tongues get in contact with each other, as the messages should be translated to the dominant, offi cial language, which oft en has to be paid for.
-Th ey have a legitimacy and rhetorical advantage. Th e speakers of the dominant language have an advantage in reasoning and bargaining, because the conversation takes place in their own language.
Th is phenomenon is further reinforced by the fact that in Romania the offi cial, "titular", default language is Romanian, and minority languages are of a lower, "marked" status (Brubaker et al. 2006). Th e latter also implies that the lack of adequate Romanian language skills leads to worse labour market opportunities. Worse positions, in their turn, may cause income disadvantages.
Several studies have already drawn attention to the inequalities of employment and income between the Transylvanian Hungarians and Romanians. According to these, Hungarians are under-represented in higher-status occupations (Veres 2015), and their income level is lower than the Transylvanian average (Kiss 2014).
Using the international Labour Force Survey data we can follow the ethnic diff erences of income conditions and their evolution over time (Table 1). Th e earning statistics confi ned exclusively to employees show that in 2012 only 9 percent of Hungarians belonged to those who are in the highest earning one-fi ft h of the population in Transylvania, while almost 32 percent of the Hungarians belong to the lowest-paid one-fi ft h. Th e results also show that between 2008 and 2013 the proportion of Hungarians signifi cantly increased in the bottom quintile, while fewer and fewer people are earning as much as the upper one-fi ft h of the employees. Th is refl ects a surprisingly clear and rapid deterioration of the income situation of Hungarians in Romania. do not provide an opportunity for this. It is possible, however, to examine whether incomplete Romanian language skills cause earning disadvantages among the Hungarians. A descriptive analysis on a 2009 survey 3 reveals that those Hungarians who speak good Romanian earn 34 percent more compared to those who do not speak well or do not speak at all. A greater part of the diff erence, however, is not explained by the disparity of language skills, but rather by the factors that correlated with the latter: the respondents' gender, level of education and place of residence (urban or rural, inside or outside Szeklerland). In order to see more clearly, we included these variables in a joint linear regression model (Table 2). Th e result shows that the lack of Romanian language skills has a signifi cant infl uence on the income, even if we control for these factors. All other things being equal, the lack of language skills or insuffi cient knowledge of Romanian negatively aff ects the incomes of native Hungarians. Moreover, this eff ect among the highly educated is not valid, it causes income diff erences, however, among those who have baccalaureate at the most. So the lack of appropriate Romanian language skills causes income loss among the less educated, further deteriorating the situation of the most disadvantaged groups. In the explanation of the diff erences in employment statistics perhaps it is even more diffi cult to determine the impact of the Romanian language skills. However, earlier qualitative studies show that Hungarian youngsters are more prone to "under-plan" their career tracks: Hungarian university graduates get hired in the competitive private sector at a lower rate (Csata -Dániel-Pop 2006) and in their career narratives an increased cautiousness related to some perceived shortcomings regarding their Romanian language skills oft en appears (Csata et al. 2009). 4 At the end of this section it should be mentioned that, obviously, the invested eff ort of the Hungarians to learn Romanian, sooner or later could pay and they could gain a comparative advantage from the fact that they know one more language. Our data is not suited to confi rm 3 "Th e Turning Points of our Life Course 2./Életünk Fordulópontjai 2." survey was carried out in 2008-2009 by the Hungarian Demographic Research Institute (Budapest) in cooperation with the Romanian Institute for Research on National Minorities (Cluj Napoca/Kolozsvár) and the Max Weber Centre for Social Research (Cluj Napoca/Kolozsvár). Th e stratifi ed, multistage random sample was representative for the Hungarians living in Transylvania by gender, age groups and territorial distribution. A total of 4017 persons was interviewed, and the margin of error was +/-1,5% with a confi dence level of 95%. We used this not-so-recent survey data because of a higher number of cases and higher reliability. 4 Th us, it would be a simplistic and superfi cial reasoning to explain the occupational inequalities with an (intended or unintended) institutionalized discrimination of the minorities.
this hypothesis and there is a need for a more nuanced further analysis. Th e results only show that: 1. the income and occupational situation of the Hungarians is less favourable compared to the Romanians', 2. among the Hungarians the higher income is partially explained by a better knowledge of the Romanian language. Th ese two observations, however, seem to be enough to conclude that those Hungarians who have less exposure to the Romanian language and have more modest conditions to learn it, are disadvantaged beyond their control, compared to the members of the majority society.

Alternatives for the elimination of chance inequalities
Th e asymmetry of the status of languages thus generates chance inequalities between the linguistic minority and majority. Th e economics of language literature suggests that there are two real solutions for this equity problem: -the use of a third language, a "lingua franca" -diff erent from the language of the minority and majority -both in formal and informal communication -symmetric multilingualism: when the languages used in the region are given equal status in all interactions. 5 Th ere is a need for any other alternative program that decreases the aforementioned chance disadvantages (e.g. more effi cient teaching of the Romanian language, bearing the costs of the realignment as a common charge, introducing bilingualism in public institutions etc.), they could improve the equity, but they do not fully solve the problem. So let us examine these two options above in the case of the Hungarian minority in Romania.
Th e use of a third language in communication (e.g. English) is quite costly; and a series of professional arguments warn us that it could endanger the organic development or even the survival of local and regional languages (Skutnabb -Kangas 1999, Phillipson 2010. In spite of this, it is a rapidly spreading practice in Transylvania as well, especially at multinational companies and in intellectual professions. Th e fact that more and more people are getting familiar with the everyday use of English can further accelerate this process and it can show up in other areas as well 6 .
However, the introduction of the English as a lingua franca -along with the exclusivity of the Romanian as the sole offi cial language in the country -does not necessarily solve the chance inequality problem mentioned above. For example, if in addition to English, the Hungarian children must continue to learn Romanian as well in order to succeed in life, the asymmetrical relation will persist, since the Hungarian children will still need to learn one more language compared to the Romanians. Even if we assume that aft er learning a second language it is easier to learn a third one, the aggregate eff orts of Hungarians to get along with inter-ethnic communication will be higher than those of the Romanians'. In other words: if the Hungarian and Romanian children spend the same amount of time to learn other languages than their mother tongue, the Romanian children will probably speak better English, because in the case of Hungarians part of this time should be spent learning Romanian. Th is is one of the alternative costs which we referred to in the previous chapter.
Studies on foreign language skills among youth in Romania support this hypothesis: Hungarian-Romanian bilingual high school students under-perform in English tests compared to monolingual Romanians (Iatcu 2005, Molnár 2008). In relation to this, however, a number of alternative explanations must be taken into account. One is that it is easier for a Romanian pupil to learn English because Romanian closely resembles English. But both studies revealed that Romanian children even knew signifi cantly more English words that do not resemble their Romanian equivalents (non-cognate words). A further alternative explanation is that in Romanian schools/classes the quality of foreign language education could be higher. Finally, slight diff erences in the exposure to English could exist. 7 Th ese latter two hypotheses therefore need to be verifi ed. Nevertheless, in all of these assumptions the initial one appears to be the strongest, that the English language skills of Hungarian students are worse than the Romanians', because they have to learn two foreign languages instead of one, and they spend time learning Romanian at the expense of English. Th eir disadvantage could only be partially compensated by the fact that due to the early compulsory learning of Romanian, their meta-linguistic competencies could be more developed, thanks to which it might be easier to learn a third language (in this case English).
Further analysis shows that this disadvantage does not even out aft er studies either. According to the Mozaik 2001 survey data, among those Hungarian youngsters who live in central and western Transylvania and who started to work, only 21.8 percent declared that they spoke English, compared to 26.1 per cent registered among the Transylvanian Romanians. Among the Hungarians in Eastern Transylvania -Szeklerland -this ratio was even less, only 7 percent.
According to the 2009 Etnobarometer survey on an adult sample, 36 percent of the Romanians and only 22 percent of the Hungarians considered that their English was good enough to participate in a conversation. 8 As a conclusion, we can state that as long as the Romanian language will continue to be compulsory subject to learn for the minorities, the introduction of English as a lingua franca probably would not solve the fairness problem caused by the asymmetrical bilingualism, since it seems that the Hungarians accumulate a further disadvantage compared to the Romanians, regarding their foreign language skills. Th e situation would radically change, however, if Romanian would be permitted to be completely replaced by English in inter-ethnic communication. Th is option -although it might sound absurd -is already in practice in some countries and brings us closer to linguistic equality. 9 Th e institutionalization of symmetrical multilingualism in contemporary Romania seems just as unrealistic as the offi cial introduction of English as a lingua franca. But this is not only due to the fact that the political context is unfavourable and the majority of Romanians in Transylvania are reluctant to learn the language of the minorities. Th e following factors have also a great infl uence: -Th e institutional costs of multilingual education for the dominant language group are usually overestimated. Yet, depending on the region's characteristics, these represent only 5-10 per cent of the budget for education (Grin 2004) 10 . So making the education multilingual is primarily not a question of money.
-At various levels of political and public life, representatives of minorities formulate almost exclusively confrontational ethical arguments based on some kind of perceived historical legitimacy (e.g. "Romanians promised autonomy for the minorities aft er Trianon and they should keep their word"), or existing international practices based on progressive perceptions of democracy (Catalonia, South-Tyrol etc.). Economic arguments -referring to cooperation instead of interest struggles and destructive competition, focusing on the chance disadvantages of linguistic minorities mentioned above and the possible benefi ts of ethno-linguistic diversity listed below -occur signifi cantly less frequently in these debates. Instead, the "us" and "them" dichotomy, the obsession of the zero-sum game where the minority can only win at the expense of the majority and vice versa, continues to be symptomatic.
-In the Hungarian public space in Romania there is a strong expectation that these problems should be solved by political representatives primarily through legislation. Yet it is obvious that favourable laws alone cannot solve the problem without local will: for example the law about multilingual administration beyond the 20 percent threshold is impossible to be fully respected if resources for its implementation are not assigned (e.g. to increase the number of local government employees who speak Hungarian) 11 .
-Th ese expectations are usually system-wide, they expect solutions from macro-level reforms. Less attention is paid to those successful local, grass-roots civic and individual initiatives that could become possible models for a wider diff usion.

Th e demand side
A "grateful" terrain for these bottom-up, grass-roots attempts, articulated along these pragmatic arguments, could be the consumer market. Here buyers could effi ciently (and since they have the money: with impunity and without risks) signal if an important element of their identity -in this case the linguistic aspect -is not endorsed properly in business policy. Th ese kinds of consumer actions could start with addressing the seller in minority language, through expressing dissatisfaction if there is no multilingual information, customer service etc., to the point that they choose other products or services that meet their expectations in this regard. 12 Although this practice seems fairly simple, they occur very rarely and sporadically. 13 Behind the lack this proactive attitude, the "no action", we could identify three main reasons: -Some of the customers belonging to a minority language group do not call for multilingualism, it is not important for them. Th ey might get along in Romanian, and it is oft en the case that the use of Romanian is taken for granted, seems natural, and most importantly, it is thought as legitimate. Th is produces linguistic hegemony on the market.
-Th ey are afraid that their action (e.g. asking for a product in Hungarian in a Romanian 10 In Canada the full multi-language "operation" of the Quebec region costs less than 0.5 per cent of the federal budget. (Canadian Heritage Languages Institute Act 1991). 11 For a review of the practical application of the administrative language policy, see Horváth 2009. 12 In the economics, this practice is also called ethical consumption (Shimp 1984). 13 For further details, see: Szilágyi 2014.

. .
shop) will have unpleasant consequences. Th e same may apply to the provider, who may fear that if he/she addresses the consumers in Hungarian as well, he/she will lose Romanian customers. It is an important task for upcoming research in economic sociology and economic psychology to estimate the probability of the occurrence of these cases in diff erent regions and socio-demographic groups in Transylvania.
-Th ese actions have opportunity costs: it is time-consuming to wait for a seller who speaks Hungarian, to write a complaint, to post on a forum, to look for another shop/provider, etc. Th ese alternative costs are heavily dependent on 1) the proportion of minority language speakers in a given town/region, 2) how well the minority groups speak the language of the majority.
It is also important to know that at a given number of service providers, the marginal cost of these additional eff orts decreases rapidly. Th e introduction of multilingualism in the economy could be perceived as an innovation and in every business innovation there are signifi cant expenses/eff orts in the initial stages, until their diff usion as good practices do not accelerate. During the stages of innovation and early adoption (Rogers 2003) the individual alternative costs could be eff ectively reduced with a community campaign, which would draw the attention on the benefi ts of multilingualism in the economy, both on the provider and consumer side. Such a campaign would probably succeed, because at least at the level of dispositions it is clearly palpable an economic ethnocentrism among the Hungarians in Transylvania: if it is possible to choose, they are rather willing to buy products and services manufactured and/or commercialized by "Hungarian" fi rms (Csata -Deák 2010).
Compared to the relative indiff erence of the customers on the "demand" side, on the "supply" side a number of successful initiatives have appeared in recent years. Using a rather utilitarian logic, these initiatives promote multilingualism in the economy because they either expect extra revenues (fi rms) or they recognized the aggregate welfare benefi ts associated with linguistic diversity (NGO-s). In the rest of the paper we will display a few of these initiatives, will highlight the antecedents of their appearance, and we will present their underlying narratives. Following the above mentioned logic, these initiatives will be considered social innovations, and if they prove to be successful, chances are good that others will copy them. Th eir signifi cance therefore goes beyond whether they are viable as business models or not: as a "latent eff ect" or a "positive externality" they succeed to introduce multilingualism in the economy through the "back door".

Th e supply side: good practices in the promotion of multilingualism through the market
Th e appearance of new transnational regulations in the last decade set the ground for new institutional conditions, "opportunity structures" for ethnic Hungarians in Romania, to use the economic institutions and cooperation as effi cient means for promoting multilingualism in Transylvania. It seems that the supranational control over market regulations apparently off ers more room for the articulation of ethno-specifi c needs in the economy including a more frequent use of minority languages in marketing communication, consumer service, the linguistic landscape of trade etc.
In the promotion of minority languages through the economy we distinguished two distinct narratives. According to the fi rst, multiculturalism and ethno-linguistic diversity could be a comparative advantage for the economy, its professional management, exploitation and marketing can contribute to the improvement of the aggregate welfare of the whole society. Th e other approach is built on the conviction that the collective experience of living in a minority could have economically convertible advantages. Th ese perspectives lie on diff erent theoretical grounds, therefore we will start with their presentation, and then we will turn to those economic and civic initiatives which -consciously or spontaneously -are using these tools in their activities.

... Multiculturalism and aggregate welfare
Th e social science research of the relationship between ethno-linguistic diversity and economic development has begun to make dynamic developments in recent decades. Among the scholars there is a convincing consensus that developed democracies and economies are able to productively "handle" ethnic and linguistic diversity, and reduce or even nullify its negative eff ects (Collier 2000). Or, as Page (2008: 14) put it simply: "We fi nd that in advanced economies, ethnic diversity proves benefi cial. In poorer countries, it causes problems." Th is positive relationship has been confi rmed at lower levels of analysis as well: in major cities (Ottaviano -Peri 2006, Putnam 2007, Bellini et al. 2013, markets (Levine et al. 2014) and workplaces (De Vaan et al. 2011, Kochan et al 2003. Representatives of a strongly interdisciplinary approach known as the "economics of language" consider that multilingualism, by itself, is a value and can have a positive eff ect on economic development. For instance, according to Grin, Sfreddo and Vaillancourt (2010), about 10 percent of Switzerland's GDP is due to linguistic diversity. Th us, the 0.5 percent that is spent on children's multilingual education appears to be a rather good investment (Grin -Vaillancourt 1997), even if the development of multilingual communication in institutions may incur further costs.
In those Central and Eastern European countries that joined the European Union in the last decade, signifi cant changes have taken place regarding the institutional enforcement of the common law on economic cooperation. Th ese changes sought the implementation and standardization of the EU regulations and the institutional improvement of control, prevention of corruption, opportunism, the free rider and rent seeking activities. It seems that the abovementioned institutional conditions for the exploitation of aggregate economic benefi ts stemming from ethno-linguistic diversity have been created.
Here, we are not talking about the direct promotion of multiculturalism in the European Union. Cultural and human resource development programs supported by the EU are recommended to have components that promote multiculturalism, and the benefi ciaries are urged to meet this requirement. Th erefore, following Romania's EU accession in 2007 the public diff usion of this offi cial canon has started, it became widespread in a growing number of political and public declarations, and it is one of the buzzwords of diff erent development projects and their host institutions. Beyond public communication and PR materials, however, this approach is barely present in the practical implementation of the programs. 14 Th ese initiatives are also ineff ective because they lack the allusion to the aggregate welfare benefi ts multilingualism/multiculturalism can have for the society as a whole (Grin 1999).
Th ere are very few attempts that go beyond considering multiculturalism as a value in itself and try to attribute economic benefi ts to diversity. One of these initiatives is the "Igen, tessék!" ("Yes, please!") movement, 15 which has been launched in 2012 in Cluj Napoca/Kolozsvár, but it is expanding to other Transylvanian towns as well. Th e movement aims to encourage communication in native language in commercial life and public spaces (shops, markets, coff ee houses, 14 A good example for this is the Cluj-Napoca Youth Capital 2015 project. One of its websites which promotes multiculturalism (http://clujmulticultural.ro) had no Hungarian version for a long time and while they showcased detailed reports about immigrant families and communities in Cluj, they omitted to mention the 50,000 Hungarians living in the city. Aft er several online protests a Hungarian translation of the website became available. Th e offi cial website of the city's administration, however, continues to be monolingual, Romanian. 15 "Da, poft iți!" in Romanian.

BELVEDERE M E R I DI ON A L E
. . cashiers etc.). An economic network has been established of those fi rms that serve their customers in their mother tongue. Th e initiative is based on the philosophy that the use of native language in everyday interactions gives extra comfort for the participants/consumers, leaving them with favourable impressions about shopping. Th erefore they urge the commercial businesses to recognize: if they are trying to communicate with their customers in their native language -and if they take into account the needs of diff erent regions with diff erent cultures -then they can expect more revenues. Being supporters of culturally sensitive marketing, they use economic arguments and strategies to encourage sellers to practice multilingualism and they are off ering a variety of marketing interfaces for this purpose (stickers in the windows of shops, monthly community magazine, website and interactive mobile application). Until now the movement has only focused on marketing among Hungarian-speaking customers in Romania. Although they intend to expand their services to other minorities in Romania and to the Romanian communities in neighbouring countries (Ukraine, Hungary), this has not yet happened.
In their communication strategy towards the Romanian majority, the members of the organization emphasize that multilingualism and multiculturalism is an important source of returns. As a result of diversity, products and services containing greater added value and thus more attractive to consumers, become accessible. As there is an increasing demand for varied, innovative services (e.g. in gastronomy, music, etc.) ethnic diversity increases consumer satisfaction and it generates a positive amenity eff ect as an externality. Th is same positive eff ect can prevail in the development of public goods and public services as well, through which the amenity value of cities and regions could increase.
... Economic advantages of minority status Economic sociologists, especially those around the "new economic sociology" school believe that the success of the economic exploitation of ethno-linguistic diversity depends not just on the existence of appropriate institutional conditions (democracy and well-functioning bureaucracies). Cultural characteristics and social organizational patterns of individual minorities in the same region can also lead to varying economic strategies and outcomes (Aldrich 1990, Granovetter 1995, Portes -Sensenbrenner 1999, and Light -Gold 2000. Th eir success or failure, in their turn will determine the aggregate welfare of the whole society. Th us, more attention has to be paid to the structural, networking and cultural resources minorities possess, and to what extent and how do they succeed in utilizing these in the economy, for their benefi t and for the benefi t of the society as a whole. Granovetter (1995) localized those minority community resources which could represent a comparative advantage vis-à-vis the majority and other immigrant minorities. He classifi ed these benefi ts into four categories: 1. cultural advantages: some social norms might be relevant to the majorities, but not to the minorities, which creates an unadulterated market opportunity for the latter. 2. networking advantages: valuable market information and opportunities, access to multiple resources through bridging or broker positions (Kim -Aldrich 2005) in "structural holes" (Burt 1992. 3). advantages stemming from solidarity: "bounded solidarity", "enforceable trust" (Portes 1998), "bonding" social capital (Putnam 2000). 4. advantages arising from marginal situations: In some cases the minority is not bound to satisfy local traditional obligations. It can employ new, more competitive commercial techniques without risking the danger of (further) ostracism and sanctioning.
Since the historical minorities in Central and Eastern Europe signifi cantly diff er from the American immigrant groups in many respects, some of the resources cited by Granovetter are not relevant for the Hungarians in Romania. Aft er centuries-old coexistence there still might

. .
be some diff erences in everyday culture of Hungarians and Romanians in Transylvania -most of them attributable to the diff erences of language and religion -but in the context of globalization these are far from being spectacular. Furthermore, aft er the change of regime the formally institutionalized discrimination against the Transylvanian Hungarians was withdrawn. So in these dimensions (1,4) we can hardly talk about economically useful comparative advantages. Th e bonding and bridging capital of ethnic Hungarians, however can carry signifi cant economic potential. In the following, we will take a look to a few relevant examples in this respect.
In Transylvania there is an increasing number of entrepreneurial initiative under way, which, appealing to the ethnic solidarity of Hungarians, attempt to gain a competitive market advantage (Gáll 2011). Th e popularity of local products specifi cally positioned as Hungarian brands in Szeklerland, for example, shows that consumer ethnocentrism is not only present at the level of dispositions, but rather increasingly determines the purchasing decisions of locals as well.
Th e biggest player in this market is undoubtedly the Merkúr supermarket chain, created in 2007 from a formal local convenience store chain in Odorheiu Secuiesc (Hun. Székelyudvarhely). With its twelve locations in fi ve towns in Szeklerland, Merkúr is a successful competitor of multinational retail stores in the region (Kaufl and, Lidl, Penny Market etc.). Th e company currently has more than 700 employees and in 2014 their turnover reached 50 million euros.
A signifi cant part of the revenues comes from the Góbé product line, which is commercialized as a kind of own brand using a unique marketing concept. Under the same image they bring together 350 products made by 64 individual producers exclusively from Szeklerland. Although the Góbé behaves like a trademark, since it promotes and sells the products of locals, it is rather a community brand. Th e Merkúr advertises itself as the Szekler store chain, the logo and the image elements on Góbé products (containing stylized elements about a fi ctional traditional Szekler village) refer exclusively to this region.
Although the management defi nes Góbé as a regional brand, which any producer from Szeklerland could join, for now the only suppliers are local Hungarians. So it is not a coincidence that the products carry an added value based on Hungarian ethnicity. Th is identity is further reinforced by the fact that in Merkúr supermarkets Hungarian products made in Hungary are also commercialized in a large scale.
An even more pronounced commodifi cation of ethno-regional solidarity takes place in the case of a recently created Szekler product, the "Igazi Csíki Sör" ("Th e Real Ciuc Beer" in Hungarian). Th is brand was established in 2014 and in the creation of product image they used symbols and narratives inspired exclusively from the history, folklore and cultural heritage of Szeklerland. Although the business model was a success from the very beginning, the beer brewery became widely popular when the Dutch multinational company, Heineken sued the producers for using the Igazi Csíki Sör brand name. Heineken argued that they patented earlier the Romanian translation of the Csíki Sör, the Ciuc Premium which has been produced for decades in the Szekler town of Miercurea Ciuc (Hun. Csíkszereda).
For the Igazi Csíki Sör, the dispute seems to come in handy, because it gave the impression to the consumers that this is a David-Goliath battle between the Szekler beer brewery and the profi t-oriented, heartless multinational company who is pushing the local producers to inability (Sipos 2014). Th anks to the cleverly constructed narrative and campaign, an online community advocacy group was quickly formed around the Igazi Csíki Sör, who called for a collective boycott of Heineken products. Th e boycott became widespread aft er the Heineken management committed a mistake making a statement where they were questioning the very existence of Szeklerland.
Th e sales of Ciuc Premium dropped, 16 the legal dispute around the brand is still going on and the Igazi Csíki Sör enjoys an unbroken popularity among Hungarian consumers in Szeklerland.
Both of these examples illustrate well that the business model built on the bounded solidarity and enforceable trust of Hungarian population in the region is successful in Szeklerland and it has more and more followers. Th e success could also be attributed to the fact that beyond the ethnic markers, these product concepts successfully integrate other elements of ethical consumption as well (regionalism, anti-globalization, environmental awareness, biodiversity etc.). Th erefore the phenomenon could be legitimately considered as a form of collective manifestation of ethical consumption. 17 Th ese examples show that Transylvanian Hungarians (and particularly those from Szeklerland) enjoy advantages stemming from "bounded solidarity" and it seems that the "bonding" type of social capital has an increasing economic utility. Moreover, from an anthropological perspective, it is particularly interesting that viable Hungarian companies, brands, products and economic cooperation practices also contribute to the further reinforcement of ethnic-regional identity. Furthermore -and this is important from our point of view -on the ground of market deregulation, using classical instruments of consumer marketing they spontaneously contribute to the development of multilingualism in the economy. We think that without the enforcement of the laws of free competition by the European Union, this process would have encountered more obstacles from the Romanian authorities.

Conclusions
In this study we argued that the problem of multilingualism is not just a matter of principles, it is a practical issue as well, since the acquisition of the majority language puts a disproportionately higher burden on the members of the linguistic minority, and the status diff erences of languages generate chances inequalities.
An equitable solution to this problem would be 1. either to replace the lingua franca with English (or with another "third" language), 2. or to create the conditions for a symmetric multilingualism. Th e latter would mean that the languages used in the region would be given equal status in all interaction situations. Th is requires the majority to learn the language of minority(ies). Although this possibility doesn't seem to be realistic at the time being, along with the diff usion of local, grass-roots initiatives it can have more serious chances.
For this, however, a rhetorical shift is also necessary: besides the confrontational, militant discourse of ethical nature against the hegemony of majority language, greater emphasis should be given to economic considerations, which illustrate along rational arguments that the linguistic asymmetry creates chance disadvantages for minorities. An even more important component of this argumentation -based on co-operation and the search of consensus -is a more effi cient presentation of the fact that multilingualism is a social resource that could generate economic and welfare benefi ts for everyone, so it is worth investing in it, regardless of nationality.
Th e market deregulation that followed the EU membership opened up new possibilities for a 16 Another interesting element of the story is that aft er Heineken realized that they insulted the Hungarian customers questioning the existence of Szeklerland, they decided to use Szekler symbols on the label of their cheaper local brand, the Harghita. 17 Regarding the research perspective on ethical consumption and conscious consumerism see a recent analysis by Bartley et al. (2015).
"grass-roots", spontaneous expansion of multilingualism in the economy. Th ese "new" strategies on one hand argue that multilingualism is benefi cial not only for the minority, but for society as a whole in terms of aggregate welfare; and in an appropriate institutional environment, diversity has tangible (economic) benefi ts for everyone. On the other hand, they encourage economic actors to take the advantages given by the market and "capitalize" the solidarity of ethnic minorities: to communicate with consumers in their native language in the commercialization of their products and services and thereby to gain additional revenues. Th ese strategies are promising, because along legitimate customer needs, as "a latent eff ect" or a "positive externality" they introduce multilingualism in the economy "through the back door". ❋